eprints.soton.ac.uk · web viewuniversity of southampton. faculty of business and law. management...
TRANSCRIPT
UNIVERSITY OF SOUTHAMPTON
FACULTY OF BUSINESS AND LAW
Management School
The effect of Telepresence and Anthropomorphic attributes on consumers comprehension of RNPs: A study on consumer innovativeness and
anthropomorphism (measurement and application)
by
Mona Seyed Esfahani
Thesis for the degree of Doctor of Philosophy
December 2016
1
UNIVERSITY OF SOUTHAMPTON
Abstract
FACULTY OF BUSINESS AND LAW
Management School
Doctor of Philosophy
The effect of Telepresence and Anthropomorphic attributes on consumers
comprehension of RNPs: A study on consumer innovativeness and
anthropomorphism (measurement and application)
By Mona Seyed Esfahani
The advancement of technology imposes an inevitable pressure on companies to
introduce new products and services into the marketplace, to stay competitive or
survive. One product category that is growing increasingly within the marketplace is
Really New Products (RNPs), which refers to very innovative products. Businesses
therefore need to be aware of the ways they can promote RNPs in order for consumer
to efficiently understand RNPs and form a positive attitude and intention towards
these products. This study is concerned with the product promotion element within
the domain of RNPs. Various presentation techniques such as telepresence (vividness
and interactivity) and anthropomorphic attributes are discussed within this thesis.
Furthermore, the influence of targeting different groups of consumers (innovative
consumers) is examined. Anthropomorphism is analysed in more depth to get a better
understanding of how this factor influences consumers learning and online behaviour.
Three papers are formed to investigate each category further. Two online experiments
and one survey are designed. The first experiment recruited 800 participants to
examine the effect of presentation formats on individual responses towards RNPs
within Paper 1. The findings from Paper 1 indicate that various presentation elements
have a different impact upon consumer learning, attitude and purchase intention for
2
RNPs. The second set of online experiments within Paper 2 with 500 participants
investigated the impact of various anthropomorphic attributes and its influence on
consumer response towards RNPs. The result indicates that the inclusion of human-
like avatars increases an individual’s perceived anthropomorphism. Furthermore,
perceived anthropomorphism has a significant positive influence upon consumer
learning, attitude and purchase intention towards RNPs. Paper 3 studied the influence
of consumer innovativeness and how consumers differ in their learning and behaviour
towards RNPs. 300 participants were recruited to answer an online survey. The
findings indicate that various types of innovative consumers react towards and learn
about RNPs in different directions. Each paper is thoroughly discussed and the
limitations, managerial implications and future research recommendations are
considered.
3
ContentsAbstract....................................................................................................................2
Declaration of Authorship.....................................................................................11
Acknowledgement.................................................................................................12
Abbreviation..........................................................................................................13
Chapter 1: Introduction..........................................................................................15
1.1 Thesis overview and objectives...................................................................15
1.2 Thesis Structure...........................................................................................16
1.3 Contribution to Knowledge.........................................................................17
1.4 Literature Review........................................................................................19
1.4.1 Product Categorization – Understanding Really New Products...........20
1.4.2 RNP and Product Presentation.............................................................24
1.4.3. Virtual Reality.....................................................................................25
1.4.3.1 Vividness.......................................................................................26
1.4.3.2. Interactivity...................................................................................28
1.4.4. Anthropomorphism..............................................................................30
1.4.5 Consumer Grouping – innovator adopters...........................................34
1.4.5.1 Diffusion of Innovation Theories - Background...........................34
1.4.5.2. Rogers’ Diffusion of Innovation Model.......................................36
4
1.4.5.2.1 Innovation...............................................................................37
1.4.5.2.2 Communication......................................................................41
1.4.5.2.3 Time........................................................................................41
1.4.5.2.4 Social System.........................................................................41
1.4.5.2.5 Consumer characteristics........................................................42
1.4.5.2.6 Product attributes....................................................................43
1.4.5.2.7 Communication Channels......................................................43
1.4.5.2.8 Social networks......................................................................44
1.4.5.3. Consumer innovativeness.............................................................44
1.4.6 Consumer Learning..............................................................................47
1.4.6.1 Theory of Multimedia Learning....................................................50
1.4.6.2 Social learning Theory..................................................................54
1.4.7. The Research Gap................................................................................56
1.5 Methodology................................................................................................59
1.5.1. Philosophy of Research.....................................................................59
1.5.2. Methodological design.......................................................................64
1.5.3. Research Design.................................................................................65
1.5.3.1. Experimental Design..................................................................66
1.5.3.2 Survey Design..............................................................................67
1.5.3.3. Study 1.........................................................................................67
1.5.3.4. Study 2.........................................................................................68
5
1.5.3.5. Study 3.........................................................................................68
1.6 Conclusion...................................................................................................69
1.7 Structure of the thesis..................................................................................71
Chapter 2: The effect of Telepresence and Anthropomorphic attributes on
consumer’s comprehension of RNPs...........................................................................73
2.1 Introduction.................................................................................................73
2.2 Theoretical Background..............................................................................75
2.2.1. New/innovative products.....................................................................76
2.2.2. Virtual Reality.....................................................................................77
2.2.2.1 Vividness.......................................................................................78
2.2.2.2 Interactivity....................................................................................80
2.2.3 Anthropomorphism...............................................................................81
2.2.4 Consumer differences that impact on learning.....................................83
2.3 Methodology................................................................................................84
2.3.1 Research Design...................................................................................85
2.3.2 Measurement........................................................................................88
2.3.2.1 Manipulation Checks.....................................................................88
2.3.2.2 Dependent Variables.....................................................................89
2.3.2.3 Covariate - Consumer Innovativeness...........................................90
2.3.2.4 Participants....................................................................................90
2.4 Results.........................................................................................................91
2.5 General Discussion and Conclusion............................................................98
6
2.6 Theoretical Implications............................................................................100
2.7 Managerial Implications............................................................................101
2.8 Limitations and Directions for Future Research.......................................101
Chapter 3: Mindful and Mindless Anthropomorphism; how to facilitate consumer
comprehension, its measurement and application......................................................104
3.1 Introduction...............................................................................................104
3.2 Theoretical Background............................................................................105
1.1.1 Really New Products.......................................................................105
3.2.2 Anthropomorphism.............................................................................106
3.2.3 Content Interactivity...........................................................................108
3.2.4 Online agents - Avatars......................................................................109
3.2.5 Human-like Avatars............................................................................111
3.2.6 Static Avatars......................................................................................111
3.2.7. Mindful and Mindless Anthropomorphism.......................................114
3.3 Methodology..............................................................................................119
3.3.1. Perceived Anthropomorphism - Measurement..................................119
3.3.2. Research Design................................................................................120
3.3.3. Experimental Conditions...................................................................121
3.3.4. Data Collection and Sample..............................................................122
3.3.5. The Stimuli and Measures.................................................................122
3.4 Results.......................................................................................................123
3.4.1. Anthropomorphism manipulation.....................................................125
7
3.4.2. Mindless Anthropomorphism............................................................129
3.4.3. Mindful Anthropomorphism.............................................................129
3.4.4. Correlation between dependent variables..........................................129
3.5 General Discussion and Conclusion..........................................................130
3.6 Limitations and Directions for Future Research.......................................132
Chapter 4: Examining the impact of consumer innovativeness on RNP
comprehension attitude and purchase intention.........................................................134
4.1 Introduction...............................................................................................134
4.2 Theoretical Background............................................................................135
4.2.1. Diffusion Theory...............................................................................135
4.2.2. Consumer Innovativeness..................................................................138
4.2.2.1 Consumer learning and innovation..............................................140
4.2.3. Innate Innovativeness and Consumer Motivation.............................141
4.3 Methodology..............................................................................................145
4.3.1. Product selection...............................................................................145
4.3.2. Participants........................................................................................145
4.3.3. Questionnaire development...............................................................146
4.4 Results.......................................................................................................147
4.4.1. Measurement development and assessment......................................147
4.4.2. Hypotheses testing.............................................................................150
4.5 General Discussion and Conclusion..........................................................151
4.6 Managerial Implications............................................................................154
8
4.7 Limitations and Direction for Future Research.........................................154
Chapter 5: Conclusion.........................................................................................156
5.1 Review of Findings....................................................................................156
5.2 Contributions, Limitations and Future Research....................................158
5.2.1. Contribution; Paper 1- Telepresence.................................................158
5.2.2. Contribution; Paper 2 - Anthropomorphism......................................162
5.2.3. Contribution; Paper 3 – Consumer innovativeness...........................164
5.2.4 Un-hypothesized analysis...................................................................167
5.2.4.1. Statement 1- Utilitarian and Hedonic Product Nature................168
5.2.4.2. Statement 2- Gender...................................................................169
5.2.5 Common Contribution........................................................................171
5.2.6 Contribution to Theory.......................................................................172
5.2.7. Limitations and future research.........................................................175
5.2.7.1 Methodological Limitations........................................................176
5.2.7.2 Paper 1.........................................................................................176
5.2.7.3 Paper 2.........................................................................................178
5.2.7.4 Paper 3.........................................................................................178
5.3. Managerial Implications...........................................................................179
Reference.............................................................................................................187
Appendices..........................................................................................................216
9
10
List of Tables
Table 1 Learning Theories Supporting Study’s Variables 42
Table 2 Sample Demographics 80
Table 3 Reliability, Discriminant/convergence validity for
final model.
82
Table 4 The effect of covariate in vividness conditions 86
Table 5 The effect of covariate in interactivity conditions 86
Table 6 Demographics of Sample 111
Table 7 Summary of items removed from the model. 114
Table 8 Reliability, Discriminant/convergence validity for
final model.
114
Table 9 Test of hypotheses – Anthropomorphism
manipulation
115
Table 10 Test of hypotheses – Correlation between
dependent variables
117
Table 11 Final models 137
Table 12 Confirmatory factor analysis: loadings and t-value 137
Table 13 Descriptive statistics 138
Table 14 Test of hypotheses 140
Table 15 Review of Findings 146
11
Declaration of Authorship
I, Mona Seyed Esfahani declare that this thesis and the work presented in it are
my own and has been generated by me as the result of my own original research.
The effect of Telepresence and Anthropomorphic attributes on consumers
comprehension of RNPs: A study on consumer innovativeness and anthropomorphism
(measurement and application)
I confirm that:
1.1 This work was done wholly or mainly while in candidature for a research degree at this University;
1.2 Where any part of this thesis has previously been submitted for a degree or any other qualification at this University or any other institution, this has been clearly stated;
1.3 Where I have consulted the published work of others, this is always clearly attributed;
1.4 Where I have quoted from the work of others, the source is always given. With the exception of such quotations, this thesis is entirely my own work;
1.5 I have acknowledged all main sources of help;
1.6 Where the thesis is based on work done by myself jointly with others, I have made clear exactly what was done by others and what I have contributed myself;
1.7 None of this work has been published before submission.
Signed: Mona Seyed Esfahani
12
Date: February 2017
Acknowledgement
I would like to express my sincere gratitude to my supervisor, Professor Nina
Reynolds, for guiding me throughout my PhD journey. She supported me whole-
heartedly, giving feedback in her busiest time. My sincere thanks to Dr. Melanie
Ashleigh for all her guidance. Without her support I couldn't finish my thesis.
I am grateful to Dr. Finola Kerrigan, for all her guidance and kindness. Also
special thanks to all my wonderful colleagues at Bournemouth University Media
School. I greatly appreciate Dr. Richard Scullion’s support and encouragement
during this difficult time. I am greatly indebted to Shenel McLawrence for her
invaluable feedback and eagle eyes.
Finally, I would like to thank my husband and my daughter, for always being
there for me and their unconditional love, encouragement and support. I couldn't have
finished this journey without their help and patience. I am grateful to my parents, for
believing in me and supporting me throughout my education.
13
Abbreviation
RNP Really New Product
INP Incrementally New Product
CME Computer Mediated Environment
PDA Personal Digital Assistants
VR Virtual Reality
2D 2 Dimensional
3D 3 Dimensional
MCI Motivated Consumer Innovativeness
fMCI Functional Motivated Consumer Innovativeness
hMCI Hedonically Motivated Consumer Innovativeness
sMCI Socially Motivated Consumer Innovativeness
cMCI Cognitive Motivated Consumer Innovativeness
CATLM Cognitive-Affective Theory of learning with media
ANOVA Analysis of Variance
MANOVA Multivariate Analysis of Variance
MANCOVA Multivariate Analysis of Covariance
SPSS Statistical Package for the Social Sciences
CFA Confirmatory Factor Analysis
AMOS Analysis of Moment Structures
SEM Structural Equation Modelling
Mturk Amazon Mechanical Turk
CFI Comparative Fir Index
GFI Goodness of Fir Index
NFI Normed Fit Index
NNFI Non-normed Fit Index
IFI Incremental Fir Index
14
AGFI Adjusted Goodness-of-fit Index
RMSEA Root Mean Square Error of Approximation
CR Composite Reliability
AVE Average Variance Extracted
MSV Maximum Shared Variance
ASV Average Shared Variance
SW Shapiro-Wilk
KS Kolmogorov-Smirnov
MLE Maximum Likelihood Estimation
15
Chapter 1: Introduction
1.1 Thesis overview and objectives
Really New Products (RNP) are a new product category, referring to innovative
products that expand the existing product category significantly (Lehmann, 1994).
RNPs are rapidly growing in various industries and have become a priority for many
companies such as Procter and Gamble and Electrolux1 (Hoeffler, 2003; Brown and
Anthony, 2011). To date however, RNPs’ failure rate is very high. One reason for this
could be that existing market research strategies and frameworks are designed for
Incrementally New Products (INPs), rather than the RNPs to which they are applied.
Yet, there are major differences between INPs and RNPs. INPs are understood to be
based on the consumer’s existing knowledge base (Hoeffler, 2003); whereas for
RNPs, consumers have no prior information regarding these and therefore product
comprehension occurs at the time of product evaluation (Hoeffler, 2003). As such,
more learning is required for RNPs due to their innovative nature (Feiereisen et al.,
2008). Frameworks and models that apply to INPs cannot automatically be assumed
to be suitable for RNPs. Hoeffler (2003) explained, for instance, how existing market
research techniques were developed for INPs, so it was always assumed that the
results were only valid when consumers had some core knowledge relevant to the
product or product category (Hoeffler, 2003). Overall, as RNPs are evaluated at the
time of presentation, offering efficient and clear information about RNPs in a
competent format is crucial in facilitating consumer learning about RNPs.
The studies in this thesis consider the unique nature of RNPs in examining
existing consumer behaviour learning models and frameworks concerned with
consumers’ learning about RNPs in a Computer Mediated Environment (CME).
Variables explored in the study include the information presentation formats of
vividness, interactivity and anthropomorphic attributes. These are investigated to
understand how they can facilitate consumer learning and improve their online
behaviour. Learning theories have been reviewed to identify frameworks supporting
consumers learning about innovative products within different presentation formats.
Furthermore, consumers are being examined as technology adopters, in order to
1 http://www.electrolux.co.uk/Innovation/
16
understand how various presentation formats could affect different consumer
categories’ understanding of RNPs.
To summarise, the objectives of the present research which will be acknowledged
within the literature review are:
To identify the effect of various presentation formats of vividness,
interactivity and anthropomorphic attributes on consumer responses to RNPs
(i.e. product comprehension, attitude and purchase intention).
To examine whether the insertion of various types of anthropomorphic
attributes influence consumers’ perceived mindless and mindful
anthropomorphism while learning about RNPs.
To determine the relationship between consumers’ perceived mindless and
mindful anthropomorphism and consumers’ response towards RNPs (i.e.
product comprehension, attitude and purchase intention).
To examine the impact of consumer innovativeness (i.e. functional, hedonic,
social, cognitive) on consumers’ comprehension and attitude towards RNPs.
1.2 Thesis Structure
This thesis has a three-paper structure. The first paper considers the various
presentation techniques and how inserting vividness, interactivity and
anthropomorphic attributes to these presentation techniques, influence an individual’s
learning, attitude and intention towards RNPs (Objective 1). The second paper
investigates the anthropomorphic attributes further, in order to understand how
anthropomorphism is perceived (as mindless and mindful) by individuals within the
RNP domain, and how it facilitates consumers’ comprehension (Objectives 2 and 3).
The third paper is concerned with consumer innovativeness and its impact on RNP
comprehension and attitude towards RNPs (Objective 3).
Papers are presented in a stand-alone format. The studies are all based on the
same underlying literature. Academic literature, as well as theories and frameworks,
support the insertion of vividness, interactivity and anthropomorphic attributes in a
CME, and the impact each factor can have on consumer learning and online behaviour
towards RNPs. The literature review also reveals how various consumers, as
innovation adopters, can comprehend RNPs more favourably due to their underlying
17
motivational characteristics. The findings of each study add to the body of knowledge,
which is discussed further.
1.3 Contribution to Knowledge
The dissertation is concerned with three aspects. Paper 1 is about the effects of
adding telepresence attributes (vividness and interactivity) as well as
anthropomorphic attributes in a CME. Telepresence is a feeling that is experienced
when individuals are immersed within a virtual environment through Virtual Reality
(Steuer, 1992). Anthropomorphic attributes refer to characteristics enabling a virtual
object to behave or look like a human (Koda, 1996; Nowak and Biocca, 2003;
Nowak, 2004). The result of this study indicates that the manipulation of
anthropomorphic attributes was not successful; this suggests a need for a deeper
investigation into the operationalization of the anthropomorphic attributes within
CMEs, as well as a further exploration of anthropomorphism measurement scales.
Another suggested line of research arising from Paper 1’s results was to investigate
consumer innovativeness in more depth. Although academic literature indicates that
innovator consumers are more likely to try new products (Rogers, 1995; Goldsmith
and Lafferty, 2002), Paper 1’s results do not support this statement. Consumer
innovativeness was considered as a covariate within Paper 1, and this study’s result
indicated that although types of innovativeness were significantly different across the
conditions, the effect of the independent variables were the same when the analysis
was done with or without the covariate. The elements influencing consumer
innovativeness (such as motivation attributes) can therefore be explored, in order to
understand whether there are underlying factors impacting consumer behaviour
towards RNPs.
As a result of the above suggestions, Paper 2 performs an advance investigation
into online anthropomorphism, examining various ways of using anthropomorphic
attributes online, and how each attribute influences consumer perceived
mindless/mindful anthropomorphism. Paper 3 addresses the second research
suggestion; that is, it looks into the impact of consumer innovativeness upon RNP
comprehension and online attitude. The core contribution of the three papers within
this dissertation to the body of knowledge can be summarised as below:
18
Firstly, it adds to the Virtual Reality and specifically telepresence literature by
indicating the difference between this study’s results and previous findings,
which can be due to the unique nature of RNPs. The findings specify how
vividness, as the first characteristic enabling telepresence, has no effect on
consumers’ learning, attitude and purchase intention in the context of RNPs;
though vividness was proved to improve consumer learning in various
contexts (e.g. Paivio, 1971; Alesandrini, 1982; Rayport and Jaworski, 2001;
Zhang et al., 2006). Interactivity, as the second characteristic enabling
telepresence, proves to be influential on consumers’ learning, attitude and
intention towards RNPs, but it differs across RNPs. Interactivity was however
supported as an attribute improving consumers’ learning and online behaviour
within various contexts (e.g. (Brandt, 1997; Schlosser et al., 2003; Fiore et al.,
2005; Park et al., 2005; Zhang et al., 2006; Pantano and Naccarato, 2010). To
conclude, telepresence is not necessarily an influential attribute upon
consumers’ comprehension, attitude and purchase intention towards RNPs.
Secondly, it adds to the anthropomorphism literature considerably. The result
clearly indicates that within the RNP context, the inclusion of a human-like
avatar containing all four types of social cues identified within the consumer
behaviour literature (Nass and Steuer, 1993), results in an increased perceived
overall anthropomorphism; this is the direct result of a significant perceived
mindless anthropomorphism and not a perceived mindful anthropomorphism.
Consumers are therefore mindlessly perceiving the CME, including a human-
like avatar, as more anthropomorphic and not mindfully. Another significant
contribution is to the academic body of consumer learning and online
behaviour, linked to the anthropomorphism literature. The result clearly
indicates that perceived anthropomorphism has a positive relationship on
consumers’ learning, attitude and purchase intention towards RNPs.
Thirdly, it adds to the innovation adoption literature. Looking into various
motivational aspects of innovation adoption, the results indicate a positive
relationship between hedonic innovativeness and comprehension. Other
functional and cognitive motivational aspects did not have any impact on
consumer learning of RNPs. Hedonic and cognitive innovativeness were also
proved to positively influence attitude towards RNPs, supporting the positive
but weak relationship between innate innovativeness and consumers adoption
19
behaviour in previous literature (e.g. Goldsmith et al., 1995; Citrin et al.,
2000; Im et al., 2007). Social innovativeness proved contradictory to previous
findings, by indicating a negative impact on consumer comprehension of
RNPs. As no study to date explored the motivational aspects of consumer
innovativeness within the context of RNPs, the findings contribute greatly to
the innovation adoption literature by specifying the different direction each
motivational innovativeness source can take in impacting comprehension and
attitude towards RNPs.
Fourthly, the findings of this thesis contribute considerably to the RNP
literature. It sheds light on the academic body of knowledge concerning the
learning of RNPs, by examining presentation elements (e.g. telepresence and
anthropomorphic attributes) as well as consumer innovativeness factors (e.g.
motivational innovativeness of hedonic, functional, social and cognitive) on
consumer comprehension and online behaviour. This thesis also supplies
evidence to support the positive influence of comprehension towards attitude
and purchase intention within the RNP context.
1.4 Literature Review
In this section, the main factors of Product Categorization, Product Presentation,
Virtual Reality, Anthropomorphism, Consumer Grouping and Consumer Learning are
explained and analysed in more detail.
Product Categorization is concerned with RNPs as a new group of products,
which emerged due to introduction of innovativeness. This section explores other
product categories and how RNPs differ from other groups of products. The section
elaborates the importance of studying RNPs, by considering their unique
characteristics within consumer learning and behaviour context. The Product
Presentation section follows on the unique nature of RNPs and how they can be
promoted in the most effective way to facilitate consumers’ comprehension and
behaviour towards these products. Various product presentation methods are
explained, such as the use of virtual reality elements and anthropomorphic attributes,
which leads to the next section, Virtual Reality. This section discusses Virtual Reality
technologies and how they can facilitate RNP promotion. Section 1.4.4 explains the
inclusion of Anthropomorphism as another element that can improve RNP promotion
20
and therefore assist consumers’ learning and online behaviour. Section 1.4.5
emphasises on Consumer Grouping and how different groups of consumers can adopt
and comprehend RNPs differently. The Diffusion of Innovation Theory and consumer
innovativeness are the particular foci of this section. Finally, section 1.4.6 discusses
Consumer Learning, exploring learning theories within the different paradigms of
cognitivism, behaviourism, and constructivism. It focuses on the Theory of
Multimedia Learning and Social Learning Theory as a baseline in supporting the
product promotion attributes proposed within the study.
1.4.1 Product Categorization – Understanding Really New Products
Typically consumers categorize products into groups by comparing product
attributes and outlining categorization criteria. However, scholars have introduced
various product categorizations in different fields of study. For instance within
conventional marketing research, the level of information asymmetry perceived by
consumers is used to differentiate products into three types: search goods, experience
goods and credence goods (Nelson, 1970; Nelson, 1974). Search goods refer to
products where full information about their attributes can be obtained prior to product
purchase; whereas experience goods refer to products where their attributes cannot be
known until the actual purchase and use of the product. Credence goods, which are
added to this categorization by Darby and Karni (1973), however refer to attributes,
which are not verifiable by consumers even after using the product. Additionally,
products have been categorized into three groups according to their characteristics of
durability, tangibility and use goods (Kotler, 2003). Due to the advancement of
technology, new innovative products are rising in different industries, especially
within high-technology industries (Moore, 1991; Feiereisen, 2009; Mohr et al., 2010).
New product development is a priority for most companies as without innovative
products, companies cannot compete in the marketplace (Hoeffler, 2003). Academics
have therefore classified new products according to their newness along a continuum,
from Incrementally New Products (INPs) to Really New Products (RNPs) ((Zhao et
al., 2009; Kuester et al., 2015). This three product grouping – RNPs, INPs and
Existing Products – is suitable when analysing products due to their newness. In this
research, the newness grouping is of interest as the study is concerned with RNPs, as
one of the products identified within this categorization.
21
Existing products are already available within the marketplace. This group
contains the lowest level of innovation such as convenience goods (e.g. grocery). In
contrast to existing products, incrementally new products (INPs) contain some level
of newness, as they are built upon established products (Garcia and Calantone, 2002).
For example, new versions of mobile phones or smartphones, such as the iPhone 5s,
at the time of their introduction were considered as INPs (Kuester et al., 2015). INPs
are products that allow consumers to do some new things, but the new functionality is
not something you can achieve using other similar products. Consumers do not need
to change their behaviour in order to enjoy the promised benefits of INPs (Alexander
et al., 2008).
The final group is Really New Products (RNPs), which contain the highest level
of newness. RNPs can be explained from different perspectives. There is academic
literature investigating RNPs from both firm and consumer perspectives. Gregan-
Paxton et al.’s (2002) definition of RNPs from a consumer point of view states they
are products which are firstly either very new and where consumers do not have any
information about them, or secondly, are existing products which have been expanded
upon significantly. RNPs allow consumers to experience something novel, that they
have not experienced before, therefore consumers need to gather information in order
to understand the functionality of RNPs. Some successful RNP examples are digital
cameras, personal digital assistants (PDAs) and MP3 Players (Feiereisen et al., 2008);
however these products have now all been superseded by the introduction of
smartphones and gadgets such as Apple’s iPad (Kuester et al., 2015). Hoeffler (2003)
described RNPs as products with greater benefits than INPs, although consumers
would need to change their behaviour in order to achieve the potential benefits of
them (Hoeffler, 2003). Based on Hoeffler’s findings, Alexander (2008) identified four
main differences between RNPs and INPs; (i) RNPs allow consumers to perform tasks
that they cannot currently do using existing products; (ii) consumers are more
uncertain of the consumption benefits of RNPs compared to INPs; (iii) consumers are
also more uncertain about the cost-benefit trade-offs in utility functions for RNPs than
INPs; and (iv) as stated earlier, consumers would need to change their behaviour in
order to achieve the potential benefits of RNPs, rather attain the benefits of INPs (e.g.
(Gourville, 2006).
22
The process of RNP and radical innovation adoption is an inevitable part of an
individual’s consumption activities. Consumers face making decisions to purchase an
innovation more frequently, due to the rise of RNPs in industries. Knowledge
adoption is therefore an important part of innovation decision making. The knowledge
concerning RNPs is often derived from what is known about Incrementally New
Products (INPs) and Existing Products (eg. (Hoeffler, 2003; Feiereisen et al., 2008);
however, this is done without taking into account the unique characteristics of RNPs.
It is assumed that consumers relate the innovation information presented to a similar
product, and deduce the concept conclusively; though this is not always the case. The
more radical the innovation, the harder it is to relate the information to one’s existing
knowledge base. It is also heavily influenced by the way RNP-related information is
presented. This lack of understanding is not only evident in business strategies in
promoting RNPs, but also in academic literature. Scholars need to adapt various
marketing techniques and frameworks to match the different characteristics of RNPs,
in order to firstly help consumers’ perception and understanding of them (e.g.
(Hoeffler, 2003; Feiereisen et al., 2008), and secondly to assist firms presenting
radical innovations effectively. The importance of understanding business and
scholarly aspects of RNPs and radical innovations are due to the increase in number
of these products within various industries.
RNPs are becoming an important factor in a firm’s success and sometimes even
the business’ existence. As technology is advancing and consumer demand is
changing, innovation has become a key factor for company growth. There is a
tangible and intangible aspect to innovation development in firms. Intangibly,
innovation is good for a company’s morale (Iacobucci and Hoeffler 2015). Not only is
a business seen as an up-to-date technologically strong firm, but furthermore,
employees enjoy being involved in innovation development and idea creation for new
products and services (Iacobucci and Hoeffler 2015). From the tangible perspective,
new product development is a priority for firms in order to maintain and potentially
increase their market share (Hoeffler 2003). Innovation pays off during the growth
and maturity from new products’ life cycles (Iacobucci and Hoeffler 2015). Radical
innovations proved to not only influence businesses in a positive way, but also to
revolutionize an industry; for instance, the introduction of MP3 players changed the
dynamic of the music industry massively. Before their introduction, Sony and Phillips
23
were the major players within the industry, however after the launch of MP3 players,
and Apple’s radical innovation of Ipod, Apple became the leader in the music
industry, overtaking Sony and Philips.
There are debates over the success of innovation, especially radical innovation
and RNPs. The proportion of RNPs is rising in various industries (Moore, 1991), but
40% to 90% of RNPs fail in the marketplace (such as Apple Newton, Sony Betamax),
with highly innovative products failing at an even greater rate (Cierpicki, 2000;
Feiereisen et al., 2008; Frattini et al., 2012). Radical innovations mean that firms can
be faced with financial insecurity, given that huge amounts of money and resources
are invested in the Research and Development phase of RNPs, need to be recovered.
There is no benchmark in order to evaluate radical innovation performance, which
results in more uncertainty. Scholars have tackled this issue and concluded that the
solution to increase the success rate of RNPs lies in understanding the nature of these
products and understanding consumers’ behaviour and difficulties towards radical
innovation adoption (Gregan-Paxton et al. 2002, Hoeffler 2003, Iacobucci and
Hoeffler 2015). As a result, it is important for firms to consider the most effective
marketing strategies to promote these products to facilitate consumer RNP adoption.
Gregan-Paxton et al.’s (2002) study indicated that designing such a marketing
strategy is a key challenge for marketers. Companies need to perform efficient market
research in order to have an effective marketing strategy; however existing market
research tools were designed for INPs, which are based on a consumer’s existing
knowledge base. Over the past two decades, academics have started to develop ways
to modify methods and marketing strategies to meet the higher-risk, higher-reward
domain of RNPs (Lehmann, 1994; Urban et al., 1996; Moreau, 1997; Hoeffler, 2003).
For example, Hoeffler proposed a framework for new product evaluation and
described the differences between RNPs and INPs; he also explained the role of visual
processing in RNP evaluation (Hoeffler, 2003). Feiereisen et al. (2008) studied how
different types of learning strategies can either enhance or damage the effect of
marketing communications upon product understanding, whilst using different
presentation formats. This thesis aims to add to the RNP literature by examining RNP
presentation techniques, and how consumer learning can be facilitated through these.
This will benefit businesses massively in order to (i) promote innovation effectively,
24
(ii) attain and retain customers, and (iii) indirectly understand consumer demand and
response by developing new products. In the following sections, product presentation
formats and various consumer learning strategies are analysed, with existing
frameworks being introduced.
1.4.2 RNP and Product Presentation
The unique characteristics of RNPs (Alexander et al., 2008) mean that they need
to be presented in a way to facilitate consumer understanding. Consumers learn about
RNPs at the time of product evaluation (Hoeffler, 2003), as consumers do not have
any core knowledge about these products. Conversely, consumers do have a base
knowledge or experience in the related domain of INPs and existing products, due to
their nature, at the time of product evaluation (Hoeffler, 2003). The information
consumers receive at the time of evaluation is crucial in influencing their behaviour
towards RNPs. More learning is required therefore, to understand RNPs in
comparison to other product types (Hoeffler, 2003). Consequently, facilitating the
process of RNP comprehension is likely to lead to greater RNP adoption. Product
information should be promoted in the most effective way to enable consumer
learning. This is particularly important for RNPs, as consumer understanding begins
at the time of product evaluation.
The Internet is used as a promotion channel in creating new product awareness
(Bickart and Schindler, 2001) and increasing new product adoption rate (Prince and
Simon, 2009). Internet advertising and product promotion has received a considerable
amount of attention from academics and managers. More than a quarter of Internet
users in the world, spend above 30 hours per month on average on the Internet
( European Advertising and Media Forecast 2007 ). It is not surprising therefore, that
companies are employing online advertising strategies and presentation techniques, in
order to present product/service related information online and compete in CMEs.
Studies comparing the effectiveness of online advertising strategies and promotional
techniques against traditional strategies (such as print or traditional media), and those
that examine new technologies in web advertising and information presentation have
been undertaken (e.g. Robinson et al., 2007; Lohtia et al., 2013). The findings
indicate that traditional media’s main pitfall was its inability to interact with the
observer (e.g. newspaper, paper magazines) (e.g. Coyle and Thorson, 2001; Nagar,
25
2009). However, with the introduction of the Internet (e.g. interactive TV and online
magazines) interactivity has become possible.
Technological advancement alongside maturing e-consumer behaviour has
changed the purpose of surfing the Internet. Product related information needs to be
presented in a way that not only attracts consumers (Reiman, 2001; Cunningham et
al., 2007), but also meets their high expectations (Palmer and Griffith, 1998;
Chevalier and Ivory, 2003). As a result, online advertising strategies need to be
revised in light of new technologies and demands (Hemp, 2006). Companies are
constantly looking for advanced advertising and promotional technologies. Liu et al.
(1997) indicated that there is a strong relationship between presenting information
effectively in a CME and revenue; given this, companies need to design an attractive
interface in order to convert online surfers into final purchasers. One asset that can
help distinguish a company’s website from its competition is Virtual Reality. This is
emerging as being significant in developing and maintaining competitive advantage
(Lee and Chung, 2008; Bruno and Muzzupappa, 2010).
1.4.3. Virtual Reality
VR technology is one which generates a virtual environment that appears as a
more realistic environment to users, and where they can interact with this
environment easily through watching, hearing and touching (Teresa, 2008; Qi, 1999).
In the communication literature, VR has been defined as “a real or simulated
environment in which a perceiver experiences telepresence” (Steuer, 1992, p.7).
Telepresence is described as a sense of ‘being there’ in a CME, which creates the
feeling of the direct product experience, where direct experience is the optimal
method for a consumer to learn about a product (Klein, 2003). There are two media
characteristics, which enable telepresence in a CME: user control (interactivity) and
vividness (media richness). Steuer (1992) also identified the attributes of VR, noting,
“Perceptual experience in the virtual reality can be categorized along interactivity
and immersion” (Bhatt, 2004, p.3). Interactivity is defined by its speed, range, and
significance, whereas immersion is deliberated through the breadth of immersion (i.e.
the sum of sensory organs that are influenced by the VR) and depth of immersion (i.e.
the degree of resolution). As Ragusa and Bochenek (2001) state, immersion refers to
26
“the feeling of being deeply engaged in a virtual world as if it were the real one”
(Ragusa and Bochenek, 2001 cited in Bhatt, 2004, p.5).
Due to the above-mentioned attributes and benefits, VR is an important
technology for enhancing user interfaces to provide more realistic and engaging
consumer experiences (Wann and Mon-Williams, 1996; Lee and Chung, 2008).
Gonzalez and Kasper (1997, cited in Lee and Chung, 2008, p.91) stated:
“[u]ser interface usually indicates a computer program or environment that
displays options on the screen as icons, i.e., picture symbols, by which users enter
commands by selecting an icon.”
As computers are considered as objects generally associated with a knowledge
function, it follows that user interfaces operate as a knowledge transfer agent.
Schlosser (2003) found that the mere presence of a computer increased the need for
information search. When a product is presented on a computer screen, compared to
one presented on paper, people feel the need to seek information in order to evaluate
that product. The more understandable and clearly presented the information, the
easier it becomes for consumers to find and comprehend the information.
Understanding the effective formats of product information presentation on the
Internet is necessary for business success. Consumer interactions with the company’s
products via an effective user interface can improve consumers’ shopping experience,
and more importantly, facilitate consumer learning which might increase the
probability of final product purchase. By inserting virtual reality languages into
CMEs, consumers are able to interact with the user interface in a realistic and
engaging manner. As a result, they will search and understand information better, and
have a more enjoyable shopping experience than an environment without VR
applications (e.g. Holzwarth, Janiszewski, and Neumann 2006). VR as the generator
of telepresence is enabled by the two characteristics of vividness (media richness) and
interactivity (user control). Sections 1.4.3.1 and 1.4.3.2 explain these two attributes.
1.4.3.1 Vividness
Vividness (media richness) is:
27
“the representational richness of a mediated environment as defined by its
formal features; that is, the way in which an environment presents information to
the senses” (Steuer, 1992, p.81).
Vividness represents two characteristics of the communication medium: sensory
depth and sensory breadth (Steuer, 1992). Sensory depth refers to the quality and
resolution of information transmitted to the senses (e.g. monitor resolution); whereas
sensory breadth is the sum of the sensory avenues that a medium utilizes (e.g. aural,
visual). Multimedia communication therefore has greater breadth than a single media
communication (e.g. magazine, radio). In the case of greater breadth, immersion
occurs (Steuer, 1992). In some studies, vividness was deployed via sensory breadth,
as manipulating sensory depth was not always feasible (e.g. Coyle and Thorson, 2001;
Sukoco and Wu, 2011).
As sensory breath is the sum of sensory channels that a medium utilizes, inserting
information in the form of various sensory representations has been supported within
several theoretical frameworks and has been operationalized by practitioners (e.g.
(Coyle and Thorson, 2001; Fortin and Dholakia, 2005). Imagery theories such as
Paivio’s (1971) Dual Coding Theory, explain the mental processing of words and
images. Learning theories, such as Mental Simulation (Taylor and Schneider, 1989) or
the Theory of Multimedia Learning (Mayer, 1997), support the insertion of various
sensory information to influence an individual’s learning. This is explained further in
section 1.4.6.
From the processing aspect within the pictorial models, imagery theories explain
that an image is a pictorial type of representation. Paivio (1965) was one of the first
psychologists to examine the mnemonic effects of imagery, and he exhibited one of
the first theories of imagery within the pictorial discipline, known as the Dual Coding
Theory of mental representation (Paivio, 1965; Paivio, 1971a; Paivio, 1986; Paivio,
2006). The Dual Coding Theory (Paivio, 1971a) states that an image has the property
of concreteness, and that it is better than words in "representing the way things look
or appear" (Paivio, 1971 cited in Morris and Hampson, 1983, p.120). Since words
and images are different in nature, they are supported by different processing systems.
Whereas images could be processed in parallel, words permit sequential or serial
processing only. Paivio (1971b) claimed that, "when both codes are used in certain
28
memory tasks, performance will be facilitated compared with when only one is
employed" (cited in Morris and Hampson, 1983, p.121). Pictorials have also been
employed in CMEs as a means of transferring information to users. The most
common product presentation format in e-commerce is a 2-dimensional (2D) image
based and text based presentation, which is used in most websites and advertising
materials (Pan et al., 2004). One of the main reasons for using a 2D format is that
design and implementation costs are low (Pan et al., 2004). With the advancement of
technology however, users’ expectations are changing, with consumers/users now
expecting various sensory representations on CMEs, such as audio and video.
Therefore in this thesis, sensory breadth is operationalized by inserting pictorial,
audio and video into the website.
1.4.3.2. Interactivity
Interactivity is the other media characteristic that enables telepresence in a CME.
Interactivity is defined as “the extent to which users can participate in modifying the
form and content of a mediated environment in real time” (Steuer, 1992, p.84). The
Theory of Multimedia Learning (Mayer and Moreno, 1998) elucidates interactivity as
a reciprocal activity between a learner and a multimedia learning system, where the
[re]action of the learner is reliant on the [re]action of the system and vice versa.
Moreno and Mayer (2007) explained five common types of interactivity:
Dialoguing
Controlling
Manipulating
Searching
Navigating
Dialoguing is when the learner either asks questions in order to receive an answer, or
gives an answer in order to receive feedback (Moreno and Mayer 2007). Some
examples of dialoguing are when a learner clicks on a highlighted text to get
additional information or asks a question to a virtual agent. Controlling is where the
learner is able to regulate the pace and/or order of the learning episode; for example,
when interacting with a video clip, the learner can control the pace by using the
pause/play keys, or fast forward key. To control the order, the learner can use the
forward and backward (rewind) keys, the slider bar or menu (select screen) for direct
29
access to a particular section. Manipulation is where the learner can control some
aspects of the presentation, such as zooming in/out, rotating or moving objects
around. Interactivity by searching refers to the learner’s ability to look up information,
for example, by entering a query, receiving options, or selecting an option.
Interactivity by navigating however refers to a learner being able to select the content
they are interested in by selecting from various available sources, such as by clicking
on a menu. Interactivity is operationalized using one or many of the types identified
above. In some cases, the focus is on a few of the interactivity types explained by
Moreno and Mayer (2007). One example is the guided activity principle, which
explains the way students are asked to engage in mixed initiative problem-solving
with a pedagogical agent (Lester et al. 1999). This principle focuses mainly on
manipulating and dialoguing, however it is explained that if the ability to manipulate
is allowed but there is no ability to use dialoguing, meaningful learning might not
occur (Plass et al. 2010, Mayer 2004). Therefore, based on the context of this study,
various types of interactivity might work while accessed together or separately. As
there is no recent study examining the effect of interactivity types identified by Mayer
and Moreno (2007), all types are employed in this study.
Within various product contexts, different types of interactivity have been studied
(e.g. Schlosser et al., 2003; Park et al., 2005). In operationalizing interactivity within
a CME in order to interact with RNP related information, Controlling and
Manipulating types of interactivity appear more suitable. Searching and Navigating
cannot be analysed as for most of RNPs, there is not much product related
information accessible through searching and navigating the web, due to these
products’ unique nature. Dialoguing is more appropriate within the anthropomorphic
attributes, which will be discussed in section 1.4.4
Content interactivity has been of interest to academics for decades. For example,
Schlosser (2003) and Fiore et al. (2005) included interactivity in websites in the form
of image/object interactivity. They explained that by adding image interactivity to
websites, approach response (e.g. returning to the site) and purchase intention are
positively influenced (Fiore et al., 2005), and visual sensory information is improved.
Park et al. (2005) examined the effect of product presentation (movement and image
size) on consumer shopping experience and purchase intention on an apparel website.
30
They concluded that when consumers experience pleasure from product movement, it
results in reduced perceived risk and increased purchase intention (Park et al., 2005).
Interactivity can also be engaged within CMEs by manipulating Dialoguing.
Dialoguing has been operationalized via anthropomorphic attributes such as
interactive avatars (e.g. Keeling et al., 2010). Anthropomorphism is explained in more
detail in section 1.4.4, to unpack its importance, characteristics and influence that it
has upon consumers’ comprehension and online behaviour.
1.4.4. Anthropomorphism
Anthropomorphism is an attribute, proven to have an influence upon an
individual’s behaviour towards CMEs. It is referenced within literature through
various terminologies of humanness, social cues and virtual entities (Nass and Steuer,
1993; Nass and Moon, 2000; Nowak and Rauh, 2006). Humanness was introduced by
early philosophers (Haslam et al., 2013), though existentialists think there is no
definition of humanness that works (Koval et al., 2009). Throughout history, there
have been various explanations of what makes a human being, human. Plato suggests
human is a “featherless biped”; Socrates’ approach was that human nature is the cause
of what human is becoming (Aristotle, 1078b). Over time, there has been some
success on defining humanness within various contexts (e.g. Nass and Moon, 2000;
Haslam, 2006); however research on humanness has either been the result of theories
(e.g. Wrightsman, 1992), or limited to certain kinds of characteristics such as traits
and emotions (e.g., Leyens et al., 2000; Haslam et al., 2005; Bain et al., 2009).
According to Haslam (2006), there are two distinct dimensions of humanness, human
uniqueness and human nature; the former refers to the attributes that distinguish
humans from other animals, whereas the latter refers to the attributes that are seen as
fundamentally, essentially or typically human. Human uniqueness attributes are about
refinement, civility and rationality. They appear rather late in human development as
they are related to socialization and learning, and they signify how humans exceed
animality through reason and culture (Haslam, 2006). Human nature attributes
however distinguish humans from machines (Koval et al., 2009). The attributes of
warmth, emotion, openness and desire, arise in the early stages of human
development, because they are widespread within the population and are cross-
31
culturally universal; they establish our “shared community” (Koval et al., 2009).
Further studies support the findings above; for instance, Haslam et al. (2013), Haslam
et al. (2008), and Loughnan and Haslam (2007) argue that people lacking human
uniqueness attributes are implicitly associated with animals, and those lacking human
nature attributes are implicitly associated with robots. The humanness attributes of
human uniqueness and human nature, can be identified and analyzed for human
beings, but are not as apparent within a CME. Any presence of humanness within a
CME has limitations; it is impossible to study or even implement all of the humanness
attributes in an online setting.
Within a CME, anthropomorphism is defined as the extent to which a virtual
object behaves or looks like a human (Koda, 1996; Nowak and Biocca, 2003; Nowak,
2004). Anthropomorphic attributes can include concepts such as Artificial Intelligence
(McCarthy, 1955) and anthropomorphized interface representations like faces
(Sproull et al., 1996; Gong and Nass, 2007) and voices (Gong and Lai, 2003; Nass
and Brave, 2005). Anthropomorphic attributes can be represented via a simple tool
such as a human image (static avatar), to a talking, interactive virtual agent (human-
like avatar). In studies about avatars, it is the human morphology, or appearance,
which has been examined, not its behaviour (e.g. Nowak and Rauh, 2005; Holzwarth
et al., 2006; Keeling et al., 2010). Morphology refers to physical dimensions (such as
attractiveness, height or size) (Meiselwitz, 2014), and it is also identified as a social
cue (such as gender, group or status symbol). For an avatar, morphology is also
named “Physical Morphology”, which indicates any appearance element within the
avatar’s body (Meiselwitz, 2014). Within morphology, there are two main perceptions
about avatars; avatar humanness (the extent to which the avatar looks human) and
androgyny (a rating of the avatar's (lack of) masculinity or femininity) (Nowak and
Rauh, 2005; Nowak et al., 2008). Instead of androgyny, Nowak (2008) explained it as
where the image falls on the masculinity-femininity continuum.
According to Social Cognition Theory, the classification of objects as human,
animal or object, and the recognition of anthropomorphic characteristics, are basic
cognitive functions (Kunda, 1999). Anthropomorphic attributes have been employed
in computers as a way to improve human-computer interaction and they refer to the
“technological efforts of imbuing computers with human characteristics and
32
capabilities” (Gong, 2008), p.1495). Studies in various fields have considered
anthropomorphic attributes, including retailing (e.g. Holzwarth et al., 2006), tourism
(e.g. Guttentag, 2010) and education (Franceschi et al., 2009). For example, Guttentag
(2010) demonstrated how avatars have been used in the tourism industry with Rome
Reborn (Rome Reborn Brochure, 2008). He mentioned the difference between using
avatars usually as a sales agent in retailing, and avatars as a virtual agent who act as
actors or guides (like in traditional museums), that have an educational effect as well
as entertainment value. Within education, using an avatar-based environment provides
“a shared visual space for students to meet and interact via avatars, providing a strong
sense of group presence, leading to engaging groups learning interactions”
(Franceschi et al., 2009), p.74).
From a consumer behaviour perspective, studies have examined consumer
perceptions of avatars within CMEs. Nass et al. (1997) for instance realised
computers with voice output provoked gender stereotypes. Keeling (2010) looked at
different communication styles used by avatars and Holzwarth (2006) studied how
perceived attractiveness and expertise of the avatar influence consumer online
behaviour. Employing anthropomorphic attributes is supported within various
theoretical frameworks. For instance, according to Social Learning Theory, having an
appropriate degree of social presence promotes user learning (Bandura, 1977);
therefore employing anthropomorphic attributes, which results in an increase in social
cues and consequently social presence, can have a positive effect upon a consumer’s
learning and behaviour. Anthropomorphic attributes may be introduced by adding
static or interactive avatars to the CME. Dialoguing, as an interactivity attribute,
initiated by Theory of Multimedia Learning (Mayer and Moreno, 1998) can be
employed via interactive avatars.
Individuals perceive anthropomorphism while interacting with various forms of
social cues presented within CMEs. However, some forms of social cues might be
preferred by individuals. For instance, when there is an interactive avatar, individuals
might focus on social cues presented in the avatar, and as a result, perceive
anthropomorphism. They might ignore other forms of social cues presented in the
website, which the avatar is part of; whereas in a website with no avatar, individuals
perceive social cues presented in the website as humanness elements. There is no up-
33
to-date research investigating this inconsistency and the present study attempts to
shed light into this area.
Kim and Sundar (2012) introduced another aspect of anthropomorphism,
explaining how individuals perceive it either consciously (mindfully) or
unconsciously (mindlessly). Mindful anthropomorphism refers to when participants
consciously rate a CME as anthropomorphic. Nass and Moon (2000, p.93) justify the
mindful nature of anthropomorphism by referring to the definition of
anthropomorphism as “a thoughtful, sincere belief that the object has human
characteristics”. Much of the research investigating mindful anthropomorphism is
within the Human Robot Interaction (HRI) literature (e.g. Powers and Kiesler, 2006;
Bartneck et al., 2009). However these categorizations have also been employed in the
CME literature while studying the employment of social cues such as avatars (e.g.
Nowak and Rauh, 2005).
Mindless anthropomorphism is explained via a heuristic systematic model
(Chaiken, 1987) and the MAIN model (Sundar, 2008). These models describe how
heuristic cues trigger a consumer’s unconscious responses towards anthropomorphic
attributes. Kim (2012) explains how richer modality cues elicit a “realism heuristic”
(Kim and Sundar, 2012, p242), which imitates face-to-face communication. Richer
modality cues further provoke agency affordance, which refers to any evidence of the
existence of an intelligent entity within the context of interaction; this leads to a
“social presence heuristic” (Kim and Sundar, 2012, p.242). Studies have manipulated
anthropomorphism by including an avatar in the website and directly asking
consumers if the character is humanlike, machine-like, natural and so on (e.g. Groom
et al., 2009; Keeling et al., 2010). Kim and Sundar (2012) explained that this
approach denies the mindless anthropomorphism perceived by individuals within
CMEs. They further developed perceived mindful and mindless scales to measure an
individual’s reaction towards various social cues, which are employed in this study.
Consumers themselves are also important when considering the comprehension of
RNPs, as their characteristics might be influential upon their perception of vividness,
interactivity and anthropomorphic attributes within CMEs towards RNPs. The next
section analyses consumer grouping and characteristics, specifically looking into their
34
role as innovator adopter and diffusor of technology.
1.4.5 Consumer Grouping – innovator adopters
Amichai-Hamburger (2002) indicated that an online consumer’s personality plays
an important role in their online behaviour. Consumer characteristics influence online
shopping acceptance and technology adoption (Amichai-Hamburger et al., 2002). As
RNPs cannot rely on a consumer’s prior knowledge, it is important to understand
consumers as technology adopters and what factors influence their innovation
acceptance. Given this, diffusion of innovation theories are firstly discussed to
identify how individuals learn about innovation and technology, and then Roger’s
Diffusion of Innovation Model (1962) is introduced and analysed. The consumer is
then studied as an innovation adopter, and the consumer innovativeness concept is
investigated further.
1.4.5.1 Diffusion of Innovation Theories - Background
Theories of innovation have been established across many fields of research over
many decades (Rogers, 1983). The starting point for innovation theories in business is
mainly from the work of economist Schumpeter (1939). He viewed innovation as
distinctly different from invention, and further proposed innovation as being
characterized by “1) construction of new plants and equipment, 2) introduction of
new firms and 3) the rise to leadership of new men” (Schumpeter, 1939 cited in
Robertson, 1967, p.14). Anthropologist Barnett (1953, p.7) explained innovation as
“the basis of cultural change … any thought, behaviour or thing that is new because
it is qualitatively different for existing forms”. The work of sociologist Rogers (1962)
broadened the definition by referring to innovation as “an idea perceived as new by
the individuals” (Rogers, 1962, p.13).
Defining innovation leads onto the concept of diffusion of innovation. Diffusion
of innovation is about how, why and at what rate an innovation spreads through
cultures and is communicated through certain channels (Rogers, 2003). Robertson
(1967) claimed that a theoretical model of the diffusion process was not possible, due
to various variables involved within the process. The diffusion theory originated
within sociology and anthropology, with sociologist Rogers being one of the first
35
researchers to propose a diffusion curve. He introduced the process of diffusion
within the agriculture industry, as farmers adopted new innovations. He identified
groups of people on a continuum from those adopting innovation first to those who
adopt it last. Robertson (1967) analysed the validity of Rogers’ (1962) model and the
Diffusion of Innovation Theory. He compared Rogers’ rural-diffusion model with an
industrial-diffusion process done by Mansfield (1961), with Robertson’s study
confirming Rogers’ model to a considerable extent. Robertson also applied Rogers’
Diffusion of Innovation model amongst physicians in four cities (Robertson 1967).
The findings validated Rogers’ Diffusion of Innovation model.
Although Rogers’ work established that the Diffusion of Innovation model is a
valid and reliable model to be used within various fields and studies, there are
limitations. His research only considered the diffusion process of a small percentage
of innovations; that is, where the innovations were better products than any existing
products at the time, and where it was only a matter of time before everyone adopted
these innovations. This is not the real world scenario for most marketing innovations
today. Not all innovations are superior to existing products, or of interest to the
consumer, and therefore will not be adopted by everyone (Robertson 1967).
Robertson (1967, p.17) concluded that there is “an ever-incomplete curve of
adoption” for innovation in marketing.
Lowrey (1991) continued analysing Rogers’ Diffusion of Innovation model,
identifying two specific problems for marketing researchers (Lowrey, 1991). Firstly,
“the general lack of necessity implicit in the majority of consumer offerings” meant
that categorizing non-adopters (or slow adopters), led “to an incorrect placement of
blame” (Lowrey, 1991, p.644). Lowrey discussed how non-adopters do not adopt the
innovation simply because it does not fit their needs. This point was raised by Rogers
himself:
“we have often assumed that all adopters perceive an innovation in a positive
light, as we ourselves may perceive it. Now we need to question this assumption
of the innovation's advantage for adopters” (1983, p.100).
Secondly, Lowrey (1991) questioned the validity of Rogers’ model, by explaining that
individuals are not consistently innovative; the same individual can be put into one
36
innovation category or another within Rogers’ diffusion model, depending on the
product s/he is dealing with. Robertson (1971) also questioned whether or not a
general innovator may even exist.
Other studies in the innovation adoption field have introduced consumer
categories; for instance, Feick and Price (1987) introduced market mavens as
knowledgeable individuals in the field of common household products. However,
when dealing with more technologically-driven products, finding individuals with a
high level of knowledge and information became more difficult (Lowrey, 1991). This
fits well with Robertson’s (1971) Theory of Innovation within a product category.
Robertson emphasized the definition of product category and introduced a simple
innovation continuum (explained further in 1.4.5.2.1).
To summarise, the two main criticisms of Rogers’ Diffusion of Innovation model
were consumer innovators and product categorization issues. These two elements
were addressed by Rogers in the later stages of developing the Diffusion of
Innovation Theory. Rogers’ model has also proven to be a valid and reliable model
within various industries and has been used in academic research to analyse RNPs
(e.g. Feiereisen, 2009). This indicates the suitability of Rogers’ model within the
context of RNPs, hence this research considers Rogers’ Diffusion of Innovation
model as a baseline for investigating how individuals adopt RNPs as innovation. The
study contributes to innovation theories by investigating RNPs as the chosen product
category, and addressing consumer differences (explained further in 1.4.5.3).
1.4.5.2. Rogers’ Diffusion of Innovation Model
Rogers (1995, p.5) explained diffusion as a process “by which an innovation is
communicated through certain channels over time among the members of a social
system”. Diffusion is a special type of communication where the message is about a
new idea. The newness injects some degree of uncertainty into the diffusion process.
Diffusion is also a kind of social change, described as “the process by which
alteration occurs in the structure and function of a social system” (Rogers, 1995,
p.6). Marketing and consumer behaviour theorists have adopted the general paradigm
introduced within sociology, in order to explain new product acceptance and the
diffusion of innovation within these fields. Although there are differences between,
for example, sociological innovation and new consumer products, the general theory
37
has been accepted within the marketing field (Gatignon and Robertson, 1985). Only a
few studies have criticized the basic conceptual frameworks. Rogers (1983) was one
of the pioneers who was critical of diffusion research, especially in relation to the pro-
innovation bias of most diffusion researchers.
Rogers (1983, p.92) explained pro-innovation bias, as how research suggested
that:
“innovation should be diffused and adopted by all members of the social system,
that it should be diffused more rapidly, and that the innovation should be neither
re-invented nor rejected”.
This bias indicates that two areas within diffusion research need more attention;
firstly, the individual categorization into adopter categories based on innovativeness,
and secondly, the characteristics of innovation. These were in line with the main
criticisms elaborated by Robertson (1967) and Lowrey (1991). Rogers further
advanced diffusion theory by analysing these two factors alongside other factors
within the diffusion of innovation process. Rogers (1995) named four main elements
in the diffusion of innovation: innovation, communication channels, time and social
system; these are addressed and critically appraised in the following sub-sections.
1.4.5.2.1 Innovation
Robertson (1967) initially classified innovation into three groups, based on their
effects upon established consumption patterns: Continuous Innovation, Dynamically
Continuous Innovation and Discontinuous Innovation. A continuous innovation is a
modified existing product, which results in the minor disruption of consumer
behaviour; this is a very similar definition to INPs. A dynamically continuous
innovation can either be a modified existing product, or a new product. This group of
products cause some, but not a considerable amount of disruption in behaviour
patterns. A discontinuous innovation is a new product, which causes major disruption
in established behaviour, resulting into a new behaviour pattern (Robertson, 1971);
again, this fits well with the definition of RNPs.
Rogers (1995) defined innovation as an idea, object or practice that is perceived
as new by an individual. ‘Newness’ of an innovation may be conveyed in terms of
knowledge, persuasion or a decision to adopt. Technological innovation creates an
38
uncertainty in the mind of adopters, but simultaneously represents an opportunity to
resolve a problem that existing technologies are unable to solve, thus reducing this
uncertainty. RNPs are an example of such innovation as they are very new in nature,
with consumers having no information about them (Gregan-Paxton et al., 2002);
however, they do promise greater benefits in comparison to existing or incrementally
new products. The advantage of solving an adopter’s potential problem is the reason
adopters exert effort to learn about innovation (Rogers, 1995). Consequently adopters
search for information, which reduces the uncertainty to a tolerable level, and
ultimately results in the adoption/rejection decision. Thus, the innovation decision-
making process is basically an information seeking and processing activity, which
motivates adopters to reduce the uncertainty about the advantages/disadvantages of
the innovation (Rogers, 1995).
Scholars within the field of innovation recognized that one’s decision about an
innovation is not an instantaneous act, but a rational process that occurs over time
(Rogers 1983). This process of innovation decision-making has five main steps of
knowledge, persuasion, decision, implementation and confirmation. Knowledge refers
to the time an individual became aware of the innovation and learns about how the
innovation functions. Persuasion happens when an individual forms a favourable or
unfavourable attitude towards the innovation. Decision refers to the time an individual
participates in activities that lead to a choice of either innovation adoption or
rejection. Implementation is when an individual uses the innovation and finally
confirmation happens when an individual obtains reinforcement of an innovation
decision they have made. If the decision-maker faces conflicting messages about the
innovation, they may reverse their previous decision (Rogers 1983).
Steps identified within the innovation decision-making process are different to the
steps identified in a classical decision-making process, even though both are rational
decision-making processes. In a traditional model, the steps are need recognition,
information search, evaluation of alternatives, action and purchase decision, and post-
purchase evaluation (Wright 2006). Within the innovation context, the need
recognition step might not occur as some innovations such as RNPs, although they are
designed to solve consumers’ problems, the problem might not be recognised by the
individual before the solution in form of innovation, is introduced to them.
39
Information search in the classical decision-making model is similar to the knowledge
step in the innovation decision-making process. For some types of innovation such as
RNPs, individuals might not be able to access a lot of information, the information
might be very limited and in some cases, technical. The information that adopters
seek in this process can be divided into process information or “software
information”, and result information or “innovation-evaluation information” (Rogers,
1983). Rogers (1983, p.6) explained software information as information “…
embodied in a technology and serves to reduce uncertainty about the cause-effect
relationships involved in achieving a desired outcome”. Innovation-evaluation
information is defined as the information “which is the reduction in uncertainty about
an innovation's expected consequences” (Rogers, 1983, p14). Hence, the
knowledge/information search step is crucial due to individuals’ lack of knowledge.
Again in the innovation decision- making process, there might not be any similar
alternatives available, depending on the product’s level of newness. Evaluation of
alternatives in the classical decision-making process might therefore not be
applicable.
Persuasion is the process after knowledge within the innovation context.
Individuals are either convinced by the information or not, leading them to decide on
the innovation’s adoption or rejection (the decision step). In classical decision-
making, after deciding on the purchase and purchasing it, consumers go through a
post-purchase evaluation, whereas in the innovation context, there is also an
implementation stage after purchasing the product. Implementation is an important
stage as individuals might not have any opportunities to experience the product use
before the actual purchase. For example, for a product such as Google Glass,
consumers needed to pre-order the product and the only experience of product use
was through video clips or product demonstration. After using the innovation,
consumers reach the confirmation stage of decision-making process (the last stage),
which has similarities to post-purchase evaluation in classical decision-making. This
confirmation step depends on the compatibility between the product’s perceived and
actual functionality, as well as elements of post-sales support, ease of maintenance
etc. (Rogers 1983).
40
It is evident that there are many similarities between the classical and innovation
decision-making processes, but there are also some differences. The differences are
mainly based upon the different nature of innovative products; therefore the more
innovative the product, the wider the gap between the two decision-making processes.
Another aspect affecting adoption rate is innovation characteristics; this has been
an element of debate between academics (e.g. Robertson, 1967; Rogers, 1995). Most
research on the diffusion of innovation within the consumer domain is concerned with
identifying differences between adopter groups; little has been focussed upon the
characteristics of innovations (Gatignon and Robertson, 1985; Mahajan et al., 1990).
Only limited research has empirically considered the role of product innovation
characteristics within the marketing literature (Ostlund, 1972; Ostlund, 1974;
Dickerson and Gentry, 1983; LaBay and C., 1983; Golsar, 1987). In order to address
one of the main criticisms that have impacted the pro-innovation area, Rogers (1995)
outlined five product innovation characteristics: relative advantage, compatibility,
complexity, trialability and observability. An innovation, which is perceived by
individuals as having higher relative advantage, compatibility, trialability and
observability with less complexity, will therefore be adopted more rapidly than other
innovations. Rogers’ innovation characteristics have though been criticized by
academics (e.g. Lowrey, 1991, Robertson, 1967). Lowrey (1991, p.645) explained
how technological innovation had additional factors which were not “adequately
addressed by existing Diffusion theory”. Another criticism was that the impact of
consumers’ perception of innovation attributes was not considered in previous
literature. Dickerson and Gentry (1983) studied various demographics,
psychographics and technical product experience variables and compared them
between adopter and non-adopter categories. They concluded that the difference
between their profile and Rogers’ adopter categorization stem from the nature of
innovation. This indicates the necessity of studying one innovation, or innovative
products within the same domain, in order to minimize the effect of innovation
characteristics and consumers’ perception, on the process of diffusion of innovation.
As a result, the RNPs employed within all three studies in this research were selected
from the kitchen appliance domain, to control for the process of innovation adoption.
1.4.5.2.2 Communication
41
Communication is the process in which participants share information with one
another to reach a mutual understanding. In the diffusion of innovation process
however, the message is a new idea. The diffusion process involves 1) an innovation,
2) an individual with knowledge/experience about innovation, 3) an individual
without knowledge of innovation, and 4) a communication channel (Rogers, 1995). A
communication channel is a means by which an individual sends/receives messages.
Mass media channels are the most rapid means of communication and are mainly
used to create awareness/knowledge. Conversely, interpersonal channels appear to be
more effective in persuading individuals to adopt a new idea. Given this, diffusion is a
very social process (Rogers, 1995), and can happen within interpersonal channels for
RNP adoption.
1.4.5.2.3 Time
Time is another element in the diffusion process. There are three aspects of time
as variables. Firstly, it is the innovation-decision process that shows the amount of
time an individual passes from the point of receiving the first information to the
adoption/rejection decision. Secondly is the rate an innovation is adopted by an
individual (earliness/lateness), which refers to the adopter categories. Lastly, is the
rate an innovation is adopted by a system (a group of individuals); this rate can be
varied depending on the social systems or the type of innovation (Rogers, 1995).
1.4.5.2.4 Social System
A social system is defined as “a set of interrelated units that are engaged in joint
problem-solving to accomplish a common goal” (Rogers, 1995, p.23). This social
system affects the innovation diffusion process in many ways, such as the effect of
norms on diffusion, group conformity, the role of leadership opinion, types of
innovation-decisions and the consequences of innovation (Rogers, 1995). In the case
of studying innovative products that do not exist in the marketplace, the social system
and social communication cannot be examined.
Rogers (1995) further defined five factors as important elements for the diffusion
and adoption of innovations, namely: consumer characteristics, product attributes,
communication channels, the social system and time (Dowling, 1999; Rogers, 2003).
1.4.5.2.5 Consumer characteristics
42
Consumer characteristics have been criticized as being overlooked by Rogers’
(1995) Theory of Innovation Diffusion. Academics studied various socio-
demographic factors to understand the influence of consumer characteristics within
the diffusion of innovation and adopter non-adopter groups (e.g. Hirschman, 1980;
Dickerson and Gentry, 1983). Consumer characteristics were analysed further by
Rogers (2003) and other academics in order to address the initial debates around this
aspect. Generally, findings indicated that males are more innovative than females; that
a negative relationship between age and consumer innovativeness exists; and that a
positive relationship exists between consumer innovativeness and education/income
(Rogers, 2003; Weijters and Roehrich, 2004; Tellis et al., 2009). Furthermore, studies
revealed that there is compatibility between innovation with the dominant values,
goals, experiences and the needs of potential adopters (Midgley and Dowling, 1978;
Daghfous et al., 1999). Previous studies demonstrated a positive relationship between
consumer innovativeness and values (such as self-enhancement and openness to
change), and a negative relationship between empathy and conservatism (e.g.
Daghfous et al., 1999; Rogers, 2003; Weijters and Roehrich, 2004). One important
value is creativity (Hirschman, 1980); the more creative the consumer, the better
product concepts are understood, facilitating the actual adoption of new products
(Rogers, 2003). Other values, which can be satisfied by innovation adoption, are the
need for stimulation (e.g. Raju, 1980; Baumgartner and Steenkamp, 1996) and the
need for uniqueness (Lynn and Harris, 1997; Tian et al., 2001). The important
outcome is that the debate around the impact of consumer characteristics on consumer
innovation adoption is ongoing; therefore Roger’s model can be applied in innovation
adoption, whilst considering the characteristics alongside adopter types, to improve
the accuracy and validity of findings. In the present research, consumer characteristics
are investigated in more detail. Study 1 controlled for consumer characteristics
(Chapter 2) and in Study 3, consumer characteristics were examined in more depth
(Chapter 4). The findings try to fill the gap in the literature relating to the impact
consumer characteristics can have in consumers’ adoption and online behaviour,
within the innovation adoption process.
1.4.5.2.6 Product attributes
Rogers (2003) explains the most important factor within product attributes is
perceived product/service advantage, which supports previous arguments that the
43
higher the perceived advantageousness, the shorter the time of adoption (Hoyer and
Ridgway, 1984). Other product attributes identified by academics are perceived
usefulness (Irani, 2000) and the perception of innovation by potential adopters
(Gatignon and Robertson, 1985). Furthermore studies have revealed that hedonic and
functional innovations attract different types of consumers (Hirschman, 1984). Later,
Grewal et al. (2000) discovered that some consumers buy innovations in order to
distinguish themselves. Product category is also an important factor. Goldenberg et
al. (2001) realized categories such as durable vs. non-durable, high-tech vs. low-tech
and low vs. high level of newness are perceived differently by consumers; as such,
these perceptions may affect their adoption intention. The arguments within product
attributes are closely linked to innovation characteristics. It is therefore important to
consider the characteristics of new products under study, by considering domain
specific products, to minimize the various impacts it might have upon a consumer’s
diffusion of innovation. By selecting RNPs within the same domain of kitchen
appliances throughout this research, the impact of product characteristics on diffusion
process is reduced.
1.4.5.2.7 Communication Channels
Selecting the right communication channel is essential in the diffusion of
innovation. For example Gatignon and Robertson (1985) and Rogers (2003) revealed
that innovative consumers have a higher interest towards mass media. Others (e.g.
Blythe, 1999; Summers, 1972) found that innovative consumers read more print
media and are attracted to magazines, rather than television and radio. Due to
technological advancements and changes in consumer expectations however, online
platforms have become an important search as well as shopping tool. The Internet is
used as a promotion channel in creating new product awareness (Bickart and
Schindler, 2001) and appears to be a suitable platform when promoting RNPs.
1.4.5.2.8 Social networks
A social network, within the diffusion of innovation concept, refers to a network
of interpersonal influences on innovative consumers (Midgley and Dowling, 1978).
Studies demonstrated a positive correlation between consumer innovativeness, and
social participation and opinion leadership (Gatignon and Robertson, 1985b;
Goldsmith and Foxall, 2003; Im et al., 2003). Innovative consumers are therefore
44
more likely to be amongst the first buyers of a new product (Blake et al., 2003; Im et
al., 2003) and are keener on word-of-mouth communication rather than mass
marketing (Midgley and Dowling, 1978; Midgley and Dowling, 1993). Paper 3
(Chapter 4) looks deeper into various innovative consumers to understand the
differences.
Considering all the factors and elements of innovation diffusion in this thesis,
consumer characteristics as innovation adopters will be examined further in Section
1.4.5.3.
1.4.5.3. Consumer innovativeness
Consumer innovativeness has been conceptualized into two main streams “Innate
Innovativeness” and “Actualized Innovativeness” (Midgley, 1977). Midgley (1977
p.75) explains innate innovativeness as “the degree to which an individual makes
innovation decisions independently from the communicated experience of others”;
whereas actualized innovativeness is the actual adoption of a new product. The RNPs
examined within this thesis are not available to be actually adopted by consumers and
they will be introduced to the marketplace within the next decade. This time of
adoption has an impact on the accuracy of actual product adoption; therefore the
actualized adoption cannot be examined correctly in the present research. Innate
innovativeness, on the other hand, is a cognitive trait based on an inherently
innovative personality, tendency and cognitive style (Hirschman, 1980; Midgley and
Dowling, 1993; Steenkamp et al., 1999). This thesis is concerned with the innate
perspective of innovativeness. The Diffusion of Innovation Theory sheds light on
understanding individuals as innovators, by examining their characteristics, which is
in line with innate innovativeness. This theory considers individuals’ differences in
adopting new products and categorizes individuals into five groups depending on their
adoption rate. The five groups are innovators, early adopters, early majority, late
majority and laggards (Rogers, 2003).
There are several prerequisites identified within research for consumers to be
innovators. Innovators should be financially capable of bearing the monetary burden
of the possible loss, if the innovation happens to be unprofitable for consumers.
Innovators should be able to cope with the high uncertainty involved in innovation
adoption. The most important characteristic of an innovator is that they are
45
venturesome, that is, they should have “a willingness to take risks and to accept that
sometime innovations fail” (Rogers, 2003, p.283). Although innovators might not be
the most respected members of the social system, they play the most important role in
the diffusion process by importing the innovation from outside the social system’s
boundaries. Rogers (1995) also mentioned that individuals with higher levels of
innovativeness are active information seekers with regard to new ideas. Innovators
view online shopping as quicker, cheaper, safer and more fun than traditional
shopping (Goldsmith and Lafferty, 2002). Therefore, if an innovation is promoted for
the first time, by targeting individuals with a higher level of innovativeness, there
might be a better chance for innovation adoption in comparison to other adopter
groups. In this study, innovative consumers are controlled for (Study 1 in Chapter 2)
and examined in more depth (Study 3 in Chapter 4), in order to understand the impact
various innovators can have in adopting and understanding RNPs. For this reason,
consumer innate innovativeness is further analysed, in order to identify individuals
with a high level of innovativeness.
Consumer innovativeness as an innate innovativeness is explained by
Vandecasteele and Geuens (2010) in the Motivated Consumer Innovativeness (MCI)
model. The model considers four aspects of consumer innovativeness, namely,
hedonic, functional, cognitive and social. Functional Motivated Consumer
Innovativeness (fMCI) explains how consumers are motivated by the functional
property of the innovation; Hedonically Motivated Consumer Innovativeness (hMCI),
looks at how consumers are motivated by affective or sensory stimulation; Socially
Motivated Consumer Innovativeness (sMCI) refers to when consumers are motivated
by “the self-assertive social need for differentiation” (Vandecasteele and Geuens,
2010, p.4); and finally Cognitive Motivated Consumer Innovativeness (cMCI)
explains how consumers are motivated by mental stimulation (Vandecasteele and
Geuens, 2010). MCI is claimed to provide a more accurate prediction of innovative
buying behaviour, as it contains a more complete range of motivations for
innovativeness. This model has proved to be more useful in the new product
development field hence this is the category of products considered in this study
(Vandecasteele 2010). Various consumer innovation adopters are also examined in
the context of RNPs to understand the influence different product presentation
46
formats have on each group of consumers’ learning and online behaviour. Section
1.4.6 will now discuss consumer learning within the RNP context.
47
1.4.6 Consumer Learning
Learning about new concepts can be explained by various learning theories,
which can be used to help understand the relationship between learning about new
products and online presentation formats.
4 1: Learning Theories Supporting Study’s Variables
Variables
Paradigm
New/Innovative Products Vividness Interactivity Anthropomorphism
Behaviourist*Stimulus-response Framework (Thorndike, 1931)
*Classical Conditioning (Pavlov, 1927)
*Social Learning Theory (Bandura, 1977)
*Social Exchange Theory*Social Learning Theory
Cognitivist
*Socio-Cognitive Theory of Product Market (Rosa et al., 1999)*Analogical Learning (Gentner, 1989)
*Mental Simulation (Taylor and Schneider, 1989).*Theory of Multimedia Learning (Mayer, 1997)
*Theory of Multimedia Learning (Mayer, 1997)
*Social Learning Theory (Bandura, 1977)*Interactive Multimodal Learning Environment (Moreno and Mayer, 2007)
Constructivist
*Contagion Theory (Burt, 1982)*Socio-Cognitive Theory of Product Market (Rosa et al., 1999)
*Social Development Theory (Vygotsky, 1978)
Learning theories fall into three philosophical frameworks: behaviourism,
cognitivism and constructivism. Behaviourism defines learning as an acquisition of
new behaviour through conditioning. The learning theories developed in this
paradigm were proposed by (Watson, 1924), (Thorndike, 1931) and Skinner (1974).
Some of the important theories in this approach are classical conditioning (Pavlov,
1927), operant (instrumental) conditioning (Skinner, 1953) and Social Learning
Theory (Bandura, 1977). Cognitivist theories were developed from Gestalt
psychology. There are two key assumptions in this approach: first, the memory
system is an active, organized information processing system and second, prior
knowledge plays an important role in learning. Consequently cognitive theorists
attempt to understand how the process of learning occurs (e.g. Chomsky, 1957;
Simon, 1957; Bruner, 1960), with Attribution Theory (Weiner, 1974), Cognitive Load
Theory (Sweller, 1988), Cognitive Theory of Multimedia Learning (Mayer, 1997) and
48
Elaboration Theory (Reigeluth and Stein, 1983) being some of the main theories in
this approach. The constructivist approach holds that knowledge is not passively
learned from the outside world, nor does it exists as a set of information within the
mind; rather, the learner actively constructs their own knowledge, based upon their
previous interaction with the world and their interpretation of this information into a
personalized knowledge base (Piaget, 1950). Social Network Theory (Barnes, 1954),
Contagion Theory (Burt, 1982), Social Development Theory (Vygotsky, 1978),
Problem Based Learning (Schmidt, 1983) and Situated Learning (Lave, 1988) are
some examples of significant theories within this approach.
When learning about RNPs, each of the philosophical approaches to learning may
be considered appropriate. Table 1 contains a list of various theories and frameworks
that are applied within academic research, looking into the main variables considered
in this study. These include new products/innovation, vividness, interactivity and
anthropomorphism. Various theories within the cognitivist, behaviourist and
constructivist frameworks were successful in supporting individual elements, but in
order to be applied within this research, theories need to be analysed in accordance
with all the main elements of the thesis. Within the behaviourist paradigm for
instance, the stimulus-response framework can explain imitation effects, when late
adopters accept an innovation as a result of observing the behaviour of early adopters
(Thorndike, 1931). It is indicated within research that RNP consumers are likely to be
innovators/early adopters and as such, are unable to observe other consumers using
the product. Benefits of RNPs are also hard to observe, so this model is of little
interest for the study of consumer response to RNPs.
The Socio-Cognitive theory of product market (Rosa et al., 1999), within the
constructivist domain, integrates social and cognitive learning processes to recognize
the importance of learning in the diffusion of new products. In this theory, product
markets are where consumers and product producers meet up and are constructed
through sense making and market stories. Market stories refer to word-of-mouth
between parties (e.g. producers, dealers or consumer communities) or journal articles.
Consumers and producers gather information about new products and this knowledge
is then integrated in the consumer’s and producer’s conceptual systems through
illiterate learning (Rosa et al., 1999). Contagion theory, also within the constructivist
49
paradigm, elucidates the probability of consumers learning about new products
through contagion by cohesion, or contagion by structural equivalence (Burt, 1982).
Contagion by cohesion occurs through interpersonal communication on a one-to-one
basis. Contagion by structural equivalence describes how symbolic influences, such as
peer pressure, affect adoption decisions (Burt, 1982). When a RNP is first introduced
to the market, contagion by cohesion operates through word-of-mouth, and contagion
by structural equivalence happens when innovative actors perceive that other
innovators expect them to purchase the new innovative product, which is a form of
indirect pressure (Burt, 1982). Both of these constructivist theories are relevant to
learning about existing products and INPs, but they are less useful when studying
RNPs as the products are largely unavailable in the marketplace. However, the
theories’ emphasis on social interaction in order to learn about new concepts
highlights the importance of social cue presence, when RNPs are being presented.
These can be operationalized by employing various anthropomorphic attributes.
In the cognitivist paradigm, the Mental Simulation strategy (Taylor and
Schneider, 1989) encourages using pictures to help with comprehension, as it helps
individuals to deal with uncertainty and knowledge development; pictures have been
shown to help with the comprehension of RNPs (e.g. Feiereisen et al., 2008). This
theory has two main components: the self and the consumption situation. When a
RNP is being presented in an advertisement for example, then individuals need to
imagine themselves in the situation portrayed (Phillips, 1996). Research in cognitive
psychology suggests that pictures are remembered better than words (Paivio, 1971;
Alesandrini, 1982), hence the use of images is supported as an attribute facilitating
consumer learning. Another theory within the cognitivist paradigm is Analogical
Learning (Gentner, 1989); this introduces verbal analogies (analogies using text) to
facilitate learning about new products (Feiereisen et al., 2008). This strategy uses a
familiar domain (the base) to understand an unfamiliar one (the target) (Gregan-
Paxton et al., 2002). Analogical transmission happens in three steps – access,
mapping and transfer. When learning about RNPs, consumers comprehend the
concept of a new domain (e.g. RNP) based on the similarities to a known one (e.g.
existing products) (e.g. Feiereisen et al., 2008). Many cognitive learning theories
support the use of vividness and interactivity in the context of information processing
and learning; for example, the Cognitive Theory of Multimedia Learning (Mayer,
50
1997) supports the use of visual as well as auditory stimuli and interactivity.
Looking back at Section 1.4.6, it is clear that theories within the behaviourist
paradigm are concerned with observing other individuals’ behaviour; something
which is not always possible when dealing with RNPs. If the theory looks into both
cognitivist and behaviourist aspects of knowledge development (such as the Social
Learning Theory which is considered a theory in between cognitivist and behaviourist
paradigms), then it might be applicable within the RNP context (Franks and Oliver,
2012). Within the constructivist paradigm, relevant learning theories are more suitable
for learning about existing products and INPs; they cannot explain learning about
RNPs, as these products are largely unavailable in the marketplace. Evidently, the
cognitivist framework appears to be more suitable in supporting the various factors
considered within this study, as they are dealing with the intellectual argument of
knowledge, rather than environmental factors. Various theories explained within the
cognitivist framework were employed to explain how new concepts (such as RNPs)
are learnt, such as mental imagery and analogical learning (Feiereisen et al., 2008)
and the Theory of Multimedia Learning (Domagk et al., 2010). As a result, the
cognitivist framework and related theories are explored further in the context of
learning via vividness and interactivity within the RNP context.
1.4.6.1 Theory of Multimedia Learning
The Theory of multimedia learning is designed to tackle a very pragmatic
educational issue: use of new technologies in student learning. Mayer and Moreno’s
(1998) research on the implication of new educational technologies resulted in a
disappointing history. They realized that there was always a strong message in regards
to use of new technologies, but that reality was not what was predicted. There have
been many examples to highlight this. Edison’s proclamation of “the motion picture is
destined to revolutionize our educational system and that in a few years it will
supplant … the use of textbooks” (cited in Cuban, 1986, p.9) never happened, with
Cuban (1986) stating they were not common in classrooms. The future of education
was said to be in the introduction of game-like computer-assisted instruction (CAI)
programs (Mayer and Morano 1998), but these proved to be ineffective. Mayer and
Moreno (1998, p.1) based their research around tackling the following question:
51
“How can we avoid a trail of broken promises concerning the educational benefits of
new educational technologies such as multimedia learning environments?” They
believed the only logical solution was to use instructional technology according to
research based theories and frameworks. The present study aims to investigate a
similar issue of whether new technologies such as multimedia use has any influence
on information promotion and an individual’s learning; this theory can hence be an
appropriate theory as a base for this research.
The Cognitive Theory of Multimedia Learning (Mayer, 1997) supports the
employment of both auditory and visual stimuli, and interactivity. The theory draws
on the Dual Coding Theory (Paivio, 1986), Baddeley (1992) Working Memory
Model, the Cognitive Load Theory (Sweller et al., 1990), Wittrock's (1989)
Generative Theory and the SOI (Select, Organize, Integrate) model of meaningful
learning (Mayer et al., 1996). The model of working memory was first introduced by
Baddeley and Hitch in 1974 in order to explain a more accurate model of short-term
memory (Figure 1). This model was further advanced by Baddeley in 2000. The
Cognitive Load Theory refers to the amount of cognitive effort used within the
working memory (Sweller et al., 1990). The theory is designed:
“ to provide guidelines intended to assist in the presentation of information in a
manner that encourages learner activities that optimize intellectual
performance” (Sweller et al., 1998, p.251).
Wittrock’s model (1989, p.348) focused on the:
“generative learning process … to selectively attend to events … and
generating relations both among concepts … and between experiences”.
52
Figure 1: Baddeley’s Working Memory Model (1974)
According to Mayer’s Cognitive Theory of Multimedia Learning (1997), there is
a ‘multimedia principle’ stating “people learn more deeply from words and pictures
than from words alone” (Mayer 2005 p.47). According to this theory, learners possess
both a visual and verbal information processing system (Mayer 1997). The Theory of
Multimedia Learning is based upon this principle and has three main assumptions
(Mayer, 2005):
1. There are two separate channels for processing information, namely auditory
and visual.
2. Each channel has a limited capacity (as in Sweller’s (1988) cognitive load
theory).
3. Learning is an active process that contains filtering, selecting, organizing and
integrating information based on preceding knowledge.
Further expanding upon assumption 3, Mayer (1997) explained the three main
cognitive processes of selecting, organizing and integrating. Selecting refers to a
learner’s ability to produce text base via incoming verbal information, and to produce
image base via incoming visual information. Organizing is applied to “the word base
to create a verbally-based model of the to-be-explained system and is applied to the
image base to create a visually-based model of the to-be-explained system” (Mayer
and Moreno 1998 p.2). Integration is when the learner connects the visual and verbal
based models.
Mayer and Moreno (1998, p.2-4) concluded that:
“(i) it is better to present an explanation in words and pictures than solely in
words; (ii) when giving a multimedia explanation, present corresponding words
and pictures contiguously rather than separately. (iii) when giving multimedia
explanation, present words as auditory narration rather than visual on-screen
text; (iv) the foregoing principles are more important for low-knowledge than
53
high-knowledge learners, and for high-spatial rather than low-spatial learners;
(v) when giving multimedia explanation, use few rather than many extraneous
words and pictures.”
This model is further advanced in Moreno and Mayer (2007) under the name of the
Interactive Multimodal Learning Environment. With this theory, the interactive
learning environment uses two different presentation modes – verbal and non-verbal –
to represent the information. It also distinguishes between two sensory modalities –
auditory and visual (Penney, 1989). This model supports the use of mixed-modality
presentations as it facilitates the expansion of effective working memory capacity
(Moreno, 2006). Conclusively the employment of vividness and interactivity is
supported within the cognitivist framework.
Within cognitivist learning theories, it is assumed that the learner’s attention is
limited and selective. The more interactive and richer the media, the quicker a learner,
who prefers an interactive learning style, would meet their needs without information
overloading (Zhang et al., 2006). The Theory of Multimedia Learning (Mayer, 1997)
elucidates how individuals learn in a multimedia environment. Applied in the context
of multimedia learning, interactivity is defined as:
“reciprocal activity between a learner and a multimedia learning system, in
which the [re]action of the learner is dependent upon the [re] action of the
system and vice versa.” (Domagk et al., 2010, p.1025).
This theory recognizes that a multimedia environment has the potential to engage the
learner and therefore creates an interactive learning environment. As explained, the
Interactive Multimodal Learning Environment (Moreno and Mayer, 2007), which is
based on the Cognitive Theory of Multimedia Learning (Mayer, 1997), further
supports the employment of interactivity in facilitating learning.
Anthropomorphic attributes are also important to learning. The employment of
social cues online has been supported by Interactive Multimodal Learning
Environments (Moreno and Mayer, 2007). One of the instructional design principles
for Interactive Multimodal Learning Environments is the ‘guided activity principle’,
54
which enables the interaction between students (information receiver) and a
pedagogical agent “who guides their cognitive processing during learning” (Moreno
and Mayer, 2007 p.315). According to this principle, when students (information
receivers) engage with a guided activity, it leads to deeper understanding (Mayer,
2004). The Cognitive-Affective Theory of Learning with Media (CATLM) (Moreno,
2005) expands the Cognitive Theory of Multimedia Learning (Mayer and Chandler,
2001; Mayer, 2005) to media such as virtual reality, agent-based learning and case-
based learning. This theory argues that guiding the learning helps prevent cognitive
overload.
In order to elucidate on the influence of anthropomorphic attributes in consumers’
learning, Social Learning Theory, which is a bridge between behaviourist and
cognitive learning theories (Franks and Oliver, 2012), is discussed further as it
concerns how learning is promoted via social interaction.
1.4.6.2 Social learning Theory
According to Social Learning Theory, having an appropriate degree of social
presence promotes user learning. Bandura (1977) explains how individuals learn from
one another via observation, imitation and modelling. This theory fits within cognitive
and behaviourist paradigms due to its emphasis on attention, memory and motivation
(Franks and Oliver, 2012). There are four processes of observational learning that
explain Social Learning Theory. Firstly with attention processes, learners need to
notice the significant features of the model’s behaviour in order to acquire
observational learning (Bandura, 1977). This attention occurs within the interpersonal
relationship, and is principally influenced by the degree of social presence in an
online environment (Walther, 1997). The retention process, the second of the four
factors, refers to information recall being a requirement for observational learning.
Motor reproduction processes (the third factor) transform symbolic representation to
appropriate actions. Lastly, the fourth factor, motivational processes refer to the
learner having a good reason to imitate the behaviour they value. Social Learning
theory therefore supports the process of learning via social interaction, which can be a
base introducing anthropomorphism. Furthermore, Social Learning Theory is closely
connected to diffusion theory, which makes this theory a suitable framework for the
55
present study.
Employing anthropomorphic attributes results in an increase in social cues and
consequently social presence; this can then have a positive effect on consumer
learning. Bandura (1977) explained three basic models of observational learning
which include i) a Live Model where a real person performs the behaviour to be
learned, ii) a Verbal Instruction Model, which gives verbal information about the
behaviour or/and instructs the learner how to engage in behaviour and iii) a Symbolic
Model, which is a real or fictional character demonstrating the behaviour via any
media sources. Online embodied agents can act as a Verbal Instruction Model or a
Symbolic Model, guiding the users in how to acquire and understand the information
in an online setting (Bandura, 1977). As a result, employing various avatars
(interactive/static) is supported by the Social Learning Theory (Bandura, 1977) as
well as Interactive Multimodal Learning Environments (Moreno and Mayer, 2007).
Social learning theory is linked to the diffusion theory. According to Rogers
(1995) both theories are concerned with an individual’s change of behaviour as a
result of communicating with another individual (virtual or real). Both theories focus
on information exchange as an essential element of the individual’s learning and
behaviour change, but neither theories claim that identical imitation must occur
(Rogers 1995). Bandura (1977) was the first to talk about Diffusion of Innovation and
Social Learning Theory (Bandura 1977 p50-55). Social Learning Theory was also
applied within the context of sociology and Diffusion of Innovation by various
scholars (e.g. Pitcher et al. 1978, Kunkel 1977). The result indicated “diffusion
models portray society as a huge learning system, where individuals are continually
behaving and making decisions through time, but not independently of one another
…” (Hamblin et al. 1979 cited in Rogers 1983 p.305).
Social Learning Theory and Diffusion of Innovation theory have been influencing
each other from the beginning. Two main differences have been identified by
scholars. Firstly, in contrast to Social Learning Theory, Diffusion Theory is more
collective in measuring the influence of modelling as a crude dichotomy of either
innovation adoption or rejection. Social Learning Theory encouraged Diffusion
Theory to measure exactly what is learnt by an adopter while interacting with the
56
network of models and individuals. Secondly, Diffusion Theory brings the element of
time into Social Learning Theory, therefore making social learning scholars to look
into behaviour change as a process. There are also two main similarities. Firstly, both
social learning theories and diffusion theories explain that a behaviour is not always
exactly imitated by individuals; rather learners/adopters might abstract or generalize
the information observed. Secondly, both theories have an emphasis on the
convergence aspects of behaviour change. Whereas conventional learning research
and mass communication research focus on the individualistic aspects of learning,
Social Learning Theory and Diffusion Theory are concerned about interpersonal
information exchange. It is believed that Social Learning theory has led Diffusion
Theory into this promising direction (Rogers 1995). As a result of a close interaction
between Social Leaning Theory and Diffusion of Innovation, and as these theories
have been influencing and in some parts, improving each other, Social Learning
Theory seems to be an appropriate base framework in support of social interaction,
while employing anthropomorphic attributes in learning about innovation (RNPs).
1.4.7. The Research Gap
RNPs are increasing in various industries, but at the same time, the majority of
these products are failing to succeed (Moore, 1991). The reason could be the unique
characteristics of RNPs as explained in Section 1.4.1. Over the past two decades,
academics have started to develop ways to modify methods and marketing strategies
to meet the higher-risk, higher-reward domain of RNPs (Lehmann, 1994; Urban et al.,
1996; Moreau, 1997; Hoeffler, 2003). The distinctive characteristics of RNPs, such as
high uncertainty and the lack of individuals’ knowledge (Hoeffler, 2003; Alexander
et al., 2008) mean they need to be presented in a way to facilitate consumers’
understanding. This is particularly important for RNPs, as consumer understanding
begins at the time of product evaluation. Facilitating the process of RNP
comprehension is likely to lead to greater adoption of RNPs. Product information
should be promoted in the most effective way to enable consumer learning, and hence
improve their adoption and online behaviour towards RNPs. Very little attention has
been directed towards RNPs’ presentation formats as a new product category. This
thesis intends to fill this gap by examining various presentation formats as a means to
57
facilitate consumer understanding and improve consumer online behaviour towards
RNPs.
Telepresence, consisting of vividness and interactivity, is an attribute that
improves an individual’s learning and online behaviour; this is supported within the
academic literature. From an information processing perspective (for instance,
Paivio’s (1977) imagery theory of Dual Coding) and from a learning perspective,
theories such as Multimedia Learning (Mayer, 1997) encourage using vividness and
interactivity attributes to enhance an individual’s learning and online behaviour.
There is no study to date examining the impact of adding telepresence (vividness and
interactivity) within the context of RNPs. This gap is identified and examined further
in Paper 1 (Chapter 2).
Anthropomorphic attributes are another aspect that can improve an individual’s
learning and online behaviour (Holzwarth et al., 2006; Keeling et al., 2010). It is
supported within various learning theories as a means to enhance individual learning,
such as the Theory of Multimedia Learning (Mayer and Moreno, 1998) and the Social
Learning Theory (Bandura, 1977). However, there are inconsistencies in previous
findings in using various levels of anthropomorphic attributes, and how influential
each can be on consumer online behaviour. Some studies debated that when
individuals are dealing with a higher level of anthropomorphism (such as a speaking
human-like avatar), they perceive the information as more credible and they perceive
the avatar as more trustworthy and friendly in comparison to that which has a lower
level of anthropomorphism (e.g. Nowak and Rauh, 2005; Donath, 2007; Jin, 2009).
On the other hand, there is a body of literature reporting that a more human-like
avatar can be evaluated more negatively (e.g. Groom et al., 2009; Keeling et al.,
2010).
Therefore there is a gap in understanding what the impact of different
anthropomorphic attributes upon an individual’s learning and online behaviour is, and
how anthropomorphism is perceived by individuals. This gap has been identified and
further investigated, within the context of RNPs, in Papers 1 and 2 (Chapters 2 and 3).
58
Rogers’ (1995) Diffusion of Innovation theory is adopted within the present
research as explained in Section 1.4.5. There are two main criticisms for Rogers’
model of innovation diffusion: 1) individual categorization into adopter categories
based on innovativeness and 2) the characteristics of innovation (Robertson, 1967;
Lowrey, 1991) which are addressed within this research. The debate around the
impact of consumer characteristics on consumers’ innovation adoption is ongoing.
This has been criticized as being overlooked by Rogers’ (1995) Theory of Innovation
Diffusion. Different consumer characteristics such as socio-demographics, and
individuals’ values, goals, experiences and knowledge have been investigated in order
to address the initial debates around consumer characteristics (Midgley and Dowling,
1978; Daghfous et al., 1999; Rogers, 2003; Weijters and Roehrich, 2004; Tellis et al.,
2009). In the present research, consumer characteristics were investigated in more
detail in order to understand the influence this factor could have on individuals’
learning and online behaviour towards RNPs. Paper 1 controlled for consumer
characteristics (Chapter 2) and in Paper 3, consumer characteristics were examined in
more depth (Chapter 4). In Paper 3, consumers’ motivational sources, as a distinct
attribute impacting the innovation diffusion process was examined further. No
research to date has examined consumers’ motivational sources within innovation
diffusion therefore Paper 3 aims to fill this gap, within the context of RNPs, by
examining this element further. For this purpose, the Motivated Consumer
Innovativeness (MCI) model, introduced by Vandecasteele and Geuens (2010) was
adopted.
Rogers’ (1995) innovation characteristics have also been criticized by academics
(e.g. Robertson, 1967; Lowrey, 1991). It is claimed that most research on the
diffusion of innovation in the consumer domain is concerned with the identification of
differences between adopter groups and little has been done to address the
characteristics of innovations (Gatignon and Robertson, 1985b; Mahajan et al., 1990).
In order to address this criticism, the RNPs selected for this research are within the
same product domain of kitchen appliances. This is done in order to minimize the
effect of innovation characteristics on consumers’ perception, through the process of
diffusion of innovation.
Overall a key research gap identified highlights the lack of research examining
59
RNPs as a new product category, in order to understand consumers’ learning and
online behaviour towards this unique product category. RNP promotion and
presentation elements (vividness, interactivity and anthropomorphic attributes) have
also received little attention so far, which is examined further within this research.
Within the anthropomorphic context, a further research gap emerged as in how and
why it influences consumers’ perception, comprehension and online behaviour
towards RNPs. Consumer innovativeness needed more attention as no research to date
has examined the motivational factors influencing consumers’ innovation adoption.
Therefore this research aims to fill these gaps by investigating each factor within the
three papers presented.
1.5 Methodology
This section looks into the research design elements of the project. It starts by
introducing the philosophy of research, then explaining the methodological and
research design.
1.5.1. Philosophy of Research
In order to uncover the origin of Research Paradigms, it is important to
understand the concept of reality. Reality was first questioned by philosophers as
early as 6th century BC. Milesians were claimed to be the first true philosophers
asking the question of “what is reality?” (Lee and Lings 2008), and were the first to
be interested in understanding reality and not relying on supernatural explanation.
Pythagoras (571-496 BC) also questioned reality. He believed “truth should not be
accepted but instead proved” (Lee and Lings, 2008 p.26). Heraclitus (circa 500 BC)
debated further on the concept of knowledge, believing that knowledge coming from
one’s senses is not trusted, and that true knowledge comes from reason and not
observation. The progress in understanding reality and knowledge in a pre-Socratian
era is a scientific progress leading to the birth of Western Philosophy by Socrates
(470-399 BC). His main contribution was to move philosophy from questioning
reality to morality. Aristotle (384-322 BC) followed Socrates’ debates and eventually
introduced the concepts of induction and deduction. Induction is concerned with
moving from observation into generating theory, whereas deduction is moving from
60
theory into observation based on theory. Aristotle’s arguments on induction and
deduction are the baseline for the concept of ‘generalisation’ (Lee and Lings 2008).
After the domination of the Christian church, it was no earlier than the 14 th
century that philosophers started to investigate science. Galileo (1564-1642) applied
experimentation, induction and observation as his research approach. Descartes
(1596-1650) was the initiator of modern philosophy, which is also known as
rationalism. Descartes argues that observed data are not trustworthy in comparison
with pure reason. Locke (1632-1704), on the other hand, introduced empiricism which
explains that the knowledge humans acquire comes from their observations. Hume
(1711-1776) advanced empiricism by opposing the idea that induction can lead to
knowledge generation. Many other philosophers such as Kant (1724-1804), Hegel
(1770-1831) and Russell (1872-1970) adopted the empiricism philosophy and
advanced the paradigm (Lee and Lings 2008).
The establishment of the Vienna Circle in 1924 was when positivism and the
scientific method were born (Aliyu et al. 2014). The primary members of the Vienna
Circle were Schlick (1882-1936), Neurath (1882-1945) and Carnap (1891-1970).
They introduced logical positivism, which is concerned with how true knowledge is
obtained through science. Logical positivism is concerned with the verifiability of
ideas through observation of elements, and the ability to testing it empirically. Logical
positivist philosophers also believed in reductionism. Reductionism explains how
theories within social sciences can be reduced to those of more fundamental sciences.
Feigl (1902-1988) introduced a logical empiricist philosophy of science, which is also
known as realism. Feigl deviated the idea that theoretical terms and concepts are only
defined based on empirical observation, arguing that although many variables are not
directly observable, they can be measured in the context of theoretical explanation.
Positivism and Realism are the most common research approaches within social
sciences, however a contradictory view emerged simultaneously with the introduction
of empiricism. Bishop Barkeley (one of the key empiricists and a student of Locke)
(1685-1753) introduced the idea that everything exists within an individual’s mind.
He believed reality is one’s interpretation of the world. This is the root of the
interpretivist research approach. The main difference between interpretivist and
traditional research approaches is that the emphasis on understanding rather than
61
explaining a phenomenon (Kaboub 2008; Lee and Lings 2008). Furthermore, a
phenomenon is always bound to the time and context it is being studied within.
This research is concerned with the explanation of a phenomenon based on
established theories. The study’s aim is to examine relationships and identify the
effect of variables on each other. Given this, an interpretivist approach is not
appropriate to this study. The study also intends to add to the body of knowledge by
generalizing the findings, which is contradictory to an interpretivist approach.
Positivism and Realism both fit within the aims of the study and are suitable
approaches. However in order to justify the best approach, the arguments on both
approaches are investigated further.
The realist approach is developed based on positivism and empiricism. The main
criticism by realist scientists of a positivist approach is towards the positivist belief
that observation could disprove a theory. This condemnation is the main difference
between the realism and positivism (Lee and Lings 2008). This criticism is based on
the idea that even one observation contradictory to all existing observation can reject
a theory. Popper (1902-1994) argued that researchers should look into falsifying a
theory, instead of proving a theory, with observations disproving the theory. Another
criticism of the positivist approach is the notion that anything observable exists,
unobservable things don’t exist and only a proposition that can be empirically tested
is verifiable (Lee and Lings 2008). In order to address this criticism, positivism’s
conceptualisation of theory needs to be explained. Positivists argue that theories are
formulated inductively from observation and can also be tested deductively against
observation (Thomas 2006 p.63). “Theory serves to explain and predict phenomena.
The creation of theory, rather than the production of descriptions, is the ultimate
purpose of science” (Thomas 2006 p63). Kerlinger (1986) explains that science is
practiced by moving back and forth between observations and theories, therefore even
a contradictory observation does not result in the theory being rejected.
On the concept of unobservable variables, Kerlinger (1986) explains that a theory
has three important characteristics: (i) it has a set of propositions, which contain
interrelated ‘constructs’; (ii) a theory “provides a systematic view of the interrelations
amongst the constructs” (p.9); and (iii) a theory explains ideas based on the
relationship between variables. Positivists’ idea of construct is anything that is
62
observable or presumed to exist in the world. A construct is either constitutively or
operationally defined. A constitutively defined construct is defined in terms of other
constructs, whereas the latter is defined empirically. In order to understand an
unobserved variable, positivists identify latent variables. A latent variable is “an
unobserved entity presumed to underline observed variables” (Kerlinger 1986, p.37).
Positivists can only measure observed behaviours, “which are taken as indicants of
the existence of the non-observable latent variable” (Thomas 2006 p.65).
Looking back at social sciences literature, both from a historical perspective and
contemporary research, positivism remains as the dominant research paradigm within
social sciences (Schon 1983; Fraser et al. 2012). Historically it is regarded as the
“received view” that dominated research in physical and social sciences for more than
400 years (Guba and Lincoln 1994). Some scholars even argue that the dominant
research paradigm for most of the 20th century was in fact positivism (Gray 2014).
Conclusively it seems a suitable paradigm to be considered in a social science study.
However, the suitability relies heavily on the nature of the research and the variables
under investigation.
This study is concerned with relationship between observable and latent variables.
The author believes any observable behaviour can be an indication of a latent
underlying variable. Furthermore, the aim is to test established theories, via
observation and to debate over models and existing theories based on observation. As
indicated, the positivism paradigm argues that theories are framed inductively from
observation and can be tested deductively against observation (Thomas 2006).
Crowther and Lancaster (2008) explain that positivist studies generally adopt a
deductive approach. In order to adopt an inductive or deductive approach, the nature
of the present study needs to be elucidated. This study is within the innovation context
and aims to understand the novel phenomenon of RNP promotion. There are two
aspects that make this study innovative. Firstly, various variables such as virtual
reality and anthropomorphism are studied to understand the effectiveness of these
new promotional technologies on consumer behaviour and learning of RNPs. The
variables examined are all new elements that are inspected recently in consumer
behaviour literature, and the relationships between these variables and RNP
comprehension and consumer behaviour is under-investigated.
63
Secondly, it is the nature of the products under study within this research, that of
RNPs. RNPs are a product category that has been studied in more depth over the
recent years due to an increase in popularity of these innovative products in various
industries (e.g. Feiereisen 2008). RNPs are introduced increasingly to the marketplace
as an important element for business to maintain their competitive position (e.g. Chao
et al. 2012). This study aims to examine RNPs, and how using new advanced
technologies such as virtual reality and anthropomorphism can improve consumers
learning and behaviour towards them. As the variables in this study are novel, and the
relationships are not investigated as such, the inductive approach is not suitable at
first. It appears that the only way these relationships and variables can be investigated
is via observation; therefore the deductive approach is a suitable method. However, in
studying RNPs, scholars mainly referred back to existing frameworks and theories
that were adopted for other product types, especially INPs (Incrementally New
Products). The study challenges existing frameworks and models in the RNP context,
as there are contradictory findings within innovation literature in regards to the
suitability and effectiveness of existing theories. This matter is rightly debated within
the innovation literature. For instance, Hoeffler (2003) explains that existing
frameworks and models cannot automatically be assumed to be suitable for RNPs,
further elaborating that standard preference measurement techniques (e.g. conjoint),
cannot effectively predict the functionality of RNPs (Hoeffler 2003). The learning
process for RNPs is also different to other products (e.g. Feiereisen 2008); hence this
study aims to base observations on existing frameworks, with an intention to deduct
from these observations. This will improve or challenge existing theories and
frameworks. As a result, the study follows Thomas’s (2006) indication that positivism
believes theories are framed inductively from observation and can be tested
deductively against observation. The study also intends to generalize findings into the
body of knowledge and aims to formulate new statements within the context of RNPs.
Conclusively, the positivism approach is adopted in this study and the research is
designed based on this paradigm.
64
1.5.2. Methodological design
In order to select the most suitable methodology for conducting this study, the
researcher needs to not only consider the research paradigm adopted, but also the
research questions. Based on an objective underpinning of this study a deductive
approach is selected. The objectives of the present study (Section 1.1) can also
determine whether qualitative or quantitative data is most suitable to achieve these
objectives and test the hypotheses developed within the study. Qualitative methods
are concerned with behaviour as a consequence of an individual’s interpretation of the
world. Qualitative research techniques, in the early stages of a research project, can
help generate hypotheses (Bartos, 1986). These techniques are more inductive than
deductive in nature. This means the study begins with understanding interactions,
which are examined for broader patterns, in order to build theory (Deshpande, 1983).
Qualitative methods are criticized for their ambiguity when assessing the value of the
research for reliability, validity, objectivity and relevance (Gummesson, 2000).
Quantitative research is traditionally associated with a logical positivist approach
(Deshpande, 1983). This method is considered objective and outcome oriented
(Feiereisen et al., 2008) and is suitable for the purpose of theory testing (Deshpande,
1983) whilst a qualitative method is more appropriate if the aim is to generate the
theory. Quantitative methods are more suitable when testing hypotheses, and asking
particular questions (Hackley, 2003). The data generated via a quantitative method
can be used for the purpose of generalization. In line with the objectives of the present
study, the researcher is interested in examining the relationship and effect of
constructs and latent variables. The aim is to test established theories based on
observation within a certain context. Quantitative methodology is best fitting the
purpose of the study, and is also the most suitable within the positivism research
paradigm.
1.5.3. Research Design
There are three papers within this thesis. Two papers (Papers 1 and 2) follow an
experimental design and one paper (Paper 3) uses an online survey. Samples used
across all three papers was drawn from an online crowdsourcing website commonly
used in consumer/psychological experimental research namely, Amazon Mechanical
65
Turk2(Mturk). Mturk has proved to be a great opportunity for data gathering, with
Mturk participants generating high quality reliable results consistent with standard
decision-making biases (Goodman et al., 2013). The Mturk participants used are US
based individuals who have access to CMEs, and were therefore considered a suitable
sample for the purpose of the studies.
In order to identify suitable RNPs, relevant information on RNPs was gathered
from different websites and forums (e.g. TechCrunch.com, mashable.com, extracted
from technorati.com`s ‘Top 100 most read blogs’). Seventeen products were selected
initially that had enough information and pictures available publicly and were
considered as very innovative. An initial screening of these products resulted in
eliminating eight of them. Three had become available in the marketplace and five
were assessed as less innovative and harder to understand. The remaining nine
products were tested using an online questionnaire to confirm their status as RNPs.
For this purpose, perceived product newness was measured using a scale based upon
Gregan-Paxton et al.’s definition of RNPs (2002) and Hoeffler’s framework of RNP
evaluation (e.g. Alexander et al., 2008). The scale was modified by Alexander et al.
(2008) to assess the unique characteristics of RNPs, such as the uncertainty in
estimating the benefits of RNPs. Four questions developed from Alexander et al.’s
study, where two of the questions discussed whether participants understood and liked
the product (Group 1), and the other two implied participants needed to change their
behaviour to do new things (Group 2).
For Group 1, the Analysis of Variance (ANOVA) result indicated a significant
difference among products means with products E-tomb, Dismount Washer and Bio
Robot Fridge coming as significantly more understandable and likable. For Group 2,
the ANOVA result also showed a significant difference between means across the
products with iDropper, Washing Machine in Wardrobe and Digital Make up Mirror,
as the products participants were willing to change their behaviour for, in order to do
something with them. In order to address the impact of product characteristics upon
consumers’ online behaviour, Dismount Washer and Washing Machine in a
Wardrobe, from the domain of electrical kitchen appliances were selected for Studies
2 www.Mturk.com
66
1 and 2, and Bio Robot Fridge from the same domain was chosen for Study 3.
Sections 1.5.3.1 and 1.5.3.2 explain the research designs employed within studies.
1.5.3.1. Experimental Design
Online Experiments were designed for the purpose of Studies 1 and 2. An
experiment is a causal research design, hence is a strong method to use when causal
connections need to be identified. They have the capacity to differentiate between the
supposed causes from the observed effects (Churchil and Iacobucci, 2004), and are
commonly used for hypothesis testing. Experimental designs provide an element of
control for the researcher, and are capable of providing convincing evidence of
causality (Churchil and Iacobucci, 2004). This method enables the researcher to
manipulate various variables, and display various stimuli. As this study is concerned
with the manipulation of vividness, interactivity and anthropomorphic attributes, the
experimental design is a suitable design for the purpose of the study.
One of the main advantages of experiment design is its high internal validity (Lee
and Lings, 2008). As the present studies are examining CMEs, undertaking online
experiments is deemed suitable. Online experiments are becoming a mainstream
method in the field of psychology (Birnbaum, 2004). Online experiments have the
following advantages. By running a web experiment, a large sample can be employed
in a short period of time which makes “statistical tests powerful and model fitting
very clean” (Birnbaum, 2004, p.813); furthermore, using web experiments increases
the generalizability of the study and also provides the opportunity to recruit
specialized participants.
Factorial designs are commonly used in experiments. These involve the variation
of several independent variables within a single study and are rich with information
(Keppel and Wickens, 2004). There are three types of factorial designs: within-
subject design, within-subject factors/between-subject factors and between-subject
design. The within-subject design is when a single sample of subjects serves in every
condition. The disadvantage of this design is that the experiments become sensitive to
the order of the conditions, and exposure to all treatments may be cumbersome for
individuals (Keppel and Wickens, 2004). The between-subject design is when every
condition uses a unique sample of subjects. The main disadvantage of this design is
67
the requirement of a large sample, as samples are less sensitive in comparison to some
other approaches (Keppel and Wickens, 2004). The within-between/subject factors
are a mixture of both designs. Due to the purpose of this study, and the availability of
participants, a between-subject design was selected.
1.5.3.2 Survey Design
Study 3 followed a survey design. The purpose of a survey is to produce statistics,
“that is quantitative or numerical description about some aspects of the study
population” (Fowler, 2008, p.1). The survey method can be divided into two broad
categories: questionnaires and interviews. Questionnaires are more suitable when
gathering quantitative data, whereas interviews are a more suitable method for
gathering qualitative data. Due to investigating variables within CMEs, online surveys
are appropriate. Online surveys are deemed to have a lower cost in comparison to
other methods (Porter, 2004). Data entry via the online survey is close to the natural
setting when analysing online variables. Next the research design elements related to
each study are explained.
1.5.3.3. Study 1
Paper 1 is concerned with presentation formats of vividness, interactivity and
anthropomorphic attributes. Three online experiments were designed with dependent
variables consisting of consumers’ comprehension, attitude and purchase intention
towards two RNPs. The independent variables were vividness (Experiment 1),
interactivity (Experiment 2), and anthropomorphic attributes (Experiment 3); each of
which were presented at two levels (Low vs. High). Each online experiment was a
single factor between-subject design for two separate products. Online surveys were
used for data collection and data was analysed using Multivariate Analysis of
Covariance (MANCOVA) and Confirmatory Factor Analysis (CFA) techniques.
MANCOVA is the Multivariate Analysis of Variance (MANOVA) with a covariate,
and is a method used to “simultaneously explore the relationship between several
categorical independent variables … and two or more dependent variables” (Hair et
al., 2010, p.18). MANOVA is a useful method in experimental situations when the
researcher is concerned with hypotheses testing (Hair et al., 2010). MANCOVA can
be used in combination with MANOVA to remove the influence of any uncontrolled
68
metric independent variables (namely covariates) on the dependent variables (Hair et
al. 2010). Therefore, this technique was an appropriate method to be used in this
study. CFA are “well measured variables represent[ing] a smaller number of
constructs” (Hair et al., 2010, p663). CFA indicates how well the theoretical
specification of the factors match reality. CFA enables the researcher to either
approve or reject the preconceived theory (Hair et al., 2010). As the scales used in this
study, for the purpose of variable measurement, were all extracted from existing
academic literature, there was only a need to check if the measurement model was
valid. CFA was therefore an appropriate technique for the purpose of the study. SPSS
(v22) and AMOS were the statistical tools used for data analysis. More detail on each
experiment, the manipulation checks and sampling are provided in Chapter 2.
1.5.3.4. Study 2
This study was concerned with various levels of online anthropomorphism.
Online experiments were selected for the purpose of this study. The dependent
variables were perceived mindless and mindful anthropomorphism, comprehension,
attitude and purchase intention. The independent variables consisted of content
interactivity and avatars. There were five conditions where the independent variables
were manipulated. Condition 1 was a base level (no content interactivity and no
avatar), condition 2 was low content interactivity, condition 3 was high content
interactivity, condition 4 was static avatar and condition 5 was human-like avatar.
CFA and Independent Sample T-tests were performed using SPSS (v22) and AMOS.
More detail on each experiment, the manipulation checks and sampling are explained
in Chapter 3.
1.5.3.5. Study 3
This study looked at the relationships between the variables of innate
innovativeness (consisting of Functional Motivated Consumer Innovativeness (fMCI),
Hedonically Motivated Consumer Innovativeness (hMCI), Socially Motivated
Consumer Innovativeness (sMCI) and Cognitive Motivated Consumer Innovativeness
(cMCI)), comprehension and attitude. The study was not concerned with the
experimental manipulation of the variables therefore an experiment design was not a
suitable method for this study. Given this, an online survey in form of an online
69
questionnaire was selected and designed as the study required quantitative data in
light of the nature of the study.
For data analysis, CFA and Structural Equation Modelling (SEM) were used.
SEM is a popular technique used within various disciplines due to its generality and
flexibility (Lei and Wu, 2007). SEM statistically “represents an extension of general
linear modelling (GLM) procedure, such as ANOVA and multiple regression
analysis” (Lei and Wu, 2007, p.33). SEM is mainly used due to its capability of
studying relationships among latent constructs (Hair et al., 2010). The method takes a
confirmatory approach to the “multivariate analysis of structural theory, one that
stipulates causal relations among multiple variables” (Lei and Wu, 2007, p.34). The
aim of SEM is to conclude if a hypothesized theoretical model is consistent with the
data collected to reflect the theory (Kline, 2005; Lei and Wu, 2007). This technique
was suitable for this study, as the relationships between multiple variables were the
centre of the analysis. SPSS (v22) and AMOS were the statistical tools used for data
analysis. More information about the research design for this paper is explained in
Chapter 4.
1.6 Conclusion
This section revisits the objectives introduced at the beginning of the thesis,
which follow from previous discussions. The objectives have been designed based on
the gaps evident in the existing academic literature. By achieving these objectives,
this thesis contributes to the body of knowledge, by filling the gaps identified within
the academic literature. The objectives are as follows:
To identify the effect of various presentation formats of vividness,
interactivity and anthropomorphic attributes on consumer responses to RNPs
(i.e. product comprehension, attitude and purchase intention).
To achieve this objective, study 1 manipulated vividness, interactivity and
anthropomorphic attributes in an online experiment design in order to investigate the
impact each independent factor had upon the dependent variables of product
comprehension, attitude and purchase intention. The findings add to the telepresence
literature, the anthropomorphism body of knowledge, as well as the RNP literature.
The findings also add to the related learning theories of the Theory of Multimedia
70
Learning (Mayer, 1997) and the Social Learning Theory (Bandura, 1977) within the
context of RNP.
To examine whether the insertion of various types of anthropomorphic
attributes influence consumers perceived mindless and mindful
anthropomorphism while learning about RNPs.
To determine the relationship between consumers’ perceived mindless and
mindful anthropomorphism and consumers’ response towards RNPs (i.e.
product comprehension, attitude and purchase intention).
To achieve these objectives, Study 2 investigated the conditions in which the
independent variables of content interactivity (low vs. high) and avatar (static vs.
human-like) were manipulated, following an online experimental design. The
impact was then analysed towards the dependent variables of mindful and
mindless anthropomorphism, product comprehension, attitude and purchase
intention. The findings add to the anthropomorphism literature considerably.
Another significant contribution is made to the academic literature on consumer
learning and online behaviour, linking the anthropomorphism literature within the
context of RNP.
To examine the impact of consumer innovativeness (i.e. functional, hedonic,
social, cognitive) on consumers’ comprehension and attitude towards RNPs.
To achieve this objective, Study 3 examined the relationship between functional,
hedonic, social and cognitive consumer innovativeness, and product comprehension
and attitude. An online survey was employed for the purpose of this study. The
findings add to the innovation adoption literature. It shed lights on the Diffusion of
Innovation Model (Rogers 1962) within the context of RNPs. Furthermore, the
findings contribute towards the literature on the Motivated Consumer Innovativeness
Model (MCI) (Vandecasteele and Geuens, 2010).
1.7 Structure of the thesis
The next three chapters introduce each study, in an independent academic paper
format. The final Chapter (Chapter 5) reviews and discusses the findings of each
paper and concludes with contributions to academic knowledge.
71
This thesis is structured into 5 chapters, including the present chapter.
Chapter One: This chapter is an introduction to the topic, including thesis
overview and objectives, contribution to knowledge, literature review of the main
elements of the research (product categorization, RNP and product presentation, VR,
consumer grouping and consumer learning), methodology and conclusion.
Chapter Two: In this chapter, Study 1 is introduced. This study looked into the
presentation elements of RNP promotion and the influence each element had on
consumers’ comprehension, attitude and purchase intention. The chapter is in form of
a stand-alone academic paper. It introduces the theoretical background supporting the
arguments around the presentation elements of vividness, interactivity and
anthropomorphic attributes, and consumer learning and online behaviour. The online
experiments performed are explained and the findings discussed. Finally, the
theoretical and managerial implications are presented, whilst acknowledging the
paper’s limitations and proposing future lines of research.
Chapter Three: This chapter focuses on the second study, which investigated
mindful and mindless anthropomorphism, how it is perceived, and how various
manipulations of the anthropomorphic attributes influence consumers’ perception,
comprehension and online behaviour. The chapter is in form of a stand-alone
academic paper. This chapter digs deeper into the area of anthropomorphism and
gives a more in-depth investigation of anthropomorphism from theoretical and
practical points of views. Content interactivity is also introduced and various types of
anthropomorphic traits such as static and human-like avatars are discussed. The
experiments performed are explained and the results are discussed. Finally there is a
general discussion and conclusion alongside the study’s limitations and future
research recommendations.
Chapter Four: This chapter presents the third study, which looks in more depth
into the topic of consumer innovativeness and the impact it has on consumers’
comprehension, attitude and purchase intention. The chapter is in form of a stand-
alone academic paper. The study investigates the theoretical background of
innovation diffusion, in particular consumer innovativeness, and further investigates
the motivational sources that are influential upon consumers’ innovation adoption.
72
Online surveys employed within this study are then introduced and discussed, with
the results presented. Finally, there is a general discussion followed by the study’s
managerial implications, its limitations and future research proposals.
Chapter Five: This chapter presents this thesis’ conclusions, including a review
of findings separated for each paper, the core contribution, limitations and future
research.
73
Chapter 2: The effect of Telepresence and Anthropomorphic
attributes on consumer’s comprehension of RNPs
2.1 Introduction
As technology advances, new products are being introduced to the marketplace.
These new products include products that challenge the classification of existing
categories. This group of products are called “really new products”. Really New
Products (RNPs) are products that are either (i) very new and consumers do not have
any information about them, or (ii) exist, but have been expanded significantly
(Gregan-Paxton et al., 2002). Hoeffler (2003) described RNPs as products with
greater benefits than Incrementally New Products (INPs), although consumers need to
change their behaviour in order to achieve the potential benefits (Hoeffler, 2003). The
prevalence of RNPs is increasing in various industries, especially in high-technology
industries (Moore, 1991). The development of RNPs is also a strategic priority for
most companies (Feiereisen et al., 2008). However, 40% to 90% of new products fail,
and compared to less innovative products, highly innovative products fail at even
greater rate (Cierpicki, 2000). One of the reasons RNPs fail in the marketplace, in
comparison to Incrementally New or existing products, is that companies spend less
attention and money on understanding the learning processes required for RNPs and
the consequences for communication message strategies (Feiereisen et al., 2008).
Furthermore, the standard preference measurement techniques, such as conjoint,
cannot predict the utility of RNPs as accurately as other product types (Hoeffler,
2003). In the case of RNPs, where consumers have limited/no knowledge about the
products, they need to “construct preferences at the time of measurement” (Hoeffler,
2003). Consequently, consumers need to acquire enough information to be convinced
about the potential benefits of using RNPs. In contrast to INPs, where consumers have
baseline knowledge/experience about the product (or a related domain) which assist
their learning, more learning is required for RNPs as consumers do not have any base
knowledge/experience (Hoeffler, 2003).
Consumers learn about RNPs at the time of product evaluation. The information
consumers receive at the time of evaluation is crucial in influencing their behaviour
74
towards RNPs. According to persuasion theorists, the consumer’s comprehension of
the RNP is a precondition in the formation of their attitudes, intentions and behaviours
(e.g. Ratneshwar and Chaiken, 1991). As more learning is required to understand
RNPs in comparison to other product types (Hoeffler, 2003), facilitating the process
of RNP comprehension is likely to lead to greater RNP adoption. As a result, RNP
information should be promoted in the most effective way to enable consumers’
learning.
The Internet is used as a promotion channel in creating new product awareness
(Bickart and Schindler, 2001) and increasing the new product adoption rate (Prince
and Simon, 2009); it is therefore employed in this study as a medium to present RNP
related information. The information needs to be presented in a way to attract
consumers (Reiman, 2001; Cunningham et al., 2007) and meet their high expectations
(Palmer and Griffith, 1998; Chevalier and Ivory, 2003). Therefore through employing
various online technologies, information can be offered in an attractive, convincing,
and clear format. One such technology is that of Virtual Reality (VR), which is
emerging as being significant in competitive advantage (Lee and Chung, 2008; Bruno
and Muzzupappa, 2010). Another attribute is anthropomorphism, that is proved to
facilitate consumer’s learning and increase information credibility (Holzwarth,
Janiszewski and Neumann 2006) (e.g. Franceschi et al., 2009; Guttentag, 2010;
Keeling et al., 2010).
The aim of this study is to examine website design factors, which not only make
the information more appealing for customers, but also impact upon consumer
comprehension of RNPs, their attitude formation and their purchase intention towards
such products. This paper starts by examining different learning theories and how
they relate to RNP comprehension, attitude formation and purchase intentions. It then
considers the impact of how various online presentation formats influence consumer
learning and online behaviour, leading to the development of specific hypotheses.
Finally the experimental design is detailed and the measurement methods are
explained.
75
2.2 Theoretical Background
Although many theories within various paradigms of behaviourism, cognitivism
and constructivism support learning about RNPs, not all paradigms fit with the
concept of this study. Behaviourist theories are concerned with the acquisition of new
behaviour through conditioning. For instance, the stimulus-response framework can
explain imitation effects, when late adopters accept an innovation as a result of
observing the behaviour of early adopters (Thorndike, 1931). This does not fit with
the existing study, when examining the effect of telepresence, as imitation is not
possible; however looking into Anthropomorphism, as social interaction occurs, this
framework can be applied. The constructivist paradigm explains how the learner
constructs their own knowledge, based on their previous interaction with the world
and their interpretation of this information into a personalized knowledge base
(Piaget, 1950); again, this is not possible due to the nature of the product under study.
Consequently, as cognitive theorists attempt to understand how the process of
learning occurs (e.g. Chomsky, 1957; Simon, 1957; Bruner, 1960), this study is
mainly concerned with theories within the cognitive paradigm of learning as it has
been deemed more suitable.
Many cognitive learning theories support RNP comprehension, such as the
Mental Simulation strategy (Taylor and Schneider, 1989) which encourages the use of
mental imagery (i.e. encouraging individuals to imagine themself using the new
product) to facilitate learning about RNPs (e.g. Feiereisen et al., 2008). Cognitive
psychology studies (Paivio, 1971; Alesandrini, 1982) support the argument that
pictures are better remembered than words, hence this strategy helps individuals to
deal with uncertainty and knowledge development through this mental process.
Verbal analogies within analogical learning also have been proven to facilitate
learning about RNPs (Feiereisen et al., 2008). This strategy helps by encouraging
consumers to refer to their familiar existing knowledge base, in order to understand
the unfamiliar RNP concept (Gregan-Paxton et al., 2002). This strategy is proved to
facilitate consumer’s learning about RNPs (e.g. Feiereisen et al., 2008). Furthermore,
many cognitive learning theories support the use of vividness and interactivity in the
context of information processing and learning; for example, the Cognitive Theory of
Multimedia Learning (Mayer, 1997) supports the use of visual as well as auditory
76
stimuli and interactivity. In this study, related learning theories are therefore drawn
upon when forming the hypotheses.
Learning about RNPs leads to the prediction of consumer behaviour towards
these products. Persuasion Theory explains how comprehension is a prerequisite to
the formation of attitudes, intention and behaviour, especially under the central or
systematic processing route (Ratneshwar and Chaiken, 1991). According to the
Elaboration Likelihood Model (Petty and Cacioppo, 1981), there are two routes to
persuasion. Consumers are either involved consumers, who process arguments in the
persuasion attempt and elaborate on the arguments resulting in a positive or negative
evaluation; or uninvolved consumers, who will use peripheral cues to form an
empirical judgment, which can be positive or negative. Involved consumers put more
cognitive effort towards the central route, characterized by placing attention to the
message content; whereas uninvolved consumers follow the peripheral route, paying
attention to peripheral cues (Petty and Cacioppo, 1981; Petty and Cacioppo, 1986).
RNPs are by definition complex, ‘high-involvement’ products (Hoeffler, 2003), and if
individuals do not pay attention to the message content and follow the peripheral
route, they will have difficulty understanding RNPs. It is therefore predicted that
consumers will be highly-involved in the process, and as a result, will develop
positive attitudes and behavioural intent for the new product, provided they learn
about product-relevant information (Bettman, 1979) and the benefits of the product
(Lehmann, 1994; Urban et al., 1996). Therefore, it is hypothesized that product
comprehension will have a positive effect on consumer attitude and intention towards
RNPs.
H1: Comprehension will be positively related to both attitude and purchase
intention towards RNPs.
2.2.1. New/innovative products
Although both INPs and RNPs are considered as new/innovative products,
Hoeffler (2003) explained that RNPs have greater benefits compared to INPs. There
are four main differences between INPs and RNPs (Alexander et al., 2008). Firstly,
unlike INPs, RNPs enable consumers to perform tasks they cannot do using existing
77
products, or they introduce new ways of doing tasks. Secondly, consumers are more
uncertain of the benefits of RNP consumption compared to INP consumption.
Thirdly, consumers are more uncertain about the cost benefit trade-offs in utility
functions for RNPs, and finally, consumers need to change their behaviour more in
order to achieve the potential benefits of RNPs (e.g. Gourville, 2006).
Cognitive learning theories support the use of various stimuli such as pictures and
texts in order to facilitate consumer comprehension of RNPs (e.g. Gregan-Paxton et
al., 2002; Feiereisen et al., 2008). Consequently, when considering RNP presentation,
using multi-sensory information is likely to be advantageous. As more learning is
required to understand RNPs, using attributes such as telepresence (multi-sensory
information) and anthropomorphic attributes (social cues) in websites can facilitate
the process of learning. Learning theories within the cognitive framework support the
employment of telepresence (vividness and interactivity) and anthropomorphic
attributes as means of improving consumer comprehension. In order to discuss
theories supporting the employment of telepresence, they need to be examined within
its components of vividness and interactivity. But first, Virtual Reality (VR), which is
the source of telepresence, is introduced.
2.2.2. Virtual Reality
VR has been defined as “a real or simulated environment in which a perceiver
experiences telepresence” (Steuer, 1992). Telepresence itself is described as a sense
of presence in a mediated environment, which approaches the direct product
experience, where direct experience is the optimal method for a consumer to learn
about the product (Klein, 2003). There are two media characteristics which enable
telepresence in a computer-mediated environment (CME): interactivity (user control)
and vividness (media richness). Learning theories support the employment of
vividness and interactivity in CMEs as means of improving consumer comprehension
(e.g. Paivio, 1971; Alesandrini, 1982; Rayport and Jaworski, 2001; Zhang et al.,
2006). Furthermore, recent studies in online marketing and consumer behaviour
confirm the role of interactivity as a means of enhancing product attitudes and buyer
behaviour (e.g. Schlosser et al., 2003; Fiore et al., 2005; Park et al., 2005; Pantano
and Naccarato, 2010). Consequently increasing telepresence will produce a more
78
enjoyable shopping experience and might increase purchase intention (Park et al.,
2005).
2.2.2.1 Vividness
Vividness (media richness) is "the representational richness of a mediated
environment as defined by its formal features; that is, the way in which an
environment presents information to the senses" (Steuer, 1992, p.81). Vividness
represents two characteristics of communication medium: sensory depth and sensory
breadth (Steuer, 1992). Sensory depth refers to the quality and resolution of
information transmitted to the senses (e.g. monitor resolution); whereas sensory
breadth is the sum of sensory avenues that a medium utilizes (e.g. aural, visual).
Multimedia communication therefore has greater breadth than a single media
communication (e.g. television versus radio). In the case of greater breadth,
immersion occurs. Immersion refers to “the feeling of being deeply engaged in a
virtual world as if it were the real one” (Bhatt, 2004, p.5).
Visual information is an important sensory input, which can result in vividness.
The use of pictorials has been found effective in various learning theories. Mental
simulation under the cognitive framework (Taylor and Schneider, 1989), which is
closely related to imagery information processing, supports the use of vivid images to
improve consumers’ online experience (e.g. Rayport and Jaworski, 2001).
Furthermore, research in cognitive psychology suggests that pictures are remembered
better than words (Paivio, 1971; Alesandrini, 1982). Cognitive learning models
propose the richer the media, the easier it becomes for learners who prefer an
interactive learning style, to meet their individual needs. Moreover, employing
animation to enhance the richness of experience has been proven to improve attention
attraction (Rothschild, 1987; Zeff and Aronson, 1997; Coyle and Thorson, 2001).
Vividness therefore is proposed to be influential in consumers’ understanding of
products.
The Cognitive Theory of Multimedia Learning is particularly relevant in this
context (Mayer, 1997). It draws on Dual Coding Theory (Paivio, 1986), Baddeley’s
(1992) Working Memory Model, Cognitive Load Theory (Sweller, 1990) and
79
Wittrock’s (1989) Generative Theory and SOI model of meaningful learning (Mayer
et al., 1996). According to Mayer (2005), there is a ‘multimedia principle’ stating,
“people learn more deeply from words and pictures than from words alone” (Mayer,
2005, p47). The Theory of Multimedia Learning is based on this principle and has
three main assumptions (Mayer, 2005):
1 There are two separate channels for processing information, namely auditory and
visual.
2 Each channel has a limited capacity (as in Sweller’s (1988) Cognitive Load
Theory
3 Learning is an active process that contains filtering, selecting, organizing and
integrating information based on preceding knowledge.
Mayer and Moreno (1998, p.2-4) further conclude, “(i) it is better to present an
explanation in words and pictures than solely in words; … (iii) when giving
multimedia explanation, present words as auditory narration rather than visual on-
screen text; (iv) the foregoing principles are more important for low-knowledge than
high-knowledge learners, … (v) when giving multimedia explanation, use few rather
than many extraneous words and pictures”. The Cognitive Theory of Multimedia
Learning is further advanced by Moreno and Mayer (2007) under the name of
Interactive Multimodal Learning Environment. With the Interactive Multimodal
Learning Environment Theory, the interactive learning environment uses two
different presentation modes, verbal and non-verbal, to represent the information. It
also distinguishes between two sensory modalities: auditory and visual (Penney,
1989).
In line with the Cognitive Theory of Multimedia Learning and Interactive
Multimodal Learning Environment (Mayer and Moreno, 1998; Moreno and Mayer,
2007), this study uses a mixed-modality presentation as it facilitates the expansion of
effective working memory capacity. Furthermore, as consumers would have limited
or no knowledge about the RNPs presented, using multi-sensory information is more
beneficial for this group of people. However, the information should be designed in a
way to prevent the consumer being overload with unnecessary information. It is
therefore proposed that:
80
H2a: A website with a high level of vividness will have higher levels of consumer
comprehension of RNPs than a website with a low level of vividness.
By adding vividness, the mediated experience will have a greater similarity to the
direct product experience, therefore this learning is likely to lead to more positive
attitudes (Klein, 2003). Moreover, Street et al., (1997) suggested that adding
multimedia results in having a positive attitude towards a system. Multimedia
presence is believed to facilitate greater engagement and involvement amongst users
(Tran, 2010). The use of multimodality has been proven to create a more positive
attitude towards websites in extra-text enhancement (Tran, 2010). It is therefore
proposed:
H2b: A website with a high level of vividness will have more positive consumer
attitudes towards RNPs than a website with a low level of vividness.
Comprehension is a precondition of the formation of attitude, intention and
behaviour (e.g. Ratneshwar and Chaiken, 1991), with the most important factor in
predicting purchase intention being attitude (Zakersalehi and Zakersalehi, 2012);
therefore:
H2c: A website with a high level of vividness will have higher consumer purchase
intention towards RNPs than a website with a low level of vividness.
2.2.2.2 Interactivity
Interactivity is defined as “the extent to which users can participate in modifying
the form and content of a mediated environment in real time” (Steuer, 1992), p.84).
Interactivity manifests as an effective attribute, improving consumer learning and has
been supported in different learning frameworks (e.g. Brandt, 1997; Zhang et al.,
2006). Within the behaviourist paradigm, interactivity is encouraged between
individuals and their environment (e.g. Social Learning Theory; Bandura, 1977).
Constructivist learning theories similarly argue that interactive activities motivate the
learner more effectively than activities where the learner is passive, as a self-directed
interactive learning process improves learning outcomes (Zhang et al., 2006).
81
However, this study is concerned with the cognitive paradigm. In cognitive learning
theories, it is assumed that the learner’s attention is limited and selective. If the
information is presented in an interactive and rich format, a learner who prefers an
interactive learning style would meet their needs without information overloading
(Zhang et al., 2006). The Theory of Multimedia Learning (Mayer, 1997) elucidates
how individuals learn in a multimedia environment. Applied in the context of
multimedia learning, interactivity is defined as “reciprocal activity between a learner
and a multimedia learning system, in which the [re]action of the learner is dependent
upon the [re]- action of the system and vice versa” (Domagk et al., 2010, p.1025).
This theory recognizes that a multimedia environment has the potential to engage the
learner, thereby creating an interactive learning environment. Hence, the Interactive
Multimodal Learning Environment (Moreno and Mayer, 2007), which is based on the
Cognitive Theory of Multimedia Learning (Mayer, 1997) further supports the
employment of interactivity in facilitating learning. Also, interactivity results in the
closeness of the experience to direct product experience, therefore it is influential
upon consumer online behaviour. Hence it is predicted:
H3a: A website with a high level of interactivity will have higher levels of
consumer comprehension of RNPs than a website with a low level of
interactivity.
H3b: A website with a high level of interactivity will have more positive consumer
attitude towards RNPs than a website with a low level of interactivity.
H3c: A website with a high level of interactivity will have higher consumer
purchase intention towards RNPs than a website with a low level of
interactivity.
2.2.3 Anthropomorphism
According to Social Cognition Theory, the classification of objects as human,
animal or object, and the recognition of anthropomorphic characteristics, are basic
cognitive functions (Kunda, 1999). Anthropomorphism is defined as the extent to
which a virtual object behaves or looks like a human (Kunda, 1999; Nowak and
82
Biocca, 2003; Nowak, 2004). Anthropomorphic attributes have been employed in
computers as a way to improve human-computer interaction and they refer to
“technological efforts of imbuing computers with human characteristics and
capabilities” (Gong, 2008, p.1495). Anthropomorphic attributes can include concepts
such as artificial Intelligence (McCarthy, 1955) and anthropomorphized interface
representations like faces (Sproull et al., 1996; Gong and Nass, 2007) and voices
(Gong and Lai, 2003; Nass and Brave, 2005). Nass et al. (1997) for instance realized
computers with voice output provoked gender stereotypes (Nass et al., 1997).
Sociology maintains that learning occurs through social interaction. In an online
setting, this social interaction occurs when consumers interact with other social actors
or participate in any form of interactive social environment. Studies have
demonstrated that people react to computers as social actors (e.g Reeves and Nass,
1996) as according to Social Response Theory, people automatically tend to treat
computer technologies as social actors. Furthermore, ‘‘the more computers present
characteristics that are associated with humans, the more likely they are to elicit social
behaviour’’ (Nass and Moon, 2000, p.97). One particularly relevant theory,
supporting the concept of the present study, is Social Learning Theory (Bandura,
1977). Social Learning Theory explains “human behaviour in terms of a continuous
reciprocal interaction between cognitive, behavioural and environmental
determinants” (Tu, 2000, p.30). This theory emphasizes behaviour, which is a result
of social interaction of people and their environment (Walther, 1992). Social
interaction between learner and role models are required and this interaction is
affected by social presence. Behaviour, personal factors and an ideal social
environment can promote learning, providing an appropriate degree of social presence
(Tu, 2000).
According to the Social Learning Theory, having an appropriate degree of social
presence promotes user learning; therefore employing anthropomorphic attributes,
which result in an increase in social cues and consequently social presence, can have a
positive effect on consumer learning. Furthermore, Interactive Multimodal Learning
Environments (Moreno and Mayer, 2007) support the interaction between students
(information receiver) and a pedagogical agent. Principle within this theory explains
how guided activity enables the interaction between students and the agent “who
83
guides their cognitive processing during learning” (Moreno and Mayer, 2007). All
these theories support the notion that employing anthropomorphic attributes in an
interface will improve consumer learning of RNPs. As such, it is hypothesised that:
H4a: A website with a high level of anthropomorphic attributes will have higher
levels of consumer comprehension of RNPs than a website with a low level of
anthropomorphic attributes.
H4b: A website with a high level of anthropomorphic attributes will have more
positive consumer attitude towards RNPs than a website with a low level of
anthropomorphic attributes.
H4c: A website with a high level of anthropomorphic attributes will have higher
consumer purchase intention towards RNPs than a website with a low level of
anthropomorphic attributes.
2.2.4 Consumer differences that impact on learning
RNPs are high-involvement products (Feiereisen, 2009) hence consumers need to
understand product related information and benefits, in order to develop a positive
attitude and behavioural intent towards RNPs. This indicates the importance of
considering consumers as an element that can influence RNP comprehension. The
Elaboration Likelihood Model posits that involved consumers put more cognitive
effort into central processing of product information (Petty and Cacioppo, 1981; Petty
and Cacioppo, 1986), and will develop positive attitudes and behavioural intentions,
towards RNPs. Involved consumers focus their attention on message relevant
information and draw on prior experience and knowledge. Nevertheless, due to the
nature of RNPs, consumers have no or limited knowledge about RNPs.
According to the ‘individual differences principle’ in the Theory of Multimedia
Learning, principles such as “multimedia effects” and “contiguity effects” are more
applicable for individuals who lack prior knowledge, compared to the individuals with
a high level of prior knowledge. Furthermore, in the Cognitive-affective Theory of
Learning with media (Moreno, 2005), it is emphasized that “differences in learners’
prior knowledge and abilities may affect how much is learned with specific media
(Kalyuga et al., 2003; Moreno, 2004; Moreno and Durán, 2004)” (Moreno and
84
Mayer, 2007, p.313). Personal factors within the Social Learning Theory are
mentioned as important elements that promote learning via a computer-mediated
communication, providing there is an appropriate degree of social presence (Tu,
2000). Furthermore, CME users should know how to use online communication to
increase sensory simulation (Tu, 2000). Consumers who are (i) motivated to seek
information regarding new products and follow the central route, (ii) familiar with
innovation and new technologies, and (iii) have a good knowledge of using online
technologies, appear to comprehend RNPs better than other types of consumers. This
type of consumer is comparable with the innovators’ profile in Diffusion Theory
(Rogers, 1995).
Diffusion Theory (Rogers, 1995) introduces five adopter types based on how long
it takes for consumers to try a new product: innovators, early adopters, early majority,
late majority, and laggards. Innovators have the quickest time-of-adoption and
laggards the slowest. Innovator consumers are more likely to try new products. They
believe they are more knowledgeable about online shopping and they purchase more
products online. Additionally, they are more likely to view online shopping as
quicker, cheaper, safer and more fun than traditional shopping (Goldsmith and
Lafferty, 2002). As it appears that consumer innovativeness has an impact upon
consumer comprehension and consequently attitude and intention, this study controls
consumer innovativeness, considering it as a covariate.
2.3 Methodology
Three experiments were used to test the effectiveness of alternative product
presentation techniques in an online setting. The main advantages of experimental
research design are its high internal validity (Lee and Lings, 2008), and its ability to
test a hypothesis in a convincing way; though an important disadvantage is the
complexity of the methodology. In this study, the experiments were designed to
mimic commercial scenarios in a realistic environment, which helps towards the
external validity of the findings. Online (web) experiments are becoming a
mainstream method in the field of psychology (Birnbaum, 2004). The most important
advantages of web experiments, compared to laboratory experiments, have been
85
identified as follows. Firstly, by running a web experiment, a large sample can be
employed in a short period of time which makes “statistical tests powerful and model
fitting very clean” (Birnbaum, 2004, p.813). Furthermore, using web experiments
increases the generalizability of the study and also provides the opportunity to recruit
specialized types of participants.
2.3.1 Research Design
Three online experiments were used to test consumer comprehension of, attitudes
towards and purchase intentions related to RNPs in different online presentations.
These presentations varied by level of vividness, interactivity and anthropomorphic
attributes. However, before the online presentational manipulations, suitable RNPs
were selected to help increase the realism of the experiment.
Product selection
Stage 1
In order to identify suitable RNPs, relevant information on RNPs was gathered
from different websites and forums (e.g. TechCrunch.com, mashable.com, extracted
from technorati.com’s ‘Top 100 most read blogs’). Seventeen products were selected
according to the amount of information available, the pictures available and the level
of product innovativeness (Appendix 1). An initial screening of these products
resulted in the elimination of eight products. Three had become available in the
marketplace and five were assessed as less innovative than the other products, or were
considered difficult to understand.
Stage 2
The remaining nine products were tested using an online questionnaire to confirm
their status as RNPs. For this purpose, perceived product newness was measured
(Appendix 2). The scale was a combination of Gregan-Paxton et al.’s definition of
RNPs (Gregan-Paxton et al., 2002) and Hoeffler’s framework that is used by recent
86
academics in RNP related research (e.g. Alexander et al., 2008) respectively.
Alexander et al. (2008) modified existing measures to assess the unique
characteristics of RNPs, such as uncertainity in estimating the benefits of RNPs.
There are four questions developed from Alexander et al.’s study, where two of the
questions discuss if participants understand and like the product (group 1), and the
other two imply a participant’s need to change their behaviour to do new things
(group 2).
For group 1, the result of the Analysis of Variance (ANOVA) indicated a
significant difference amongst products means, with products E-tomb, Dismount
Washer and Bio Robot Fridge coming in as significantly more understandable and
likable. For group 2, the result also showed a significance difference between means
across the products with iDropper, Washing Machine in Wardrobe and Digital Make
up Mirror, as products participants were willing to change their behaviour for, in
order to do something with them. Looking at the selected products in each group, to
minimize the differences between the two final product groups, Dismount Washer
(referred to as product 1) and Washing Machine in a Wardrobe (referred to as product
2), which address similar consumer needs, were selected (see further details in
Appendix 3).
Main Experiments
Experiment 1- vividness
Experiment 1 was concerned with the impact of vividness and followed a single
factor between-subject design for two separate products. Steuer (1992) divided
vividness into sensory breadth and sensory depth. In many studies, vividness is
deployed via sensory breadth, as the manipulation of sensory depth was not feasible
(e.g. Coyle and Thorson, 2001; Sukoco and Wu, 2011). Sensory breadth was
manipulated in this study, and operationalized by inserting audio and video into the
website for the high condition (Appendix 4). Participants were first asked to explore
the website and learn about the product as much as possible. They were then directed
to the online survey and asked to fill in the demographics measures, followed by the
87
questions in regards to other variables. 400 participants were recruited and paid to
complete the survey, resulting in 100 participants for each cell of either condition low
or high for either product 1 or product 2. Participation in the survey was limited to
adults living in the United States.
Experiment 2- interactivity
This experiment tested the impact of interactivity and followed a single factor
between-subject design for two separate products. Schlosser et al. (2003) and Fiore et
al. (2005) included interactivity in websites in the form of image/object interactivity.
They explained that by adding image interactivity to websites, approach response (e.g.
willingness to purchase) and purchase intention will be positively influenced (Fiore et
al., 2005) and visual sensory information will be improved. Moreover, Park et al.
(2005) examined the effect of product presentation (movement and image size) on
consumer shopping experience and purchase intention on an apparel website. They
concluded that purchase intention may increase, where consumers experience pleasure
from product movement, resulting in a reduced perceived risk (Park et al., 2005).
Interactivity was therefore operationalized in the form of objective interactivity (3D),
which allows movement and re-sizing. In the high-interactivity website, consumers
were given the option to click on the 3D product picture and rotate, zoom in/out of the
product freely. The low-interactivity website only allowed the user to see specific
views of the product with no zoom or pace control (Appendix 5). 400 participants
were recruited and paid to complete the survey, resulting in 100 participants for each
cell of either condition low or high, for either product 1 or product 2. Participation in
the survey was limited to adults living in the United States.
Experiment 3- anthropomorphic attributes
This experiment tested the result of adding anthropomorphic attributes to the
online setting and followed a single factor between-subject design for two separate
products. An example of the employment of anthropomorphism (social cues) in an
88
online setting is the embodied agent (avatar), which according to Social Response
Theory, results in better social interaction. Therefore, anthropomorphism was
operationalized by adding an avatar to the base website for the high condition. The
avatar was a 3D interactive virtual agent, which enabled users to ask their questions
using a question/comment box provided. The avatar had limited facial expressions,
voice and Artificial Intelligence (Appendix 6). The full description of the avatar can
be found in Appendix 7. 400 participants were recruited and paid to complete the
survey, again resulting in 100 participants for each cell of either condition low or
high, for either product 1 or product 2. Participation in the survey was limited to
adults living in the United States.
2.3.2 Measurement
Measurement for the manipulation checks, dependent variables and covariates are
explained below with scale items being detailed in Appendix 8. In addition,
demographic information was also collected. All the Likert scales were converted to a
7-point scale for the purpose of uniformity.
2.3.2.1 Manipulation Checks
Vividness
Many vividness scales focus on visual sensory information (e.g Fortin and
Dholakia, 2005; Lee, 2012; Shen and Khalifa, 2012). However, as in this study aural
sensory information is also used to increase vividness, the manipulation check needs
to be able to capture the impact of aural as well as visual information. Six items based
on Steuer’s (1992) dimensions of sensory breadth and depth, constructed by Fortin
and Dholkia (2005) were employed in this study. Fortin and Dholakia (2005) reported
a Cronbach alpha of .90 for this scale. In addition, the statement regarding aural
information from the scale developed by Sukoco and Wu (2011) was added in order
to study the effect of the aural information.
Interactivity
89
To check the manipulation for interactivity, the perceived interactivity scale from
Kim and Sundar (2012) was adopted. These five items were measured by a 10 point
Likert scale. The reliability of the scale was tested and described by Kim and Sundar
(2012) as α = .83, M = 24.92, SD = 7.20.
Anthropomorphism
In measuring anthropomorphism, both mindful and mindless anthropomorphism
were considered. The mindful anthropomorphic scale consists of three statements
measured by a 10-point scale (α = .86, M = 15.53, SD = 5.94), whereas the mindless
anthropomorphism scale comprises four items measured on a 10-point semantic
differential scale (α = 0.85, M = 24.90, SD = 7.25) (Kim and Sundar, 2012).
2.3.2.2 Dependent Variables
Product Comprehension
Feiereisen et al. (2008) combined a 2-item semantic differential scale by (Phillips,
2000) with a 4-item 7-point Likert scale by Moreau et al., (2001) to give the 6-item
measure of product comprehension was employed in this study. Phillips (2000)
reported a Cronbach alpha of 0.89, with the Cronbach alpha for Moreau’s original
scale being 0.80. The KMO and Bartlett’s test both support the combination of the
scale items as appropriate (Feiereisen et al., 2008).
Attitude
A ten-item consumer attitude measurement scale developed by Voss et al. (2003)
was applied in this study to measure attitude towards RNPs. This 7-point semantic
differential scale demonstrated “solid performance in several psychometric tests and
in multiple test of criterion and discriminant validity” (Voss et al., 2003), p.318). The
reliability of the scale was tested via Fornell and Larcker’s (1981) index of construct
reliability. The index indicates that reliability is acceptable when higher than 0.70
(Fornell and Larcker, 1981), which for this scale, was above 0.90 within different
categories.
90
Purchase Intention
In order to measure consumer purchase intention, a 4-item 7-point Likert scale
(α=0.86), adopted from (Moon et al.;s (2008) was employed. The scale is a
modification of scales by Dodds et al. (1991) and Sweeney et al. (1999)
2.3.2.3 Covariate - Consumer Innovativeness
Motivated Consumer Innovativeness (MCI) is a 20-item 7-point Likert scale,
which takes into account four different motivations of innovative consumers (social,
functional, hedonic, and cognitive). The dimensionality, reliability, and convergent,
discriminant, and predictive validity of the MCI scale is reported to be satisfactory
(Vandecasteele and Geuens, 2010), with the Cronbach alphas of the four dimensions
all above 0.9 (alphasMCI=.929; alphafMCI=.907; alphahMCI=.928; alphacMCI=.902).
2.3.2.4 Participants
The participants for all stages of the research were drawn from an online crowd
sourcing website commonly used in consumer/psychological experimental research
namely Amazon Mechanical Turk3(Mturk). Mturk has proven to be a great
opportunity for data gathering and Mturk participants generate high quality reliable
results, consistent with standard decision-making biases (Goodman et al., 2013).
However, it is recommended to avoid questions with factual answers and consider
individuals’ differences such as financial and social conditions (Goodman et al.,
2013). In this study, information regarding participant income and education is
gathered, as they can be indicators of an individual’s financial and social conditions.
Furthermore, the questionnaire did not include any questions with a factual answer. A
pool of participants was randomly assigned to each experimental treatment.
Participant demographic data of gender, age, education and monthly income were
gathered for all experiments.
3 www.Mturk.com
91
Table 2 includes the demographic information of the recruited sample. More
information is available in Appendix 9.
Table 2: Sample Demographics
Factor PercentageGender Male: 50.7%
Female: 49.2%Age 20-29: 40.2%
30-29: 31.2%40-49: 15.6%Over 50: 12.8%
Monthly Income less than $999 pm: 22.4%$1000-$1999 pm: 24.6%$2000-$2999 pm: 21.7%$3000-$3999 pm: 24.6%More than $4000 pm: 16.6%
Education Primary school: 0.8%Secondary/high school: 27%Undergraduate college/university: 51.7%Graduate college/university: 18.9%PhD: 1.6%
2.4 Results
Measures
To test the hypotheses, constructs were measured using existing multi item scales
to evaluate respondents’ replies to the stimuli. Although the scales were all
established scales extracted from related studies, some scales were slightly altered to
suit the context of the study. The measurement model was tested using Confirmatory
Factor Analysis (CFA). The CFA model was designed according to the stages defined
by Hair et al. (2010), with individual constructs designed and coded (see Appendix 10
for further details). For the purpose of CFA, a sample of 400 participants was
employed. This was a sub-set of experimental participants that were required to
complete all the questionnaire items. Figure 3 is the structural model of the study.
Before CFA, KMO and Bartlett’s test for sphericity were performed to determine
that the sample was satisfactory for factor analysis. The result showed KMO=0.910
which is satisfactory (more than 0.5); Bartlett's test of sphericity (tests whether the
92
correlation matrix is an identity matrix) was 13155.128 with P value of 0.000 and df
=1378. It rejects the null hypothesis that the variables in the population correlation
matrix are uncorrelated; therefore factor analysis can be performed.
The initial model fit was not acceptable (χ2=(1238)=3465.137 p=.000; CFI= .822;
GFI=.649; AGFI= .649; RMSEA=.078; Pclose=.000). The model therefore was
examined and items with factor loadings less than 0.5 and with the biggest
modification indices were removed, as in this study, the error terms were not
correlated as a solution to improve model fit. More detail is available in Appendix 11.
According to Hair et al. (2010, p.646), usually three to four fit indices provide
adequate evidence of model fit. However a researcher should consider at least one
incremental index (NFI, CFI etc.) and one absolute index (GFI, RMSEA, SRMR etc).
Therefore by looking at CMIN/DF, AGFI, CFI and RMSEA, the final model had a
good fit (χ2(630)=1174.86 p=.000 ; CFI=.936; GFI= .82; AGFI=.80; RMSEA= .05;
Pclose=.095). Furthermore, all items loaded significantly on their respective
constructs, which supports the convergent validity of the measurement items. All final
scales were tested for Unidimensionality and Homogeneity and were all satisfactory.
More detail can be found in Appendix 12.
The internal consistency estimates and reliability of the scales were tested and
deemed satisfactory. Consequently Composite Reliability (CR), Average Variance
Extracted (AVE), Maximum Shared Variance (MSV), Average Shared Variance
(ASV) and Cronbach alpha were calculated and evaluated (Fornell and Larcker,
1981); Hair et al., 2010). Table 3 shows that all the conditions necessary to
demonstrate reliability, convergent and discriminant validity were met.
Table 3: Reliability, Discriminant/convergence validity for final model.
CR (>0.7) AVE (>0.5)(<CR)
MSV(<AVE
)ASV α
Comprehension 0.921 0.702 0.186 0.093 0.75Vividness 0.853 0.595 0.240 0.093 0.87Anthropomorphism 0.901 0.606 0.032 0.008 0.90Social Innovativeness 0.909 0.715 0.484 0.175 0.91
93
Purchase Intention 0.939 0.836 0.412 0.183 0.90Interactivity 0.883 0.656 0.097 0.060 0.93Functional Innovativeness 0.826 0.542 0.392 0.181 0.93
Hedonic Innovativeness 0.901 0.648 0.484 0.215 0.75Cognitive Innovativeness 0.925 0.755 0.271 0.134 0.87Attitude 0.919 0.696 0.412 0.184 0.91
Descriptive Analysis of Individual Scales
This section explains the final scales after confirmatory factor analysis. The scales
have also been examined in order to determine whether the measures were appropriate
for further use in hypothesis testing. Bartlett’s test for sphericity and KMO were
performed for individual scales for the purpose of Homogeneity, which all appeared
to be satisfactory (see Appendix 13).
Looking at the frequency distribution of individual scales, all of them appeared to
be close to normality (more information in Appendix 14). The results of the
Kolmogorov-Smirnov (KS) and Shapiro-Wilk (SW) tests were considered for each
scale. The Shapiro-Wilk test appeared to be the most powerful test for all types of
distribution and sample sizes (especially for sample sizes larger than 30), in
comparison with other tests such as Kolmogorov-Smirnov and Lilliefors (Razali and
Wah, 2011). However, for samples larger than 200, in order to make sure the sample
is normal, rather than looking only at SW or KS tests, the kurtosis and skewness
needs to be examined (Field, 2005, p.72). This is due to the KS test being extremely
sensitive to minor departure from normality (Sharma, 1996) and also violating the
assumption of normality is quite common in larger samples (Pallant, 2005, p.57).
Moreover, it has been argued that skewness and kurtosis below the threshold of 1.96
will not cause significant issues in the dataset (Field, 2005, p.72 cited in Feiereisen,
2009 p.226-227). There were no serious concerns regarding the normality of any of
the variables and they were all preserved without transformation for future analysis.
Detailed information about each variable’s tests of normality and histograms can be
found in Appendix 14.
Study 1: Experimental Manipulation of Vividness
94
Manipulation checks
This experiment followed a single factor between-subject design for two separate
products. Treatments were tested using independent sample t-tests to examine the
manipulations in various treatments. An independent sample t-test for product 1,
revealed a higher level of vividness was perceived by participants in high condition
(Mean=4.07, SD= 1.02) than in low condition (Mean=3.93, SD=1.07). However the
difference was not significant t(198)= -0.927, p>0.05 (p=0.355), indicating that
participants in high condition did not perceive a significantly higher vividness level
than participants in low condition. Levene’s test for equality of variances was non-
significant therefore the assumption of homoscedasticity was met for this group. An
independent sample t-test for product 2 under vividness treatment, revealed a higher
level of vividness perceived by participants in high condition (Mean=4.26, SD= 1.08)
than in low condition (Mean=3.76, SD=1.14). The difference was significant t(198)= -
3.1, p<0.05. Levene’s test for equality of variances was non-significant, therefore the
assumption of homoscedasticity was met for this group.
Outcomes
A MANCOVA, with consumer’s comprehension, purchase intention and attitude
as the dependent variables, and the level of manipulated vividness as the independent
variable was performed for product 2 solely, as manipulation for product 1 did not
show a significant difference. For product 2, the assumption of homoscedasticity was
upheld (Box’s M=10.03, f=1.64, p=0.130), however there was no significant main
effect observed for vividness (F(3, 196) = 0.602, p=0.615). Consequently H2a, H2b
and H2c are not supported.
Study 2: Experimental Manipulation of Interactivity
Manipulation checks
This experiment followed a single factor between-subject design for two separate
products. Treatments were tested using independent sample t-tests to examine the
manipulations in various treatments. An independent sample t-test for product 1,
under Interactivity treatment revealed a higher level of interactivity perceived by
participants in high condition (Mean=3.86, SD= 1.33) than in low condition
(Mean=3.07, SD=1.63); the difference was significant t(198)= -3.75, p<0.05.
95
However, Levene’s test for equality of variances was significant, hence the
assumption of homoscedasticity was not met for this group. The t-test result was
therefore changed to “equal variance not assumed” which is still significant t(190)= -
3.75, p<0.05. The fact that interactivity for product 1 showed heteroscedasticity can
be ignored as according to Hair et al. (2010) in relatively large sample sizes (i.e. 200),
when the rest of the groups show homoscedasticity, corrective remedies are not
needed for the heteroscedastic group. An independent sample t-test for product 2,
under Interactivity treatment revealed a higher level of interactivity perceived by
participants in high condition (Mean=3.83, SD= 1.4) than in low condition
(Mean=3.19, SD=1.4). The difference was significant t(198)= -3.23, p<0.05. Levene’s
test for equality of variances was non-significant, therefore the assumption of
homoscedasticity was met for this group.
Outcomes
A MANCOVA, with consumer’s comprehension, purchase intention and attitude
as the dependent variables, and the level of manipulated interactivity as the
independent variable was performed for both products 1 and 2. Looking into
Interactivity for product 2, the assumption of homoscedasticity was upheld as Box’s
M= 7.42, f=1.21, p=0.294. There was no significant main effect observed for
Interactivity for this product, F(3, 196)=1.534, p=0.207. For product 1, the
assumption of equality of covariance matrices was satisfied, as Box’s M=2.71,
F=0.445, P=0.849. Furthermore, a significant main effect was observed F(3, 196)=
2.71, p<0.05. The multivariate effect size was estimated at .04, which implies that 4%
of the variance in the canonically derived dependent variable was accounted for by
the interactivity level. Interactivity therefore has a weak effect on consumer
comprehension, purchase intention and attitude collectively towards product 1. As a
result H3a, H3b and H3c are partially supported, for product 1.
Study 3: Experimental Manipulation of Anthropomorphism
Manipulation checks
This experiment followed a single factor between-subject design for two separate
products. Treatments were tested using independent sample t-tests to examine the
manipulations in various treatments. An independent sample t-test for product 1,
96
under Anthropomorphism treatment, revealed a higher level of anthropomorphism
perceived by participants in high condition (Mean=4.94, SD= 2.13) than in low
condition (Mean=4.41, SD=1.71); the difference was not significant t(98)= 1.38,
p>0.05. However, Levene’s test for equality of variances was non-significant,
therefore the assumption of homoscedasticity was met for this group. An independent
sample t-test for product 2 under Anthropomorphism treatment, revealed a higher
level of anthropomorphism perceived by participants in high condition (Mean=4.88,
SD= 1.94) than in low condition (Mean=4.75, SD=1.87); the difference was not
significant t(98)= -0.356, p>0.05). Levene’s test for equality of variances was non-
significant, therefore the assumption of homoscedasticity was met for this group.
Consequently, MANCOVA was not performed for the anthropomorphic conditions,
as the manipulation did not work.
Covariate
MANCOVA was performed for both vividness and interactivity. Consumer
innovativeness was divided into four sub-scales of Social Innovativeness (Insoc),
Functional Innovativeness (Infun), Hedonic Innovativeness (Inhed) and Cognitive
Innovativeness (Incog). Covariates were added one by one to the model. Tables 4 and
5 summarize the results for vividness and interactivity respectively.
Table 4: The effect of covariate in vividness conditions
No covariate Social Innovativeness
Functional Innovativeness
Hedonic Innovativeness
Cognitive Innovativeness
Vividness product 1 .114 .145/.000 .214/.000 .176/.000 .176/.000Power .511 .465/.973 .395/.997 .432/1 .432/.994Effect size .030 .027/.093 .023/.129 .025/.153 .025/.116Vividness product 2 .615 .5071/.0002 .859/.000 .377/.000 .601/.007Power .174 .2161/12 .098/1 .279/1 .179/.850Effect size .009 .0121/.1932 .004/.275 .016/.184 .009/.060
1. Main effect; 2. Covariate
97
Table 5: The effect of covariate in interactivity conditions
No covariate Social Innovativeness
Functional Innovativeness
Hedonic Innovativeness
Cognitive Innovativeness
Interactivity product 1 .047 .044/.000 .070/.000 .033/.000 .055/.000Power .651 .659/1 .592/.989 .695/1 .628/1Effect size .040 .041/.186 .036/.109 .044/.240 .038/.205Interactivity product 2 .215 .232/.002 .195/.000 .200/.000 .175/.000Power .401 .378/.909 .412/1 .408/.996 .433/.997Effect size .023 .022/.071 .024/.170 .023/.123 .025/.128
1. Main effect; 2. Covariate
Results above indicate that, although the types of innovativeness were
significantly different across the conditions, the effect of vividness and interactivity
was the same when the analysis was done with or without the covariate. Differences
in consumer comprehension, attitude and purchase intention cannot therefore be
attributed to consumer innovativeness. However, an exception is evident for social
innovativeness and hedonic innovativeness when controlled for within the
interactivity conditions for product 1. By controlling for these covariates, the main
effect improves.
Correlation between dependent variables
The correlation between the dependent variables of comprehension, attitude and
purchase intention, were checked. Correlation showed a moderate (Dancey and Reidy,
2004) positive correlation between purchase intention and comprehension (r=0.40,
N=800, p=.000), and comprehension and attitude (r=0.40, N=800, p=.000). There was
a relatively strong positive correlation between purchase intention and attitude
(r=0.58, N=800, p=.000). The correlation was statistically significant between all
three variables.
2.5 General Discussion and Conclusion
The study provides an insight into the role of vividness and interactivity (as two
characteristics enabling telepresence), and anthropomorphic attributes in consumer
learning about RNPs. Three sets of online experiments were designed to examine
vividness, interactivity and anthropomorphic attributes. Study 1 examined the role of
98
vividness on consumer online behaviour towards two RNPs. This study shows that
vividness is perceived significantly higher in high condition for product 2, but the
difference was not significant for product 1. Examining the effect of vividness on
consumer comprehension, purchase intention and attitude collectively, it is evident
that vividness has no significant effect on the dependent variables collectively or
individually for either of the products. In summary, vividness has no effect on
consumer comprehension, attitude and purchase intention for the RNPs used in this
study. This finding is in contradiction to previous academic literature. The Cognitive
Theory of Multimedia Learning (Mayer, 1997) supports the presentation of
information in a multimedia format as an attribute influencing comprehension.
Vividness is supported as a tool to improve learning and online behaviour within
various contexts (e.g. Paivio, 1971; Alesandrini, 1982; Rayport and Jaworski, 2001;
Zhang et al., 2006). The result of this study adds to the body of knowledge of
telepresence, by indicating how vividness effectiveness is not the same when
individuals are learning about RNPs.
Interactivity, as the second characteristic enabling telepresence, was examined in
Study 2. The effect of interactivity on consumer online behaviour and comprehension
was investigated. The findings indicate that participants perceive a significantly
higher level of interactivity for high conditions for all products. For product 1,
interactivity proved to influence consumer comprehension, purchase intention and
attitude collectively towards the product. However, interactivity did not have any
influence upon consumer comprehension, attitude and intention towards the product
2. This is in contradiction to previous literature, indicating interactivity is an attribute
positively influencing consumer learning and online behaviour (e.g. Schlosser et al.,
2003; Fiore et al., 2005; Park et al., 2005; Pantano and Naccarato, 2010). In
summary, the findings of this study suggest that interactivity can be an influential
attribute on consumer comprehension, attitude and purchase intention within the RNP
context; however this cannot be generalized for all products within the RNP category,
as it differes across RNPs.
Study 3 examined the effect of anthropomorphic attributes on consumer online
behaviour. Although a higher level of anthropomorphism was perceived by
participants in high condition than in low condition, the difference was not significant
99
therefore the manipulation had not worked. This could be due to the fact that
participants in low condition, analysed the website (without avatar) as
anthropomorphic, whereas in high condition (with avatar), they evaluated the avatar
and not the website as anthropomorphic. Another explanation could be the
measurement scales available. Anthropomorphic attributes have been manipulated by
comparing various formats of the same type of social cue, such as different avatars
(McGoldrick et al., 2008; Keeling et al., 2010). There is no study up to date that has
compared various types of social cues, therefore the scales developed within previous
studies might not be suitable when comparing different types of social cues in an
online setting.
The dissimilarity in consumer’s perceived vividness and the influence of
vividness and interactivity on consumer comprehension, attitude and purchase
intention, could be related to various factors such as product nature or consumer
characteristics. In terms of product nature, a possible explanation could be that
product 1 belonged to the group where participants indicated the product to be
significantly more understandable and likable; whereas product 2 belonged to the
group of products which participants were willing to change their behaviour in order
to interact with them (according to product selection process). The nature of the
product therefore could be the reason for this mixed result. A preliminary analysis
showed that product nature has an effect on consumer comprehension, purchase
intention and attitude collectively. There were also indications that product nature is
influential in consumer comprehension and attitude as separate variables, but it does
not have a significant effect on consumer purchase intention.
Examining consumer characteristics, factors such as consumer knowledge and
ability could be influential in consumer perceived vividness, interactivity and
anthropomorphism. According to the Cognitive Theory of Multimedia Learning
(Mayer and Moreno, 1998), presenting information in a multimedia format is more
important for low-knowledge than high-knowledge learners, and for high-spatial
rather than low-spatial learners. Furthermore, in the Cognitive-Affective Theory of
Learning with media (Moreno, 2005), it is emphasized that differences in learners’
prior knowledge and abilities may affect how much is learnt with specific media
(Kalyuga et al., 2003; Moreno, 2004; Moreno and Durán, 2004). Socio-demographic
100
factors could also have an influence on consumer online behaviour. A preliminary
study showed overall that males express a significantly more positive attitude towards
product 2 than females.
The study also looked into the relationship between consumer comprehension,
attitude and purchase intention within the context of RNPs. It was first hypothesized
that comprehension is positive in relation to attitude towards and purchase intention of
RNPs. The results of the study indicate that there is a moderate positive correlation
amongst purchase intention and comprehension, as well as comprehension and
attitude; whereas there is a relatively strong positive correlation between purchase
intention and attitude towards RNPs. The findings confirm the previous literature that
not only does comprehension lead to attitude formation (e.g. Goldsmith and Newell,
1997; Pagani, 2007), but also there is a positive relationship between comprehension,
attitude and purchase intention (Ratneshwar and Chaiken, 1991). The findings add to
the body of knowledge on RNPs, as well as online consumer behaviour, by
elaborating this relationship within the RNP context.
2.6 Theoretical Implications
The study contributes to learning theories within the concept of RNPs. First, it
complements the Cognitive Theory of Multimedia Learning (Mayer, 1997) by
exploring the efficiency of presenting information in both verbal and non-verbal
forms when dealing with RNPs. As vividness did not have any influence upon
consumer learning, attitude and purchase intention, it raises the question of whether
multimodality enhances consumer comprehension when dealing with a very new
concept. In contrast to vividness, interactivity had an impact on consumer
comprehension for one of the products, therefore it can be argued that interactivity
can facilitate a consumer’s comprehension of RNPs, but it depends on the product
nature, as per inconsistency of the results.
The study also sheds light onto the concept of telepresence. Due to the
inconsistent results, it can be argued whether telepresence experience, as a close to
“direct” product experience, facilitates consumer’s comprehension of RNPs. The
101
study also contributes to the product presentation literature, arguing how inserting
various technologies such as audio, pictorial, video and 3D design can improve
consumer’s perceived vividness and perceived interactivity towards RNPs.
2.7 Managerial Implications
The study provides valuable information for managers in promoting RNPs. From
the presentation point of view, the study explains how employing various online
technologies, facilitating telepresence, may improve consumer understanding of
RNPs. According to this study’s findings, the presentation format that can influence
consumer understanding of RNPs is interactivity. Managers can therefore include 3D
design of the RNPs within CME, in order to facilitate consumer learning and improve
their online behaviour towards RNPs. The study also emphasizes the importance of
consumer learning, as it has a positive relationship with consumer attitude and
purchase intention towards RNPs. Managers can therefore utilize this study’s findings
while promoting RNPs, in order to facilitate consumer learning, which then results
into a more positive attitude and purchase intention.
2.8 Limitations and Directions for Future Research
Although this study examined RNPs, the difference in the nature of RNPs were
not controlled or measured. This may be a reason why some findings are inconsistent
across the two RNPs used. Product attribute is an influential factor on consumers’
diffusion of innovation (Rogers, 2003), which includes many factors. These factors
are related to the product’s functional aspect (such as compatibility, relative
advantage, and perceived usefulness) or the product’s hedonic aspect (such as
appearance) (Rogers, 1995; Irani, 2000). This paper’s findings support the criticism
towards the diffusion of innovation literature; that is, the impact of consumer
perception of innovation attributes needs to be considered in related studies (Lowrey,
1991). The consideration of various aspects of RNP attributes were outside the scope
of this study, due to the focus being on presentation formats and consumer behaviour.
However, it is recommended for future research to examine various aspects of RNP
attributes, in order to get a deeper insight into the influence of consumer perception of
102
innovation attributes upon consumer perception of presentation formats and consumer
online behaviour. Further product related limitations could be that the study has
explored only two types of RNPs, so investigating various RNPs within other product
domains will help towards generalizability of this study.
With respects to information presentation, product promotion websites were
designed due to the availability and feasibility of accessible technologies; this needs
to be reconsidered in future research, as technology is advancing rapidly. Sensory
depth, as a presentation element identified by Steuer (1992) was not controlled in this
study. As the products are different in colour (product 1 was lighter in colour than
product 2) and shape (product 1 was smaller than product 2), the presentation element
might have been affected by the RNP's image attributes, and consequently have
affected individual's perception of them. Further exploration of the sensory breadth
element can assist academics in understanding the different results observed for two
RNPs within this study.
From the consumer aspect, one limitation is that US participants were recruited
via Mturk for this study. This was due to the availability and accessibility of the
researcher to a suitable sample pool from the US and the Mturk website. Expanding
the study to participants from other parts of the world can therefore add to the
literature of demographic differences when learning about RNPs. Another factor for
future research is to examine participant’s differences in more detail, from various
aspects. For instance, learning differences or background knowledge of the RNP’s
domain can be investigated. This might help to discover the reason behind the
ineffectiveness of vividness, as employed in this study for the selected products.
103
Chapter 3: Mindful and Mindless Anthropomorphism; how to
facilitate consumer comprehension, its measurement and application
3.1 Introduction
Online advertising and product presentation have been of interest to academics,
due to consumers’ increased attention and preference towards information presented
via Computer Mediated Environment (CME) (e.g. Robinson et al., 2007; Van Noort,
2012). The internet has been used for years as a promotional channel for product
awareness, especially new product awareness (Bickart and Schindler, 2001); this has
resulted in an increase in the new product adoption rate (Prince and Simon, 2009).
CME is therefore a suitable platform for promoting new products and services, in
order to influence consumer awareness and adoption. Academics are looking into
various elements that differentiate online product promotion from traditional media
promotion. For example, interactivity has been deemed to be a key difference between
traditional and new media (Chung and Zhao, 2004). Anthropomorphism is another
attribute that has been proven to influence consumer behaviour within CMEs (e.g.
Keelings et al. 2010), which can also implicate the interactivity characteristics that are
a central theme in the marketing communication literature (Kim and McMillan,
2008). Anthropomorphic attributes have been used in CMEs as a way to improve
human-computer interactions (Gong, 2008). Studies around anthropomorphism
revealed an impact that they have upon consumer online behaviour and learning (e.g.
Holzwarth et al., 2006; Franceschi et al., 2009; Keeling et al., 2010).
Inserting anthropomorphic attributes within a CME for product promotion is
particularly important for presenting new and innovative product information. One
reason is that innovation adoption involves a higher level of uncertainty in
comparison to existing product adoption (Alexander et al., 2008). Really New
Products (RNPs) are a category of products, which are considered as very innovative.
The insertion of anthropomorphic attributes can therefore improve consumer learning
and online behaviour for RNPs. However, the question would be whether consumers
perceive websites presenting anthropomorphic cues as more anthropomorphic. If
individuals perceive anthropomorphism, this will then have an effect on their level of
trust; this will reduce uncertainty and positively affect consumers’ behaviour
104
(Hoeffler, 2003; Castano and Giner-Sorolla, 2006). The higher level of perceived
anthropomorphism can also lead to better learning (e.g. Bandura’s (1977) Social
Learning Theory; ). However, there are contradictory findings suggesting the diverse
influence of application of various forms of anthropomorphic attributes within CMEs.
This indicates the importance of exploring different types of anthropomorphic
attributes, in order to understand the influence of each on consumer perception, and
therefore comprehension and behaviour within the RNP context.
The aim of this study is hence to examine the difference in individuals’
perception, while dealing with different types of anthropomorphic attributes, with the
intention of understanding and adopting RNPs. For this purpose, the study looks into
RNPs to uncover the importance of using anthropomorphic attributes for RNP
promotion and learning. Anthropomorphism is then explored and categorized further.
3.2 Theoretical Background
In this section, the identification of the research gap and previous academic
research/frameworks are explained. The study is approached via looking into various
product categories, anthropomorphism, consumer comprehension and online
behaviour, building up to the formation of hypotheses.
1.1.1 Really New Products
Really New Products (RNPs) allow consumers to experience something they have
not experienced before, as well as performing tasks they are unable to perform using
existing products. Nevertheless, consumers face more uncertainty dealing with RNPs
than with familiar products. RNPs are considered a higher-risk, higher-reward product
category (Hoeffler, 2003). Consequently, individuals need to put more cognitive
effort into understanding RNPs in order to reduce their uncertainty. This indicates
Hoeffler’s (2003) finding that more learning is required to understand RNPs.
Furthermore, consumers need to change their behaviour in order to achieve the
potential benefits of RNPs (Alexander et al., 2008). Consumers do not have any
existing knowledge base when learning about RNPs, therefore comprehension occurs
at the time of product evaluation (Hoeffler, 2003). This indicates the importance of
presenting the RNP related information in a way to facilitate consumers learning, and
reduce uncertainty.
105
The employment of anthropomorphic attributes increases information credibility
(Holzwarth et al., 2006) leading to lower perceived risk and reducing uncertainty for
consumers towards RNPs (Hoeffler, 2003; Castano and Giner-Sorolla, 2006).
Moreover, individuals experience a high level of performance uncertainty and risk
when learning and browsing the information online, which exacerbates the
uncertainty of online RNP comprehension. Anthropomorphic attributes are likely to
facilitate consumer learning about RNPs, as well as decrease consumer uncertainty,
thus resulting in more positive consumer behaviour, such as intention and attitude.
Learning theories within different frameworks support the employment of
anthropomorphic attributes as a means of improving consumer comprehension (e.g.
Bandura’s (1997) Social Learning Theory). Academic literature also pinpoints that
inserting anthropomorphic attributes in websites facilitate the process of learning (e.g.
Tu, 2000; Keeling et al., 2010). Furthermore, according to Persuasion Theory,
consumer comprehension of a RNP is a precondition in the formation of their
attitudes, intentions and behaviours (e.g. Ratneshwar and Chaiken, 1991); adding
anthropomorphic attributes can therefore improve consumer attitude and intention
towards RNPs. Conclusively, employing anthropomorphism in RNP promotion, can
improve product comprehension and attitude. However, within the RNP domain, the
impact of inserting different types of anthropomorphic attributes on individuals’
anthropomorphism perception, comprehension and attitude is not yet explored. To
analyse the matter further, anthropomorphism is explained and categorized.
3.2.2 Anthropomorphism
Social Cognition Theory, which is based on Social Learning Theory (Bandura,
1977), explains the classification of objects as humans, animals or objects, and the
fact that the recognition of anthropomorphic characteristics is a basic cognitive
function (Kunda, 1999). Anthropomorphism is here defined as “an interpretation of
what is not human or personal in terms of human or personal characteristics”
(Merriam-Webster Collegiate Dictionary, 2004, p.53). Anthropomorphism is used to
describe any kind of “dispositional inference about a nonhuman agent” (Waytz et al.,
2014, p.220). ‘Social cues’ is the terminology commonly used within the field of
Information Technology, which refers to various types of anthropomorphic attributes
(Wang et al., 2007). In CMEs, individuals experience perceived anthropomorphism
106
when dealing with different types of social cues. Social Response Theory explains
“the more computers present characteristics that are associated with humans, the
more likely they are to elicit social behaviour” (Nass and Moon, 2000, p.97).
Therefore, the more computers possess social cues, the more people respond to them
with their social scripts (Sundar and Nass, 2000). As a result, people apply social
rules to computers and then respond to them accordingly (Nass and Steuer, 1993).
Social cues can stimulate social response (Nass and Steuer, 1993). Within the
consumer behaviour context, four cues relevant to eliciting social responses are
identified, namely human voice, language, interactivity and social role (Nass and
Steuer, 1993). Human voice, as a social cue, encourages individuals to apply human-
human interaction rules to their relationship with a CME (Reeves and Nass, 1996).
Voice is a distinctive human attribute as humans are the only creatures capable of
speaking. A study by Nass and Steuer (1993) showed how people react to different
voices on the same computer as if they are different social actors. Other types of
social cues are language, interactivity and social role. Language has been employed
in almost all computer-mediated interfaces as an important tool for communication,
but mainly as text and not spoken language (Wakefield et al., 2011). Nass et al.
(1995) explained how the use of strong or weak language on a screen, created the
perception of two different personalities. Interactivity consists of two-way
communication, active control and immediate feedback (Liu and Shrum, 2002).
Interactivity has been employed within CMEs and can increase perceived
anthropomorphism (Liu and Shrum, 2002; Wakefield et al., 2011). Finally social role
cues, explained within the social development literature, are about how individuals
define entities, including themselves, as human by examining the role that other
entities play (Wallace, 1983).
Various forms of social cues, or anthropomorphic attributes, have been employed
within CMEs as a way to improve human-computer interaction. In the computer
technology literature, they are referred to as the “technological efforts of imbuing
computers with human characteristics and capabilities” (Gong, 2008, p.1495). Within
this context, anthropomorphic attributes are operationalized in various forms such as
Artificial Intelligence (McCarthy, 1955) to anthropomorphized interface
representations, like faces (Sproull et al., 1996; Gong and Nass, 2007) and voices
107
(Gong and Lai, 2003; Nass and Brave, 2005). Nass et al. (1997) for instance realized
computers with voice output provoked gender stereotypes (Nass et al., 1997). Moon
(2000) claimed that when consumers are interacting with computers and have
provided some information about themselves, they are involved in an intimate self-
disclosure (Moon, 2000).
Although many studies have looked into individual anthropomorphic attributes
and how they are perceived by individuals (e.g. Nass et al., 1997) ; Moon, 2000),
there is no study to date explaining individual perceived anthropomorphism, when
comparing various degrees of social cues; for example, using an interactive virtual
agent instead of just employing language and voice. Moreover, inserting various
social cues for the purpose of facilitating consumer comprehension towards RNPs has
not been explored to date. In order to address this gap, this study examines how using
various anthropomorphic attributes within a CME, influence individuals perceived
anthropomorphism, which can then impact upon an individual’s comprehension and
change of attitude towards RNPs.
In order to differentiate between various forms of social cues, the common types
of social cues identified within the consumer behaviour context are explored further.
Language is used as part of every interface. Human voice is studied as a characteristic
of online agent (avatars), which is explained in section 3.2.4. Interactivity is looked at
from two points of view: (i) content interactivity, when it is inserted within the
content of the website, such as a 3D design, and (ii) online agent interactivity which is
explained further under the avatar section. Social role is considered as a characteristic
of avatars.
3.2.3 Content Interactivity
Steuer (1992, p.84) defined interactivity as “the extent to which users can
participate in modifying the form and content of a mediated environment in real
time”. Liu and Shrum (2002) described interactivity as a “multidimensional
construct that consists of active control, two-way communication, and
synchronicity” (cited in Wang et al. 2007, p.145). Synchronicity refers to the
immediate feedback, the simultaneousness of users input into a communication,
108
and the response they receive. Wang et al. (2007) expressed that two-way
communication and immediate feedback are the main characteristics in
interpersonal communication. In a user-computer interaction, individuals perceive
more interactivity when communication more closely resembles interpersonal
communication (Ha and James, 1998); therefore if synchronicity and two-way
communications are implemented within a CME, individuals should perceive the
communication as interpersonal and interactive.
Due to the nature of web browsing, all websites have some degree of interactivity.
However, by considering synchronicity and two-way communication (Wang et al.,
2007), websites can become more interpersonal. Content interactivity can provide a
more interpersonal platform for consumers, being anything from an interactive video
to an interactive 3D design. It can improve immediate feedback to the person, and
produce a two-way communication within the CME, resulting in a more interpersonal
communication. Furthermore, content interactivity can resemble the human-like
interactivity of a CME and create website socialness, which then leads to improved
perceived anthropomorphism (e.g. Liu and Shrum, 2002; Wakefield et al., 2011). It is
therefore predicted:
H1: A website with interactive content will be perceived as higher in
anthropomorphism than a website with no interactive content.
Another approach towards employing interactivity online would be to recruit an
interactive online agent; for example, using embodied agents (e.g. recommendation
agents) would indicate a richer interactive environment for the consumer (e.g.
Prendinger and Ishizuka, 2004). Interactivity is therefore further explored within an
online agent context.
3.2.4 Online agents - Avatars
An example of anthropomorphism employed in an online setting is the online
agent (avatar). This would be consistent with the Social Response Theory, where the
result would be a better social interaction. In CMEs, an avatar is a representation of an
entity; for example, a “pictorial representation of a human in a chat environment”
(Bahorsky et al., 1998, p.8) or “general graphic representations that are personified
by means of computer technology” (Holzwarth et al., 2006, p.20). Avatars can also be
109
defined as “virtual characters that can be used as company representatives and can
serve as identification figures, personal shopping assistants, web site guides or as
conversation partners” (Holzwarth et al., 2006, p.19). Holzwarth et al. (2006) then
explains “avatars can anthropomorphize the interaction and make the shopping
experience more interpersonal” (Holzwarth et al., 2006, p. 19). This definition is in
line with Nass and Steuer (1993) identification of social cues, such as interactivity and
social role cues, therefore has been employed in this research. Avatars have also been
referred to in different terms, such as embodied conversational agents (ECAs)
(Cassell et al., 2000), virtual agents (Abbattista et al., 2002), synthetic personae
(McBreen and Jack, 2001), interactive characters (Isbister and Nass, 2000), animated
pedagogical agents (Lester et al., 1997), artificial shopping agents (Redmond, 2002)
and animated interface agents (Dehn and van Mulken, 2000) in related literature.
Studies about avatars have generally examined the human morphology/
appearance, not the behaviour (e.g. Nowak and Rauh, 2005). Within avatar
morphology, there are two main dimensions: androgyny, a rating of the avatar's (lack
of) masculinity or femininity and avatar humanness, or the extent to which the avatar
looks human (Nowak and Rauh, 2005; Nowak et al., 2008). Androgyny is sometimes
operationalized as the position on which the avatar’s image falls on the masculinity-
femininity continuum (Nowak et al., 2008). Human-like and visibly gendered avatars
are more attractive and credible than androgynous avatars (Jin, 2009), and feminine
avatars are more attractive than masculine avatars (Nowak and Rauh, 2005). Avatars
can vary in the extent to which they resemble humans. Two main types of avatar -
human-like (interactive avatar) and static (image avatar) – are common. However,
which is preferable is unclear as there are inconsistent findings with respect to avatar
humanness. Donath (2007), for example, found how head and eye movement that
matched the conversation flow, increased participants’ perception of avatar
trustworthiness and friendliness. Other studies have found that a human-like spokes
avatar is perceived as more attractive and credible than a static avatar (e.g. Nowak and
Rauh, 2005; Jin, 2009): however, research has also reported a more human-like avatar
can be evaluated more negatively (e.g. Groom et al., 2009; Keeling et al., 2010).
Consequently, it is hypothesised that:
110
H2: A website with a human-like (interactive) avatar will have a different level of
perceived anthropomorphism, from a website with a static (image) avatar.
3.2.5 Human-like Avatars
Human-like avatars have been employed in order to improve consumer
experience online. They are perceived as attractive (Nowak and Rauh, 2005; Jin,
2009), credible (Nowak and Rauh, 2005), and are easily understood (Burgoon et al.
2009). While interacting with a human-like avatar, participants feel more positive
about the information presented to them in comparison to a no-avatar condition.
Further studies support how introducing a human-like avatar will result in improved
perceived anthropomorphism as oppose to a no-avatar condition. Research has shown
that the use of human-like avatars increases the level of social presence in an online
setting (Hale and Stiff, 1990) and creates an automatic social response (Lee and Nass,
2003). It is therefore predicted:
H3a: A website with a human-like (interactive) avatar will have higher perceived
anthropomorphism, than a website with no human-like avatar.
3.2.6 Static Avatars
Static avatars are anthropomorphic virtual images, which represent human
characteristics. They are known to be engaging, interesting and attractive (Nowak,
2004), though there are conflicting results in regards to preference levels of static
avatars to human-like avatars. Some studies support that the use of human-like avatars
results in a better comprehension and an increased feeling of social presence for
individuals, though some argue in a completely opposite doctrine (e.g. Mori, 1970;
Nowak and Biocca, 2003). The Uncanny Valley Theory (Mori, 1970; Groom et al.,
2009), which was first introduced within the human-robot context, explains this
conflict. The theory was developed within the domain of embodied humanoid robots.
111
Mori (1970) observed that while changing the robots from being totally non-realistic
to more human-like within the experimental conditions, this improved individuals’
perceived likability of them; however, when changing the robots more to a very life-
like machine, they were perceived as disturbing by individuals (Mori, 1970). He
therefore concluded that highly realistic agents receive more negative evaluations,
than agents that demonstrate only moderate realism (Mori, 1970; Groom et al., 2009).
Later studies confirmed that avatars which are perceived as anthropomorphic
create an expectation of sociability, and stimulate judgments reserved for social
entities like credibility and homophily (Bente et al., 2008). These expectations can
lead to more negative attributions, when the expectations are not met (Reeves and
Nass, 1996; Nowak and Biocca, 2003; Nowak, 2004). Furthermore, evidence suggests
that although participants tend to prefer a more realistic, human-like avatar, there is
also a higher risk of disappointment due to a higher expectation of participants, in
comparison to avatars with a more cartoon-like or abstract appearance (Keeling,
2010). Kim (2012) examined how mindless anthropomorphism can be evoked using a
simple cartoon-like avatar. Consequently, it is hypothesised that:
H3b: A website presenting RNPs with a static avatar (image) will have higher
perceived anthropomorphism, than a website with no static avatar.
Anthropomorphism and Comprehension
Various social cues, as identified by and Nass and Steuer (1993), have been
studied to understand their impact on individuals’ learning. Interactivity, as one
common social cue presented in different studies, is supported by many learning
theories and research for improving consumer’s learning (e.g. Brandt, 1997; Zhang et
al., 2006). One of these learning theories, the Theory of Multimedia Learning (Mayer,
1997), interprets how individuals learn in a multimedia environment. In this
framework, interactivity is defined as “reciprocal activity between a learner and a
multimedia learning system, in which the [re]action of the learner is dependent upon
the [re]action of the system and vice versa” (Domagk et al., 2010, p.1025). This
112
definition of interactivity is compatible with the usage of 3D product design, allowing
users to interact with the product. The Interactive Multimodal Learning Environment
(Moreno and Mayer, 2007), which is based on the Cognitive Theory of Multimedia
Learning (Mayer, 1997), further supports the employment of interactivity in
facilitating learning.
Interactivity along with the social role cue is also supported within Sociology. It
explains the fact that individuals learn through social interaction. People react to
computers as social actors (e.g. Reeves and Nass, 1996) and therefore interact with
computers as much as possible in order to learn. If computers elicit social behaviour,
this interaction increases (e.g. Social Response Theory). Social Learning Theory
(Bandura, 1977) explains how behaviour is a result of social interaction of people and
their environment (Walther, 1992). Social interaction between a learner and role
model is required and this interaction is affected by social presence. Behaviour,
personal factors and an ideal social environment can promote learning, providing an
appropriate degree of social presence (Tu, 2000). Employing an avatar improves
social presence and is likely to facilitate learning. As more learning is required in
order to understand RNPs, inserting social cues into CMEs promoting RNPs is
beneficial for individuals in order to learn about these products. As avatars possess
various social cues such as interactivity and social role, it is predicted that:
H4: Perceived anthropomorphism will have a positive relationship with consumer
comprehension of RNPs.
Website socialness has been proven to have a positive effect on consumer attitude
and intention (Wakefield et al., 2011). Content interactivity (e.g. using an interactive
3D product presentation) can also result in an improved perceived anthropomorphism,
which will influence consumer comprehension, attitude and purchase intention.
Furthermore, according to Persuasion Theory, learning leads to attitude and purchase
intention. This relationship is further supported within the CME literature. Previous
research indicates that interaction with a human-like agent influences consumer
attitude and intention towards the product (e.g. Choi et al., 2001; Jin, 2009; Wakefield
et al., 2011). Therefore:
113
H5: Perceived anthropomorphism will have a positive relationship with consumer
attitude and purchase intention towards RNPs.
3.2.7. Mindful and Mindless Anthropomorphism
Anthropomorphism is considered within various topics, including social presence,
socialness, and human touch (Nowak and Rauh, 2006; Sivaramakrishnan et al., 2007).
In CMEs, anthropomorphism is perceived while individuals interact with the interface
presenting the information. Various social cues within the interface can result in
individuals perceiving anthropomorphism. . However, some social cues might be
preferred by individuals. For instance, when an avatar is presented, individuals might
only perceive the avatar as anthropomorphic and not the content of the website;
whereas when no avatar is presented, the perceived anthropomorphism is a result of
interacting with the website’s content. It is important to understand how individuals
perceive anthropomorphism within CMEs, and also what the point of reference is
towards the interaction, content, avatar or both. There is no study exploring this area
of online anthropomorphism, therefore this study investigated the area further.
Another important factor in uncovering individuals’ perceived anthropomorphism
is the way people perceive this attribute. Anthropomorphism can be perceived
consciously (mindfully) or unconsciously (mindlessly) (e.g. Kim and Sundar, 2012).
Mindful anthropomorphism refers to when participants consciously rate a CME as
anthropomorphic. Nass and Moon (2000, p.93) justify the mindful nature of
anthropomorphism by referring to the definition of anthropomorphism as “a
thoughtful, sincere belief that the object has human characteristics”. Much of the
research examining mindful anthropomorphism is within the Human Robot
Interaction (HRI) literature (e.g. Powers and Kiesler, 2006; Bartneck et al., 2009).
However these categorizations have also been employed within the CME literature
while studying employment of social cues such as avatars (e.g. Nowak and Rauh,
2005).
Mindless anthropomorphism is explained via a heuristic systematic model
(Chaiken, 1987) and the MAIN model (Sundar, 2008). These models describe how
heuristic cues trigger a consumer’s unconscious responses towards anthropomorphic
114
attributes. Kim (2012) explains how richer modality cues elicit a “realism heuristic”
(Kim and Sundar, 2012, p242), which imitates face-to-face communication. Richer
modality cues further provoke agency affordance, which refers to any evidence of the
existence of an intelligent entity within the context of interaction, leading to a “social
presence heuristic” (Kim and Sundar, 2012, p.242). Studies have manipulated
anthropomorphism by employing an avatar and directly asking consumers if the
character is humanlike, machine-like, natural and so on (e.g. Groom et al., 2009;
Keeling et al., 2010). Kim and Sundar (2012) argued that this approach denies the
mindless anthropomorphism individuals perceived within CMEs. He further
developed perceived mindful and mindless scales to measure individuals’ reaction
towards various social cues.
Another type of perceived anthropomorphism discussed within the CME context,
is perceived website anthropomorphism, also known as website socialness. Perceived
website anthropomorphism is strongly linked to the Social Response Theory (Moon,
2000), and is used in studies to “describe the phenomenon of users treating
technology or technology interface such as websites as social actors” (Wakefield et
al., 2011, p.119). Perceived website anthropomorphism refers to “the extent to which
consumers detect socialness on a website” such as politeness and friendliness
(Wakefield et al., 2011, p.119). Wakefield (2011) further links back to literature
suggesting that perceived anthropomorphism is a mindless act (e.g. Langer, 1989;
Nass and Moon, 2000). This statement can be linked to the fact that an individual’s
automatic reaction towards human-computer interaction is based on the heuristic
aspect of the information provided (e.g. Reeves and Nass, 1996). Any interactive
content enhancing the visual modality results in an improved mindless
anthropomorphism similar to virtual agents, as rich media can transit social cues
(Wang et al., 2007). Nass and Carney (1999) explored the use of interactivity in a
CME and realized that increased interactivity escalates the possibility of users’
automatic social response towards the technology. Kim and Sundar (2012) further
echoed that mindless anthropomorphism increases when individuals deal with a high
interactivity condition therefore:
H6a: A website presenting RNPs with interactive content will result in a higher
mindless anthropomorphism, than a website with no interactive content.
115
As well as interactive content, an avatar will enrich the audio and visual
modalities in a CME by making the experience as close as possible to one-to-one
communication. Avatars have been considered as a “way of improving interactivity
and giving a more human touch” to the CME (Mimoun et al., 2012, p.605).
Employing avatars will also lead to a better social presence heuristic, leading to an
increased unconscious response towards the social cues presented. Research has
explained how inserting a static avatar online has improved consumers’ perceived
level of mindless anthropomorphism (Kim and Sundar, 2012). Kim and Sundar
(2012) further explained “anthropomorphic cues do not have to be fancy in order to
elicit human-like attribution” (Kim and Sundar, 2012, p.249). He explained that
mindless anthropomorphism was induced using a simple static cartoon-like avatar
online. Therefore:
H6b: A website presenting RNPs with a static avatar will result in a higher
mindless anthropomorphism, than a website with no static avatar.
Conversely, a more human-like interactive avatar is attractive and the information
provided by them is perceived as credible (e.g. Nowak and Rauh, 2005; Jin, 2009).
There are studies supporting the employment of an interactive avatar in order to
improve an individual’s perceived anthropomorphism (Lee and Nass, 2003; Jin, 2009;
Kang and Gratch, 2014). Furthermore, employing a close to real virtual agent will
result in realism, as well as social presence, heuristics. Therefore:
H6c: A website presenting RNPs with a human-like avatar will result in a higher
mindless anthropomorphism, than a website with no human-like avatar.
RNPs are complex, high-involvement products (e.g. Feiereisen, 2009); however
there are disagreements in the definition of high vs. low involvement product.
According to Antil (1984), it is the individual’s perception that determines if the
product is high or low involvement, further explaining that there are many stimuli that
can influence the involvement level, such as the message, product, spokesperson and
setting. It is therefore almost impossible to identify a product as high or low
involvement by merely looking at the product’s characteristics (Antil, 1984).
116
Conversely, other studies have applied the definition of high and low involvement
towards products. The involvement level is explained as the individual’s time and
effort in their buying decision. Individuals spend more time and effort on purchasing a
high-involvement product, whereas they place much less cognitive effort and time on
deciding to buy low-involvement products (Bloch and Richins, 1983). Low-
involvement products normally do not have a major impact on consumer’s expenses,
lifestyle and self-concept, whereas high-involvement products are not only more
expensive, but also affect one’s lifestyle and self-concept. High-involvement products
are also owned and utilized for a longer period of time (Subhani et al., 2012).
RNPs are new concepts therefore individuals need to spend a lot of time and
effort in order to understand these products, making them high-involvement products
(Feiereisen, 2009). As such, individuals need to follow a systematic route to
comprehend them. Having said this, individuals might not be motivated enough to
learn about RNPs; that is, they might not be socially motivated, as the product is not
available in the marketplace and they are aware they cannot possess one in the near
future. Additionally, individuals might not be cognitively motivated as the
information about the products is abstract; individuals may therefore attempt to limit
their cognitive effort by reaching for necessary information in regards to RNPs.
Excessive social cues might therefore present themselves as distractions. For example
using human-like avatars, which try to interact with individuals and explain the
product using human voice, might get ignored or may irritate unmotivated
individuals. This might decrease the level of unconscious anthropomorphism towards
human-like avatars, as a representation of social cues that need individual interaction.
However, a static avatar might result in a better mindless anthropomorphism, as this
can be easily ignored by individuals if preferred. Therefore:
H7a: A website presenting RNPs with a static avatar will result in a higher
mindless anthropomorphism, than a website with a human-like avatar.
Comparing the use of an avatar and a 3D design, although interacting with a 3D
design takes extra effort, if consumers perceive it as a tool facilitating their learning,
they might attempt to take advantage of this technology. Furthermore, a 3D design is
directly related to the product related information, which is clearly observable.
117
However, an avatar represents a virtual agent that might facilitate consumer
understanding, by indirectly presenting the product-related information, which is not
directly observable. Therefore:
H7b: A website presenting RNPs with interactive content will result in a higher
mindless anthropomorphism, than a website with an avatar (static or human-like).
Kim and Sundar (2012) challenged the use of mindful scales within this context.
He explained that if anthropomorphism occurs mindlessly, participants who are
exposed to a human-like avatar should mindlessly perceive it as more
anthropomorphic and not mindfully. The result of Kim and Sundar’s (2012)
experiment showed that participants reported a lesser degree of mindful
anthropomorphism (human-likeness) when dealing with a human-like avatar. They
concluded that individuals intentionally deny treating the website as human-like when
they were dealing with a human-like avatar or highly interactive content. Further
experiments revealed that participants, who reported a low mindful
anthropomorphism when exposed to human-like avatar or high interactivity, reported
a high mindless perceived anthropomorphism compared to the conditions of no-avatar
or low interactivity (Kim and Sundar, 2012). The more social cues were forced, the
less likely individuals mindfully perceived it as anthropomorphic. The interactive
avatar, as opposed to a static avatar, can be seen as a forced social cue, and according
to the Uncanny Valley Theory (Mori, 1970), can result in a more negative evaluation.
Interactive content can be seen as the least obvious presentation of social cues online.
Therefore:
H8a: A website presenting RNPs with a static avatar will result in a higher level
of mindful anthropomorphism, than a website with a human-like avatar.
H8b: A website presenting RNPs with interactive content will result in a higher
level of mindful anthropomorphism, than a website with an avatar (static or human-
like).
118
3.3 Methodology
3.3.1. Perceived Anthropomorphism - Measurement
For a long time, anthropomorphism was evaluated as a mindful phenomenon in
the social response literature. Anthropomorphism was manipulated using mainly the
same type of social cues, such as comparing various forms of avatars. Nowak and
Rauh (2006) for instance employed a 3-item semantic differential scale asking
participants to rate whether the image “looks very human/does not look human”,
“looks very realistic/does not look realistic” and “looks very cartoon-like/does not
look like a cartoon” (Nowak and Rauh, 2006, p.160). This scale was used to compare
various avatars from animals to human-like. Powers and Kiesler (2006) developed a
3-item 10-point semantic differential scale asking participants to rate if they perceived
the website as “human-like/machinelike, natural/unnatural, lifelike/artificial” (cited
in (Kim and Sundar, 2012, p.247). Kim and Sundar (2012) employed the scale for
evaluating individual mindful anthropomorphism in various websites, by
manipulating the presence of a human-like agent and interactivity.
Mindless anthropomorphism has been evaluated by researchers in various settings.
Wakefield et al. (2011) developed a scale consisting of adjectives derived from the
social response literature discussed by Nass and Steuer (1993), Reeves and Nass
(1996) and McMillan and Hwang (2002). Wakefield’s measure is a 7-item scale
including the adjectives: friendly, helpful, polite, informative, likeable, intelligent,
and interactive. Kim and Sundar (2012) have employed a similar scale of 8-items 10-
points scale adjectives of: attractive, exciting, pleasant, interesting, likeable, sociable,
friendly and personal, which was then reduced to the four items of likeable, sociable,
friendly and personal. More recent studies distinguish mindless anthropomorphism
alongside mindful anthropomorphism in their measurement (e.g. Wakefield et al.
2011; Kim and Sundar, 2012). This study is concerned with an individual’s perception
of anthropomorphism when dealing with different websites displaying various social
cues, including content interactivity and static/human-like avatars. Kim and Sundar’s
(2012) mindful and mindless scale are therefore selected for the purpose of this study,
as it looks at both website and avatar factors.
119
3.3.2. Research Design
The research design is a single factor between-subject design for two separate
products. The study manipulation involves anthropomorphism (mindful and mindless)
within five settings of a base level (no interactivity and no anthropomorphism),
low/high content interactivity conditions and static/human-like avatars conditions.
Products were selected by conducting an online experiment with 50 participants, in
order to confirm the products to be RNPs. In this study, a human-like image was used
as the static avatar because a human-like virtual image will result in a stronger social
response than a non-human image.
Product selection
To identify suitable RNPs for this study, nine RNPs were selected from various
websites, where there was enough information and pictorials available online to
provide a presentation. The products were tested in order to confirm their
characteristics as RNPs, via an online experiment. Perceived product newness was
measured by combining Gregan-Paxton et al.’s (2002) definition of RNPs and
Hoeffler’s framework that has been used in recent RNP related research (e.g.
(Alexander et al., 2008) (Appendix 1). Gregan-Paxton et al.’s definition is related to
the level of product newness; whereas Hoeffler’s framework is concerned with
whether participants understand and like the product (group 1), or whether
participants are aware that they need to change their behaviour in order to do new
things with the product (group 2). Products that were scored highly in either group
were identified and two products, which fit a similar domain were selected, namely
the Dismount Washer (referred to as product 1) and the Washing Machine in a
Wardrobe (referred to as product 2).
120
3.3.3. Experimental Conditions
There are five conditions designed for the purpose of this study for products 1 and
2: (1) Base website (low interactivity, low anthropomorphism), (2) low interactivity
(low level of content interactivity), (3) high interactivity (high level of content
interactivity), (4) low anthropomorphism (static avatar), and (5) high
anthropomorphism (human-like avatar). Both static and human-like avatars were
used for this study. The static avatar is a picture of the anthropomorphic avatar. The
anthropomorphic avatar was purchased from sitepal4. Sitepal is a provider of online
virtual agents and avatars can be purchased on an annual contract. The avatar
possesses artificial intelligence and contains a general database of commands and
responses. In order to customize the avatar according to the nature and purpose of this
study, a focus group was employed so as to develop a close-to-reality database for the
human-like avatar. The focus group consisted of six participants (2 female) between
the ages of 20-49. They were asked to think about the questions participants within
the human-like settings could ask from the avatar, and what would be the best
answers the avatar could reply with. In addition, participants were encouraged to
think about any ideas that would make the human-like avatar more anthropomorphic.
They suggested changing the presentation format of the human-like avatar. There
were also suggestions in regard to how the human-like avatar should start her
introduction. The complete explanation of questions and suggestions can be found in
Appendix 15. The questions and answers extracted from the focus group were added
to the existing database of the purchased avatar.
Content interactivity was implemented in the form of a 3D design of the selected
RNPs for the high interactivity condition. The 3D design enabled participants to
interact with the product, rotate it, zoom in/out of it, and move the model. In the low
interactivity condition there were static pictures and text. Snapshots of various
experimental conditions can be found in Appendix 16.
4 www.sitepal.com
121
3.3.4. Data Collection and Sample
250 experimental participants from the USA were recruited via Amazon
Mechanical Turk (Mturk)5for each product, making a total of 500 participants.
Participants were randomly assigned to one of two products and one of the five
experiment conditions (i.e., webpage). After examining the webpage, participants
were directed to the online survey, which was designed and distributed using iSurvey.
Participants’ demographic data consisting of gender, age, education and monthly
income were gathered for all experiments.
Table 6: Demographics of Sample
Factor PercentageGender Male: 50.6%
Female: 49.4%Age 20-29: 36%
30-39: 39.8%40-49: 12.2%Over 50: 12%
Monthly Income less than $999 pm: 20.6%$1000-$1999 pm: 26.4%$2000-$2999 pm: 21%$3000-$3999 pm: 15.8%More than $4000 pm: 16.2%
Education Primary school: 1%Secondary/high school: 26.6%Undergraduate college/university: 48.4%Graduate college/university: 22.6%PhD: 1.4%
3.3.5. The Stimuli and Measures
Data was collected via an online questionnaire, developed using existing scales
within the related literature. The questionnaire contained measures for the dependent
variables of comprehension, attitude and purchase intention, followed by measures of
mindless and mindful anthropomorphism.
Comprehension
5 www.mturk.com
122
In order to measure comprehension, a 6-item 7-point scale developed by
Feiereisen et al. (2008) was selected. The scale itself is combination of a 2-item
semantic differential scale by (Phillips, 2000) and a 4-item 7-point Likert scale by
Moreau et al. (2001). The KMO and Bartlett’s test both support the combination of
the scale items as appropriate (Feiereisen et al., 2008). The scale can be found in
Appendix 17.
Attitude
In order to measure attitude, a 10-item scale developed by Voss et al. (2003) was
used. The scale was deemed reliable as further studies reported a Cronbach alpha of
0.91 for the utilitarian part of the scale, and 0.89 for the hedonic part of the scale
(Ogertschnig and van der Heijden, 2004). The scale was measured by a 7-point
semantic differential scale and is presented in Appendix 17.
Purchase Intention
To measure purchase intention, a 4-item 7-point Likert scale adopted from Moon
et al. (2008) was used. A Cronbach alpha of 0.86 supported the reliability of the scale.
The scale can be found in Appendix 17.
Anthropomorphism
In measuring anthropomorphism, both mindful and mindless anthropomorphism
were considered. The mindful anthropomorphic scale consisted of three statements
measured by a 10-point scale (α = .86, M = 15.53, SD = 5.94); the mindless
anthropomorphism scale consisted of four items measured on a 10-point semantic
differential scale (α = 0.85, M = 24.90, SD = 7.25) (Kim and Sundar, 2012; see
Appendix 17 for more details).
3.4 Results
To test the hypotheses, constructs were measured using existing multi item scales
to evaluate respondent responses to the stimuli. The scales were all established scales
extracted from related studies, though some scales were slightly altered to suit the
context of the study. Confirmatory Factor Analysis (CFA) was a suitable method of
testing the measurement model. However, before performing CFA, KMO and
123
Bartlett’s test for sphericity were performed to determine that the sample was
satisfactory for factor analysis. The result showed KMO=0.92 which is satisfactory
(more than 0.5), with Bartlett's test of sphericity (tests whether the correlation matrix
is an identity matrix) was 11511.033 with P value of 0.000 and df 351. It rejects the
null hypothesis that the variables in the population correlation matrix are
uncorrelated; therefore factor analysis can be performed. The CFA model was then
designed according to the stages defined by Hair et al. (2010). More information is
available in Appendix 18. For the purpose of CFA, the full data set (500 participants)
was employed.
The initial model fit was not acceptable (χ2(314)=2720.707 p=.000; CFI= .789;
GFI=.635; AGFI= .561; RMSEA=.124; Pclose=.000), therefore the model was
examined. Items with factor loadings less than 0.5 and with the biggest modification
indices were removed. In this study, the error terms were not correlated as a solution
to improve model fit. According to Hair et al. (2010, p.646), three to four fit indices
are usually enough to provide adequate evidence of model fit; however researchers
should consider at least one incremental index (NFI, CFI etc.) and one absolute index
(GFI, RMSEA, SRMR etc). By looking therefore at CMIN/DF, AGFI, CFI and
RMSEA, the final model has a good fit (χ2(158)=391.026 p=.000 ; CFI=.971;
GFI= .925; AGFI=.90; RMSEA= .05; Pclose=.141). Furthermore, all items loaded
significantly on their respective construct, which supports the convergent validity of
the measurement items. All final scales were tested for uni-dimensionality and
homogeneity, and were all satisfactory. Detailed information for each scale’s uni-
dimensionality and homogeneity can be found in Appendix 19.
Table 7 below indicates the number of items removed during CFA. The scales
have also been examined in order to determine whether the measures were appropriate
for further use in hypothesis testing. Bartlett’s test for sphericity and KMO were
performed for individual scales for the purpose of homogeneity, which all appeared to
be satisfactory.
124
Table 7: Summary of items removed from the model.
Number of items removed during CFA
Attitude 5Purchase Intention 1Mindless Anthropomorphism 1
The scales’ internal consistency estimates and reliability were tested and deemed
satisfactory. Consequently Composite Reliability (CR), Average Variance Extracted
(AVE), Maximum Shared Variance (MSV), Average Shared Variance (ASV) and
Cronbach alpha were calculated and evaluated (Fornell and Larcker, 1981; Hair et al.,
2010). Table 8 shows that all the conditions necessary to demonstrate reliability,
convergent and discriminant validity were met.
Table 8: Reliability, Discriminant/convergence validity for final model.
CR
(>0.7)
AVE
(>0.5)
(<CR)
MSV
(<AVE)ASV α
Mindful Anthropomorphism 0.852 0.658 0.364 0.181 0.85
Mindless
Anthropomorphism0.916 0.784 0.364 0.249 0.91
Anthropomorphism 0.895 0.712 0.364 .0203 0.88
Purchase Intention 0.940 0.840 0.281 0.209 0.94
Comprehension 0.923 0.668 0.176 0.129 0.93
Attitude 0.896 0.635 0.281 0.214 0.90
All scales’ distribution appears to be close to normal. There were no serious
concerns regarding the normality of any of the variable (Field, 2005). Detailed
information about each variable’s tests of normality and histograms can be found in
Appendix 20.
3.4.1. Anthropomorphism manipulation
The first section of this paper is concerned with perceived anthropomorphism as a
combination of mindful and mindless anthropomorphism. The hypotheses then related
to correlation between anthropomorphism and comprehension, attitude and purchase
intention is discussed. Table 4 provides a summary of the hypotheses’ results.
125
Interactive Content
An independent sample t-test for products 1 and 2 revealed a non-significant
difference in participants’ perceived anthropomorphism, when comparing the high
interactive content condition to a no interactive content condition. Levene’s test for
equality of variances was non-significant for both products, indicating that the
assumption of homoscedasticity was met. H1 is not supported.
Static vs. Human-Like Avatar
The results indicated that participants did not perceive a significantly higher level
of anthropomorphism in websites with a human-like avatar for either product 1 or 2,
in comparison with websites with a static avatar; therefore H2 is rejected. However,
participants perceived a significantly higher level of anthropomorphism in a website
with a human-like avatar (Mean=4.09, SD= 1.29) in comparison with a website with
no avatar (Mean=3.43, SD= 1.12) for product 1, where t(98)=-2.73 p=.007. For
product 2, the difference was not significant. Levene’s test for equality of variances
was not significant, therefore the assumption of homoscedasticity was met for both
products. H3a is thus partially supported. Individuals did not perceive
anthropomorphism to be significantly different when comparing the static avatar
condition with no static avatar condition, therefore H3b is rejected.
Table 9: Test of hypotheses – Anthropomorphism manipulation
Hypotheses Conditions Dependent Variable t-test result
H1 High Interactive content
vs.No interactive content
Perceived Anthropomorphism
P1*: Non-significant – Rejected(t(98)= -1.11, p>0.05)
P2*: Non-significant – Rejected(t(98)= 0.77, p>0.05)
H2Human like avatar
vs.Static avatar
Perceived Anthropomorphism
P1: Non-significant – Rejected(t(98)= -1.63, p>0.05)
P2: Non-significant – Rejected(t(98)= -.643, p>0.05)
H3a
Human-like avatar vs.
No human-like avatar
Perceived Anthropomorphism
P1: Significant – Supported(MeanHA=4.09, SD= 1.29;
MeanNHA=3.43, SD=1.12, t(98)= -2.73, p=.007)
P2: Non-significant – Rejected(t(98)= -1.11, p>0.05)
H3b
Static avatar vs.
No static avatar
Perceived Anthropomorphism
P1: Non-significant – Rejected(t(98)= -.916, p>0.05)
P2: Non-significant – Rejected
126
(t(98)= .262, p>0.05)
H6a
High Interactive contentvs.
No Interactivity
Perceived mindless Anthropomorphism
P1: Non-significant – Rejected(t(98)= -1.28, p>0.05)
P2: Non-significant – Rejected(t(98)= .97, p>0.05)
H6b
Static avatar vs.
No static avatar
Perceived mindless Anthropomorphism
P1: Non-significant – Rejected(t(98)= -.954, p>0.05)
P2: Non-significant – Rejected(t(98)= .636, p>0.05)
H6c
human-like avatarvs.
No human-like avatar
Perceived mindless Anthropomorphism
P1: Significant – Supported(MeanHA=4.59, SD= 1.24;
MeanNHA=3.57, SD=1.44, t(98)= -3.78, p=.000)
P2: Non-significant – Rejected(t(98)= -1.64, p>0.05)
H7a
Static avatarvs.
Human-like avatar
Perceived mindless Anthropomorphism
P1: Significant – Supported(MeanHA=4.59, SD= 1.24;
MeanSA=3.85, SD=1.58, t(98)= -2.57, p=.012)
P2: Significant – Supported(MeanHA=4.62, SD= 148;
MeanSA=3.98, SD=1.46, t(98)= -2.16, p=.033)
H7b
High Interactive contentvs.
Static or human-like avatar
Perceived mindless Anthropomorphism
P1 (Static avatar): Non-significant – Rejected
(t(98)= .237, p>0.05)P1 (human-like avatar): Significant
(reverse) – Rejected(MeanHA=4.59, SD= 1.24;
MeanSA=3.92, SD=1.36, t(98)= -2.54, p=.013)
P2 (Static avatar): Non-significant – Rejected
(t(98)= -.494, p>0.05)P2 (human-like avatar): Non-
significant – Rejected(t(98)= -161, p>0.05)
H8a
Static avatarvs.
Human-like avatar
Perceived mindful Anthropomorphism
P1: Non-significant – Rejected(t(98)= -.436, p>0.05)
P2: Non-significant – Rejected(t(98)= .882, p>0.05)
H8b
Static avatarvs.
High Interactive content
Perceived mindful Anthropomorphism
P1: Non-significant – Rejected(t(98)= .084, p>0.05)
P2: Non-significant – Rejected(t(98)= .170, p>0.05)
* P1: Product 1 (Dismount Washer) ; P2: Product 2 (Washing Machine in Wardrobe)
Table 10: Test of hypotheses – Correlation between dependent variables
Hypotheses Conditions Dependent
Variable Result
H4
Correlation between perceived
anthropomorphism And
Comprehension
Perceived Anthropomorphism
P1: Moderate Positive Correlation – Supported
(r=0.40, N=250, p=.000) P2: Weak Positive Correlation –
Supported(r=0.36, N=250, p=.000)
127
H4
Correlation between perceived mindless anthropomorphism
AndComprehension
Perceived mindless Anthropomorphism
P1: Moderate Positive Correlation – Supported
(r=0.43, N=250, p=.000) P2: Weak Positive Correlation –
Supported(r=0.32, N=250, p=.000)
H4
Correlation between perceived mindful anthropomorphism
AndComprehension
Perceived mindful Anthropomorphism
P1: Weak Positive Correlation – Supported
(r=0.32, N=250, p=.000) P2: Weak Positive Correlation –
Supported(r=0.34, N=250, p=.000)
H5
Correlation between perceived
anthropomorphism And
Attitude
Perceived Anthropomorphism
P1: Moderate Positive Correlation – Supported
(r=0.61, N=250, p=.000) P2: Moderate Positive Correlation –
Supported(r=0.46, N=250, p=.000)
H5
Correlation between perceived mindless anthropomorphism
AndAttitude
Perceived mindless Anthropomorphism
P1: Moderate Positive Correlation – Supported
(r=0.61, N=250, p=.000) P2: Moderate Positive Correlation –
Supported(r=0.47, N=250, p=.000)
H5
Correlation between perceived mindful anthropomorphism
AndAttitude
Perceived mindful Anthropomorphism
P1: Moderate Positive Correlation – Supported
(r=0.43, N=250, p=.000) P2: Weak Positive Correlation –
Supported(r=0.35, N=250, p=.000)
H5
Correlation between perceived
anthropomorphism And
Purchase Intention
Perceived Anthropomorphism
P1: Moderate Positive Correlation – Supported
(r=0.53, N=250, p=.000) P2: Moderate Positive Correlation –
Supported(r=0.56, N=250, p=.000)
H5
Correlation between perceived mindless anthropomorphism
AndPurchase Intention
Perceived mindless Anthropomorphism
P1: Moderate Positive Correlation – Supported
(r=0.53, N=250, p=.000) P2: Moderate Positive Correlation –
Supported(r=0.56, N=250, p=.000)
H5
Correlation between perceived mindful anthropomorphism
AndPurchase intention
Perceived mindful Anthropomorphism
P1: Moderate Positive Correlation – Supported
(r=0.42, N=250, p=.000) P2: Moderate Positive Correlation –
Supported(r=0.43, N=250, p=.000)
* P1: Product 1 (Dismount Washer) ; P2: Product 2 (Washing Machine in Wardrobe)
128
3.4.2. Mindless Anthropomorphism
Table 4 shows descriptive statistics for mindless anthropomorphism across all
conditions for both products. T-test results show that there are no significant
differences between the means in the no and high interactivity conditions with either
products. T-tests were also non-significant when the no static avatar mindless
anthropomorphism was compared to the static avatar mindless anthropomorphism.
Consequently, neither H6a nor H6b can be supported. In contrast, the t-tests for product
1 found significant differences (t(98)= -3.78, p=.000) in mindless anthropomorphism
between the website with a human-like avatar (MeanHA=4.59, SD= 1.24) and the base
website (no human-like) (MeanNHA=3.57, SD=1.44). This was not significant for
product 2; therefore H6c is partially supported.
When interacting with a static avatar in comparison to a human-like avatar,
participants in the static-avatar condition experienced a significantly lower perceived
mindless anthropomorphism (MeanP1SA=3.85, SD=1.58; MeanP2SA=3.98, SD=1.46),
compared to participants interacting with a human-like avatar (MeanP1HA=4.59, SD=
1.24; MeanP2HA=4.62, SD= 148). The t-test results are (t(98)= -2.57, p=.012) for
product 1 and (t(98)= -2.16, p=.033) for product 2; therefore H7a is supported. In
contrast, the mindless anthropomorphism was not significantly higher for high
interactive content, in comparison to the static or human-like avatar; therefore H7b is
rejected.
3.4.3. Mindful Anthropomorphism
Looking into mindful anthropomorphism, individuals dealing with the static
avatar in comparison to the human-like avatar did not experience a significantly
higher mindful anthropomorphism for any of the products. Furthermore, T-test results
show that there are no significant differences between the means in the static avatar
and high interactive content condition, for either of the products; therefore H8 a and
H8b are not supported.
3.4.4. Correlation between dependent variables
In this section, the correlation between variables of perceived anthropomorphism
and comprehension, attitude and purchase intention is examined and reported. As per
129
Table 5, the correlation showed a moderate (Dancey and Reidy, 2004) positive
correlation between perceived anthropomorphism and comprehension (r=0.40,
N=250, p=.000) for product 1, and a weak positive correlation (r=0.36, N-250,
p=.000) for product 2. The correlation was moderate for both products 1 (r=0.61,
N=250, p=.000) and 2 (r=0.46, N=250, p=.000) between perceived
anthropomorphism and attitude. Again, there was a moderate positive correlation
observed between perceived anthropomorphism and purchase intention for product 1
(r=0.53, N=250, p=.000) and product 2 (r=0.56, N=250, p=.000).
Due to categorising anthropomorphism into mindful and mindless in later stage of
this study, the correlation between the separate components and variables are also
examined. The correlation between mindless anthropomorphism and comprehension
was moderate positive for product 1 (r=0.43, N=250, p=.000) and weak positive for
product 2 (r=0.32, N=250, p=.000). For mindful anthropomorphism, the correlation
with comprehension showed a weak positive result for both products 1 (r=0.32,
N=250, p=.000) and 2 (r=0.34, N=250, p=.000). The result of the correlation test
between mindless anthropomorphism and attitude was moderate positive for product 1
(r=0.61, N=250, p=.000) and product 2 (r=0.47, N=250, p=.000). The correlation
between mindful anthropomorphism and attitude showed a moderate positive
correlation (r=0.43, N=250, p=.000) for product 1 and a weak positive correlation
(r=0.35, N=250, p=.000) for product 2. For purchase intention, the correlation
between mindless anthropomorphism and purchase intention was moderate positive
for both products 1 (r=0.53, N=250, p=.000) and 2 (r=0.56, N=250, p=.000). The
same moderate positive correlation was observed between mindful anthropomorphism
and purchase intention for product 1 (r=0.42, N=250, p=.000) and product 2 (r=0.43,
N=250, p=.000).
3.5 General Discussion and Conclusion
This study adds to the body of knowledge regarding anthropomorphism,
exploring the impact of anthropomorphic attributes on consumers’ perceived
anthropomorphism, comprehension, attitude and purchase intention. Mindless and
mindful aspects of anthropomorphism were also investigated. When looking at
anthropomorphism as a collective concept including mindful and mindless, the results
130
indicate that participants perceived significantly higher anthropomorphism when
dealing with a human-like avatar in comparison to the no avatar condition. Human-
like avatars that provide all four types of social cues identified within the consumer
behaviour literature (Nass and Steuer, 1993) of language, human voice, interactivity
and social role, therefore create the sense of anthropomorphism amongst observers
when learning about RNPs. This adds to the anthropomorphism literature as well as
Social Response Theory (Moon, 2000) within the area of RNP or innovation
comprehension. Previous studies within Social Response Theory in CMEs were
inconsistent in observing whether human-like avatars can result in a better sense of
anthropomorphism (e.g. Mori, 1970; Nowak and Biocca, 2003). This study provides
empirical support for the influence of social cues in participants’ perceived
anthropomorphism.
Categorising anthropomorphism into mindless and mindful aspects produced
inconsistent results. Where mindless anthropomorphism was significantly higher as
participants were interacting with a human-like avatar in comparison to a base website
(no avatar, no interactivity), there was no difference within the mindful perceived
anthropomorphism of participants. Furthermore, when dealing with a human-like
avatar in comparison to a static-avatar, participants perceived a significantly higher
mindless anthropomorphism; but their perceived mindful anthropomorphism was not
significantly different. In fact, mindful anthropomorphism was not perceived
significantly different in any conditions hypothesized within the study. This result has
a few implications. Firstly, it adds to the Social Response Theory that the influence of
inserting various social cues within CMEs are not perceived the same when
anthropomorphism is looked into mindlessly as opposed to mindfully. Secondly, it
adds to the anthropomorphism literature, by emphasizing the necessity of looking into
anthropomorphism in two distinct category of mindless and mindful, as the perception
amongst these two groups were not similar. The result also indicates the importance
of advancing measurement scales for perceived anthropomorphism, in order to
include all types of perceived anthropomorphism within a comprehensive scale.
The result of this paper adds evidence to the findings of Kim and Sundar (2012)
that analysed various effects of mindless vs. mindful anthropomorphism within
CMEs. They concluded that people “intentionally denied treating the website in a
131
human way” (Kim and Sundar, 2012, p.248). As is evident, individuals did not report
any significantly different mindful anthropomorphism in this study as well. Kim and
Sundar (2012, p.249) also explained how people who mindfully “denied treating the
website in human-terms when exposed to the character or the high level of
interactivity, tended more to attribute personal character to the website”. The result
of this study supports this statement, as participants reported a significantly higher
level of mindless anthropomorphism when dealing with human-like avatar in
comparison to no-avatar/no-interactivity; whereas no mindful anthropomorphism was
perceived by the same individuals.
Additionally, by looking into perceived anthropomorphism and participants’
comprehension of RNPs, attitude and purchase intention, a positive moderate
correlation between perceived anthropomorphism and comprehension, attitude and
purchase intention was apparent. The result is almost similar for the correlation of
mindless anthropomorphism and comprehension, attitude and purchase intention;
however the correlation between mindful anthropomorphism and comprehension was
weak, and was inconsistent with the other relationships. Overall the findings indicate
the benefits of including social cues in a form that participants perceive
anthropomorphism, as then it will influence individuals’ comprehension, attitude and
purchase intention towards RNPs. According to this study’s results, inserting a
human-like avatar within the RNP promotion platforms can improve an individual’s
understanding of RNPs, as well as their attitude and intention towards these products.
Finally, this study contributes to consumer online behaviour literature within the
field of RNPs. The result of the study adds to the RNP promotion literature, as to how
to use various social cues to improve consumer understanding and attitude towards
such products. This result is useful for managers in understanding the
anthropomorphic factors affecting consumer understanding and online behaviour
towards RNPs, and highly innovative concepts.
3.6 Limitations and Directions for Future Research
Although both RNPs used in this study are described as utilitarian, there are still
some inconsistencies in the result of the studies across RNPs. This could be due to the
importance of studying the innovation attribute, which is indicated in previous
literature (e.g. Rogers, 2003) when investigating innovation diffusion. The lack of
132
considering product attributes is one of the main criticisms of diffusion of innovation
literature, as consumers’ perception of innovation attribute is not examined in related
studies (Lowrey, 1991). Understanding various attributes such as those identified by
Rogers (1995) of relative advantage, compatibility, complexity, trialability and
observability, can thus shed light into why this paper’s findings are inconsistent and
different across various RNPs. Another element to explore is the differences amongst
participants in order to understand the impact consumer characteristics might have on
the results. For example, a preliminary examination of the gender factor revealed
some differences amongst male and female’s perceived anthropomorphism, attitude
and comprehension.
One main limitation within this paper is the lack of exhaustive literature on
perceived anthropomorphism scales. The result of this study clearly indicates that
scales measuring perceived anthropomorphism need to take into account the fact that
perceived mindless anthropomorphism is different to mindful anthropomorphism.
There is also another category looking into the ‘socialness’ of the websites; for
example Wakefield et al. (2011) has developed a scale namely ‘website socialness’
which is very similar to the mindless scale used within this study. Website socialness
aims to capture individuals’ perceived anthropomorphism as a result of any socialness
inserted within a website; this can be the inclusion of any social cues, such as Nass
and Steuer’s (1993) categorization of language, human voice, interactivity and social
role. Website socialness can therefore be a website with interactive content, and not
necessarily an interactive avatar. Therefore, a comprehensive scale that includes
mindless and mindful anthropomorphism as well as website socialness might result in
a better understanding of any anthropomorphism perceived by individuals, as a result
of the insertion of social cues. Preliminary research into this area indicated a good set
of results on socialness scales.
133
Chapter 4: Examining the impact of consumer innovativeness on
RNP comprehension attitude and purchase intention
4.1 Introduction
Innovation is considered as a crucial competitive power for firms (Andries et al.,
2009) and is an important factor in global economic growth (Golder and Tellis, 1997).
New and innovative products can contribute to a company’s growth, competitive
advantage and profitability (Steenkamp et al., 1999) and improve consumer
satisfaction (Wilke and Sorvillo, 2005). Due to an extremely competitive market,
firms need to introduce innovation in the form of really new products (RNPs) rather
than incrementally new products (INPs) (Chao et al., 2012). RNPs are products that
are very different to existing products, and offer greater benefits than INPs; however
consumers need to change their behaviour in order to achieve the potential benefits of
RNPs (Hoeffler, 2003).
Although RNPs are being introduced into the marketplace, 40% to 90% of new
products fail, with highly innovative products failing at an even greater rate
(Cierpicki, 2000). In order to understand RNPs’ high failure rates, new product
adoption and diffusion of innovation factors need to be examined (Hauser et al.,
2006). By improving consumers’ innovation adoption rate, RNPs will be more
successful within the marketplace. Early research established a link between the
adoption of new products and innovativeness (Midgley and Dowling, 1978;
Hirschman, 1980). Consumer innovativeness, conceptualized into Innate
Innovativeness and Actualized Innovativeness (Midgley, 1977), has been proven to
significantly impact consumers’ adoption of new products (Im et al., 2003; Rogers,
2003; Weijters and Roehrich, 2004; Im et al., 2007). This study was concerned with
innate innovativeness, as actualized innovativeness was not possible for RNPs, due to
their unavailability in the marketplace. A conceptual model based on existing
innovation frameworks has been designed, that supports the links between innate
innovativeness, and a consumer’s comprehension of, attitude towards and purchase
intention towards RNPs (Hirschman, 1980; Goldsmith and Hofacker, 1991;
Hirunyawipada and Paswan, 2006; Hoffmann and Soyez, 2010; Bartels and Reinders,
2011).
134
Studies around innate innovativeness have indicated a positive but weak
relationship between consumer innovativeness and new product adoption (Goldsmith
et al., 1995; Citrin et al., 2000; Im et al., 2007); however the impact of various
consumer characteristics (such as motivated consumer innovativeness elements) upon
consumer comprehension, attitude and purchase intention towards RNPs has received
little examination. This research aims to fill this gap and provide insight into the
influence of motivated consumer innovativeness elements upon consumer RNP
comprehension, attitude change and purchase intention.
4.2 Theoretical Background
This section examines diffusion of innovation and consumer innovativeness.
Theories and frameworks in support of consumers’ adoption of innovation and
different types of innovativeness are explored in detail. Additionally, learning theories
in line with innovation adoption are investigated in order to understand the
relationship between innovativeness and comprehension.
4.2.1. Diffusion Theory
Diffusion is described as a process “by which an innovation is communicated
through certain channels over time among the members of a social system” (Rogers,
1995), where the message is about a new idea. The main elements of the diffusion
process are innovation, communication, time and the social system. Innovation is an
idea, object or practice that is perceived as new (e.g., a RNPs) by individuals. The
newness of the innovative product generates some degree of uncertainty in the mind
of the adopter, but it also provides them with an opportunity to resolve a problem that
they are currently unable to solve using existing products. Potential adopters will
search for information in order to reduce their uncertainty and conclusively adopt the
innovation (Rogers, 1995). Communication refers to the process in which participants
share information with one another to reach a mutual understanding of the RNP.
Potential adopters seek information via various communication channels, indicating
that diffusion is a very social process (Rogers, 1995).
The adoption process is not the same for all people. One aspect of difference is
time. Time is defined by three elements: firstly, the innovation-decision process;
135
secondly the rate an innovation is adopted by an individual; and, thirdly, the rate an
innovation is adopted within a system. As this study is concerned with individuals as
adopters, the time it takes for an individual to adopt an innovation is of interest to this
paper. This rate of innovation adoption determines which consumer adopter category
they lie within (e.g. innovator, early adopter...etc.). The last element of diffusion is the
social system, an interrelated unit that interacts in order to solve a problem. As RNPs
are not yet in the marketplace, a social system that includes individual interactions is
not possible and cannot be analysed. The diffusion process in this study is defined as
adopting RNPs as innovation by individuals as adopters, recognizing adopter
differences due to the variability of the element of time.
Various factors influence innovation diffusion and adoption, one of which is
consumer characteristics. Many studies have addressed how consumer characteristics
influence innovation: for example, studies have found that males are more innovative
than females (Tellis et al., 2009), and that features such as creativity, influence the
actual adoption of new products (Hirschman, 1980; Roger, 2003). However, this
study focuses more upon the cognitive aspect of individuals as adopters, such as
motivational factors; therefore such factors will be explored further.
Motivational factors can impact an individual’s innovation adoption
(Vandecasteele and Geuens, 2010), decision-making (Rogers, 1995) and learning
(Bandura, 1977; Franks and Oliver, 2012). Rogers (1995) illustrated that the
innovation decision-making process is an information seeking and processing activity,
motivating individuals to decrease uncertainty about the advantages/disadvantages of
the innovation (Rogers, 1995). Vandecasteele and Guenes (2010) explained how
individuals are motivated by various forms of sensory information or needs, to adopt
an innovation. Motivational sources were explained as the need for sensory
stimulation (hedonic), the drive to search for solutions for consumption-related
problems (functional), the need to have a good social status and support (social), and
the need for being involved in mentally demanding activities (cognitive). Bandura
(1977) explained via observational learning, how motivational sources result in
individuals having a good reason to imitate behaviour they value. Motivational
sources are explained further in Section 4.2.3.
136
Another element impacting the diffusion of innovation is product attributes.
Studies have revealed how various product attributes, such as perceived
advantageousness (Roger, 2003) or perceived usefulness (Irani, 2000), influence the
time of adoption. RNPs’ unique characteristics can therefore contribute to the time of
adoption. RNPs are perceived as high risk products; hence perceiving RNPs as
advantageous and useful requires more cognitive effort by individuals, in order to
understand these products, as more learning is required to comprehend them (e.g.
(Gregan-Paxton et al., 2002; Feiereisen et al., 2008). The communication channel also
impacts upon innovation adoption. Consequently, selecting the right communication
channel is essential in the diffusion process (Midgley and Dowling, 1978). The
Internet has been used as a promotional channel for RNPs, in order to create product
awareness and increase new product adoption rate (Bickart and Schindler, 2001;
Prince and Simon, 2009); therefore online platforms have been deemed suitable as
communication channels for RNPs.
As explained, one important element, which has an impact upon innovation
diffusion and adoption, is time. The element of time within Diffusion Theory explains
how individuals differ in their rate of innovation adoption. Consistent with Diffusion
Theory, innovativeness is the degree to which an individual adopts an innovation
earlier than other members of the social system (Rogers, 1995). Within consumer
behaviour research, innovativeness is described as a “willingness to change” (Hurt et
al., 1977; Im et al., 2003), a “predisposition to buy new products” (Hirschman, 1980;
Midgley and Dowling, 1993) and a “preference for new and different experiences”
(Pearson, 1970; Hirschman, 1980; Raju, 1980 cited in Tellis et al., 2009, p.2).
Consumer innovativeness is explained as a tendency to purchase different and new
products, rather than remain with former choices and consumption patterns
(Venkatesh and Nicosia, 1997). The shorter the adoption time, the greater the
indication is that the consumer is more willing to change their behaviour and adopt
the new concept or product.
Diffusion Theory further categorizes individuals into five groups, according to the
length of time they take to adopt: innovators, early adopters, early majority, late
majority and laggards (Rogers, 1983). Within the Diffusion of Innovation model,
innovators are referred to as venturesome (Rogers, 2003); this is the most important
137
characteristic that an innovator can have. Innovators take risks and are willing to
accept that sometimes innovations fail. They play the most important role in the
diffusion process by importing the innovation from outside the social system’s
boundaries. There are several other common characteristics of innovators. For
instance, they should be financially capable of bearing the monetary burden of the
possible loss, if the innovation happens to be unprofitable; furthermore, they should
be able to cope with the high uncertainty involved in innovation adoption.
Individuals with higher levels of innovativeness are active information seekers
with regards to new ideas. They can cope with a higher level of uncertainty in
comparison to other adopter types (Rogers, 1995). They have the quickest time-of-
adoption and are more likely to try new products. They believe they are more
knowledgeable about online shopping and they purchase more products online.
Additionally, they are more likely to view online shopping as quicker, cheaper, safer
and more fun than traditional shopping methods (Goldsmith and Lafferty, 2002).
Therefore, initial promotion of innovations (such as RNPs) through targeting
individuals with a higher level of innovativeness, gives them a better chance of
adoption into a new social system.
4.2.2. Consumer Innovativeness
Innovativeness is “the degree to which an individual or other unit is relatively
earlier in adopting new ideas than other members of a system” (Rogers, 1995, p.22).
Innovativeness can be measured as a behaviour, as a global personality trait, and/or as
a domain-specific personality trait (Goldsmith and Foxall, 2003). Consumer
innovativeness has been conceptualized in two main streams: “Innate
Innovativeness” and “Actualized Innovativeness” (Midgley, 1977, cited in Roehrich,
2004, p.672). Midgley (1977, p.75) defines innate innovativeness as “the degree to
which an individual makes innovation decisions independently from the
communicated experience of others”. Innate Innovativeness, or “global
innovativeness” (Goldsmith and Foxall, 2003), is a cognitive trait based on one’s
inherently innovative personality, tendency and cognitive style (Hirschman, 1980;
Midgley and Dowling, 1993; Steenkamp et al., 1999). Innate Innovativeness refers to
innovative consumers as being independent decision-makers (Midgley and Downling,
1978) and inherent novelty seekers (Manning et al., 1995). Innately innovative
138
consumers have a tendency to buy new and different products, and rather than staying
with current consumption patterns and existing products (Steenkamp et al., 1999),
they have a tendency to change (Hurt et al., 1977). Innate Innovativeness is further
grouped into two types of tendency to be innovative: Life Innovativeness and
Adoptive Innovativeness (Roehrich, 2004). The Life Innovativeness tendency
explains “the ability to introduce newness in one’s life” (Vandecasteele and Geuens,
2010, p.7), which is beyond merely adopting an innovation; whereas the Adoptive
Innovativeness tendency is about one’s predisposition to adopt innovation, such as a
consumer’s interest and behavioural tendency towards innovation. This study takes
into account Adoptive Innovativeness to further explore attitudes and behaviours
towards RNPs. Innate Innovativeness can lead to attitudes, as an Adoptive
Innovativeness tendency.
Actualized Innovativeness is a behavioural perspective. It translates into new
product adoption behaviour and is the actual acquisition of new information, ideas
and products (Hirschman, 1980; Midgley and Dowling, 1978). Actualized
Innovativeness is an authentic innovative behaviour rather than a cognitive trait
(Midgley, 1977). Hirschman (1980 cited in Vandecasteele and Geuens, 2010, p.7)
categorized Actualized Innovativeness into three concepts: Use Innovativeness
behaviour, Vicarious Innovativeness behaviour and Adoptive Innovativeness
behaviour.
Use Innovativeness behaviour refers to using previously adopted products to solve
an existing consumption problem. Vicarious Innovativeness behaviour concerns the
acquisition of information and learning about new concepts. Adoptive Innovativeness
is the actual adoption of new products. Due to the nature of RNPs (they are not
currently in the marketplace), Adoptive Innovativeness is not applicable. Use
Innovativeness is also not possible due to the really new and innovative nature of the
products. Therefore, only Vicarious Innovativeness behaviour can be explored further
when considering RNPs. This study is concerned with the cognitive aspect of
innovativeness, namely Innate Innovativeness. Furthermore, a model based on
innovativeness theories is developed to analyse the relationship between various
innovative consumers and individuals learning (Fig.2). It considers Innate
139
Innovativeness as a “cognitive style, a personality-like construct” (Pagani, 2007
p.711), leading to Vicarious Innovativeness behaviour that involves comprehension.
4.2.2.1 Consumer learning and innovation
Learning models also provide support for the relationship between
comprehension, attitude and purchase intention. According to Persuasion Theory,
learning will result in a higher purchase intention and more positive attitude
(Ratneshwar and Chaiken, 1991). Persuasion Theory explains how comprehension is
a prerequisite to the formation of attitudes, intention and behaviour, especially under
the central or systematic processing route (Ratneshwar and Chaiken, 1991). Two
routes to persuasion may be relevant according to the Elaboration Likelihood Model
(Petty and Cacioppo, 1981). Consumers are either involved consumers (central route),
who process arguments in the persuasion attempt and elaborate on the arguments
resulting in a positive or negative evaluation; or uninvolved consumers, who will use
peripheral cues to form an empirical judgment, which can be positive or negative.
Involved consumers put more cognitive effort towards the central route, which is
characterized by attention being paid to the message content; whereas uninvolved
consumers follow the peripheral route, paying more attention to peripheral cues (Petty
and Cacioppo, 1981; Petty and Cacioppo, 1986). RNPs are by definition complex,
‘high-involvement’ products (Hoeffler, 2003). If individuals do not pay attention to
the message content and only follow the peripheral route, they will have difficulty
understanding RNPs (Hoeffler, 2003). If individuals learn about RNPs better through
using the central route, they will have a more positive attitude and higher purchase
intention towards these products. As such, it is hypothesised that:
H1a: Comprehension has a positive influence on attitude towards RNPs.
H1b: Comprehension has a positive influence on purchase intention towards
RNPs.
The relationship between attitude and purchase intention within the innovation
context is further supported by the innovation decision process, a process that
comprises five stages: knowledge, persuasion, decision, implementation and
confirmation. Knowledge is when an individual learns about an innovation.
Persuasion happens when an individual forms a “favourable or unfavourable attitude
140
towards innovation” (Rogers, 1995, p.20). Decision happens when an individual gets
involved in activities leading to a choice to either adopt or reject (or their intention to
adopt or reject) the innovation. Implementation refers to using the innovation, and
finally confirmation “seeks reinforcement of an innovation-decision that has already
been made” (Rogers, 1995, p.20). In this theory, comprehension leads to attitude
formation. This statement is further supported in the innovation adoption literature,
that indicates the acquisition of knowledge is linked to adoption attitude (Goldsmith
and Newell, 1997; Pagani, 2007). Therefore:
H2: Attitude has a positive influence on purchase intention towards RNPs.
Figure 2: Theoretical framework
4.2.3. Innate Innovativeness and Consumer Motivation
Innate innovativeness can be divided up into four distinct aspects – hedonic,
functional, cognitive and social (Vandecasteele and Geuens, 2010). Many consumer
innovativeness traits take into account the hedonic dimension of consumer
innovativeness. Hedonic needs include needs for sensory stimulation and novelty
(Hoyer and Stokbuger-Sauer, 2012). These needs replicate an individual’s inherent
desire for sensory pleasure (Hoyer and MacInnis, 2010). The Hedonic dimension is
examined and employed in many studies such as Baumgartner and Steenkamp’s
Innate Innovativeness
Comprehension
Attitude
Intention
141
(1996) Exploratory Consumer Buying Behaviour (ECBB) study. This research
focused on behaviour that provides “the consumer with a satisfactory level of
stimulation” (Baumgartner and Steenkamp 1996, p.122). The Hedonic dimension is
also reflected in the innovativeness concept introduced by Venkatraman and Price
(1990) as the “sensory innovativeness dimension” (cited in Roehrich, 2004, p.672).
Roehrich (1994) has also employed hedonic innovativeness as the drive to adopt
innovativeness for hedonic causes, such as the appreciation of a product’s newness.
Therefore, as an innate perspective, individuals adopt innovations to satisfy their
hedonic needs.
On the other hand, from a product perspective, it is claimed that hedonic
innovations and functional innovations attract dissimilar types of innovative
consumers (Hirschman, 1984). It is therefore important to consider this distinction
when examining RNPs and how consumers perceive and adopt these products.
Various promotion factors can also influence the hedonic value perceived by
individuals; for example, according to literature by adding a 3D picture within a
CME, the product’s hedonic value is more observable. Furthermore, a richer interface
makes the shopping experience more enjoyable, that can upsurge consumers’ attitude
(e.g. (Lee, 2010). Therefore by experiencing the product’s aesthetic aspects
hedonically, motivated individuals might be encouraged to learn more about the
product. This is in line with Skinner’s (1953) Dichotomy of Intrinsic-Extrinsic
motivations, where intrinsic motivations were introduced as “a task being
interesting” (cited in Ryan and Deici, 2000, p.56). In intrinsically motivated
activities, the reward is the activity itself. Intrinsic motivations are “the doing of an
activity for its inherent satisfaction rather than for some separable consequence”
(Skinner 1953 cited in Ryan and Deici, 2000, p.56). Hull (1943) relates intrinsic
motivation to learning theories and explains it as the satisfaction gained from intrinsic
engagement; the intrinsic activities thus result in the satisfaction of inner
psychological needs (Hull, 1943) such as hedonic needs. According to Hull (1943), by
increasing the product’s hedonic value, consumers will be intrinsically motivated,
resulting in increased satisfaction and better learning. The positive feeling
experienced as well as the feeling of enjoyment, might facilitate consumer
understanding of the product, and influence their attitude. Therefore:
142
H3a: Hedonic innovativeness has a positive influence on consumers’
comprehension of RNPs.
H3b: Hedonic innovativeness has a positive influence on consumers’ attitude
towards RNPs.
Functional needs are an individual’s drive to search for products that solve
consumption-related problems. Functional reasons to adopt a product are mentioned
in several studies such as Hirschman (1984), Venkatraman and Price (1991), Griffin
(1994) and Voss et al. (2003). Hirschman (1984) and Venkatraman and Price (1991)
explained how some innovative consumers are attracted to functional new products.
Babin et al. (1994) and Voss et al. (2003) proposed a utilitarian shopping value that
motivates the consumer to buy innovative products. RNPs are products, which are
designed to solve consumers’ existing problems, which cannot be solved using
existing products. They allow consumers to perform tasks they are unable to perform
with existing products (Alexander et al., 2008). According to the Dichotomy of
Intrinsic-Extrinsic Motivations Theory (Skinner, 1953), functional needs are satisfied
when an activity is undertaken in order to attain some separable outcome. It can be a
feeling of accomplishment (Vallerand, 1997), such as when an individual understands
a new concept, resulting in a consumer’s problem being solved. Therefore
highlighting the product’s functionality, by using various stimuli via product
promotion, is influential on consumers’ understanding and attitude towards RNPs.
Thus:
H4a: Functional innovativeness has a positive influence on consumers’
comprehension of RNPs.
H4b: Functional innovativeness has a positive influence on consumers’ attitude
towards RNPs.
Social needs are related to other individuals and include concepts such as status
and support. In order to fulfil these needs, the presence and actions of other
individuals are required (Vandecasteele and Geuens, 2010). In the product innovation
field, Brown and Venkatesh (2005) and Foxall (1988) explained how consumers also
want to buy products to impress others and advance their social status. Researchers
have also emphasized the importance of the social components within consumer
143
innovativeness (Venkatraman and Price, 1990; Rogers, 2003; Weijters and Roehrich,
2004). By possessing new products, consumers can build a clear identity (Tian et al.,
2001; Tian and McKenzie, 2001) and this possession is a socially acceptable way of
making a unique impression (Simonson and Nowlis, 2000). When a product is not yet
available (RNP), consumers are not able to imagine possessing it in near future, and
do not have the opportunity to impress the others. As a result, they might not perceive
possessing the RNP as a way to improve their social status; thus possessing it might
have an inverted effect on their learning and behaviour. Therefore:
H5a: Social innovativeness has a negative influence on consumers’
comprehension of RNPs.
H5b: Social innovativeness has a negative influence on consumers’ attitude
towards RNPs.
Cognitive needs and stimulation also influence motivation and behaviour.
Consumers with a cognitive need enjoy being involved in mentally demanding
activities, such as reading, and/or deeply processing information (Vandecasteele and
Geuens, 2010). Venkatraman and Price (1990, p.294) defined cognitive
innovativeness in their study as “the desire for new experiences with the objective of
stimulating the mind”. Other closely related concepts include exploratory information
seeking defined as providing stimulation for the mind (Baumgartner and Steenkamp,
1996, cited in Roehrich, 2004, p.674), and cognitive motivations within intrinsic
motivations, also known as motivations to know (Vallerand, 1997 cited in
Vandecasteele and Geuens, 2010, p.4). When dealing with RNPs, consumers have
limited existing cognitive structures for the products (Feiereisen et al., 2008). This is
due to the level of newness of RNPs (as consumers do not have any existing
knowledge base of the product and product category), and the level of uncertainty
they experience. Various learning techniques that stimulate a consumer’s cognitive
processing can be employed to facilitate their comprehension of RNPs, such as mental
simulation and analogies. On the other hand, promoting RNPs using various
information presentation formats, such as text, imagery and 3D pictures, can help
individuals to understand the information better, thus reducing uncertainty.
Furthermore, due to the highly innovative nature of RNPs, cognitively motivated
144
individuals may find this new experience stimulating and as a result, put more effort
into understanding the product, and find the experience more enjoyable. Therefore:
H6a: Cognitive innovativeness has a positive influence on consumers’
comprehension of RNPs.
H6b: Cognitive innovativeness has a positive influence on consumers’ attitude
towards RNPs.
4.3 Methodology
This study adopted a quantitative approach, gathering data via online surveys. A
sample of 300 participants was recruited via an online platform namely Mturk6. The
first step was to select an appropriate RNP, and then a website was designed for the
purpose of product presentation. An online questionnaire was designed and data was
analysed using SPSS.
4.3.1. Product selection
For the purpose of this study, nine new products were initially selected. These
products were ones with information in the form of text and pictures freely available
online, at the time of selection. The products claimed to be very new and innovative
and relatively easy to understand. The nine products were tested using an online
questionnaire to confirm their status as RNPs in the population of interest. For this
purpose, perceived product newness was measured (Appendix 2). The scale is a
combination of Gregan-Paxton et al.’s (2002) definition of RNPs and Hoeffler’s
(2003) framework that was used in RNP related research on new product evaluation
(e.g. Alexander et al., 2008). As a result, Bio-robot Fridge was selected as a product
for this study, which was perceived as new, understandable and likeable.
4.3.2. Participants
Data was collected using an online survey tool of iSurvey7. Existing scales,
extracted from related literature were employed in order to evaluate the variables in
6 www.Mturk.com7 www.isurvey.soton.ac.uk
145
the model. A website was designed presenting the information of the product with the
enclosure of 2D and 3D pictures.
Data was collected via Mturk, an online crowdsourcing Internet marketplace. The
sample was limited to adult participants from the USA. A sample of 300 participants
was collected. As SEM was employed, it was important to determine the sample size
according to this method. When using SEM, five factors need to be considered
namely model complexity, estimation technique, multivariate normality, missing data
and average error variance of indicators. In order to prevent data deviation from
multivariate normality, an accepted ratio of 15 respondents for each parameter
estimated in our model was followed (Hair et al., 2010). The most common SEM
estimation procedure is maximum likelihood estimation (MLE), which recommends a
minimum sample size of 50 respondents. However, it is suggested that a model
becomes more sensitive when a large sample (more than 400 respondents) are
recruited; therefore a sample size of 100 to 400 is suggested. Model complexity refers
to the fact that simpler models can be tested with a smaller sample. Sample size is
important as it provides stability and the general preference is to have a larger sample
size within the recommended range (Lei and Wu, 2007).
In dealing with Average Error Variance of Indicators, research supports the use of
communality as a way to approach the sample size issue. Communality refers to “the
average amount of variation among the measured variables” identified by the model
(Hair et al., 2010, p.636). The communality is calculated as the square of
standardized construct loadings. The larger the sample size, the smaller the
communalities.
“Models containing constructs with communalities less than 0.5 (i.e.
standardized loading estimates less than 0.7) require larger sample size for
convergence and model stability” (Hair et al., 2010, p.636).
Considering all these factors, a sample size of 300 was deemed suitable for this study.
4.3.3. Questionnaire development
A questionnaire was developed using existing scales.
146
Innate Innovativeness
To measure innate innovativeness, Vandecasteele and Geuens’ (2010) Motivated
Consumer Innovativeness (MCI) was used. MCI takes into account the three different
motivations of innovative consumers considered in this study, as well as a fourth
dimension not central to the study. The 20-item scale measures the dimensions of
Functional Motivated Consumer Innovativeness (fMCI), Hedonically Motivated
Consumer Innovativeness (hMCI), Socially Motivated Consumer Innovativeness
(sMCI) and Cognitive Motivated Consumer Innovativeness (cMCI) (Vandecasteele
and Geuens, 2010). Vandecasteele and Guenes (2010) reported satisfactory Cronbach
alphas for each group of items that indicated their reliability (αsMCI=.929; αfMCI=.907;
αhMCI=.928; αcMCI=.902). In this study, the MCI scale was presented via a 7-point
Likert scale. The full scale can be found in Appendix 21.
Comprehension
In order to measure comprehension a 6-item 7-point scale developed by
Feiereisen (2008) was used. The scale itself is a 2-item semantic differential scale by
(Phillips, 2000) and a 4-item 7-point Likert scale by Moreau et al. (2001). The KMO
and Bartlett’s test both support the combination of the scale items as appropriate
(Feiereisen et al., 2008). The scale can be found in Appendix 21.
Attitude
In order to measure attitude, a 10-item scale developed by Voss et al. (2003) was
used. The scale was deemed reliable as reported by Fornell and Larcker (1981); it was
measured by a 7-point semantic differential scale and is presented in Appendix 21.
Purchase Intention
To measure purchase intention, a 4-item 7-point Likert scale adopted from Moon
et al. (2008) was used. A Cronbach alpha of 0.86 reported by Moon et al. (2008)
supported the scale’s reliability. The scale can be found in Appendix 21.
147
4.4 Results
4.4.1. Measurement development and assessment
All constructs were measured using 7-point scales where 1 represented “strongly
disagree” and 7 “strongly agree”. All the scales were modified to fit the purpose of
this study. A Confirmatory Factor Analysis was performed to confirm the suitability
of the scales. The initial model fit was not acceptable (χ2=(545)=1894.98 p=.000;
CFI= .862; GFI=.69; AGFI= .649; RMSEA=.091; Pclose=.000), therefore the model
was examined and the items with factor loadings less than 0.5 and with the biggest
modification indices were removed.
Table 11: final models
Measure CFA Model SEM modelChi-square statistics (x2) 609.77 736.98Degree of freedom (Marks) 329 334Probability < 0.01 < 0.01CMIN/DF 1.85 2.20Root mean square error of approximation (RMSEA)
.05 .059
90% confident interval of RMSEA .047-.060 .057-0.70GFI .87 .86AGFI .85 .83Comparative fit index (CFI) .96 .94Incremental fit index (IFI) .96 .94Normed fit index (NFI) .92 .90Pclose .19 .024
The final model, as demonstrated in Table 11, indicates a good fit for the
construct measurement model. The final models are illustrated in Appendix 22.
Furthermore, the factor loadings are all higher than 0.70, indicating that all items
loaded significantly onto their respective constructs; this supports the convergent
validity of the measurement items. A complete list of the final measurement items and
their factor loading are presented in Table 12.
Table 12: Confirmatory factor analysis: loadings and t-value
Construct items Loading t-valueFunctional Innovativeness Functional Innovativeness (Infun5) .84Functional Innovativeness (Infun4) .77 15.15Functional Innovativeness (Infun3) .83 17.04
148
Functional Innovativeness (Infun2) .84 17.23Hedonic Innovativeness Hedonic Innovativeness (Inhed5) .80Hedonic Innovativeness (Inhed4) .90 18.35Hedonic Innovativeness (Inhed3) .90 18.40Hedonic Innovativeness (Inhed2) .75 14.40Hedonic Innovativeness (Inhed1) .79 15.20Social Innovativeness Social Innovativeness (Insoc1) .91Social Innovativeness (Insoc2) .92 25.61Social Innovativeness (Insoc3) .91 25.13Social Innovativeness (Insoc5) .70 14.89Cognitive Innovativeness Cognitive Innovativeness (Incog1) .78 16.61Cognitive Innovativeness (Incog2) .83 18.46Cognitive Innovativeness (Incog3) .90 21.36Cognitive Innovativeness (Incog4) .92 21.86Cognitive Innovativeness (Incog5) .85Comprehension Comprehension (Comp1C) .92Comprehension (Comp2C) .93 26.74Comprehension (Comp3) .77 17.98Comprehension (Comp4) .75 17.10Purchase Intention Purchase Intention (PI1) .94Purchase Intention (PI2) .77 18.81Purchase Intention (PI3) .98 36.37Attitude Attitude (Att1) .86Attitude (Att4R) .80 16.73Attitude (Att5) .90 20.17Attitude (Att7) .71 13.98
The internal consistency and reliability of the scales were also tested by
calculating the Cronbach’s alpha, ranging from .88 to .93, which was considered
satisfactory. The composite reliability (CR) for each construct was calculated along
with Average Variance Extracted (AVE), Maximum Shared Variance (MSV) and
Average Shared Variance (ASV) (Fornell and Larcker, 1981; Hair et al., 2010). As
shown in Table 3, all the conditions necessary to demonstrate reliability, convergent
and discriminant validity, were met.
Table 13: Descriptive statistics
CR(>0.7) AVE(>0.5)(<CR) MSV ASV
(<AVE) 1 2 3 4 5 6 7 Mean SD α
1:Functional Innovativeness 0.892 0.675 0.419 0.219 0.821 4.7 1.22 .89
2:Hedonic Innovativeness 0.916 0.686 0.581 0.295 0.647 0.828 4.54 1.34 .91
149
3:Social Innovativeness 0.922 0.748 0.581 0.227 0.491 0.762 0.865 3.8 1.48 .92
4:Cognitive Innovativeness 0.933 0.738 0.342 0.221 0.573 0.580 0.585 0.859 3.86 1.36 .93
5: Comprehension 0.872 0.693 0.331 0.099 0.151 0.196 0.064 0.107 0.833 5.11 1.37 .91
6:Purchase Intention 0.930 0.816 0.332 0.206 0.454 0.421 0.359 0.456 0.431 0.904 3.2 1.55 .92
7:Attitude 0.891 0.673 0.332 0.191 0.316 0.465 0.257 0.317 0.575 0.576 0.820 4.95 1.36 .88
4.4.2. Hypotheses testing
In order to test the hypotheses, structural equation modelling (SEM) was
employed. AMOS (Version 20) was used. The goodness of fit for SEM model is
marginally different from CFA model (Table 1). Model fit measurements for the
structural model are (χ2=(334)=736.978 p=.000; CFI= .942; GFI=.857; AGFI= .827;
RMSEA=.059; Pclose=.024). The Chi-square difference test of the structural model
(χ2(334)=736.978) and the construct measurement model (χ2(329) = 609.769) was
performed in order to formally check whether the difference was significant
(Anderson, 1988). The critical chi-square value for 5 df is 15.08 (p<0.1), indicating
that this chi-square value is significant. However, it has been argued that the chi-
square test is very sensitive to multivariate normality and sample size (Bentler and
Bonnet, 1980; Jöreskog and Long, 1993; McIntosh, 2006). Due to the restrictiveness
of chi-square, alternative indices were used to access model fit. One example is
Wheaton et al., (1977) relative/normed chi-square. The relative/normed chi-square
showing χ2/df=2.2 which is less than 5, is therefore acceptable (Wheaton et al., 1977).
Another index is NFI, which should be larger than 0.9; this applies to this study
(NFI=0.91), therefore the model is acceptable for hypotheses testing.
According to Hair et al., (2010), a p-value of less than 0.05 is considered as
significant. Therefore looking at Table 14, most of the hypotheses are supported. H1a,b
which indicate a positive influence of comprehension towards attitude is supported
(β=.509, p<.001); as well as the positive influence of comprehension towards
purchase intention (β=.162, p<.05). The positive impact of attitude towards purchase
intention is supported (β=.613, p<.001) which supports H2. Hedonic innovativeness
seems to have a positive effect upon participant comprehension (β=.309, p<.01) and
attitude (β=.458, p<0.001); however functional innovativeness does not influence
participant comprehension (β=.043, n.s.) or attitude (β=-.027, n.s.), thus H4a,b are not
150
supported. Results also support H5a that social innovativeness has a negative impact
on consumer comprehension (β=-.175, p<0.05), but it does not support H5b, as social
innovativeness does not have a negative effect on attitude (β=-.147, n.s.). Cognitive
innovativeness has a positive significant effect on an individual’s attitude (β=.136,
p<.05), therefore H6b is supported; but there is no significant effect observed between
cognitive innovativeness and comprehension (β=.029, n.s.), which rejects H6a.
Table 14: Test of hypotheses
Hypotheses Unstandardized Estimates
S.E P value
H1a (+): Comprehension Attitude .509 .061 .001H1b (+): Comprehension Purchase Intention .162 .083 .050H2 (+): Attitude Purchase Intention .613 .077 .001H3a (+): Hedonic Innovativeness Comprehension .309 .134 .010H3b (+): Hedonic Innovativeness Attitude .458 .106 .001H4a (+): Functional Innovativeness Comprehension .043 .094 .663H4b (+): Functional Innovativeness Attitude -.027 .081 .735H5a (-): Social Innovativeness Comprehension -.175 .094 .050H5b (-): Social Innovativeness Attitude -.147 .078 .058H6a (+): Cognitive Innovativeness Comprehension .029 .084 .727H6b (+): Cognitive Innovativeness Attitude .136 .068 .047
To summarize, the results indicate a positive relationship between hedonic
innovativeness and comprehension and attitude towards RNPs. A negative
relationship between social innovativeness and comprehension is observed; whilst
cognitive innovativeness showed a positive relationship towards consumer attitude.
There was no significant relationship between social innovativeness and attitude,
cognitive innovativeness and comprehension, and functional innovativeness and
learning and attitude towards RNP.
4.5 General Discussion and Conclusion
Within the innovation adoption literature, there is evidence suggesting
comprehension leads to attitude formation (e.g. Goldsmith and Newell, 1997; Pagani,
151
2007). In addition, the relationship between comprehension, attitude formation and
purchase intention is supported by persuasion theories (Ratneshwar and Chaiken,
1991). The result of this study confirms the positive influence of comprehension
towards attitude and purchase intention within the RNP context, which is a form of
innovation. Furthermore, the innovation decision-process explains a relationship
between attitude and purchase intention within the innovation context (Rogers, 1995),
which is yet again supported within the RNP domain, by the results of this study.
Turning to the innovativeness hypotheses, the result from the structural model
provides evidence of a direct positive relationship between hedonic innovativeness
and the comprehension and attitude of individuals towards RNP. This finding is in
line with previous studies stating a positive but weak relationship between innate
innovativeness and adoption behaviour (e.g. Goldsmith et al., 1995; Citrin et al.,
2000; Im et al., 2007). This weak relationship found in previous studies can be an
indication of the different directions (positive or negative) each category of innate
innovativeness takes, in relation to comprehension and attitude, as evident in this
study. The results also indicate the possibility of using 3D pictures as a variable
positively influencing the product’s hedonic aspect, which can be investigated further.
This could result in a more enjoyable experience for consumers, hence a more
positive attitude; this is in line with previous findings of Lee et al. (2010). Another
important finding was the negative relationship between social innovativeness and
comprehension. As predicted, socially innovative consumers were not motivated
enough to learn about the product, a point supported by the results of this study. This
can be due to the fact that consumers cannot imagine possessing the product in the
near future, as it is not yet in the marketplace.
Functional reasons to adopt a product are mentioned in several studies such as
Hirschman (1984), Venkatraman and Price (1990), Babin et al. (1994) and Voss et al.
(2003). Hirschman (1984) and Venkatraman and Price (1990) explained how some
innovative consumers are attracted to functional new products. Babin et al. (1994) and
Voss et al. (2003) proposed a utilitarian shopping value that motivates consumers to
buy innovative products. RNPs are products, which are designed to solve consumers
existing problems, which cannot be solved using existing products. They allow
consumers to perform tasks they are unable to perform with existing products
152
(Alexander et al., 2008). According to the Dichotomy of Intrinsic-Extrinsic
Motivations Theory (Skinner, 1953), functional needs are satisfied when an activity is
undertaken in order to attain some separable outcome. It can be a feeling of
accomplishment (Vallerand, 1997), such as when an individual understands a new
concept, resulting in a consumer’s problem being solved.
There was no significant relationship between functional innovativeness and
consumer learning and attitude towards RNP. Although this was supported in
previous studies, that by satisfying individuals’ functional needs, they become
motivated to learn and approach innovative products (e.g. Babin et al. 1994; Voss et
al. 2003). One reason for the inconsistency of this study’s findings against the
previous body of literature could be that there is relatively little information about the
RNP, and it does not allow individuals to fully understand the product’s functionality;
or on the contrary, participants already know the functionality of the selected RNP
(which is very close to a conventional fridge). For the cognitive aspect of
innovativeness, the literature states that cognitively motivated individuals have a
desire for new experiences (Venkatraman and Price, 1990). They have a tendency for
exploratory information seeking (Baumgartner and Steenkamp, 1996). The result of
this study supports a positive impact of cognitive innovativeness on an individual’s
attitude, which may be the result of their positive attitude towards a new experience;
this is indicated within the innovation literature (Rogers, 2003; Feiereisen et al.,
2008). However, there was no evidence is this study to support the hypothesis that
cognitive innovativeness leads to better comprehension. This might be due to the lack
of information, which does not stimulate consumers seeking to satisfy their cognitive
needs as consumers have limited existing cognitive structures for the products
(Feiereisen et al., 2008). This limited cognitive structure can be due to the level of
newness of RNPs and the level of uncertainty that consumer experiences.
Overall, this study provides key insights into the relationship between various
innate innovativeness needs of hedonic, functional, cognitive and social, and an
individual’s learning and attitude towards RNPs. The results indicate the difference in
direction (positive or negative) within the relationship between innate innovativeness
and consumer comprehension and attitude within the RNP context. This adds to the
theories on innovation-adoption and diffusion of innovation by looking into each
153
aspect of innate innovativeness individually. It also adds to the body of knowledge
concerning RNP promotion, by providing evidence that the use of various imagery
(2D and 3D) might result in a positive relationship between hedonic innovativeness
and consumer comprehension and attitude.
4.6 Managerial Implications
The key question for practitioners and managers is how to present RNP related
information, in order to transfer knowledge in the most effective way to the target
customer. The reason why this is needed is clearly articulated in the literature; more
learning is required to understand RNPs (e.g. Gregan-Paxton et al., 2002; Feiereisen
et al., 2008) and the consumer learns about RNPs at the time of product evaluation
(Hoeffler, 2003). The result of this study provides evidence on how innovative
individuals, motivated by various factors, respond to RNP promotion; therefore
practitioners should take this into account while promoting RNPs online. For
example, to target consumers motivated by social needs, practitioners might observe a
negative effect on consumer learning towards RNPs, which is not useful in an
introductory phase for new products. Therefore, practitioners might benefit from
using various techniques such as celebrity endorsement, in order to encourage
individuals to learn about RNPs.
When targeting consumers with more hedonic needs, it might be useful to
consider highlighting the product’s hedonic aspects. In order to attract consumers with
more functional needs, as there is no evidence explaining how to encourage this
group, it might be useful to provide as much information as possible in various forms
to facilitate their understanding about the functionality of RNPs. For a more
cognitively motivated group of consumers, practitioners should consider presenting
RNPs as not a very new concept. There are also various learning strategies, such as
mental imagery or analogies, as explained by Feiereisen et al. (2008), that might
facilitate cognitively motivated consumers by linking their limited knowledge about
RNPs to their existing knowledge base of similar products or domains; this would
facilitate their learning and reduce the uncertainty they experience. Furthermore, the
result of this study provides evidence that learning is key to the intention to adopt
RNPs. If consumers learn about RNPs, it positively influences both their attitude and
purchase intention towards these products.
154
4.7 Limitations and Direction for Future Research
Although the research contributed to both consumer innovativeness and consumer
behaviour literature, the survey methodology has its limitations, which provide
recommendations for future research. The first issue is generalizability. The study is
based on one RNP (Gel Fridge), which is within a specific domain (Kitchen ware
electronic). The study needs to be replicated using various RNPs within dissimilar
domains. Also, the study was carried out with limited product presentation techniques.
Other product presentation formats (such as videos, chat boxes etc.) could be
examined to understand the influence they have on consumer learning and behaviour
towards RNPs.
Another area that can be investigated further is targeting isolated innovative
consumer groups, while inserting various attributes facilitating consumer
comprehension and online behaviour; for instance, using anthropomorphic attributes
(such as different types of avatars) and analysing if this has any influence on socially
motivated consumers to learn about RNPs better. Furthermore, this study is not
exhaustive in its scope. Further research should look at other factors such as other
individuals’ characteristics, learning style and motivational sources. This would
provide further insight into developing an integrated framework of consumer
innovativeness in consumer behaviour contexts, within the RNP domain.
155
Chapter 5: Conclusion
This thesis is concerned with three aspects, which are examined further via three
separate papers. Firstly, it is concerned with various RNP presentation elements and
how these characteristics influence consumers’ comprehension, attitude and purchase
intention for RNPs. Secondly, it investigates consumers further by examining whether
there is any dissimilarity between different groups of innovative consumers in
understanding RNPs and their attitude towards RNPs. Thirdly, the thesis discusses the
element of anthropomorphism as a presentation element, and the impact it has upon
consumer RNP comprehension, attitude and purchase intention, as well as an
individual’s perception of anthropomorphism when presented through various
formats. This chapter summarises the findings of each paper as well as their
individual contributions, limitations and recommendations for future research.
5.1 Review of Findings
The conclusion discusses the findings from each of the three papers in turn. Paper
1 examined the impact of telepresence (vividness and interactivity) and
anthropomorphic attributes on consumers’ comprehension and online behaviour;
Paper 2 investigates the insertion of various forms of anthropomorphism and the way
this attribute is perceived and influences consumers’ comprehension and online
behaviour; and finally Paper 3 investigated consumer innovativeness and the effect it
has upon consumer learning and online behaviour. Table 5.1 summarises each paper’s
results, linking them to the objectives of this thesis. Each conclusion is expanded
upon in the form of contributions, in Sections 5.2.1, 5.2.2 and 5.2.3. Un-supported
hypotheses are explained in section 5.2.4.
156
Table 15: Review of Findings
Study Objectives Findings
Paper 1
1. To identify the effect of various presentation formats (Vividness, interactivity and anthropomorphic attributes) on consumer responses to RNPs.
1.1 Vividness has no effect on consumers’ learning, attitude and purchase intention in the context of RNPs.
1.2 Interactivity has a significant positive influence upon consumers’ learning, attitude and intention towards RNPs, but it differs across RNPs.
1.3 There is no perceived difference across conditions for Anthropomorphism in relation to consumers’ learning, attitude and purchase intention towards RNPs.
Paper 2
2. To examine whether the insertion of various types of anthropomorphic attributes influence consumers perceived mindless and mindful anthropomorphism while learning about RNPs.
3. To determine the relationship between consumers’ perceived mindless and mindful anthropomorphism and consumers response towards RNPs.
2.1 Inclusion of a human-like avatar, containing all four types of social cues (human voice, language, interactivity, social role), results in an increased perceived overall anthropomorphism; this is the direct result of a significant perceived mindless anthropomorphism.
3.1 Perceived anthropomorphism has a significant positive influence on consumers’ learning, attitude and purchase intention towards RNPs.
Paper 3
4. To examine the impact of consumer innovativeness (functional, hedonic, social and cognitive) on consumers’ comprehension and attitude towards RNPs.
4.1 Functional innovativeness does not impact on consumers’ comprehension and attitude towards RNPs.
4.2 Hedonic innovativeness has a positive influence on consumers’ comprehension and attitude towards RNPs.
4.3 Social innovativeness has a negative influence on consumers’ comprehension towards RNPs and no impact on consumers’ attitude.
4.4 Cognitive innovativeness has a positive influence on consumers’ attitude towards RNPs and no impact on consumers’ comprehension.
157
5.2 Contributions, Limitations and Future Research
This section first discusses each paper’s contributions towards the body of
literature followed by a generic contribution of three papers. Next, limitations and
suggestions for future lines of research are discussed for each paper.
5.2.1. Contribution; Paper 1- Telepresence
Vividness
Paper 1 examined the influence of presenting information in a multimedia format,
which resulted in higher vividness towards consumers’ comprehension and online
behaviour within the RNP context. Vividness is one of the characteristics enabling
telepresence in a CME. Inserting multimodal information is supported via the
Cognitive Theory of Multimedia Learning (Mayer, 1997), which explains the
effectiveness of presenting information in a multimedia format for individuals’
comprehension. Within various contexts, vividness is supported to improve consumer
learning (e.g. Paivio, 1971; Alesandrini, 1982; Rayport and Jaworski, 2001; Zhang et
al., 2006). In examining consumer online behaviour, vividness was found to create a
similar experience to direct product experience, which can lead to a more positive
attitude (Klein, 2003). This is supported within various contexts such as CME (Street
et al., 1997; Tran, 2010). The findings of Paper 1 appear to contradict previous
findings by exposing the ineffectiveness of inserting vividness in order to improve
consumers’ comprehension, attitude and purchase intention within the context of
RNP. The findings raise questions as to whether multimodality enhances consumer’s
comprehension when dealing with a very new concept. This may be due to the unique
nature of RNPs. For instance, RNPs are high-involvement products (Hoeffler, 2003,
Feiereisen et al., 2008), therefore individuals need to pay attention to the message
content in order to learn about these products (central route) (Petty and Cacioppo,
1981; Petty and Cacioppo, 1986). This may result into individuals not being interested
in multisensory information presentation, and instead prefer straightforward text, that
simply presents information.
Another characteristic of RNPs, similar to other products, is the level of utilitarian
or hedonic nature. This product attribute can have an effect on consumers’ behaviour
towards RNPs similar to other product categories (e.g. Okada 2005, Feiereisen 2009,
158
Choi et al. 2014). A preliminary study by the author indicated that product nature is
influential in consumer comprehension and attitude as separate variables, but it does
not have a significant effect on consumer purchase intention. When product nature
was investigated further, it was apparent that both products used in this study were
more on the utilitarian side of the spectrum. However, this needs further investigation
and can be a base for future studies. Another possible explanation is connected with
Davies et al.’s (1989) Technology Acceptance Model (TAM); an individual’s
perception of a RNP’s usefulness and ease of use of RNPs influences online
behaviour. These factors are amongst the attributes identified by scholars as new
product characteristics that need to be considered while studying a new concept (e.g.
Rogers, 1995). Analysing the RNP related attributes might therefore give the answer
to this question of why vividness did not make any difference on an individual’s
comprehension, attitude and purchase intention.
Examining consumer characteristics, factors such as consumer knowledge and
ability could be influential in consumer perceived vividness. According to the
Cognitive Theory of Multimedia Learning (Mayer and Moreno, 1998), presenting
information in a multimedia format is more important for low-knowledge than high-
knowledge learners, and for high-spatial rather than low-spatial learners. Furthermore,
in the Cognitive-Affective Theory of Learning with media (Moreno, 2005), it is
emphasized that differences in learners’ prior knowledge and abilities may affect how
much is learnt from a specific media (Kalyuga et al., 2003; Moreno, 2004; Moreno
and Durán, 2004). Socio-demographic factors could also have an influence on
consumer online behaviour (e.g. gender). For instance, a preliminary study by the
author showed that overall, males express a significantly more positive attitude
towards product 2 than females. This finding is explained further in section 5.2.4.
Interactivity
Paper 1 investigated the impact of interactivity on individuals’ learning, attitude,
and purchase intention towards RNPs. Interactivity is a factor enabling telepresence,
and is supported as an attribute improving consumers’ learning and online behaviour
within various contexts (e.g. Brandt, 1997; Schlosser et al., 2003; Fiore et al., 2005;
Park et al., 2005; Zhang et al., 2006; Pantano and Naccarato, 2010). It is also
159
supported in different learning frameworks (e.g. Bandura, 1977; Brandt, 1997; Zhang
et al., 2006). Inserting interactivity resulted in a consumer experience that was as
close to a direct product experience, the latter proving to be the ideal method for a
consumer to learn about a product (Klein, 2003). Paper 1 indicates that interactivity
can be an influential attribute on consumers’ comprehension, attitude and purchase
intention within the RNP context, but this cannot be generalized for all the products
within the RNP category, as it differes across the two RNPs used in the study. The
impact of interactivity upon consumer comprehension and online behaviour can thus
be linked to new product attributes. Scholars emphasized the importance of
considering new product attributes, while studying about a new concept (e.g. Davis et
al., 1989; Rogers, 1995). Furthermore, the utilitarian and hedonic nature of products
can also be an influential factor as explained in ‘vividness’ section above. Another
justification could be that product 1 belonged to the group where participants
indicated the product to be significantly more understandable and likable; whereas
product 2 belonged to the group of products which participants were willing to change
their behaviour, in order to interact with them (according to product selection process
and the perceived product newness scale; Alexander et al. 2008).
Another element explained above is consumer characteristics, which is also a
source of debate within the Diffusion of Innovation theory. In order to enrich the
discussion above, the related literature needs to be explained. Rogers (2003) was one
of the main scholars who continued to criticise the Diffusion of Innovation theory
(Rogers 1995), specifically in relation to product attributes and second consumer
characteristics. Various scholars looked into the relationship between consumer
characteristics and their adoption of innovation. Overall, it appears that males are
more innovative than females; there is a negative relationship between age and
consumer innovativeness; and a positive relationship between education/income and
consumer innovativeness (Rogers, 2003; Weijters and Roehrich, 2004; Tellis et al.,
2009). Hirschman (1980) looked into creativity and innovativeness and realized that,
more creative individuals understand the concept of an innovation better, which then
facilitates the innovation adoption (Rogers 2003; Hirschman 1980). Studies revealed a
positive relationship between consumer innovativeness and consumers’ values (such
as self-enhancement and openness to change) and a negative relationship between
empathy and conservatism (e.g. Daghfous et al., 1999; Rogers, 2003; Weijters and
160
Roehrich, 2004). Hence consumer attributes, alongside product attributes, such as
perceived ease of use and perceived usefulness as explained in Davies’s TAM, or
hedonic and utilitarian product nature can be the reasons of the inconsistency
observed in the manipulation of Interactivity.
Anthropomorphic Attributes
This paper also examined the impact of anthropomorphic attributes on consumer
learning, attitude, and purchase intention towards RNPs. The result indicated that
individuals did not perceive a higher anthropomorphism when interacting with a high
level of anthropomorphism; therefore, the impact of this attribute is not further
analysed for consumer learning, attitude and purchase intention. There are two
justifications, in addition to the elements of product attributes and consumer
characteristics as main criticism of Diffusion of Innovation theory, explained above.
The first justification is related to individual’s percieved anhropomorphism. Based on
consumer behaviour literature, there are four social cues that elicite social responses
which are human voice, language, interactivity and social role (Nass and Steuer
1993). Studies have demonstrated that people react to computers as social actors (e.g
Reeves and Nass, 1996) as according to Social Response Theory, people
automatically tend to treat computer technologies as social actors. Nass and Moon
(2000, p.97) explain ‘‘the more computers present characteristics that are associated
with humans, the more likely they are to elicit social behaviour”. Hence, any level of
social cues presented within a CME can provoke a level of anthropomorphism
perceived by individuals.
In this study, it is possible that participants in the low condition (no avatar),
perceived the website as anthropomorphic considering various social cues presented
(such as language); whereas in the high condition (with avatar), they evaluated the
avatar as the element of humanness and ignored the website. As a result, the avatar
may not be perceived as anthropomorphic. Looking back at literature, avatars which
are perceived as anthropomorphic create an expectation of sociability, and stimulate
judgments reserved for social entities like credibility and homophily (Bente et al.,
2008). These expectations can lead to more negative attributions, when the
expectations are not met (Reeves and Nass, 1996; Nowak and Biocca, 2003; Nowak,
2004). Studies have also revealed that although individuals prefer a more realistic,
161
human-like avatar, there is a higher risk of disappointment as individuals have higher
expectations from humanlike avatars (Keeling 2010). Hence, in the high condition,
individuals might have had an instant high expectation when learning about the
existence of an interactive, human-like avatar, which resulted into their
disappointment and lower level of anthropomorphism. This could be an interesting
area for future research.
The second justification is the measurement scale used in this study. In most
studies, in order to manipulate the level of anthropomorphism, the same level social
cue is compared (use of different forms of avatars) (e.g. McGoldrick et al. 2008;
Keeling et al. 2010). There is no recent study that compares different levels of social
cues such as a condition including interactivity in comparison to a condition including
human voice. In this study various conditions include various elements of social cues.
However, this study recruited an existing perceived anthropomorphism scale, which
might not be the most suitable for the purpose of the study. As developing a new
measurement scale was outside the scope of the present study, it is suggested as an
area for future research.
To summarise, the findings of this study contradict previous academic literature.
Not all characteristics enabling telepresence (vividness and interactivity) proved to be
influential on an individual’s comprehension of and online behaviour towards RNPs.
Vividness attributes did not improve an individual’s comprehension, attitude or
purchase intention towards RNPs; the influence of interactivity on comprehension,
attitude and purchase intention across RNPs was inconsistent; furthermore,
individuals didn't perceive a significantly different level of anthropomorphism for
various conditions. Anthropomorphism was therefore examined further in Paper 2.
5.2.2. Contribution; Paper 2 - Anthropomorphism
Paper 2 investigated the impact of anthropomorphic attributes on consumers’
perceived anthropomorphism, comprehension, attitude and purchase intention. The
study explored mindful and mindless aspects of anthropomorphism, and measured
how each aspect influenced consumer learning and online behaviour. The findings
clearly indicated that the inclusion of a human-like avatar, which provided all four
types of social cues identified within the consumer behaviour literature (Nass and
162
Steuer, 1993) of language, human voice, interactivity and social role, creates a sense
of anthropomorphism amongst observers when learning about RNPs. The results
however illustrated that perceived anthropomorphism while dealing with a human-
like avatar, is a direct result of mindless perceived anthropomorphism and not mindful
perceived anthropomorphism. This adds to the anthropomorphism literature as well as
to Social Response Theory (Moon, 2000) within the area of RNP or innovation.
Previous studies within Social Response Theory in CMEs have been inconsistent in
observing whether human-like avatars can result in an increased sense of
anthropomorphism (e.g. Nowak and Biocca, 2003; Groom et al., 2009). This study
provides empirical support for the influence of social cues in participants’ perceived
anthropomorphism; specifically, it sheds light on how different aspects of mindless
and mindful anthropomorphism influence overall perceived anthropomorphism
individuals experience.
Paper 2’s findings add to the anthropomorphism literature, by emphasizing the
necessity of looking into anthropomorphism as two distinct categories of mindless
and mindful. An individual’s mindless and mindful perception is dissimilar within
these two groups. The results further support findings of Kim and Sundar (2012) by
emphasizing how people intentionally reject the consideration of a website in a human
way, whereas they perceive it mindlessly. The results indicate the importance of
advancing measurement scales for perceived anthropomorphism, in order to include
all types of perceived anthropomorphism within a comprehensive scale. This can
significantly improve the understanding of how and why individuals perceive various
anthropomorphic attributes, whether they are presented as a human-like avatar, or as a
less recognizable mode of interactive content. The second theoretical contribution of
this paper is that it links anthropomorphic literature to the consumer comprehension
and online behaviour literature. When individuals perceive a CME as more
anthropomorphic, they learn more about the information presented; this results in an
improved attitude and purchase intention towards RNPs. The results clearly indicate
that perceived anthropomorphism has a positive relationship with consumers’
learning, attitude and purchase intention towards RNPs, which contributes to
consumer learning and online behaviour research.
163
To summarise, the insertion of a human-like avatar, including all four attributes
identified within Social Response Theory, results in increased perceived
anthropomorphism by individuals, within the context of RNPs. This is the direct result
of perceived mindless anthropomorphism and not perceived mindful
anthropomorphism. The findings contribute to both anthropomorphism and social
response research. Furthermore, perceived anthropomorphism has a positive
relationship with an individual’s comprehension, attitude and purchase intention
towards RNPs, which also adds to the consumer learning and online behaviour
literature.
5.2.3. Contribution; Paper 3 – Consumer innovativeness
Paper 3 examined the influence of four motivational aspects of consumer
innovativeness (hedonic, functional, social, and cognitive) upon consumers’
comprehension, attitude and purchase intention. Previous academic literature supports
the positive but weak relationship between innate innovativeness and consumers’
adoption behaviour (e.g. Goldsmith et al., 1995; Citrin et al., 2000; Im et al., 2007).
This weak relationship can be an indication of the different directions (positive or
negative) each category of innate innovativeness takes towards comprehension and
attitude. Paper 3 attempted to shed light on whether each motivational source impacts
consumer learning and their adoption behaviour differently. The four motivational
sources of hedonic, functional, cognitive and social (Vandecasteele and Geuens,
2010) were examined; however, the findings of Paper 3 were found not to be in line
with previous academic literature.
Hedonic innovativeness is explained as the drive to adopt innovativeness for
hedonic causes such as the appreciation of the product’s newness (Roehrich, 1994).
As RNPs are promoted using 3D, it was predicted that the hedonic value of the RNP
should be more observable, hence resulting in a more enjoyable experience that can
upsurge consumers’ attitude (e.g. Lee, 2010). Furthermore, it was predicted that by
experiencing the aesthetic aspects of the product hedonically, motivated individuals
might be encouraged to learn more about the product, which is in line with Skinners’
(1953) Dichotomy of Intrinsic-Extrinsic Motivations. Paper 3’s findings support
previous research by concluding a significant positive relationship between hedonic
innovativeness, and the comprehension and attitude of individuals towards RNPs.
164
Social innovativeness relates to when individuals adopt an innovation in order to
satisfy their social needs such as status, impressing others, etc. (Vandecasteele and
Geuens, 2010). It is suggested that possessing new products enables consumers to
build a clear identity (Tian et al., 2001; Tian and McKenzie, 2001); this possession is
a socially acceptable way of making a unique impression (Simonson and Nowlis,
2000). However, it was predicted in Paper 3 that due to the unavailability of the RNPs
used, individuals would not be able to satisfy their social needs, which might have an
inverted effect on their learning and behaviour. The findings of Paper 3 support this
proposition by indicating a significant negative relationship between social
innovativeness and comprehension. This finding is contradictory to previous findings
of innate innovativeness and consumers’ adoption behaviour (e.g. Goldsmith et al.,
1995; Citrin et al., 2000; Im et al., 2007). The findings contribute to the innovation
adoption literature by introducing a possible explanation of the weak relationship
observed in previous research, as the social aspect of innate innovativeness takes a
negative direction towards comprehension.
Consumers with a cognitive need, enjoy being involved in mentally demanding
activities like reading and deeply processing information (Vandecasteele and Geuens,
2010). The academic literature is contradictory in understanding the impact cognitive
innovativeness can have on consumers’ comprehension and online behaviour. Some
academics believe that due to the newness of RNPs, consumers have a limited
existing cognitive structure for the products (Feiereisen et al., 2008), and experience a
high level of uncertainty (e.g. Rogers, 1995; Rogers, 2003; Alexander et al., 2008);
as such, they cannot satisfy their cognitive needs. Conversely, due to the highly
innovative nature of RNPs, cognitively motivated individuals may find this new
experience stimulating; therefore they may attempt to learn about RNPs, which can
influence their online behaviour. The findings of Paper 3 elaborate the latter point by
indicating a significant positive relationship between cognitive innovativeness and
attitude towards RNPs. This finding supports previous academic literature indicating
perceived positive online behaviour towards RNPs. For example, according to the
Elaboration Likelihood Model (Petty and Cacioppo, 1981), due to RNPs being high-
involvement products, individuals need to follow the central route in order to
understand RNPs. Individuals will be highly-involved in the process and as a result,
will develop positive attitudes and behavioural intent for the new product, provided
165
they learn product-relevant information (Bettman, 1979) and the benefits of the
product (Lehmann, 1994; Urban et al., 1996). Findings of the paper 3 however
disagree with the comprehension evidence of previous academic literature, as the
results found no evidence to support that cognitively motivated individuals
comprehend RNPs any better than other innovative consumers.
Cognitively innovative individuals have the desire to learn about new experiences
with the objective of being stimulated (Baumgartner and Steenkamp, 1996). They are
motivated to know (Vallerand 1997). However individuals have limited existing
cognitive structure for RNPs due to their level of newness and high uncertainty
(Feiereisen et al. 2008). Although, it is explained within the innovation literature that
cognitively innovative consumers might find learning about RNPs stimulating and
exciting, it is also predicted that they follow a central route and need to not only be
highly involved, but also have a preference of accessing rich text based information
(according to Elaboration Likelihood Model; Petty and Cacioppo, 1981). As text
based information presented about RNPs in this study is limited, due to the
unavailability of information, this can be an explanation of no significant relationship
beteen conginitvely motivated innovators and comprehension of RNPs.
Functional innovativeness relates to satisfying functional needs, which are an
individual’s drive to search for products that solve consumption-related problems
(Vandecasteele and Geuens, 2010). Previous innovation literature highlights the
importance of functional aspects of RNPs by explaining how RNPs are products,
designed to solve consumers existing problems. They allow consumers to perform
tasks, which they are unable to perform with existing products (Alexander et al.,
2008). Functional needs are claimed to be satisfied when an activity is undertaken to
attain a separable outcome, such as a feeling of accomplishment (Vallerand, 1997). If
the functional aspects of RNPs are highlighted, then it can result in improved
comprehension and online behaviour. The findings of paper 3 do not support these
propositions, as there is no significant relationship between functional innovativeness,
and learning of and attitude towards RNPs. One reason may be that there is relatively
little information given about RNPs, thus this lack of information does not allow
individuals to fully understand the product’s functionality. According to the
Dichotomy of Intrinsic-Extrinsic Motivations Theory (Skinner, 1953), functional
166
needs are satisfied when an activity is undertaken in order to achieve a separable
outcome, such as a feeling of accomplishment (Vallerand, 1997) or understanding a
new concept, resulting into individuals’ problem solving. When a cognitively
innovative individual is unable to understand the concept, it may result in them not
being motivated to change their attitude and/or intention towards RNPs. An
alternative explanation is that participants already knew the functionality of the
selected RNP (which was very close to a conventional fridge), therefore they are not
motivated to learn more about the product, hence their lack of motivation results in no
change in attitude or intention towards these products.
To summarise, the findings of Paper 3 contribute to the innovation adoption
literature, by specifying the different direction each motivational innovativeness
source can take, in impacting comprehension and attitude towards RNPs. Where
hedonic motivation aspects of consumer innovativeness positively influence learning,
social innovativeness takes a negative direction towards learning, within the context
of RNPs. Hedonic and cognitive innovativeness were found to positively influence
consumers’ attitude towards RNPs, when functional innovativeness had no impact on
consumers’ learning and attitude.
5.2.4 Un-hypothesized analysis
This section investigates the un-hypothesized statements, which were formed as a
result of unsupported hypotheses. The present research started with paper 1, as the
base study. Hence the un-hypothesized statement 1 was formed after the results of
paper 1 indicated inconsistency between the results of product 1 and product 2.
Therefore, product nature is tested in order to understand the utilitarian or hedonic
nature of the RNPs used in the study. Un-hypothesized statement 2 was formed due
to the majority of hypotheses on paper 1 deemed unsupported. Hence, various socio-
demographic elements are identified within the literature as influential: learning,
online behaviour and innovation adoption (e.g. Rogers, 2003; Weijters and Roehrich,
2004; Tellis et al., 2009). As the sample was divided into nearly half male half
female, it provided a good opportunity for the author to look into gender. Gender was
examined as an un-hypothesized statement based on the evidence within the
innovation and consumer behaviour literature.
167
5.2.4.1. Statement 1- Utilitarian and Hedonic Product Nature
The result of paper 1 indicated some inconsistency between two products. For
example, interactivity happened to be influential on consumers’ comprehension and
attitude on one of the products. One justification for this inconsistency was the
differences in product nature and characteristics. The author looked into the products’
utilitarian and hedonic nature as an un-hypothesized statement, to see if they are
perceived differently. In order to evaluate, participants were asked to assign the
product into one of the categories of utilitarian or hedonic. A brief description of
hedonic and utilitarian products had been given beforehand. The description is a
combination of two statements extracted from studies by Feiereisen (2009) and Dhar
et al. (2000). Feiereisen (2009, p.174) evaluated the product nature by asking
participants to group products based on the explanation below:
“Utilitarian product: A product which is utilitarian is primarily functional and
instrumental, and practical;
Hedonic: A product which IS hedonic is primarily experiential, provides
enjoyment-related benefits, and is able to create feelings” .
Dhar et al. (2000, p.7) described utilitarian and hedonic products as:
Primarily utilitarian is defined as “useful, practical, functional, something that
helps you achieve a goal (e.g. a vacuum cleaner”)
Primarily hedonic is defined as “pleasant and fun, something that is enjoyable
and appeals to your senses (e.g. perfume)”
The final description was designed by combining the narratives above.
Please fit the product (product name) into one of the two categories.
1. Utilitarian product: This product is primarily functional, instrumental and
practical. The product is something that helps you achieve a goal.
168
2. Hedonic product: This product is primarily experiential, provides enjoyment-
related benefits, and is able to create feelings. The product is pleasant, fun and appeals
to your senses.
The result of the questionnaire indicated that out of 250 participants, for product 1
(washing machine in wardrobe), 91.6% voted the product to be utilitarian and only
8.4% voted for hedonic. For product 2, the result indicated that from 250 participants,
90.4% believed the product to be utilitarian and 9.6% agreed on it being hedonic in
nature. Although the result clearly indicates both product are utilitarian in nature,
there is still a possibility that the level of products’ functionality is different from one
another. Furthermore, hybrid product nature was not analysed. A hybrid product is
defined as a product that is hedonic in outcomes, but utilitarian in process. Hybrid
products “... provide hedonic benefits (aesthetic, experiential and enjoyment-related)
but …require consumers to go through a utilitarian process to achieve these benefits”
(Feiereisen 2009 p.23). Hence, this can be an interesting area for future research.
5.2.4.2. Statement 2- Gender
Investigating socio-demographic characteristics of consumers was one of the
possible justifications for unsupported hypotheses and inconsistency in findings, as
explained in previous sections. The author investigated the role of gender for media
richness and interactivity conditions in paper 1, after discovering the literature
supporting the role of gender in consumer behaviour and innovation adoption. Recent
literature demonstrated how understanding the antecedents of consumer intention
towards adoption of an innovative and new product would benefit academics in
understanding consumer adoption behaviour (e.g. Alexander, Lynch and Wang 2008;
Castano et al. 2008). As a result, understanding the adopter’s characteristics as well as
the innovation’s characteristics will be beneficial for both academics and managers.
Adopter characteristics can be divided into demographic and psychographic. Many
studies have considered a wide range of socio-demographic characteristics (Gatignon
& Robertson, 1985; Rogers, 2003; Tornatzky & Klein, 1982). The most common
socio-demographic features, which have been considered in many studies, are
consumers’ age, level of education and income. Other frequently used factors are
household size, gender and family life-cycle. Additionally adopter psychographics,
which are considered more frequently, are innovativeness, opinion leadership, media
169
proneness and involvement (Arts et al. 2011), where innovativeness is examined in
this study as a covariate. From the socio-demographic point of view, information on
gender, income, education and age is collected, but “gender” is examined further.
Gender was an area studied by many academics, having been explored in the
concept of technology adoption and use (e.g. Brosnan 1998; Durndell & Haag, 2002;
King, Bond, & Blandford, 2002; Schottenbauer, Rodriguez, Glass, & Arnkoff, 2004;
Schumacher & Morahan-Martin, 2001; Stephan & El-Ganainy, 2007; Venkatesh &
Morris, 2000; Venkatesh, Morris, & Ackerman, 2000; Whitley, 1997); the results
however have been mixed (Kay, 1992; Rosen & Maguire, 1990). Gender is also an
important attribute in the innovation adoption literature, such as consumer
innovativeness. This study took into consideration “consumer innovativeness” as a
covariate. The results of Tellis et al.’s (2009) study found that a global profile of
innovators across countries can be identified as wealthy, young, mobile, educated and
male. There is evidence in literature pinpointing how male adopt innovative products
more than female (e.g. Morris 1999). The type of product is also influential as Tellis
et al. (2009) discussed how the eagerness to adopt innovation varies by product
category. For example, females are more eager to buy home appliances, cosmetics,
and food and grocery products, whereas males are keener on buying automobiles and
sporting goods. Therefore, in this study, females might be more interested in the
selected RNPs as they are two types of Laundry/Washing machines which can be
categorized as home appliances.
From an attitudinal point of view, there are inconsistent results. In some studies
males appear to be more positive towards computers than females (e,g, Anderson
1987; Schumacher & Morhan-Martin, 2001); whereas in others it is the opposite (e.g.
Loyd, Loyd & Gressard, 1987; Siann, Macleod, Glissov, & Durndell, 1990).
Venkatesh et al. (2000) found that attitude towards technology adoption influenced
men more strongly than women in the US. On the learning side, males and females
learn differently (William 1968). For instance, a study by Barmeyer (2004) revealed
that females are more represented in diverging, accommodating and converging
learning styles, whereas males are more represented in assimilating learning style
(Barmeyer 2004). Therefore in this study the effect of gender on purchase intention,
attitude and comprehension of RNPs in media richness and interactivity conditions is
examined.
170
In the Media Richness treatment, for product 1, there were 103 female and 97
male participants. There was no main effect evident (F(3, 196)=2.15 , p=0.09) and the
assumption of equality of covariance matrices was not satisfied as; Box’s M=13.9,
F=2.27, p=0.034. Looking into individual ANOVA, there was a significant difference
observed for attitude between female and male (F=4.1, p=0.04). For product 2, 88
female and 112 male participated. There was no main effect observed
(F(3,196)=0.343, p=0.79) but the assumption of homoscedasticity was upheld (Box’s
M=1.41, F=0.232, p=0.966). No statistically significant difference was observed when
performing multiple ANOVAs.
In the Interactivity condition, for product 1, the sample consists of 104 female and
96 male participants. There was a significant main effect observed for purchase
intention, comprehension and attitude collectively towards product 1 (F(3,196)=6.1 ,
p=0.001). The assumption of homoscedasticity was upheld (Box’s M=3.279, F=0.538,
p=0.780). In performing multiple ANOVAs, there was a significant difference
observed for attitude (F=17.5, p=0.000). For product 2, there are 93 female and 107
male participants. Again there was a main effect observed (F(3,196)=2.68, p=0.04)
with the assumption of homoscedasticity being upheld (Box’s M=2.153, F=0.353,
p=0.909). In performing ANOVAs, there was no significant effect observable.
The result revealed that there can be a difference between female and male
attitude (male being more positive) towards RNPs, but the result is inconsistent
amongst products as it is only observed for product 1. This might be due to
differences in product characteristics, which needs to be investigated further.
5.2.5 Common Contribution
There is a contribution towards the body of knowledge of innovation adoption,
learning, and online behaviour, which is evident across all three papers. Within the
innovation adoption literature, there is general agreement that comprehension leads to
attitude formation (e.g. Goldsmith and Newell, 1997; Pagani, 2007). The relationship
between comprehension, attitude formation and purchase intention is supported by
persuasion theories (Ratneshwar and Chaiken, 1991). The result of these three papers
171
confirms the positive influence of comprehension towards attitude and purchase
intention within the RNP context. Furthermore, the innovation-decision process
explains a relationship between attitude and purchase intention within the innovation
context (Rogers, 1995), which is again extended to the RNP domain, by the results of
all three papers. This adds to the body of academic literature, as the findings confirm
that comprehension leads to attitude and purchase intention, and that attitude is
positively related to purchase intention, within the context of RNPs. These findings
complement the previous body of literature on consumer learning and online
behaviour.
5.2.6 Contribution to Theory
Persuasion Theory
The first theoretical contribution of this thesis is towards the Persuasion Theory
(Ratneshwar and Chaiken, 1991). The theory explains how comprehension is a
prerequisite to the formation of attitudes, intention and behaviour, especially under
the central or systematic processing route. According to Elaboration Likelihood
Model (Petty and Cacioppo, 1981), the central route is one of the routes towards
persuasion, alongside the peripheral route. Consumers either process arguments in
the persuasion attempt and elaborate on the arguments resulting in a positive or
negative evaluation, or use peripheral cues to form an empirical judgment (Petty and
Cacioppo, 1981). As RNPs are high-involvement products by nature (Feiereisen et
al., 2008), consumers need to follow the central route to persuasion. Therefore if
consumers learn about RNPs, they develop positive attitudes and behaviour intention
towards these products (Bettman, 1979). The findings of this thesis confirm this
prediction, and add to the Persuasion Theory, by indicating how within the context of
RNPs, better comprehension leads to a more positive attitude and purchase intention.
Cognitive Theory of Multimedia Learning
The second theoretical contribution is towards the Cognitive Theory of
Multimedia Learning (Mayer, 1997). The theory supports the effectiveness of
inserting multimedia information in order to improve consumers’ comprehension and
attitude (e.g. Paivio, 1971; Alesandrini, 1982; Rayport and Jaworski, 2001; Zhang et
al., 2006). The theory is concerned with use of multimedia learning as a tool to
172
improve learning (Mayer and Moreno 1998). Learning, or consumer comprehension
was an important factor studies in this research. The inclusion of vividness and
interactivity as characteristics of multimedia information is supported in this theory;
however, the findings of this thesis contradict previous academic literature by
exhibiting that vividness does not improve individuals’ comprehension, attitude and
purchase intention towards RNPs. Furthermore, the influence of interactivity on
comprehension, attitude and purchase intention was inconsistent across RNPs.
Looking back at the theory of multimedia learning, the finding of this study
challenges the compatibility of this theory within the innovation adoption literature,
especially when learning about RNPs. Is this theory generalizable for RNPs and
radical innovations? In order to answer this question, more research is needed, to first
manipulate multimedia information presentation further, and second to include
different types of RNPs in the experiments and examine RNP nature. The present
study employed sensory information in a certain format, such as use of audio with a
female voice, that needs to be played by the user (it does not start automatically and
user can not manipulate it). Therefore, the multimedia information presentation might
be contradictory to the interactivity explanation of Theory of Multimedia Learning as
controlling and manipulating are two main characteristics defining interactivity. This
challenges the Theory as the role of interactivity in sensory information presentation
needs more clarification.
The second element in need of more in-depth investigaion is the characteristics of
RNPs. It is explained in a ‘multimedia principle’ from the Cogntivie Theory of
Multimedia Learning that “people learn more deeply from words and pictures than
from words alone” (Mayer 2005 p.47). On the other hand, RNPs are high-
involvement product in nature (Feiereisen 2009). According to the Elaboration
Likelihood Model (ELM), individuals need to follow a central route in order to
comprehend RNPs (Petty and Cacioppo, 1981). Individuals following a central route
prefer a rich text-based information presentation format (Petty and Cacioppo, 1981).
In the case of radical innovation, product comprehension occurs at the time of product
information evaluation (Hoeffler 2003), and information presented via various senses
is proven to improve individuals’ comprehension. There needs to be further
investigation to understand whether inefficiency of multisensory information
173
presentation on RNP comprehension is due to the information preference based on
RNPs unique characteristics, or if it varies based on the availability of information or
other RNP characteristics. This then clarifies the ambiguity surrounding the Theory of
Multimedia Learning applicability within the radical innovation context.
Social Learning Theory
The third theoretical contribution is to the Social Learning Theory (Bandura,
1977) and Social Response Theory (Moon, 2000). According to the Social Learning
Theory, having an appropriate degree of social presence promotes users’ learning.
Social Response Theory further explains how people automatically tend to treat
computer technologies as social actors. However, the academic literature is
inconsistent in supporting whether the inclusion of a higher degree of social cues
(such as a human-like avatar) can result in an increased sense of anthropomorphism,
and conclusively an improved consumer comprehension. The findings of this thesis
contribute to the theories of Social Learning and Social Response by stating how the
inclusion of a high degree of social cues (four social cues identified within consumer
behaviour literature (Nass and Steuer, 1993) has a positive influence upon consumers’
perceived anthropomorphism. The findings of the thesis also supply empirical support
for how various aspects of mindless and mindful anthropomorphism influence overall
perceived anthropomorphism individuals’ experience.
The result of this study indicates a moderately positive correlation between
perceived anthropomorphism and comprehension, attitude and purchase intention.
The result is almost similar for the correlation of mindless anthropomorphism and
comprehension, attitude and purchase intention; however the correlation between
mindful anthropomorphism and comprehension was weak, and was inconsistent with
the other relationships. Overall, the findings indicate the benefits of including social
cues in a form that participants perceive anthropomorphism, as then it will influence
individuals’ comprehension, attitude and purchase intention towards RNPs. This
consistent result might be due to the close relationship between Social Learning
Theory and Diffusion of Innovation Theory. This adds to the Social Learning Theory
literature by confirming the theory’s compatibility within the innovation context, and
reiterates the connection between Social Learning Theory and The Diffusion of
Innovation Theory.
174
Diffusion of Innovation Theory
The fourth theoretical contribution is towards the Diffusion of Innovation Theory
(Rogers, 1962). This theory explains how over time an innovation is diffused by
individuals. Individuals as innovation adopters are one of the main elements
identified within the Diffusion of Innovation Theory, with individuals being
categorized into five groups based on their adoption rate: innovators, early adopters,
early majority, late majority and laggards (Rogers, 2003). Therefore consumers as
innovators are the quickest group to adopt an innovative product. Consumer
innovativeness as an innate innovativeness is explained by Vandecasteele and Geuens
(2010) in their Motivated Consumer Innovativeness (MCI) model as four distinct
aspects of hedonic, functional, cognitive and social. Previous academic literature
supports the positive but weak relationship between innate innovativeness and
consumers’ adoption behaviour (e.g. Goldsmith et al., 1995; Citrin et al., 2000; Im et
al., 2007). Furthermore, according to the innovation literature, innate innovativeness
leads to comprehension and attitude.
The result of this study contributes significantly to the Diffusion of Innovation
Theory (Rogers, 1962) by explaining how various hedonic, functional, cognitive and
social motivational sources influence comprehension and attitude (adoption
behaviours) either positively or negatively. Where hedonic motivation aspects of
consumer innovativeness positively affect learning, social innovativeness takes a
negative direction towards learning, within the context of RNPs. Hedonic and
cognitive innovativeness were found to positively impact consumers’ attitude towards
RNPs, when functional innovativeness had no influence on consumers’ learning and
attitude. This can be a possible explanation for the positive but weak relationship
between innate innovativeness and adoption behaviour identified by academics.
5.2.7. Limitations and future research
This section looks into the limitations of each paper and proposes future research
recommendations, based on the limitations identified within each paper.
175
5.2.7.1 Methodological Limitations
The first limitation is inherent in experimental studies (Papers 1 and 2) and
mainly relates to the generalisation of the findings beyond the research’s product
choice. The thesis is has considered particular RNPs within the domain of Kitchen
Appliances, therefore studying other RNP categories is an interesting topic for future
research. Also, the choice of web experimentation, which resulted into lack of control
over participants’ action, may upsurge external validity but decrease the study’s
internal validity. However, web experimentation is a close to commercial scenario for
the purpose of this thesis and considered a good method to be used.
Online surveys were also used for the purpose of Paper 3 within this thesis.
Although online surveys are an appropriate method when gathering data via CMEs,
some limitations have been identified by scholars. For example, Carbonaro and
Bainbridge (2000), pointed out that participants need to have certain skills in order to
take part in online surveys. This factor shouldn’t jeopardize the result of this study as
the sample is recruited via an online survey tool (Mturk), hence participants should
already have a minimum skillset to be able to partake via this online platform.
However, it may be beneficial for future research to measure participants’ IT skills to
identify any bias as a result of participants’ technological skills. Next, the limitations
and future research for each paper is discussed.
There are some common limitations and future research recommendations for all
three papers. RNP attributes and characteristics were not controlled for in either Paper
1 or 2; therefore, the same justification and future research have been identified for
both papers and explained in Section 5.2.7.2. Sensory depth and consumer
characteristics were not controlled for in all papers therefore, the justification and
future research recommendation are explained in Section 5.2.7.2
5.2.7.2 Paper 1
The first limitation of Paper 1 is that RNP characteristics, such as those identified
by Rogers (1995) of relative advantage, compatibility, complexity, trialability and
observability, were not measured or controlled. This may be a reason why some of the
findings were inconsistent across the two RNPs used within Paper 1. For instance, the
findings of this paper indicated that vividness was perceived higher by individuals for
176
Product 2 when dealing with a CME consisting of multimedia presentation, in
comparison to using a text and image interface. The same result did not apply to
Product 1. This may have been due to an element of the product, such as product
attributes. Product attributes are influential factors on consumers’ diffusion of
innovation (Rogers, 1999). These factors are related to the functional aspect of the
product (such as compatibility, relative advantage, perceived usefulness) and/or the
hedonic aspect of the product (such as appearance) (Rogers, 1995; Irani, 2000). Each
of these elements can be influential upon not only the presentation of the product, but
also the individuals’ perception towards the new concept. This is due to the fact that
one of the main criticisms of diffusion of innovation was that the impact of
consumers’ perception of innovation attributes was not considered in previous
literature (Lowrey, 1991). Paper 1 is concerned about the presentation elements of
RNP promotion, thus considering RNP attributes would have made the study
overcomplicated; this element is therefore not examined within this paper.
Another reason for the inconsistency across the findings is the presentation
elements (sensory depth as explained by Steuer, 1992). This includes the quality of
the pictures or the resolution of the information transmitted to the senses (Steuer,
1992), which was not controlled for in Paper 1. As the products are different in colour
and shape (Product 1 was lighter in colour and smaller than Product 2), the
presentation element might be affected by the RNP’s image attributes and therefore
the perception of individuals. Finally, although consumer innovativeness as an
attribute is controlled for, the results indicated that differences in consumer
comprehension, attitude and purchase intention cannot be attributed to consumer
innovativeness. Considering various characteristics identified within the innovation
literature (e.g. Rogers, 2003; Weijters and Roehrich, 2004; Tellis et al., 2009) could
shed more light into this element of the study, but this was outside the scope of this
study.
Future research could firstly examine product attributes identified by scholars
within the innovation adoption literature, such as Roger’s (1995) relative advantage,
compatibility, complexity, trialability and observability. Secondly, sensory depth
(Steuer, 1992) could be examined to understand if this presentation element is
influential on individuals’ online behaviour and perception towards various websites.
177
Thirdly, future research could further explore consumer differences and
characteristics. Although consumer innovativeness was controlled for, various other
characteristics, such as participants’ different learning styles (e.g. Feiereisen et al.,
2008) or their background knowledge of the RNP domain (according to innovation
adoption literature) may be influential on their online behaviour towards RNPs.
5.2.7.3 Paper 2
As a common limitation, RNP attributes and characteristics, sensory depth and
consumer characteristics are not controlled for in Paper 2. However Paper 2 examined
consumers’ perceived anthropomorphism. One main limitation was the lack of
exhaustive literature on perceived anthropomorphism scales. The result of this study
clearly indicates that scales measuring perceived anthropomorphism need to take into
account mindless and mindful aspects of anthropomorphism. Another factor to
consider is that of website socialness; this is considered as the way a website displays
any form of social cues (not necessarily in the form of an avatar). For example,
Wakefield et al. (2011) developed a website socialness scale, which is very similar to
the mindless scale used in Paper 2. As the effect of content interactivity is not
captured within the perceived anthropomorphism of participants, one area for future
study could be to measure perceived website anthropomorphism using scales such as
Wakefield et al.’s (2011) website socialness scale. This may help our understanding
of whether content interactivity (or the insertion of other forms of social cue), is
actually perceived differently by participants. Preliminary research into this area
indicated a good set of results for socialness scales
5.2.7.4 Paper 3
As a common limitation, sensory depth and consumer characteristics are not
controlled in Paper 3. A further limitation for Paper 3 is the generalizability of the
findings, as it only considered one RNP, within a specific domain of kitchen
electronic appliances. This study needs to be replicated using various RNPs within
different domains. Paper 3 can be expanded upon by using various types of product
presentation. This was not implemented in Paper 3 as the focus was on consumer
characteristics rather than presentation formats. Presentation formats of interactivity
or anthropomorphic attributes could be inserted and examined to understand whether
178
the influence of consumer innovativeness on comprehension and online behaviour
differs across various presentation formats within CME.
5.3. Managerial Implications
The study provides valuable information for managers in promoting RNPs. Not
only does the study assist managers in information presentation, but also in targeting
innovator consumers as adopters of RNPs. From the information promotion aspect,
elements of vividness, interactivity and anthropomorphism were investigated.
However, the findings were not always consistent with existing literature. Product
managers need to be cautious in developing marketing communication strategies for
RNPs. It is evident from the innovation literature that individuals constantly
undervalue innovations, whereas companies overvalue their innovative products as
suggested by an objective analysis (Gourville 2005). Hence, assisting consumers to
reach a high level of RNP comprehension proved to be a difficult task. Also, existing
marketing strategies are designed for existing products and/or incrementally new
products (INP) (Hoeffler 2003). This is consistent with the findings of this study as
some factors, such as vividness, proved to improve consumer’s learning and online
behaviour in academic literature, weren’t influential as a result of our experiments.
Comprehension of RNPs may be very low if inappropriate communication strategies
are used for RNP promotion (Feiereisen 2008). It is evident from literature that
learning RNPs differs from learning other product types. RNP comprehension occurs
at the time of product evaluation (Hoeffler, 2003). Hence, it is important and
massively influential to facilitate consumers learning about RNPs. The result of this
study supports the challenging task of product managers. It also provides some
practical suggestions for product managers to understand the most suitable strategies
to present RNPs and radical innovations.
The first element investigated in this study from the information promotion aspect
is vividness. Vividness is supported to improve consumer learning (e.g. Paivio, 1971;
Alesandrini, 1982; Rayport and Jaworski, 2001; Zhang et al., 2006). Vividness is also
proved to influence consumer’s attitude positively as it provides a very similar
179
experience to direct product experience (Klein 2003). The findings of this study are
inconsistent with existing literature. Therefore, inclusion of vividness might not be
sufficient for RNP comprehension. Roehm et al. (2001) explained that presenting
RNPs using pictures might not be enough for product comprehension, as RNPs
possess benefits that might not be apparent from products’ surface attributes (Roehm
et al. 2001). On the other hand, as RNPs are high involvement products (Feiereisen
2008); according to ELM, individuals are following the central route in order to
understand RNPs (Petty and Cacioppo, 1981; Petty and Cacioppo, 1986). By
following the central route, individuals need to pay attention to the message content
and they prefer rich text based information. It is advisable for product managers not to
focus merely on the vividness element of information presentation, as vividness alone
might not be sufficient for improvement of RNP comprehension and consumer
attitude. It is also recommended to include sufficient verbal information to accompany
visual information.
Interactivity is the second presentation element investigated. Interactivity is
supported as an attribute improving consumers’ learning and online behaviours (e.g.
Brandt, 1997; Schlosser et al., 2003; Fiore et al., 2005; Park et al., 2005; Zhang et al.,
2006; Pantano and Naccarato, 2010). Direct product experience is an ideal form of
consumer-product experience that digital marketers attempt to achieve in CMEs.
Inclusion of interactivity is proved to resemble direct product experience, hence
improving consumer learning (Klein 2003). The findings of present research partially
support existing literature. The inclusion of interactivity improved consumers’
understanding of RNPs, as well as their attitude and purchase intention. It is
recommended that managers take advantage of inserting interactivity for RNP
presentation. This can be in form of including a 3D design of the product, or a video
clip that consumers can manipulate.
The 3D design enables individuals to interact with product by rotating, zooming
in/out and moving the product. With advanced technologies, accessing companies
providing 3D design services is easy. There are many software packages such as CAD
or Solid Work, enabling 3D designing. The cost is also manageable which makes this
technology affordable. A 3D design can be easily displayed in any CME and be
accessible for consumers. The inclusion of a video clip can also be beneficial, as it
180
can illustrate the product being used, or a sales agent instructing individuals how to
use a product. In this study, as RNPs are not yet available in the marketplace, and
there is not enough information available, video clips are made using existing
pictures. A female voice explained RNPs details over the video clip. Production of a
video clip is easy and affordable. Due to advancement of technology, making a video
clip can be done using any camera, and edited via numerous software packages and
applications. A video clip is easily embedded in a chosen CME and consumers can
access the video clip effortlessly. There are many applications and programmes
already installed in CMEs, and individuals can run/pause/forward and backward the
video clip as they wish. Because of their affordability, ease of production and
accessibility for consumers, businesses can take advantage of these technologies as
multimedia tools to improve consumers learning and attitude towards RNPs.
Anthropomorphism was the third element studied. Studies have demonstrated that
people react to computers as social actors (e.g Reeves and Nass, 1996).
Anthropomorphism attributes were included in CMEs as an element encouraging
social interaction and social learning (e.g. Nass and Moon 2000; Keeling 2010).
According to the Social Learning Theory, having an appropriate degree of social
presence promotes user learning; therefore employing anthropomorphic attributes,
which result in an increase in social cues and consequently social presence, can have a
positive effect on consumer learning. Furthermore, Interactive Multimodal Learning
Environments (Moreno and Mayer, 2007) support the interaction between students
(information receiver) and a pedagogical agent. A medium, which includes any type
of social cue, can enhance the level of anthropomorphism perceived by individuals.
There are four types of social cues identified within the consumer behaviour
literature: human voice, language, interactivity and social role (Nass and Steuer
1993). According to the literature, the more computers possess social cues, the more
people respond to them. The findings of this study support inclusion of a human-like
avatars, which possess all four social cues of human voice, language, interactivity and
social role. Therefore it is recommended to product managers to take advantage of
human-like avatars as an agent assisting consumers to gain understanding of the
innovative product.
181
Businesses have started recruiting human-like avatar as a facilitator for
company’s online consumer communication. Anna, Ikea’s chat-bot, is a well-known
example. Avatars were also used frequently within the tourism industry. It is claimed
that avatars can successfully mimic interpersonal interactions with online consumers
and apply social influence (Wang and Fodness 2010). It also appears that a human-
like interactive avatar facilitates consumers’ attitude and purchase intention.
Therefore, including avatars for the purpose of promoting RNPs can indirectly affect
consumers’ attitude towards the product and conclusively upsurge consumer’s
purchase intention. Using human-like avatars within anthropomorphism literature has
also proved to improve trust, which is influential on consumers purchase intention and
sale of the product. Hence, according to the related literature and findings of this
study, it is beneficial for companies to include a human-like interactive avatar in their
website in order to promote RNPs, and improve individual’s attitude and purchase
intention.
There are various companies selling virtual agents or avatars to businesses. They
offer a variety of virtual agents with different capabilities. According to technavio8, a
market research company, virtual assistants will take over traditional customer contact
centres very soon. Technavio claims the virtual agent market is predicted a growth
rate of 42.07% by 2017. Companies such as Anboto Group, Creative Virtual Ltd.
eGain Group are amongst the top 17 companies offering a wide range of avatars to the
marketplace. Main capabilities highlighted by users and companies are 3D persona
with emotions, mobile solutions, concept understanding, multiple questions, analytics
and social dialogue. These abilities are all offered by many companies. At the time of
this study, the avatar was purchased from sitepal9. This company has been operating
from 2003 and was one of the reputable companies offering avatars at that time.
Sitepal still offers a variety of avatars and based on the abilities required, they offer
packages from as little as $9.95 per month. This study employed an agent with
Artificial Intelligence with a monthly payment of $39.95. Nowadays it is even more
affordable and accessible for businesses to use avatars. Managers can program virtual
agents to communicate with consumers as they wish and benefit from an
improvement in consumers comprehension and attitude towards RNPs.
8 www.technavio.com9 www.sitepal.com
182
Another element, which proved to be influential on business successes for new
products, is targeting innovator customers. Innovation is considered as a crucial
competitive power for firms (Andries et al., 2009) and is an important factor in global
economic growth (Golder and Tellis, 1997). Although RNPs are being introduced to
the marketplace, 40% to 90% of new products fail, with highly innovative products
failing at an even greater rate (Cierpicki, 2000). Therefore it is important to
communicate with the right group of customers to introduce innovation to the
marketplace. Attracting correct customers results in other customers being encouraged
to consider product adoption. Early research established a link between the adoption
of new products and innovativeness (Midgley and Dowling, 1978; Hirschman, 1980).
Innovators have the quickest time of adoption, and although they are a very small
percentage of adopters, other adopter groups (early adopters, early majority, late
majority and laggard) follow innovators (Rogers 1995). Innovators possess a high
level of innate innovativeness. Innate innovativeness was investigated in this study.
Innate innovativeness is the degree to which an individual makes an innovation
decision “independently from the communicated experience of others” (Midgley
1977, p.75). Individuals with Innate innovativeness have a tendency to buy novelty,
new and different products (Manning et al., 1995, Steenkamp et al., 1999). Innate
innovativeness is further categorized into hedonic, cognitive, functional and social.
Looking back at the innovation literature, there is a weak relationship between innate
innovativeness and adoption behaviour (e.g. Goldsmith et al., 1995; Citrin et al.,
2000; Im et al., 2007). This study’s findings support the relationship identified within
literature. Hence by identifying and targeting innovative customers, companies
increase the possibility of product adoption.
As identified in innovation literature, innovative individuals are motivated by
different sources. Hedonic innovativeness is a drive to adopt innovation for hedonic
reasons such as appreciation of the product’s newness (Roehrich 1994). Having an
enjoyable experience when learning about RNPs via the website, the possibility of
them adopting the innovation increases (e.g. Lee 2010). The findings indicate that,
product managers can include elements such as use of 3D product design, to motivate
hedonic innovators. Furthermore, paying attention to aesthetic elements of product
and information presentation may be advantageous in attracting hedonic innovators.
This might be useful especially for products with a more hedonic nature, as
183
highlighting the hedonic attributes of products, results in a more hedonic product
presentation. According to the findings of this research, hedonic innovator customers
comprehend the information better and have a more positive attitude towards RNPs.
Hence improving the hedonic aspect of product presentation and targeting hedonic
innovators while promoting RNPs is strongly recommended to managers.
Radical innovations with a more functional nature seem to be better received by
functional innovative customers. Functional innovativeness refers to satisfying
functional needs such as solving consumption-related problems (Vandecasteele and
Geuens, 2010). RNPs are defined as products designed to solve consumers’ problem.
RNPs with an apparent functional nature are predicted to be easily adopted by
functional innovators. The findings of this study do not support this proposition.
Despite a weak relationship between innate innovativeness and adoption behaviour
(e.g. Goldsmith et al., 1995; Citrin et al., 2000; Im et al., 2007), no significant
relationship was observed for functional innovativeness. RNPs are complex products,
and it might not be an easy task to explain their functionality in plain, easy text.
Hence, product managers need to consider recruiting strategies to highlight
functionality of RNPs in a clear way, or they need to target different innovator
consumer types.
RNPs are challenging products, which can make learning about them stimulating.
Cognitively innovative customers are motivated to learn about new experience with
the aim of being stimulated (Baumgartner and Steenkamp, 1996). Due to the level of
newness of RNPs, individuals have limited cognitive structure for these products
(Feiereisen 2008). Also RNPs are products with a high level of uncertainty (e.g.
Rogers, 1995; Rogers, 2003; Alexander et al., 2008), hence cognitively motivated
individuals might not be able to satisfy their cognitive needs while acquiring
information about RNPs. This is supported with the findings of this study, as there
was no significant relationship between cognitive innovativeness and RNP
comprehension. However, the element of enjoyment and stimulation due to newness
of the RNP might influence cognitive innovators attitude, which is supported by the
findings of this research. For a more cognitively motivated group of consumers,
practitioners should consider presenting RNPs as a new concept but not a radical
innovation. This strategy can instantly reduce customers’ perceived uncertainty. There
184
are also various learning strategies, such as mental imagery or analogies, as explained
by Feiereisen et al. (2008), that might facilitate cognitively motivated consumers by
linking their limited knowledge base of RNPs to their existing knowledge base of
similar products or domains; this would facilitate their learning and reduce the
uncertainty they experience.
Possessing a radical innovation enables individuals to build a clear identity and
make a unique impression in social community ( Simonson and Nowlis, 2000, Tian et
al., 2001; Tian and McKenzie, 2001). According to the findings of this study, there is
a negative relationship between social innovators and RNP comprehension. The main
reason could be the unavailability of RNPs in marketplace. Therefore customers are
aware of products inaccessibility; hence they are unable to satisfy their social needs.
Product managers are not recommended to contact socially innovative customers,
unless very close to the time of the launch of RNPs and avoid contacting them in early
stages of product promotion. Another strategy is to involve social innovators into
exclusive communities that enable them to share information about RNPs, or to use
celebrity endorsement.
Looking back at the consumer behaviour literature, reference groups have always
been an influential factor on consumers’ attitude towards products (e.g. McCracken
1989, Eze et al. 2012). Reference groups were first introduced by Hyman (1942), and
includes a group consisting of friends, family, celebrities, work colleagues etc.
Consumers’ decision to purchase a particular product can be influenced by reference
groups (Eze et al. 2012). Celebrities as a reference group can have a positive impact
on individuals’ attitude towards products. As elaborated by Kelman (1961), reference
groups can have informational, utilitarian and value expressive influences on
individuals. Consumers either seek information to reduce uncertainty (informational),
or follow the wishes of others to achieve rewards (utilitarian), or are willing to
express themselves to the group they would like to appear similar to or belong to
(value expressive) (Bearden & Etzel 1982; Childers & Rao 1992). Celebrities as a
reference group have been employed by managers for decades. “A celebrity endorser
is an individual who is known to the public (actor, sports figure, entertainer, etc.) for
his or her achievements in areas other than that of the product class
endorsed”(Friedman & Friedman, 1979 p.63). Therefore, the inclusion of a celebrity
185
might increase individuals’ effort to learn about RNPs, especially if they have a strong
social motivation to be associated with a particular celebrity. Other reference groups
such as family, friends, and work colleagues will be influential in the later stages of
RNP promotion when the product has passed the introduction stage.
To sum up, the findings of this study assist product managers and practitioners
from two aspects. Firstly, it gives them direction to promote RNP related information
in a way to facilitate consumers’ understanding of the radical innovations. By
applying suitable techniques of product promotion, consumers not only understand
products, but also have a more positive attitude towards RNPs; this can result into a
higher possibility of product purchase. Secondly, is the targeting of appropriate
innovation adopters. Based on the nature of RNPs and the objectives of the company,
product managers need to consider the most suitable innovative consumer group to
upsurge the chance of RNPs being adopted by innovators.
186
Reference
Abbattista, F., Anderson, V., Anderson, H.H.K., Lops, P. and Sameraro, G. (2002)
Evaluating Virtual Agents for E-Commerce. Paper presented at First
International Joint Conference on AAMAS, Bologna, Italy.
Alesandrini, K.L. (1982) Imagery-Eliciting Strategies and Meaningful Learning.
Journal of Mental Imagery, 6, 125-140.
Alexander , D.L., Lynch , J.G. and Wang , Q. (2008) As Time Goes By : Do Cold
Feet Follow Warm Intentions for Really New Versus Incrementally New
Products? Journal of Marketing Research, 45 (3), 307-319.
Aliyu, A. Bello, M. Kasim R. Martin, D. (2014) Positivist and Non-Positivist
Paradigm in Social Science Research: Conflicting Paradigms or Perfect
Partners? Journal of Management and Sustainability, 4(3)
Amichai-Hamburger, Y., Wainapel, G. and Fox, S. (2002) "On the internet no one
knows I'm an introvert": Extroversion, neuroticism, and internet interaction.
CyberPsychology and Behaviour, 5 (2), 125-128.
Anderson, R. E. (1987). Females surpass males in computer problem solving: findings
from the Minnesota computer literacy assessment. Journal of Educational
Computing Research, 3(1), 39–51.
Anderson, J.C. (1988) Structural equation modeling in practice: A review and
recommended two-step approach. Psychological Bulletin, 103 (3), 411-423.
Andries, P., Czarnitzki, D. and Hottenrott, H. (2009) Innovation efforts of the Flemish
companies: Essential results of the CIS-2007 Survey IN: Debackere, K. and
Veugelers, R. (eds.) Flemish indicators book 2009. Science, Innovation,
Technology (in Dutch). 117-128.
Antil, J.H. (1984) Conceptualization and operationalization of involvement. IN:
Kinnear, T.C. (ed.) Advances in consumer research. Provo, UT: Association
for Consumer, 203−209.
Aristotle (1078b) Metaphysics.
187
Arts, Joep W. C., Frambach, Ruud T., & Bijmolt, Tammo H. A. (2011).
Generalizations on consumer innovation adoption: A meta-analysis on drivers
of intention and behavior. International Journal of Research in Marketing,
28(2), 134–144.
Babin, B.J., Darden, W.R. and Fgriffin, M. (1994) Work and or Fun: Measuring
Hedonic and Utilitarian Shopping Value. Journal of Consumer Research, 20
(4), 644-656.
Baddeley, A. (1992) Working Memory. Science, 31.
Baddeley, A. and Hitch, G. (1974) Working memory IN: Bower, G.H. (ed.) The
psychology of learning and motivation: Advances in research and theory. New
York: Academic Press, 47–89.
Bahorsky, R., Graber, J. and Mason, S. (1998) Official Internet Dictionary: A
Comprehensive Reference for Professionals. Rockville MD: Government
Institutes.
Bain, P.G., Park, J., Kwok, C. and Haslam, N. (2009) Attributing human uniqueness
and human nature to cultural groups: Distinct forms of subtle dehumanization.
Group Processes and Intergroup Relations, 12, 789-805.
Bandura, A. (1977) Social Learning Theory. General Learning Press.
Barmeyer C.I., (2004) Learning styles and their impact on cross-cultural training: An
international comparison in France, Germany and Quebec, International
Journal of Intercultural Relations, 28, 577–594
Barnes, J. (1954) Class and Committees in a Norwegian Island Paris. Human
Relations, 7, 39-58.
Bartels, J. and Reinders, M. (2011) Consumer innovativeness and its correlates: A
propositional inventory for future research. Journal of Business Research, 64
(6), 601-609.
Bartneck, C., Croft, E. and Zoghbi, S. (2009) Measuring the anthropomorphism,
animacy, likeability, perceived intelligence and perceived safety of robots.
International Journal of Social Robotics, 1 (1), 71-81.
188
Bartos, R. (1986) Qualitative Research: What It Is and Where It Came From. Journal
of Advertising Research, 26 (3), 3-6.
Baumgartner, H. and Steenkamp, J.E.M. (1996) Exploratory consumer buying
behavior: conceptualization and measurement. International Journal of
Research Marketing, 13, 121-137.
Bearden, W.O. and Etzel, M.J. (1982), Reference group influence on product and
brand purchase decisions, Journal of Consumer Research, vol. 9, pp. 183-194.
Bente, G., Rüggenberg S., Kramer, N. and Eschenburg F. (2008) Avatar-Mediated
Networking: Increasing Social Presence and Interpersonal Trust in Net-Based
Collaborations. Human Communication Research, 34 (2), 287-318.
Bentler, P.M. and Bonnet, D.C. (1980) Significance Tests and Goodness of Fit in the
Analysis of Covariance Structures. Psychological Bulletin, 88 (3), 588-606.
Bettman, J.R. (1979) An Information Processing Theory of Consumer Choice.
Reading: MA: Addison-Wesley.
Bhatt, G. (2004) Bringing virtual reality for commercial Web sites.
International.Journal of Human-Computer.Studies, 60 (1), 1-15.
Bickart, B. and Schindler, R.M. (2001) Internet forums as influential sources of
consumer information. Journal of Interactive Marketing, 15 (3), 31-40.
Birnbaum, M.H. (2001) Introduction to Behavioral Research on the Internet Upper
Saddle River, NJ: Prentice Hall.
Birnbaum, M.H. (2004) Human Research and Data Collection via the Internet.
Annual. Review of Psychology, 55 (1), 803-832.
Blake, B.F., Neuendorf, K.A. and Valdiserri, C.M. (2003) Innovativeness and Variety
of Internet Shopping. Internet Reseaarch: Elletronic Networking Appliations
and Policy, 13 (3), 156-169.
Bloch, P.H. and Richins, M.L. (1983) A theoretical model of the study of product
importance perceptions. Journal of Marketing, 69-81.
Blythe, J. (1999). Innovativeness and Newness in High-Tech Consumer Durables.
Journal of Product & Brand Management, 8(5), 415-429.
189
Bogden, R. and Taylor, S.J. (1975) Introduction to Qualitative Research Methods.
New York: Wiley.
Brandt, D.A. (1997) Constructivism: teaching for understanding of the Internet.
Communications of the ACM, 40 (10), 112-117.
Brosnan, M. J. (1998). The role of psychological gender in the computer-related
attitudes and attainments of primary school children (aged 6–11). Computers
in Human Behavior, 30(3-4), 203–208.
Brown, B. and Anthony, S. (2011) How PandG Tripled Its Innovation Success Rate.
Harvard Business Review, (June), 66-72.
Brown, S.A. and Venkatesh, V. (2005) Model of Adoption of Technology in
Households: A Baseline Model Test and Extension Incorporating Household
Life Cycle. MIS Quarterly Journal of Electronic Commerce, 29 (3), 399-426.
Bruner, J.S. (1960) The Process of Education: A Landmark in Educational theory.
Harvard University Press.
Bruno, F. and Muzzupappa, M. (2010) Product interface design: A participatory
approach based on virtual reality. International Journal of Human-Computer
Studies, 68 (5), 254-269.
Burt, R. (1982) Toward a Structural Theory of Action. New york: Academic Press.
Carbonaro, M. and Bainbridge, J. (2000) Design and development of a process for
web-based survey research. Paper presented at Annual Meeting of the
American Educational Research Association, New Orleans, LA.
Cassell, J., Sullivan, J., Prevost, S. and Churchill, E. (2000) Embodied
Conversational Agents. Cambridge: MIT Press.
Castano, E. and Giner-Sorolla, R. (2006) Not quite human: Infra-humanization as a
response to collective responsibility for intergroup killing. Journal of
Personality and Social Psychology, 90, 804–818.
Chaiken, S. (1987) The heuristic model of persuasion IN: Zanna, M.P., Olson, J.M.
and Herman, C.P. (eds.) Social influence: The Ontario Symposium Hillsdale,
NJ: Erlbaum, 3-39.
190
Chao, C., Reid, M. and Mavondo, F.T. (2012) Consumer innovativeness influence on
really new product adoption. Australasian Marketing Journal, 20, 211-217.
Chevalier, A. and Ivory, M.Y. (2003) Web site designs: Influences of designer's
expertise and design constraints. International.Journal of Human-
Computer.Studies, 58 (1), 57-87.
Choi, Y.K., Miracle, G.E. and Biocca, F. (2001) The Effects of Anthropomorphic
Agents on Advertising Effectiveness and the Mediating Role of Presence.
Journal of Interactive Advertising, 2 (1), 3-21.
Chomsky, N. (1957) Syntactic structures. The Hague, the Netherlands: Mouton.
Chung, H. and Zhao, X. (2004) Effects of Perceived Interactivity on Web Site
Preference and Memory: Role of Personal Motivation. Journal of Computer-
Mediated Communication, 10 (1), 00-00.
Churchil, G.A. and Iacobucci, D. (2004) Marketing Research: Methodological
Foundations, 9 ed. Cincinnati, OH.: Southwestern Publications.
Childers, T.L. and Rao, A.R. (1992) The influence of familial and peer-based
reference groups on consumer decision, Journal of Consumer Research, vol.
19, 198-211.
Cierpicki, S.W. (2000) Managers’ Knowledge of Marketing Principles: The Case of
New Product Development. Journal of Empirical Generalizations in
Marketing Science, 5, 771-790.
Citrin, A.V., Sprott, D.E., Silverman, S.N. and Donald E. Stem, J. (2000) Adoption of
Internet shopping: the role of consumer innovativeness. Industrial
Management and Data Systems, 100 (7), 294-300.
Crowther, D. & Lancaster, G. (2008) Research Methods: A Concise Introduction to
Research in Management and Business Consultancy, Butterworth-Heinemann
Coyle, J.R. and Thorson, E. (2001) The Effects of Progressive Levels of Interactivity
and Vividness in Web Marketing Sites. Journal of Advertising, 30 (3), 65-78.
Cunningham, D., Thach, L. and Thompson, K. (2007) Innovative E-Commerce Site
Design: A Conceptual Model to Match Consumer MBTI Dimensions to
Website Design Journal of Internet Commerce, 6 (3), 1-27.
191
Daghfous, N., Petrof, J.V. and Pons, F. (1999) Values and Adoption of Innovations: a
Cross-Cultural Study. Journal of Consumer Marketing, 16 (4/5), 314-331.
Dancey, C. and Reidy, J. (2004) Statistics without Maths for Psychology: using SPSS
for Windows. London: Prentice Hall.
Darby, M.R. and Karni, E. (1973) Free competition and the optimal amount of fraud.
Journal of Law and Economics, 16, 67-88.
Davis, F.D., Bagozzi, R.P. and Warshaw, P.R. (1989) User acceptance of computer
technology: A comparison of two theoretical models. Management Science, 5
(8), 982-1003.
Dehn, D. and Van Mulken, S. (2000) The Impact of Animated Interface Agents: A
Review of Empirical Research. International Journal of Human-Computer
Studies, 52 (1), 1-22.
Deshpande, R. (1983) Paradigms Lost': On Theory and Method in Research in
Marketing. Journal of Marketing, 47 (4), 101-110.
Dhar, Ravi and Klaus Wertenbroch (2000), "Consumer Choice Between Hedonic and
Utilitarian Goods", Journal of Marketing Research, 37 (1), 60-71.
Dickerson, M.D. and Gentry, J.W. (1983) Characteristics of adopters and nonadopters
of home computers. Journal of Consumer Research, 10, 225-235.
Dodds, W.B., Monroe, K.B. and Grewal, D. (1991) Effects of Price, Brand, and Store
Information on Buyers' Product Evaluations. Journal of Marketing Research,
28 (3), 307-319.
Domagk, S., Schwartz, R.N. and Plass, J.L. (2010) Interactivity in multimedia
learning: An integrated model. Computers in Human Behavior, 26 (5), 1024-
1033.
Donath, J. (2007) Virtually Trustworthy. Science, 317 (5834), 53-54.
Dowling, G.R. (1999) Consumer Innovativeness. IN: Earl, P. and Kemp, S. (eds.) The
Elgar Companion to Consumer Research and Economic Psychology.
Cheltenham, UK:: Edward Elgar Publishing, 111-115.
European Advertising and Media Forecast 2007
192
Eze, U., Chin C.H., Lee, C.H. (2012) Purchasing Designer Label Apparels: The Role
of Reference Groups, Asian Journal of Business Research, 2(2), 52-74
Feiereisen, S. (2009) Consumer Response to Really New Products: Examining the
Impacts of Learning Strategies and Presentation Formats on Product
Comprehension and Attitude Formation, Doctor of Philosophy, Aston.
Feiereisen, S., Wong, V. and Broderick, A.J. (2008) Analogies and Mental
Simulations in Learning for Really New Products: The Role of Visual
Attention. Journal of Product Innovation Management, 25 (6), 593-607.
Field, A. (2005) Discovering Statistics Using SPSS, 2 ed. Thousand Oaks, CA: SAGE
Publications.
Fiore, A.M., Jin, H.J. and Kim, J. (2005) For fun and profit: Hedonic value from
image interactivity and responses toward an online store. Psychology and
Marketing, 22 (8), 669-694.
Floyd J. Fowler, J. (2008) Survey Research Methods (Applied Social Research
Methods), 4 ed.: Sage.
Fornell, C. and Larcker, D.F. (1981) Evaluating Structural Equation Models with
Unobservable Variables and MeasurementError. Journal of Marketing
Research, 28, 39-50.
Fortin, D.R. and Dholakia, R.R. (2005) Interactivity and vividness effects on social
presence and involvement with a web-based advertisement. Journal of
Business Research, 58 (3), 387-396.
Foxall, G.R. (1988) Consumer Innovativeness: Novelty-Seeking, Creativity and
Cognitive Style. Research in Consumer Behavior,, 3, 79-113.
Franceschi, K., Lee, R.M., Zanakis, S.H. and Hinds, D. (2009) Engaging Group E-
Learning in Virtual Worlds. Journal of Management Information Systems, 26
(1), 73-100.
Franks, P.C. and Oliver, G.C. (2012) Experiential learning and international
collaboration opportunities: virtual internships. Library Review, 61 (4), 272 -
285.
193
Fraser, B. Tobin, K. McRobbie C.J. (2012), Second International Handbook of
Science Education (Part one), Dordrecht, Springer Netherland
Frattini, F., De Massis, A., Chiesa, V., Cassia, L. and Campopiano, G. (2012)
Bringing to Market Technological Innovation: What Distinguishes Success
from Failure. , 2012, 4:15. International Journal of Engineering Business
Management, 4 (15), 1-11.
Friedman, H. H., & Friedman, L. (1979). Endorser Effectiveness by Product Type.
Journal of Advertising Research, 63-71.
Garcia, R. and Calantone, R. (2002) A critical look at technological innovation
typology and innovativeness terminology: a literature review. Journal of
Product Innovation Management, 19 (2), 110-132.
Gatignon, H. and Robertson, T.S. (1985a) A Propositional Inventory for New
Diffusion Research. Journal of Consumer Research, 11, 849-867.
Gatignon, H. and Robertson, T.S. (1985b) A Propositional Inventory for New
Diffusion Research. Journal of Consumer Research, 11 (4), 849-867.
Gentner, D. (1989) The Mechanisms of Analogical Learning IN: Vosniadou, S. and
Ortony, A. (eds.) Similarity and Analogical Reasoning. Cambridge University
Press.
Goldenberg, J., Lehmann, D.R. and Mazursky, D. (2001) The Idea Itself and the
Circumstances of Its Emergence as Predictors of New Product Success.
Management Science, 47 (1), 69-84.
Golder, P.N. and Tellis, G.J. (1997) Will It Ever Fly? Modeling the Growth of New
Consumer Durables. Marketing Science, 16 (3), 256-270.
Goldsmith, R.E. and Foxall, G.R. (2003) The Measurement of Innovativeness. IN:
Shavinina, L.V. (ed.) The International Handbook on Innovation. Oxford:
Elsevier, 321-330.
Goldsmith, R.E., Freiden, J.B. and Eastman, J.K. (1995) The generality/specificity
issue in consumer innovativeness research. Technovation, 15 (10), 601-612.
Goldsmith, R.E. and Hofacker, C.F. (1991) Measuring consumer innovativeness.
Journal of the Academy of Marketing Science, 19 (3), 209-221.
194
Goldsmith, R.E. and Lafferty, B.A. (2002a) Consumer respons to Web sites and their
influence on advertising effectiveness. Internet Reseaarch: Elletronic
Networking Appliations and Policy, 12 (4), 318-328.
Goldsmith, R.E. and Lafferty, B.A. (2002b) Consumer response to Web sites and their
influence on advertising effectiveness. Internet Reseaarch: Elletronic
Networking Appliations and Policy, 12 (4), 318-328.
Goldsmith, R.E. and Newell, S.J. (1997) Innovativeness and price sensitivity:
managerial, theoretical and methodological issues. Journal of Product and
Brand Management, 6 (3), 163–174.
Golsar, M.D. (1987) Marketing and The Adoption of Microcomputers: An application
of Diffusion Theory. Journal of Marketing Science, 15 (2), 42-48.
Gong, L. (2008) How social is social responses to computers? The function of the
degree of anthropomorphism in computer representations. Computers in
Human Behavior, 24 (4), 1494-1509.
Gong, L. and Lai, J. (2003) To mix or not to mix synthetic speech and human speech?
Contrasting impact on judge-rated task performance versus self-rated
performance and attitudinal responses. International Journal of Speech
Technology, 6, 123–131.
Gong, L. and Nass, C. (2007) When a Talking-Face Computer Agent Is Half-Human
and Half-Humanoid: Human Identity and Consistency Preference. Human
Communication Research, 33 (2), 163-193.
Gonzalez, C. and Kasper, G.M. (1997) Animation in user interfaces designed for
decision support systems: the effects of image abstraction, transition, and
interactivity on decision quality. Decision Science, 28, 793-823.
Goodman, J.K., Cryder, C.E. and Cheema, A. (2013) Data Collection in a Flat World:
The Strengths and Weaknesses of Mechanical Turk Samples. Journal of
Behavioral Decision Making, 26 (3), 213-224.
Gourville, J.T. (2006) Eager sellers stony buyers: understanding the psychology of
new-product adoption. Harvard Business Review, 99-106.
Gray D. E. (2009) Doing Research in the Real World, 2nd Edition, Sage
195
Gregan-Paxton, J., Hibbard, J.D., Brunel, F.F. and Azar, P. (2002) So That's What
That Is: Examining the Impact of Analogy on Consumers' Knowledge
Development for Really New Products. Psychology and Marketing, 19 (6),
533-550.
Grewal, R., Mehta, R. and Kardes, F.R. (2000) The Role of the Social-Identity
Function of Attitudes in Consumer Innovativeness and Opinion Leadership.
Journal of Economic Psychology, 21 (3), 233-252.
Groom, V., Nass, C., Chen, T., Nelson, A., Scarborough, J.K. and Robles, E. (2009)
Evaluating the effects of behavioural realism in embodied agents.
International Journal of Human-Computer Studies, 67 (10), 842–849.
Guba, E.G. and Lincoln, Y.S. (1994) Competing Paradigms in Qualitative Research",
in N.K. Denzin and Lincoln, , Sage Publications. IN: Denzin, N.K. and
Lincoln, Y.S. (eds.) Handbook of Qualitative Research,. Thousand Oaks:
Sage.
Gummesson, E. (2000) Qualitative Methods in Management Research, 2 ed.
Thousand Oaks: Sage.
Guttentag, D.A. (2010) Virtual reality: Applications and implications for tourism.
Tourism.Management, 31 (5), 637-651.
Ha, L. and James, E.L. (1998) Interactivity Reexamined: A Baseline Analysis of
Early Business Web Sites. , 42(4), 456. Journal of Broadcasting and
Electronic Media, 42 (4), 456-463.
Hackley, C. (2003) Doing Research Projects in Marketing Management and
Consumer Research. Routledge.
Hair, J.F., Black, W.C., Bebin, B.J. and Anderson, R.E. (2010) Multivariate Data
Analysis, 7 ed. New Jersey:: Prentice Hall.
Hale, J.L. and Stiff, J.B. (1990) Nonverbal primacy in deception detection.
Communication Reports 3,75–83.
Haslam, N. (2006) Dehumanization: An Integrative Review. Personality and Social
Psychology Review (Lawrence Erlbaum Associates), 10 (3), 252-264.
196
Haslam, N., Bain, P.G., Dough, L., Lee, M. and Bastian, B. (2005) More human than
you: Attributing humanness to self and others. Journal of Personality and
Social Psychology, 89, 937-950.
Haslam, N., Kashima, Y., Loughnan, S., Shi, J. and Suitner, C. (2008) Subhuman,
inhuman, and superhuman: Contrasting humans with nonhumans in three
cultures. Social Cognition, 26, 248-258.
Haslam, N., Loughnan, S. and Holland, E. (2013) The psychology of humanness IN:
Gervais, S.J. (ed.) Objectification and (de)humanization: The 60th Nebraska
Symposium on motivation. New York: Springer, 25-51.
Hauser, J., Tellis, G.J. and Griffin, A. (2006) Research on innovation: a review and
agenda for marketing science. Marketing science, 25, 687–717.
Hemp, P. (2006) Avatar-Based Marketing. Harvard Business Review, 84 (6), 48-57.
Hirschman, E.C. (1980) Innovativeness, Novelty Seeking and Consumer Creativity.
Journal of Consumer Research, 7 (3), 283-295.
Hirschman, E.C. (1984) Experience Seeking - a Subjectivist Perspective of
Consumption. Journal of Business Research, 12 (1), 115-136.
Hirunyawipada, T. and Paswan, A.K. (2006) Consumer innovativeness and perceived
risk: Implications for high technology product adoption. Journal of Consumer
Marketing, 23 (4), 182–198.
Hoeffler, S. (2003) Measuring Preferences for Really New Products. Journal of
Marketing Research (JMR), 40 (4), 406-420.
Hoffmann, S. and Soyez, K. (2010) A cognitive model to predict domain-specific
consumer innovativeness. Journal of Business Research, 63 (7), 778-785.
Holzwarth, M., Janiszewski, C. and Neumann, M.M. (2006) The Influence of Avatars
on Online Consumer Shopping Behavior. Journal of Marketing, 70 (4), 19-36.
Hoyer, W. and Stokbuger-Sauer, N. (2012) The role of aesthetic taste in consumer.
Journal of Academy of Marketing Science, 40 (1), 167-180.
Hoyer, W.D. and Macinnis, D.J. (2010) Consumer Behavior. Mason: South-Western:
CENGAGE Learning.
Hoyer, W.D., and Ridgway, N.M. (1984). Variety Seeking as an Explanation for
197
Exploratory Purchase Behavior: A Theoretical Model. Advances in Consumer
Research, 11, 114-119.
Hull, C.L. (1943) Principles of behavior. New York, NY: Appleton–Century–Crofts.
Hurt, H.T., Joseph, K. and Cook, C.D. (1977) Scales for the measurement of
innovativeness. Human Communication Research, 4 (1), 58-65.
Hyman, H.H. (1942), The psychology of status, Archives of Psychology, vol. 269, pp.
94-102.
Im, S., Bayus, B. and Mason, C.H. (2003) An Empirical Study of Innate Consumer
Innovativeness, Personal Characteristics and New-Product Adoption Behavior.
Journal of the Academy of Marketing Science, 31 (1), 61-73.
Im, S., Mason, C. and Houston, M. (2007) Does innate consumer innovativeness
relate to new product/service adoption behavior? The intervening role of social
learning via vicarious innovativeness. Journal of the Academy of Marketing
Science, 35 (1), 63-75.
Irani, T. (2000) Prior Experience, Perceived Usefulness and the Web: Factors
Influencing Agricultural Audiences’ Adoption of Internet Communication
Tools. Journal of Applied Communications, 84 (2), 49-63.
Isbister, K. and Nass, C. (2000) Consistency of personality in interactive characters:
verbal cues, non-verbal cues, and user characteristics. International Journal of
Human-Computer.Studies, 53 (2), 251-267.
Jin, S.-a.A. (2009) Modality Effects in Second Life: The Mediating Role of Social
Presence and the Moderating Role of Product Involvement. CyberPsychology
and Behaviour, 12.
Jöreskog, K. and Long, J.S. (1993) Introduction IN: Kenneth, A.B. and Scott, J. (eds.)
Testing Structural Equation Models. Newbury Park, CA: Sage.
Kaboub F (2008). Positivist Paradigm. Retrieved from http://
personal.denison.edu/~kaboubf/Pub/2008-Positivist-Paradigmpdfon in 25
August 2016
198
Kalyanaraman, S. and Sundar, S.S. (2006) The Psychological Appeal of Personalized
Content in Web Portals: Does Customization Affect Attitudes and Behavior?
Journal of Communication, 56 (1), 110-132.
Kalyuga, S., Ayres, P., Chandler, P. and Sweller, J. (2003) The expertise reversal
effect. Educational Psychologist, 38, 23-31.
Kang, S. and Gratch, J. (2014) Exploring Users' Social Responses to Computer
Counseling Interviewers' Behavior. Journal of Computers in Human
Behavior, 34, 120-130.
Kay, E. (1992). An analysis of methods used to examine gender differences in
computer-related behavior. Journal of Educational Computing Research, 8(3),
277–290.
Keeling, K.A., Mcgoldrick, P.J. and Beatty, S. (2010) Avatars as salespeople:
communication style, trust, and intentions. Journal of Business Research, 63
(8), 793-800.
Kelman, H.C. (1961), Processes of opinion change. Public Opinions Quarterly, vol.
25, pp. 57-
Keppel, G. and Wickens, T.D. (2004) Design and Analysis: A Researcher's
Handbook, 4 ed. New Jersey: Prentice Hall.
Kerlinger, Fred N. (1986), Foundations of Behavioural Research, 3rd edition, Holt,
Kim, J. and Mcmillan, S.J. (2008) Evaluation of Internet advertising research: a
bibliometric analysis of citations from key sources. Journal of Advertising, 37
(1), 99-112.
Kim, Y. and Sundar, S.S. (2012) Anthropomorphism of computers: Is it mindful or
mindless? Computers in Human Behavior, 28 (1), 241-250.
King, J., Bond, T., & Blandford, S. (2002). An investigation of computer anxiety by
gender and grade. Computers in Human Behavior, 18(1), 69–84.
Klein, L.R. (2003) Creating virtual product experiences: The role of telepresence.
Journal of Interactive Marketing, 17 (1), 41-55.
199
Kline, R.B. (2005) Principles and practice of structural equation modeling, 2 ed.
New York: Guilford Press.
Kline, S., Jhally, S. and Leiss, W. (1986) Social Communication in Advertising:
Persons, Products and Images of Well-Being. Toronto, Canada: Methuen.
Koda, T. (1996) Agents with faces: The effects of personification Proceedings of
HCI 1996, London.
Kotler, P. (2003) Marketing Management. Prentice Hall.
Koval, P., Park, J. and Haslam, N. (2009) Human, or Less Human? The inquisitive
mind magazine.
Kuester, S., Feurer, S., Schuhmacher, M.C. and Reinartz, D. (2015) Comparing the
incomparable? How consumers judge the price fairness of new products.
International Journal of Research in Marketing, (0).
Kunda, Z. (1999) Social cognition: Making sense of people. Cambridge, MA. MIT
Press.
Labay, D.G. and C., K.T. (1983) Exploring the Consumer Diffusion Process in the
Adoption of Solar energy Systems. Journal of Consumer Research, 8 (4), 271-
278.
Langer, E.J. (1989) Mindfulness. Reading: Addison-Wesley.
Lave, J. (1988) Cognition in Practice: Mind, mathematics, and culture in everyday
life. Cambridge: UK: Cambridge University Press.
Lee, K.Y. (2012) Consumer processing of virtual experience in e-commerce: A test of
an integrated framework. Computers in Human Behavior, 28 (6), 2134-2142.
Lee, K.C. and Chung, N. (2008) Empirical analysis of consumer reaction to the virtual
reality shopping mall. Computers in Human Behavior, 24 (1), 88-104.
Lee, K.M. and Nass, C. (2003) Designing social presence of social actors in human
computer interaction. Paper presented at SIGCHI conference on Human
factors in computing systems, Lauderdale, Florida, USA.
Lee, N. and Lings, I. (2008) Doing Business Research; A guide to theory and
practice. Sage, London.
200
Lee, S. (2010) Two Essays on Product Design and Consumer Evaluation, PhD,
University of Central Florida.
Lehmann, D. (1994) Characteristics of "Really" New Products. Paper presented at
Marketing Science Institute Conference, Boston.
Lei, P.-W. and Wu, Q. (2007) Introduction to Structural Equation Modeling: Issues
and Practical Considerations. Educational Measurement: Issues and Practice,
26 (3), 33-43.
Lester, J.C., Barlow, S.T., Converse, S.A., Stone, B.A., Kahler, S.E. and Bhogal, R.S.
(1997) The Persona Effect: Affective Impact of Animated Pedagogical Agents
CHI ‘97, Atlanta, GA. ACM Press,, 359-366.
Leyens, J.P., Paladino, M.P., Rodriguez, R.T., Vaes, J., Demoulin, S., Rodriguez, A.P.
and Gaunt, R. (2000) The emotional side of prejudice: The attribution of
secondary emotions to in groups and outgroups. Personality and Social
Psychology Review (Lawrence Erlbaum Associates), 4, 186-197.
Liu, C., Arnett, K.P., Capella, L.M. and Beatty, R.C. (1997) Web sites of the Fortune
500 companies: Facing customers through home pages. Information and
Management, 31 (6), 335-345.
Liu, Y. and Shrum, L. (2002) What is interactivity and is it always such a good thing?
Implications of definition, person and situation for the influence of
interactivity on advertising effectiveness. Journal of Advertising, 4, 53-64.
Lohtia, R., Donthu, N. and Guillory, M. (2013) The Impact of Advertising,
Trustworthiness, and Valance on the Effectiveness of Blogs. International
Journal of Electronic Marketing and Retailing, 5 (4), 317-339.
Loughnan, S. and Haslam, N. (2007) Animals and androids: Implicit associations
between social categories and nonhumans. Psychological Science, 24 (3-4),
433-461.
Lowrey, T.M. (1991) The Use cf Diffusion Theory in Marketing: A Qualitative
Approach to Innovative Consumer Behavior. Advances in Consumer
Research, 18.
201
Loyd, B. H., Loyd, D. E., & Gressard, C. P. (1987). Gender and computer experience
as factors in the computer attitudes of middle school students. Journal of
Early
Adolescence, 7(1), 13–19.
Lynn, M. and Harris, J. (1997) The Desire for Unique Consumer Products: A New
Individual Differences Scale. Psychology and Marketing, 14 (6), 601-616.
Mahajan, V., Muller, E. and Bass, F. (1990) New product diffusion models: A review
and directions for research. . Journal of Marketing, 54, 1-26.
Manning, K.C., Bearden, W.O. and Madden, T.J. (1995) Consumer Innovativeness
and the Adoption Process. Journal of Consumer Psycology, 4 (4), 329-345.
Mansfield, E. (1961) Technical Change and the Rate of Imitation,. Econometrica, 29,
741-766.
Mayer, R. (2005) Cambridge handbook of multimedia learning. New York:
Cambridge University Press.
Mayer, R. and Moreno, R. (1998) A Cognitive theory of multimedia learning:
Implications for design principles IN: Naryanan, N.H. (ed.) CHI'98 Workshop
on Hyped-media to Hyper-media: Toward theoretical foundations of design,
use and evaluation.
Mayer, R.E. (1997) Multimedia learning: Are we asking the right questions.
Educational Psychologist, 32, 1-19.
Mayer, R.E. (2004) Should there be a three-strikes rule against pure discovery
learning? American Psychologist, 59, 14-19.
Mayer, R.E., Bove, W., Bryman, A., Mars, R. and Tapangco, L. (1996) When less is
more: Meaningful learning from visual and verbal summaries of science
textbook lessons. Journal of Educational Psychology, 88, 64-73.
Mayer, R.E. and Chandler, P. (2001) When learning is just a click away: Does simple
user interaction foster deeper understanding of multimedia messages? Journal
of Educational Psychology, 93, 390-397.
202
McBreen, H.M. and Jack, M.A. (2001) Experimental Assessment of Effectiveness of
Synthetic Personae for Multi-Modal E-Retail Applications 4th International
Conference on Autonomous Agents, Barcelona,. 39-45.
McCarthy, E. (1955) Mathematical Proceedings of the Cambridge Philosophical
Society. Cambridge Philosophical Society, 131-140.
McCracken , G. (1989), “Who Is the Celebrity Endorser? Cultural Foundations of the
Endorsement Process,” Journal of Consumer Research, 16 (December), 310–
21
McGoldrick, P.J., Keeling, K.A. and Beatty, S.F. (2008) A typology of roles for
avatars in online retailing. Journal of Marketing Management, 24 (3/4), 433-
461.
Mcintosh, C. (2006) Rethinking fit assessment in structural equation modelling: A
commentary and elaboration on Barrett (2007). Personality and Individual
Differences, 42 (5), 859-867.
Mcmillan, S.J. and Hwang, J. (2002) Measures of perceived interactivity: an
exploration of the role of direction of communication, user control, and time in
shaping perceptions of interactivity. . Journal of Advertising, 31 (3), 29-42.
Meiselwitz, G. (2014) Social Computing and Social Media: 6th International
Conference, SCSM 2014. Springer.
Midgley, D. (1977) Innovation and New Product Marketing. New York: Halsted
Press, John Wiley and sons, Inc.
Midgley, D. and Dowling, G. (1978) Innovativeness: the concept and its
measurement. J Consum Res 4,229-242.
Midgley, D. and Dowling, G. (1993) A Longitudinal-Study of Product Form
Innovation - the Interaction between Predispositions and Social Messages.
Journal of Consumer Research, 19 (4), 611-625.
Mimoun, M.S., Poncin, I. and Garnier, M. (2012) Case study—Embodied virtual
agents: An analysis on reasons for failure. Journal of Retailing and Consumer
Services, 19 (6), 605-612.
203
Mohr, J., Sengupta, S. and Slater, S. (2010) Marketing of High-Technology Products
and Innovations. , 3 ed. New Jersey: Pearson.
Moon, J., Chadee, D. and Tikoo, S. (2008) Culture, product type, and price influences
on consumer purchase intention to buy personalized products online. Journal
of Business Research, 61 (1), 31-39.
Moon, Y. (2000) Intimate Exchanges: Using Computers to Elicit Self-Disclosure from
Consumers. Journal of Consumer Research, 26 (4), 323-339.
Moore, G.A. (1991) Crossing the Chasm: Marketing and Selling High-Tech Products
to Mainstream Customers. New York: HarperBusiness.
Moreau, C.P. (1997) A Different Game: Really-New Products, Evolving Markets, and
Responsive Organizations
Moreau, C.P., Lehmann, D.R. and Markman, A.B. (2001) Entrenched Knowledge
Structures and Consumer Response to New Products. Journal of Marketing
Research (JMR), 38 (1), 14-29.
Moreno, R. (2004) Decreasing cognitive load for novice students: Effects of
explanatory versus corrective feedback on discovery-based multimedia.
Instructional Science, 32, 99-113.
Moreno, R. (2005) Instructional technology: Promise and pitfalls IN: Pytlikzillig, L.,
Bodvarsson, M. and R., B. (eds.) Technology-based education: Bringing
researchers and practitioners together. Greenwich: CT: Information Age
Publishing., 1-19.
Moreno, R. (2006) Does the modality principle hold for different media? A test of the
method-affects-learning hypothesis. Journal of Computer Assisted Learning,
22 (3), 149-158.
Moreno, R. and Durán, R. (2004) Do multiple representations need explanations? The
role of verbal guidance and individual differences in multimedia mathematics
learning. Journal of Educational Psychology, 96, 492-503.
Moreno, R. and Mayer, R. (2007) Interactive Multimodal Learning Environments.
Educational Psychology Review, 19 (3), 309-326.
Mori, M. (1970) The uncanny valley. Energy, 7 (4), 33–35.
204
Morris, P.E. and Hampson, J. (1983) Imagery and consciousness. London: Academic.
Nagar, K. (2009) Advertising effectiveness in different media: A comparison of web
and television advertising. IIMB management review,, 21 (3).
Nass, C. and Brave, S.B. (2005) Wired for Speech: How Voice Activates and
Advances the Human-Computer Relationship. Cambridge, MA.: MIT Press.
Nass, C. and Carney, P. (1999) Are People Polite to Computers? Responses to
Computer-Based Interviewing Systems. Journal of Applied Social
Psychology, 29 (5), 1093-1110.
Nass, C. and Moon, Y. (2000) Machines and mindlessness: Social responses to
computers. Journal of Social Issues, 56 (1), 81-103.
Nass, C., Moon, Y., Fogg, B.J., Reeves, B. and Dryer, D.C. (1995) Can computer
personalities be human personalities? International Journal of Human-
Computer Studies, 43 (2), 223-239.
Nass, C., Moon, Y. and Green, N. (1997) Are computers gender-neutral? Gender
stereotypic responses to computer. Journal of Applied Social Psychology, 27
(10), 864–876.
Nass, C. and Steuer, J. (1993) Voices, Boxes, and Sources of Messages. Human
Communication Research, 19 (4), 504-527.
Nelson, P. (1970) Information and consumer behavior. Journal of Political Economy,
78 (2), 311-329.
Nelson, P. (1974) Advertising as information. Journal of Political Economy, 82, 729-
754.
Nowak, K.L. (2004) The Influence of Anthropomorphism and Agency on Social
Judgment in Virtual Environments. Journal of Computer Mediated
Communication, 9 (2).
Nowak, K.L. and Biocca, F. (2003) The effect of the agency and anthropomorphism
on users' sense of telepresence, copresence, and social presence in virtual
environments. Presence: Teleoperators and Virtual Environments, 12, 481-
494.
205
Nowak, K.L., Hamilton, M.A. and Hammond Chelsea, C. (2008) The Effect of
Viewer Perceptions of Avatar Anthropomorphism and Realism on Judgments
of Credibility and Homophily, and Avatar Choice 11th Annual International
Workshop on Presence, Padova.
Nowak, K.L. and Rauh, C. (2005) The Influence of the Avatar on Online Perceptions
of Anthropomorphism, Androgyny, Credibility, Homophily, and Attraction.
Journal of Computer-Mediated Communication, 11, 153–178.
Nowak, K.L. and Rauh, C. (2006) The influence of the avatar on online perceptions of
anthropomorphism, androgyny, credibility, homophily, and attraction. Journal
of Computer-Mediated Communication, 11 (1), 153-178.
Ogertschnig, M. and Van Der Heijden, H. (2004) A short-form measure of attitude
towards using a mobile information service. IN: Tan, Y.H. and Vogel, D.
(eds.) Proceedings of the 17th Bled Electronic Commerce conference. Bled.
Ostlund, L.E. (1972) Identifying Early Buyers. Journal of Advertising, 12 (1), 25-30.
Ostlund, L.E. (1974) Perceived Innovation Attributes as Predictors of Innovativeness.
Journal of Consumer Research, 1 (2), 23-29.
Pagani, M. (2007) A Vicarious Innovativeness Scale for 3G Mobile Services:
Integrating the Domain Specific Innovativeness Scale with Psychological and
Rational Indicators. Technology Analysis and Strategic Management, 19 (6),
709-728.
Paivio, A. (1965) Abstractness, imagery, and meaningfulness in paired-associate
learning. Journal of Verbal Learning and Verbal Behaviour, 4, 32-38.
Paivio, A. (1971) Imagery and verbal processes. New York: Holt, Rinehart, and
Winston.
Paivio, A. (1971a) Imagery and verbal processes. New York: Holt, Rinehart, and
Winston.
Paivio, A. (1971b) Imagery and Language IN: Segal, S.J. (ed.) Imagery: Current
cognitive approaches. New York: Academic Press.
Paivio, A. (1986) Mental Representations: A Dual Coding Approach. New York:
Oxford University Press.
206
Paivio, A. (2006) Dual Coding Theory and Education. Paper presented at Pathways to
Literacy Achievement for High Poverty Children, The University of Michigan
School of Education, September 29-October. Available from:
http://www.csuchico.edu/~nschwartz/paivio.pdf. (accessed June 7, 2015)
Pallant, J. (2005) SPSS Survival Manual: A Step by Step Guide to Data Analysis using
SPSS version 12. New York: Open University Press.
Palmer, J.W. and Griffith, D.A. (1998) An emerging model of web site design for
marketing. Communications of the ACM, 41 (3), 44-51.
Pan, Z., Chen, T., Zhang, M. and Xu, B. (2004) Virtual Presentation and
Customization of Products Based on Internet. International Journal of
CAD/CAM, 4 (1).
Pantano, E. and Naccarato, G. (2010) Entertainment in retailing: The influences of
advanced technologies. Journal of Retailing and Consumer Services, 17 (3),
200-204.
Park, J., Lennon, S.J. and Stoel, L. (2005) On-line product presentation: Effects on
mood, perceived risk, and purchase intention. Psychology and Marketing, 22
(9), 695-719.
Pavlov, I.P. (1927) Conditioned Reflexes: An Investigation of the Physiological
Activity of the Cerebral Cortex. London: Oxford University Press.
Pearson, P.H. (1970) Relationships Between Global and Specific Measures of
Novelty Seeking. Journal of Consulting and Clinical Psychology, 34 (2), 199-
204.
Penney, C.G. (1989) Modality effects and the structure of short-term verbal memory.
Memory and Cognition, 17, 398-422.
Petty, R.E. and Cacioppo, J.T. (1981) Attitudes and Persuasion: Classic and
Contempormy Approaches. Dubuque: IA: Wm. C. Brown.
Petty, R.E. and Cacioppo, J.T. (1986) Communication and Persuasion: Central and
Peripheral Routes to Attitude Change. New York: Springer-Verlag.
Phillips, B.J. (2000) The Impact of Verbal Anchoring on Consumer Response to
Image Ads. Journal of Advertising, 29 (1), 15-24.
207
Phillips, D.M. (1996) Anticipating the Future: The Role of Consumption Visions in
Consumer Behavior. Advances.in Consumer.Research, 23 (1), 70-75.
Piaget, J. (1950) The Psychology of Intelligence. London: Routledge.
Plass, J.L. Moreno, R. Brünken, R. (Ed) (2010) Cognitive Load Theory, Cambridge
University Press
Porter, S.R. (2004) Pros and cons of paper and electronic surveys. New Directions for
Institutional Research, 2004 (121), 91-97.
Powers, A. and Kiesler, S. (2006) The advisor robot: tracing people's mental model
from a robot's physical attributes. Paper presented at Proceedings of the 1st
ACM SIGCHI/SIGART conference on Human-robot interaction, Salt Lake
City, Utah, USA.
Prendinger, H. and Ishizuka, M. (2004) Life-Like Characters: Tools, Affective
Functions and Applications. . Berlin: Springer.
Prince, J. and Simon, D. (2009) Has the Internet accelerated the diffusion of new
products? Research Policy, 38 (8), 1266-1277.
Qi, Y.X. (1999) Application of virtual reality in safety engineering. China Safety
Science Journal, 9 (6), 49-53.
Ragusa, J.M. and Bochenek, G.M. (2001) Collaborative Virtual Design
Environments. Communications of the ACM, 44 (12), 40-43.
Raju, P.S. (1980) Optimum Stimulation Level - Its Relationship to Personality,
Demographics, and Exploratory-Behavior. Journal of Consumer Research, 7
(3), 272-282.
Ratneshwar, S. and Chaiken, S. (1991) Comprehension's Role in Persuasion: The
Case of Its Moderating Effect on the Persuasive Impact of Source Cues.
Journal of Consumer Research, 18 (1), 52-62.
Rayport, J.F. and Jaworski, B.J. (2001) Cases in E-Commerce. McGraw-Hill/Irwin
MarketspaceU.
Razali, N.M. and Wah, Y.B. (2011) Power comparisons of Shapiro-Wilk,
Kolmogorov-Smirnov, Lilliefors and Anserson-Darling tests. Journal of
Statistical Modeling and Analysis 2(1), 21-33.
208
Redmond, W.H. (2002) The Potential Impact of Artificial Shopping Agents in E-
Commerce Markets. Journal of Interactive Marketing, 16 (1), 56-66.
Reeves, B. and Nass, C. (1996) The Media Equation: How People Treat Computers,
Televisions, and New Media Like Real People and Places New York:
Columbia University Press.
Reigeluth, C.M. and Stein, R. (1983) Elaboration theory IN: Reigeluth, C.M. (ed.)
Instructional-design theories and models: An overview of their current status.
Hillsdale: NJ: Erlbaum.
Reiman, P. (2001) Net Ad Sector Moderates. AdvertisingAge, 72 (January 1), 72-74.
Robertson, T.S. (1967) The Process of Innovation and the Diffusion of Innovation.
Journal of Marketing, 31, 14-19.
Robertson, T.S. (1971) Innovative Behaviour and Communication. New York: Holt,
Rinehart and Winston.
Robinson, H., Wysocka, A. and Hand, C. (2007) Internet advertising effectiveness:
the effect of design for click-through rates for banner ads. International
Journal of Advertising, 26 (4), 527-541.
Roehm, Michelle L. and Brian Sternthal (2001), "The Moderating Effect of
Knowledge and Resources on the Persuasive Impact of Analogies", Journal of
Consumer Research, 28 (2), 257-272.
Rogers, E. (1962) Diffusion of innovations, 1 ed. New York: NY: Free Press.
Rogers, E. (1983) Diffusion of Innovations, 3 ed. New York: Free Press.
Rogers, E. (1995) Diffusion of Innovations, 4 ed. New York: The Free Press.
Rogers, E. (2003) Diffusion of innovations, 5 ed. New York: NY: Free Press.
Rome Reborn Brochure. (2008).
Rosa, J.A., Porac, J.F., Runser-Spanjol, J. and Saxon, M.S. (1999) Sociocognitive
Dynamics in a Product Market. Journal of Marketing, 63, 64-77.
Rosen, L. D., & Maguire, P. (1990). Myths and realities of computerphobia: A
metaanalysis. Anxiety Research, 3(3), 175–191.
209
Rothschild, M.L. (1987) Marketing Communications. Lexington, MA: D.C. Heath
and Company.
Ryan, R.M. and Deici, E.L. (2000) Intrinsic and extrinsic motivations: Classic
definitions and new directions. Contemporary Educational Psychology, 25,
54-67.
Schlosser, A.E., Mick, D.G. and Deighton, J. (2003) Experiencing Products in the
Virtual World: The Role of Goal and Imagery in Influencing Attitudes versus
Purchase Intentions. Journal of Consumer Research, 30 (2), 184-198.
Schmidt, H.G. (1983) Problem-based learning rationale and description. Medical
Education, 17 (1), 11-16.
Schottenbauer, M. A., Rodriguez, B. F., Glass, C. R., & Arnkoff, D. B. (2004).
Computers, anxiety, and gender: An analysis of reactions to the Y2K
computer problem. Computers in Human Behavior, 20(1), 67–83.
Schumacher, P., & Morahan-Martin, J. (2001). Gender, internet and computer
attitudes and experiences. Computers in Human Behavior, 17(1), 95–110.
Schumpeter, J.A. (1939) Business Cycles. New York: McGraw-Hill Book Company.
Sharma, S. (1996) Applied Multivariate Techniques. New York: John Wiley.
Shen, K.N. and Khalifa, M. (2012) System design effects on online impulse buying.
Internet Research, 22 (4), 396 - 425.
Siann, G., Macleod, J., Glissov, P., & Durndell, A. (1990). The effect of computer use
on gender differences in attitudes to computers. Computers in Education,
14(2),183–191.
Simon, H.A. (1957) Models of Man. New York: Wiley.
Simonson, I. and Nowlis, S.M. (2000) The Role of Explanations and Need for
Uniqueness in Consumer Decision Making: Unconventional Choices Based on
Reasons. Journal of Consumer Research, 27 (1), 49-68.
Sivaramakrishnan, S., Wan, F. and Tang, Z. (2007) Giving an “e-human touch” to e-
tailing: The moderating roles of static information quantity and consumption
210
motive in the effectiveness of an anthropomorphic information agent. Journal
of Interactive Marketing, 21 (1), 60-75.
Skinner, B.F. (1953) Science and Human Behavior. New York: Macmillan.
Skinner, B.F. (1974) About Behaviorism. New York: Knopf.
Sproull, L., Subramani, M., Kiesler, S., Walker, J.H. and Waters, K. (1996) When the
Interface Is a Face. Human-Computer Interaction, 11 (2), 97-124.
Steenkamp, J.E.M., Ter Hofstede, F. and Wedel, M. (1999) A crossnational
investigation into the individual and national cultural antecedents of consumer
innovativeness. Journal of Marketing Intelligence and Planning, 63, 55–69
Stephan, P. E., & El-Ganainy, A. (2007). The entrepreneurial puzzle: Explaining the
gender gap. Journal of Technology Transfer, 32(5), 475–487.
Steuer, J. (1992) Defining Virtual Reality: Dimensions Determining Telepresence.
The Journal of Communication, 42 (4), 73-93.
Street, R.L., Gold, W.R. and Manning, T. (1997) Health promotion and interactive
technology: theoretical applications and future directions, 1 ed. New Jersey:
Routledge.
Subhani, M.I., Hasan, S.A., Osman, S.W. and Rudhani, A. (2012) The Crux of Green
Marketing: An Empirical Effusive Study. European Journal of Social
Sciences, 27 (3), 425-435.
Sukoco, B.M. and Wu, W.-Y. (2011) The effects of advergames on consumer
telepresence and attitudes: A comparison of products with search and
experience attributes. Expert Systems with Applications, 38 (6), 7396-7406.
Summers, J. O. (1972). Media Exposure Patterns of Consumer Innovators. Journal of
Marketing, 36(1), 43-49.
Sundar, S.S. (2008) The MAIN model: A heuristic approach to understanding
technology effects on credibility. IN: Metzger, M.J. and Flanagin, A.J. (eds.)
Digital media, youth, and credibility. Cambridge, MA: The MIT Press, 72-
100.
211
Sundar, S.S. and Kim, J. (2005) Interactivity and persuasion: Influencing attitudes
with information and involvement. . Journal of Interactive Advertising, 5 (2),
6-29.
Sundar, S.S. and Nass, C. (2000) Source orientation in human–computer interaction:
Programmer, networker, or independent social actor? Communication
Research, 27, 683–703.
Sweeney, J.C., Soutar, G.N. and Johnson, L.W. (1999) The role of perceived risk in
the quality-value relationship: A study in a retail environment. Journal of
Retailing, 75 (1), 77-105.
Sweller, J. (1988) Cognitive load during problem solving: Effects on learning.
Cognitive Science, 12 (2), 257-285.
Sweller, J., Chandler, P., Tierney, P. and Cooper, M. (1990) Cognitive load as a factor
in the structuring of technical material. Journal of Experimental Psychology,
119 (2), 176-192.
Sweller, J., Van Merrienboer, J.G. and Paas, F.W.C. (1998) Cognitive Architecture
and Instructional Design. Educational Psychology Review, 10 (3), 251-296.
Taylor, S.E. and Schneider, S.K. (1989) Coping and the Simulation of Events. Social
Cognition, 7 (2), 174-194.
Tellis, G.J., Yin, E. and Bell, S. (2009) Global consumer innovativeness: Cross-
country differences and demographic commonalities. Journal of International
Marketing, 17 (2), 1-22.
Teresa, M. G. (2008) Virtual reality for collaborative e-learning. Computers and
Education, 50 (4), 1339-1353.
Thomas, B.A. (2006) Research Concepts for Management Studies, Routledge
Thorndike, E.L. (1931) Human Learning. New York: Century.
Tian, K.T., Bearden, W.O. and Hunter, G.L. (2001) Consumers' Need for Uniqueness:
Scale Development and Validation. Journal of Consumer Research, 28 (1),
50-66.
212
Tian, K.T. and Mckenzie, K. (2001) The Long-Term Predictive Validity of the
Consumers' Need for Uniqueness Scale. Journal of Consumer Psychology, 10
(3), 171-193.
Tran, H.L. (2010) Multimedia exemplification: The interplay of representation and
modality in the processing of online news and perceptions of international
issues. , PhD, University of North Carolina.
Tornatzky, G., Klein, K. (1982) Innovation characteristics and innovation adoption-
implementation: A meta-analysis of findings, IEEE Transactions on
Engineering Management EM 29(1) · February 1982
Tu, C.-H. (2000) On-line learning migration: from social learning theory to social
presence theory in a CMC environment. Journal of Network and Computer
Applications, 23 (1), 27-37.
Urban, G.L., Weinberg, B.D. and Hauser, J.R. (1996) Premarket Forecasting of
Really-New Products. Journal of Marketing, 60 (1), 47-60.
Vallerand, R.J. (1997) Toward a Hierarchical Model of Intrinsic and Extrinsic
Motivation IN: Zanna, M.P. (ed.) Advances in Experimental Social
Psychology, 29 ed. New York: Academic Press, 271-360.
Van Noort, G. (2012) Interactivity in Brand Web Sites: Cognitive, Affective, and
Behavioural Responses Explained by Consumers’ Online Flow Experience.
Journal of Interactive Marketing, 26 (4), 223-234.
Vandecasteele, B. and Geuens, M. (2010) Motivated consumer innovativeness:
concept, measurement, and validation. International Journal of Research in
Marketing, 27 (4), 308-318.
Venkatesh, A. and Nicosia, F. (1997) New Technologies For the Home-Development
of a Theoretical Model of Household Adoption and Use. Advances in
Consumer Research, 24, 522-528.
Venkatesh, V., & Morris, M. G. (2000). Why don’t men ever stop to ask for
directions? Gender, social influence, and their role in technology
acceptanceand usage behavior. MIS Quarterly, 24(1), 115–139.
Venkatesh, V., Morris, M. G., & Ackerman, P. L. (2000). A longitudinal field
investigation of gender differences in individual technology adoption
213
decisionmaking processes. Organizational Behavior and Human Decision
Processes, 83(1), 33–60.
Venkatraman, M.P. and Price, L.L. (1990) Differentiating between cognitive and
sensory innovativeness: Concepts, measurement, and implications. Journal of
Business Research Policy, 20 (4), 293-315.
Voss, K.E., Spangenberg, E.R. and Grohmann, B. (2003) Measuring the Hedonic and
Utilitarian Dimensions of Consumer Attitude. Journal of Marketing Research,
40 (3), 310-320.
Vygotsky, L.S. (1978) Mind and society: The development of higher mental
processes. Cambridge: MA: Harvard University Press.
Wakefield, R.L., Wakefield, K.L., Baker, J. and Wang , L.C. (2011) How website
socialness leads to website use. European Journal of Information Systems, 20,
118-132.
Wallace, W. (1983) Principles of Scientific Sociology. New York: Aldine.
Walther, J.B. (1992) Interpersonal effects in computer-mediated interaction: A
relational perspective. Communication Research, 19 (1), 52-90.
Walther, J.B. (1997) Group and interpersonal effects in international computer-
mediated collaboration. Human Communication Research, 23 (3), 342-369.
Wang, L., Baker, J., Wagner, J.A. and Wakefield, K. (2007) Can a Retail Web Site Be
Social? Journal of Marketing, 71 (3), 143-157.
Wang L., Fodness, D. (2010) Can avatars enhance consumer trust and emotion in
online retail sales? Int. J. Electronic Marketing and Retailing, Vol. 3, No. 4,
341
Wann, J. and Mon-Williams, M. (1996) What does virtual reality need?: human
factors issues in the design of three-dimensional computer environments.
International.Journal of Human-Computer.Studies, 44 (6), 829-847.
Watson, J.B. (1924) Behaviourism. New York: Norton.
Waytz, A., Cacioppo, J.T. and Epley, N. (2014) Who Sees Human? The Stability and
Importance of Individual Differences in Anthropomorphism. Perspective
Psychology Science, 5 (3), 219-232.
214
Weijters, B. and Roehrich, G. (2004) Consumer Innovativeness and Personal Values
of Openness and Self-enhancement. Unpublished manuscript.
Weiner, B. (1974) Achievement motivation and attribution theory. Morristown, N.J.:
General Learning Press.
Wheaton, B., Muthén, D.F., Alwin, B.B. and Summers, G.F. (1977) Assessing
reliability and stability in panel models. IN: Heise, D.R. (ed.) Sociological
methodology. San Francisco:: Jossey-Bass,, 84–136.
Whitley, B. E. Jr., (1997). Gender differences in computer-related attitudes and
behavior: A meta-analysis. Computers in Human Behavior, 13(1), 1–22.
Wilke, J. and Sorvillo, N. (2005) Targeting Early Adopters – A Means for New
Product Survival. ACNielsen BASES publication.
Wittrock, M.C. (1989) Generative processes of comprehension. Educational
Psychologist, 24, 345-376.
Wright, R. (2006). Consumer behaviour, Cengage Learning EMEA, p.90-92
Wrightsman, L. (1992) Assumptions about human nature: Implications for
researchers and practitioners 2ed. Newbury Park, CA: Sage.
Zakersalehi, M. and Zakersalehi, A. (2012) Consumers’ attitude and purchasing
intention toward green packaged foods; A Malaysian perspective. Paper
presented at International Conference on Economics Marketing and
Management, Singapore.
Zeff, R.L. and Aronson, B. (1997) Advertising on the Internet. New York: Wiley
Computer Publishing.
Zhang, D., Zhou, L., Briggs, R.O. and Nunamaker Jr, J.F. (2006) Instructional video
in e-learning: Assessing the impact of interactive video on learning
effectiveness. Information andamp; Management, 43 (1), 15-27.
Zhao, M., Hoeffler, S. and Dahl, D.W. (2009) The Role of Imagination-Focused
Visualization on New Product Evaluation. Journal of Marketing Research, 46
(1), 46-55.
215
216
Appendices
Appendix 1
RNPs initial selection
Product ReferenceBio-Robot Refrigerator Yanko DesignDismount Washer, Wash and Go Laundry Yanko DesignRibbon Electrolux Design LabHelpReaders Yanko designPowerkiss SpringwiseE-Tomb Yanko DesignDigital Makeup Mirror SpringwiseWashing Machine-in Wardrobe Yanko Designi dropper Yanko DesignGruve SpringwiseElectrolux Eco Cleaner Yanko DesignYour Liquid Email Box Yanko DesignLupe Yanko DesignTypeFace: SpringwiseDouble Drum Wash Yanko DesignReplicator of food Yanko DesignTenna Yanko Design
217
Appendix 2
Perceived Product Newness scale
Please answer the questions with Yes/No.
Have you seen this product before? Do you have any information about this product (According to (Gregan-
Paxton et al., 2002))?
Participants are asked to score each of the below statements on a five point Likert
scale ranging from “strongly disagree” to “strongly agree”:
I feel quite certain of the benefits I could expect to get if I adopted this product. (reverse coded)
I’m quite sure of what the relevant trade-offs are among the costs and benefits of buying and using this product. (reverse coded)
I’ll have to change my behavior significantly to attain the potential benefits of this new product.
Using this product would allow me to do things that I can’t easily do now. (Alexander et al., 2008)
218
Appendix 3
RNP Selection
Data Cleaning:
Incomplete questionnaires were deleted from the sample pool (2 questionnaires).
Furthermore, careless participants, those who completed the questionnaire in a very
short time (less than 2 standard deviation of the total participant’s mean) were deleted
(5 participants), and ones filling a similar answer for all 9 products were eliminated (3
participants). The complete sample of 50 participants is shown in Table 3.1.
Descriptive Analysis:
The database contained demographic information and the answers for six
questions in the questionnaire as follow:
1. I have not seen this product before (Yes/No)
2. I have no information about this product (Yes/No)
3. I feel quite certain of the benefits I could expect to get if I bought (adopted) this
product/service (5 point Likert Scale)
4. I’m quite sure of what the relevant trade-offs are among the costs and benefits
of buying and using this product/service (5 point Likert Scale)
5. I’ll have to change my behaviour significantly to attain the potential benefits of
this new product/service (5 point Likert Scale)
6. Using this new product/service would allow me to do things that I can’t easily
do now (5 point Likert Scale)
The first 2 questions with the answers of yes/no were number coded to yes=1,
no=0. The Likert style items were coded from strongly disagree=1, strongly agree=5
for questions 5and6 and reversed coded for questions 3and4. After recoding the
reverse-coded items, SPSS was used to analyse the data.
Demographic profile of Respondents:
219
50% of the participants were from India, 28% US, 10% Europe, 6% Canada, 2%
Latin America and 2% Asia.
Table 3.1: Complete Sample of 50.
Participant ID Location1 478003 India2 478011 US3 478020 India4 478023 US5 478024 India6 478026 India7 478029 India8 478030 India9 478035 India
10 478039 Canada11 478041 India12 478043 India13 478056 India14 478057 India15 478066 India16 478072 India17 478087 US18 478088 US19 478092 Canada20 478105 India21 478139 US22 478141 Ukraine23 478166 US24 478167 Macedonia25 478177 US26 478208 US27 478211 US28 478214 Jamaica29 478228 Canada30 478230 US31 478233 US32 478255 Singapore33 478266 India34 478276 India35 478278 US36 478281 India37 478286 US38 478305 India39 480504 India
220
40 480526 UK41 480539 Romania42 480559 India43 480571 India44 480584 India45 480919 US46 480930 India47 481006 Northern Ireland48 481358 India49 481366 US50 481375 India
Gender:
60% of the respondents were men and 40% were women.
Age:
The main age group comprised individuals between the ages of 20 and 29, which
included 44% of the sample. 30% of the respondents were between 30 and 39 years
of age. Out of the rest of participants, 12% were within 40-49, 8% were older than 50
year and 6% were younger than 19 years old.
221
Education Level:
The majority of the respondents were either graduate college or university (46%),
or undergraduate college or university (44%) levels. Only 6% were in the secondary
school group and 4% held a PhD.
Monthly Income:
The majority of the participants expressed an income of under $2000 (36% under
$1000 and 34% between $1000 to $2000). 22% were in the $2000-$3000 category
and only 2% were in $3000-$4000 group. The remaining (6%) were achieving an
income more than $4000.
222
223
Appendix 4
Website designs for vividness experimental conditions
Figure 4.1 Vividness manipulation
Low Vividness product 1
Low Vividness product 2
224
High Vividness product 1
225
High Vividness product 2
226
Appendix 5
Website designs for interactivity experimental conditions
Figure 5.1 Interactivity manipulations
Low Interactivity product 1
227
Low Interactivity product 2
High Interactivity product 1 (By clicking on the red button user will be directed to the interactive
3D design page)
228
Interactive 3D design page:
High Interactivity product 2 (By clicking on the red button, user will be directed to the
229
interactive 3D design page)
Interactive 3D design page:
230
231
Appendix 6
Website designs for interactivity experimental conditions
Figure 6.1 Anthropomorphism Manipulations
Human-like avatar for product 1
Static avatar for product 2
232
233
Appendix 7
Avatar Style
In this study, a combination of task oriented and social oriented avatar style is
employed, as Keeling (2010) indicated that “warm, enthusiastic style of interaction
enhance the benefits of the components of the task-oriented style” (Keeling 2010,
p.798). An avatar with Artificial Intelligence (AI) features is recruited that makes
communication with consumers more enjoyable. Furthermore, this results in a more
personal relationship between te avatar and consumer. One of the best companies
offering Avatar services is Sitepal. Sitepal provides many avatar styles in different
packages, which can be customized subject to user needs. A package with the
following features was selected for the purpose of this study:
Publish to Web pages, Emails, PowerPoint, Facebook Use SitePal's Client API Get Unlimited Email Support Use the Editor to Customize your Characters. Add Voice by Phone, Mic or Upload MP3 files. Create and Edit Unlimited Characters and Audios!
The advanced features are as below:
Text-to-Speech“With SitePal's Text-to-Speech (TTS) functionality, you can make your
SitePal characters speak written text in a natural sounding voice. Simply type in a
script in an input box to create audio for your character to speak.
SitePal supports TTS in two useful ways, which we refer to as static and
dynamic. Static TTS audio is audio which you assign to your Scene, to be played
back later on. Dynamic TTS audio is audio that is generated in real-time when a
visitor interacts with your web page. Using dynamic TTS allows your SitePal
character to speak a message that is generated "on the fly" for a particular visitor!
Note: dynamic TTS requires programming.
3D PhotoFaceAll packages get 70+ built-in 3D avatars. Creating your own 3D avatar from
an uploaded photo (using SitePal's facial recognition technology) is included from
the Silver package and up.
Lead Generation
234
Lead Generation allows your SitePal speaking character to collect contact
information from your web site visitors - information such as phone numbers or
email addresses. The collected information is emailed to you in real time, and
collected in a report you can review anytime. No programming is required to use
this feature.
FAQ SolutionUsing the FAQ feature your SitePal character can be preconfigured to answer
your website visitors' questions. Simply enter your questions and setup the
corresponding audio answers.
Publish to eBayWith SitePal's "publish to eBay" functionality, you can easily incorporate
your SitePal virtual salesperson into your eBay auctions without programming!
Imagine being able to add your own speaking salesperson to promote your
auctions.
Integrate with FlashYou can integrate your SitePal characters into Flash movies and create
seamless presentations. Advanced users can extend this functionality by using
SitePal'sActionScript API to create very rich interactive multimedia experiences.
SitePal supports AS2 and AS3 integration.
Artificial Intelligence (AI) / ChatSitePal's AI knowledge base comes loaded with more than 40,000 different
topics, which you can customize for your own needs. This technology lets your
users look for answers in a much more intuitive and enjoyable way by having a
two way conversation with your SitePal virtual character.
Remove SitePal BrandingEvery SitePal Scene briefly displays the SitePal logo when it is being loaded.
With the Gold Package and above you can remove SitePal branding from all of
your scenes. 10”
The package above was recruited for the price of $40 per month and the result is
as the pictured below:
10 www.sitepal.com
235
The avatar automatically starts explaining about the product (the same text used
in all other website formats), then the user is able to type in questions into the avatar’s
“Ask me” box. The avatar is able to communicate with the user around general
knowledge area comfortably (such as weather, greetings etc.). Furthermore, there is a
limited database, which is designed by the author for the purpose of the dissertation,
containing relevant information about the product with possible questions/keywords
typed in and appropriate replies by avatar.
236
Appendix 8
Measurement Scales
Vividness Scale
Response options: Strongly Agree 1 ------------- 4 (Neutral) ------------- 7 Strongly Disagree
Please select the appropriate answer for each statement.
I thought this site was very colourful.
I could perceive a lot of dynamism in this site.
I thought this site was visually attractive.
I found many graphical elements on this site.
The quality of the images on the site was good enough for me to actually see the
product featured.
The quality of the images on the site was good enough for me to visualize the
physical product featured.*
Overall, I thought this site was highly vivid visually.
The content of advertisement can stimulate my audio senses (hearing) (Sukoco
and Wu, 2011)
* The original scale was: “The quality of the images on the site was good enough
for me to actually see and visualize the physical product featured.” which was divided
into two questions in order to make the sentence more clear for participants.
Interactivity Scale
Please select the appropriate rating for each statement.
Response options: Strongly Agree 1 --------------- 4 (Neutral) --------------- 7 Strongly Disagree
(Adopted from Kalyanaraman and Sundar, 2006)
I interacted with the content of this website
I interacted with the structure of this website
(from Sundar and Kim, 2005)
This website enabled two-way communication
This website enabled synchronous (simultaneous) communication
This website enabled active control
237
Anthropomorphic Scale
Mindful anthropomorphism was measured by asking: whether participants
perceived the website as being “human-like/machinelike, natural/unnatural or
lifelike/artificial” on a 10-point semantic differential scales (Powers and Kiesler,
2006) (α = .86, M = 15.53, SD = 5.94, Kim and Sundar, 2012).
Please choose a rating between 1-10 for each item.
I perceive the website to be:
1 10
Machinelike Human-like
1 10
Unnatural Natural
1 10
Artificial lifelike
For measuring mindless anthropomorphism, the four items of likeable, sociable,
friendly and personal were rated using a 10-point scale (1=very poorly and 10=very
well), (α = 0.85, M = 24.90, SD = 7.25, Kim and Sundar, 2012).
How well each adjective describes the website (1=very poorly and 10=very well)
1- Likeable2- Sociable3- Friendly4- Personal
Comprehension Scale
Please complete the following statement
“I found the product description:
1- Difficult to understand/easy to understand2- Confusing/straightforward” (from Phillips, 2000, p.20)
“Please indicate the extent to which you agree with the following statements (1=
totally disagree, 7=totally agree):
238
1- After reading/listening to the product description, I have a very strong understanding of how this product works.
2- After reading/listening to the product description I would be able to use the product.
3- After reading/listening to the product description, I understand what the main features of this product are.
4- After reading/listening to the product description, I understand what the main benefits of this product are.”
* “the product description” is a replacement for the phrase “the advert” and “reading/listening to”
is a replacement for the phrase “reading” in the original scale (from Hoeffler, 2003)
Attitude Scale
I evaluate the usage of the product as:
Purchase Intention Scale
Please select the appropriate rating for each statement (1= totally disagree, 4=
Neutral, 7=totally agree)
(1) I will purchase the RNP
(2) Given a choice, my friends will choose the RNP
(3) There is a strong likelihood that I will buy the RNP
(4) I would like to recommend the RNP to my friends.
Consumer Innovativeness Scale
Please select the appropriate rating for each statement (1= totally disagree,
5=totally agree)
20-item Motivated Consumer Innovativeness (MCI) scale.
239
HED/UT Scale1 ---------------------------- 7
Not Fun FunExciting DullNot Delightful DelightfulThrilling Not thrillingUnenjoyable Enjoyable
Effective 1 ---------------------------- 7 IneffectiveUnhelpful HelpfulFunctional Not functionalUnnecessary NecessaryPractical Impractical
Factor Item
Social I love to use innovations that impress others.
I like to own a new product that distinguishes me from others who do not own
this new product.
I prefer to try new products with which I can present myself to my friends and
neighbours.
I like to outdo others, and I prefer to do this by buying new products, which my
friends do not have.
I deliberately buy novelties that are visible to others and which command
respect from others.
Functional If a new time-saving product is launched, I will buy it right away.
If a new product gives me more comfort than my current product, I would not
hesitate to buy it.
If an innovation is more functional, then I usually buy it.
If I discover a new product in a more convenient size, I am very inclined to buy
this.
If a new product makes my work easier, then this new product is a “must” for
me.
Hedonic Using novelties gives me a sense of personal enjoyment.
It gives me a good feeling to acquire new products.
Innovations make my life exciting and stimulating.
Acquiring an innovation makes me happier.
The discovery of novelties makes me playful and cheerful.
Cognitive I mostly buy those innovations that satisfy my analytical mind.
I find innovations that need a lot of thinking intellectually challenging and
therefore I buy them instantly.
I often buy new products that make me think logically.
I often buy innovative products that challenge the strengths and weaknesses of
my intellectual skills.
I am an intellectual thinker who buys new products because they set my brain
to work.
240
Appendix 9
Demographic Profile of Respondents for the sample of 800
Respondent’s Gender
50.75% of the respondents are Male and 49.25% are female.
Figure 9.1 Respondent’s gender
Respondent’s Age
The sample is a fairly young sample with 20-29 years old in first place
by 40.25% and 30-39 years old in second place by 31.25%.
Figure 9.2 Respondent’s age
241
Respondent’s monthly Income
Most of the respondents gain a salary between $1000 and $1999
(24.63%) whereas 22.38% claimed to earn less than $999, 16.63% more
than $4000 and only 14.63% in the range of $3000-$3999.
Figure 9.3 Respondent’s Monthly Income
Education
The respondents are highly educated, with 51.75% holding an
undergraduate degree and 16% a Graduate degree. 32% were graduated up
to high school.
Figure 9.4 Respondent’s Education
242
Appendix 10
Defining individual constructs
The observed indicators used in the model are as detailed in Table 10.1.
Table 10.1: The Observed indicators
Construct Response format Description Item Code
Media Richness (MR)
1-7 Likert
Strongly Agree-Strongly Disagree
I thought this site was very colourful. MRcolor
I could perceive a lot of dynamism in this site. MRdyn
I thought this site was visually attractive. MRVatt
I found many graphical elements on this site. MRgraph
The quality of the images on the site was good enough for me to actually see the physical product featured.
MRimagsee
The quality of the images on the site was good enough for me to actually visualize the physical product featured.
MRimgvis
Overall, I thought this site was highly vivid visually. MRvivid
The content of advertisement can stimulate my audio senses (hearing) MRaudio
Anthropomorphism 10 points semantic differential
I perceive the website to be:
Machinelike ----- Human-like
AntMF1
I perceive the website to be:
Unnatural ----- Natural
AntMF2
I perceive the website to be:
Artificial ----- Lifelike
AntMF3
1 – 10 Likert
Very Poor – Very Well
How well each adjective describes the website: Likeable AntML1
How well each adjective describes the website: Sociable AntML2
How well each adjective describes the website: Friendly AntML3
How well each adjective describes the website: Personal AntML4
Interactivity (Intr) 1-7 Likert
Strongly Agree-Strongly Disagree
I interacted with the content of this website Intcont
243
Construct Response format Description Item Code
I interacted with the structure of this website Intstru
This website enabled two-way communication Int2way
This website enabled synchronous (simultaneous) communication
Intsynch
This website enabled active control Intcont_A
Attitude (Attit) 10 points semantic differential
I evaluate the usage of the product as:
Not Fun ----- Fun
Att1
I evaluate the usage of the product as:
Exciting ----- Dull
Att2R
I evaluate the usage of the product as:
Not Delightful -----Delightful
Att3
I evaluate the usage of the product as:
Thrilling ----- Not Thrilling
Att4R
I evaluate the usage of the product as:
Unenjoyable ----- Enjoyable
Att5
I evaluate the usage of the product as:
Effective ----- Ineffective
Att6R
I evaluate the usage of the product as:
Unhelpful ----- Helpful
Att7
I evaluate the usage of the product as:
Functional ----- Not Functional
Att8R
I evaluate the usage of the product as:
Unnecessary ----- Necessary
Att9
I evaluate the usage of the product as:
Practical ----- Impractical
Att10R
Comprehension (Comp) 10 points semantic differential
I found the product description:
Difficult to understand ----- Easy to understand
Comdes1
244
Construct Response format Description Item Code
I found the product description:
Confusing ----- Straightforward
Compdes2
After reading/listening to the product description, I have a very strong understanding of how this product works
Compundut
After reading/listening to the product description I would be able to use this product.
Compuse
After reading/listening to the product description, I understand what the main features of this product are
Compundfeat
After reading/listening to the product description, I understand what the main benefits of this product are
Compundbenf
Purchase Intention (Pint) 1-7 Likert
Strongly Agree-Strongly Disagree
I will purchase the product PI
Given a choice, my friends will choose the product Pifr
There is a strong likelihood that I will buy the product Pilikelihd
I would like to recommend the product to my friends PIRecom
Innovativeness
(Innov)
Social Innovativeness (InSoc)
1-7 Likert
Strongly Agree-Strongly Disagree
I love to use innovations that impress others. InSoc1
I like to own a new product that distinguishes me from others who do not own this new product.
InSoc2
I prefer to try new products with which I can present myself to my friends and neighbours.
InSoc3
I like to outdo others, and I prefer to do this by buying new products which my friends do not have.
InSoc4
I deliberately buy novelties that are visible to others and which command respect from others.
InSoc5
Functional Innovativeness (InFun)
1-7 Likert
Strongly Agree-Strongly Disagree
If a new time-saving product is launched, I will buy it right away.
InFun1
If a new product gives me more comfort than my current product, I would not hesitate to buy it.
InFun2
If an innovation is more functional, then I usually buy it. InFun3
If I discover a new product in a more convenient size, I am very inclined to buy this.
InFun4
If a new product makes my work easier, then this new product is a “must” for me.
InFun5
Hedonic Innovativeness (InHed)
1-7 Likert
Strongly Agree-Strongly Disagree
Using novelties gives me a sense of personal enjoyment. InHed1
It gives me a good feeling to acquire new products. InHed2
Innovations make my life exciting and stimulating. InHed3
Acquiring an innovation makes me happier. InHed4
The discovery of novelties makes me playful and cheerful.
InHed5
245
Construct Response format Description Item Code
Cognitive Innovativeness (InCog)
1-7 Likert
Strongly Agree-Strongly Disagree
I mostly buy those innovations that satisfy my analytical mind.
InCog1
I find innovations that need a lot of thinking intellectually challenging and therefore I buy them instantly.
InCog2
I often buy new products that make me think logically. InCog3
I often buy innovative products that challenge the strengths and weaknesses of my intellectual skills.
InCog4
I am an intellectual thinker who buys new products because they set my brain to work.
InCog5
Figure 10.1 The new model after dividing the Innovativeness Scale
246
Appendix 11
CFA Model FitTable 11.1: Summary of items removed from the model.
Round Construct Reason
1 MRaudio Factor loading lower than 0.5
2 MRQimgsee MRQimgvis According to MIs
3 1. PIRecom2. Att6 According to MIs
4 Att10-8-7 According to MIs
5 Att9According to MIs and standard residuals covariances
6 Intcont Mgraph According to MIs
7
Compundbenf InSoc5 InCog2 InFun1 AntMF2
According to MIs and standard residuals covariances
After each amendment to the model, a model fit test was performed. Table 7.2 is a
review of model fit indices in each stage of the model improvement.
Table 11.2: Summary of model fit in each stage.
Round Model fitting result1 Chi-square: 3465.137
DF: 1238CMIN/DF: 2.799GFI: 0.649AGFI: 0.609CFI: 0.822RMSEA: 0.078PCLOSE: 0.000
2 Chi-square: 3020.92DF: 1139CMIN/DF: 2.652GFI: 0.675AGFI: 0.636CFI: 0.843RMSEA: 0.074PCLOSE: 0.000
3 Chi-square: 2645.997DF: 1044CMIN/DF: 2.534GFI: 0.707AGFI: 0.670CFI: 0.857
247
RMSEA: 0.072PCLOSE: 0.000
4 Chi-square: 2096.102DF: 909CMIN/DF: 2.306GFI: 0.758AGFI: 0.724CFI: 0.886RMSEA: 0.066PCLOSE: 0.000
5 Chi-square: 1936.565DF: 866CMIN/DF: 2.236GFI: 0.766AGFI: 0.732CFI: 0.895RMSEA: 0.064PCLOSE: 0.000
6 Chi-square: 1614.195DF: 783CMIN/DF: 2.06GFI: 0.788AGFI: 0.756CFI: 0.914RMSEA: 0.060PCLOSE: 0.000
7 Chi-square: 1174.860DF: 630CMIN/DF: 1.86GFI: 0.82AGFI: 0.8 CFI: 0.936RMSEA: 0.05PCLOSE: 0.095
248
Appendix 12
Test of Unidimensionality and Homogeneity for individual scales
Table 12.1 KMO and Bartlett's Test
Scales KMO Measure of Sampling Adequacy.
Approximate Chi-Square
Bartlett's Test of Sphericity
df
Significance
Media Richness .829 1179.701 28 .000
Interactivity .792 1076.038 10 .000
Anthropomorphism
.865 1717.379 21 .000
Attitude .901 2099.135 45 .000
Comprehension .872 1466.275 15 .000
Purchase Intention .823 1280.786 6 .000
Innovativeness .920 4420.175 190 .000
249
Appendix 13
Final Scales Evaluation for loadings and Homogeneity
Table 13.1: CFA result for Vividness
Scale item Factor LoadingMRvivid 0.799MRVatt 0.839MRDyn 0.782MRColor 0.652four items were removed during CFA. KMO=0.799; Bartlett’s test= 525.820; df:6 , p= 0.000After adding MRaudioKMO=0.812; Bartlett’s test= 573.308; df:10 , p= 0.000
Table 13.2: CFA result for Interactivity
Scale item Factor LoadingIntcont_A 0.764Intstru 0.670Int2way 0.877Intsynch 0.907One item removed during CFA.KMO=0.784; Bartlett’s test= 688.094; df: 6, p=0.000
Table 13.3: CFA result for Anthropomorphism
Scale item Factor LoadingAntML4 0.866AntML3 0.844AntML2 0.840AntML1 0.777AntMF1 0.646AntMF3 0.669One item removed during CFA.KMO= 0.848; Bartlett’s test= 1355.134; df: 15, p= 0.000
Table 13.4: CFA result for Attitude
Scale item Factor LoadingAtt1 0.862Att2 0.813Att3 0.894Att4 0.767Att5 0.8305 items removed during CFAKMO= .871; Bartlett’s test= 1092.363; df: 10, p= 0.000
250
Table 13.5: CFA result for Comprehension
Scale item Factor LoadingCompuse 0.734Compundfeat 0.787Compdes2 0.897Compundut 0.846Comdes1 0.913One item was removed during CFAKMO= .852; Bartlett’s test= 1177.078; df: 10, p=0.000
Table 13.6: CFA result for Purchase Intention
Scale item Factor LoadingPilikelihd 0.955Pifr 0.829PI 0.954One item removed during CFA.KMO= 0.736; Barttlet’s test= 843.545; df:3, p=0.000
Table 13.7: CFA result for Consumer Innovativeness
Scale item Factor LoadingInSoc1 0.799InSoc2 0.903InSoc3 0.886InSoc4 0.787InFun2 0.734InFun3 0.761InFun4 0.701InFun5 0.748InHed1 0.706InHed2 0.793InHed3 0.834InHed4 0.881InHed5 0.800InCog1 0.773InCog3 0.878InCog4 0.926InCog5 0.891Three items removed during CFA.KMO= 0.911; Bartlett’s test= 3606.252; df: 136, p= 0.000
251
Appendix 14
Scales test of normality
o Media Richness
Figure 14.1 displays the frequency distribution of the Media Richness scale. The
distribution appears to be close to normality. The result of the Kolmogorov-Smirnov
(KS) test is significant (z=0.068, p=0.002), indicating that the observed distribution is
non-normal. On the other hand, Shapiro-Wilk (SW) is non-significant (w=0.991,
p=0.055), indicating the distribution is normal. The Shapiro-Wilk test appears to be
the most powerful test for all types of distribution and sample sizes (especially for
sample sizes larger than 30) in comparison with other tests such as Kolmogorov-
Smirnov and Lilliefors (Razali and Wah, 2011). Therefore, according to SW, the
variable is normal. However, for samples larger than 200, in order to make sure the
sample is normal, rather than looking only at SW or KS tests, the kurtosis and
skewness needs to be examined (Field, 2005, p. 72) as KS test is extremely sensitive
to minor departure from normality (Sharma, 1996) and also violating the assumption
of normality is quite common in larger samples (Pallant, 2005, p.57). Moreover, it has
been argued that skewness and kurtosis below the threshold of 1.96 will not cause
significant issues in the dataset (Field, 2005, p. 72 cited in Feiereisen, 2009, p.226-
227).
Looking at the skewness and Kurtosis values, they are well below the threshold of
1.96 (-0.208, -0.089 respectively); it is considered that there are no serious concerns
regarding the normality of this variable and it is preserved without transformation for
future analysis.
252
Figure 14.1. Histogram of Media Richness
o Interactivity
Figure 14.2 displays the frequency distribution of the Interactivity scale. A skew
towards lower values is evident. Both KS (KS) (z=.117, p=.000) and SW tests
(w=.938, p=.000) are significant for Interactivity. The variable returned values of
0.598 and -0.361 for skewness and kurtosis respectively. These values are below the
threshold of 1.96 therefore there is no serious concern regarding the normality of this
scale and it will be retained for future analysis.
Figure 14.2. Histogram of Interactivity
o Anthropomorphism
Figure 14.3 displays the frequency distribution of the Anthropomorphism scale.
The distribution appears to be close to normality, although some skew towards the
lower values. Both KS (KS) (z=.0.68, p=.024) and SW tests (w=.986, p=.049) are
253
significant for Anthropomorphism. The variable returned values of 0.150 and -0.621
for skewness and kurtosis respectively. These values are below the threshold of 1.96
therefore there is no serious concern regarding the normality of this scale and it will
be retained for future analysis. Figure 14.3. Histogram of Interactivity
o Purchase Intention
Figure 14.4 displays the frequency distribution for the Purchase Intention scale.
The distribution appears to be close to normality, although some skew towards the
lower values. A significant KS test (z=.086, p=.000) and SW test (w=.0959, p=.000)
are returned. However the skewness and kurtosis values are both lower than the
threshold of 1.96 (0.598, -0.361 respectively). Therefore there are no serious concerns
regarding scale’s normality and the scale is retained for future analysis.
254
Figure 14.4. Histogram of Purchase Intention
o Comprehension
Figure 14.5 displays the frequency distribution of the Comprehension scale. A
skew towards higher values is evident. Both KS (KS) (z=.107, p=.000) and SW tests
(w=.942, p=.000) are significant for the Comprehension scale. The variable returned
values of -0.762 and -0.021 for skewness and kurtosis respectively. These values are
below the threshold of 1.96, therefore there is no serious concern regarding the
normality of this scale and it will be retained for future analysis.
Figure 14.5. Histogram of Comprehension
o Attitude
Figure 14.6 displays the frequency distribution for the Attitude scale. The
distribution appears to be close to normality, although some skew towards the lower
255
values. A significant KS test (z=.092, p=.000) and SW test (w=.0981, p=.000) are
returned. However the skewness and kurtosis values are both lower than the threshold
of 1.96 (0.335, -0.063 respectively). Therefore there is no serious concerns regarding
scales normality and the scale is retained for future analysis.
Figure 14.6. Histogram of Attitude towards product
o Innovativeness Social Innovativeness
Figure 14.7 displays the frequency distribution for the Social Innovativeness
scale. The distribution appears to be close to normality, although some skew towards
the higher values. A significant KS test (z=.072, p=.001) and SW test (w=.0978,
p=.000) are returned. However the skewness and kurtosis values are both lower than
the threshold of 1.96 (0.076, -0.762 respectively). Therefore there is no serious
concern regarding scale’s normality and the scale is retained for future analysis.
256
Figure 14.7. Histogram of Social Innovativeness
Functional Innovativeness
Figure 14.8 displays the frequency distribution for the Functional Innovativeness
scale. The distribution appears to be close to normality, although some skew towards
the higher values. A significant KS test (z=.107, p=.000) and SW test (w=.963,
p=.000) are returned. However the skewness and kurtosis values are both lower than
the threshold of 1.96 (-0.691, 1.003 respectively). Therefore there is no serious
concern regarding scale’s normality and the scale is retained for future analysis.
Figure 14.8. Histogram of Functional Innovativeness
Hedonic Innovativeness
Figure 14.9 displays the frequency distribution for the Hedonic Innovativeness
scale. The distribution appears to be close to normality, although some skew towards
257
the higher values. A significant KS test (z=.094, p=.000) and SW test (w=0.971,
p=.000) are returned. However the skewness and kurtosis values are both lower than
the threshold of 1.96 (-0.466, -0.369 respectively). Therefore there is no serious
concern regarding scale’s normality and the scale is retained for future analysis.
Figure 14.9. Histogram of Hedonic Innovativeness
Cognitive Innovativeness
Figure 14.10 displays the frequency distribution for the Cognitive Innovativeness
scale. The distribution appears to be close to normality. A significant KS test (z=.053,
p=.043) and SW test (w=0.987, p=.008) are returned. However the skewness and
kurtosis values are both lower than the threshold of 1.96 (-0.110, -0.492 respectively).
Therefore there is no serious concern regarding scale’s normality and the scale is
retained for future analysis.
258
Figure 14.10. Histogram of Cognitive Innovativeness
Appendix 15
259
Focus Group
There were six participants, aged between 19 and 43 years (2 males). The focus
group started with a brief introduction to the study. The participants were then guided
to take a look at the websites, and were introduced to the avatar. They learnt about the
process of participation in the survey and how participants can interact with the
avatar. They were then asked: for their thoughts on the possible questions a
participant may ask from the avatar. They were asked to consider the limited time,
and possible intentions of the participants, as they would be being paid to complete
the survey.
The focus group came up with many possible questions; some indicated their
curiosity about the product. The group was then asked to think about other aspects of
the product, rather than the hedonic aspect or the functionality. They came up with
following questions:
Dismount Washer
1. How does it work?
2. Is it noisy?
3. How quickly does the machine, wash the cloths?
4. What is the capacity of the machine?
5. Can you stop it if you need to add more clothing?
6. How heavy is this?
7. How do you clean the washing machine?
8. Is it safe for children?
9. How do you use the product?
10. Where do you put the detergent?
11. Can you use softener with this machine?
12. How big is the product?
13. How does it get fixed to the wall?
14. How does the door secure?
15. How do you program it?
16. How does the water drain?
17. Can you use it in the office?
260
18. Is it portable?
19. Is the end result comparable/better than conventional washing machine?
Washing Machine in Wardrobe
1. How big is it?
2. Is the end product better/comparable with conventional dry cleaning?
3. How heavy is it?
4. Is it noisy?
5. Can you carry it?
6. How does it fit in the wardrobe?
7. How much space does it take?
8. How long does it take to dry clean?
9. Does it come in other colours?
10. Is it easy to use?
11. How many programs does it have?
12. Can it dry clean big items?
13. Does it get hot?
14. Does it smell?
The focus group was then encouraged to think about the avatar, and if potential
participants will ask the avatar any questions which were not directly related to the
product.
Avatar
1. What is your name?
2. Do you like working?
3. Where are you from?
4. What is the weather like?
5. How many people have you spoken to today?
6. Who are you?
7. What is your job?
8. Are you married?
9. How long you have been working for this company?
10. What is your favourite colour
261
11. How old are you?
12. Is it the best product for me?
13. What do you think about this product?
14. Will you buy this product?
15. Do you have this product?
16. Do you like this product?
17. Do you recommend this product to a friend?
18. Are you sitting down?
19. How did you find out about this product?
20. Do you have any children?
21. Do you ever rest?
22. Do you have a break?
The group was then asked to think about how the avatar could start talking about
the product, and how she could introduce herself, in the most natural way. The
recommendations are as below:
They suggested if she can start by the following sentence:
“Hello, I am Julie ….”
At the end of the introduction, they suggested that she could add:
“If you want to know more about the …. I will try to answer your questions.”
The group suggested that I could add some personal touches to the introduction
speech of the avatar such as “green is my favourite colour” (when talking about
the colour of the product).
They suggested that the avatar was better to moved back, and to be smaller with
more visibility of the head and neck. They suggested it was better for the avatar to
be in an office setting.
They suggested that participants might ask inappropriate questions, as some might
take the opportunity to have some fun! Therefore, I could add a database of
possible inappropriate keywords and a default answer of:
262
I don’t think this is an appropriate question; can we change the subject back to the
product?
263
Appendix 16
Website designs for experimental conditions
Base website for product 1
Low Interactivity for product 2
264
High Interactivity for product 1
265
266
With a 3D design showing when clicking on the red button
Static avatar for product 2
267
Human-like avatar for product 1
268
269
Appendix 17
Measurement Scales
Comprehension Scale
Please complete the following statement
“I found the product description:
3- Difficult to understand/easy to understand
4- Confusing/straightforward” (from Phillips, 2000 p20)
“Please indicate the extent to which you agree with the following statements (1=
totally disagree, 7=totally agree):
5- After reading/listening to the product description, I have a very strong
understanding of how this product works.
6- After reading/listening to the product description I would be able to use the
product.
7- After reading/listening to the product description, I understand what the main
features of this product are.
8- After reading/listening to the product description, I understand what the main
benefits of this product are.”
* “the product description” is a replacement for the phrase “the advert” and “reading/listening to”
is a replacement for the phrase “reading” in the original scale (from Hoeffler, 2003)
Attitude Scale
I evaluate the usage of the product as:
270
HED/UT Scale1 ---------------------------- 7
Not Fun FunExciting DullNot Delightful DelightfulThrilling Not thrillingUnenjoyable Enjoyable
Effective 1 ---------------------------- 7 IneffectiveUnhelpful HelpfulFunctional Not functionalUnnecessary NecessaryPractical Impractical
Purchase Intention Scale
Please select the appropriate rating for each statement (1= totally disagree, 4=
Neutral, 7=totally agree)
(1) I will purchase the RNP
(2) Given a choice, my friends will choose the RNP
(3) There is a strong likelihood that I will buy the RNP
(4) I would like to recommend the RNP to my friends.
Anthropomorphism scale
Please choose a rating between 1-10 for each item.
I perceive the website to be:
1 10
Machinelike Human-like
1 10
Unnatural Natural
1 10
Artificial lifelike
How well each adjective describes the website (1=very poorly and 10=very well)
5- Likeable
6- Sociable
7- Friendly
8- Personal
271
Appendix 18
Defining individual constructs
The model is designed according to Hair et al (2010) steps. Step 1 is to define
individual constructs. Table below indicates the constructs and the observed
indicators.
Construct Scale Type Description ItemMindful Anthropomorphism (AntMF)
10 points semantic differential
Please choose a number rating between 1-10 to indicate how you perceived the website.Machinelike ---- Human-like
Ant MF1
Please choose a number rating between 1-10 to indicate how you perceived the website.Unnatural ---- Natural
AntMF2
Please choose a number rating between 1-10 to indicate how you perceived the website.Artificial ---- Lifelike
AntMF3
Mindless Anthropomorphism (AntML)
1 – 10 LikertVery Poor – Very Well
Looking at each of the adjectives below indicate how well they describe the website.Likeable
AntML1
1 – 10 LikertVery Poor – Very Well
Sociable AntML2
1 – 10 LikertVery Poor – Very Well
Friendly AntML3
1 – 10 LikertVery Poor – Very Well
Personal AntML4
Attitude (Att) 10 points semantic differential
I evaluate the usage of the product as:Not Fun ----- Fun
Att1
I evaluate the usage of the product as:Exciting ----- Dull
Att2R
I evaluate the usage of the product as:Not Delightful -----Delightful
Att3
I evaluate the usage of the product as:Thrilling ----- Not Thrilling
Att4R
I evaluate the usage of the product as:Unenjoyable ----- Enjoyable
Att5
I evaluate the usage of the product as:Effective ----- Ineffective
Att6R
I evaluate the usage of the product as:Unhelpful ----- Helpful
Att7
I evaluate the usage of the product as:Functional ----- Not Functional
Att8R
I evaluate the usage of the product as:Unnecessary ----- Necessary
Att9
I evaluate the usage of the product as:Practical ----- Impractical
Att10R
272
Comprehension (Comp)
10 points semantic differential
I found the product description:Difficult to understand ----- Easy to understand
Compdes1
10 points semantic differential
I found the product description:Confusing ----- Straightforward
Compdes2
1-7 Likert Strongly Agree-Strongly Disagree
After reading/listening to the product description, I have a very strong understanding of how this product works
Comp3
1-7 Likert Strongly Agree-Strongly Disagree
After reading/listening to the product description I would be able to use this product.
Comp4
1-7 Likert Strongly Agree-Strongly Disagree
After reading/listening to the product description, I understand what the main features of this product are
Comp5
1-7 Likert Strongly Agree-Strongly Disagree
After reading/listening to the product description, I understand what the main benefits of this product are
Comp6
Purchase Intention (PI)
1-7 Likert Strongly Agree-Strongly Disagree
I will purchase the product PI1
1-7 Likert Strongly Agree-Strongly Disagree
Given a choice, my friends will choose the product
PI2
1-7 Likert Strongly Agree-Strongly Disagree
There is a strong likelihood that I will buy the product
PI3
1-7 Likert Strongly Agree-Strongly Disagree
I would like to recommend the product to my friends
PI4
Step 2 is to develop the overall measurement model. The initial and final
measurement models are shown below.
273
The initial CFA Model.
274
Final Model
275
Appendix 19
CFA Model Fit
Table 19.1: CFA result for Mindful Anthropomorphism
Scale item Factor LoadingAntMF1C 0.799AntMF2C 0.751AntMF3C 0.878No item is removed during CFA. KMO=0.715; Bartlett’s test= 653.869; df:3 , p= 0.000
Table 19.2: CFA result for Mindless Anthropomorphism
Scale item Factor LoadingAntML2C 0.918AntML3C 0.904AntML4C 0.831One item removed during CFA.KMO=0.746; Bartlett’s test= 1069.085; df:3, p=0.000
Table 19.3: CFA result for Attitude
Scale item Factor LoadingAtt1 0.813Att2R 0.712Att3 0.893Att4R 0.686Att5 0.859Five items removed during CFAKMO= .849; Bartlett’s test= 1541.108; df: 10, p= 0.000
Table 19.4: CFA result for Comprehension
Scale item Factor LoadingCompdes1C 0.789Compdes2C 0.784Comp3 0.875Comp4 0.806Comp5 0.825Comp6 0.822No item was removed during CFAKMO= .856; Bartlett’s test= 2670.894; df: 15, p=0.000
276
Table 19.5: CFA result for Purchase Intention
Scale item Factor LoadingPI1 0.953PI2 0.830PI3 0.961One item removed during CFA.KMO= 0.736; Barttlet’s test= 1442.672; df:3, p=0.000
277
Appendix 20
Test of normality for each scale
Figure 20.1 displays the frequency distribution of the Mindful Anthropomorphism
scale. The distribution appears to be close to normality. The result of the
Kolmogorov-Smirnov (KS) test is significant (z=0.086, p=0.000) and Shapiro-Wilk
(SW) is also significant (w=0.980, p=0.000), which indicates that the observed
distribution is non-normal. Shapiro-Wilk test appears to be the most powerful test for
all types of distribution and sample sizes (especially for sample sizes larger than 30),
in comparison with other tests such as Kolmogorov-Smirnov and Lilliefors (Razali
and Wah, 2011); therefore according to SW, the variable is normal. However, for
samples larger than 200, in order to make sure the sample is normal, rather than
looking only at SW or KS tests, the kurtosis and skewness needs to be examined
(Field, 2005, p.72) as KS test is extremely sensitive to minor departure from
normality (Sharma, 1996) and also violating the assumption of normality is quite
common in larger samples (Pallant, 2005, p.57). Moreover, it has been argued that
skewness and kurtosis below the threshold of 1.96 will not cause significant issues in
the dataset (Field, 2005, p.72 cited in Feiereisen 2009, p.226-227).
Looking at the skewness and kurtosis values, they are well below the threshold of
1.96 (0.243, -0.700 respectively); it is considered therefore that there is no serious
concern regarding the normality of this variable, and it is preserved without
transformation for future analysis.Figure 20.1. Histogram of mindful anthropomorphism
278
Figure 20.2 displays the frequency distribution of the AntML scale. The
distribution appears to be close to normality. Both KS (KS) (z=.058, p=.000) and SW
test (w=.984, p=.000) are significant for AntML. The variable returns values of -0.216
and -0.465 for skewness and kurtosis respectively. These values are below the
threshold of 1.96 therefore there is no serious concern regarding the normality of this
scale and it will be retained for future analysis.
Figure 20.2. Histogram of Mindless Anthropomorphism
Figure 20.3 displays the frequency distribution of the Comp scale. A skew
towards higher values is evident. Both KS (KS) (z=.101, p=.000) and SW test
(w=.928, p=.000) are significant for Comp. The variable returns values of -1.032 and
1.044 for skewness and kurtosis respectively. These values are below the threshold of
1.96 therefore there is no serious concern regarding the normality of this scale and it
will be retained for future analysis.
279
Figure 20.3. Histogram of Comprehension
Figure 20.4 displays the frequency distribution of the PI scale. The distribution
appears to be normal. Both KS (KS) (z=.091, p=.000) and SW test (w=.960, p=.000)
are significant for PT. The variable returns values of -0.031 and -1.028 for skewness
and kurtosis respectively. These values are below the threshold of 1.96 therefore there
is no serious concern regarding the normality of this scale and it will be retained for
future analysis. Figure 20.4. Histogram of Purchase Intention
Figure 20.5 displays the frequency distribution of the Attitude scale. The
distribution appears to be normal. Both KS (KS) (z=.073, p=.000) and SW test
(w=.981, p=.000) are significant for Attitude. The variable returns values of -0.356
and -0.089 for skewness and kurtosis respectively. These values are below the
280
threshold of 1.96 therefore there is no serious concern regarding the normality of this
scale and it will be retained for future analysis.
Figure 20.5. Histogram of Attitude
281
Appendix 21
Measurement Scales
Consumer Innovativeness Scale
Please select the appropriate rating for each statement (1= totally disagree,
7=totally agree)
20-item Motivated Consumer Innovativeness (MCI) scale.Factor ItemSocial I love to use innovations that impress others.
I like to own a new product that distinguishes me from others who do not own this new product.I prefer to try new products with which I can present myself to my friends and neighbours.I like to outdo others, and I prefer to do this by buying new products which my friends do not have.I deliberately buy novelties that are visible to others and which command respect from others.
Functional If a new time-saving product is launched, I will buy it right away.If a new product gives me more comfort than my current product, I would not hesitate to buy it.If an innovation is more functional, then I usually buy it.If I discover a new product in a more convenient size, I am very inclined to buy this.If a new product makes my work easier, then this new product is a “must” for me.
Hedonic Using novelties gives me a sense of personal enjoyment.It gives me a good feeling to acquire new products.Innovations make my life exciting and stimulating.Acquiring an innovation makes me happier.The discovery of novelties makes me playful and cheerful.
Cognitive I mostly buy those innovations that satisfy my analytical mind.I find innovations that need a lot of thinking intellectually challenging and therefore I buy them instantly.I often buy new products that make me think logically.I often buy innovative products that challenge the strengths and weaknesses of my intellectual skills.I am an intellectual thinker who buys new products because they set my brain to work.
Comprehension Scale
Please complete the following statement
“I found the product description:
282
Difficult to understand/easy to understand
Confusing/straightforward” (from Phillips, 2000 p20)
“Please indicate the extent to which you agree with the following statements (1=
totally disagree, 7=totally agree):
After reading/listening to the product description, I have a very strong
understanding of how this product works.
After reading/listening to the product description I would be able to use the
product.
After reading/listening to the product description, I understand what the main
features of this product are.
After reading/listening to the product description, I understand what the main
benefits of this product are.”
* “the product description” is a replacement for the phrase “the advert” and “reading/listening to”
is a replacement for the phrase “reading” in the original scale (from Hoeffler, 2003)
Attitude Scale
I evaluate the usage of the product as:
Purchase Intention Scale
Please select the appropriate rating for each statement (1= totally disagree, 4=
Neutral, 7=totally agree)
(1) I will purchase the RNP
(2) Given a choice, my friends will choose the RNP
283
HED/UT Scale1 ---------------------------- 7
Not Fun FunExciting DullNot Delightful DelightfulThrilling Not thrillingUnenjoyable Enjoyable
Effective 1 ---------------------------- 7 IneffectiveUnhelpful HelpfulFunctional Not functionalUnnecessary NecessaryPractical Impractical
(3) There is a strong likelihood that I will buy the RNP
(4) I would like to recommend the RNP to my friends.
284
Appendix 22
Final CFA model
285
Final SEM model
286