phd thesis a.y.m. atiquil islam submission...
TRANSCRIPT
iii
ABSTRACT
In this age of exponential knowledge growth, where Information and Communication
Technology (ICT) has been playing a dominant role, the authorities of Higher Education
concerned have to ensure that the ICT facilities remain within the reach of the lecturers
for their teaching and research purposes. However, despite a decade of existence, the
ICT facilities were found to be underutilized by lecturers. Constructs such as computer
self-efficacy, perceived ease of use, perceived usefulness, use and gratification were
hypothesized to be major barriers. In doing so, the prime objective of this study was to
develop and validate the Technology Adoption and Gratification (TAG) Model to assess
the lecturers’ adoption and gratification in using ICT facilities in higher education. The
second purpose of this study was to evaluate the cross-cultural validation of the causal
structure of the TAG model. The respondents were collected from two public
universities, namely, University of Malaya in Malaysia and Jiaxing University in China.
A total of 396 lecturers were selected using a stratified random sampling procedure. A
questionnaire consisting of items validated from prior studies were put together and
modified to suit the current study. A seven-point Likert scale asking the respondents of
the extent of their agreement/disagreement to the items constituting the construct in the
questionnaire was used. The questionnaire’s validity and reliability were established
through a Rasch model using Winsteps version 3.94. The sample size was judged
adequate for the application of Structural Equation Modeling (SEM) to develop and
validate the Technology Adoption and Gratification (TAG) Model as well as assess all
the research hypotheses. The data was analyzed using AMOS version 18. The findings
of the TAG model discovered that computer self-efficacy had a statistically significant
direct influence on perceived usefulness and perceived ease of use. Subsequently,
lecturers’ perceived ease of use and perceived usefulness had also statistically
significant direct influence on their gratification in using ICT facilities in higher
iv
education. Moreover, perceived usefulness and perceived ease of use revealed
significant direct influence on lecturers’ intention to use ICT facilities for their teaching
and research. Similarly, lecturers’ intention to use had statistically significant direct
effect on their gratification as well as actual use of ICT facilities, respectively. On the
other hand, computer self-efficacy had a significant indirect influence on gratification
mediated by perceived ease of use and perceived usefulness, respectively. Meanwhile,
computer self-efficacy had also significant indirect influence on lecturers’ intention to
use mediated by perceived ease of use and perceived usefulness, respectively.
Eventually, lecturers’ perceived ease of use and perceived usefulness had statistically
significant indirect influence on their gratification in using ICT facilities in higher
education mediated by intention to use. The results of the invariance analysis of the
TAG model demonstrated that the model was valid for measuring lecturers’ adoption
and gratification in using ICT facilities. However, the TAG model works differently in
the cross-cultural settings. The findings contribute to the body of knowledge by
developing and validating the TAG Model as proposed in this study. The results will
also help the Universities’ higher authority in taking and providing necessary steps
towards training to the personnel regarding the application of the ICT facilities in higher
education.
v
ABSTRAK
Dalam zaman pertumbuhan eksponen pengetahuan, di mana Teknologi Komunikasi
Maklumat (ICT) dan telah memainkan peranan yang dominan, pihak berkuasa
Pengajian Tinggi berkenaan perlu memastikan bahawa kemudahan ICT berada di dalam
jangkauan pensyarah untuk tujuan pengajaran dan penyelidikan mereka. Walau
bagaimanapun, walaupun satu dekad kewujudan, kemudahan ICT didapati kurang
digunakan oleh pensyarah. Konstruk seperti komputer sendiri keberkesanan, mudah
dilihat penggunaan, tanggapan kegunaan, penggunaan dan kepuasan hipotesis telah
menjadi halangan utama. Dengan berbuat demikian, objektif utama kajian ini adalah
untuk membangunkan dan mengesahkan Penggunaan Teknologi dan Wang Suapan
(TAG) Model untuk menilai penerimaan pensyarah dan kepuasan dalam menggunakan
kemudahan ICT dalam pendidikan tinggi. Tujuan kedua kajian ini adalah untuk menilai
pengesahan silang budaya struktur yang menyebabkan model TAG. Responden telah
dikumpulkan dari dua universiti awam, iaitu Universiti Malaya di Malaysia dan
Universiti Jiaxing di China; pensyarah yang bekerja di fakulti yang berbeza. Seramai
396 pensyarah telah dipilih menggunakan persampelan berstrata prosedur rawak. Satu
soal selidik yang terdiri daripada item disahkan daripada kajian sebelum telah bersama-
sama dan diubah suai untuk disesuaikan dengan kajian semasa. A tujuh mata skala
Likert meminta responden setakat perjanjian mereka / tidak bersetuju dengan perkara-
perkara yang menjadi binaan di dalam soal selidik yang telah digunakan. Kesahan soal
selidik dan kebolehpercayaan telah ditubuhkan melalui model Rasch menggunakan
Winsteps versi 3.94. Saiz sampel dihakimi mencukupi untuk aplikasi pemodelan
persamaan berstruktur (SEM) untuk membangunkan dan mengesahkan Penggunaan
Teknologi dan Wang Suapan (TAG) Model serta menilai semua hipotesis penyelidikan.
Data dianalisis menggunakan Amos versi 18. Dapatan model TAG mendapati bahawa
komputer self-efficacy mempunyai pengaruh langsung yang signifikan secara statistik
vi
pada manfaat dan tahap kemudahan dilihat penggunaan. Selepas itu, pensyarah dilihat
kemudahan penggunaan dan kegunaan dilihat mempunyai pengaruh langsung juga
statistik yang signifikan pada kepuasan mereka dalam menggunakan kemudahan ICT
dalam pendidikan tinggi. Selain itu, manfaat dan tahap kemudahan dilihat penggunaan
mendedahkan pengaruh langsung yang signifikan terhadap niat pensyarah menggunakan
kemudahan ICT untuk pengajaran dan penyelidikan mereka. Begitu juga dengan niat
pensyarah untuk penggunaan mempunyai kesan yang signifikan secara statistik
langsung pada suapan mereka serta penggunaan sebenar kemudahan ICT, masing-
masing. Sebaliknya, komputer self-efficacy mempunyai pengaruh yang ketara tidak
langsung pada suapan pengantara dengan mudah dilihat penggunaan dan tanggapan
kegunaan masing-masing. Sementara itu, komputer self-efficacy juga mempunyai
pengaruh yang ketara tidak langsung kepada niat pensyarah untuk menggunakan
pengantara dengan mudah dilihat penggunaan dan tanggapan kegunaan masing-masing.
Akhirnya, pensyarah dilihat kemudahan penggunaan dan kegunaan dilihat mempunyai
pengaruh langsung statistik yang signifikan pada kepuasan mereka dalam menggunakan
kemudahan ICT dalam pendidikan tinggi pengantara dengan niat untuk digunakan.
Keputusan analisis invariance yang model TAG menunjukkan bahawa model itu adalah
sah untuk mengukur penggunaan pensyarah dan kepuasan dalam menggunakan
kemudahan ICT. Walau bagaimanapun, model TAG bekerja secara berbeza dalam
tetapan silang budaya. Penemuan menyumbang kepada badan pengetahuan dengan
mengembangkan dan mengesahkan Model TAG seperti yang dicadangkan dalam kajian
ini. Keputusan juga akan membantu pihak berkuasa Universiti 'yang lebih tinggi dalam
mengambil dan menyediakan langkah-langkah yang perlu ke arah latihan kepada
kakitangan mengenai permohonan daripada kemudahan ICT dalam pendidikan tinggi.
vii
DEDICATION
This research is dedicated to my parents, A.Y.M. Nazrul Islam and Noor Aptap Pear
Zahan Chowdhury, as well as my wife, Qian Xiuxiu, for their encouragement, sacrifice,
endeavor and ideas when I was conducting this research. Moreover, this work is also
devoted to human beings as my contribution to the development of human capital which
is of utmost significance in Islam.
viii
ACKNOWLEDGEMENT
The greatest thank to Allah, for his consents, strengths and opportunities that enable me
to complete this research.
I would like to express my sincere appreciation and profound gratitude to my supervisor,
Dr. Chin Hai Leng, for all her guidance, motivation and patience in supervising this
research endeavor. Most of all I am obliged to her as She always helped me with wise
suggestions and corrections. Thus, I have benefited a lot and she has contributed greatly
towards my intellectual aspirations and academic development. I will be forever grateful
to her for taking the time to read, correct and verify my TAG model. My gratitude also
goes to Associate Professor Dr. Diljit Diljit Singh A/L Balwant Singh for his knowledge,
skills, and ideas on the application “modeling”, which I would not have learnt otherwise.
He taught me to be a creative and critical thinker in teaching, learning and research.
I would like to thank the lecturers, and administrative staff for their help and support. A
special vote of thanks goes to authorities of University of Malaya and Jiaxing
University in China, for giving me the chance to collect the appropriate data.
Lastly, I would like to extend my utmost gratitude to my beloved wife Qian Xiuxiu and
parents A.Y.M. Nazrul Islam and Noor Aptap Pear Zahan Chowdhury for their love,
sympathy, inspiration, and support throughout my study.
ix
TABLE OF CONTENTS
Abstract…………………………………………………………………………….. iii
Abstrak…………………………………………………………………………….. v
Dedication………………………………………………………………………...... vii
Acknowledgement…………………………………………………………………. viii
List of Figures..……………………………………………………………………. xi
List of Tables ……………………………………………………………………… xiv
List of Appendices…………………………………………………………………. xvi
CHAPTER 1: INTRODUCTION
1.1 Background of the Study ……………………………………………... 1
1.2 Conceptual Model……… …………………………………………….. 8
1.3 Statement of the Problem………………………………………………. 11
1.4 Objectives of the Study ………………………………………………. 13
1.5 Research Hypotheses ………………………………………………… 13
1.6 Significance of the Study ...…………………………………………… 17
1.7 Limitation of the Study ………………………………………………. 18
1.8 Operational Definition of Terms..……………………………………... 18
CHAPTER 2: LITERATURE REVIEW
2.1 Introduction …………………………………………………………… 21
2.2 Information and Communication Technology Usage ….……………… 21
2.3 Technology Acceptance Model (TAM)….…………………………… 33
2.4 Technology Satisfaction Model (TSM)…..…………………………… 54
2.5 Online Database Adoption and Satisfaction (ODAS) Model…………. 57
2.6 Construction of Hypotheses………………………..…………………. 60
2.6.1 Cross-cultural Studies ………………………………………… 71
CHAPTER 3: METHODOLOGY
3.1 Introduction ………………………………………………………….. 77
3.2 Population and Sample ………………………………………………. 77
3.2.1 Lecturers’ Demographic Information in University of Malaya.. 78
3.2.2 Lecturers’ Demographic Information in Jiaxing University ….. 83
3.3 Sampling Procedure…………………………………………………… 87
3.4 Research Instruments…..……………………………………………… 89
3.4.1 Use of ICT facilities…………………………………………... 90
3.4.2 Perceived Ease of Use……..………………………………….. 91
3.4.3 Perceived Usefulness………………………………………….. 92
3.4.4 Computer Self-Efficacy……………………………………….. 93
3.4.5 Intention to Use………………………………………………... 94
3.4.6 Gratification…………………………………………………… 95
3.4.7 Pilot Test………………………………………………………. 96
3.4.8 Reliability and Validity of Instrument………………………… 104
3.5 Data Analysis Procedures……………………………………………... 113
3.6 Structural Equation Modeling (SEM)…………………………………. 114
x
3.7 Model Evaluation ……………………………………………………. 115
CHAPTER 4: RESULTS
4.1 Introduction …………………………………………………………… 117
4.2 One-Factor Measurement Model of Perceived Ease of Use (PEU)..….. 117
4.3 One-Factor Measurement Model of Perceived Usefulness (PU)..…….. 121
4.4 One-Factor Measurement Model of Computer Self-efficacy (CSE)....... 124
4.5 One-Factor Measurement Model of Intention to Use (INT)…………… 127
4.6 One-Factor Measurement Model of Use (USE)……………………….. 130
4.7 One-Factor Measurement Model of Gratification (GRAT)..………….. 134
4.8 The Hypothesized Technology Adoption and Gratification (TAG)
Model…………………………………………………………………. 137
4.8.1 Hypotheses Testing ………………………………………….. 141
4.9 The Revised Technology Adoption and Gratification (TAG) Model..... 147
4.10 Cross-cultural Validation of TAG Model…………………………… 154
4.10.1 Configural and Metric Invariance Analyses of the TAG
Model…………………………………………………………. 166
CHAPTER 5: DISCUSSION AND CONCLUSION
5.1 Introduction …………………………………………………………... 170
5.2 Discussion of the Findings……………………………………………. 171
5.3 Conclusion ……………………………………….................................. 184
REFERENCES......………………………………………………………………. 193
APPENDICES …………………………………………………………………… 211
xi
LIST OF FIGURES
Figure No. Page No.
1.1 Technology Acceptance Model 9
1.2 Conceptual Model 10
2.1 Technology Acceptance Model 33
2.2 Technology Satisfaction Model (TSM) 54
2.3 Online Database Adoption and Satisfaction (ODAS)
Model 58
2.4 The Hypothesized Technology Adoption and
Gratification (TAG) Model 59
3.1 Sample of Study According to Gender 79
3.2 Sample of Study According to Faculties 79
3.3 Sample of Study According to Educational Background 80
3.4 Sample of Study According to Age Groups 81
3.5 Sample of Study According to Nationality 81
3.6 Sample of Study According to Work Experience 82
3.7 Sample of Study According to Designation 82
3.8 Sample of Study According to Gender 83
3.9 Sample of Study According to Colleges 84
3.10 Sample of Study According to Educational Background 84
3.11 Sample of Study According to Age Groups 85
3.12 Sample of Study According to Nationality 85
3.13 Sample of Study According to Work Experience 86
3.14 Sample of Study According to Designation 86
3.15 Summary of Statistics 98
xii
3.16 Item Polarity Map 99
3.17 Item Map 101
3.18 Item Fit order 102
3.19 Principal Components 103
3.20 Summary of Statistics 104
3.21 Item Polarity Map 105
3.22 Item Map 107
3.23 Item Fit order 108
3.24 Principal Components 113
3.25 The Hypothesized Technology Adoption and
Gratification (TAG) Model 114
4.1 One-Factor Measurement Model for
Perceived Ease of Use 119
4.2 One-Factor Measurement Model for Perceived
Usefulness 122
4.3 One-Factor Measurement Model for Computer
Self-efficacy 125
4.4 One-Factor Measurement Model for Intention to Use 128
4.5 One-Factor Measurement Model for Use 132
4.6 One-Factor Measurement Model for Gratification 135
4.7 The Hypothesized Technology Adoption and
Gratification (TAG) Model 139
4.8 The Revised Technology Adoption and
Gratification (TAG) Model 148
4.9 The Revised TAG Model for Malaysian Participants 155
4.10 The Revised TAG Model for Chinese Participants 161
xiii
4.11 The Constrained TAG Model 167
4.12 The Unconstrained TAG Model 168
xiv
LIST OF TABLES
Table No. Page No.
1.1 The list of Specific Objectives and Hypotheses 15
2.1 The Summary of TAM Literature Matrix 40
3.1 Stratified Random Sampling Procedure 89
3.2 The number of Components Measured 90
3.3 Use items 91
3.4 Perceived Ease of Use Items 92
3.5 Perceived Usefulness Items 93
3.6 Computer Self-efficacy Items 94
3.7` Intention to Use Items 95
3.8 Gratification items 96
3.9 The Valid Items Measuring Perceived Ease of Use (PEU),
Perceived Usefulness (PU), Computer Self-Efficacy (CSE),
Intention to Use (INT), Actual Use (USE) and
Gratification (GRAT) 109
3.10 The Recommended Goodness-of-Fit Indices 116
4.1 The List of Removed Items 118
4.2 The Loadings for Perceived Ease of Use 120
4.3 The Squared Multiple Correlations 120
4.4 The List of Removed Items 121
4.5 The Loadings for Perceived-Usefulness Items 123
4.6 The Squared Multiple Correlations 123
4.7 The List of Removed Items 124
4.8 The Loadings for Computer Self-efficacy 126
xv
4.9 The Squared Multiple Correlations 127
4.10 The List of Removed Items 128
4.11 The Loadings for Intention-to-Use Items 129
4.12 The Squared Multiple Correlations 130
4.13 The List of Removed Items 131
4.14 The Loadings for Use Items 133
4.15 The Squared Multiple Correlations 133
4.16 The List of Removed Items 134
4.17 The Loadings for Gratification Items 136
4.18 The Squared Multiple Correlations 137
4.19 The Summary of the Hypotheses Testing 146
4.20 The List of Removed Items 147
4.21 Valid Items of the Revised TAG Model and Their
Corresponding Loadings, Means, Standard
Deviations and Alpha Values 151
4.22 Valid Items of the Revised TAG Model for Malaysian
Participants and Their Corresponding Loadings,
Means, Standard Deviations and Alpha Values 158
4.23 Valid Items of the Revised TAG Model for Chinese
Participants and Their Corresponding Loadings,
Means, Standard Deviations and Alpha Values 164
4.24 Results of Critical Value of Chi-squared 169
xvi
LIST OF APPENDICES
Appendix No. Page No.
A The Original Model of Computer Self-efficacy (CSE) and
Outputs 211
B The original Model of Perceived Usefulness (PU) and
Outputs 220
C The original Model of Perceived Ease of Use (PEU) and
Outputs 232
D The original Model of Intention to Use (INT) and
Outputs 242
E The original Model of Actual Use (USE) and Outputs 246
F The original Model of Gratification (GRAT) and Outputs 257
G The Hypothesized TAG Model’s Outputs 264
H The Revised TAG Model’s Outputs 282
I The Outputs of Invariance Analyses 300
J The Survey Questionnaire 329
1
CHAPTER 1
INTRODUCTION
1.1 BACKGROUND OF THE STUDY
Information and Communication Technology (ICT) constitutes a key dimension in the
process of the wider development of a country. The estimation of the ICT evolution
involves use of an appropriate metric for assessing the information society in a
country on the sub-indices of access, use and skills (Kyriakidou, Michalakelis &
Sphicopoulos, 2013). Their study of the influential components of the ICT maturity
level revealed considerable discrepancies in their influence in developed and
developing countries. The use factor was found to be considerably more important in
developed countries than in developing countries. Notwithstanding the fact that,
during previous years ICT witnessed an escalating dissemination. The distinctions in
the level of use, access and skills of ICT can be identified both within and between
countries. The policy and decision makers have stated that these dissimilarities cause
an ICT gap and thus strategies aiming at the expansion of ICT have been applied in
many countries. Hence, assessing and analyzing the digital split among countries is of
overriding consequence for researchers as well as managers.
As information and communication technology (ICT) is increasingly used in
education, its integration to teaching, learning and research have had immense
significance in fostering technology-based education among university lecturers,
students and staff. Hong and Songan (2011) contend that in Southeast Asia, ICT is
being utilized more to deal with the challenges that are faced by higher education
systems. In this regard, they have discussed a series of likely applications of ICT. For
instance, what and how students learn, when and where students learn, how new
2
students and lecturers interact, and how to reduce the cost of education. However,
there is very little research on effective utilization of ICT in tertiary education
programs in Southeast Asia. Thus, it is immensely significant for related and advanced
staff of higher education to learn from success stories, experiences and lessons from
the application of ICT in countries within the region. Among those, particular stress
should be laid on the various practical and research-based experiences with respect to
teaching and learning, which include e-learning and blended learning, pedagogic
innovation in using ICT for increased usefulness, accessibility and efficiency of higher
education, applying ICT for online approaches for professional improvement in higher
education, and implementing ICT projects in higher education to enhance access and
quality of learning. Furthermore, they affirmed that the higher education systems need
to keep up with the rapid progress in knowledge and skills that followed the
introduction of new advanced technologies rather than the requirements on relevancy
of employees. Moreover, to be competitive in the gradually challenging global
economy circumstance, it is very essential for university students in this region to be
armed with the appropriate knowledge, skills and aptitudes.
Malaysia is a developing country that is progressing towards development. It
follows the models implemented and practiced in developed countries to maintain
right and steady pace towards development. Malaysia extracts its ideology for the
development process, especially for the education sector, from developed countries
such as Australia, where about 80% of academics use technology in regular ways as
part of the teaching and learning program (Oliver and Towers, 2000). In Malaysia, the
implementation of technology in teaching and learning activity has attracted great
interest from the practitioners in higher education institutions (HEIs), which have
started to adopt and implement information and communication technology (ICT)
3
solutions as a source of a flexible teaching and learning process (Azizan, 2010).
With rapid expansion of information technology, information access is
expected to become central to life in the 21st century (Huang, Liu, Chang, Sung,
Huang, Chen, Shen, Huang, Liao, Hu, Luo, & Chang, 2010). The incorporation of
information and communication technology into classroom practices can integrate a
wide range of activities. From those designed to inspire learners to devour knowledge
(e.g. teachers’ use of presentation software, DVDs or podcasting) to those designed to
increase students’ capabilities to generate their own knowledge (e.g. development of a
reflective blog, collaborative wiki site or e-portfolio). Thus, ICT offers the prospect of
improving both teaching and learning and it is for the educator to decide when and
how it can be applied. Nevertheless, there is extensive agreement in the literature
demonstrating that teachers are not willing to acquire complete benefit of these
opportunities (Groff & Mouza, 2008; Sutherland, Robertson & John, 2009; Levin &
Wadmany, 2008).
In a study, Bate (2010) evaluated how new teachers applied ICT in the first
three years of their teaching and understood the role that ICT played in developing
pedagogical practices of the educators engaged. The findings demonstrated that fresh
teachers expressed pedagogical beliefs that entail their learners in dynamic meaning
making. These teachers were capable of using a basic suite of ICT software.
Nevertheless, pedagogical beliefs that resonate with modern learning theory and
operational ICT capability did not translate into practices that synergised pedagogical
content and technological knowledge. Moreover, educators did not use ICT in ways
that are consistent with their affirmed pedagogical beliefs. Liu (2012) asserted that
pre-service teachers must be prepared with the aptitude to assimilate technology to
facilitate their prospective students’ learning. Furthermore, Liu (2012) concluded that
4
teacher education courses did not facilitate teachers’ use of technology when teaching
students in practicum environments.
According to Blackwell, Lauricella, Wartella, Robb and Schomburg (2013),
the enhanced access to technology with sustained under-utilization in education
necessitates comprehending the obstacles faced by teachers in incorporating
technology into their classrooms. It is remarkable that many models assume that
educators have a moderate R2 value when using technology and anticipate a
reasonable proportion of the variance in how frequently teachers use a certain
technology. Their findings showed that models for the teachers’ use of computers,
iPod touch devices, and tablet computers accounted for around 27–35% of the
variance. The inclusion of attitudes and beliefs constructs as additional dimensions
altered these models from insignificant to significant and enhanced the assumption of
teachers’ actual use of technology. They also suggested that the reliability and
significance of the affordances and obstacles scales discovered in their study makes
their application practical in future quantitative studies of teachers’ use of technology.
Zhou, Zhang and Li (2011) stated that over the last two decades, the
incorporation of ICT into teaching and learning has become an indispensable topic in
education. Studies have revealed that ICT can improve teaching and learning
outcomes. As such, they investigated how well secondary pre-service teachers were
equipped to use technology for teaching in China and evaluated teachers’ experiences
with views of and anticipations about the use of technology. Their findings showed an
overall low level of capability to use technology and few mutual concerns shared by
technology training teachers observed in the teacher education programs. The results
also discovered that universities fall behind in terms of ICT facilities. More than forty
percent of teachers desired the university to update its hardware, and over fifty percent
5
recommended that the university expand its quantity of computers and labs. Over half
of the respondents wished the software were updated and roughly seventy-six percent
demanded a higher-speed campus network. The insufficient access to technology was
also demonstrated by the disappointments voiced by teachers having restricted access
to the computer labs.
According to Soffer, Nachmias and Ram (2010), in the recent decade, the
information technologies have penetrated each field of our daily life, including
education. The incorporation of instructional technology is no longer an extravagance
in higher education but a necessity for all universities. The main contribution of this
study is highlighting the necessity of acquiring better comprehension of the
procedures of the diffusion of web-based learning among lecturers in higher education
and the reasons that influence adoption rate.
Lai (2011) asserted that the use of digital technologies might assist a shift of
cultural practices in teaching and learning, to serve the needs of the 21st century
higher education learners. As a result, higher education institutions require improved
efficiency, additional transparent accountability and better performance in both
teaching and research. However, from the studies that have been undertaken to assess
the overall impact of ICT on teaching and learning in higher education in the last two
decades, one can conclude that higher education institutions have been slow in
acquiring the complete benefit of the prospective advantages offered by the use of ICT.
In this regards, several obstacles have been identified, as to why such a little impact
has been made on the use of ICT in teaching and learning in higher education. The
lack of understanding of why and how technology should be included in pedagogy by
university educators has been recognized as the key cause, and the lack of professional
expansion opportunities is typically quoted as the reason for this lack of understanding.
6
According to Lau and Yuen (2013), social revolution is inevitable with the
emergence of ICT. Further, technology also leads to numerous momentous alterations
in a number of school practices; and in this process, teachers are the main players in
the introduction and continuity of those changes in daily classroom practices. Thus, in
order to assist their use of technology in teaching, educational technology training
becomes indispensable. However, this leads to another paramount problem of
identifying the factors that influence technology adoption and the formulation of
workable solutions to sustainably integrate technology into teaching practices. In fact,
teachers’ perception towards technology is a significant element that affects the
decision on technology adoption despite that training was found to influence
perceptions, especially in terms of teachers’ self-efficacy, attitudes and beliefs.
Furthermore, the actual gap between research and practice hedges the application of
educational research in daily routines which causes criticism. Although a lot of effort
has been put into identifying both contributing and inhibiting factors of technology
use, teachers are still reluctant and opposed to adopting technology in their classrooms.
Nevertheless, teachers are the most important factor in the series of educational
changes. And, teachers’ professional development is the most valuable solution to
stimulate their technological adoption. Moreover, they claimed that effective
integration of technology in education is a complex and multi-faceted issue. They
conducted their study among in-service mathematics teachers and identified a variety
of technology adoption parameters. Meanwhile, their study demonstrated that senior
teachers’ attitudes towards educational benefits of utilizing technology are not
consistent with their perceived efficacy. Thus, it not only reveals that improvement on
educational technology training for in-service teachers is needed, but also there is a
gap between practice and research in educational research which is a serious issue.
7
Despite the benefits that increased technological options can provide, there are
still many barriers to the successful integration and usage of emerging educational
technology such as Internet within educational environments (Roblyer, 2006; Wozney,
Venkatesh & Abrami, 2006). According to Zejno and Islam (2012), determinants such
as perceived ease of use, benefits, and ability to use the ICT facilities were assumed to
be major barriers by postgraduate students in higher education. Islam (2011a) claimed
that one of the ICT facilities, namely, the research database was discovered to be
underutilized, especially by postgraduate students. Factors such as perceived benefits,
computer self-efficacy and satisfaction were hypothesized to be major barriers.
Despite a decade of existence, literatures have demonstrated that the majority of
studies were concerned with the development of integration of information and
communication technology (ICT) in teaching and learning and the acceptance or
adoption of the use of ICT applications from the perspective of preschool teachers and
students. Little research has been conducted on discovering lecturers’ adoption and
gratification in using ICT facilities for research and teaching in higher education.
Moreover, in its endeavor to foster the use of technology in higher education, the
University of Malaya (UM) in Malaysia and Jiaxing University (JU) in China provide
extensive ICT facilities to cater to the needs of the lecturers. However, since its
execution, no research has been conducted to assess the lecturers’ adoption and
gratification of using this emerging educational technology within their environments
for teaching, learning and research purposes. Therefore, there exists a gap for further
investigation.
8
1.2 CONCEPTUAL MODEL
Extant literatures that examined relationships among variables related to technology
acceptance model abound extensively (Kaba & Osei-Bryson, 2013; Kreijns,
Vermeulen, Kirschner, Buuren and Acker, 2013; Moses, Wong, Bakar and Rosnaini,
2013; Islam, 2011a; Zejno & Islam, 2012; Saleh, 2009; Usluel, Askar & Bas, 2008;
Ferdousi, 2009; Shroff, Deneen & Ng, 2011; Shittu, Basha, Rahman & Ahmad, 2011;
Gibson, Harris & Colaric, 2008; Yuen & Ma, 2008; Martínez-Torres, Marín, García,
Vázquez, Oliva & Torres, 2008; Lee, Hsieh & Hsu, 2011; Teo & van Schalk, 2009;
Teo, 2009; Liu, Liao & Pratt, 2009; Agarwal & Prasad, 1998; Dillon & Morris, 1996;
Taylor & Todd, 1995). As such, numerous theoretical models have been propounded
to provide detailed explanations on users’ intention to use technology, and actual
technology use (Venkatesh, Morris, Davis, & Davis, 2003). Following the
introduction of the Technology Acceptance Model (TAM) by Davis (1989), the model
has been widely used to explain computer-usage behavior and factors associated with
acceptance of technology.
According to the model, the technology adoption is influenced by the user’s
intention to use which consequently determined his or her attitudes towards
technology. Perhaps, user’s perceptions— perceived ease of use (PEU) and perceived
usefulness (PU) determine the variability in these attitudinal and behavioural
dimensions (Ahmad, Basha, Marzuki, Hisham and Sahari, 2010). While PEU reveals
the degree to which technology might be free of efforts in its application, PU on the
other hand indicates the extent to which technology usage can help in improving
users’ performance in order to enhance one’s task (Davis, Bagozzi, & Warshaw,
1989). The suggested Technology Acceptance Model (TAM) indicates that,
behavioral intention mediates the influence of the two extrinsic motivation
9
components, namely; PEU and PU as presented in Figure 1.1.
Figure 1.1: Technology Acceptance Model (Davis et al., 1989)
According to Ahmad et al. (2010), TAM has attracted the interest of decision
makers, researchers and practitioners due to the fact that, it is simple, robust and
reasonable. Though, it has unanimously gained empirical support in terms of its
strength through applications and validations in assuming the use of information
systems (Davis, 1993; Taylor & Todd, 1995; Venkatesh & Morris, 2000), current
meta-analyses demonstrates that, if various overriding issues are considered, it could
be resolved through proper understanding (Ma & Liu, 2004; Schepers & Wetzels,
2007; Yousafzai, Foxall and Pallister, 2007).
The Online Database Adoption and Satisfaction Model, proposed by Islam
(2011a) is developed and validated by incorporating two intrinsic motivation attributes,
namely, computer self-efficacy and satisfaction into the original TAM. The findings
of the model affirm a better understanding of the TAM (p. 50). Additional studies on
factors related to technology acceptance and refinement of acceptance models that can
facilitate its generalizability has been suggested (Sun & Zhang, 2006; Thompson,
Compeau, & Higgins, 2006). Considering the evolving of new technologies, perceived
ease of use and perceived usefulness are not the only suitable constructs that
determine technology acceptance (Thompson et al., 2006). Many researchers have
shown that, including more dimensions with other IT acceptance models in order to
10
enhance its specificity and explanatory utility would perform better for a particular
context and focus (Agarwal & Prasad, 1998; Mathieson, 1991). Moreover, there is a
suggestion by previous researchers (Legris, Ingham & Collerette, 2003) that, TAM
deserves to be extended by integrating additional factors that could facilitate the
explanation of more than 40 percent of technology acceptance and usage. Against this
backdrop, the proposed conceptual model was analyzed to develop and validate the
Technology Adoption and Gratification (TAG) model of ICT usage as shown in
Figure 1.2, two other intrinsic motivation attributes – gratification and self-efficacy –
were incorporated into the original TAM based on prior studies (Islam, 2011b &
2011b; Islam, 2014; Wu, Chang & Guo, 2008; Ahmad et al., 2010; Huang, 2008;
Shamdasani, Mukherjee & Malhotra, 2008).
Figure 1.2: Conceptual Model
Note: PEU= “perceived ease of use”, PU= “perceived usefulness”, CSE= “computer
self-efficacy”, INT= “Intention to Use”, USE= “ICT Usage”, GRAT= “Gratification”.
11
1.3 STATEMENT OF THE PROBLEM
Several studies have been and are being conducted on TAM in the attempt to measure
the usage of Information and Communication Technology (ICT) among teachers and
students (Islam, 2011b; Zejno & Islam, 2012; Saleh, 2009; Wozney et al., 2006;
Usluel et al., 2008; Ferdousi, 2009; Kaba & Osei-Bryson, 2013; Kreijns et al., 2013;
Moses et al., 2013). However, despite a decade of existence, the usage of ICT was
found to be underutilized, especially by postgraduate students (Islam, Ahmad,
Madziah & Sahari, 2011; Islam, Leng & Singh, 2015; Islam, 2011a). Besides this,
Goktas, Gedik and Baydas (2013) discovered that the major obstacles that primary
school teachers confronted in the assimilation of ICT as of 2011 emerged to be the
lack of hardware and appropriate software materials, restrictions of hardware, lack of
in-service training and technical support. Kim, Choi, Han and So (2012) asserted that
the appearance and adoption of cutting-edge technologies generate demands for
developing roles and competencies of teachers in the new knowledge society.
However, it is ambiguous how well teachers have been prepared to face the challenges
and difficulties of teaching and learning in the 21st century; and what directions must
be acquired to better practice the new generation of teachers. In reality, the issue of
human use and acceptance of innovative technologies is much more difficult and
demanding (Spector, 2013). Moreover, the effective combination of ICT into
education is still occasionally complex and challenging (Zhao, Paugh, Sheldon, &
Byers, 2002; Keengwe, Onchwari, & Wachira, 2008).
According to Bringula and Basa (2011), even though teachers were skilled in
executing diverse computer and Internet-associated activities, have a high sense of
dedication to the use of the faculty web portal, and have internet access at home. The
faculty web portal which comprised all online research materials, namely, Electronic
12
Journals, Books, Articles and Case Studies were still not used frequently. Buabeng-
Andoh (2012) stated that the worldwide investment in ICT to enhance teaching and
learning in schools have been instigated by many governments. Despite all these
investments on ICT infrastructure, equipments and professional development to
extend education in many countries, ICT adoption and incorporation in teaching and
learning have been limited. If educators do not utilize technology the way it was
planned to serve, the affordances of technology cannot be maximised for effectual
teaching and learning to take place. As such, numerous studies on technology
acceptance have been conducted over the years. Literature revealed that several
acceptance studies had focused on the exploration of factors that affected technology
acceptance among teachers and students (Teo, 2011). However, most studies have
been conducted to assess the students (Stone & Baker-Eveleth, 2013; Alenezi, Karim
& Veloo, 2010; Martínez-Torres et al., 2008; Shroff et al., 2011; Shipps & Phillips,
2013; Terzis, Moridis, Economides & Rebolledo-Mendez, 2013; Chang & Tung, 2008;
Guo & Stevens, 2011; Zejno & Islam, 2012; Islam, 2011a; Rasimah, Ahmad & Zaman,
2011) and teachers (Al-Ruz & Khasawneh, 2011; Tezci, 2011; Yuena, & Ma, 2008;
Wong, Teo & Russo, 2012) acceptance or adoption in using ICT facilities for their
teaching and learning, little is done to investigate on the lecturers’ adoption and
gratification in using ICT facilities in higher education. In so doing, determinants such
as perceived ease of use, perceived usefulness, computer self-efficacy, intention to use,
use and gratification were postulated to be major barriers. To examine the effect of
these factors on lecturers’ ICT usage, the technology adoption and gratification (TAG)
model was developed and validated. Finally, there is a lack of adequate facilities for
the adoption and gratification of the infrastructure considered for this study to
determine its implementation and use as experienced by the lecturers.
13
1.4 OBJECTIVES OF THE STUDY
This study was conducted to explore and understand factors that influenced lecturers’
adoption and gratification of the comprehensive Public Universities’ (University of
Malaya in Malaysia and Jiaxing University in China) ICT usage, the relationships
among perceived ease of use, perceived usefulness, computer self-efficacy, intention
to use, use and gratification and how these constructs interacted to influence adoption
and gratification framed within the TAG Model. Specifically, the objectives were to
(1) develop and validate the Technology Adoption and Gratification (TAG) Model
consisting of the constructs, namely, computer self-efficacy and gratification into the
original TAM, (2) measure the lecturers’ views on the ease of use of ICT facilities, (3)
evaluate the lecturers’ perception regarding the usefulness of ICT facilities, (4)
examine the lecturers’ computer self-efficacy to access ICT facilities, (5) examine the
lecturer’s intention to use of ICT facilities, (6) determine the lecturers’ actual use of
ICT facilities, (7) assess the lecturers’ gratification in using ICT services at the public
universities in Malaysia and China, (8) examine the cross-cultural validation of the
TAG model in higher education.
1.5 RESEARCH HYPOTHESES
Based on existing literature review, the following research hypotheses were generated:
1. Computer self-efficacy (CSE) will have a positive direct influence on
lecturers’ perceived ease of use (PEU) and perceived usefulness (PU) of ICT
facilities in higher education.
2. Perceived ease of use (PEU) will have a positive direct influence on lecturers’
intention to use (INT), actual use (USE) and gratification (GRAT) in using
ICT facilities in higher education.
14
3. Perceived usefulness (PU) will have a positive direct influence on lecturers’
intention to use (INT), actual use (USE) and gratification (GRAT) in using
ICT facilities in higher education.
4. Intention to use (INT) will have a positive direct influence on lecturers’
gratification (GRAT) and actual use (USE) of ICT facilities in higher
education.
5. Actual use (USE) will have a positive direct influence on gratification (GRAT)
in using ICT services in higher education.
6. Computer self-efficacy (CSE) will have a positive indirect influence on
lecturers’ intention to use (INT) of ICT facilities mediated by perceived ease
of use (PEU) and perceived usefulness (PU), respectively.
7. Computer self-efficacy (CSE) will have a positive indirect influence on
lecturers’ gratification (GRAT) in using ICT facilities mediated by perceived
ease of use (PEU) and perceived usefulness (PU), respectively.
8. Computer self-efficacy (CSE) will have a positive indirect influence on
lecturers’ use (USE) of ICT facilities mediated by perceived ease of use (PEU)
and perceived usefulness (PU), respectively.
9. Perceived ease of use (PEU) and Perceived usefulness (PU) will have a
positive indirect influence on lecturers’ gratification (GRAT) in using ICT
facilities mediated by intention to use (INT), respectively.
10. Intention to use (INT) will have a positive indirect influence on lecturers’
gratification (GRAT) mediated by actual use (USE) of ICT facilities in higher
education.
11. There will be a cross-cultural invariant of the causal structure of the TAG
model.
15
The broad objective of this study was to develop and validate the Technology
Adoption and Gratification (TAG) Model by incorporating two intrinsic motivation
attributes, namely, computer self-efficacy and gratification into the original
Technology Acceptance Model (TAM). The broad objective was achieved for this
study through the specific objectives. To address the specific objectives, the present
study generated eleven hypotheses as shown in Table 1.1.
Table 1.1: The list of Specific Objectives and Hypotheses
List of Specific Research Objectives related to Hypotheses
The specific objectives were to
measure the lecturers’ views on the
ease of use of ICT facilities and
evaluate the lecturers’ perception
regarding the usefulness of ICT
facilities.
To address the objectives 2 and 3, the
present study hypothesized that:
H2: Perceived ease of use (PEU) will
have a positive direct influence on
lecturers’ intention to use (INT), actual
use (USE) and gratification (GRAT) in
using ICT facilities in higher
education.
H3: Perceived usefulness (PU) will
have a positive direct influence on
lecturers’ intention to use (INT), actual
use (USE) and gratification (GRAT) in
using ICT facilities in higher
education.
H9: Perceived ease of use (PEU) and
Perceived usefulness (PU) will have a
positive indirect influence on lecturers’
gratification (GRAT) in using ICT
facilities mediated by intention to use
(INT), respectively.
16
Table 1.1, continued
The specific objective was to examine
the lecturers’ computer self-efficacy to
access ICT facilities.
To address the objective 4, the present
study hypothesized that:
H1: Computer self-efficacy (CSE) will
have a positive direct influence on
lecturers’ perceived ease of use (PEU)
and perceived usefulness (PU) of ICT
facilities in higher education.
H6: Computer self-efficacy (CSE) will
have a positive indirect influence on
lecturers’ intention to use (INT) of ICT
facilities mediated by perceived ease of
use (PEU) and perceived usefulness
(PU), respectively.
H7: Computer self-efficacy (CSE) will
have a positive indirect influence on
lecturers’ gratification (GRAT) in
using ICT facilities mediated by
perceived ease of use (PEU) and
perceived usefulness (PU),
respectively.
H8: Computer self-efficacy (CSE) will
have a positive indirect influence on
lecturers’ use (USE) of ICT facilities
mediated by perceived ease of use
(PEU) and perceived usefulness (PU),
respectively.
The specific objective was to examine
the lecturer’s intention to use of ICT
facilities.
To address the objective 5, the present
study hypothesized that:
H4: Intention to use (INT) will have a
positive direct influence on lecturers’
gratification (GRAT) and actual use
(USE) of ICT facilities in higher
education.
H10: Intention to use (INT) will have a
positive indirect influence on lecturers’
gratification (GRAT) mediated by
actual use (USE) of ICT facilities in
higher education.
17
Table 1.1, continued
The specific objective was to
determine the lecturers’ actual use of
ICT facilities.
To address the objective 6, the present
study hypothesized that:
H5: Actual use (USE) will have a
positive direct influence on
gratification (GRAT) in using ICT
services in higher education.
The specific objective was to assess the
lecturers’ gratification in using ICT
services at the public universities in
Malaysia and China.
To address the objective 7, the present
study may not necessarily generate the
hypothesis, which is an endogenous
variable tested using Confirmatory
Factor Analyses as well as Structural
Equation Modeling like other
objectives.
The specific objective was to examine
the cross-cultural validation of the
TAG model in higher education.
To address the objective 8, the present
study hypothesized that:
H8: There will be a cross-cultural
invariant of the causal structure of the
TAG model.
1.6 SIGNIFICANCE OF THE STUDY
This study has provided useful information on the lecturers of the public universities
in Malaysia and China adoption of and gratification in using the universities’ ICT
facilities. This information can give insight to the universities authority and
administration to better gauge the utility of the ICT services provided. This study has
contributed significantly in getting the lecturers’ perspectives on their adoption of and
gratification in using the ICT facilities. This research exercise has comprised lecturers
of all faculties of the University of Malaya in Malaysia and Jiaxing University in
China; it is, therefore, comprehensive and representative in nature. The results will
help the Universities’ higher authority in taking and providing necessary steps
towards training personnel regarding the application of this ICT service in higher
education. Besides, this study will also contribute to the body of the knowledge by
18
developing and validating the technology adoption and gratification (TAG) model in
higher education.
1.7 LIMITATION OF THE STUDY
The study was confined and limited to two public universities, namely, University of
Malaya in Malaysia and Jiaxing University in China. Lecturers who work in the
different faculties/colleges, namely, Arts and Social Sciences, Built Environment,
Business and Accountancy, Computer Science and Information Technology, Dentistry,
Economics and Administration, Education, Engineering, Language and Linguistics,
Law, Medicine, Sciences, Academy of Islamic Studies, Foreign Language Studies,
Mathematics and Engineering, Biology and Chemistry, Material Engineering,
Education, Civil Engineering & Architecture and Jiaxing University Nanhu were
sampled to evaluate their adoption of and gratification in using the ICT facilities for
their teaching and research. However, this study did not include all lecturers due to the
restriction of time.
1.8 OPERATIONAL DEFINITION OF TERMS
The following operational definitions were adopted in the study:
1. Perceived Ease of Use
According to Davis (1989), perceived ease of use refers to the degree to which
the user believes that using the technology would be free of effort. In this
study, perceived ease of use (PEU) refers to the lecturers’ perception of how
easy it is to use the ICT facilities measured through items in the questionnaire.
2. Perceived Usefulness
Perceived usefulness refers to the degree to which a person believes that using
19
the technology would enhance his or her performance (Davis, 1989). In this
study, perceived usefulness (PU) refers to the lecturers’ perception of the
benefits derived from using the ICT services measured through items in the
questionnaire.
3. Computer self-efficacy
In this study, computer self-efficacy (CSE) refers to the lecturers’ beliefs in
their computer ability to use the ICT facilities measured through items in the
questionnaire.
4. Intention to Use
In this study, intention to use (INT) refers to the lecturers’ perception of how
the ICT facilities are examined in terms of the likelihood of lecturers’ use of
such facilities for enhancing their teaching, learning and researching ability as
measured through items in the questionnaire.
5. Gratification
In this study, the extent to which the usage of ICT is consistent with existing
values, needs and the experience of the lecturers’ which will be measured
through items in the questionnaire.
6. Actual Use
In present study, actual use (USE) refers to the lecturers’ view of how
frequently they use ICT facilities for their research and academic-related
activities in higher education as measured through items in the questionnaire.
7. ICT facilities
The ICT facilities have been provided by the authorities of University of
Malaya in Malaysia and Jiaxing University in China, respectively. ICT
20
facilities include computers, web browser (WWW), search engine (e.g. Google,
Yahoo, etc.), Microsoft office applications, document processing, editing and
composing, email and wireless internet, research related software, computer
graphics, computer labs, multimedia facilities (CD-ROM, VCD, DVD), and
university research databases. The research databases comprise volumes of
research articles and reports published in journals. It also includes thesis and
dissertations that would be helpful for lecturers and staff to conduct thorough
literature reviews. In addition, it also helps lecturers to get access to their
subject related research materials which give them extra assistance in their
teaching, learning and research.
8. University Lecturers
This refers to the lecturers of the University of Malaya and Jiaxing University
who are currently working as professors, associate professors, assistant
professors or senior lecturers and lecturers at the different faculties or colleges
which were measured through items in the questionnaire.
21
CHAPTER 2
LITERATURE REVIEW
2.1 INTRODUCTION
This chapter provides an overview of the information and communication technology
usage, Technology Acceptance Model (TAM), and its constructs. It then briefly
defines the Technology Satisfaction Model (TSM) as well as Online Database
Adoption and Satisfaction (ODAS) Model. It is followed by a detailed description of
the conceptual model as proposed in this study, and its constructs to develop and
validate the Technology Adoption and Gratification (TAG). This section is capped
with an overview of the construction of hypotheses.
2.2 INFORMATION AND COMMUNICATION TECHNOLOGY USAGE
In a study by Elen, Clarebout, Sarfo, Louw, Pöysä-Tarhonen, and Stassens (2010), the
definition of information communication technology (ICT) was cited from Wikipedia
(retrieved May 25 2009) and stated that “Information and communication
technologies (ICT) is an umbrella term that includes all technologies for the
manipulation and communication of information. The term is sometimes used in
preference to Information Technology (IT), particularly in two communities:
education and government. In the common usage it is often assumed that ICT is
synonymous with IT; ICT in fact encompasses any medium to record information
(magnetic disk/tape, optical disks (CD/DVD), flash memory etc. and arguably also
paper records); technology for broadcasting information - radio, television; and
technology for communicating through voice and sound or images - microphone,
22
camera, loudspeaker, telephone to cellular phones. It includes the wide variety of
computing hardware (PCs, servers, mainframes, networked storage), the rapidly
developing personal hardware market comprising mobile phones, personal devices,
MP3 players, and much more; the gamut of application software from the smallest
home-developed spreadsheet to the largest enterprise packages and online software
services; and the hardware and software needed to operate networks for transmission
of information, again ranging from a home network to the largest global private
networks operated by major commercial enterprises and, of course, the Internet. Thus,
"ICT" makes more explicit that technologies such as broadcasting and wireless mobile
telecommunications are included.
According to Zhou et al., (2011), like many other countries, the Chinese
government also granted enormous attention and significance to educational
technology. In order to accelerate their post-secondary institutions’ technology
popularizing rate, the China Educational Technology in Higher Education Committee
was established by its Ministry of Education in December 1999. One year later, a
program of Connecting Every School which devoted to connect 90% of their schools
onto the Internet was launched for school level. So, high quality online educational
resources can be obtained by their students and teachers within the next five to ten
years. Further, in 2002, a series of educational technology norms were formulated by
the Ministry for the purpose of guiding the practical technology application for school
teachers, facilitators, as well as administrators, which was followed by the publication
of National Educational Technology Standards put forward firstly by the International
Society for Technology in Education (ISTE) for teachers and students in 2000. The
research they conducted revealed some positive viewpoints as well as problems of
technology and its use among school teachers. In detail, the positive perspectives can
23
be reflected from the following findings: nearly 90% of participators confirmed the
useful role of technology in teaching process; more than 80% agreed teachers need to
learn how to integrate technology into teaching practices, and 77% identified that it
was significant for school students to learn technology. On the contrary, only 11%
stated that they are proficient in using technology and lower than 17% were self-
confident with the use of technology. In addition, compared with the fact that most of
the participators knew technology could be a helpful teaching tool, only 9% felt
skillful to adopt it in actual teaching activities. Thus, strong confidence and ability in
using technology did not assist in direct proportion to the optimistic opinions.
Moreover, the insufficient technological preparation for teaching was frequently
pointed out by focus group participants. They also uncovered that the participants’
education program did not work well in helping them to apply technology in teaching.
However, concerning the use of technology in teaching, they shared similar
viewpoints, and their suggestions on how to integrate technology in teacher education
was not very different to what had been discussed in other countries. Consequently,
the technology course should be re-constructed by the teacher education program to
tackle the participants’ concerns. In order to acquire technology skills more effectively,
it is necessary for teaching candidates to seek better ways of compounding the lecture
and lab components.
Du Plessis and Webb (2012) claimed that there is a growing number of schools
that have been established with ICT infrastructure, and there is an increasing necessity
for the teachers’ professional development in ICT. As a result information on ways to
assist teachers, especially those who are digital immigrants, is needed to help them
deal with the demands of the curriculum and the 21st century skills implied therein.
Nevertheless, they highlighted the caveat that the execution of ICT associated
24
techniques may not be a one-off event, and that the participants require dissimilar
levels of support and training (professional ICT teacher development), generally in a
planned manner, to attain the confidence and competence obligatory to use innovative
ICT strategies effectively. Against this backdrop, they examined whether revelation to
an Internet based extended Cyberhunt strategy allows teachers to achieve a set of
outcomes related to Prensky’s ‘Essential 21st Century Skills’ and the ‘Critical
Outcomes of the South African National Curriculum Statement (NCS)’. The outcomes
comprised effective planning, designing, decision making and goal setting; improved
computer and data searching skills; enhanced confidence, interest, reflective ability,
collaboration, judgment and creative and critical thinking; as well as effective problem
solving and the ability to communicate and interact with individuals and groups. The
quantitative and qualitative data were collected from 26 school teachers, even though
the sample size was not adequate. The findings of their study demonstrated
statistically significant gains in increasing each teacher's decision making, searching
skills, ability to compose questions on dissimilar cognitive levels, planning ability,
notion of audience, computer skills, confidence, reflective ability, interest, and
collaboration skills. The results also show that learning by the design context prepared
the respondents conscious of the inconsistencies that are linked with the traditional
teaching environment and were made aware of the affirmative likelihoods that a
diverse ICT related approach might provide for learning.
Eyal (2012) observed teacher’s skills, abilities, and perceptions required in the
digital environment through assessment, and showed the significance of adapting the
different technologies to the dissimilar estimation purposes. In the study, Eyal (2012)
concluded the significances of teachers in the digital learning environment are: a) the
role of teachers who appreciate the digital learning environments is primary and
25
significant; b) wise use of technological tools to assess learners is essential for the
students, teachers, and for other students participating in educational processes; c)
teachers in the 21st century prefer to use technologies that advance the assessment
methods which emphasize the learning process, enable peer assessment and develop
reflective abilities. Eyal also articulated that teachers are obliged to have some
abilities and skills in basic, advance and intermediate digital assessment literacy,
although there was no empirical assessment.
Afshari, Bakar, Luan and Siraj (2012) affirmed that despite the significance of
computer use in education and the role of the principals of the schools in assisting
technology assimilation, there has been little study on the use of ICT by the principal
themselves, their computer competencies as well as their transformational leadership
role in implementing ICT in schools. As such, they conducted an empirical research to
observe whether the transformational leadership role of principals in implementing
ICT in secondary schools was affected by the computer competence, level of
computer use, and the professional development activities of the principals. The data
for their study were gathered from 320 secondary school principals in Iran. To
develop and validate the research model, a structural equation modeling was applied
to test five hypotheses. Their findings exhibited that computer competence had a
significant effect on secondary school principals’ computer use and it indirectly
influenced the transformational leadership role of principals in implementing ICT in
schools, while it did not show any significant direct effect on transformational
leadership as hypothesized. Thus, It recommended that continuing professional
development opportunities on the subject of leadership and technology must be
provided for principals to enhance their levels of ability in computer use which might
assist future study comprehending the significance of the use of technology in
26
education which may further lead to the discovery of model that include the
transformational leadership dimensions of charisma, inspirational motivation,
intellectual stimulation and individualized contemplation in their schools.
Kregor, Breslin and Fountain (2012) conducted a study with regards to e-
learning practice, demand and capacity which aims to improve planning, decision
making and the quality of the online experience. A total of 2,300 students and 250
staff members participated in the study conducted at the University of Tasmania
(UTAS) in Australia. The findings of their study on technology use presented
information about the overall familiarity on various regular and irregular technology
usages from students and staff. An indication can be perceived from the usage records
and trail on how easily adopted a technology might be if it is known as an innovative
way of teaching and learning, or on the contrary, how much endeavor would be
required to increase usage and acceptance. Most of the staff and students were web
users on a daily or at least regular basis for information searching and general
reference, online transactions like shopping, banking and email. On the other hand, a
large number of students showed satisfaction on the availability of computers and the
wireless network at UTAS. A slightly greater minority (30%) found that applying
technology in courses sometimes challenged their IT skills. Staff were obviously
unsatisfied with their experience because 50% of the staff believed that the workload
associated with e-learning did not involve adequate recognition and planning;
compared to only 18% who did and 42% did not consider learning spaces to be
physically appropriate for “innovative use of learning technology”, while 18% held
the opposite opinion. They also found teaching staff were somewhat less convinced
that the integration of ICT and the online environment in teaching and learning is
beneficial to learners. Therefore, in order to advance more effective and broader use
27
of technology at the university, the support of teaching staff is crucially significant.
Jang and Tsai (2013) studied the technological pedagogical and content
knowledge (TPACK) of science teachers in secondary schools employing a new
contextualized TPACK model. The research model consisted of four factors: content
knowledge (CK), pedagogical content knowledge in context (PCKCx), technology
knowledge (TK) and technological pedagogical content knowledge in context
(TPCKCx). The contextualized TPACK model was tested using a total of 1292
science teachers from Taiwan. Their findings discovered that the four underlying
factors were consistent with prior study. The results also showed that (the percentage
of use of computer technology) computer technology was employed by 94.6% of
secondary school science teachers in Taiwan. However, the technology knowledge
and technological content knowledge in context rated by science teachers with less
teaching experience tended to be considerably higher than that by teachers with more
teaching experience.
According to Kreijns et al., (2013), the introduction of novel pedagogical
practices which complied with 21st knowledge society’s educational demands can be
enabled, supported and reinforced by Information and communication technology
(ICT). Thus, teachers are expected to employ ICT pedagogically. On the contrary,
they feel more often reluctant rather than willing to use ICT, even though the level of
ICT infrastructure increases and this potential skills-based professional development
is delivering. The Technology Acceptance Model (TAM) and its descendants TAM2
and TAM3 are frequently applied to expand insight into the dimensions effecting
acceptance of a specific technology. TAM’s fundamental constructs are perceived
ease of use and perceived usefulness. TAM is primarily a technology-oriented model
which overlooks various factors that might explicate why teachers are not willing to
28
use ICT. In doing so, they alternatively proposed the application of Fishbein’s (2000)
Integrative Model of Behaviour Prediction (IMBP). The IMBP consisted of attitude,
self-efficacy and subjective norm that are the direct and indirect predecessors of
intention to use of ICT and actual ICT usage. Nevertheless, there was no empirical
validation for their proposed research model in the domain of teachers’ intention to
pedagogically use an ICT object. They suggested that if teachers are short of
knowledge and skills, then it would hint that the most significant concerns for
interfering are self-efficacy and its antecedent factors.
According to Fu (2013), Information and Communication Technology (ICT) is
broadly employed in the current educational fields, including computers, the Internet,
and electronic delivery systems like televisions, radios, projectors, and more. As such,
Fu’s study made a summarization of the pertinent researches on the application of
information and communication technology (ICT) in educational fields. Particularly,
the research looked at studies that mentioned the advantages of integrating ICT into
schools, the factors affecting successful ICT integration, the challenges and
difficulties faced in using ICT, the significance of school culture in using ICT as well
as pre-service and in-service teachers’ perceptions, attitudes, and confidence in the use
of ICT. Based on the information gathered, Fu (2013) concluded that it is obvious that
ICT integration is not one-off task, but is adjustable and it is an evolving process.
Moreover, only few studies have focused on challenges of ICT integration faced by
teachers, students and administrative when they participated in specific teaching
activities. Thus, further investigation is required in this area by future researchers.
Buabeng-Andoh (2012) examined teachers’ adoption and integration of
information and communication technology in teaching and learning processes. The
author explained several factors that prevent teachers from using ICT such as: the lack
29
of ICT skills, confidence, pedagogical teacher training, suitable educational software,
access, rigid structure of traditional education systems and restrictive curricula.
However, there was no empirical assessment to generalize this research. The study
concluded that the adoption and integration of technologies by classroom teachers has
been complicated because of its fast progress. Teachers are challenged by the efficient
integration of technology into classroom practices which required more than just
connecting computers to a network. In addition, teachers’ intentions to apply
technology in their teaching practice or attitude toward technology is the key factor in
the study. Providing teachers who have negative attitudes toward technology with
excellent ICT facilities may not have an influence on their use of it in their teaching.
Thus, it is necessary to assure teachers that their teaching can be made interesting,
easier, more fun, more motivating and more enjoyable through technology.
Lim and Pannen (2012) carried out a research to find out how the Indonesian
teacher education institutions (TEIs) applied strategic planning to improve their
capability in developing pre-service teachers’ educational competence in ICT usage.
A holistic method was adopted towards strategic planning by these TEIs through
utilizing the six dimensions of the Capacity Building Toolkit for TEIs in the Asia
Pacific. Among them, the pre-service teacher education program (practicum,
curriculum and assessment) is the key dimension of the strategic planning which is
motivated by the philosophy and vision of the TEI. ICT plan, professional learning,
research and evaluation, and communication and partnerships are four other
dimensions supporting the program. The development of pre-service teacher education
programs were focused on by three of the four TEIs’ strategic planning, while
research and evaluation was focused on by one. During this process, support from
management was essential to the TEIs, but they also faced challenges from senior staff
30
because they were reluctant to change, lack of funding, and less qualified.
On the basis of Gibson, Stringer, Cotton, Simoni, O’Neal, and Howell-
Moroney (2014), helping students to use technology as a learning tool is a significant
aim of incorporating computer usage into daily classroom tutoring. Except access to
computers, teachers’ implementation of computing into learning can also lead to
fruitful classroom integration. The teachers’ belief on classroom computing affects
successful implementation greatly. Therefore, they determined the impacts of a
teacher-focused technology intervention on the students’ computer usage, anxiety and
attitudes towards learning tools, while Teachers’ classroom computer usage, attitudes
and anxiety were investigated as mediators of this relationship. They predicted that
increased teacher participation will have a statistically significant influence on
students’ educational use of computers and attitudes towards classroom computing,
and reduce student anxiety towards computing. On the other hand, the relationship
between the student outcomes and interventions were mediated by teachers’ computer
usage, anxiety and attitudes towards classroom computing. To test the hypotheses, a
total of 696 students and 60 teachers were collected using survey questionnaire. The
findings of their study found that the technology intervention had a significant
influence on the students’ computer usage and attitudes for educational purposes.
Nonetheless, there was no confirmation that teachers’ attitudes or use had any
mediating impact on this relationship. The findings also suggest that it is possible to
enhance learners’ attitudes toward computer use through intense interventions aimed
at their educators.
Kim, Choi, Han and So (2012) stated that since the initiation of the first IT
Master Plan of Korea in 1996, teachers there are commonly deemed well trained to
incorporate ICT into their teaching practice. Therefore, they investigated novel
31
educational measures to improve teachers’ ICT capacities in the contemporary
learning atmosphere. The literature showed that innovative and efficient adaptive
expertise and novel media literacy skills are the new requirements for teachers. Under
this consideration, three cases were observed: (1) learning Scratch for computational
and creative thinking, (2) learning robotics as emerging technology requires
convergent and divergent thinking, and (3) learning to design with ICT systems
thinking. Thus, some new methods helped teachers to fostering adaptive expertise, for
example, offering a variety of contexts unlike usual classroom lessons, and
emphasizing on thinking skills instead of technical skills. However, difficulties were
also encountered by participants in the above three cases, such as problems in
handling different course activities, incorporating new ideas, and comprehending new
design contexts in their comprehensive projects. They suggested that in order to
validate the challenges and potentials of those methods, it is compulsory to look into
teachers’ learning processes and their outcomes with a wider variety of cases with
more depth and with more sources of data.
Bringula and Basa (2011) studied the determinants that may significantly
influence faculty web portal usability. The data for their study was collected from 82
faculty members of the University of the East-Manila in Philippine. Descriptive
analysis demonstrated that the majority of the faculty members were moderately
young, were holders of Master degree, were experts in using the internet and
computers, were dedicated in usage of the web portal, and had internet access at home.
The outcome of their multiple regression analysis discovered the only way to predict
web portal usability was information content, as a web portal design-related factor.
Therefore, the null hypothesis, defining that web portal design-related and faculty-
related dimensions do not have significant impact on faculty web portal usability, is
32
well-received. They recommended that the search for accounts of low usability of
faculty web portals by the internet-connected, dedicated and expert faculty staff was
worth further investigation.
Goktas, Gedik and Baydas (2013) conducted a research to explore the
difficulties faced in the integration of ICT by Turkish primary school teachers, to
make a comparison between the present (in 2011) condition of ICT integration and
that in 2005, and to recommend possible enablers to solve those difficulties. A total of
1373 teachers were collected from 52 schools in 39 provinces in Turkey. The findings
of their study exhibited that the lack of hardware, appropriate software materials,
limitations of hardware, in-service training, and technical support were the most
significant difficulties. Besides this, the maximum ordered enablers were allocation of
more budgets, specific units for peer support, support offices and personnel for
teachers and higher quality pre-service training for ICT. Additional leading enablers
were supporting teachers to enable effective ICT use, having technology plans,
offering higher quality and more quantity of in-service training and designing
appropriate course content/instructional programs. The results also showed that the
majority of barriers indicated statistically significant differences and most enablers
demonstrated moderate or low differences between teachers’ views of their situation
in 2005 and in 2011. Their study recommended that in order to help teachers
incorporate ICT into their teaching practice more efficiently, technical support should
be provided to them. Moreover, online system supports, special ICT units, and
personnel allocation should be focused to facilitate them (besides tutor teachers). And,
to improve the quality of in-service training programs, further approaches should be
taken. Additionally, it should be mandatory for both teachers and administrators to
attend the in-service training programs for classroom ICT application.
33
2.3 TECHNOLOGY ACCEPTANCE MODEL (TAM)
The technology acceptance model (TAM) was rooted in the theory of reasoned action
(TRA), a model concerned with determinants of consciously intended behaviors
(Fishbein & Ajzen, 1975). TRA states that beliefs influence attitudes, which in turn
lead to intentions and then generate behaviors. TAM assumes that beliefs about
usefulness and ease of use are always the primary determinants of Information
Technology/Information System adoption in organizations. According to TAM, these
two determinants serve as the basis for attitudes towards a particular system, which in
turn determines the intention to use, and then generates the actual usage behavior. The
model is shown in Figure 2.1. However, the Technology Acceptance Model (TAM)
developed by Davis (1989) ignored the issue of computer self-efficacy and
satisfaction (Islam, 2014). Moreover, the findings of the Online Database Adoption
and Satisfaction (ODAS) Model (Islam, 2011a) contributed to a better understanding
of the TAM and technology adoption among postgraduate students by incorporating
two intrinsic motivation attributes, namely, computer self-efficacy and satisfaction.
Figure 2.1: Technology Acceptance Model (Davis et al. 1989)
Furthermore, many researchers have extended the TAM by incorporating
additional constructs for measuring e-Learning systems (Lee et al., 2011; Alenezi et
al., 2010; Martínez-Torres et al., 2008; Yuena & Ma, 2008), ICT usage (Tezci, 2011;
Ahmad et al., 2010; Terzis et al., 2013; Zejno & Islam, 2012; Usluel et al., 2008;
34
Wong et al., 2012; Al-Ruz & Khasawneh, 2011), Social Networks (Shipps & Phillips,
2013; Shittu, Basha, Rahman & Ahmad, 2013; Shittu et al., 2011), mobile learning
(Chang, Yan & Tseng, 2012; Wang, Lin & Luarn, 2006; Aljuaid, Alzahrani & Islam,
2014), online learning (Chang & Tung, 2008; Guo & Stevens, 2011; Rasimah et al.,
2011), Internet banking (Amin, 2007), e-Commerce (Devaraj, Fan & Kohli, 2002;
Lallmahamood, 2007; Lee & Park, 2008), wireless internet (Islam, 2011b; Islam,
2014), online shopping (Barkhi, Belanger & Hicks, 2008; Stone & Baker-Eveleth,
2013), mobile banking (Kumar & Ravindran, 2012; Chitungo & Munongo, 2013;
Hanafizadeh, Behboudi, Koshksaray & Tabar, 2014), online research databases (Islam,
2011a; Islam et al., 2015) and YouTube (Lee & Lehto, 2013). The summary of these
studies are shown in Table 2.1. However, most of these studies are concerned with the
acceptance or adoption of technological applications from the perception of school
teachers, students as well as consumers, while, very few studies examined the
lecturers’ adoption and gratification in using ICT facilities for their teaching and
research purposes in higher education.
Teo (2011) extended the technology acceptance model (TAM) by including
two additional constructs, namely, subjective norm and facilitative conditions to
measure school teachers’ intention to use technology. The research model for this
study incorporated six factors, namely, perceived usefulness, perceived ease of use,
subjective norm, facilitative conditions, attitude and behavioural intention to test all
nine hypotheses. The sample used in Teo’s study consisted of 592 teachers from
schools in Singapore. The research model was estimated using the structural equation
modeling. The findings of this study discovered that perceived usefulness, attitude
towards use, and facilitating conditions had direct influences on teachers’ behavioural
intention to use technology, while subjective norm and perceived ease of use had
35
significant indirect effect on behavioural intention to use technology mediated by
attitude towards use and perceived usefulness, respectively. From the direct effects, it
was obvious that when teachers perceive technology to be useful and that applying
technology would enhance their productivity, similarly, their intention to use will be
drastically improved. A positive attitude also showed a positive and direct effect on
teachers’ behavioural intention to use technology, while subjective norm did not exert
any significant influence on behavioural intention. Teo (2011) also suggested that with
the use of technology in education becoming pervasive internationally, proportional
studies across countries or cultures could be conducted to discover the culture-
invariant variables that effect teachers’ intention to use technology.
Lai and Chen (2011) asserted that there has been recently a remarkable
propagation in the number of teaching blogs; nonetheless, little has been discovered
about what encourages teachers to adopt teaching blogs. As a result, they conducted a
study to assess secondary school teachers’ adoption of teaching blogs. To examine
teachers’ adoption, they extended Rogers’ (1995) Innovation Diffusion Theory (IDT)
to develop and validate their research model which integrated four main dimensions,
namely, individual (codification effort, loss of knowledge power, reputation,
enjoyment in helping others, knowledge self-efficacy and personal innovativeness),
technological (perceived usefulness, perceived ease of use, compatibility and
perceived enjoyment), school (school support and school incentives) and
environmental (supervisor influences and peer influence) characteristics to test the
hypotheses. To estimate the research model, survey data were collected from a total
325 secondary school teaching blog adopters and secondary school teachers and
analyzed using discriminant analysis. The findings of their study demonstrated that the
secondary school teachers’ decisions to adopt teaching blogs were strongly related to
36
eight constructs, namely, perceived enjoyment, codification effort, compatibility,
perceived ease of use, personal innovativeness, enjoyment in helping others, school
support and perceived usefulness. It was found that self-efficacy did not significantly
influence teaching blog adoption among others. The results also showed that
perceived ease of use and usefulness were important factors for teachers to adopt
teaching blogs. If teachers observe teaching blogs easy to use and useful, they are
much more likely to use and update their blogs, effecting decisions regarding teaching
blog adoption. However, their study indicated that still very few teachers were used
teaching blogs.
Lee and Lehto (2013) modified the TAM to examine users’ behavioral
intention to use YouTube. Their conceptual model integrated user satisfaction, task-
technology fit, vividness, YouTube self-efficacy, as well as one second-order factor of
content richness into the TAM. A total of 432 participants were collected in their
study. The data for their study were analyzed using structural equation modeling to
validate the extended TAM. The findings of their study showed that users’ perceived
usefulness of YouTube was significantly influenced by task-technology fit, vividness
and content richness, while behavioral intention to use was affected by user
satisfaction and perceived usefulness. Users’ perceived usefulness had a significant
effect on satisfaction whereas YouTube self-efficacy showed a moderate effect on
usefulness. However, perceived ease of use did not have a significant influence on
either perceived usefulness or intention to use of YouTube.
Mac Callum and Jeffrey (2013) investigated how ICT skills influenced
university students' adoption of mobile technology in the tertiary learning
environment. An extended TAM was applied as a theoretical framework of their study.
The hypothesized model included ICT self-efficacy (basic ICT skill, advanced ICT
37
skill & advanced mobile skill) into the TAM. A total of 446 students were collected
using stratified convenience sampling procedure from three tertiary institutions in
New Zealand. The questionnaire was distributed via electronic methods. The
hypothesized model was validated using structural equation modeling to test twelve
hypotheses and their causal relationships. The findings of their study confirmed that
students' behavioral intention to adopt mobile learning was influenced by three factors
which were: perceived ease of use, perceived usefulness and basic ICT skill. However,
perceived ease of use and basic ICT skill had a moderate influence on intention to
adopt. Besides this, basic ICT skill and advanced mobile skill were moderately
significant factors of mobile learning adoption.
In a recent study, Suki, Ramayah, Nee and Suki (2014) studied consumer
intention to use anti-spyware software. Their theoretical framework was developed
based on the innovation diffusion theory and the theory of planned behavior, which
consisted of relative advantage, moral compatibility, ease of use, subjective norms,
image, computing capacity, perceived cost, trialability and intention to use the anti-
spyware software. The theoretical framework was tested using data collected from the
survey questionnaire that engaged 200 students studying in a comprehensive public
university in Malaysia. Data were analyzed using structural equation modeling to
validate eight hypotheses of the model. The findings of their study revealed that
students’ perceived cost, image, moral compatibility and ease of use were significant
determinants of students’ intention to use anti-spyware software, while relative
advantage and subjective norms negatively influenced students’ intention to use anti-
spyware software. However, computing capacity and trialability did not have an
influence on consumer intention to use anti-spyware software in higher education.
38
Niu (2014) examined adolescents’ shopping in cyberspace. One of the
purposes of this study was to identify the influence extent of the perceived use of
technology applied by shopping websites on buyers’ buying behavior. In doing so, the
TAM was used as a moderating variable. She hypothesized that perceived ease of use
and perceived usefulness moderate the relationship between teenage consumers’
decision-making styles and purchasing behavior. The results of her study confirmed
the moderating influences of perceived usefulness and ease of use. However,
perceived ease of use and usefulness moderately influenced adolescents’ purchasing
behavior.
Lee, Hsiao and Purnomo (2014) extended the TAM by including additional
factors of internet self-efficacy, computer self-efficacy, instructor attitude toward
students, learning content and technology accessibility to investigate students’
acceptance of e-learning systems. Their proposed research model was validated using
Structural Equation Modeling to test altogether eleven hypotheses. The findings of
their model confirmed that students’ perceived usefulness and ease of use had
significant effect on intention to use e-learning systems, while perceived usefulness
and ease of use were positively influenced by learning content. On the other hand,
computer self-efficacy had a significant effect on perceived ease of use but it had a
negative insignificant effect on perceived usefulness. Subsequently, perceived ease of
use and usefulness were moderately influenced by internet self-efficacy. However,
instructor attitude towards students showed an insignificant impact on perceived
usefulness. Eventually, perceived ease of use was moderately influenced by
technology accessibility, while it did significantly influence perceived usefulness.
Their suggestions for researchers demonstrated that the individual characteristics of
computer self-efficacy and internet self-efficacy play crucial roles in influencing user
39
beliefs about ease of use.
In a related study, Pynoo, Tondeur, Braak, Duyck, Sijnave and Duyck (2012)
developed and validated their theoretical framework based on a combination of
Theory of Planned Behavior (TPB) and Technology Acceptance Model (TAM) to
examine teachers’ acceptance and use of an educational portal. Their theoretical
framework was consisted of perceived usefulness, perceived ease of use, subjective
norms, perceived behavioral control, attitude, behavioral intention and self-reported
frequency of use. A total of 919 teachers were collected for their study. The research
model was estimated using path analyses. The results of their study indicated that
perceived usefulness, perceived ease of use, subjective norms, perceived behavioral
control, attitude and behavioral intention to use influenced teachers’ acceptance of a
portal. Teachers’ perceived usefulness and attitude were the most significant
predictors of intention to use the portal. The findings also showed that rather than
uploading or sharing information, the portal was used to search for and download
information by the teachers. While, few teachers indicated to browse through the
portal for fun, or without a precise goal.
40
Tab
le 2
.1:
The
Sum
mar
y o
f T
AM
Lit
erat
ure
Mat
rix
Auth
ors
T
itle
s S
ample
C
onst
ruct
s M
easu
red
M
ajor
Fin
din
gs
Dav
is,
Bag
ozz
i an
d
War
shaw
(1989)
Use
r ac
cepta
nce
of
com
pute
r-
tech
nolo
gy:
A
com
par
ison o
f tw
o
theo
reti
cal
model
s.
107 M
BA
studen
ts
Ex
tern
al v
aria
ble
s,
Per
ceiv
ed u
sefu
lnes
s,
Per
ceiv
ed e
ase
of
use
,
Att
itude
tow
ard u
sing,
Beh
avio
ral
inte
nti
on t
o
use
, A
ctual
syst
em u
se
Per
ceiv
ed
use
fuln
ess
stro
ngly
in
fluen
ced
studen
ts’
inte
nti
ons,
ex
pla
inin
g m
ore
than
hal
f of
the
var
iance
in
inte
nti
ons.
P
erce
ived
ea
se
of
use
had
a
smal
l but
signif
ican
t ef
fect
on i
nte
nti
ons
as w
ell,
whil
e th
is e
ffec
t
subsi
ded
over
ti
me.
A
ttit
udes
only
m
od
erat
ely
med
iate
d
the
effe
cts
of
thes
e b
elie
fs
on
in
ten
tions.
Subje
ctiv
e norm
s had
no e
ffec
t on i
nte
nti
ons.
Lee
and P
ark
(2008)
Mobil
e te
chnolo
gy
usa
ge
and B
2B
mar
ket
per
form
ance
under
man
dat
ory
adopti
on
86 u
sefu
l
resp
onse
s
from
alco
holi
c
bev
erag
e
whole
sale
r
Per
ceiv
ed u
sefu
lnes
s,
Per
ceiv
ed e
ase
of
use
,
Per
ceiv
ed l
oss
of
contr
ol,
Use
r sa
tisf
acti
on,
Per
ceiv
ed m
arket
per
form
ance
Com
bin
ing
per
ceiv
ed
loss
of
contr
ol
wit
h
use
r
sati
sfac
tion a
nd t
he
TA
M i
n a
sin
gle
model
can
bet
ter
expla
in
the
Busi
nes
s-to
-Busi
nes
s (B
2B
) tr
ansa
ctio
n
mar
ket
per
form
ance
m
odel
. T
he
findin
gs
sugges
ted
that
per
ceiv
ed l
oss
of
contr
ol
has
a n
egat
ive
effe
ct o
n
use
r sa
tisf
acti
on a
nd p
erce
ived
mar
ket
per
form
ance
is
infl
uen
ced
by
use
r sa
tisf
acti
on
and
per
ceiv
ed
use
fuln
ess.
Lee
and
Leh
to (
2013)
Use
r ac
cepta
nce
of
YouT
ube
for
pro
cedura
l le
arnin
g:
An e
xte
nsi
on
of
the
Tec
hnolo
gy
Acc
epta
nce
Mod
el
432 Y
ouT
ub
use
rs
Per
ceiv
ed e
ase
of
use
,
task
-tec
hnolo
gy f
it,
conte
nt
rich
nes
s,
viv
idnes
s, a
nd Y
ouT
ube
spec
ific
sel
f-ef
fica
cy,
beh
avio
ral
inte
nti
on,
per
ceiv
ed u
sefu
lnes
s,
and u
ser
sati
sfac
tion
Beh
avio
ral
inte
nti
on
was
si
gnif
ican
tly
infl
uen
ced
by
both
per
ceiv
ed u
sefu
lnes
s an
d u
ser
sati
sfac
tion.
Tas
k-
tech
nolo
gy
fit,
co
nte
nt
rich
nes
s,
viv
idnes
s,
and
YouT
ube
self
-eff
icac
y
wer
e dis
cover
ed
to
be
signif
ican
t p
redic
tors
of
per
ceiv
ed
use
fuln
ess.
Nev
erth
eles
s,
per
ceiv
ed
ease
of
use
w
as
not
signif
ican
tly p
redic
tive
of
eith
er p
erce
ived
use
fuln
ess
or
beh
avio
ral
inte
nti
on.
41
Tab
le 2
.1, co
nti
nued
Auth
ors
T
itle
s S
ample
C
onst
ruct
s M
easu
red
M
ajor
Fin
din
gs
Han
afiz
adeh
,
Beh
boudi,
Kosh
ksa
ray
and T
abar
(2014)
Mobil
e-ban
kin
g
adopti
on
by
Iran
ian
ban
k c
lien
ts
361 b
ank
clie
nts
Per
ceiv
ed u
sefu
lnes
s,
per
ceiv
ed e
ase
of
use
,
nee
d f
or
inte
ract
ion,
per
ceiv
ed r
isk, per
ceiv
ed
cost
, co
mpat
ibil
ity
wit
h l
ife
style
, per
ceiv
ed
cred
ibil
ity, tr
ust
and
inte
nti
on t
o u
se m
obil
e
Per
ceiv
ed u
sefu
lnes
s, p
erce
ived
eas
e of
use
, nee
d f
or
inte
ract
ion,
per
ceiv
ed
risk
, per
ceiv
ed
cost
,
com
pat
ibil
ity w
ith l
ife
style
, p
erce
ived
cre
dib
ilit
y a
nd
trust
su
cces
sfull
y
expla
in
the
adopti
on
of
mobil
e
ban
kin
g
among
Iran
ian
cl
ients
. A
dap
tati
on
wit
h
life
style
and t
rust
wer
e re
vea
led t
o b
e th
e m
ost
sig
nif
ican
t
ante
ced
ents
ex
pla
inin
g t
he
adopti
on o
f m
obil
e ban
kin
g.
Gult
ekin
(2011)
Tec
hnolo
gy
Acc
epta
nce
and t
he
Eff
ect
of
Gen
der
in
the
Turk
ish N
atio
nal
Poli
ce:
The
Cas
e of
the
poln
et s
yst
em
407 p
oli
ce
off
icer
s
Per
ceiv
ed u
sefu
lnes
s,
per
ceiv
ed e
ase
of
use
,
subje
ctiv
e norm
and
inte
nti
on t
o u
se
The
findin
gs
dem
onst
rate
d
that
gen
der
has
no
signif
ican
t ef
fect
s on
beh
avio
ral
inte
nti
on
to
use
PO
LN
ET
. B
esid
es,
the
effe
cts
of
per
ceiv
ed u
sefu
lnes
s,
per
ceiv
ed e
ase
of
use
an
d s
ubje
ctiv
e norm
do
not
dif
fer
bet
wee
n m
ale
and f
emal
e poli
ce o
ffic
ers.
Al-
Ruz
and
Khas
awn
eh
(2011)
Jord
ania
n P
re-
Ser
vic
e T
each
ers'
and T
echnolo
gy
Inte
gra
tion:
A
Hum
an R
esourc
e
Dev
elopm
ent
Appro
ach
1,0
08
pre
-ser
vic
e
teac
her
s
Model
ing o
f te
chnolo
gy
in T
each
er e
du
cati
on
cours
es, te
chn
olo
gy s
elf-
effi
cacy, te
chnolo
gy
pro
fici
ency, use
fuln
ess
of
tech
nolo
gy, over
all
support
, te
chnolo
gy
inte
gra
tion a
nd
tech
nolo
gy
avai
labil
ity
The
resu
lts
indic
ated
that
the
model
ing o
f te
chnolo
gy
was
a h
ighly
infl
uen
tial
const
ruct
eff
ecti
ng p
re-s
ervic
e
teac
her
s’
tech
nolo
gy
self
-eff
icac
y,
tech
nolo
gy
pro
fici
ency,
and u
sefu
lnes
s of
tech
nolo
gy.
Tec
hn
olo
gy
self
-eff
icac
y w
as th
e m
ost
im
port
ant
fact
or
wit
h th
e
hig
hes
t dir
ect
effe
ct
on
tech
nolo
gy
inte
gra
tion.
More
over
, th
e su
pport
st
ruct
ure
w
as
the
most
infl
uen
tial
fa
cto
r w
ith
the
hig
hes
t dir
ect
effe
ct
on
tech
nolo
gy i
nte
gra
tion.
42
Tab
le 2
.1, co
nti
nued
Auth
ors
T
itle
s S
ample
C
onst
ruct
s M
easu
red
M
ajor
Fin
din
gs
Sto
ne
and
Bak
er-
Evel
eth
(2013)
Stu
den
ts’
inte
nti
ons
to p
urc
has
e
elec
tronic
tex
tbooks
1,3
82
studen
ts f
rom
a m
id-s
ized
univ
ersi
ty
Eas
e of
use
, at
titu
de
tow
ard e
-tex
tbooks,
beh
avio
ral
inte
nti
ons
to
purc
has
e e-
tex
tbooks,
outc
om
e
expec
tancy/u
sefu
lnes
s,
ver
bal
per
suas
ion/s
oci
al
norm
s, s
elf-
effi
cacy a
nd
pre
vio
us
com
pute
r
exper
ience
The
findin
gs
dem
onst
rate
d t
hat
both
eas
e of
e-te
xtb
ook
use
an
d
ver
bal
per
suas
ion/s
oci
al
norm
posi
tivel
y
infl
uen
ced
beh
avio
ral
inte
nti
ons
on
purc
has
ing
e-
tex
tbooks
thro
ugh
both
se
lf-e
ffic
acy
and
outc
om
e
expec
tancy/u
sefu
lnes
s. P
revio
us
com
pute
r ex
per
ience
posi
tivel
y i
nfl
uen
ces
beh
avio
ral
inte
nti
ons
to p
urc
has
e
an e
-tex
tbook o
nly
thro
ugh s
elf-
effi
cacy
Lee
, H
sieh
and H
su
(2011)
Addin
g I
nnov
atio
n
Dif
fusi
on T
heo
ry t
o
the
Tec
hnolo
gy
Acc
epta
nce
Mod
el:
Support
ing
Em
plo
yee
s'
Inte
nti
ons
to u
se E
-
Lea
rnin
g S
yst
ems
552 b
usi
nes
s
emplo
yee
s
Per
ceiv
ed u
sefu
lnes
s,
per
ceiv
ed e
ase
of
use
,
inte
nti
on t
o u
se,
com
pat
ibil
ity,
com
ple
xit
y, re
lati
ve
advan
tages
, obse
rvab
ilit
y
and t
rial
abil
ity
The
findin
gs
rev
eale
d th
at fi
ve
per
cepti
ons,
nam
ely,
com
pat
ibil
ity,
com
ple
xit
y,
rela
tive
adv
anta
ges
,
obse
rvab
ilit
y
and
tria
labil
ity
of
innovat
ion
char
acte
rist
ics
signif
ican
tly in
fluen
ced em
plo
yee
s’ e-
lear
nin
g s
yst
em b
ehav
iora
l in
tenti
on.
The
effe
cts
of
the
com
pat
ibil
ity,
com
ple
xit
y,
rela
tive
adv
anta
ge,
an
d
tria
labil
ity o
n t
he
per
ceiv
ed u
sefu
lnes
s ar
e si
gnif
ican
t.
More
over
, th
e co
mp
lex
ity,
rela
tive
advan
tage,
tria
labil
ity,
and co
mple
xit
y on th
e per
ceiv
ed ea
se of
use
hav
e a
signif
ican
t in
fluen
ce.
Chit
ungo
and
Munongo
(2013)
Ex
tendin
g t
he
Tec
hnolo
gy
Acc
epta
nce
Mod
el
to M
obil
e B
ankin
g
Adopti
on i
n R
ura
l
Zim
bab
we
275 r
ura
l
consu
mer
s
Per
ceiv
ed e
ase
of
use
,
per
ceiv
ed u
sefu
lnes
s,
rela
tive
adv
anta
ges
,
per
ceiv
ed r
isk, per
son
al
innovat
iven
ess,
soci
al
norm
s an
d p
erce
ived
cost
The
resu
lts
show
ed
that
per
ceiv
ed
use
fuln
ess,
per
ceiv
ed
ease
of
use
, re
lati
ve
advan
tages
, per
sonal
innovat
iven
ess
and
soci
al
norm
s hav
e si
gnif
ican
t
infl
uen
ce on use
rs’
atti
tude
thus
effe
ct th
e in
tenti
on
tow
ard
mobil
e ban
kin
g,
whil
st
per
ceiv
ed
risk
s an
d
cost
s det
erre
d t
he
adopti
on o
f th
e se
rvic
e.
43
Tab
le 2
.1, co
nti
nued
Auth
ors
T
itle
s S
ample
C
onst
ruct
s M
easu
red
M
ajor
Fin
din
gs
Ale
nez
i,
Kar
im a
nd
Vel
oo
(2010)
An E
mpir
ical
Inv
esti
gat
ion i
nto
the
Role
of
Enjo
ym
ent,
Com
pute
r A
nx
iety
,
Com
pute
r S
elf-
Eff
icac
y a
nd I
nte
rnet
Ex
per
ience
in
Infl
uen
cin
g t
he
Stu
den
ts' I
nte
nti
on t
o
Use
E-L
earn
ing:
A
Cas
e S
tud
y f
rom
Sau
di
Ara
bia
n G
ov
ernm
enta
l
Univ
ersi
ties
402
gover
nm
enta
l
univ
ersi
ties
'
studen
ts
Enjo
ym
ent,
com
pute
r an
xie
ty,
per
ceiv
ed u
sefu
lnes
s,
per
ceiv
ed e
ase
of
use
, co
mpute
r se
lf
effi
cacy, in
tern
et
exper
ience
, at
titu
de
tow
ard u
sing E
-
lear
nin
g a
nd
beh
avio
ral
inte
nti
on
to u
se E
-lea
rnin
g
The
findin
gs
show
ed t
hat
com
pute
r an
xie
ty,
com
pute
r
self
-eff
icac
y
and
Enjo
ym
ent
signif
ican
tly
infl
uen
ced
the
studen
ts'
inte
nti
on
to
use
E
-lea
rnin
g,
whil
e th
e
Inte
rnet
ex
per
ience
was
of
insi
gnif
ican
tly i
nfl
uen
ce.
In
addit
ion,
the
signif
ican
ce o
f A
ttit
ude
in m
edia
ting t
he
rela
tionsh
ip bet
wee
n per
ceiv
ed use
fuln
ess,
p
erce
ived
ease
of
use
and t
he
stud
ents
' b
ehav
iora
l in
tenti
on w
as
confi
rmed
.
Dav
is (
1993)
Use
r ac
cepta
nce
of
info
rmat
ion
tech
nolo
gy:
syst
em
char
acte
rist
ics,
use
r
per
cepti
ons
and
beh
avio
ral
impac
ts
122
pro
fess
ional
and m
anag
eria
l
emplo
yee
s
Syst
em d
esig
n
feat
ure
s, p
erce
ived
use
fuln
ess,
per
ceiv
ed
ease
of
use
, at
titu
de
tow
ard u
sing a
nd
actu
al s
yst
em u
se
The
resu
lts
show
ed t
hat
TA
M c
om
ple
tely
med
iate
d t
he
effe
cts
of
syst
em
char
acte
rist
ics
on
usa
ge
beh
avio
r.
Poss
ibly
the
most
str
ikin
g f
indin
g w
as t
hat
per
ceiv
ed
use
fuln
ess
was
50
% m
ore
sig
nif
ican
t th
an e
ase
of
use
in d
eter
min
ing u
sage.
Tez
ci (
2011)
Fac
tors
th
at i
nfl
uen
ce
pre
-ser
vic
e te
ach
ers’
ICT
usa
ge
in
educa
tion
1898
pre
-ser
vic
e
teac
her
s
Com
pute
r at
titu
de,
inte
rnet
att
itude,
sel
f-
confi
den
ce, sc
hool
clim
ate/
support
, IC
T
usa
ge,
IC
T
know
ledge
The
findin
gs
show
ed
that
m
ost
pre
-ser
vic
e te
acher
s
report
ed t
hat
they u
se o
nly
bas
ic I
CT
appli
cati
ons
for
educa
tional
purp
ose
s.
Inte
rnal
an
d
exte
rnal
fa
ctors
wer
e fo
und t
o b
e re
late
d t
o e
ach o
ther
and t
o I
CT
usa
ge
level
s.
More
over
, m
ale
pre
-ser
vic
e te
acher
s’
know
ledge
and
usa
ge
level
s of
ICT
wer
e hig
her
than
fem
ale
teac
her
s.
44
Tab
le 2
.1, co
nti
nued
Auth
ors
T
itle
s S
ample
C
onst
ruct
s M
easu
red
M
ajor
Fin
din
gs
Shin
(2012)
3D
TV
as
a so
cial
pla
tform
for
com
munic
atio
n a
nd
inte
ract
ion
229 u
sers
of
3D
TV
Soci
al p
rese
nce
,
per
ceiv
ed e
njo
ym
ent,
per
ceiv
ed q
ual
ity,
flow
, per
ceiv
ed
use
fuln
ess,
att
itude,
inte
nti
on, usa
ge
The
resu
lts
indic
ated
that
the
signif
ican
t ro
les
for
soci
al
pre
sence
and f
low
, both
of
whic
h i
nfl
uen
ced a
ttit
ude
as
wel
l as
per
ceiv
ed u
sefu
lnes
s an
d p
erce
ived
enjo
ym
ent.
This
se
t of
fact
ors
is
k
ey
to
use
rs’
expec
tati
ons
of
3D
TV
.
Mar
tínez
-
Torr
es,
Mar
ín,
Gar
cía,
Váz
quez
,
Oli
va
and
Torr
es
(2008)
A t
echnolo
gic
al
acce
pta
nce
of
e-
lear
nin
g t
ools
use
d i
n
pra
ctic
al a
nd
labora
tory
tea
chin
g,
acco
rdin
g t
o t
he
Euro
pea
n h
igher
educa
tion a
rea
220 s
tuden
ts
Use
, In
tenti
on o
f use
,
Per
ceiv
ed u
sefu
lnes
s,
Eas
e of
use
,
Met
hodolo
gy,
Acc
essi
bil
ity,
Rel
iabil
ity,
Enjo
ym
ent,
Use
r
adap
tati
on,
Com
munic
ativ
enes
s,
Fee
db
ack,
Form
at,
Inte
ract
ivit
y a
nd
contr
ol,
Dif
fusi
on
and U
ser
tools
The
findin
gs
stro
ngly
su
pport
th
e ex
tended
T
AM
in
pre
dic
ting a
stu
den
t’s
inte
nti
on t
o u
se e
-lea
rnin
g a
nd
def
ine
a se
t of
exte
rnal
var
iable
s w
ith
a si
gnif
ican
t
infl
uen
ce
in
the
ori
gin
al
TA
M
var
iable
s.
Ho
wev
er,
per
ceiv
ed e
ase
of
use
did
not
posi
t a
signif
ican
t im
pac
t
on s
tuden
t at
titu
de
or
inte
nti
on t
ow
ards
e-le
arnin
g t
ool
usa
ge.
Ahm
ad,
Bas
ha,
Mar
zuki,
His
ham
and
Sah
ari
(2010)
Fac
ult
y’s
acc
epta
nce
of
com
pute
r-b
ased
tech
nolo
gy:
cross
-
val
idat
ion o
f an
exte
nded
model
731 f
acult
y
mem
ber
s
Sel
f-ef
fica
cy,
per
ceiv
ed u
sefu
lnes
s,
inte
nti
on, use
The
resu
lts
support
ed th
e ad
equac
y o
f th
e ex
tended
tech
nolo
gy a
ccep
tance
model
(T
AM
E).
Alt
hough t
he
TA
ME
’s c
ausa
l st
ruct
ure
was
val
id t
o b
oth
mal
e an
d
fem
ale
staf
f, ag
e gro
up
s sh
ow
ed m
oder
ate
stru
ctura
l
rela
tionsh
ips
among t
he
const
ruct
s of
inte
rest
.
45
Tab
le 2
.1, co
nti
nued
Auth
ors
T
itle
s S
ample
C
onst
ruct
s M
easu
red
M
ajor
Fin
din
gs
Shro
ff,
Den
een a
nd
Ng (
2011)
Anal
ysi
s of
the
tech
nolo
gy a
ccep
tance
model
in e
xam
inin
g
studen
ts’
beh
avio
ura
l
inte
nti
on t
o u
se a
n e
-
port
foli
o s
yst
em
72 s
tuden
ts
Per
ceiv
ed u
sefu
lnes
s,
per
ceiv
ed e
ase
of
use
, at
titu
des
tow
ards
usa
ge
and
beh
avio
ura
l in
tenti
on
to u
se
The
resu
lts
confi
rmed
that
stu
den
ts’
per
ceiv
ed e
ase
of
use
had
a
signif
ican
t in
fluen
ce
on
atti
tude
tow
ards
usa
ge.
C
onse
qu
entl
y,
per
ceiv
ed ea
se o
f use
had
th
e
stro
nges
t si
gnif
ican
t in
flu
ence
on p
erce
ived
use
fuln
ess.
Ship
ps
and
Phil
lips
(2013)
Soci
al N
etw
ork
s,
Inte
ract
ivit
y a
nd
Sat
isfa
ctio
n:
Ass
essi
ng
Soci
o-T
echnic
al
Beh
avio
ral
Fac
tors
as
an E
xte
nsi
on t
o
Tec
hnolo
gy
Acc
epta
nce
164 s
tuden
ts
Per
ceiv
ed u
sefu
lnes
s,
per
ceiv
ed e
ase
of
use
, at
titu
des
,
sati
sfac
tion,
conce
ntr
atio
n f
ocu
s,
per
sonal
involv
emen
t,
per
ceiv
ed
inte
ract
ivit
y
com
munic
atio
n a
nd
per
ceiv
ed
inte
ract
ivit
y c
ontr
ol
The
resu
lts
show
ed
that
per
ceiv
ed
inte
ract
ivit
y
and
level
of
focu
s/co
nce
ntr
atio
n do af
fect
an
en
d
use
r’s
sati
sfac
tion
wit
h
a so
cial
net
work
al
on
g
wit
h
ante
ced
ents
fr
om
th
e te
chnolo
gy
acce
pta
nce
m
odel
.
Thes
e fi
ndin
gs
sugges
ted
that
both
T
AM
re
late
d
fact
ors
and m
arket
ing r
elat
ed f
acto
rs b
oth
im
pac
t th
e
use
r ex
per
ience
on a
soci
al n
etw
ork
ing s
ite.
Chan
g, Y
an
and T
sen
g
(2012)
Per
ceiv
ed c
onven
ien
ce
in a
n e
xte
nded
tech
nolo
gy a
ccep
tance
model
: M
obil
e
tech
nolo
gy a
nd
Engli
sh l
earn
ing f
or
coll
ege
studen
ts
158
coll
ege
studen
ts
Per
ceiv
ed
conven
ience
,
per
ceiv
ed e
ase
of
use
, per
ceiv
ed
use
fuln
ess,
att
itude
and c
onti
nuan
ce
inte
nti
on t
o u
se
The
findin
gs
show
ed
that
per
ceiv
ed
conven
ience
,
per
ceiv
ed e
ase
of
use
and p
erce
ived
use
fuln
ess
wer
e
ante
ced
ent
fact
ors
th
at
affe
cted
th
e ac
cepta
nce
o
f
Engli
sh
mobil
e le
arnin
g.
Per
ceiv
ed
conven
ience
,
per
ceiv
ed e
ase
of
use
and p
erce
ived
use
fuln
ess
had
a
signif
ican
t posi
tive
effe
ct o
n t
hei
r at
titu
de.
Per
ceiv
ed
use
fuln
ess
and
att
itude
tow
ards
use
had
a s
ignif
ican
t
posi
tive
effe
ct o
n c
onti
nuan
ce o
f in
tenti
on t
o u
se.
46
Tab
le 2
.1, co
nti
nued
Auth
ors
T
itle
s S
ample
C
onst
ruct
s M
easu
red
M
ajor
Fin
din
gs
Ter
zis,
Mori
dis
,
Eco
nom
ides
and
Reb
oll
edo-
Men
dez
(2013)
Com
pute
r B
ased
Ass
essm
ent
Acc
epta
nce
: A
Cro
ss-
cult
ura
l S
tud
y i
n
Gre
ece
and M
exic
o
168 s
tuden
ts
from
Gre
ece
and M
exic
o
Per
ceiv
ed
pla
yfu
lnes
s,
Per
ceiv
ed u
sefu
lnes
s,
Per
ceiv
ed e
ase
of
use
, C
om
pute
r se
lf
effi
cacy, G
oal
expec
tancy, C
onte
nt,
faci
lita
ting
condit
ions,
beh
avio
ral
inte
nti
on
and s
oci
al i
nfl
uen
ce
The
resu
lts
indic
ated
th
at
the
Com
pute
r B
ased
Ass
essm
ent
Acc
epta
nce
M
odel
(C
BA
AM
) w
as
val
id
for
both
countr
ies.
How
ever
, th
ere
are
som
e cu
ltura
l
dif
fere
nce
s.
Gre
ek
stud
ents
’ beh
avio
ral
inte
nti
on
is
trig
ger
ed
mai
nly
b
y
Per
ceiv
ed
Pla
yfu
lnes
s an
d
Per
ceiv
ed
Eas
e o
f U
se,
whil
e M
exic
an
stu
den
ts’
beh
avio
ral
inte
nti
on i
s ca
use
d b
y P
erce
ived
Pla
yfu
lnes
s
and P
erce
ived
Use
fuln
ess.
Chan
g a
nd
Tung (
2008
)
An e
mpir
ical
inves
tigat
ion o
f
studen
ts’
beh
avio
ura
l
inte
nti
ons
to u
se t
he
onli
ne
lear
nin
g c
ours
e
web
site
s
212
under
gra
du
ate
studen
ts
Com
pute
r se
lf-
effi
cacy, P
erce
ived
syst
em q
ual
ity,
Per
ceiv
ed e
ase
of
use
, P
erce
ived
use
fuln
ess,
Com
pat
ibil
ity a
nd
inte
nti
on t
o u
se
The
findin
gs
show
ed
that
co
mpat
ibil
ity,
per
ceiv
ed
use
fuln
ess,
per
ceiv
ed
ease
of
use
, p
erce
ived
sy
stem
qual
ity
and
com
pute
r se
lf-e
ffic
acy
wer
e im
port
ant
fact
ors
for
studen
ts’
beh
avio
ura
l in
tenti
ons
to u
se t
he
onli
ne
lear
nin
g c
ours
e w
ebsi
tes.
Yuen
a an
d
Ma
(2008)
Ex
plo
ring t
each
er
acce
pta
nce
of
e-
lear
nin
g t
echnolo
gy
152 i
n-s
ervic
e
teac
her
s
Subje
ctiv
e norm
,
com
pute
r se
lf-
effi
cacy, per
ceiv
ed
use
fuln
ess,
per
ceiv
ed
ease
of
use
, in
tenti
on
The
resu
lts
found t
hat
subje
ctiv
e norm
and c
om
pute
r
self
-eff
icac
y
serv
e as
th
e tw
o
signif
ican
t per
cepti
on
anch
ors
of
the
fundam
enta
l co
nst
ruct
s in
T
AM
.
How
ever
, co
ntr
ary
to
pre
vio
us
lite
ratu
re,
per
ceiv
ed
ease
o
f use
bec
ame
the
sole
d
eter
min
ant
for
the
pre
dic
tion
of
inte
nti
on
to
use
, w
hil
e per
ceiv
ed
use
fuln
ess
was
not
signif
ican
t in
th
e pre
dic
tion
of
inte
nti
on t
o u
se.
47
Tab
le 2
.1, co
nti
nued
Auth
ors
T
itle
s S
ample
C
onst
ruct
s M
easu
red
M
ajor
Fin
din
gs
Guo a
nd
Ste
ven
s
(2011)
Fac
tors
in
fluen
cin
g
per
ceiv
ed u
sefu
lnes
s
of
wik
is f
or
gro
up
coll
abora
tive
lear
nin
g
by f
irst
yea
r st
uden
ts
205 s
tuden
ts
Att
itude
tow
ards
wik
is, W
iki
use
fuln
ess,
Wik
i use
,
studen
t’s
pri
or
exper
tise
, ea
se o
f
acce
ss,
The
findin
gs
dem
onst
rate
d t
hat
wik
i use
infl
uen
ced b
y
the
studen
t’s
pri
or
exper
tise
. T
hei
r per
ceiv
ed
use
fuln
ess
of
wik
is w
ere
stro
ngly
infl
uen
ced
by t
hei
r
teac
her
s’ a
ttit
udes
tow
ards
the
tech
nolo
gy a
s w
ell
as
the
ease
of
acce
ss t
o t
he
wik
is.
Att
itude
tow
ards
wik
is
was
lar
gel
y i
nfl
uen
ced b
y t
he
exte
nt
to w
hic
h s
tuden
ts
saw
wik
is a
s a
sourc
e of
hel
p w
ith t
hei
r as
signm
ent
work
, an
d t
hei
r in
tenti
on t
o u
se w
ikis
in t
he
futu
re w
as
dri
ven
by t
hei
r per
cepti
on o
f w
iki’
s use
fuln
ess.
Am
in (
2007)
In
tern
et B
ankin
g
Adopti
on a
mong
Young I
nte
llec
tual
s
250 s
tuden
ts
Per
ceiv
ed u
sefu
lnes
s,
per
ceiv
ed e
ase
of
use
, per
ceiv
ed
cred
ibil
ity a
nd
com
pute
r se
lf-
effi
cacy a
nd
beh
avio
ral
inte
nti
on
The
resu
lts
show
ed t
hat
per
ceiv
ed u
sefu
lnes
s, e
ase
of
use
an
d
per
ceiv
ed
cred
ibil
ity
had
a
signif
ican
t
rela
tionsh
ip
wit
h
beh
avio
ral
inte
nti
on.
More
over
,
per
ceiv
ed u
sefu
lnes
s an
d e
ase
of
use
had
a s
ignif
ican
t
rela
tionsh
ip
wit
h
com
pute
r se
lf-e
ffic
acy.
How
ever
,
com
pute
r se
lf-e
ffic
acy d
id n
ot
asso
ciat
e w
ith p
erce
ived
cred
ibil
ity.
Ev
entu
ally
, per
ceiv
ed
ease
of
use
had
a
signif
ican
t re
lati
onsh
ip w
ith p
erce
ived
use
fuln
ess
and
cred
ibil
ity w
hic
h s
how
s th
at t
hes
e sc
ales
are
rel
ated
to
per
ceiv
ed
ease
of
use
in
ex
pla
inin
g
under
gra
duat
e
pre
fere
nce
.
Zej
no a
nd
Isla
m (
20
12)
Dev
elopm
ent
and
val
idat
ion o
f li
bra
ry
ICT
usa
ge
scal
e fo
r th
e
IIU
M p
ost
gra
du
ate
studen
ts
285
Post
gra
duat
e
studen
ts
Per
ceiv
ed u
sefu
lnes
s,
per
ceiv
ed e
ase
of
use
, co
mpute
r se
lf-
effi
cacy a
nd u
se
The
findin
gs
of
the
stud
y s
how
ed t
hat
per
ceiv
ed e
ase
of
use
, per
ceiv
ed u
sefu
lnes
s an
d c
om
pute
r se
lf-e
ffic
acy
are
signif
ican
t p
redic
tors
of
post
gra
duat
e st
uden
ts’
ICT
usa
ge.
48
Tab
le 2
.1, co
nti
nued
Auth
ors
T
itle
s S
ample
C
onst
ruct
s M
easu
red
M
ajor
Fin
din
gs
Usl
uel
,
Aşk
ar a
nd
Baş
(2008
)
A S
truct
ura
l E
quat
ion
Model
for
ICT
Usa
ge
in H
igher
Educa
tion
814 f
acult
y
mem
ber
s
ICT
fac
ilit
ies
(cla
ssro
om
, la
b a
nd
off
ice)
, p
erce
ived
attr
ibute
s (r
elat
ive
advan
tages
,
com
pat
ibil
ity,
ease
of
use
, obse
rvab
ilit
y,
trai
abil
ity),
IC
T
usa
ge
(inst
ruct
ional
and m
anag
eria
l
purp
ose
s)
The
resu
lts
rev
eale
d th
at th
e p
erce
ived
at
trib
ute
s o
f
ICT
and I
CT
fac
ilit
ies
in t
he
univ
ersi
ties
pre
dic
t th
e
ICT
use
. T
he
facu
lty m
ember
s use
IC
T m
ost
ly as
a
mea
ns
of
sear
chin
g f
or
info
rmat
ion a
bout
the
cours
e
thro
ugh t
he
Inte
rnet
, an
d a
s a
mea
ns
of
com
munic
atio
n
such
as
: publi
shin
g
thei
r le
cture
note
s,
annou
nci
ng
cours
e as
signm
ents
, an
d p
roje
cts-
on W
WW
.
Dev
araj
, F
an
and K
ohli
(2002)
Ante
ced
ents
of
B2C
Chan
nel
Sat
isfa
ctio
n
and P
refe
rence
:
Val
idat
ing e
-
Com
mer
ce M
etri
cs
134 c
onsu
mer
s
Ass
et s
pec
ific
ity,
unce
rtai
nty
,
use
fuln
ess,
eas
e of
use
, ti
me,
pri
ce
savin
gs,
em
pat
hy,
reli
abil
ity,
resp
onsi
ven
ess,
assu
rance
,
sati
sfac
tion w
ith
elec
tronic
com
mer
ce
chan
nel
, ch
annel
pre
fere
nce
The
findin
gs
show
ed th
at per
ceiv
ed ea
se of
use
an
d
use
fuln
ess
are
import
ant
in f
orm
ing c
onsu
mer
att
itudes
and
sati
sfac
tion
wit
h
the
elec
tronic
co
mm
erce
(E
C)
chan
nel
. E
ase
of
use
was
als
o f
ound
to b
e a
signif
ican
t
det
erm
inan
t of
sati
sfac
tion.
The
stud
y a
lso d
isco
ver
ed
empir
ical
su
pport
fo
r th
e as
sura
nce
dim
ensi
on
of
SE
RV
QU
AL
as
a
det
erm
inan
t in
E
C
chan
nel
sati
sfac
tion.
More
over
, th
e re
sult
s al
so
indic
ated
gen
eral
su
pport
fo
r co
nsu
mer
sa
tisf
acti
on
as
a
det
erm
inan
t of
chan
nel
pre
fere
nce
.
49
Tab
le 2
.1, co
nti
nued
Auth
ors
T
itle
s S
ample
C
onst
ruct
s M
easu
red
M
ajor
Fin
din
gs
Wan
g,
Lin
and L
uar
n
(2006)
Pre
dic
ting c
onsu
mer
inte
nti
on t
o u
se m
obil
e
serv
ice
258 c
onsu
mer
s
Sel
f-ef
fica
cy,
Per
ceiv
ed f
inan
cial
reso
urc
e, p
erce
ived
use
fuln
ess,
eas
e of
use
, per
ceiv
ed
cred
ibil
ity a
nd
inte
nti
on
The
resu
lts
exhib
ited
th
at
per
ceiv
ed
use
fuln
ess,
per
ceiv
ed
ease
of
use
, per
ceiv
ed
cred
ibil
ity,
self
-
effi
cacy
and
per
ceiv
ed
finan
cial
re
sourc
es
all
had
a
signif
ican
t in
fluen
ce
on
beh
avio
ura
l in
tenti
on.
Furt
her
mo
re,
self
-eff
icac
y
was
fo
und
to
hav
e a
signif
ican
t ef
fect
on p
erce
ived
eas
e of
use
, w
hic
h,
in
turn
, had
a p
osi
tive
effe
ct o
n b
oth
per
ceiv
ed u
sefu
lnes
s
and
per
ceiv
ed
cred
ibil
ity.
Fin
ally
, both
per
ceiv
ed
finan
cial
re
sourc
e an
d
per
ceiv
ed
cred
ibil
ity
had
a
signif
ican
t ef
fect
on p
erce
ived
use
fuln
ess.
Isla
m
(2011b)
Via
bil
ity o
f th
e
exte
nded
tec
hnolo
gy
acce
pta
nce
model
: an
empir
ical
stu
dy
285 s
tuden
ts
Per
ceiv
ed e
ase
of
use
, per
ceiv
ed
use
fuln
ess,
com
pute
r
self
-eff
icac
y a
nd
Sat
isfa
ctio
n
The
resu
lts
indic
ated
th
at s
tuden
ts’
per
ceiv
ed e
ase
of
use
had
a
stat
isti
call
y
signif
ican
t ef
fect
on
th
eir
sati
sfac
tion.
The
stud
y a
lso d
isco
ver
ed t
hat
gen
der
did
not
exer
t an
y
infl
uen
ce
as
a m
oder
atin
g
var
iable
tow
ards
studen
ts’
sati
sfac
tion
in
usi
ng
Wir
eles
s
Inte
rnet
.
Bar
khi,
Bel
anger
and
Hic
ks
(2008)
A m
odel
of
the
det
erm
inan
ts o
f
purc
has
ing F
rom
vir
tual
sto
res
277 s
tuden
ts
Act
ual
purc
has
e,
Per
ceiv
ed s
ecuri
ty,
Att
itude
tow
ard
onli
ne
shoppin
g,
Per
ceiv
ed u
sefu
lnes
s,
Per
ceiv
ed p
eer
infl
uen
ce a
nd
per
ceiv
ed b
ehav
iora
l
contr
ol
The
findin
gs
show
ed
that
per
ceiv
ed
use
fuln
ess,
per
ceiv
ed
beh
avio
ral
contr
ol,
an
d
per
ceiv
ed
pee
r
infl
uen
ce at
titu
de
tow
ards
purc
has
ing fr
om
a
vir
tual
store
. O
n
the
oth
er
han
d,
atti
tude
infl
uen
ced
actu
al
purc
has
ing f
rom
a v
irtu
al s
tore
.
50
Tab
le 2
.1, co
nti
nued
Auth
ors
T
itle
s S
ample
C
onst
ruct
s M
easu
red
M
ajor
Fin
din
gs
Lal
lmah
amo
od (
2007)
An e
xam
inat
ion o
f
indiv
idual
’s p
erce
ived
secu
rity
and p
rivac
y o
f
the
inte
rnet
in
Mal
aysi
a an
d t
he
infl
uen
ce o
f th
is o
n
thei
r in
tenti
on t
o u
se e
-
com
mer
ce:
Usi
ng a
n
exte
nsi
on o
f th
e
tech
nolo
gy a
ccep
tance
model
187 s
tuden
ts
Per
ceiv
ed u
sefu
lnes
s,
per
ceiv
ed e
ase
of
use
, per
ceiv
ed
secu
rity
& p
rivac
y
and i
nte
nti
on t
o u
se
inte
rnet
ban
kin
g
The
resu
lts
dem
onst
rate
d t
hat
per
ceiv
ed s
ecu
rity
an
d
pri
vac
y
had
a
signif
ican
t im
pac
t on
per
ceiv
ed
use
fuln
ess,
per
ceiv
ed e
ase
of
use
and i
nte
nti
on t
o u
se
inte
rnet
ban
kin
g,
whil
e per
ceiv
ed
ease
o
f use
infl
uen
ces
on
per
ceiv
ed
use
fuln
ess
and
inte
nti
on
to
use
. M
ore
ov
er,
per
ceiv
ed
use
fuln
ess
show
ed
a
signif
ican
t im
pac
t on
inte
nti
on
to
use
, w
hic
h is
th
e
most
im
port
ant
pre
dic
tor
among o
ther
s.
Ras
imah
,
Ahm
ad a
nd
Zam
an
(2011)
Eval
uat
ion o
f use
r
acce
pta
nce
of
mix
ed
real
ity t
echnolo
gy
63 s
tuden
ts
Per
sonal
innovat
iven
ess,
per
ceiv
ed e
njo
ym
ent,
per
ceiv
ed e
ase
of
use
, per
ceiv
ed
use
fuln
ess
and
inte
nti
on t
o u
se
Res
ult
s in
dic
ated
posi
tive
corr
elat
ions
bet
wee
n
per
sonal
in
novat
iven
ess,
per
ceiv
ed
enjo
ym
ent,
per
ceiv
ed
ease
of
use
, per
ceiv
ed
use
fuln
ess
and
inte
nti
on t
o a
ccep
t m
ixed
rea
lity
tec
hnolo
gy.
Ho
wev
er,
findin
gs
from
re
gre
ssio
n
anal
ysi
s su
gges
ted
th
at
per
ceiv
ed u
sefu
lnes
s w
as t
he
most
im
port
ant
pre
dic
tor
in d
eter
min
ing u
sers
’ in
tenti
on t
o u
se t
his
tec
hn
olo
gy
in t
he
futu
re.
Kum
ar a
nd
Rav
indra
n
(2012)
An e
mpir
ical
stu
dy o
n
serv
ice
qu
alit
y
per
cepti
ons
and
conti
nuan
ce i
nte
nti
on
in m
obil
e ban
kin
g
conte
xt
in I
ndia
184 c
ust
om
ers
Per
ceiv
ed s
ervic
e
qual
ity, per
ceiv
ed
use
fuln
ess,
eas
e of
use
, per
ceiv
ed
cred
ibil
ity,
sati
sfac
tion,
per
ceiv
ed r
isk a
nd
inte
nti
on
The
resu
lts
dis
cover
ed th
at per
ceiv
ed se
rvic
e q
ual
ity
had
a s
ignif
ican
t ef
fect
on s
atis
fact
ion.
Per
ceiv
ed r
isk
had
an
ad
ver
se
impac
t on
serv
ice
qual
ity
and
sati
sfac
tion.
Inte
nti
on t
o u
se o
f m
-ban
kin
g w
as f
ound
to
be
sole
ly
dep
end
ent
on
sati
sfac
tion.
How
ever
,
per
ceiv
ed e
ase
of
use
an
d p
erce
ived
ris
k d
id n
ot
exer
t
any s
ignif
ican
t in
fluen
ce o
n s
atis
fact
ion.
51
Tab
le 2
.1, co
nti
nued
Auth
ors
T
itle
s S
ample
C
onst
ruct
s M
easu
red
M
ajor
Fin
din
gs
Gib
son,
Har
ris
and
Cola
ric
(2008)
Tec
hnolo
gy
Acc
epta
nce
in a
n
Aca
dem
ic C
onte
xt:
Fac
ult
y A
ccep
tance
of
Onli
ne
Educa
tion
110 f
acult
y
mem
ber
s
Per
ceiv
ed e
ase
of
use
and p
erce
ived
use
fuln
ess
The
resu
lts
show
ed t
hat
TA
M w
as a
ppli
cable
to f
ost
er
a re
ason
able
ex
pla
nat
ion o
f th
e in
tenti
ons
of
ph
ysi
cian
s
in u
sing t
elem
edic
ine
tech
nolo
gy.
Per
ceiv
ed u
sefu
lnes
s
was
found t
o h
ave
a si
gn
ific
ant
and s
tron
g i
nfl
uen
ce o
n
the
ph
ysi
cian
s’
inte
nti
on
to
use
te
lem
edic
ine
tech
nolo
gy;
how
ever
, per
ceiv
ed e
ase
of
use
pro
vid
es
litt
le
addit
ional
pre
dic
tive
pow
er
beyond
th
at
contr
ibute
d
by
per
ceiv
ed
use
fuln
ess
of
onli
ne
educa
tion t
echnolo
gy.
Shit
tu,
Bas
ha,
Rah
man
and
Ahm
ad
(2013)
Det
erm
inan
ts o
f so
cial
net
work
ing s
oft
war
e
acce
pta
nce
: A
mult
i-
theo
reti
cal
app
roac
h
500
under
gra
du
ate
studen
ts
Per
ceiv
ed u
sefu
lnes
s,
soci
al i
den
tity
,
com
pat
ibil
ity,
gro
up
norm
, obse
rvab
ilit
y
and a
ccep
tan
ce t
o
use
soci
al
net
work
ing
The
resu
lts
indic
ated
th
at co
mpat
ibil
ity,
gro
up norm
and
obse
rvab
ilit
y
po
siti
vel
y
infl
uen
ced
soci
al
net
work
ing a
ccep
tan
ce.
How
ever
, per
ceiv
ed u
sefu
lnes
s
and s
oci
al i
den
tity
did
not
hav
e a
signif
ican
t ef
fect
on
the
acce
pta
nce
in u
sing s
oci
al n
etw
ork
ing.
Alj
uai
d,
Alz
ahra
ni
and I
slam
(2014)
Ass
essi
ng m
obil
e
lear
nin
g r
eadin
ess
in
Sau
di
Ara
bia
hig
her
educa
tion:
An
empir
ical
stu
dy.
140 l
ectu
rers
P
erce
ived
use
fuln
ess,
per
ceiv
ed e
ase
of
use
and r
eadin
ess
of
mobil
e le
arnin
g
The
resu
lts
show
ed
that
per
ceiv
ed
use
fuln
ess
and
per
ceiv
ed e
ase
of
use
wer
e si
gnif
ican
t v
alid
pre
dic
tors
of
lect
ure
rs’
read
ines
s of
mobil
e le
arnin
g
in
Sau
di
Ara
bia
n h
igh
er e
duca
tion
. M
ore
over
, per
ceiv
ed e
ase
of
use
an
d
use
fuln
ess
had
a
signif
ican
t im
pac
t on
read
ines
s of
mobil
e le
arn
ing.
52
Tab
le 2
.1, co
nti
nued
Auth
ors
T
itle
s S
ample
C
onst
ruct
s M
easu
red
M
ajor
Fin
din
gs
Isla
m (
20
14)
V
alid
atio
n o
f th
e
tech
nolo
gy s
atis
fact
ion
model
(T
SM
)
dev
eloped
in H
igher
Educa
tion:
The
Appli
cati
on o
f
Str
uct
ura
l E
quat
ion
Model
ing
285 s
tuden
ts
Com
pute
r se
lf-
effi
cacy, per
ceiv
ed
use
fuln
ess,
per
ceiv
ed
ease
of
use
and
sati
sfac
tion
The
resu
lts
dis
cover
ed t
hat
per
ceiv
ed e
ase
of
use
and
per
ceiv
ed
use
fuln
ess
had
a
stat
isti
call
y
signif
ican
t
posi
tive
dir
ect
infl
uen
ce
on
sa
tisf
acti
on.
Bes
ides
,
com
pute
r se
lf-e
ffic
acy
show
ed
a si
gnif
ican
t posi
tive
dir
ect
infl
uen
ce o
n p
erce
ived
use
fuln
ess
and
per
ceiv
ed
ease
of
use
. In
addit
ion,
the
findin
gs
also
indic
ated
that
com
pute
r se
lf-e
ffic
acy
had
a
signif
ican
t in
dir
ect
infl
uen
ce
on
sa
tisf
acti
on
med
iate
d
by
per
ceiv
ed
use
fuln
ess.
F
inal
ly,
com
pute
r se
lf-e
ffic
acy
also
exhib
ited
a s
tati
stic
ally
sig
nif
ican
t in
dir
ect
infl
uen
ce o
n
sati
sfac
tion m
edia
ted b
y t
he
per
ceiv
ed e
ase
of
use
of
wir
eles
s in
tern
et.
Wong, T
eo
and R
uss
o
(2012)
Infl
uen
ce o
f gen
der
and c
om
pute
r te
achin
g
effi
cacy o
n c
om
pute
r
acce
pta
nce
am
on
g
Mal
aysi
an s
tuden
t
teac
her
s: A
n e
xte
nded
tech
nolo
gy a
ccep
tance
model
302 s
tuden
t-
teac
her
s
Com
pute
r te
achin
g
effi
cacy, per
ceiv
ed
use
fuln
ess,
per
ceiv
ed
ease
of
use
, at
titu
de
tow
ard c
om
pute
r use
,
and b
ehav
ioura
l
inte
nti
on
Com
pute
r te
achin
g e
ffic
acy s
how
ed s
ignif
ican
t ef
fect
on
per
ceiv
ed
use
fuln
ess,
per
ceiv
ed
ease
o
f use
an
d
atti
tude.
On t
he
oth
er h
and,
per
ceiv
ed u
sefu
lnes
s had
a
signif
ican
t in
fluen
ce
on
atti
tude
and
beh
avio
ura
l
inte
nti
on,
whil
e at
titu
de
infl
uen
ced
beh
avio
ura
l
inte
nti
on.
Fin
ally
, per
ceiv
ed e
ase
of
use
was
fou
nd t
o
hav
e a
signif
ican
t ef
fect
on p
erce
ived
use
fuln
ess.
Shit
tu,
Bas
ha,
Rah
man
and
Ahm
ad
(2011)
Inv
esti
gat
ing s
tuden
ts’
atti
tude
and i
nte
nti
on
to u
se s
oci
al s
oft
war
e
in h
igher
inst
ituti
on o
f
lear
nin
g i
n M
alaysi
a
151 s
tuden
ts
Att
itude,
Subje
ctiv
e
norm
, B
ehav
iora
l
inte
nti
on, P
erce
ived
use
fuln
ess
and
Per
ceiv
ed e
ase
of
use
The
resu
lts
show
ed
that
per
ceiv
ed
use
fuln
ess,
subje
ctiv
e norm
, an
d
per
ceiv
ed
ease
of
use
had
signif
ican
t ef
fect
on th
e at
titu
de
of
studen
ts to
war
ds
soci
al s
oft
war
e ad
opti
on
; w
hil
e at
titu
de
was
fou
nd t
o
be
the
stro
nger
fa
ctor
of
studen
ts’
inte
nti
on
to
use
soci
al s
oft
war
e.
53
Tab
le 2
.1, co
nti
nued
Auth
ors
T
itle
s S
ample
C
onst
ruct
s M
easu
red
M
ajor
Fin
din
gs
Isla
m,
Len
g
and S
ingh
(2015)
Eff
icac
y o
f th
e
tech
nolo
gy s
atis
fact
ion
model
(T
SM
): A
n
empir
ical
stu
dy
180
post
gra
du
ate
studen
ts
Com
pute
r se
lf-
effi
cacy, per
ceiv
ed
use
fuln
ess,
per
ceiv
ed
ease
of
use
and
sati
sfac
tion
The
findin
gs
show
ed t
hat
com
pute
r se
lf-e
ffic
acy h
ad a
stat
isti
call
y
signif
ican
t dir
ect
infl
uen
ce
on
per
ceiv
ed
ease
of
use
an
d
per
ceiv
ed
use
fuln
ess.
O
n
the
oth
er
han
d,
studen
ts’
per
ceiv
ed ea
se of
use
an
d per
ceiv
ed
use
fuln
ess
had
a s
tati
stic
ally
sig
nif
ican
t posi
tive
dir
ect
infl
uen
ce o
n t
hei
r sa
tisf
acti
on i
n u
sing o
nli
ne
rese
arch
dat
abas
es.
Mo
reover
, co
mpute
r se
lf-e
ffic
acy
had
a
signif
ican
t in
dir
ect
infl
uen
ce o
n s
atis
fact
ion m
edia
ted
by
per
ceiv
ed
ease
of
use
. F
inal
ly,
com
pute
r se
lf-
effi
cacy
was
al
so
rev
eale
d
to
hav
e a
stat
isti
call
y
signif
ican
t in
dir
ect
infl
uen
ce o
n s
atis
fact
ion m
edia
ted
by p
erce
ived
use
fuln
ess
of
dat
abas
es.
54
2.4 TECHNOLOGY SATISFACTION MODEL (TSM)
The gratification of ICT usage by lecturers, like many cases of technology
gratification, may be understood from the perspective of the Technology Satisfaction
Model (TSM). The TSM, proposed by Islam (2014) is developed and validated by
incorporating two additional intrinsic motivation attributes, namely, computer self-
efficacy and satisfaction into the original Technology Acceptance Model (Davis et al.,
1989). The TSM posits that three factors influence technology satisfaction, the first
being users’ perception of the benefits derived from using the technology (Perceived
Usefulness or PU), the second, how easy they perceive the use of technology
(Perceived Ease of Use or PEU) and the third, technology requires a considerable
degree of computer self-efficacy (CSE) to perform effective searches. These factors
then determine users’ satisfaction in using the technology in higher education. The
TSM contains one exogenous variable (CSE), two mediated variables (PEU, PU) and
one endogenous variable (SAT), which are shown in Figure 2.2.
Figure 2.2: Technology Satisfaction Model (TSM) (Islam, 2014)
The initial TSM was applied to assess the satisfaction on wireless internet
usage with a particular focus to the students studying in Higher Education. However,
TSM ignored the issues of two extrinsic motivation components, namely, intention to
use and actual use, even though they are important constructs in measuring users’
technology acceptance or adoption. The findings of the study have in many ways
55
contributed to the body of knowledge on the Technology Satisfaction Model (TSM).
In addition, the findings of TSM indicated that perceived ease of use and perceived
usefulness had a statistically significant positive direct influence on satisfaction.
Subsequently, computer self-efficacy was found to have a significant positive direct
influence on perceived usefulness and perceived ease of use. Moreover, the results
also demonstrated that computer self-efficacy had a significant indirect influence on
satisfaction mediated by perceived usefulness. Eventually, computer self-efficacy was
also revealed to have a statistically significant indirect influence on satisfaction
mediated by perceived ease of use of wireless internet. However, Islam (2014)
recommended that the usage of internet and the users hold a global pattern; thus this
model has a great potential to work globally regardless of geographical, economical,
social and cultural patterns in a particular country or region. As such, this study
included TSM as one of the theoretical frameworks to develop and validate the
Technology Adoption and Gratification (TAG) model to assess lecturers’ ICT usage
in higher education.
On the other hand, Lee and Park (2008) examined the relationship between the
mandatory adoption of mobile technology and market performance in the business-to-
business (B2B) setting. They presented and validated the B2B technology satisfaction
model, adding perceived loss of control with satisfaction and two extrinsic motivation
components of TAM, namely, perceived usefulness and perceived ease of use (Davis,
1989). However, they ignored the issue of computer self-efficacy even though this
was demonstrated to be the most significant predictor of TSM developed and
validated by Islam (2014). The proposed B2B technology satisfaction model showed
that the construct of user satisfaction was a mediator variable, while perceived market
performance was an endogenous variable. In this case, it would be better to have user
56
satisfaction as an endogenous variable to claim or generalize the B2B TSM. The
findings demonstrated that perceived usefulness had a significant direct effect on user
satisfaction and perceived market performance. Subsequently, perceived ease of use
has a significant direct effect on user satisfaction. However, Lee and Park (2008) did
not assess the indirect relationships between the latent variables. User satisfaction was
represented by only two valid items which was not sufficient for measuring or
validating the B2B TSM. In other related studies, Adamson and Shine (2003)
regarded perceived ease of use as a marginal influencer of the end user satisfaction in
a mandatory environment of a Bank treasury. Their findings suggested that computer
self-efficacy and user satisfaction played an important role in new technology
acceptance, thus the two constructs require more thought when designing information
systems in mandatory environments. Along this line, Al-Jabri and Roztocki (2010)
proposed a conceptual mandatory information technology implementation framework.
It included the TAM with user satisfaction and explained based on literature review,
the complexity of issues of technology adoption and use, especially in mandatory
settings. However, there is no validation to their proposed framework with empirical
data to examine the causal relationships among the constructs. Shipps and Phillips
(2013) found that concentration and perceived interactivity influenced an end user’s
satisfaction with a social network along with antecedents from the TAM. The
structural model indicated that perceived usefulness poorly influenced satisfaction (β
= .16).
57
2.5 ONLINE DATABASE ADOPTION AND SATISFACTION (ODAS)
MODEL
Islam (2011a) developed and validated the Online Database Adoption and Satisfaction
(ODAS) Model by consisting of the constructs computer self-efficacy and satisfaction
into the TAM (Davis, 1989). The ODAS model containing three exogenous variables
(perceived ease of use, perceived usefulness and computer self-efficacy), one
mediated variable (intention to use) and one endogenous variable (satisfaction) as
shown in Figure 2.3. The findings revealed significant influences of perceived
usefulness and computer self-efficacy on postgraduate students’ intention to use the
database and their satisfaction in using it. In addition, perceived ease of use and
intention to use were found to have a significant direct influence on satisfaction. The
findings contributed to a better understanding of the TAM and technology adoption
among postgraduate students. Moreover, the ODAS Model (Islam, 2011a) also
demonstrated that perceived ease of use found to be most significant predictor
compared to the behavioural intention influence on the students’ satisfaction in using
the online database. Besides this, the results of ODAS model discovered that the two
exogenous variables (PU and CSE) explained almost 53% of the variance intention to
use, and the three exogenous variables (PEU, PU and CSE) and one mediated (INT)
variable together explained approximately 61% of the variability of the students’
satisfaction and adoption in using the online database. However, ODAS model also
ignored the issue of an extrinsic motivation attribute, namely, actual use, which is a
vital factor in assessing users’ technology adoption.
58
Figure 2.3: Online Database Adoption and Satisfaction (ODAS) Model (Islam, 2011a)
In a recent study, Ahmad et al., (2010) demonstrate that computer self-efficacy
influences technology use. In fact, the influence of this construct is evident in many
cases involving the employment of computer technology (Ball & Levy, 2008;
Charalambous & Papaioannou, 2010; Kenzie, Delecourt & Power, 1994; Moos &
Azevedo, 2009). Eastin and LaRose (2000) studied computer self-efficacy in the
context of Internet use among experienced and novice users, and concluded that it is a
determining factor in closing the digital divide between the two groups of users.
Moreover, TSM (Islam, 2014) and ODAS model (Islam, 2011a) also explored that
computer self-efficacy was the most important construct in measuring students’
technology satisfaction and adoption in higher education. Based on this observation
and taking into account that the universities ICT facilities entail a considerable degree
of computer self-efficacy to perform effective usage, therefore this study proposed the
inclusion of computer self-efficacy (CSE) as an important predictor of lecturers’
adoption and gratification of ICT facilities for their teaching and research.
Lecturers’ intention to use and actual use of the ICT are more likely to persist
if they are satisfied with the resources and facilities provided therein, and if their
educational or research needs are met. Togia and Tsigilis (2009) found satisfaction to
be one of the barriers hampering graduate students’ utilization of the electronic
resources made available to them. While TSM discovered that students’ satisfaction
59
was influenced computer self-efficacy, perceived usefulness and perceived ease of use
of wireless internet in higher education (Islam, 2014). Along this line, ODAS model
also found that satisfaction was determined by intention to use, computer self-efficacy,
perceived usefulness and perceived ease of use (Islam, 2011a). Based on this
observation, the present study included gratification (GRAT) as an endogenous
variable explaining ICT facilities gratification among the universities lecturers. This
study predicted that gratification would also directly and indirectly be affected by the
ICT’s ease of use, usefulness, computer self-efficacy, intention to use and actual use.
Therefore, in the study’s hypothesized model as shown in Figure 2.4, two other
intrinsic motivation attributes – gratification and computer self-efficacy – were taken
from TSM and ODAS model and incorporated into the original TAM to develop and
validate the Technology Adoption and Gratification (TAG) model to examine
lecturers’ adoption of and gratification in using the universities ICT facilities.
Figure 2.4: The Hypothesized Technology Adoption and Gratification (TAG) Model
60
2.6 CONSTRUCTION OF HYPOTHESES
In this section, this study provided a concise but relevant literature review pertaining
to the constructs impacting lecturers’ utilization of ICT facilities, as well as the use of
the TAM, TSM and ODAS model in predicting technology adoption and gratification.
Based on the review, the study proposed eleven hypotheses to develop and validate
the TAG model and elucidated the dimensions influencing lecturers’ adoption and
gratification of ICT facilities provided by the universities. These hypotheses clarified
the causal relationships among all the constructs under study: computer self-efficacy,
perceived usefulness, perceived ease of use, intention to use, actual use and
gratification in using ICT facilities in higher education.
According to Bandura and Schunk (1981), self-efficacy is defined as one's
judgments about how well one can perform various courses of actions in different
prospective situations fraught with many unpredictable and stressful elements. This
definition was echoed by Aiken (1993), who opined that self-efficacy stands for a
person's expectation of his/her capability in learning or performing certain behaviours
that will translate into desirable outcomes in a particular situation. In the learning
process, it conveys students’ judgments about their cognitive capabilities to
accomplish a specific academic task or particular goals (Schunk, 1991). In the context
of Internet use, it denotes a user's belief in his/her capabilities to organize and execute
courses of Internet actions required to produce given attainments. This, Eastin and
LaRose (2000) asserts, is a determining factor in efforts to close the digital divide
between the experienced Internet users and the novices.
Research confirms propositions that self-efficacy influences the choice as to
whether or not to engage in a task, the effort made in performing it, and the level of
persistence required in accomplishing it (Bandura, 1977; Bandura & Schunk, 1981;
61
Delcourt & Kinzie, 1993). It influences how people feel, think, and behave (Bandura,
1986: 393). According to Ahmad et al., (2010, p. 271), the inclusion of self-efficacy as
an intrinsic motivation construct offers ‘deeper and richer understanding of why and
how the technology is used’.
In their study, Compeau and Higgins (1995) empirically tested a 10-item
measure of computer self-efficacy (CSE) and found a significant relationship between
CSE and outcome expectations and use. In a further empirical validation of the CSE
instrument developed in this study, Compeau, Higgins, and Huff (1999) reaffirmed the
evidence that CSE influences an individual’s behavioral orientations to adopt an
information system.
The findings of TSM (Islam, 2014) discovered that computer self-efficacy is
the most important predictor in measuring the students’ satisfaction in using wireless
internet for their learning purposes. Moreover, TSM also showed that computer self-
efficacy revealed a significant positive direct influence on perceived usefulness in
using wireless internet. On the other hand, Ahmad et al., (2010) indicated that
computer self-efficacy indirectly influences faculty’s use of computer mediated
technology through perceived usefulness and intention to use. Subsequently, Islam
(2011a) revealed the presence of significant indirect relationship between computer
self-efficacy and satisfaction through the mediating influence of intention to use
online research database. Along this line, in his book entitled “Online Database
Adoption and Satisfaction Model”, Islam, (2011a) also discovered the significant
interrelationships among the exogenous variables were measured in the revised
structural model, and it was found that the interrelationships between computer self-
efficacy, perceived ease of use and perceived usefulness were statistically significant.
Similarly, Islam (2011b) also demonstrated significant positive interrelationships
62
between computer self-efficacy, perceived ease of use and perceived usefulness.
YouTube users’ self-efficacy showed a significant effect on perceived usefulness (Lee
& Lehto, 2013). Stone and Baker-Eveleth (2013) hypothesized that a mid-sized
university students’ self-efficacy regarding the use of an e-textbook significantly
influences outcome expectancy/usefulness regarding e-textbooks. However, their
findings showed that it did not exert statistically significant effect on outcome
expectancy/usefulness in purchasing electronic textbooks. In a previous study, pre-
service teachers’ self-efficacy levels regarding ICT usage in education were found to
be moderate. Thus, teachers’ self-efficacy or self-confidence concerning ICT usage
must be high in order for them to be highly encouraged users of ICT, which in turn
would lead to flourishing ICT technology integration (Tezci, 2011). Amin (2007)
confirmed that university students’ computer self-efficacy of internet banking had a
significant effect on their perceived usefulness, while it was not related with perceived
credibility of internet banking system. Zejno and Islam (2012) demonstrated that
computer self-efficacy is one of the significant predictors of postgraduate students’
ICT usage in higher education. In a related study, Wong et al., (2012) revealed that
computer teaching efficacy had a significant effect on perceived usefulness (β=.22,
p<.00), perceived ease of use (β=.48, p<.00) and attitude toward computer use (β=.38,
p<.00). However, the findings also showed that computer teaching efficacy
moderately effects on usefulness compare to other constructs.
On the other hand, according to Islam (2014), computer self-efficacy had a
significant positive direct influence on perceived ease of use of wireless internet.
Islam (2011a & 2011b) also exhibited significant interrelationships between computer
self-efficacy and perceived ease of use. Terzis et al., (2013) predicted that Greek and
Mexican university students’ computer self-efficacy will have a positive influence on
63
their Perceived Ease of Use. Their findings were consistent with the prediction. The
significant influence of Greek and Mexican university students’ computer self-
efficacy shows that a student who recognizes how to use computers, perhaps he/she
will find easy to use a Computer Based Assessment that needs fundamental
information technology skills. Amin (2007) found that university students’ computer
self-efficacy had a positive effect on their perceived ease of use of the Internet
banking systems. In a study involving university students’ use of an information
system, Agarwal and Karahanna (2000) discovered CSE to be a key antecedent of
PEU, while Wu et al., (2008) found a statistically significant relationship between
teachers’ CSE and their intention to use ICT in teaching. In light of such findings, it is
important to assess lecturers’ beliefs about their computer ability and the extent to
which it impacts their decision to adopt technology. Besides this, Islam et al., (2015)
revealed that postgraduate students’ computer self-efficacy had a statistically
significant direct influence on perceived ease of use and perceived usefulness in using
online research databases in higher education. These observations led to hypothesize
that in the study:
H1: Computer self-efficacy (CSE) will have a positive direct influence on lecturers’
perceived ease of use (PEU) and perceived usefulness (PU) of ICT facilities in higher
education.
In addition to perceived benefits, another important influencer is how easy
lecturers find the ICT facilities to use. In the original TAM, perceived ease of use
means the extent to which the user believes the use of technology will be free of effort
(Davis et al., 1989). Prior studies have found a significant influence of PEU on users’
intention to use computer-based systems (Chau, 1996; Davis et al. 1989; Karahanna &
Straub, 1999; Venkatesh & Davis, 1996, 2000; Venkatesh & Morris, 2000). However,
64
PEU has been found to be less influential and reliable (Ma & Liu, 2004; Wu et al.,
2008; Huang, 2008). Tenopir (2003), specifically, discovered that ease of use was a
major factor influencing her respondents’ adoption of the online database. Outside the
educational context, Lee, Fiore and Kim (2006) demonstrated that PEU significantly
enhanced consumer attitude and behavioral intention towards an online retailer, while
Ramayah and Lo (2007) revealed a significant effect of PEU on intention to use an
electronic resource planning system. Moreover, Lee and Lehto (2013) hypothesized
that perceived ease of use will have a significant effect on YouTube users’ behavioral
intention. Nevertheless, the results of their study showed that perceived ease of use
did not indicate significant effect on either behavioral intention or perceived
usefulness of YouTube. Similarly, Islam (2011a) also predicted that perceived ease of
use will have a positive direct influence on postgraduate students’ intention to use the
online research databases. However, the finding of his study revealed that perceived
ease of use had a significant direct effect on postgraduate students’ intention to use of
online research databases, but in an adverse direction (β = -.31, p < .005).
In addition, Hanafizadeh et al., (2014) found out that bank clients’ perceived
ease of use had significant influence on intention to use mobile banking. Subsequently,
Gultekin (2011) discovered that the effects of perceived ease of use on police officers’
intention to use the POLNET system were statistically significant. Meanwhile, Stone
and Baker-Eveleth (2013) confirmed that perceived ease of e-textbook use positively
influences behavioral intentions to purchase an e-textbook, while Martínez-Torres et
al., (2008) discovered that perceived ease of use, despite indicating a slightly weaker
influence compared to perceived usefulness on behavioural intention to use e-learning
tools. In other study, Shroff et al., (2011) confirmed that students’ perceived ease of
use had a significant influence on attitude towards usage of portfolio system.
65
Consequently, perceived ease of use had the strongest significant influence on
perceived usefulness. Along this line, Chang et al., (2012) demonstrated that
perceived ease of use was antecedent factor that affected intention to accept of English
mobile learning. Terzis et al., (2013) found that the students’ perceived ease of use
had a significant effect on intention to use computer based assessment in higher
education.
On the other hand, the Online Database Adoption and Satisfaction Model
(Islam, 2011) revealed that perceived ease of use found to be more significant
predictor compared to the behavioural intention influence on the students’ satisfaction
in using the online research databases. According to Langeard, Bateson, Lovelock,
and Eiglier (1981), the study discovered that, customers consider effort expended in
utilizing service delivery to be more relevant in choosing between available service
delivery options, the greater the service delivery, the more the consumer satisfaction
(Churchill & Suprenant, 1982). Szymanski & Hise (2000) and Dabholkar & Bagozzi
(2002) respectively revealed that, convenience and ease of use are important
antecedents of e-satisfaction. In a related study, Shamdasani et al. (2008) regarded
ease of use as one of the factors of service quality and investigated its impact on
consumer satisfaction in using self-service internet technologies. Along this line,
Huang (2008) indicates that, the impact of e-consumers’ perceived ease of use
mediated by their behavioral attitude on their satisfaction are statistically significant.
The TSM confirmed that perceived ease of use demonstrated a significant direct effect
on students’ satisfaction in using wireless internet (Islam, 2014). Meanwhile, Islam
(2011b) also represented that perceived ease of use had a statistically significant effect
on students’ satisfaction in using wireless internet technology in higher education.
Furthermore, Lee and Park (2008) discovered that perceived ease of use had a
66
significant effect on user satisfaction in using mobile technology. On the other hand,
the significance of perceived ease of use in predicting technology adoption is highly
emphasized in the TAM (Davis et al., 1989). In addition, Devaraj et al. (2002)
discovered that perceived ease of use was found to be a significant determinant of
satisfaction in transaction cost analysis as well as important component of TAM in
forming consumer attitude and satisfaction with the electronic commerce channel,
while Islam et al., (2015) confirmed that perceived ease of use had a significant effect
on students’ satisfaction in using online research database in higher education. These
observations led to next hypothesis, which reads as follows:
H2: Perceived ease of use (PEU) will have a positive direct influence on lecturers’
intention to use (INT), actual use (USE) and gratification (GRAT) in using ICT
facilities in higher education.
In the TAM, perceived usefulness is defined as the extent to which an
individual believes that using the technology will enhance his or her performance
(Davis 1989). PU has also been conceived as the benefits and advantages derived from
using a technology. The impact of PU on users’ intention to use (INT) various types of
computer-based and Internet-based systems has been comprehensively documented in
numerous studies (Ahmad et al. 2010; Davis, 1993; Guriting & Ndubisi, 2006;
Mathwick, Rigdon & Malhotra, 2001; Ramayah & Lo, 2007). Convenience of access,
in particular, has been frequently cited as a major advantage by users of electronic
journals in multiple studies (Maughan, 1999; Tenner & Yang, 1999; Hiller, 2002). In
Liew, Li, Foo and Chennupati (2000), the benefits of electronic resources were found
to be the facility to link to additional information, the ability to browse, and the
currency of materials. Other benefits and advantages include robust search capabilities,
hyperlinks to outside content, efficient acquisition of information (Dillon & Hahn,
67
2002; Ray & Day, 1998), time saving factor, and print and save capabilities (Togia &
Tsigilis, 2009). These benefits and advantages were realized all the more by students
when their learning and academic pursuits became greatly enhanced. Wu et al., (2008)
corroborated these findings on the impact of PU by noting that it is a significant
determinant of science teachers’ intention to integrate technology into their instruction.
In recent studies have also explored a significant effect of perceived usefulness on
users’ intention to use mobile learning (Hanafizadeh et al., 2014; Chang et al., 2012),
POLNET system (Gultekin, 2011), e-learning tools (Martínez-Torres et al., 2008) and
e-portfolio system (Shroff et al., 2011), while Terzis et al., (2013) discovered that
perceived usefulness showed a significant influence on students’ intention to use
computer based assessment in higher education.
The TSM (Islam, 2014) showed that perceived usefulness had a statistically
significant positive direct influence on satisfaction in using wireless internet in higher
education. Regarding the Online Database Adoption and Satisfaction Model, Islam,
(2011a) discovered that, perceived usefulness has a statistically significant indirect
influence on satisfaction mediated by intention to use online database, while Islam et
al., (2015) found that perceived usefulness was the most significant factor of TSM in
effecting postgraduate students’ satisfaction in using online research database in
higher education. Besides this, Lee and Park (2008) found that perceived usefulness
had a significant effect on user satisfaction, a surrogate of information system success.
Similarly, Islam (2011b) hypothesized that perceived usefulness positively influences
students’ satisfaction in their acceptance of wireless technology. Even though, the
results indicated the low path coefficients of 0.19 between perceived usefulness and
satisfaction. In addition to this, in the extant literature, numerous studies exist vis-à-
vis the effect of perceived usefulness on consumer satisfaction (Anderson, Fornell, &
68
Lehmann, 1994). According to Huang (2008), perceived usefulness was found to have
an impact on consumer satisfaction mediated by their behavioral attitudes. On the
other hand, Clemons and Woodruff (1992) reported that, perceived value might
directly lead to the formation of overall feelings of satisfaction. This argument is
corroborated by McDougall and Levesque (2000) who indicated that, perceived value
is a significant driver of customer satisfaction. Devaraj et al., (2002) revealed that
perceived usefulness was an essential factor in forming consumer attitude and
satisfaction with the electronic commerce channel. Against this backdrop, It is thus
hypothesized that:
H3: Perceived usefulness (PU) will have a positive direct influence on lecturers’
intention to use (INT), actual use (USE) and gratification (GRAT) in using ICT
facilities in higher education.
H9: Perceived ease of use (PEU) and Perceived usefulness (PU) will have a positive
indirect influence on lecturers’ gratification (GRAT) in using ICT facilities mediated
by intention to use (INT), respectively.
The result of ODAS model (Islam, 2011a) depicted that intention to use had a
positive direct effect on students’ satisfaction in using online research databases in
higher education. From various prior studies performed on the relationship between
CSE and behavioural intention to use technology or infusion (Wu et al., 2008; Ahmad
et al., 2010) and the latter with individual satisfaction (Huang, 2008; Shamdasani et
al., 2008), it is inferred that there is a relationship between intention to use and
satisfaction. Against this backdrop, one purpose of this study was to develop and
validate the TAG model by incorporating computer self-efficacy and gratification as
two attributes to examine lecturers’ adoption and gratification in using ICT facilities.
The presented study assessed the direct effects of intention to use on lecturers’ actual
69
use and gratification in using the ICT facilities for their teaching and research
purposes, and hypothesized that:
H4: Intention to use (INT) will have a positive direct influence on lecturers’
gratification (GRAT) and actual use (USE) of ICT facilities in higher education.
H5: Use (USE) will have a positive direct influence on gratification (GRAT) in using
ICT facilities.
H10: Intention to use (INT) will have a positive indirect influence on lecturers’
gratification (GRAT) mediated by actual use (USE) of ICT facilities in higher
education.
Havelka’s (2003) study perceived significant differences in CSE among
students in different areas of academic discipline. In light of such findings, it is
imperative on the part of the researchers to assess students' self-beliefs about their
academic capabilities as well as the acknowledgement of this aspect as being
instrumental on their motivation and academic achievement (Pajares, 2002). Besides,
Wu et al. (2008) found a statistically significant relationship between CSE and
intention to infuse Information Technology, although Stone and Baker-Eveleth (2013)
indicated that computer self-efficacy positively influenced students’ attitudes
regarding purchasing e-textbooks. Alenezi et al., (2010) confirmed that computer self-
efficacy was significantly influence the students' intention to use e-learning, while
Ahmad et al., (2010) hypothesized that computer self efficacy indirectly effects
faculty’s use of the computer mediated technology medicated by perceived usefulness
and intention to use. Their findings concluded that computer self efficacy was
moderately more influential than did the behavioural intention in influencing the
faculty’s use of computer technology. According to Chang and Tung (2008), self-
efficacy was an important factor for university students’ behavioural intention to use
70
the online learning course websites. Similarly, Zejno and Islam (2012) claimed that
computer self-efficacy is one of the significant predictors of postgraduate students’
ICT usage.
In addition, Islam (2014) demonstrated that computer self-efficacy is the most
significant construct of the TSM and it also exerted indirect significant influence on
satisfaction mediated by perceived ease of use and perceived usefulness, respectively.
The effect of computer self-efficacy on perceived usefulness, ease of use and
satisfaction were limited to their capabilities and requisite skills in using wireless
internet facility and online research databases, saving and printing journals or articles,
and accessing the database from the university website. However, students described
obstacles related to accessing journals from the online database. In other studies,
computer self-efficacy showed a significant indirect effect on students’ satisfaction
mediated by intention to use the online database (Islam, 2011a, p. 50). On the other
hand, Islam (2011b) hypothesized that computer self-efficacy positively influences
students’ satisfaction in accepting wireless technology. However, it did not depict the
direct impact on satisfaction; while it might be reflected through the significant
interrelationships among the exogenous variables, namely, perceived ease of use and
perceived usefulness. At the same time, Islam et al., (2015) discovered that students’
computer self-efficacy had a significant indirect influence on satisfaction mediated by
perceived ease of use and perceived usefulness of research databases. In other studies,
Ahmad et al., (2010), Gong, Xu and Yu (2004), Hu, Clark and Ma (2003) and Ellen,
Bearden and Sharma (1991) demonstrate that computer self-efficacy influences
technology acceptance and use. In fact, the influence of this construct is evident in
many cases involving the employment of computer technology (Ball & Levy, 2008;
Charalambous & Papaioannou, 2010; Kenzie, Delecourt & Power, 1994; Moos &
71
Azevedo, 2009). Eastin and LaRose (2000) studied computer self-efficacy in the
context of Internet use among experienced and novice users, and concluded that it is a
determining factor in closing the digital divide between the two groups of users. It is
therefore hypothesized that:
H6: Computer self-efficacy (CSE) will have a positive indirect influence on lecturers’
intention to use (INT) of ICT facilities mediated by perceived ease of use (PEU) and
perceived usefulness (PU), respectively.
H7: Computer self-efficacy (CSE) will have a positive indirect influence on lecturers’
gratification (GRAT) in using ICT facilities mediated by perceived ease of use (PEU)
and perceived usefulness (PU), respectively.
H8: Computer self-efficacy (CSE) will have a positive indirect influence on lecturers’
use (USE) of ICT facilities mediated by perceived ease of use (PEU) and perceived
usefulness (PU), respectively.
2.6.1 CROSS-CULTURAL STUDIES
Terzis et al., (2013) stated that the necessity to comprehend the individual acceptance
drives numerous researchers to a cross cultural analysis concerning acceptance. Many
prior studies included a cross-cultural factor by comparing the efficacy of an
acceptance model such as TAM or by comprising constructs that were distinguishing
the cultures. However, earlier cross-cultural researches exhibited that acceptance or
adoption models such as TAM were affected by culture. As such, they examined
university students’ acceptance of Computer Based Assessment (CBA) Systems by
applying Computer Based Assessment Acceptance Model (CBAAM) in the two
diverse cultures of Greece and Mexico. The CBAAM was integrated by nine factors,
namely, perceived playfulness, perceived usefulness, perceived ease of use, computer
72
self-efficacy, social influence, facilitating conditions, goal expectancy, content and
behavioral intention. To validate the model, a total of 168 students were selected from
two universities in Greece and Mexico. They consisted of 117 first-year Greek and 51
first-year Mexican students. The CBAAM was estimated using partial least squares
(PLS) to test fifteen hypotheses in a cross cultural setting. The findings of their study
demonstrated that the CBAAM was valid for both countries in general. Nevertheless,
there were some distinctions due to the culture. Greek students’ behavioral intention
was primarily influenced by perceived ease of use and perceived playfulness, although
Mexican students’ behavioral intention was affected by perceived usefulness and
perceived playfulness. However, these results may not be generalized due to the small
sample size. Besides this, this study was limited between two universities in Greece
and Mexico. They suggested that the findings of this study could be useful for
researchers, developers and educators and they should take them into contemplation
for future, especially researches interest in cultural dimensions and their influences in
education.
Teo, Luan and Sing (2008) conducted a cross-cultural study to investigate the
pre-service teachers’ intention to use technology in Singapore and Malaysia applying
the Technology Acceptance Model (TAM). However, so far, there was no cross-
cultural research to assess pre-service teachers’ acceptance by applying TAM in
Singapore and Malaysia so that it would be of interest to the research community to
observe if the TAM could be validated in the diverse culture. The data for their study
were gathered through an online survey questionnaire. The pre-service teachers for
their study consisted of 250 and 245 in Singapore and Malaysia, respectively. They
proposed research model based on TAM for this study which included the four factors,
(a) perceived ease of use, (b) perceived usefulness, (c) computer attitude, and (c)
73
behavioral intention control, influence the future intentions to use computer in the
different culture. The Structural Equation Modeling (SEM) was used to validate the
model. The results of their study revealed that there were significant dissimilarities
between Malaysian and Singaporean pre-service teachers in terms of their computer
attitude, perceived ease of use and perceived usefulness of technology. However, no
differences were identified in behavioral intention among these two countries with
regards to technology acceptance. They recommended that the SEM could be applied
to assess structural invariance of the TAM to examine of its validity in the cross-
culture. In addition, future study also can extend TAM by including additional
variables of interest to the education community.
According to Singh, Fassott, Chao and Hoffmann (2006), the technology
acceptance model (TAM) has been one of the most influential models in the
information technology literature. However, TAM has not been applied in the
marketing literature to comprehend user acceptance of international web sites. In so
doing, they extended the TAM by incorporating cultural adaptation factor to measure
international students’ acceptance and usage of multinational company web sites in
the different cultures in German, Brazil and Taiwan. They also asserted that one of the
causes they preferred to collect data from Brazil, Germany, and Taiwan that the
findings were more comparable and meaningful in the different culture. Data were
collected from 40 Brazilian, 110 German, and 100 Taiwanese online consumers.
However, the total sample size seems to be inadequate in order to generalize the
findings of their study. To evaluate the extended TAM and its causal relationships
between the constructs, the partial least squares (PLS) were applied. The result of their
study discovered that the causal relationships between the extended TAM factors,
namely, perceived ease of use, perceived usefulness, cultural adaptation, attitude and
74
behavioral intention to use international websites were presented strong evidence for
the viability of TAM in elucidating web site usage in Brazil, Germany, and Taiwan.
The findings also showed that the total influence of consumers’ perceived usefulness
was greater than that of perceived ease of use PEOU on intention to purchase, while
perceived usefulness had a greater effect on students’ attitude toward the website in
comparison to perceived ease of use. Furthermore, cultural adaptation was found to be
an important exogenous variable for the three diverse cultures in Brazil, Germany, and
Taiwan in using international web site. These three countries students’ attitude toward
using the web site was demonstrated to have a strong influence on their intention to
purchase from the international web sites. They suggested based on their findings that
future researchers may apply TAM to examine numerous forms of technology usage
and in diverse contexts to discover its validity and frontier surroundings.
On the one hand, Al-Gahtani, Hubona and Wang (2007) investigated a study
of Culture and the acceptance and use of information technology (IT) in Saudi Arabia.
Using a survey of 722 knowledge workers those who were utilizing desktop computer
applications on a voluntary basis in Saudi Arabia, they applied a modified unified
theory of acceptance and use of technology (UTAUT) to measure users’ acceptance
and use of IT. Their findings confirmed that the UTAUT model explained 39.1% of
intention to use, and 42.1% of usage of the total variance. Besides this, applying on
the theory of cultural dimensions, they examined the similarities and dissimilarities
between the Saudi Arabia and North American validations of UTAUT with regards of
cultural diversities that influenced the users’ acceptance of IT in these two societies.
Straub, Keil and Brenner (1997) estimated the technology acceptance model
(TAM) through comparison among three different countries: United States; Japan; and
Switzerland. The data for their study were collected from 99 American, 142 Japanese,
75
and 152 Swiss employees of three different airlines. Their results indicated that TAM
was influenced by the different cultures. Additionally, TAM model was significant
and explained around 10% of the variance in usage and adoption observed in both the
U.S. and Swiss respondents, it was not significant for the Japanese. However, the total
sample size may not be adequate in order to generalize their findings.
Huang, Ming-te, and Wong (2003) asserted that the culture of People’s
Republic of China (PRC) typically varies from that of North America, and is a typical
transitional economy at the accepting end of information technology transfer. Thus,
the cross-cultural validity of technology acceptance models may not be justified
without an assessment on the applicability of technology acceptance in the context of
the PRC, whereas acceptance models were explored in North America. As a result,
they conducted a study of assessing of the cross-cultural applicability of technology
acceptance model (TAM) as evident from China. The data for their study were
collected from 121 employees in Guangzhou, China. The results confirmed that the
TAM was applicable across cultures. The finding provides evidence on the
applicability of Technology Acceptance Model (TAM) across cultures. Therefore, the
validity of TAM was expanded by assessing technology acceptance beliefs in a
developing country with Chinese culture. They believed that with a broad
comprehending of cross-cultural technology adoption process and outcome,
organizations would be in a better position to utilize the usefulness of the new
communication media, particularly in culturally different non-western economies that
are at the accepting end of technology transfer. They recommended that many studies
need to be conducted in diverse countries or cultures in order to generalize the validity
and applicability of TAM.
76
In a recent cross-cultural study, Chong, Chan and Ooi (2012) extended the
technology acceptance model (TAM) and diffusion of innovation (DOI) theory by
adding more constructs, namely, trust, cost, social influence, variety of services, and
control variables such as age, educational level, and gender of consumers to develop
their conceptual model for this study. The model was validated to predict consumers’
intention to adopt mobile commerce in China and Malaysia. To develop and validate
their model, data obtained through a survey conducted with 172 Malaysian and 222
Chinese consumers, and the model was evaluated using hierarchical regression
analysis to test fourteen hypotheses. The findings of their research model
demonstrated that the factors in TAM and DOI models were not competent to predict
consumer decisions to adopt mobile commerce. However, constructs such as trust,
cost, social influence, and variety of services had a significant influence on
consumers’ intention to adopt mobile commerce. The results also showed that
consumers’ decision to adopt mobile commerce was affected by the different cultures.
They suggested that future studies can incorporate additional adoption dimensions,
namely, self efficacy and perceived enjoyment into the research models, although
their study ignored them. These observations led this study to hypothesize that:
H11: There will be a cross-cultural invariant of the causal structure of the TAG model
between Malaysian and Chinese lecturers.
77
CHAPTER 3
METHODOLOGY
3.1 INTRODUCTION
The broad objective of this study was to develop and validate the technology adoption
and gratification (TAG) model to evaluate lecturers’ ICT usage in higher education.
The factors that influenced lecturers’ adoption and gratification of the ICT facilities
were hypothesized to be the ICT’s ease of use, usefulness, computer self-efficacy,
actual use and gratification from the usage of this service.
This study also applied the method of Rasch model to develop and validate the
psychometric properties of lecturers’ ICT usage and Structural Equation Modeling
(SEM) to examine the causal relationships among the various exogenous, endogenous
and mediated constructs of TAG model. Subsequently, this chapter provides detailed
descriptions of the population and sample, sampling procedure, research instrument,
and data analysis procedures, structural equation modeling, and model evaluation.
3.2 POPULATION AND SAMPLE
The present study was conducted at two public universities, namely, University of
Malaya in Malaysia and Jiaxing University in China. As such, the target population of
this study was different from each other and the total sample size of this study was
400 academic staff, which was equally divided for both universities. The total
population of this study comprised all the academic staff of the University of Malaya
and Jiaxing University. The total sample size was recommended by Hair, Black,
Babin, and Anderson (2010). According to Hair et al., (2010), “the smallest sample
size is to have at least five times as many observations (respondents) as the number of
78
variables (items) to be analyzed”. The meaning is that each item in the questionnaire
should have minimum five respondents. In this case, the total number of items has to
be multiplied by five to compute the total sample size whereas the questionnaire of the
present study was containing almost eighty items to develop and validate the
technology adoption and gratification (TAG) Model to assess lecturers’ ICT usage.
Approximately 1965 lecturers were taken as the sampling frame; out of this, a total of
200 lecturers from thirteen faculties of the University of Malaya (Arts and Social
Sciences, Built Environment, Business and Accountancy, Computer Science and
Information Technology, Dentistry, Economics and Administration, Education,
Engineering, Language and Linguistics, Law, Medicine, Sciences and Academy of
Islamic Studies) were selected using stratified random sampling procedure.
Meanwhile, the target population of Jiaxing University was approximately 811
lecturers included as the sampling frame; out of this, a total of 200 lecturers from
seven colleges (Foreign Language Studies, Mathematics and Engineering, Biology
and Chemistry, Material Engineering, Education, Civil Engineering & Architecture
and Jiaxing University Nanhu) were collected applying stratified random sampling
procedure.
3.2.1 LECTURERS’ DEMOGRAPHIC INFORMATION IN THE
UNIVERSITY OF MALAYA
The 200 participants were lecturers from all the faculties of the University of Malaya.
They consisted of 47% male and 53% female. The gender breakdown is shown in
Figure 3.1.
79
Figure 3.1: Sample of Study According to Gender
The 200 participants were lecturers from thirteen faculties of the University of
Malaya, namely, Arts and Social Sciences (FASS), Built Environment (FBE),
Business and Accountancy (FBA), Computer Science and Information Technology
(FCSIT), Dentistry (FD), Economics and Administration (FEA), Education (FEDU),
Engineering (FE), Language and Linguistics (FLL), Law (FL), Medicine (FM),
Sciences (FS) and Academy of Islamic Studies (AIS). The sample breakdown
according to faculties is shown in Figure 3.2.
Figure 3.2: Sample of Study According to Faculties
80
The sample consisted of lecturer’s educational background. They were holding
the different levels of educational background such as undergraduate (UG), Master
and PhD and working in various faculties of the University of Malaya. The majority
of the lecturers were holding PhD degree (74%) whereas only 25% lecturers were
Master degree holders and the remaining 1% was undergraduate holders. The
breakdown of which is shown in Figure 3.3.
Figure 3.3: Sample of Study According to Educational Background
The sample consisted of different age groups. Most of the lecturers (33%)
were above 45 years of age, followed by 23%, 19% and 18% were between 35 to 39,
30 to 34 and 40 to 44 years of age groups, respectively. Lastly, only 7% of the
lecturers were between 25 to 29 years of age. The age group breakdown is shown in
Figure 3.4.
81
Figure 3.4: Sample of Study According to Age Groups
This study was conducted among local (88%) and international (12%)
academic staffs. The nationality breakdown is shown in Figure 3.5.
Figure 3.5: Sample of Study According to Nationality
82
Most of the lecturers (29%) had between 6 to 10 years of working experiences,
followed by 26% which had between 1 to 5 years, and the remaining 14%, 15% and
16% of the lecturers had between 16 to 20, 11 to 15 and 21 above years of experience,
respectively. The working experience breakdown is shown in Figure 3.6.
Figure 3.6: Sample of Study According to Working Experience
With regards to designations, majority of the academic staff surveyed were
assistant professors/senior lecturers (44%), followed by lecturers (30%), associate
professors (16%) and professors (10%), respectively. The designation breakdown is
shown in Figure 3.7.
Figure 3.7: Sample of Study According to Designation
83
3.2.2 LECTURERS’ DEMOGRAPHIC INFORMATION IN JIAXING
UNIVERSITY
The 200 participants were lecturers from seven colleges of Jiaxing University.
However, 4 participants were removed from the final study due to incomplete
responses. They consisted of 48% male and 52% female. The gender breakdown is
shown in Figure 3.8.
Figure 3.8: Sample of Study According to Gender
The 196 participants were lecturers from seven colleges of Jiaxing University,
namely, Foreign Language Studies (CFLS), Mathematics and Engineering (CME),
Biology and Chemistry (CBC), Material Engineering (CME), Education (CE), Civil
Engineering & Architecture (CCEA) and Jiaxing University Nanhu (CJUN). The
sample breakdown according to colleges is shown in Figure 3.9.
84
Figure 3.9: Sample of Study According to Colleges
The sample consisted of lecturer’s educational background. They had different
levels of educational background such as undergraduate (UG), Master and PhD and
worked in various colleges in Jiaxing University. The majority of the lecturers held
Master degree (48%) whereas only 19% of the lecturers were undergraduate holders
and the remaining 33% were PhD degree holders. The breakdown of which is shown
in Figure 3.10.
Figure 3.10: Sample of Study According to Educational Background
85
The sample consisted of different age groups. Most of the lecturers (29%)
were between 30 to 34 years of age, followed by 25%, 21%, and 17% who were
between 45 years and above, 35 to 39 and 40 to 44 years of age, respectively, whereas
only 8% of the lecturers were between 25 to 29 years of age. The age group
breakdown is shown in Figure 3.11
Figure 3.11: Sample of Study According to Age Groups
This study was conducted among local (100%) academic staff. The nationality
breakdown is shown in Figure 3.12.
Figure 3.12: Sample of Study According to Nationality
86
Most of the lecturers (32%) had between 6 to 10 years of working experience,
followed by 22% who had between 1 to 5 years, and the remaining 18%, 10% and
18% of the lecturers had between 11 to 150, 16 to 20 and 21 above years of
experience, respectively. The working experience breakdown is shown in Figure 3.13.
Figure 3.13: Sample of Study According to Working Experience
With regards to designation, majority of the surveyed academic staff were
lecturers (53%), followed by associate professors (38%), professors (7%) and assistant
professors/Senior lecturer (2%). The designation breakdown is shown in Figure 3.14.
Figure 3.14: Sample of Study According to Designation
87
3.3 SAMPLING PROCEDURE
The respondents involved in this study were from two comprehensive public
universities, namely, the University of Malaya and Jiaxing University and they are
currently working there as professors, associate professors, assistant professors or
senior lecturers and lecturers in the different faculties of the universities. The sample
size of this study was based on previous studies For instance, Terzis et al., (2013)
conducted a research on “computer based assessment acceptance: a cross-cultural
study in Greece and Mexico”. Their study was conducted at two universities in Greece
and Mexico. The data for their study was collected from 117 Greek and 51 Mexican
students. Similarly, Chong et al., (2012) investigated a cross-cultural study between
China and Malaysia to examine consumers’ decision to adopt mobile commerce. The
data was collected from 172 Malaysian and 222 Chinese consumers. Teo et al., (2008)
conducted a study on “a cross-cultural examination of the intention to use technology
between Singaporean and Malaysian pre-service teachers: An application of the
Technology Acceptance Model (TAM)”. The respondents for their study comprised of
250 and 245 pre-service teachers in Singapore and Malaysia respectively. As such, in
the current study, a total of 400 academic staffs were separated into two equal parts.
Data was collected through a survey questionnaire using stratified random sampling
procedure. The questionnaire was distributed to all the faculties (Arts and Social
Sciences, Built Environment, Business and Accountancy, Computer Science and
Information Technology, Dentistry, Economics and Administration, Education,
Engineering, Language and Linguistics, Law, Medicine, Sciences and Academy of
Islamic Studies) academic staff based on the total number of faculty members as well
as the specific respondents of each faculty. Half of lecturers of the sample were from
the University of Malaya, so the researcher was able to hand it to the lecturers of the
88
faculties: Arts and Social Sciences (12), Built Environment (8), Business and
Accountancy (7), Computer Science and Information Technology (8), Dentistry (10),
Economics and Administration (6), Education (9), Engineering (19), Language and
Linguistics (14), Law (4), Medicine (61), Sciences (32) and Academy of Islamic
Studies (10). On the other hand, the remaining 200 lecturers were collected from
seven colleges (Foreign Language Studies, Mathematics and Engineering, Biology
and Chemistry, Material Engineering, Education, Civil Engineering & Architecture
and Jiaxing University Nanhu) of Jiaxing University using stratified random sampling
procedure. To collect the data from Jiaxing University, the researcher traveled to
China and was able to hand it to the lecturers in Foreign Language Studies (20),
Mathematics and Engineering (28), Biology and Chemistry (23), Material Engineering
(16), Education (5), Civil Engineering & Architecture (17) and Jiaxing University
Nanhu (91). The formula of stratified random sampling procedure is as follows:
(expected sample size/total population)*each faculty population = each faculty sample
size. The sum of all the faculties sample size represents the total sample size of the
study. The detail of stratified random sampling procedure is shown in Table 3.1.
89
Table 3.1: Stratified Random Sampling Procedure
Universities Faculties/Colleges Each Faculty
Population
Expected Sample
Size for Each
Faculty
University of Malaya
in Malaysia
(Expected Sample
size/Total
Population)*Each
faculty population
e.g. (200/1965)*121
=Sample size for each
faculty
Arts and Social Sciences 121 12.315
Built Environment 75 7.633
Business and
Accountancy
72 7.328
Computer Science and
Information Technology
76 7.735
Dentistry 97 9.872
Economics and
Administration
60 6.106
Education 90 9.160
Engineering 186 18.931
Language and
Linguistics
140 14.249
Law 39 3.969
Medicine 598 60.865
Sciences 313 31.857
Academy of Islamic
Studies
98 9.978
Total Population of University of Malaya 1965
Total sample size for University of Malaya 199.998
Jiaxing University in
China
Foreign Language
Studies
83 20.468
Mathematics and
Engineering
113 27.866
Biology and Chemistry 95 23.427
Material Engineering 63 15.536
Education 19 4.685
Civil Engineering &
Architecture
68 16.769
Jiaxing University Nanhu
College
370 91.245
Total Population of Jiaxing University 811
Total sample size for Jiaxing University 199.996
3.4 RESEARCH INSTRUMENTS
This study used a questionnaire as the primary instrument to collect data. The
questionnaire consisted of seven sections. The first section consisted of 7 items about
the participants’ demographic information that included their gender, age, working
experience, educational background, designation, faculty/college/school and
90
nationality. The second section consisted of 16 items intended to measure the
mediator variable (use), on a 7-point Likert-type agreement scale with 1 being “never”
and 7 being “always”. The third, fourth and fifth sections comprised of 46 items
intended to measure the other two mediator variables (perceived ease of use and
perceived usefulness) and one exogenous variable (computer self-efficacy), on a 7-
point Likert-type agreement scale with 1 being “strongly disagree” and 7 being
“strongly agree”. The sixth section involved 10 questions about the participants’
intention to use of ICT facilities (mediator variable), on a 7-point Likert-type
agreement scale with 1 being “very unlikely” and 7 being “very likely”. The last
section consisted of 10 questions that addressed the participants’ perceptions of the
gratification (endogenous variable) of ICT usage. To measure the endogenous
(gratification), the study used a 7-point Likert scale survey questionnaire, with 1 being
“very unsatisfied” and 7 being “very satisfied”. The number of components is shown
in Table 3.2.
Table 3.2: The number of Components Measured
Section Components Measured No of Items
I Demographic Information
II Use 16
III Perceived Ease of Use 15
IV Perceived Usefulness 16
V Computer self-efficacy 15
VI Intention to Use 10
VII Gratification 10
Total 82
3.4.1 Use of ICT facilities
The items for actual use of ICT were adapted from previous studies (Ahmad et al.,
2010; Usluel et al., 2008; Islam, 2011a; Zejno & Islam, 2012) and modified
accordingly to suit the needs of the present study. Use was measured by sixteen items.
91
The items are shown in Table 3.3.
Table 3.3: Use Items
Items
How often do you use the following ICT facilities for research
and academic-related activities?
USE1 The University online research databases/Online
library research databases.
USE2 Laptop or Desktop computers.
USE3 Wireless Internet.
USE4 Web browser (www).
USE5 Search engine (e.g. Google, Yahoo, etc.).
USE6 Microsoft Office applications.
USE7 Research related software to analyze the data.
USE8 Document processing (e.g. PDF).
USE9 Computer graphics.
USE10 Email
USE11 Document editing/composing.
USE12 Computer labs.
USE13 Multimedia facilities (e.g. CD-ROM, VCD, DVD
etc.).
USE14 I use ICT for preparing the course and lecture notes.
USE15 I use ICT facilities for making presentations in the
course.
USE16 I use ICT facilities for carrying out studies in
laboratories or workshops, and making experiments.
3.4.2 Perceived Ease of Use
The Items for perceived ease of use were adapted from previous studies (Islam, 2011a
and 2011b; Islam 2014) and modified accordingly to address the needs of the present
study. Perceived ease of use was measured by fifteen items. The items are shown in
Table 3.4.
92
Table 3.4: Perceived Ease of Use Items
Items
PEU1 I find the university ICT facilities easy to use.
PEU2 I find it easy to access the university research databases.
PEU3 It is easy for me to become skilful at using university
databases for conducting research.
PEU4 Interacting with the research databases system requires
minimal effort on my part.
PEU5 I find it easy to get the research databases to help
facilitate my research.
PEU6 Interacting with the university research databases
system is very stimulating for me.
PEU7 I find it easy to select articles/journals of different
categories using research databases
(Education/Engineering/Business, etc.).
PEU8 With Wireless Internet, I find it easy to access
university databases to do research.
PEU9 Wireless Internet allows me to access research and
learning materials from Web Browser (WWW).
PEU10 Wireless Internet is easy to use for teaching and
research.
PEU11 I find it easy to use computers provided by the
university.
PEU12 My interaction with university ICT services available is
clear and understandable.
PEU13 I find it easy to download teaching, learning and
research materials using Wireless Internet in terms of
speed.
PEU14 It is easy for me to use multimedia facilities at the
university.
PEU15 I find it easy to use Microsoft office for teaching and
research purposes.
3.4.3 Perceived Usefulness
The Items for perceived usefulness were adapted from previous studies (Islam, 2011a
and 2011b; Islam 2014; Ahmad et al., 2010; Usluel et al., 2008) and modified
accordingly to suit the needs of the present study. Perceived usefulness was measured
by sixteen items. The items are shown in Table 3.5.
93
Table 3.5: Perceived Usefulness Items
Items
PU1 Using the ICT facilities at the university enables me to
accomplish tasks more quickly.
PU2 Using the ICT services available at the university
increases my research productivity.
PU3 The current university ICT system makes work more
interesting.
PU4 ICT facilities at the university improve the quality of
the work I do.
PU5 Using the university ICT facilities enhances my
research skills.
PU6 Using the university ICT facilities would make it easier
for me to find information.
PU7 ICT services at the university make information always
available to users.
PU8 Using the ICT facilities helps me write my journal
articles.
PU9 My job would be difficult to perform without the ICT
facilities.
PU10 Using the ICT facilities saves my time.
PU11 Using the university ICT facilities provides me with the
latest information on specific areas of research.
PU12 I can use ICT facilities from anywhere, anytime at the
campus.
PU13 I believe that the use of ICT in the classroom enhances
student learning in my discipline.
PU14 I believe that email and other forms of electronic
communication are important tools in faculty/student
communication.
PU15 I believe that web-based instructional materials enhance
student learning.
PU16 Using the ICT services enables me to download
teaching, learning and research materials from the
internet.
3.4.4 Computer Self-Efficacy
The Items for computer self-efficacy were adapted from previous studies (Islam,
2011a and 2011b; Islam 2014; Ahmad et al., 2010) and modified accordingly to suit
the needs of the present study. Computer self-efficacy was measured by fifteen items.
94
The items are shown in Table 3.6.
Table 3.6: Computer Self-Efficacy Items
Items
CSE1 I have the skills required to use computer applications
for writing my research papers.
CSE2 I have the skills and knowledge required to use
computer applications for demonstrating specific
concepts in class.
CSE3 I have the skills required to use computer applications
for presenting lectures.
CSE4 I have the skills required to communicate
electronically with my colleagues and students.
CSE5 I have the ability to e-mail, chat, download teaching,
learning and research materials, and search different
websites.
CSE6 I have the ability to use the Wireless Internet service
provided by the university.
CSE7 I have the skills required to use ICT facilities to
enhance the effectiveness of my teaching, learning
and research.
CSE8 I feel capable of using the university research
databases for writing journal papers.
CSE9 I have the ability to navigate my way through the ICT
facilities.
CSE10 I have the skills required to use the ICT facilities to
enhance the quality of my research works.
CSE11 I have the ability to save and print journals/articles
from the research databases.
CSE12 I have the knowledge and skills required to benefit
from using the university ICT facilities.
CSE13 I am capable of accessing the research databases from
the university website.
CSE14 I am capable to download and install research related
software using ICT facilities at the university.
CSE15 I am capable of using multimedia facilities for
delivering lecturers.
3.4.5 Intention to Use
The Items for intention to use were adapted from previous studies (Islam, 2011a;
Ahmad et al., 2010) and modified accordingly to suit the needs of the present study.
Intention to use was measured by ten items. The items are indicated in Table 3.7.
95
Table 3.7: Intention to Use Items
Items
INT1 I will use the ICT facilities to carry out my research
activities.
INT2 I will use computer applications to present lecture
materials in class.
INT3 I will use Microsoft office applications to write my
research papers.
INT4 I will teach a course that will be delivered totally on
the world wide web.
INT5 I intend to use ICT facilities for teaching, learning
and research purposes frequently.
INT6 I intend to download research materials from the
university databases frequently.
INT7 I will use the university research databases to update
myself on the research areas I am pursuing.
INT8 I intend to use the ICT facilities to upgrade my
research performance.
INT9 I will use the university research databases to write
outstanding journal articles.
INT10 I intend to use research related software to analyze
the data.
3.4.6 Gratification
The Items for gratification were adapted from previous studies (Islam, 2011a & 2011b;
Islam 2014) and modified accordingly to suit the needs of the present study.
Gratification was measured by ten items. The items are shown in Table 3.8.
96
Table 3.8: Gratification Items
Items
GRAT1 Overall I am satisfied with the ease of completing my
task using ICT facilities.
GRAT2 I am satisfied with the ICT facilities provided by the
university.
GRAT3 I am satisfied with the ease of use of the ICT
facilities.
GRAT4 The university ICT facilities have greatly affected the
way I search for information and conduct my
research.
GRAT5 Providing the ICT is an indispensable services
provided by the university.
GRAT6 Overall I am satisfied with the amount of time it takes
to complete my task.
GRAT7 I am satisfied with the structure of accessible
information (available as categories of research
domain, or by date of issue—of journals in particular,
or as full-texts or abstracts of theses and
dissertations) of the university research databases.
GRAT8 I am satisfied with the support information provided
by the university’s ICT facilities.
GRAT9 I am satisfied in using ICT facilities for teaching,
learning and research.
GRAT10 Overall I am satisfied with the Wireless Internet
service provided at the university.
3.4.7 Pilot Test
The validity of an instrument is concerned with the soundness and the effectiveness of
the measurement. According to Babbie (2001), validity explains a measure that
precisely reflects the concept which is intended to be measured. The questionnaire
items used in this study were adopted and modified from Islam (2011a and 2011b),
Islam (2014), Ahmad et al., (2010), Usluel et al., (2008) and Zejno and Islam (2012).
Their instruments were validated using well established statistical tools. Specially,
Islam (2011a and 2011b; 2014) developed and validated previous instruments by
applying a Rasch Model and most of the items in the present study were adopted and
97
modified from these instrument. Moreover, the Technology Acceptance Model (Davis,
1989), Technology Satisfaction Model (Islam, 2014) and Online Database Adoption
and Satisfaction (ODAS) model were applied as a theoretical framework for this study.
According to Mathieson, Peacock, and Chin (2001), the TAM has been examined and
broadly accepted among researchers as a theoretically-based model with good
predictive validity and confirmed reliability in the information technology field. The
instrument of Technology Adoption and Gratification (TAG) model was translated
from English to Chinese in order to collect data from Jiaxing University in China. The
questionnaire was translated by a young researcher Qian Xiuxiu from Jiaxing
University who was my former research partner. The translated questionnaire was
verified by Professor Dr. Zhang Quan and Professor Tao Danyu from the college of
foreign language studies, Jiaxing University.
To establish the validity and reliability of the questionnaire items, a Rasch
analysis was used. Using Winsteps version 3.49, a pilot test was conducted with 82
items. A total of 60 lecturers were equally collected from the University of Malaya
and Jiaxing University to conduct pilot tests. The findings of the Rasch model showed
that the summary statistics of the 82 items obtained from the Rasch analysis,
indicating the items measured reliability (r = .90), and persons measured reliability (r
= .95). In addition, items separation (3.05) and persons separation (4.35) are
statistically significant as shown in Figure 3.15.
98
TABLE 3.1 TAG MODEL ZOU999ws.txt Nov 16 21:43 2014
INPUT: 60 PERSONS, 82 ITEMS MEASURED: 60 PERSONS, 82 ITEMS, 7 CATS 3.49
--------------------------------------------------------------------------------
SUMMARY OF 60 MEASURED PERSONS
+-----------------------------------------------------------------------------+
| RAW MODEL INFIT OUTFIT |
| SCORE COUNT MEASURE ERROR MNSQ ZSTD MNSQ ZSTD |
|-----------------------------------------------------------------------------|
| MEAN 459.8 82.0 .97 .11 1.17 .3 1.12 .1 |
| S.D. 47.3 .0 .65 .04 .70 2.9 .64 2.9 |
| MAX. 563.0 82.0 3.38 .30 4.80 7.7 4.38 6.7 |
| MIN. 368.0 82.0 .09 .08 .27 -6.4 .24 -6.8 |
|-----------------------------------------------------------------------------|
| REAL RMSE .15 ADJ.SD .64 SEPARATION 4.35 PERSON RELIABILITY .95 |
|MODEL RMSE .12 ADJ.SD .64 SEPARATION 5.37 PERSON RELIABILITY .97 |
| S.E. OF PERSON MEAN = .09 |
+-----------------------------------------------------------------------------+
PERSON RAW SCORE-TO-MEASURE CORRELATION = .95
CRONBACH ALPHA (KR-20) PERSON RAW SCORE RELIABILITY = .96
SUMMARY OF 82 MEASURED ITEMS
+-----------------------------------------------------------------------------+
| RAW MODEL INFIT OUTFIT |
| SCORE COUNT MEASURE ERROR MNSQ ZSTD MNSQ ZSTD |
|-----------------------------------------------------------------------------|
| MEAN 336.4 60.0 .00 .13 1.02 -.1 1.12 .1 |
| S.D. 32.1 .0 .45 .02 .53 2.3 .78 2.7 |
| MAX. 390.0 60.0 1.44 .19 2.72 6.7 5.06 9.7 |
| MIN. 195.0 60.0 -1.11 .09 .40 -3.5 .42 -3.4 |
|-----------------------------------------------------------------------------|
| REAL RMSE .14 ADJ.SD .43 SEPARATION 3.05 ITEM RELIABILITY .90 |
|MODEL RMSE .13 ADJ.SD .43 SEPARATION 3.38 ITEM RELIABILITY .92 |
| S.E. OF ITEM MEAN = .05 |
+-----------------------------------------------------------------------------+
UMEAN=.000 USCALE=1.000
ITEM RAW SCORE-TO-MEASURE CORRELATION = -.98
Figure 3.15: Summary Statistics
The item polarity map exhibits most of the items measured in the same
direction (ptmea. Corr. > 0.30), which are greater than the recommended value (.30).
However, few items discovered low point measure correlations due to the small
sample size as shown in Figure 3.16.
99
Figure 3.16: Item Polarity Map
100
The item map exhibits that the map of the valid items is located in their
calibrations on a single continuum. As shown on the right hand side of Figure 3.17,
the items on this single continuum are structured from the “less difficult to be rated as
technology adoption and gratification” to the “more difficult to be rated as technology
adoption and gratification” The left hand side of this figure also matches the level of a
person’s ability in a single line, whereby the respondents are ordered from the “higher
level of agreement to endorse lecturers’ adoption and gratification with the ICT
usage” to the “lower level agreement to endorse lecturers’ adoption and gratification
with the ICT usage”. Therefore, this study demonstrated that the most difficult item
was use12 and the least difficult use2, although the majority of the lecturers elected to
use the ICT facilities in higher education as shown in Figure 3.17.
101
TABLE 12.2 TAG MODEL ZOU999ws.txt Nov 16 21:43 2014
INPUT: 60 PERSONS, 82 ITEMS MEASURED: 60 PERSONS, 82 ITEMS, 7 CATS 3.49
--------------------------------------------------------------------------------
PERSONS MAP OF ITEMS
<more>|<rare>
4 +
|
|
|
|
X |
|
|
3 +
X |
|
|
X |
|
T|
X |
2 +
X |
XX |
XXX S|
XX | 12use12
XXX |
XXXX |
|
1 XXXXXX M+
XXXXXXXX |T 9use9
XXX | 1use1
XXXXXXX | 13use13 16use16 22peu6 24peu8 29peu13 43pu12
66int4 7use7 82grat10
XXXXX |S 20peu4
XXXX S| 21peu5 25peu9 26peu10 28peu12 34pu3 36pu5
XXXXXX | 18peu2 19peu3 23peu7 27peu11 30peu14 38pu7
74grat2 75grat3 76grat4 78grat6 79grat7 80grat8
XX | 17peu1 33pu2 35pu4 73grat1 81grat9
0 +M 32pu1 37pu6 39pu8 44pu13 46pu15 8use8
| 42pu11 45pu14 47pu16 55cse8 56cse9 61cse14
71int9 72int10 77grat5
| 15use15 41pu10 48cse1 49cse2 51cse4 57cse10
60cse13 67int5 68int6 69int7 70int8
T| 31peu15 3use3 40pu9 50cse3 54cse7 59cse12
65int3
|S 11use11 14use14 53cse6 58cse11 63int1 64int2
| 4use4 52cse5 62cse15 6use6
|
|T 10use10
-1 + 5use5
| 2use2
|
|
|
|
|
|
-2 +
<less>|<frequ>
Figure 3.17: Item Map
The item fit order indicated that most of the items showed good item fit and
were constructed on a continuum of increasing intensity. However, seven items (int3,
use16, use4, use12, int4, use13, use3) showing a misfit to the Rasch model had to be
102
slightly revised in order to validate the final questionnaire as shown in Figure 3.18.
Statistically, INFIT and OUTFIT Mean square (MNSQ) statistics for items should be
greater than 0.5 and less than 1.5 for rating scale application. Similarly, mean square
(MNSQ) statistics for Pearson measures should also be less than 1.8.
Figure 3.18: Item Fit Order
103
This study discovered that the variance explained by the measures was 48%
which indicated that the items were able to endorse the lecturers’ adoption and
gratification in using ICT facilities as shown in Figure 3.19.
TABLE 23.2 TAG MODEL ZOU999ws.txt Nov 16 21:43 2014
INPUT: 60 PERSONS, 82 ITEMS MEASURED: 60 PERSONS, 82 ITEMS, 7 CATS 3.49
--------------------------------------------------------------------------------
PRINCIPAL COMPONENTS (STANDARDIZED RESIDUAL) FACTOR PLOT
Factor 1 extracts 9.8 units out of 82 units of ITEM residual variance noise.
Yardstick (variance explained by measures)-to-This Factor ratio: 7.7:1
Yardstick-to-Total Noise ratio (total variance of residuals): .9:1
Table of STANDARDIZED RESIDUAL variance (in Eigenvalue units)
Empirical Modeled
Total variance in observations = 157.6 100.0% 100.0%
Variance explained by measures = 75.6 48.0% 50.6%
Unexplained variance (total) = 82.0 52.0% 49.4%
Unexpl var explained by 1st factor = 9.8 6.2%
-2 -1 0 1 2
++---------------+---------------+---------------+---------------++
.7 + A | +
| | |
.6 + FEC BD | +
| I GHJ | |
.5 + L MK | +
| N O | |
F .4 + Q SRPT +
A | V |UW |
C .3 + Y X Z | +
T | 1 | |
O .2 + 1 1 | +
R | 1 1 3 1 3 21 |
.1 + 1 1| +
1 | |1 1 |
.0 +---------------------1----------|1--1----------------------------+
L | 1 1 1 | 1 1 |
O -.1 + | 2 +
A | | 1z |
D -.2 + y | p x +
I | r | uvs twq |
N -.3 + | mn o +
G | | k j l |
-.4 + | g i hf +
| | d e |
-.5 + | +
| | cb |
-.6 + | a +
| | |
++---------------+---------------+---------------+---------------++
-2 -1 0 1 2
ITEM MEASURE
Figure 3.19: Principal Component
104
3.4.8 Reliability and Validity of Instrument
The questionnaire’s reliability and validity were established through a RASCH Model
applying Winsteps version 3.49. A total of 396 lecturers were collected from the
different faculties of the University of Malaya and Jiaxing University for the final
validation. The findings of the Rasch analyses discovered in the summary statistics
were that: (i) the items reliability was found to be 0.98 (SD=195.5), while the persons
reliability was 0.96 (SD=57.8); (ii) the items and persons separation were 7.99 and
5.00, respectively as shown in Figure 3.20.
TABLE 3.1 TAG MODEL ZOU999ws.txt Nov 16 23:14 2014
INPUT: 396 PERSONS, 82 ITEMS MEASURED: 396 PERSONS, 82 ITEMS, 7 CATS 3.49
--------------------------------------------------------------------------------
SUMMARY OF 396 MEASURED PERSONS
+-----------------------------------------------------------------------------+
| RAW MODEL INFIT OUTFIT |
| SCORE COUNT MEASURE ERROR MNSQ ZSTD MNSQ ZSTD |
|-----------------------------------------------------------------------------|
| MEAN 455.8 82.0 1.02 .12 1.14 .1 1.10 .0 |
| S.D. 57.8 .1 .78 .04 .68 3.1 .64 3.0 |
| MAX. 568.0 82.0 4.14 .41 4.99 9.3 5.52 9.3 |
| MIN. 238.0 81.0 -.70 .08 .17 -7.3 .17 -7.5 |
|-----------------------------------------------------------------------------|
| REAL RMSE .15 ADJ.SD .76 SEPARATION 5.00 PERSON RELIABILITY .96 |
|MODEL RMSE .13 ADJ.SD .77 SEPARATION 6.12 PERSON RELIABILITY .97 |
| S.E. OF PERSON MEAN = .04 |
+-----------------------------------------------------------------------------+
VALID RESPONSES: 99.9%
SUMMARY OF 82 MEASURED ITEMS
+-----------------------------------------------------------------------------+
| RAW MODEL INFIT OUTFIT |
| SCORE COUNT MEASURE ERROR MNSQ ZSTD MNSQ ZSTD |
|-----------------------------------------------------------------------------|
| MEAN 2201.2 396.0 .00 .05 1.03 -.3 1.10 .0 |
| S.D. 195.5 .1 .44 .01 .44 4.6 .56 4.9 |
| MAX. 2533.0 396.0 1.39 .07 2.53 9.9 3.06 9.9 |
| MIN. 1372.0 395.0 -1.00 .04 .56 -6.4 .57 -5.9 |
|-----------------------------------------------------------------------------|
| REAL RMSE .05 ADJ.SD .43 SEPARATION 7.99 ITEM RELIABILITY .98 |
|MODEL RMSE .05 ADJ.SD .43 SEPARATION 8.71 ITEM RELIABILITY .99 |
| S.E. OF ITEM MEAN = .05 |
+-----------------------------------------------------------------------------+
UMEAN=.000 USCALE=1.000
Figure 3.20: Summary of Statistics
The item polarity map exhibits the majority of the items measured in the same
direction (ptmea. Corr. > 0.31), which are greater than the recommended value (.30).
However, only three items (use13, use, 12 and use3) were slightly smaller than the
recommended value as shown in Figure 3.21.
105
Figure 3.21: Item Polarity Map
106
According to Bond and Fox (2001), the Rasch model is based on the idea that
all persons are more likely to respond correctly to easy items than difficult ones, and
all items are more likely to be answered by persons who have a higher ability than a
lower one. Figure 3.22 reveals that the map of the valid items is located in their
calibrations on a single continuum. As shown on the right hand side of Figure 3.22,
the items on this single continuum are structured from the “less difficult to be rated as
technology adoption and gratification” to the “more difficult to be rated as technology
adoption and gratification” The left hand side of this figure also matches the level of a
person’s ability in a single line, whereby the respondents are ordered from the “higher
level of agreement to endorse lecturers’ adoption and gratification with the ICT
facilities” to the “lower level agreement to endorse lecturers’ adoption and
gratification with the ICT facilities”. Thus, this study discovered that the most difficult
item was use12 and the least difficult were use2 and use5. Majority of the lecturers
elected to use the ICT facilities in higher education as shown in Figure 3.22.
107
TABLE 12.2 TAG MODEL ZOU999ws.txt Nov 16 23:14 2014
INPUT: 396 PERSONS, 82 ITEMS MEASURED: 396 PERSONS, 82 ITEMS, 7 CATS 3.49
--------------------------------------------------------------------------------
PERSONS MAP OF ITEMS
<more>|<rare>
5 +
|
|
|
|
|
. |
4 . +
|
. |
. |
. |
. |
|
3 . +
# |
|
# T|
.# |
## |
.# |
2 .## +
.### S|
.## |
.### |
.#### | 12use12
.##### |
.###### |
1 ######## M+
######## |T 9use9
.############ | 1use1
######## | 13use13 20peu4 22peu6 66int4 7use7
####### |S 16use16 19peu3 24peu8 29peu13 43pu12
79grat7 82grat10
.####### S| 18peu2 21peu5 23peu7 25peu9 26peu10
28peu12 30peu14 34pu3 36pu5 38pu7
74grat2 75grat3 78grat6 80grat8
#### | 17peu1 27peu11 35pu4 76grat4 81grat9
8use8
0 #### +M 32pu1 33pu2 37pu6 42pu11 44pu13
55cse8 61cse14 73grat1
.# | 11use11 31peu15 39pu8 46pu15 48cse1
49cse2 56cse9 57cse10 60cse13 68int6
71int9 72int10 77grat5
. | 40pu9 41pu10 47pu16 50cse3 54cse7
59cse12 62cse15 67int5 69int7 70int8
. |S 14use14 15use15 3use3 45pu14 51cse4
53cse6 58cse11 63int1
. T| 52cse5
. | 4use4 65int3
|T 10use10 64int2 6use6
-1 + 2use2 5use5
<less>|<frequ>
EACH '#' IS 4.
Figure 3.22: Item Map
Item fit order demonstrated that most items showed good item fit and were
constructed on a continuum of increasing intensity. However, eight items (use12,
108
use13, use16, use9, use3, use7, use1 and use14) showing a misfit to the Rasch model
had to be removed in order to validate the final questionnaire as shown in Figure 3.23.
Figure 3.23: Item Fit Order
109
Table 3.9: The Valid Items Measuring Perceived Ease of Use (PEU), Perceived
Usefulness (PU), Computer Self-Efficacy (CSE), Intention to Use (INT), Actual Use
(USE) and Gratification (GRAT)
Constructs The List of the Valid Items
PEU
PEU1 I find the university ICT facilities easy to use.
PEU2 I find it easy to access the university research databases.
PEU3 It is easy for me to become skilful at using university
databases for conducting research.
PEU4 Interacting with the research databases system requires
minimal effort on my part.
PEU5 I find it easy to get the research databases to help facilitate
my research.
PEU6 Interacting with the university research databases system is
very stimulating for me.
PEU7 I find it easy to select articles/journals of different categories
using research databases (Education/Engineering/Business,
etc.).
PEU8 With Wireless Internet, I find it easy to access university
databases to do research.
PEU9 Wireless Internet allows me to access research and learning
materials from Web Browser (WWW).
PEU10 Wireless Internet is easy to use for teaching and research.
PEU11 I find it easy to use computers provided by the university.
PEU12 My interaction with university ICT services available is clear
and understandable.
PEU13 I find it easy to download teaching, learning and research
materials using Wireless Internet in terms of speed.
PEU14 It is easy for me to use multimedia facilities at the university.
PEU15 I find it easy to use Microsoft office for teaching and research
purposes.
PU
PU1 Using the ICT facilities at the university enables me to
accomplish tasks more quickly.
PU2 Using the ICT services available at the university increases
my research productivity.
PU3 The current university ICT system makes work more
interesting.
PU4 ICT facilities at the university improve the quality of the
work I do.
PU5 Using the university ICT facilities enhances my research
skills.
PU6 Using the university ICT facilities would make it easier for
me to find information.
PU7 ICT services at the university make information always
available to users.
PU8 Using the ICT facilities helps me write my journal articles. PU9 My job would be difficult to perform without the ICT
facilities.
110
Table 3.9, continued
PU
PU10 Using the ICT facilities saves my time.
PU11 Using the university ICT facilities provides me with the latest
information on specific areas of research.
PU12 I can use ICT facilities from anywhere, anytime at the
campus.
PU13 I believe that the use of ICT in the classroom enhances
student learning in my discipline.
PU14 I believe that email and other forms of electronic
communication are important tools in faculty/student
communication.
PU15 I believe that web-based instructional materials enhance
student learning.
PU16 Using the ICT services enables me to download teaching,
learning and research materials from the internet.
CSE
CSE1 I have the skills required to use computer applications for
writing my research papers.
CSE2 I have the skills and knowledge required to use computer
applications for demonstrating specific concepts in class.
CSE3 I have the skills required to use computer applications for
presenting lectures.
CSE4 I have the skills required to communicate electronically with
my colleagues and students.
CSE5 I have the ability to e-mail, chat, download teaching, learning
and research materials, and search different websites.
CSE6 I have the ability to use the Wireless Internet service provided
by the university.
CSE7 I have the skills required to use ICT facilities to enhance the
effectiveness of my teaching, learning and research.
CSE8 I feel capable of using the university research databases for
writing journal papers.
CSE9 I have the ability to navigate my way through the ICT
facilities.
CSE10 I have the skills required to use the ICT facilities to enhance
the quality of my research works.
CSE11 I have the ability to save and print journals/articles from the
research databases.
CSE12 I have the knowledge and skills required to benefit from
using the university ICT facilities.
CSE13 I am capable of accessing the research databases from the
university website.
CSE14 I am capable to download and install research related
software using ICT facilities at the university.
CSE15 I am capable of using multimedia facilities for delivering lecturers.
111
Table 3.9, continued
INT
INT1 I will use the ICT facilities to carry out my research activities.
INT2 I will use computer applications to present lecture materials
in class.
INT3 I will use Microsoft office applications to write my research
papers.
INT4 I will teach a course that will be delivered totally on the
world wide web.
INT5 I intend to use ICT facilities for teaching, learning and
research purposes frequently.
INT6 I intend to download research materials from the university
databases frequently.
INT7 I will use the university research databases to update myself
on the research areas I am pursuing.
INT8 I intend to use the ICT facilities to upgrade my research
performance.
INT9 I will use the university research databases to write
outstanding journal articles.
INT10 I intend to use research related software to analyze the data.
USE
USE2 How often do you use the following ICT facilities for
research and academic-related activities? - Laptop or Desktop
computers
USE4 How often do you use the following ICT facilities for
research and academic-related activities? - Web browser
(www)
USE5 How often do you use the following ICT facilities for
research and academic-related activities? - Search engine
(e.g. Google, Yahoo, etc.).
USE6 How often do you use the following ICT facilities for
research and academic-related activities? - Microsoft Office
applications.
USE8 How often do you use the following ICT facilities for
research and academic-related activities? - Document
processing (e.g. PDF).
USE10 How often do you use the following ICT facilities for
research and academic-related activities? - Email.
USE11 How often do you use the following ICT facilities for
research and academic-related activities? - Document
editing/composing.
USE15 I use ICT facilities for making presentations in the course GRAT1 Overall I am satisfied with the ease of completing my task
using ICT facilities. GRAT2 I am satisfied with the ICT facilities provided by the
university.
112
Table 3.9, continued
GRAT3 I am satisfied with the ease of use of the ICT facilities. GRAT4 The university ICT facilities have greatly affected the way I
search for information and conduct my research.
GRAT5 Providing the ICT is an indispensable services provided by
the university. GRAT6 Overall I am satisfied with the amount of time it takes to
complete my task. GRAT7 I am satisfied with the structure of accessible information
(available as categories of research domain, or by date of
issue—of journals in particular, or as full-texts or abstracts of
theses and dissertations) of the university research databases. GRAT8 I am satisfied with the support information provided by the
university’s ICT facilities. GRAT9 I am satisfied in using ICT facilities for teaching, learning
and research. GRAT10 Overall I am satisfied with the Wireless Internet service
provided at the university.
This study revealed that the variance explained by the measures was 55.3%
which demonstrated that the items were able to endorse the University of Malaya and
Jiaxing University lecturers’ adoption of and gratification in using ICT facilities in
higher education for their teaching and research purposes as shown in Figure 3.24.
113
TABLE 23.2 TAG MODEL ZOU999ws.txt Nov 16 23:14 2014
INPUT: 396 PERSONS, 82 ITEMS MEASURED: 396 PERSONS, 82 ITEMS, 7 CATS 3.49
--------------------------------------------------------------------------------
PRINCIPAL COMPONENTS (STANDARDIZED RESIDUAL) FACTOR PLOT
Factor 1 extracts 9.5 units out of 82 units of ITEM residual variance noise.
Yardstick (variance explained by measures)-to-This Factor ratio: 10.7:1
Yardstick-to-Total Noise ratio (total variance of residuals): 1.2:1
Table of STANDARDIZED RESIDUAL variance (in Eigenvalue units)
Empirical Modeled
Total variance in observations = 183.4 100.0% 100.0%
Variance explained by measures = 101.4 55.3% 57.3%
Unexplained variance (total) = 82.0 44.7% 42.7%
Unexpl var explained by 1st factor = 9.5 5.2%
-2 -1 0 1 2
++---------------+---------------+---------------+---------------++
.7 + | +
| A D BC| |
.6 + E GF | +
| JI H | |
F .5 + K | +
A | L| |
C .4 + M | +
T | P O QN | |
O .3 + RT S | +
R | V ZY XW U |
.2 + 1 1 11 2 | +
1 | 1 21 |1 |
.1 + | +
L | 1 1111 1 |
O .0 +-----------------------------1--|---------11---------------------+
A | | 1 1 |
D -.1 + 11 +
I | 1| 1 1 |
N -.2 + |z 1 1 1 +
G | | xy w |
-.3 + |suqprvt +
| | m oln |
-.4 + | gkji h +
| | de f |
-.5 + | acb +
++---------------+---------------+---------------+---------------++
-2 -1 0 1 2
ITEM MEASURE
Figure 3.24: Principal Components
3.5 DATA ANALYSIS PROCEDURES
The data analysis procedures involved in the study included the use of descriptive
statistics using SPSS version 22. The specific purpose of RASCH analysis was to
develop and validate the psychometric properties of TAG model. As such, the
questionnaire’s reliability and validity were established through a RASCH analysis
applying Winsteps version 3.49. This present study also applied the three-stage of the
structural equation modeling (SEM), namely, confirmatory factor analysis (CFA),
114
full-fledged structural model and invariance analysis (metric and configural invariance
analyses) to develop and validate the Technology Adoption and Gratification (TAG)
model in a cross-culture that addressed the objectives of the study. In order to analyse
the causal relationships among the various constructs, AMOS software version 18.0
was applied. Furthermore, the structural aspects of the validity are specially addressed
in the validation of the TAG model. Finally, a Sobel test was also conducted to
examine the indirect relationships between the constructs of the TAG model.
3.6 STRUCTURAL EQUATION MODELING (SEM)
Structural Equation Modeling (SEM) is one of the statistical models that seek to
validate, extend and propose new theories. Primarily, this study used the Structural
Equation Modeling (SEM) techniques to validate the measurement models and
estimate the fit of the hypothesized and revised models to the survey data. Specifically,
the three-step Structural Equation Modeling, involving Confirmatory Factor Analysis
(CFA), a Full-fledged Structural model and invariance analysis, were performed to
develop and validate the Technology Adoption and Gratification (TAG) model, and to
explain the cross-cultural validation in higher education. The hypothesized Structural
Equation Model (SEM) for the study is shown in Figure 3.25.
Figure 3.25: The Hypothesized Technology Adoption and Gratification (TAG) Model
115
3.7 MODEL EVALUATION
In order to analyze the structural relationships among the various constructs, namely,
computer self-efficacy, perceived ease of use, perceived usefulness, intention to use,
use and gratification, the analysis used the greatest probability estimation in
generating estimates of the six Confirmatory Factor Analysis (CFA) models and Full-
fledged structural equation model. In other words, the models were estimated by
applying a set of measures to assess the goodness-of-fit of each model. The goodness-
of-fit measures, guided by the universally accepted criteria for deciding good fit,
assessed the (i) reliability of the hypothesized model with the observed data, (ii)
justifications of variability, (iii) fairness of the estimates, and (iv) simplicity of the
measured models. The adequacy of the model was determined using five measures of
good fit. The first Chi-square (χ2) measured the compatibility of the data with the
hypothesized model. The chi-square value divided (χ2/df) by the degree of freedom
value should be less than or equal to 3.0. Second, the Root Mean Square Error of
Approximation (RMSEA) measures the intimacy of model fit, and should be less
than .08. Third, the Comparative Fit Index (CFI) evaluates the fit of the existing
model with a null model that predicts the latent variables to be uncorrelated in the
model. The value of the CFI should be greater than or equal to 0.90. Fourth, the
Goodness of Fit Index (GFI), which deals with error in imitating the variance-
covariance matrix, must be greater than .90 to indicate good fit. Finally, the Tucker-
Lewis Index (TLI) measures the estimated model against the null-model, and should
display a value of .90 or more to show good fit of the model. The summary of the
goodness-of-fit indices is shown in Table 310.
116
Table 3.10: The Recommended Goodness-of-Fit Indices
Fit Indices Recommended
Chi-square (χ2) Non-significant
p value ≥ .05
RMSEA ≤ .08
CFI ≥ .90
TLI ≥ .90
117
CHAPTER 4
RESULTS
4.1 INTRODUCTION
This chapter presents the findings of this study. The data were examined to discover
the causal relationships of the Technology Adoption and Gratification (TAG) model
between the University of Malaya (UM) and Jiaxing University (JU) lecturers’
adoption and gratification in using ICT facilities and the constructs of: computer self-
efficacy, perceived ease of use, perceived usefulness, intention to use, actual use and
gratification. Thus, A 3-step Structural Equation Modeling (SEM) approach was
adopted. Section 1 demonstrates the results of validating the Confirmatory Factor
Analysis (CFA) for 6 measurement models, namely, computer self-efficacy, perceived
ease of use, perceived usefulness, intention to use, actual use and gratification.
Section 2 displays the estimation of the full-fledged structural model of TAG. Section
3 shows the cross-cultural validation of the TAG model using invariance analyses.
Moreover, the measurement models were evaluated, followed by an explanation of the
full-fledged Structural Equation Modeling (SEM) as well as cross-cultural-invariant.
SECTION 1: VALIDATING THE CONFIRMATORY FACTOR ANALYSIS
MODELS
4.2 ONE-FACTOR MEASUREMENT MODEL OF PERCEIVED EASE OF
USE (PEU)
The specific objective was to measure the lecturers’ views on the ease of use of ICT
facilities for their teaching and research purposes in higher education. To address the
objective 2, the present study conducted a Confirmatory Factor Analysis (CFA) to
118
confirm lecturers’ perceptions on the ease of use and to test the impact of perceived
ease of use in the hypothesized TAG model; hypotheses 2 and 9 were evaluated. The
preliminary CFA model was performed on 15 items that explain the factor of
perceived ease of use. The majority of the standardized factor loadings were greater
than .70, with only few items demonstrating loadings ranging from .57 to .69.
Nevertheless, the findings of one-factor measurement model of perceived ease of use
discovered that the overall fit indices did not fulfill the statistical requirements as
indicated by the following fit indices: χ2 (df = 90) = 1152.997; p = 0.000; RMSEA =
0.173; CFI = 0.723; TLI = 0.677; SRMR = 0.185 (see Appendix C). As a result, the
model required revision due to the violations and estimation. After deleting these
items (PEU1 PEU3, PEU5, PEU8, PEU10, PEU11, PEU12, PEU13 and PEU14), the
analysis demonstrated an exceptional goodness-of-fit of the model. The items were
removed one at a time according to their modification indices, with the largest being
removed first, which showed the presence of multiple collinearity as shown in Table
4.1.
Table 4.1: The List of Removed Items
Item
Deleted
Chi-square
(χ2)
df p value RMSEA CFI TLI SRMR
PEU1 1060.716 77 .000 .180 .724 .674 .189
PEU3 932.028 65 .000 .184 .729 .675 .181
PEU5 746.723 54 .000 .180 .749 .694 .175
PEU8 513.111 44 .000 .164 .791 .739 .159
PEU10 267.821 35 .000 .130 .872 .835 .119
PEU11 216.366 27 .000 .133 .882 .842 .119
PEU12 204.390 20 .000 .153 .862 .806 .132
PEU13 66.497 14 .000 .097 .950 .924 .071
PEU14 19.151 9 .024 .053 .987 .979 .043
The following 6 items remained in the original pool of 15 items: PEU2, PEU4,
PEU6, PEU7, PEU9 and PEU15. These items were used as the indicators for the one-
119
factor measurement model of perceived ease of use. The Figure 4.1 shows the error
variances in the model represented by e2, e4, e6, e7, e9 and e15.
Figure 4.1: One-Factor Measurement Model for Perceived Ease of Use
The results of the model’s overall goodness-of-fit statistics showed an
exceptional fit; the model chi-square, χ2 (df=9) =19.151, p=.024 and the root mean
square error of approximation (RMSEA) represented a goodness-of-fit of the model
with a satisfactory value of .053; the value of the comparative fit index (CFI) was
.987; the Tucker-Lewis index (TLI) value was .979 and the value of the standardized
root mean square residual (SRMR) was .043. The CFA model fit indices were
supported by prior studies (Hu & Bentler, 1999; Marsh, Hau & Wen, 2004). The
results in Figure 4.1 indicate that the parameters were free from offending estimates.
The standardized factor demonstrated loadings ranging from .55 to .78, representing
statistically significant indicators. The item loadings are shown in Table 4.2.
120
Table 4.2: The Loadings for Perceived Ease of Use
Item
No Item label Loadings M SD α
PEU2 I find it easy to access the university
research databases.
.72 5.300 1.313
.833
PEU4 Interacting with the research
databases system requires minimal
effort on my part.
.72 4.931 1.366
PEU6 Interacting with the university
research databases system is very
stimulating for me.
.78 4.846 1.388
PEU7 I find it easy to select articles/journals
of different categories using research
databases
(Education/Engineering/Business,
etc.).
.75 5.346 1.246
PEU9 Wireless Internet allows me to access
research and learning materials from
Web Browser (WWW).
.55 5.287 1.486
PEU15 I find it easy to use Microsoft office
for teaching and research purposes.
.55 5.828 1.123
The findings of the one-factor measurement model of perceived ease of use
also revealed that the lecturers’ views on the ease of use of ICT facilities were
indicated by six valid predictors. They explained approximately 46.7% of the
variability of lecturers’ perceived ease of use of ICT facilities for their teaching and
research purposes in higher education as represented in Table 4.3.
Table 4.3: The Squared Multiple Correlations
Items Measured
PEU2 .524
PEU4 .515
PEU6 .609
PEU7 .558
PEU9 .301
PEU15 .299
The average total variance 0.467
121
4.3 ONE-FACTOR MEASUREMENT MODEL OF PERCEIVED
USEFULNESS (PU)
The specific objective was to evaluate the lecturers’ perception regarding the
usefulness of ICT facilities in higher education. To address the objective 3, this study
performed a Confirmatory Factor Analysis (CFA) to confirm lecturers’ perceptions on
the usefulness of ICT facilities for their teaching and research purposes and to
examine the influence of perceived usefulness in the hypothesized TAG model,
hypotheses 3 and 9 were tested. The initial CFA model was estimated with 16
indicators of perceived usefulness. All items exhibited a loading greater than .70 on
their respective factor. However, the results of the one-factor measurement model of
perceived usefulness exhibited that the overall fit indices were not adequate as
indicated by the following fit indices: χ2 (df = 104) = 1217.231; p = 0.000; RMSEA =
0.165; CFI = 0.783; TLI = 0.750; SRMR = 0.137. Thus, 9 items were dropped due to
violations and estimation (see Appendix B). After deleting these items (PU1, PU8,
PU9, PU10, PU12, PU13, PU14, PU15 and PU16), the overall fit indices showed an
adequate fit. The items were removed one at a time according to their modification
indices, with the largest being removed first, which showed the presence of multiple
collinearity as indicated in Table 4.4.
Table 4.4: The List of Removed Items
Item
Deleted
Chi-square
(χ2)
df p value RMSEA CFI TLI SRMR
PU1 1090.810 90 .000 .168 .784 .748 .141
PU8 1005.069 77 .000 .175 .784 .745 .147
PU9 783.672 65 .000 .167 .819 .783 .137
PU10 653.472 54 .000 .168 .837 .801 .136
PU12 573.215 44 .000 .174 .848 .810 .136
PU13 444.767 35 .000 .172 .870 .833 .124
PU14 216.395 27 .000 .133 .933 .910 .084
PU15 91.114 20 .000 .095 .973 .962 .049
PU16 44.929 14 .000 .075 .987 .981 .032
122
Eventually, only seven items were selected for this model; they were: PU2,
PU3, PU4, PU5, PU6, PU7 and PU11. These items were used as the indicators for the
one-factor measurement model of perceived usefulness with e2, e3, e4, e5, e6, e7 and
e11 representing error variances. This is shown in Figure 4.2.
Figure 4.2: One-Factor Measurement Model for Perceived Usefulness
The model’s overall goodness-of-fit statistics showed a satisfactory fit; the
model chi-square, χ2 (df=14) =44.929, p=.000 and the root mean square error of
approximation (RMSEA) signified an adequacy of the model with a value of .075; the
value of the comparative fit index (CFI) was .987; the Tucker-Lewis index (TLI)
value was .981 and the value of the standardized root mean square residual (SRMR)
was .032. The overall CFA model fit indices were supported by previous studies (Hu
& Bentler, 1999; Marsh, Hau & Wen, 2004). The results in Figure 4.6 exhibit that the
items were free from offending estimates. The standardized factor established
loadings ranging from .70 to .89, indicating statistically significant indicators. The
123
item loadings are shown in Table 4.5.
Table 4.5: The Loadings for Perceived-Usefulness Items
Item
No Item label Loadings M SD α
PU2 Using the ICT services available at the
university increases my research
productivity.
.83 5.555 1.270
.944
PU3 The current university ICT system
makes work more interesting.
.89 5.217 1.324
PU4 ICT facilities at the university improve
the quality of the work I do.
.88 5.401 1.289
PU5 Using the university ICT facilities
enhances my research skills.
.89 5.335 1.285
PU6 Using the university ICT facilities
would make it easier for me to find
information.
.86 5.558 1.304
PU7 ICT services at the university make
information always available to users.
.83 5.247 1.379
PU11 Using the university ICT facilities
provides me with the latest information
on specific areas of research.
.70 5.628 1.226
The results of the one-factor measurement model of perceived usefulness also
discovered that the lecturers’ perceptions on the usefulness of ICT facilities for their
teaching and research purposes in higher education were demonstrated by seven valid
predictors. They explained almost 71.1% of the variability of lecturers’ perceived
usefulness of ICT facilities as shown in Table 4.6.
Table 4.6: The Squared Multiple Correlations
Items Measured
PU2 .696
PU3 .799
PU4 .774
PU5 .796
PU6 .736
PU7 .688
PU11 .489
The average total variance 0.711
124
4.4 ONE-FACTOR MEASUREMENT MODEL OF COMPUTER SELF-
EFFICACY (CSE)
The specific objective was to examine the lecturers’ computer self-efficacy to access
ICT facilities in higher education for their teaching and research purposes. To address
the objective 4, the current study conducted Confirmatory Factor Analysis (CFA) to
assess the lecturers’ perceptions on their computer self-efficacy of ICT use and to
evaluate the effect of computer self-efficacy in the hypothesized TAG model,
hypotheses 1, 6, 7 and 8 were estimated. The original CFA model was estimated with
15 items of computer self-efficacy. All items demonstrated a loading greater than .59
on their respective factor. However, the findings of the one-factor measurement model
of computer self-efficacy revealed that the overall fit indices did not execute the
statistical requirements as indicated by the following fit indices: χ2 (df = 90) =
834.732; p = 0.000; RMSEA = 0.145; CFI = 0.861; TLI = 0.838; SRMR = 0.067. As
such, 8 items were removed due to violations and estimation (see Appendix A). After
deleting these items (CSE1, CSE2, CSE3, CSE5, CSE6, CSE8, CSE13 and CSE14),
the analysis showed a goodness-of-fit of the model. The items were removed one at a
time according to their modification indices, with the largest being removed first,
which showed the presence of multiple collinearity as shown in Table 4.7.
Table 4.7: The List of Removed Items
Item
Deleted
Chi-square
(χ2)
df p value RMSEA CFI TLI SRMR
CSE1 663.238 77 .000 .139 .880 .858 .062
CSE2 568.268 65 .000 .140 .885 .863 .062
CSE3 427.488 54 .000 .132 .904 .883 .060
CSE5 266.433 44 .000 .113 .934 .918 .057
CSE6 217.543 35 .000 .115 .942 .925 .059
CSE8 170.934 27 .000 .116 .947 .930 .061
CSE13 48.627 20 .000 .060 .988 .983 .028
CSE14 20.893 14 .104 .035 .997 .995 .012
125
The items that remained were: CSE4, CSE7, CSE9, CSE10, CSE11, CSE12
and CSE15. These items were used as the indicators for the one-factor measurement
model of computer self-efficacy. Figure 4.3 shows the error variances in the model
represented by e4, e7, e9, e10, e11, e12 and e15.
Figure 4.3: One-Factor Measurement Model for Computer Self-efficacy
The model’s overall goodness-of-fit statistics showed excellent fit; the model
chi-square was statistically non-significant, χ2 (df=14) =20.893, p=.104 and the root
mean square error of approximation (RMSEA) indicated the model’s adequacy at an
acceptable value of .035. The value of the comparative fit index (CFI) was .997; the
Tucker-Lewis index (TLI) value was .995 and the value of the standardized root mean
126
square residual (SRMR) was .012. The overall CFA model fit indices were
recommended by earlier researchers (Hu & Bentler, 1999; Marsh, Hau & Wen, 2004).
The results in Figure 4.3 show that the parameters were free from offending estimates.
The standardized factor demonstrated loadings ranging from .72 to .88, indicating
statistically significant indicators. The item loadings are shown in Table 4.8.
Table 4.8: The Loadings for Computer Self-efficacy
Item
No Item label Loadings M SD α
CSE4 I have the skills required to
communicate electronically with my
colleagues and students.
.72 5.974 .988
.938
CSE7 I have the skills required to use ICT
facilities to enhance the effectiveness
of my teaching, learning and research.
.85 5.931 .984
CSE9 I have the ability to navigate my way
through the ICT facilities.
.87 5.714 1.051
CSE10 I have the skills required to use the
ICT facilities to enhance the quality of
my research works.
.88 5.785 .009
CSE11 I have the ability to save and print
journals/articles from the research
databases.
.84 5.989 1.016
CSE12 I have the knowledge and skills
required to benefit from using the
university ICT facilities.
.87 5.891 .955
CSE15 I am capable of using multimedia
facilities for delivering lecturers.
.77 5.939 1.029
The findings of the one-factor measurement model of computer self-efficacy
also exhibited that the lecturers’ views on the computer self-efficacy of ICT usage
were indicated by seven valid predictors. They explained approximately 68.8% of the
variability of lecturers’ computer self-efficacy of ICT usage for their teaching and
research purposes in higher education as shown in Table 4.9.
127
Table 4.9: The Squared Multiple Correlations
Items Measured
CSE4 .517
CSE7 .721
CSE9 .760
CSE10 .771
CSE11 .699
CSE12 .763
CSE15 .586
The average total variance 0.688
4.5 ONE-FACTOR MEASUREMENT MODEL OF INTENTION TO USE
(INT)
The specific objective was to examine the lecturers’ intention to use of ICT facilities
for their teaching and research purposes in higher education. To address the objective
5, the present study performed a Confirmatory Factor Analysis (CFA) to measure the
lecturers’ perceptions on their intention to use of ICT facilities and to determine the
impact of intention to use in the hypothesized TAG model, hypotheses 4 and 10 were
tested. The primary CFA model was estimated with 10 items of intention to use. The
majority of the items exhibited a loading greater than .63 on their respective factor.
The majority of the standardized factor loadings were greater than .63, with only two
items demonstrating loadings ranging from .32 to .40 (INT4 and INT8). However, the
results of the one-factor measurement model of intention to use showed that the
overall fit indices did not confirm a satisfactory fit to the data as indicated by the
following fit indices: χ2 (df = 35) = 269.468; p = 0.000; RMSEA = 0.130; CFI = 0.892;
TLI = 0.861; SRMR = 0.083. Thus, the researcher dropped items: INT4, INT6 and
INT9 due to violations and estimation (see Appendix D). After deleting these items,
the analysis showed a goodness-of-fit of the model. The items were removed one at a
time according to their modification indices, with the largest being removed first,
128
which showed the presence of multiple collinearity as shown in Table 4.10.
Table 4.10: The List of Removed Items
Item
Deleted
Chi-square
(χ2)
df p value RMSEA CFI TLI SRMR
INT4 235.920 27 .000 .140 .900 .866 .070 INT6 162.597 20 .000 .134 .912 .876 .070 INT9 55.478 14 .000 .087 .966 .949 .069
Eventually, only seven items were taken, namely: INT1, INT2, INT3, INT5,
INT7, INT8 and INT10. These items were used as the indicators for the one-factor
measurement model of Intention to Use. Figure 4.14 shows the error variances in the
model represented by e1, e2, e3, e5, e7, e8, and e10.
Figure 4.4: One-Factor Measurement Model for Intention to Use
129
The model’s overall goodness-of-fit statistics demonstrated an adequate fit; the
model chi-square, χ2 (df=14) =55.478, p=.000 and the root mean square error of
approximation (RMSEA) showed the model’s adequacy with an acceptable value of
.087; the value of the comparative fit index (CFI) was .966; the Tucker-Lewis index
(TLI) value was .949 and the value of the standardized root mean square residual
(SRMR) was .069. The CFA model fit indices were supported by prior studies (Hu &
Bentler, 1999; Marsh, Hau & Wen, 2004). The majority of the items standardized
factor demonstrated loadings ranging from .69 to .83, indicating statistically
significant indicators. The only exception was the sixth item (INT8) which had a
loading of .30. This loading was supported by Hair et al., (2010). The results in Figure
4.4 show that the parameters were free from offending estimates. The item loadings
are shown in Table 4.11.
Table 4.11: The Loadings for Intention-to-Use Items
Item
No Item label Loadings M SD α
INT1 I will use the ICT facilities to carry
out my research activities.
.83 6.070 1.038
.700
INT2 I will use computer applications to
present lecture materials in class.
.72 6.280 1.001
INT3 I will use Microsoft office
applications to write my research
papers.
.71 6.189 1.157
INT5 I intend to use ICT facilities for
teaching, learning and research
purposes frequently.
.80 5.916 1.104
INT7 I will use the university research
databases to update myself on the
research areas I am pursuing.
.75 5.861 1.092
INT8 I intend to use the ICT facilities to
upgrade my research performance.
.30 6.108 3.724
INT10 I intend to use research related
software to analyze the data.
.69 5.755 1.221
130
The results of the one-factor measurement model of intention to use also
confirmed that the lecturers’ perceptions on the intention to use of ICT facilities were
demonstrated by seven valid predictors. They explained almost 50% of the variability
of lecturers’ intention to use of ICT facilities for their teaching and research purposes
in higher education as shown in Table 4.12.
Table 4.12: The Squared Multiple Correlations
Items Measured
INT1 .691
INT2 .524
INT3 .506
INT5 .638
INT7 .570
INT8 .093
INT10 .478
The average total variance 0.50
4.6 ONE-FACTOR MEASUREMENT MODEL OF USE (USE)
The specific objective was to determine the lecturers’ actual use of ICT facilities for
their teaching and research purposes in higher education. To address the objective 6,
this study analyzed Confirmatory Factor Analysis (CFA) to examine the lecturers’
views on their actual use of ICT facilities and to measure the effect of use in the
hypothesized TAG model, hypotheses 5 was estimated. The initial CFA model was
tested for the factor of use by 16 items. Most items demonstrated a loading greater
than .56 on their particular factor. However, few items indicated a loading smaller
than the recommended value (see Appendix E). Subsequently, the findings of the one-
factor measurement model of use exhibited that the overall fit indices did not show a
satisfactory fit to the data as indicated by the following fit indices: χ2 (df = 104) =
875.902; p = 0.000; RMSEA = 0.137; CFI = 0.644; TLI = 0.589; SRMR = 0.293. As a
131
result, the initial model required revision due to violations and estimation. After
removing 11 items (USE1, USE3, USE7, USE8, USE9, USE10, USE12, USE13,
USE14, USE15 AND USE16), the measurement model executed the statistical
requirements. The items were removed one at a time according to their modification
indices, with the largest being removed first, which showed the presence of multiple
collinearity as shown in Table 4.13.
Table 4.13: The List of Removed Items
Item
Deleted
Chi-square
(χ2)
df p value RMSEA CFI TLI SRMR
USE1 833.444 90 .000 .145 .642 .582 .305
USE3 792.271 77 .000 .153 .642 .577 .319
USE7 647.096 65 .000 .151 .680 .616 .291
USE8 562.157 54 .000 .154 .689 .620 .303
USE9 480.960 44 .000 .159 .715 .644 .258
USE10 436.691 35 .000 .170 .692 .604 .281
USE12 359.946 27 .000 .177 .730 .640 .214
USE13 319.156 20 .000 .195 .749 .648 .205
USE14 107.751 14 .000 .130 .893 .840 .174
USE15 18.893 9 .026 .053 .986 .977 .078
USE16 8.475 5 .132 .042 .995 .990 .024
The following 5 items (USE2, USE4, USE5, USE6 and USE11) were used as
the indicator for the one-factor measurement model of use. Figure 4.5 shows the error
variances in the model represented by e2, e4, e5, e6 and e11.
132
Figure 4.5: One-Factor Measurement Model for Use
The model’s overall goodness-of-fit statistics indicated a satisfactory fit; the
model chi-square was statistically non-significant, χ2 (df=5) =8.475, p=.132 and the
root mean square error of approximation (RMSEA) showed an adequate fit with a
value of .042. The value of the comparative fit index (CFI) was .995; the Tucker-
Lewis index (TLI) value was .990 and the value of the standardized root mean square
residual (SRMR) was .024. The CFA model fit indices were supported by previous
researchers (Hu & Bentler, 1999; Marsh, Hau & Wen, 2004). The results in Figure 4.5
exhibit that the items were free from offending estimates. The standardized factor
established loadings ranging from .47 to .83, representing statistically significant
indicators. The item loadings are shown in Table 4.14.
133
Table 4.14: The Loadings for Use Items
Item
No Item label Loadings M SD α
USE2 How often do you use the
following ICT facilities for
research and academic-related
activities? - Laptop or Desktop
computers.
.69 6.396 .892 .793
USE4 How often do you use the
following ICT facilities for
research and academic? - Web
browser (www).
.75 6.239 1.098
USE5 How often do you use the
following ICT facilities for
research and academic? - Search
engine (e.g. Google, Yahoo,
etc.).
.83 6.383 .902
USE6 How often do you use the
following ICT facilities for
research and academic? -
Microsoft Office applications.
.73 6.308 933
USE11 How often do you use the
following ICT facilities for
research and academic-related
activities? - Document
editing/composing.
.47 5.785 1.438
The findings of the one-factor measurement model of use also revealed that the
lecturers’ views on the actual use of ICT facilities were indicated by five valid
predictors. They explained approximately 49.5% of the variability of lecturers’ use of
ICT facilities for their teaching and research purposes in higher education as shown in
Table 4.15.
Table 4.15: The Squared Multiple Correlations
Items Measured
USE2 .479
USE4 .560
USE5 .692
USE6 .529
USE11 .219
The average total variance 0.495
134
4.7 ONE-FACTOR MEASUREMENT MODEL OF GRATIFICATION
(GRAT)
The specific objective was to assess the lecturers’ gratification in using ICT services
for their teaching and research purposes in higher education. To address the objective
7, this study conducted a Confirmatory Factor Analysis (CFA) to confirm the
lecturers’ perceptions on their gratification in using ICT facilities and to examine the
lecturers’ overall gratification in the hypothesized TAG model; the Structural
Equation Modeling (SEM) was applied. The original CFA model was performed on
10 items that explained the factor of gratification. The majority of the standardized
factor loadings were greater than .78, with only two items that demonstrated loadings
ranging from .61 to .65 (GRAT5 and GRAT10). However, the value of RMSEA and
TLI were not as expected (see Appendix F). As such, the one-factor measurement
model of gratification was revised due to violations and estimation. After deleting
these items (GRAT1, GRAT2, GRAT5 and GRAT8), the analysis demonstrated an
exceptional goodness-of-fit of the model. The items were removed one at a time
according to their modification indices, with the largest being removed first, which
showed the presence of multiple collinearity as shown in Table 4.16.
Table 4.16: The List of Removed Items
Item
Deleted
Chi-square
(χ2)
df p value RMSEA CFI TLI SRMR
GRAT1 205.309 27 .000 .129 .938 .918 .063
GRAT2 125.762 20 .000 .116 .955 .937 .056
GRAT5 47.803 14 .000 .078 .984 .976 .038
GRAT8 13.301 9 .149 .035 .997 .995 .030
135
These are the 6 items that remained after the corrections: GRAT3, GRAT4,
GRAT6, GRAT7, GRAT9 and GRAT10. These items were used as the indicators for
the one-factor measurement model of gratification. Figure 4.6 depicts the error
variances in the model represented by e3, e4, e6, e7, e9 and e10.
Figure 4.6: One-Factor Measurement Model for Gratification
The results of the model’s overall goodness-of-fit statistics showed an
exceptional fit; the model chi-square was statistically non-significant, χ2 (df=9)
=13.301, p=.149 and the root mean square error of approximation (RMSEA)
represented a goodness-of-fit of the model with a satisfactory value of .035; the value
of the comparative fit index (CFI) was .997; the Tucker-Lewis index (TLI) value was
.995 and the value of the standardized root mean square residual (SRMR) was .030.
The CFA model fit indices were recommended by earlier studies (Hu & Bentler, 1999;
Marsh, Hau & Wen, 2004). The results in Figure 4.6 indicate that the parameters were
free from offending estimates. The standardized factor demonstrated loadings ranging
136
from .65 to .86, representing statistically significant indicators. The item loadings are
shown in Table 4.17.
Table 4.17: The Loadings for Gratification Items
Item No Item label Loadings M SD α
GRAT3 I am satisfied with the ease of use
of the ICT facilities.
.82 5.346 1.294 .908
GRAT4 The university ICT facilities have
greatly affected the way I search for
information and conduct my
research.
.79 5.457 1.251
GRAT6 Overall I am satisfied with the
amount of time it takes to complete
my task.
.86 5.237 1.361
GRAT7 I am satisfied with the structure of
accessible information (available as
categories of research domain, or by
date of issue—of journals in
particular, or as full-texts or
abstracts of theses and dissertations)
of the university research databases.
.82 5.149 1.335
GRAT9 I am satisfied in using ICT facilities
for teaching, learning and research.
.85 5.467 1.239
GRAT10 Overall I am satisfied with the
Wireless Internet service provided
at the university.
.65 4.997 1.596
The findings of the one-factor measurement model of gratification showed that
the lecturers’ perceptions on the gratification of ICT facilities were represented by six
valid predictors. They explained almost 64.1% of the variability of lecturers’
gratification in using ICT facilities for their teaching and research purposes in higher
education as shown in Table 4.18.
137
Table 4.18: The Squared Multiple Correlations
Items Measured
GRAT3 .666
GRAT4 .623
GRAT6 .748
GRAT7 .667
GRAT9 .724
GRAT10 .422
The average total variance 0.641
SECTION 2: ESTIMATING THE FULL-FLEDGED STRUCTURAL
EQUATION MODELING
4.8 THE HYPOTHESIZED TECHNOLOGY ADOPTION AND
GRATIFICATION (TAG) MODEL
The structural equation modeling was adopted to evaluate the hypothesized
Technology Adoption and Gratification (TAG) model. The estimation of parameters
used the utmost likelihood technique provided by AMOS 18 to assess the model fit
statistics, structural model, and measurement model (Byrne, 2000). The hypothesized
TAG model was integrated by the six measurement models of the latent constructs,
namely, computer self-efficacy (CSE), perceived usefulness (PU), perceived ease of
use (PEU), intention to use (INT), use (USE) and gratification (GRAT) as tested
individually by the Confirmatory Factor Analysis (CFA) for their overall goodness-of-
fit of the models. The TAG model contained one exogenous variable (CSE), four
mediator variables (PU, PEU, INT and USE), one endogenous variable (GRAT),
thirty eight observed variables and forty three error variances as shown in Figure 4.7.
Perceived usefulness and perceived ease of use were determined by computer self-
efficacy. Furthermore, intention to use was determined directly by perceived ease of
use and perceived usefulness and indirectly by computer self-efficacy. Besides this,
the actual use was directly and indirectly influenced by intention to use, perceived
138
usefulness and perceived ease of use as well as computer self-efficacy. Similarly,
gratification was directly and indirectly influenced by four mediator variables (PEU,
PU, INT and USE), whereas computer self-efficacy was an indirect determinant of
gratification.
In the hypothesized TAG model, the majority of the items exhibited a loading
greater than .70. However, few items produced multiple large residuals with a low
factor loading, and furthermore, they had large modification indices suggesting the
presence of multiple collinearity (see Appendix G). Subsequently, the overall
statistical analyses demonstrated inadequate fit statistics of the hypothesized TAG
model to the empirical data: the chi-square, χ2 (df=654) =1781.099, p=.000; the value
of the comparative fit index (CFI) was .893 and the Tucker-Lewis index (TLI) value
was .885. The only exception was the root mean square error of approximation
(RMSEA) value of .066, which was smaller than the recommended value. The results
in Figure 4.7 exhibit that the parameter estimates of the hypothesized TAG model
were free from offending values. Most path coefficients of the hypothesized TAG
model’s critical ratio (CR >1.96) were statistically significant. Significant paths were
from: computer self-efficacy (CSE) to perceived usefulness (PU) and perceived ease
of use (PEU), perceived usefulness to intention to use (INT) and gratification (GRAT)
and perceived ease of use (PEU) to intention to use (INT) and gratification (GRAT).
Similarly, the paths from intention to use (INT) to use (USE) and gratification (GRAT)
were also significant. However, the paths from perceived usefulness (PU) and
perceived ease of use (PEU) to use (USE) were not significant as well as the path from
use (USE) to gratification (GRAT).
139
Figure 4.7: The Hypothesized Technology Adoption and Gratification (TAG) Model
140
The data demonstrated that computer self-efficacy was the most influential
factor by a moderate amount in the TAG model in affecting the lecturers’ perceived
usefulness and perceived ease of use of ICT facilities in higher education. Likewise,
perceived usefulness was slightly more influential than perceived ease of use in
affecting lecturers’ gratification in using ICT facilities. On the other hand, perceived
ease of use was a more influential predictor than perceived usefulness in affecting
lecturers’ intention to use ICT facilities. The total standardized effect sizes of (i)
computer self-efficacy → perceived ease of use and perceived usefulness were .566
and .528 respectively, (ii) perceived usefulness → intention to use, use and
gratification were .278, .148 and .525 respectively, (iii) perceived ease of use →
intention to use, use and gratification were .394, .250 and .378 respectively, (iv)
intention to use → use and gratification were .373 and .147 respectively, (v) use →
gratification was -.078, and (vi) computer self-efficacy → intention to use, use and
gratification were .370, .220 and .491 respectively. In addition, the analysis revealed
that the exogenous variable (CSE) explained almost 32% and 28% of the variance in
perceived ease of use and perceived usefulness respectively, while the exogenous
variable (CSE) and the two mediator variables (PU and PEU) collectively explained
approximately 30% of the variability of lecturers’ intention to use of ICT facilities in
higher education. The exogenous variable (CSE) and the three mediator variables (PU,
PEU and INT) explained almost 20% of the variance of lecturers’ actual use of ICT
facilities. Finally, the exogenous variable (CSE) and the four mediator variables (PU,
PEU, INT and USE) collectively explained approximately 56% of the variability of
lecturers’ gratification in using ICT facilities in higher education. The TAG model,
however, had to be revised due to a few items that produced multiple large residuals
and had large modification indices suggesting the presence of multi-collinearity.
141
Furthermore, the presence of a negative path coefficient that contradicted the
hypothesis as well as the TAG model also exhibited inadequate convergent and
discriminant validity. Thus, indicating that the parameters were not estimated or
revealed in detail.
4.8.1 Hypotheses Testing
Testing the hypothesized TAG model, using the structural equation modeling, gave
results that demonstrated the findings of the overall causal model, and specified the
direct and indirect effect of exogenous variables, namely, computer self-efficacy on
the mediated (perceived ease of use, perceived usefulness, intention to use and use)
and endogenous (gratification) variables. The findings of the hypotheses test are as
followed:
Hypothesis 1
Computer self-efficacy (CSE) will have a positive direct influence on lecturers’
perceived ease of use (PEU) and perceived usefulness (PU) of ICT facilities in higher
education.
The findings revealed that computer self-efficacy had a significant direct
influence on perceived ease of use (β = .57, p < .001) and perceived usefulness (β =
.53, p < .001) of ICT facilities in higher education.
Hypothesis 2
Perceived ease of use (PEU) will have a positive direct influence on lecturers’
intention to use (INT), actual use (USE) and gratification (GRAT) in using ICT
facilities in higher education.
142
The results demonstrated that perceived ease of use had a significant direct
influence on lecturers’ intention to use (β = .39, p < .001) and gratification (β = .33, p
< .001) in using ICT facilities in higher education. However, it did not exert a
statistically significant influence on use (β = .10, p < .005).
Hypothesis 3
Perceived usefulness (PU) will have a positive direct influence on lecturers’ intention
to use (INT), actual use (USE) and gratification (GRAT) in using ICT facilities in
higher education.
The results of the hypothesized TAG model showed that perceived usefulness
had a significant positive direct effect on intention to use (β = .28, p < .001) and
gratification (β = .49, p < .001) in using ICT facilities in higher education.
Nevertheless, perceived usefulness did not have a statistically significant influence on
use (β = .04, p > .005).
Hypothesis 4
Intention to use (INT) will have a positive direct influence on lecturers’ gratification
(GRAT) and actual use (USE) of ICT facilities in higher education.
The findings of the TAG Model exhibited that intention to use had a positive
direct influence on use (β = .37, p < .001) and gratification (β = .18, p < .001) in using
ICT facilities.
Hypothesis 5
Actual use (USE) will have a positive direct influence on gratification (GRAT) in
using ICT services in higher education.
143
The finding depicted that use had a direct effect on gratification, but in an
adverse direction (β = -.08, p > .005).
Hypothesis 6
Computer self-efficacy (CSE) will have a positive indirect influence on lecturers’
intention to use (INT) of ICT facilities mediated by perceived ease of use (PEU) and
perceived usefulness (PU), respectively.
The results indicated that computer self-efficacy had a statistically significant
and practically important indirect influence on intention to use mediated by perceived
usefulness (Chi-square, χ2 = 3.510; p = 0.000). Subsequently, computer self-efficacy
also had a statistically significant and practically important indirect influence on
intention to use mediated by perceived ease of use (Chi-square, χ2 = 3.599; p = 0.000)
as conducted by the Sobel test (Sobel, 1982).
Hypothesis 7
Computer self-efficacy (CSE) will have a positive indirect influence on lecturers’
gratification (GRAT) in using ICT facilities mediated by perceived ease of use (PEU)
and perceived usefulness (PU), respectively.
The findings showed that computer self-efficacy had a statistically significant
and practically important indirect influence on lecturers’ gratification in using ICT
facilities mediated by perceived ease of use (Chi-square, χ2 = 2.978; p = 0.001).
Similarly, computer self-efficacy also had a statistically significant and practically
important indirect influence on gratification mediated by perceived usefulness (Chi-
square, χ2 = 5.098; p = 0.000) as conducted by the Sobel test (Sobel, 1982).
144
Hypothesis 8
Computer self-efficacy (CSE) will have a positive indirect influence on lecturers’ use
(USE) of ICT facilities mediated by perceived ease of use (PEU) and perceived
usefulness (PU), respectively.
The results discovered that computer self-efficacy did not have a significant
indirect influence on lecturers’ use of ICT facilities mediated by perceived ease of use
(χ2 = 1.254; p = 0.104) and perceived usefulness (χ2 = 0.724; p = 0.234), respectively
as conducted by the Sobel test (Sobel, 1982).
Hypothesis 9
Perceived ease of use (PEU) and Perceived usefulness (PU) will have a positive
indirect influence on lecturers’ gratification (GRAT) in using ICT facilities mediated
by intention to use (INT), respectively.
The findings confirmed that perceived ease of use had a statistically significant
indirect influence on gratification in using ICT facilities mediated by intention to use
(Chi-square, χ2 = 2.257; p = 0.011), while perceived usefulness showed a significant
indirect influence on lecturers’ gratification mediated by intention to use of ICT
facilities (Chi-square, χ2 = 2.255; p = 0.012) as conducted by the Sobel test (Sobel,
1982).
Hypothesis 10
Intention to use (INT) will have a positive indirect influence on lecturers’ gratification
(GRAT) mediated by actual use (USE) of ICT facilities in higher education.
The results showed that intention to use did not have a significant indirect
influence on lecturers’ gratification mediated by actual use of ICT facilities due to the
145
negative path coefficient (Chi-square, χ2 = -1.015; p = 0.154) as conducted by the
Sobel test (Sobel, 1982).
Hypothesis 11
There will be a cross-cultural invariant of the causal structure of the TAG model.
The findings of the invariance analysis of the TAG model demonstrated that
the revised TAG model is valid for measuring lecturers’ adoption and gratification in
using ICT facilities in higher education for their teaching and research purposes.
However, the invariance analysis revealed that the TAG model works differently in
cross-cultures. This means that there was a cross-cultural invariant in lecturers’
adoption and gratification in using ICT facilities.
The results of the Hypotheses testing are summarized in Table 4.19.
146
Table 4.19: The Summary of the Hypotheses Testing
Hypotheses Description Results
H1 Computer self-efficacy (CSE) will have a positive direct
influence on lecturers’ perceived ease of use (PEU) and
perceived usefulness (PU) of ICT facilities in higher
education.
Supported
H2 Perceived ease of use (PEU) will have a positive direct
influence on lecturers’ intention to use (INT), actual use
(USE) and gratification (GRAT) in using ICT facilities
in higher education.
Supported
H3 Perceived usefulness (PU) will have a positive direct
influence on lecturers’ intention to use (INT), actual use
(USE) and gratification (GRAT) in using ICT facilities
in higher education.
Supported
H4 Intention to use (INT) will have a positive direct
influence on lecturers’ gratification (GRAT) and actual
use (USE) of ICT facilities in higher education.
Supported
H5 Actual use (USE) will have a positive direct influence
on gratification (GRAT) in using ICT services in higher
education.
Not
Supported
H6 Computer self-efficacy (CSE) will have a positive
indirect influence on lecturers’ intention to use (INT) of
ICT facilities mediated by perceived ease of use (PEU)
and perceived usefulness (PU), respectively.
Supported
H7 Computer self-efficacy (CSE) will have a positive
indirect influence on lecturers’ gratification (GRAT) in
using ICT facilities mediated by perceived ease of use
(PEU) and perceived usefulness (PU), respectively
Supported
H8 Computer self-efficacy (CSE) will have a positive
indirect influence on lecturers’ use (USE) of ICT
facilities mediated by perceived ease of use (PEU) and
perceived usefulness (PU), respectively.
Not
Supported
H9 Perceived ease of use (PEU), Perceived usefulness (PU)
will have a positive indirect influence on lecturers’
gratification (GRAT) in using ICT facilities mediated by
intention to use (INT), respectively.
Supported
H10 Intention to use (INT) will have a positive indirect
influence on lecturers’ gratification (GRAT) mediated
by actual use (USE) of ICT facilities in higher
education.
Not
Supported
H11 There will be a cross-cultural invariant of the causal structure of the TAG model.
Supported
147
4.9 THE REVISED TECHNOLOGY ADOPTION AND GRATIFICATION
(TAG) MODEL
The hypothesized TAG model was revised and re-estimated in order to examine its
overall adequacy. In the re-specified TAG model, the items were removed one at a
time (CSE15, PEU9, PEU15, GRAT7, GRAT10, USE2, USE11, INT8 and INT10)
which produced multiple large residuals and associated with the largest modification
indices with relatively low factor loadings, and that were found to be cross-loaded as
shown in Table 4.20.
Table 4.20: The List of Removed Items
Item
Deleted
Chi-square
(χ2)
df p value RMSEA CFI TLI
CSE15 1701.441 621 .000 .066 .894 .886
PEU9 1571.680 586 .000 .065 .901 .894
PEU15 1431.166 552 .000 .063 .909 .902
GRAT7 1361.367 519 .000 .064 .909 .902
GRAT10 1279.240 487 .000 .064 .913 .905
USE2 1226.239 456 .000 .065 .913 .905
USE11 1165.950 426 .000 .066 .915 .907
INT8 1154.941 397 .000 .070 .913 .905
INT10 1095.131 369 .000 .071 .914 .906
After removing these items, the revised TAG model exhibited adequate
convergent and discriminant validity (see Appendix H), with all loadings showing a
value greater than 0.70 as shown in Table 4.21. Thus, only 29 items now remained
from the initial TAG model set of 38 items. Figure 4.8 presents the parameter
estimates of the revised TAG, which were free from offending values. The magnitude
of the factor loadings and path coefficients were statistically significant.
148
Figure 4.8: The Revised Technology Adoption and Gratification (TAG) Model
149
With the revised TAG model, the direct paths from perceived usefulness (PU)
and perceived ease of use (PEU) to use (USE) were dropped due to low coefficients of
0.04 (PU > USE) and 0.10 (PEU > USE), respectively. The direct path from use (USE)
to gratification (GRAT) was also dropped due to a negative path coefficient of -0.08
(USE > GRAT) which indicated that the respective hypotheses were not supported.
The revised model depicted a goodness-of-fit to the empirical data as indicated by the
following fit indices: χ2 (df = 369) = 1095.131; p < 0.000; RMSEA = 0.071; CFI =
0.914; TLI = 0.906 as supported by Byrne (2000). The Figure 4.8 shows the error
variances in the revised TAG model represented by e1, e2, e3, e5, e7, e8, e9, e10, e11,
e15, e16, e17, e18, e19, e20, e21, e22, e23, e24, e25, e26, e27, e30, e31, e32, e33, e35,
e36, e37, e39, e40, e41, e42 and e43.
In addition, the data of the revised TAG model revealed standardized path
coefficients as shown in Figure 4.8. As anticipated, hypotheses H1 and H2 were
supported, the findings revealed that computer self-efficacy had a significant positive
direct influence on perceived usefulness (β = 0.54, p<0.001) and perceived ease of use
(β = 0.53, p < .001). On the other hand, perceived ease of use had a significant direct
influence on gratification (β = 0.26, p<0.001) and intention to use (β = 0.25, p<0.001).
Subsequently, hypotheses H3 and H4 were supported, the results of the revised TAG
model discovered that perceived usefulness had a significant direct influence on
intention to use (β = 0.37, p<0.001) and gratification (β = 0.49, p<0.001). Besides this,
intention to use had also a significant direct influence on use (β = 0.45, p<0.001) and
gratification (β = 0.21, p<0.001). Moreover, H6 was also supported by the findings of
the revised TAG model which indicated that computer self-efficacy had a statistically
significant and practically important indirect influence on intention to use mediated by
perceived usefulness (Chi-square, χ2 = 4.451; p = 0.000). Subsequently, computer
150
self-efficacy also had a statistically significant and practically important indirect
influence on intention to use mediated by perceived ease of use (Chi-square, χ2 =
3.224; p = 0.000) as conducted by the Sobel test (Sobel, 1982). Furthermore, The
results of the revised TAG model demonstrated that computer self-efficacy had a
statistically significant and practically important indirect influence on lecturers’
gratification in using ICT facilities mediated by perceived ease of use (Chi-square, χ2
= 3.411; p = 0.000). Similarly, computer self-efficacy also had a statistically
significant and practically important indirect influence on gratification mediated by
perceived usefulness (Chi-square, χ2 = 5.253; p = 0.000) as conducted by the Sobel
test and validated H7 (Sobel, 1982). Eventually, H9 was supported as the results of the
revised TAG model discovered that perceived ease of use had a statistically significant
indirect influence on gratification in using ICT facilities mediated by intention to use
(Chi-square, χ2 = 2.503; p = 0.006), while perceived usefulness showed a significant
indirect influence on lecturers’ gratification mediated by intention to use of ICT
facilities (Chi-square, χ2 = 2.912; p = 0.001) as conducted by the Sobel test (Sobel,
1982). Furthermore, all path coefficients of the causal structure were practically
important and statistically significant, and the strongest value of the standardized path
coefficient of computer self-efficacy on perceived usefulness was 0.54. The data also
demonstrated that computer self-efficacy was a moderately more influential factor in
the revised TAG model in directly affecting the lecturers’ perceived usefulness and
perceived ease of use as well as indirectly influencing the lecturers’ intention to use,
use and gratification in using ICT facilities in higher education for their research and
teaching purposes. Moreover, the total standardized effect sizes of (i) computer self-
efficacy → perceived ease of use and perceived usefulness were .525 and .536
respectively, (ii) perceived usefulness → intention to use, use and gratification
151
were .375, .168 and .574 respectively, (iii) perceived ease of use → intention to use,
use and gratification were .247, .111 and .310 respectively, (iv) intention to use → use
and gratification were .447 and .231 respectively, and (v) computer self-efficacy →
intention to use, use and gratification were .331, .148 and .471 respectively. In
addition, the analysis revealed that the exogenous variable (CSE) explained almost
28% and 29% of the variance in perceived ease of use and perceived usefulness
respectively, while the exogenous variable (CSE) and the two mediator variables (PU
and PEU) collectively explained approximately 25% of the variability of lecturers’
intention to use of ICT facilities in higher education. Similarly, the exogenous variable
(CSE) and the three mediator variables (PU, PEU and INT) explained almost 20% of
the variance of lecturers’ actual use of ICT facilities. Finally, the exogenous variable
(CSE) and the three mediator variables (PU, PEU and INT) collectively explained
approximately 56% of the variability of lecturers’ gratification in using ICT facilities
in higher education.
Table 4.21: Valid Items of the Revised TAG Model and Their Corresponding
Loadings, Means, Standard Deviations and Alpha Values
Constructs Items Item Measure Loadings M SD Alpha
Computer
Self-
efficacy
(CSE)
CSE4 I have the skills required to
communicate electronically with
my colleagues and students.
.72 5.974 .988
CSE7 I have the skills required to use
ICT facilities to enhance the
effectiveness of my teaching,
learning and research.
.84 5.931 .984
CSE9 I have the ability to navigate my
way through the ICT facilities. .88 5.714 1.051 .933
CSE10 I have the skills required to use
the ICT facilities to enhance the
quality of my research works.
.88 5.785 1.009
CSE11 I have the ability to save and
print journals/articles from the
research databases.
.83 5.989 1.016
152
Table 4.21, continued
Constructs Items Item Measure Loadings M SD Alpha
CSE12 I have the knowledge and skills
required to benefit from using
the university ICT facilities.
.87 5.891 .955
Perceived
Usefulness
(PU)
PU2 Using the ICT services available
at the university increases my
research productivity.
.84 5.555 1.270
PU3 The current university ICT
system makes work more
interesting.
.89 5.217 1.324 .944
PU4 ICT facilities at the university
improve the quality of the work I
do.
.88 5.401 1.289
PU5 Using the university ICT
facilities enhances my research
skills.
.89 5.335 1.285
PU6 Using the university ICT
facilities would make it easier
for me to find information.
.86 5.558 1.304
PU7 ICT services at the university
make information always
available to users.
.83 5.247 1.379
PU11 Using the university ICT
facilities provides me with the
latest information on specific
areas of research.
.70 5.628 1.226
Perceived
Ease of Use
(PEU)
PEU2 I find it easy to access the
university research databases. .70 5.300 1.313
PEU4 Interacting with the research
databases system requires
minimal effort on my part.
.74 4.931 1.366 .830
PEU6 Interacting with the university
research databases system is
very stimulating for me.
.77 4.846 1.388
PEU7 I find it easy to select
articles/journals of different
categories using research
databases
(Education/Engineering/Busines
s, etc.).
.75 5.346 1.246
153
Table 4.21, continued
Constructs Items Item Measure Loadings M SD Alpha
Intention to
Use (INT)
INT1 I will use the ICT facilities to
carry out my research activities. .83 6.070 1.038
INT2 I will use computer applications
to present lecture materials in
class.
.72 6.280 1.001
INT3 I will use Microsoft office
applications to write my research
papers.
.70 6.189 1.157 .874
INT5 I intend to use ICT facilities for
teaching, learning and research
purposes frequently.
.79 5.916 1.104
INT7 I will use the university research
databases to update myself on
the research areas I am pursuing.
.72 5.861 1.092
Actual Use
(USE)
USE4 How often do you use the
following ICT facilities for
research and academic? – Web
browser (www).
.72 6.239 1.098
USE5 How often do you use the
following ICT facilities for
research and academic? – Search
engine (e.g. Google, Yahoo,
etc.).
.85 6.383 .902 .802
USE6 How often do you use the
following ICT facilities for
research and academic? –
Microsoft Office applications.
.72 6.308 .933
Gratificatio
n (GRAT)
GRAT3 I am satisfied with the ease of
use of the ICT facilities. .80 5.346 1.294
GRAT4 The university ICT facilities
have greatly affected the way I
search for information and
conduct my research.
.80 5.457 1.251 .899
GRAT6 Overall I am satisfied with the
amount of time it takes to
complete my task.
.84 5.237 1.361
GRAT9 I am satisfied in using ICT
facilities for teaching, learning
and research.
.82 5.467 1.239
154
4.10 CROSS-CULTURAL VALIDATION OF TAG MODEL
One of the objectives of this study was to evaluate the structural invariance of the
TAG model regarding the moderating effect of the cross-cultural dimension of the
respondents. In order to test cross-cultural-invariant, a two-stage analysis i.e.,
configural and metric invariance, was conducted on both the respondents of the
University of Malaya (n1 = 200) and Jiaxing University (n2 = 196). The revised TAG
model was estimated using a total of 396 lecturers who were collected from two
comprehensive public universities, namely, the University of Malaya in Malaysia and
Jiaxing University in China. However, to perform invariance analyses between the
groups of Malaysian and Chinese participants, the present study separated the original
pool of data into two groups. The overall fit indices of the revised TAG model for
Malaysian respondents discovered a satisfactory fit to the empirical data as
demonstrated by the following fit indices: χ2 (df = 369) = 897.527; p < 0.000;
RMSEA = 0.085; CFI = 0.882; TLI = 0.870 as shown in Figure 4.9. Moreover, the
data of the revised TAG model for Malaysian respondents confirmed the standardized
path coefficients as indicated in Figure 4.30. As anticipated, hypotheses H1 and H2
were accepted, the results exhibited that Malaysian lecturers’ computer self-efficacy
had a significant positive direct effect on their perceived usefulness (β = 0.63, p<0.001)
and perceived ease of use (β = 0.57, p < .001). On the other hand, Malaysian lecturers’
perceived ease of use had a significant direct influence on their gratification (β = 0.26,
p<0.001) but it did not exert any significant direct effect on intention to use (β = 0.06,
p≥0.05). Additionally, hypotheses H3 and H4 were supported, the findings of revised
TAG model for Malaysian participants revealed that lecturers’ perceived usefulness
had a significant direct influence on their intention to use (β = 0.56, p<0.001) and
gratification (β = 0.40, p<0.001), while Malaysian lecturers’ intention to use had a
155
significant direct effect on their actual use (β = 0.32, p<0.001) and gratification (β =
0.30, p<0.001) in using ICT facilities for their teaching and research purposes in
higher education.
Figure 4.9: The Revised TAG Model for Malaysian Participants
156
Furthermore, H6 was also supported, the results of the revised TAG model for
Malaysian respondents demonstrated that Malaysiann lecturers’ computer self-
efficacy had a statistically significant and practically important indirect effect on their
intention to use mediated by perceived usefulness (Chi-square, χ2 = 4.952; p = 0.000).
However, Malaysian lecturers’ computer self-efficacy did not show a statistically
significant and practically important indirect effect on their intention to use mediated
by perceived ease of use due to the low path coefficient. In addition, The findings of
the revised TAG model for Malaysian participants showed that lecturers’ computer
self-efficacy had a statistically significant and practically important indirect influence
on their gratification in using ICT facilities mediated by perceived ease of use (Chi-
square, χ2 = 2.956; p = 0.001). Meanwhile, Malaysian lecturers’ computer self-
efficacy also had a statistically significant and practically important indirect effect on
their gratification mediated by perceived usefulness (Chi-square, χ2 = 3.273; p =
0.000) as conducted by the Sobel test and validated H7 (Sobel, 1982). Lastly, H9 was
supported as the findings of the revised TAG model for Malaysian participants
indicated that perceived usefulness had a statistically significant indirect effect on
gratification in using ICT facilities for their teaching and research purposes in higher
education mediated by intention to use (Chi-square, χ2 = 2.702; p = 0.003) as
conducted by the Sobel test (Sobel, 1982), while lecturers’ perceived ease of use did
not exert any significant indirect influence on their gratification mediated by intention
to use of ICT facilities due to the insignificant path coefficient. Moreover, all path
coefficients of the causal structure were practically important and statistically
significant, while the strongest value of the standardized path coefficient of Malaysian
lecturers’ computer self-efficacy on their perceived usefulness was 0.63. The data also
confirmed that Malaysian lecturers’ computer self-efficacy was the most influential
157
factor in the revised TAG model for Malaysian participants in directly influencing the
lecturers’ perceived usefulness and perceived ease of use as well as indirectly
affecting the lecturers’ intention to use, use and gratification in using ICT facilities for
their teaching and research purposes in higher education.
Subsequently, the total standardized effect sizes of (i) computer self-efficacy
→ perceived ease of use and perceived usefulness were .570 and .629 respectively, (ii)
perceived usefulness → intention to use, use and gratification were .563, .179
and .573 respectively, (iii) perceived ease of use → intention to use, use and
gratification were .056, .018 and .277 respectively, (iv) intention to use → use and
gratification were .318 and .301 respectively, and (v) computer self-efficacy →
intention to use, use and gratification were .386, .123 and .518 respectively.
Consequently, the analysis of the revised TAG model for Malaysian participants
discovered that the exogenous variable (CSE) explained almost 32% and 40% of the
variance in perceived ease of use and perceived usefulness respectively, while the
exogenous variable (CSE) and the two mediator variables (PU and PEU) collectively
explained almost 34% of the variability of Malaysian lecturers’ intention to use of ICT
facilities for their teaching and research purposes in higher education. Similarly, the
exogenous variable (CSE) and the three mediator variables (PU, PEU and INT)
explained almost 10% of the variance of lecturers’ actual use of ICT facilities. Finally,
the exogenous variable (CSE) and the three mediator variables (PU, PEU and INT)
collectively explained approximately 58% of the variability of lecturers’ gratification
in using ICT facilities in higher education. The revised TAG model for Malaysian
participants depicted adequate convergent and discriminant validity (see Appendix I),
with the majority of the items’ loadings showing a value greater than 0.70 as shown in
Table 4.22.
158
Table 4.22: Valid Items of the Revised TAG Model for Malaysian Participants and
Their Corresponding Loadings, Means, Standard Deviations and Alpha Values
Constructs Items Item Measure Loadings M SD Alpha
Computer
Self-
efficacy
(CSE)
CSE4 I have the skills required to
communicate electronically
with my colleagues and
students.
.69 5.974 .988
.933
CSE7 I have the skills required to use
ICT facilities to enhance the
effectiveness of my teaching,
learning and research.
.82 5.931 .984
CSE9 I have the ability to navigate my
way through the ICT facilities. .87 5.714 1.051
CSE10 I have the skills required to use
the ICT facilities to enhance the
quality of my research works.
.90 5.785 1.009
CSE11 I have the ability to save and
print journals/articles from the
research databases.
.77 5.989 1.016
CSE12 I have the knowledge and skills
required to benefit from using
the university ICT facilities.
.86 5.891 .955
Perceived
Usefulness
(PU)
PU2 Using the ICT services available
at the university increases my
research productivity.
.81
5.555 1.270
.944
PU3 The current university ICT
system makes work more
interesting.
.87
5.217 1.324
PU4 ICT facilities at the university
improve the quality of the work
I do.
.87
5.401 1.289
PU5 Using the university ICT
facilities enhances my research
skills.
.88
5.335 1.285
PU6 Using the university ICT
facilities would make it easier
for me to find information.
.82 5.558 1.304
PU7 ICT services at the university
make information always
available to users.
.81 5.247 1.379
PU11 Using the university ICT
facilities provides me with the
latest information on specific
areas of research.
.74 5.628 1.226
159
Table 4.22, continued
Constructs Items Item Measure Loadings M SD Alpha
Perceived
Ease of Use
(PEU)
PEU2 I find it easy to access the
university research databases. .82 5.300 1.313
.830
PEU4 Interacting with the research
databases system requires
minimal effort on my part.
.78 4.931 1.366
PEU6 Interacting with the university
research databases system is
very stimulating for me.
.81 4.846 1.388
PEU7 I find it easy to select
articles/journals of different
categories using research
databases
(Education/Engineering/Busines
s, etc.).
.80 5.346 1.246
Intention to
Use (INT)
INT1 I will use the ICT facilities to
carry out my research activities. .86 6.070 1.038
.874
INT2 I will use computer applications
to present lecture materials in
class.
.81 6.280 1.001
INT3 I will use Microsoft office
applications to write my
research papers.
.72 6.189 1.157
INT5 I intend to use ICT facilities for
teaching, learning and research
purposes frequently.
.78 5.916 1.104
INT7 I will use the university research
databases to update myself on
the research areas I am
pursuing.
.65 5.861 1.092
Actual Use
(USE)
USE4 How often do you use the
following ICT facilities for
research and academic? – Web
browser (www).
.82 6.239 1.098
.802
USE5 How often do you use the
following ICT facilities for
research and academic? –
Search engine (e.g. Google,
Yahoo, etc.).
.85 6.383 .902
USE6 How often do you use the
following ICT facilities for
research and academic? –
Microsoft Office applications.
.68 6.308 .933
160
Table 4.22, continued
Constructs Items Item Measure Loadings M SD Alpha
Gratificatio
n (GRAT)
GRAT3 I am satisfied with the ease of
use of the ICT facilities. .86 5.346 1.294
.899
GRAT4 The university ICT facilities
have greatly affected the way I
search for information and
conduct my research.
.75 5.457 1.251
GRAT6 Overall I am satisfied with the
amount of time it takes to
complete my task.
.89 5.237 1.361
GRAT9 I am satisfied in using ICT
facilities for teaching, learning
and research.
.83 5.467 1.239
On the other hand, the overall fit statistics of the revised TAG model for
Chinese participants showed an adequate fit to the empirical data as indicated by the
following fit indices: χ2 (df = 369) = 840.593; p < 0.000; RMSEA = 0.081; CFI =
0.882; TLI = 0.870 as depicted Figure 4.10. In addition, the data of the revised TAG
model for Chinese participants revealed the standardized path coefficients as shown in
Figure 4.10. As anticipated, hypotheses H1 and H2 were supported, the findings
discovered that Chinese lecturers’ computer self-efficacy had a significant positive
direct influence on their perceived usefulness (β = 0.39, p<0.001) and perceived ease
of use (β = 0.44, p < .001). On the other hand, Chinese lecturers’ perceived ease of use
had a significant direct influence on their gratification (β = 0.29, p<0.001) and
intention to use (β = 0.46, p<0.001). In addition, hypotheses H3 and H4 were accepted,
the results of the revised TAG model for Chinese respondents exhibited that lecturers’
perceived usefulness had a significant direct influence on their gratification (β = 0.50,
p<0.001) but it did not exert any significant influence on intention to use (β = 0.05,
p≥0.05), while Chinese lecturers’ intention to use had a significant direct effect on
their actual use (β = 0.42, p<0.001) and gratification (β = 0.15, p<0.05) in using ICT
facilities for their teaching and research purposes in higher education.
161
Figure 4.10: The Revised TAG Model for Chinese Participants
Moreover, hypothesis H6 was also supported as the results of the revised TAG
model for Chinese participants indicated that Chinese lecturers’ computer self-
162
efficacy had a statistically significant and practically important indirect effect on their
intention to use mediated by perceived ease of use (Chi-square, χ2 = 3.767; p = 0.000).
However, Chinese lecturers’ computer self-efficacy did not confirm a statistically
significant and practically important indirect effect on their intention to use mediated
by perceived usefulness due to the insignificant path coefficient. Subsequently, The
results of the revised TAG model for Chinese respondents demonstrated that lecturers’
computer self-efficacy had a statistically significant and practically important indirect
effect on their gratification in using ICT facilities mediated by perceived ease of use
(Chi-square, χ2 = 2.920; p = 0.001). Similarly, Chinese lecturers’ computer self-
efficacy also had a statistically significant and practically important indirect influence
on their gratification mediated by perceived usefulness (Chi-square, χ2 = 3.349; p =
0.000) as conducted by the Sobel test and validated H7 (Sobel, 1982). Finally,
hypothesis H9 was supported as the results of the revised TAG model for Chinese
lecturers’ revealed that perceived ease of use had a statistically significant indirect
effect on gratification in using ICT facilities for their teaching and research purposes
in higher education mediated by intention to use (Chi-square, χ2 = 1.464; p = 0.071)
as conducted by the Sobel test (Sobel, 1982), while lecturers’ perceived usefulness did
not exert any significant indirect effect on their gratification mediated by intention to
use of ICT facilities due to the low path coefficient. Additionally, all the path
coefficients of the causal structure were practically important and statistically
significant, while the strongest value of the standardized path coefficient of Chinese
lecturers’ perceived usefulness on their gratification was 0.50. The data also explored
that Chinese lecturers’ perceived usefulness of ICT facilities was the most dominant
determinant in the revised TAG model for Chinese participants in directly affecting
the lecturers’ perceived usefulness of ICT facilities for their teaching and research
163
purposes in higher education.
Furthermore, the total standardized effect sizes of (i) computer self-efficacy →
perceived ease of use and perceived usefulness were .432 and .370 respectively, (ii)
perceived usefulness → intention to use, use and gratification were .061, .024
and .597 respectively, (iii) perceived ease of use → intention to use, use and
gratification were .500, .201 and .407 respectively, (iv) intention to use → use and
gratification were .402 and .161 respectively, and (v) computer self-efficacy →
intention to use, use and gratification were .238, .096 and .397 respectively.
Subsequently, the analysis of the revised TAG model for Chinese participants
exhibited that the exogenous variable (CSE) explained almost 19% and 15% of the
variance in perceived ease of use and perceived usefulness respectively, while the
exogenous variable (CSE) and the two mediator variables (PU and PEU) collectively
explained approximately 23% of the variability of Chinese lecturers’ intention to use
of ICT facilities for their teaching and research purposes in higher education.
Meanwhile, the exogenous variable (CSE) and the three mediator variables (PU, PEU
and INT) explained almost 18% of the variance of lecturers’ actual use of ICT
facilities. Finally, the exogenous variable (CSE) and the three mediator variables (PU,
PEU and INT) collectively explained around 47% of the variability of lecturers’
gratification in using ICT facilities for their teaching and research purposes in higher
education. The revised TAG model for Chinese participants confirmed adequate
convergent and discriminant validity (see Appendix I), with most of the indicators’
loadings showing a value greater than 0.70 as demonstrated in Table 4.23.
164
Table 4.23: Valid Items of the Revised TAG Model for Chinese Participants and
Their Corresponding Loadings, Means, Standard Deviations and Alpha Values
Constructs Items Item Measure Loadings M SD Alpha
Computer
Self-
efficacy
(CSE)
CSE4 I have the skills required to
communicate electronically
with my colleagues and
students.
.69 5.974 .988
.933
CSE7 I have the skills required to use
ICT facilities to enhance the
effectiveness of my teaching,
learning and research.
.82 5.931 .984
CSE9 I have the ability to navigate
my way through the ICT
facilities.
.87 5.714 1.051
CSE10 I have the skills required to use
the ICT facilities to enhance
the quality of my research
works.
.90 5.785 1.009
CSE11 I have the ability to save and
print journals/articles from the
research databases.
.77 5.989 1.016
CSE12 I have the knowledge and skills
required to benefit from using
the university ICT facilities.
.86 5.891 .955
Perceived
Usefulness
(PU)
PU2 Using the ICT services
available at the university
increases my research
productivity.
.81
5.555 1.270
.944
PU3 The current university ICT
system makes work more
interesting.
.87
5.217 1.324
PU4 ICT facilities at the university
improve the quality of the work
I do.
.87
5.401 1.289
PU5 Using the university ICT
facilities enhances my research
skills.
.88
5.335 1.285
PU6 Using the university ICT
facilities would make it easier
for me to find information.
.82 5.558 1.304
PU7 ICT services at the university
make information always
available to users.
.81 5.247 1.379
PU11 Using the university ICT
facilities provides me with the
latest information on specific
areas of research.
.74 5.628 1.226
165
Table 4.23, continued
Constructs Items Item Measure Loadings M SD Alpha
Perceived
Ease of Use
(PEU)
PEU2 I find it easy to access the
university research databases. .82 5.300 1.313
.830
PEU4 Interacting with the research
databases system requires
minimal effort on my part.
.78 4.931 1.366
PEU6 Interacting with the university
research databases system is
very stimulating for me.
.81 4.846 1.388
PEU7 I find it easy to select
articles/journals of different
categories using research
databases
(Education/Engineering/Busine
ss, etc.).
.80 5.346 1.246
Intention to
Use (INT)
INT1 I will use the ICT facilities to
carry out my research
activities.
.86 6.070 1.038
.874
INT2 I will use computer
applications to present lecture
materials in class.
.81 6.280 1.001
INT3 I will use Microsoft office
applications to write my
research papers.
.72 6.189 1.157
INT5 I intend to use ICT facilities for
teaching, learning and research
purposes frequently.
.78 5.916 1.104
INT7 I will use the university
research databases to update
myself on the research areas I
am pursuing.
.65 5.861 1.092
Actual Use
(USE)
USE4 How often do you use the
following ICT facilities for
research and academic? – Web
browser (www).
.82 6.239 1.098
.802
USE5 How often do you use the
following ICT facilities for
research and academic? –
Search engine (e.g. Google,
Yahoo, etc.).
.85 6.383 .902
USE6 How often do you use the
following ICT facilities for
research and academic? –
Microsoft Office applications.
.68 6.308 .933
166
Table 4.23, continued
Constructs Items Item Measure Loadings M SD Alpha
Gratificatio
n (GRAT)
GRAT3 I am satisfied with the ease of
use of the ICT facilities. .86 5.346 1.294
.899
GRAT4 The university ICT facilities
have greatly affected the way I
search for information and
conduct my research.
.75 5.457 1.251
GRAT6 Overall I am satisfied with the
amount of time it takes to
complete my task.
.89 5.237 1.361
GRAT9 I am satisfied in using ICT
facilities for teaching, learning
and research.
.83 5.467 1.239
4.10.1 Configural and Metric Invariance Analyses of the TAG Model
First, without constraining the structural paths, the results derived a baseline Chi-
square value. Next, the structural paths were constrained to be equal for the University
of Malaya and Jiaxing University. The results for both constrained and unconstrained
models were found to be consistent with the data as shown in Figure 4.11 and 4.12.
The analysis of this constrained TAG model produced another Chi-square value
(1768.175) and df = 746, which was then tested against the unconstrained Chi-square
value (1738.118) and df = 738 for statistically significant differences.
167
Figure 4.11: The Constrained TAG Model
168
Figure 4.12: The Unconstrained TAG Model
169
The invariance test between the University of Malaya in Malaysia and Jiaxing
University in China resulted in a statistically significant change in the Chi-square
value, Chi-square (df = 8) = 30.057, p > 0.05, while the critical value was at 15.507 as
shown in Table 4.24. This meant that cross-culture did interact with the exogenous
variables to influence the lecturers’ adoption and gratification in using ICT facilities in
higher education in Malaysia and China. The results of the invariance analysis of the
TAG model also demonstrated that the model was valid for measuring lecturers’
adoption and gratification in using ICT facilities in the diverse context of higher
education. It is, therefore, reasonable to conclude that cross-culture was a moderating
variable, whereas the TAG model works differently in cross-cultural settings.
Table 4.24: Results of Critical Value of Chi-squared
Models Chi squared df Critical
value
Chi squared
change
A Cross-cultural
validation between
University of
Malaya (Malaysia)
and Jiaxing
University (China)
Unconstrained 1738.118 738 15.507
(p > 0.05)
30.057
Constrained 1768.175 746
170
CHAPTER 5
DISCUSSION AND CONCLUSION
5.1 INTRODUCTION
The main objective of the study was twofold. The first was to develop and validate the
Technology Adoption and Gratification (TAG) Model to assess the lecturers’ adoption
and gratification in using ICT facilities for their teaching and research in higher
education which was achieved through the specific objectives 2 to 7. The second
purpose of the study was to evaluate the cross-cultural validation of the causal
structure of the TAG model. To achieve these objectives two versatile statistical tools
were applied. Firstly, the questionnaire’s reliability and validity were established
through a Rasch analysis. Secondly, the data were analysed using a three-step
Structural Equation Modeling (SEM) involving Confirmatory Factor Analysis (CFA)
to validate the measurement models, full-fledged structural model to estimate the
hypothesized and revised models of the lecturers’ adoption and gratification in using
ICT facilities and invariance analysis (metric and configural invariance analyses) to
examine the cross-cultural-invariant of the TAG model. The data were collected
through a questionnaire, and was administered to 396 lecturers who were selected
using stratified random sampling from two comprehensive public universities, namely,
the University of Malaya in Malaysia and Jiaxing University in China. This chapter
highlights the main findings of the study, discusses them in relation to theory and
prior studies, and concludes with recommendations and limitations for future research.
171
5.2 DISCUSSION OF THE FINDINGS
The results have theoretically contributed to the existing body of knowledge and
current understanding on the Technology Adoption and Gratification (TAG) model in
numerous ways. The model was developed and validated by incorporating three
models, namely, Technology Acceptance Model (Davis et al., 1989), Online Database
Adoption and Satisfaction Model (Islam, 2011a) and Technology Satisfaction Model
(Islam, 2014), to assess the lecturers’ adoption and gratification in using ICT facilities
for their teaching and research in higher education. The broad objective of this study
was achieved through specific objectives. To address the specific objectives 2 and 3,
the present study generated and tested three hypotheses (H2, H3 and H9) as discussed
below.
First, perceived ease of use had a statistically significant direct influence on
lecturers’ intention to use and gratification but not actual use, thereby validating the
second hypothesis (H2); a finding also consistent with earlier studies which
demonstrated the influential effect of perceived ease of use on users’ intention to use
computer-based systems (Chau, 1996; Karahanna & Straub, 1999; Venkatesh & Davis,
1996, 2000; Venkatesh & Morris, 2000). However, perceived ease of use was found to
be less influential and reliable (Ma & Liu, 2004; Wu et al., 2008; Huang, 2008).
Several researches prior to this also observed no significant relationship between the
perceived ease of use and behavioral intention (Jackson, Chow, & Leitch, 1997; Hu et.
al., 1999). In other studies, Tenopir (2003) found that ease of use was a major factor
influencing her respondents’ adoption of the online database. Outside the educational
context, Lee et al., (2006) showed that perceived ease of use significantly enhanced
consumer attitude and behavioral intention towards an online retailer, while Ramayah
and Lo (2007) demonstrated a significant effect of perceived ease of use on intention
172
to use an electronic resource planning system. Along this line, Lee and Lehto (2013)
and Islam (2011a) hypothesized that perceived ease of use will have a significant
effect on intention to use. Nevertheless, the findings of their study indicated that
perceived ease of use did not exert statistically significant effect on behavioral
intention to use.
On the one hand, Hanafizadeh et al., (2014) found that bank clients’ perceived
ease of use had a significant influence on intention to use mobile banking, while
Gultekin (2011) discovered that the effects of perceived ease of use on police officers’
intention to use the POLNET system were also statistically significant. Subsequently,
Stone and Baker-Eveleth (2013) confirmed that perceived ease of e-textbook use
positively influences behavioral intentions to purchase an e-textbook, however,
Martínez-Torres et al., (2008) discovered that perceived ease of use had a slightly
weaker influence compared to perceived usefulness on behavioural intention to use e-
learning tools. Many previous studies also demonstrated that perceived ease of use
was an antecedent factor that affected intention to accept English mobile learning
(Chang et al., 2012) as well as computer based assessment in higher education (Terzis
et al., 2013). Even though, Shroff et al., (2011) confirmed that students’ perceived
ease of use had significant influence on perceived usefulness and attitude towards
usage of portfolio system.
On the other hand, the present study inferred the possible relationship between
these two variables from several studies that investigated the relationship between
perceived ease of use and satisfaction (Lee & Park, 2008; Islam, 2014; Islam, 2011a;
Islam, 2011b; Huang, 2008; Shamdasani et al., 2008). The results of Szymanski and
Hise (2000) and Dabholkar and Bagozzi (2002) found convenience (a construct very
similar in concept to ease of use) to be an important factor in e-satisfaction.
173
Furthermore, Devaraj et al., (2002) exhibited that perceived ease of use was revealed
to be a significant determinant of satisfaction in the transaction cost analysis as well as
an important component of TAM in forming consumer attitude and satisfaction with
the electronic commerce channel, while Islam et al., (2015) confirmed that perceived
ease of use had a significant effect on students’ satisfaction in using online research
database in higher education. Thus, this study concluded that lecturers are more likely
to adopt ICT facilities for its ease of use coupled with their gratification in using it,
rather than its actual use.
Second, the results of the revised TAG model confirmed the significant direct
influence of perceived usefulness on lecturers’ intention to use and gratification but
not actual use of ICT facilities, hence validating the third hypothesis (H3). This was
congruent with the findings of prior studies (Wu et al., 2008; Davis, 1989; Guriting &
Ndubisi, 2006; 2004; Ramayah et al., 2005; Ramayah & Lo, 2007; Venkatesh, 2000;
Agarwal et al., 2003; Davis, 1993; Mathwick et al., 2001; Ahmad et al., 2010), which
found the significant influence of perceived usefulness on technology adoption.
Similarly, the impact of perceived usefulness on users’ intention to use different types
of computer-based and Internet-based systems has been comprehensively documented
in numerous studies (Ahmad et al., 2010; Davis, 1993; Guriting & Ndubisi, 2006;
Mathwick et al., 2001; Ramayah & Lo, 2007). Wu et al., (2008) corroborated these
findings on the impact of perceived usefulness by noting that it is a significant
determinant of science teachers’ intention to assimilate technology into their
instruction. Recent studies have also explored a significant effect of perceived
usefulness on users’ intention to use mobile learning (Hanafizadeh et al., 2014; Chang
et al., 2012), the POLNET system (Gultekin, 2011), e-learning tools (Martínez-Torres
et al., 2008) and e-portfolio systems (Shroff et al., 2011). In line with this, Terzis et al.,
174
(2013) showed that perceived usefulness demonstrated a significant influence on
students’ intention to use computer based assessment in higher education. According
to Davis et al., (1989) perceived usefulness was an important predictor, which
strongly influenced users’ intentions, while TAM (Davis, 1993) showed that perceived
usefulness had a significant direct effect on the actual use of information technology.
Additionally, the TSM (Islam, 2014) discovered that perceived usefulness had
a significant positive direct effect on students’ satisfaction in using wireless internet in
higher education. Concerning the Online Database Adoption and Satisfaction (ODAS)
Model, Islam, (2011a) confirmed that, perceived usefulness had a statistically
significant indirect influence on postgraduate students’ satisfaction mediated by
intention to use online research databases, while Islam et al., (2015) exhibited that
perceived usefulness was the most important construct of TSM in affecting
postgraduate students’ satisfaction in using online research databases in higher
education. Subsequently, Lee and Park (2008) indicated that perceived usefulness had
a significant effect on user satisfaction. Meanwhile, Islam (2011b) found that
perceived usefulness moderately influenced students’ satisfaction in using wireless
technology. However, its impact on satisfaction might be reflected through the
significant interrelationships with other latent constructs, namely, perceived ease of
use and computer self-efficacy. In addition to this, in extant literature, numerous
studies exist vis-à-vis the effect of perceived usefulness on consumer satisfaction
(Anderson et al., 1994). According to Huang (2008), perceived usefulness was found
to have an impact on consumer satisfaction mediated by their behavioral attitudes. On
the other hand, Clemons and Woodruff (1992) reported that, perceived value might
directly lead to the formation of overall feelings of satisfaction. This argument is
corroborated by McDougall and Levesque (2000) who indicated that, perceived value
175
is a significant driver of customer satisfaction, while Devaraj et al., (2002) revealed
that perceived usefulness was an important dimension in forming consumer attitude
and satisfaction with the electronic commerce channel. The discovery was that when
lecturers understand the benefits of ICT facilities, they are more likely to use the
service provided by the universities because they believe ICT facilities can enhance
their research and teaching ability and performance. Moreover, the data exhibited that
perceived usefulness was one of the important factors of the TAG model in affecting
lecturers’ adoption and gratification in using ICT facilities in higher education.
Third, the results of the TAG model depicted that perceived ease of use and
perceived usefulness had positive indirect influences on lecturers’ gratification in
using ICT facilities mediated by intention to use separately validated the ninth
hypothesis (H9). This concurred with an earlier study (Islam, 2011a), which found
perceived usefulness indirectly influenced students’ satisfaction in using online
research databases mediated by intention to use, while perceived ease of use had a
direct effect on satisfaction. However, in their study, Shamdasani et al., (2008)
observed that ease of use mediated by service quality, did not have a big influence on
satisfaction. In recent studies, perceived usefulness and perceived ease of use showed
a significant direct effect on students’ satisfaction in using wireless internet (Islam,
2014) and online research databases in higher education (Islam et al., 2015).
Therefore, it was found that when lecturers understand the ease of use and benefits of
ICT facilities, they are more likely to use the service provided by the universities
because they believe ICT facilities can increase their research and teaching
performance.
176
To address the specific objective 4, this study generated and tested four
hypotheses (H1, H6, H7 and H8) as discussed below.
Firstly, the hypothesis (H1) that computer self-efficacy would exert a positive
direct influence on lecturers’ perceived ease of use and perceived usefulness of the
ICT facilities in higher education stands validated as the findings were congruent with
prior studies which discovered the significant influence of computer self-efficacy on
perceived usefulness and perceived ease of use, respectively (Islam, 2014; Islam et al.,
2015). Moreover, the results also revealed that computer self-efficacy was the most
significant construct of the TAG model in directly affecting lecturers’ perceived
usefulness and perceived ease of use of the ICT facilities, while the technology
satisfaction model (TSM) confirmed that computer self-efficacy was the most
influential factor in measuring students’ satisfaction in using wireless internet (Islam,
2014) and the online research databases (Islam et al., 2015) in higher education.
Previous studies confirm propositions that self-efficacy influences the choice as to
whether or not one engages in a task, the effort made in performing it, and the level of
persistence required in accomplishing it (Bandura, 1977; Bandura & Schunk, 1981;
Delcourt & Kinzie, 1993). It influences how people feel, think, and behave (Bandura,
1986: 393). According to Ahmad et al., (2010), the inclusion of self-efficacy as an
intrinsic motivation construct offers ‘deeper and richer understanding of why and how
the technology is used’. However, they discovered that computer self-efficacy
indirectly influences faculty’s use of computer mediated technology through
perceived usefulness and intention to use. Islam (2011a & 2011b) demonstrated the
significant interrelationships among the exogenous variables which were measured in
the revised structural model, and it was found that the interrelationships between
computer self-efficacy, perceived ease of use and perceived usefulness were
177
statistically significant. YouTube users’ self-efficacy indicated a significant effect on
perceived usefulness (Lee & Lehto, 2013), whereas Stone and Baker-Eveleth (2013)
hypothesized that a mid-sized university students’ self-efficacy regarding the use of an
e-textbook significantly influences outcome expectancy/usefulness regarding e-
textbooks but it did not exert statistically significant effect on outcome
expectancy/usefulness in purchasing electronic textbooks. In a related study, Tezci
(2011) asserted that pre-service teachers’ self-efficacy levels regarding ICT usage in
education were found to be moderate. Thus, teachers’ self-efficacy or self-confidence
concerning ICT usage must be high in order for them to be highly encouraged in the
use of ICT, which in turn would lead to flourishing ICT integration. Amin (2007)
confirmed that university students’ computer self-efficacy of internet banking had a
significant effect on their perceived usefulness and perceived ease of use. Similarly,
Wong et al., (2012) revealed that computer teaching efficacy had a significant effect
on perceived usefulness, perceived ease of use and attitude towards computer use.
Nevertheless, the findings of their studies showed that computer teaching efficacy
only moderately effects perceived usefulness when compared to other constructs.
According to Zejno and Islam (2012), computer self-efficacy is one of the significant
predictors of postgraduate students’ ICT usage in higher education.
On the other hand, Terzis et al., (2013) explored that Greek and Mexican
university students’ computer self-efficacy had a positive influence on their perceived
ease of use, which shows that if a student knows how to use computers, perhaps
he/she will find it easy to use a computer based assessment that needs fundamental
information technology skills. In a study involving university students’ use of an
information system, Agarwal and Karahanna (2000) discovered that computer self-
efficacy was a key antecedent of perceived ease of use, while Wu et al., (2008) found
178
a statistically significant relationship between teachers’ CSE and their intention to use
ICT in teaching. More importantly, the present study discovered that lecturers’
adoption and gratification of ICT facilities does not depend only on perceived ease of
use and perceived usefulness, but also on their beliefs in their own ability to use it.
Thus, computer self-efficacy is acknowledged as a distinct antecedent of perceived
usefulness and perceived ease of use, which is an important intrinsic motivation factor
of technology adoption and gratification.
Secondly, the findings of the TAG model discovered that computer self-
efficacy had a significant indirect effect on lecturers’ intention to use of ICT facilities
mediated by perceived ease of use and perceived usefulness separately, thereby
validating the sixth hypothesis (H6). However, Ahmad et al., (2010) found that
computer self-efficacy indirectly influences faculty’s use of computer mediated
technology through perceived usefulness and intention to use. Prior studies also
demonstrated the influential effect of computer self-efficacy on technology adoption
and usage (Wu et al., 2008; Gong et al., 2004; Hu et al., 2003; Agarwal & Karahanna,
2000; Compeau et al., 1999; Ahmad et al., 2010; Compeau & Higgins, 1995;
Dabholkar & Bogazzi, 2002; Ellen et al., 1991). Stone and Baker-Eveleth (2013)
showed that computer self-efficacy positively influenced students’ attitudes regarding
the purchase of e-textbooks. Alenezi et al., (2010) confirmed that computer self-
efficacy significantly influenced students' intention to use e-learning. According to
Chang and Tung (2008), self-efficacy was an important factor for university students’
behavioural intention to use the online learning course websites, while Zejno and
Islam (2012) claimed that computer self-efficacy is one of the significant predictors of
postgraduate students’ ICT usage. In a recent study, Islam (2011a) demonstrated that
computer self-efficacy had a direct influence on postgraduate students’ intention to
179
use online research databases. In fact, the influence of this construct is evident in
many cases involving the employment of computer technology (Ball & Levy, 2008;
Charalambous & Papaioannou, 2010; Kenzie et al., 1994; Moos & Azevedo, 2009).
More importantly, the present study discovered that lecturers’ adoption and
gratification of ICT usage does not only depend on perceived ease of use and
perceived usefulness, but also on their beliefs in their own ability to use it. Thus,
computer self-efficacy is acknowledged as a distinct antecedent of intention to use,
which is an important intrinsic motivation factor of technology adoption and
gratification. As such, computer self-efficacy can indirectly influence lecturers’
intention to use ICT facilities in higher education.
Thirdly, the results of the TAG model demonstrated that computer self-
efficacy had a statistically significant indirect influence on lecturers’ gratification in
using ICT facilities mediated by perceived usefulness and perceived ease of use
separately, which validated the seventh hypothesis (H7). The results were consistent
with prior studies (Islam, 2014; Islam et al., 2015) which exhibited that computer self-
efficacy was the most dominant factor of the TSM and it also had an indirect
significant influence on satisfaction mediated by perceived ease of use and perceived
usefulness, respectively. The ODAS model (Islam, 2011a) revealed the significant
interrelationships among the exogenous variables, namely, computer self-efficacy,
perceived ease of use and perceived usefulness in measuring postgraduate students’
adoption and satisfaction in using online research databases. Similarly, Islam (2011b)
also found significant positive interrelationships among the three exogenous variables,
namely, computer self-efficacy, perceived ease of use and perceived usefulness in
assessing students’ satisfaction in using wireless internet in higher education.
However, computer self-efficacy had a significant indirect effect on students’
180
satisfaction mediated by intention to use (Islam, 2011a, p. 50). Thus, the findings
suggested that lecturers the ICT usage for its self-efficacy because of their beliefs in
their computer ability.
Finally, in regards to the indirect effect of computer self-efficacy on lecturers’
use of ICT facilities mediated by perceived ease of use and perceived usefulness
separately, the findings had an insignificant path co-efficient, thereby invalidating the
eighth hypothesis (H8). However, several studies explored the direct relationships
between computer self-efficacy, perceived ease of use as well as perceived usefulness
(Islam, 2014; Islam et al., 2015). As a result, this hypothesis was dropped in the
revised model.
To address the specific objective 5, the present study generated and tested two
hypotheses (H4 and H10) as discussed below.
First, as revealed by the significant direct path coefficient, intention to use
significantly influenced lecturers’ gratification and actual use of ICT facilities in
higher education, thereby validating the fourth hypothesis (H4). This finding
corroborated the results of the ODAS model (Islam, 2011a), which demonstrated that
intention to use had a positive direct effect on students’ satisfaction in using online
research databases in higher education. This study had inferred the possible
relationship between intention to use and gratification from several studies
investigating the relationship between computer self-efficacy and intention to use (Wu
et al., 2008; Ahmad et al., 2010) and intention to use with individual satisfaction
(Huang, 2008; Shamdasani et al., 2008). On the other hand, Davis et al., (1989)
hypothesized and confirmed that behavioral intention was the key determinant of
actual use of computer technology. Their findings also stated that behavior should be
181
predictable from measures of users’ behavioral intention and that if there were any
additional constructs that effected user behavior, they do so indirectly by influencing
behavioral intention. However, Davis (1993) ignored the behavioral intention, which
was the most important factor of the TAM, whereas attitude toward using revealed a
significant direct effect on actual system use of information technology. Furthermore,
the data for this study explored that lecturers’ behavioral intention to use was one of
the dominant factors in affecting their adoption and gratification in using ICT facilities
in teaching and research.
Second, the indirect influence of intention to use on lecturers’ gratification
mediated by actual use of ICT facilities in higher education was found to be a
statistically insignificant path coefficient, thereby invalidating the tenth hypothesis
(H10). However, Islam (2011a) investigated the direct relationship between intention
to use and satisfaction and intention to use with individual satisfaction (Huang, 2008;
Shamdasani et al., 2008). In addition, the findings of the TAG model also confirmed
that lecturers’ intention to use of ICT facilities was one of the vital factors in
evaluating technology adoption and gratification in higher education.
To address the specific objective 6, this study generated and tested one
hypothesis (H5) as discussed below.
In regards to the effect of actual use on lecturers’ gratification in using ICT
facilities, it had an insignificant path co-efficient, thereby invalidating the fifth
hypothesis (H5). However, user satisfaction was found to be positively related to
perceived market performance (Lee & Park, 2008). According to Shipps and Phillips
(2013), attitude toward the social networking tool showed a significant effect on user
satisfaction, while Devaraj et al., (2002) indicated general support for consumer
182
satisfaction as a determinant of channel preference. Thus, the finding of this study
suggested that lecturers adopted and gratified the ICT facilities not only for its actual
usage, but rather for its ease of use, usefulness, intention to use and because of their
beliefs in their computer self-efficacy.
To address the specific objective 7, it was not necessary to generate a
hypothesis, as the endogenous variable was assessed applying Confirmatory Factor
Analyses (CFA) as well as Structural Equation Modeling (SEM) like other objectives.
For instance, the objectives 2 to 7 were firstly evaluated using CFA to confirm
lecturers’ perceptions in using ICT facilities for their teaching and research purposes
in higher education in terms of perceived ease of use, perceived usefulness, computer
self-efficacy, intention to use, actual use and gratification. Later on, to examine the
causal relationships and the effect of perceived ease of use, perceived usefulness,
computer self-efficacy, intention to use and actual use on gratification, the present
study tested nine hypotheses as discussed above. However, lecturers’ gratification did
not have any influence on other latent constructs, so the hypothesis was not required
to test for this objective. Besides this, the results of the one-factor measurement model
of gratification were analyzed using CFA and confirmed that lecturers’ views on the
gratification in using ICT facilities. The one-factor measurement model of
gratification explained almost 64.1% of the variability of lecturers’ gratification in
using ICT facilities for their teaching and research purposes in higher education.
To address the specific objective 8, the present study generated and tested one
hypothesis (H8) as discussed below.
This study exhibited that there was a cross-cultural invariant of the causal
structure of the TAG model, thereby validating the eighth hypothesis (H8). This
183
meant that cross-culture did interact with the exogenous and mediating variables to
influence lecturers’ adoption and gratification in using ICT facilities in higher
education in both Malaysia and China. The findings of this study confirmed that there
were significant differences between Malaysian and Chinese lecturers in using ICT
facilities for their teaching and research purposes in higher education. This was
consistent with a prior study (Teo et al., 2008), which found that there were
significant dissimilarities between Malaysian and Singaporean pre-service teachers in
terms of their computer attitude, perceived ease of use and perceived usefulness of
technology. Chong et al., (2012) explored that consumers’ intention to adopt mobile
commerce was affected by the different cultures, namely, Malaysian and Chinese. On
the other hand, Terzis et al., (2013) discovered that the Computer Based Assessment
Acceptance Model (CBAAM) was valid for both countries (Greece & Mexico) in
general; however, there were some distinctions due to the culture. Moreover, Singh et
al., (2006) showed that the causal relationships between the extended TAM factors,
namely, perceived ease of use, perceived usefulness, cultural adaptation, attitude and
behavioral intention to use international websites presented strong evidence for the
viability of TAM in elucidating website usage in Brazil, Germany, and Taiwan.
Along this line, Al-Gahtani et al., (2007) examined the similarities and dissimilarities
between Saudi Arabian and North American validations of UTAUT with regards to
cultural diversities that influenced the users’ acceptance of IT in these two societies.
Similarly, Keil and Brenner (1997) demonstrated that TAM was influenced by the
three different cultures in the United States, Japan, and Switzerland. According to
Huang et al., (2003), the validity of TAM was expanded by assessing technology
acceptance beliefs in a developing country with Chinese culture.
184
5.3 CONCLUSION
As information technology is increasingly used in education, its integration in
teaching, learning and research has had an immense significance in fostering
technology-based education among the lecturers of universities. The findings
contributed to a better understanding of the Technology Adoption and Gratification
(TAG) model, which was developed and validated for assessing lecturers’ adoption
and gratification in using ICT facilities for their teaching and research purposes in
higher education among the two comprehensive public universities, namely, the
University of Malaya in Malaysia and Jiaxing University in China. To develop and
validate the TAG model, data were gained through a survey questionnaire. The
psychometric properties of the TAG models’ reliability and validity were performed
by a Rasch model using Winsteps version 3.49. The advantage of using a Rasch model
was that it measured not only items’ reliability and validity but also persons’
reliability and validity, while most of the studies focused only on the items’ reliability
and validity. This study discovered that the TAG model was integrated by six valid
measurement models of the latent constructs, namely, computer self-efficacy (CSE),
perceived usefulness (PU), perceived ease of use (PEU), intention to use (INT), actual
use (USE) and gratification (GRAT) as tested individually by the Confirmatory Factor
Analysis (CFA) to obtain specific objectives. To test all the hypotheses, the structural
equation modeling (SEM) was applied to develop and validate the TAG model as well
as its causal relationships among the constructs. The findings of the TAG model
revealed that lecturers’ computer self-efficacy had a statistically significant direct
influence on perceived ease of use and usefulness of ICT facilities in higher education
for their teaching and research. On the other hand, lecturers’ perceived ease of use and
perceived usefulness had a significant direct influence on their gratification in using
185
ICT facilities in higher education. Subsequently, perceived usefulness and perceived
ease of use were found to have a significant direct influence on lecturers’ intention to
use ICT facilities for their teaching and research. Similarly, lecturers’ intention to use
had a statistically significant direct effect on their gratification as well as actual use of
ICT facilities. Moreover, computer self-efficacy had a significant indirect influence on
gratification mediated by perceived ease of use and perceived usefulness, respectively.
Meanwhile, computer self-efficacy had also a significant indirect influence on
lecturers’ intention to use mediated by perceived ease of use and perceived usefulness,
respectively. Lastly, lecturers’ perceived ease of use and perceived usefulness had a
significant indirect effect on their gratification in using ICT facilities in higher
education mediated by intention to use. The data also exhibited that computer self-
efficacy moderately more influential factor in the revised TAG model in directly
affecting the lecturers’ perceived usefulness and perceived ease of use as well as
indirectly influencing the lecturers’ intention to use and gratification in using ICT
facilities in higher education for their research and teaching purposes. In addition, the
analysis demonstrated that the exogenous variable (CSE) explained almost 28% and
29% of the variance in perceived ease of use and perceived usefulness respectively,
while the exogenous variable (CSE) and the two mediator variables (PU and PEU)
collectively explained approximately 25% of the variability of lecturers’ intention to
use of ICT facilities in higher education. Similarly, the exogenous variable (CSE) and
the three mediator variables (PU, PEU and INT) explained almost 20% of the variance
of lecturers’ actual use of ICT facilities. This implies that lecturers’ use of ICT
facilities are found to be underutilized. Finally, the exogenous variable (CSE) and the
three mediator variables (PU, PEU and INT) collectively explained approximately
56% of the variability of lecturers’ gratification in using ICT facilities in higher
186
education. The revised TSM model exhibited adequate convergent and discriminant
validity, with all loadings indicating a value greater than 0.70 as shown in Figure 4.15.
Several findings of the TAG model were consistent with the Technology
Satisfaction Model (TSM), Technology Acceptance Model (TAM) and Online
Database Adoption and Satisfaction (ODAS) model and contributed that the TAG is
able to evaluate lecturers’ adoption and gratification in using ICT facilities in teaching
and research. However, TSM focused only on students’ satisfaction in using wireless
internet (Islam, 2014) and online research databases (Islam et al., 2015) in higher
education but ignored its adoption. Similarly, TAM, developed by Davis (1989),
ignored the issues of computer self-efficacy and satisfaction. In general, TAM is used
for measuring the acceptance of technology (Islam, 2014). On the other hand, Islam
(2011a) developed and validated the ODAS model by incorporating two intrinsic
motivation attributes, namely, computer self-efficacy and satisfaction into the original
TAM. However, the ODAS model overlooked an extrinsic motivation component,
namely, actual use which is an important factor for assessing technology acceptance
or adoption as suggested by prior studies. As such, this study combined TSM, TAM
and the ODAS model to develop and validate the TAG model which would be able to
examine any kind of technology adoption and gratification simultaneously.
The findings brought forth some noteworthy information about lecturers’
adoption and gratification of the universities’ ICT facilities. First, they used the ICT
facilities largely because they found it beneficial and easy to use for their teaching and
research purposes. Moreover, they also had strong self-efficacy in using ICT facilities
in higher education. The ability (computer self-efficacy) appeared to be a stronger
factor than perceived usefulness and perceived ease of use in influencing their
adoption and gratification of ICT. As the universities provide a library information on
187
the use of ICT facilities, lecturers’ lack of ICT skills and efficacy is most likely not a
real threat to its adoption and gratification. Lecturers will persist in the use of ICT
facilities if the benefits are much greater than the difficulties encountered. To increase
the usefulness of ICT facilities, the university authorities should reexamine the utility
of resources because teaching, learning and research are facilitated only if lecturers
have access to the facilities.
Second, ease of use is a determinant for lecturers’ gratification, intention to use
and actual use of ICT, which directly and indirectly influences their adoption and
gratification in using it. In other words, the university authorities can increase
lecturers’ gratification by enhancing the user-friendliness of their ICT facilities in
terms of the user interface, hyperlinks and search facilities. Gratification may also be
directly influenced by the users’ ability to retrieve desired information from electronic
resources. To address difficulties in retrieving information, authorities should focus
their training on database-specific search skills, rather than just increasing the
awareness of their existence and utilities. In fact, the ability to retrieve information
from different databases could be a stronger predictor of adoption and gratification,
compared to computer self-efficacy because a discrepancy usually exists between
users’ beliefs in their ability to retrieve information from electronic systems and their
actual ability to do it.
Evidenced from the revised TAG model showed that the direct and indirect
effect of lecturers’ computer self-efficacy on perceived ease of use, perceived
usefulness, intention to use and gratification in using ICT facilities was limited to their
requisite skills to communicate electronically with colleagues and students, use ICT
facilities to enhance the effectiveness of teaching, learning and research, improve the
quality of research works in using ICT facilities, navigate their way through ICT
188
facilities, save and print journals or articles from the research databases. In addition,
they also had the knowledge and skills to benefit from using the university ICT
facilities. However, Islam (2011a & 2014) discovered that postgraduate students
articulated obstacles related to accessing journals from the online research databases.
The influence of lecturers’ perceived usefulness on intention to use,
gratification and actual use of ICT facilities in teaching and research manifested in
lecturers increasing research productivity, making work more interesting, improving
the quality of the work they do, enhancing research skills, making information easier
to find, making information readily available, and providing the latest information on
specific areas of research. Notwithstanding, Islam (2011a) revealed that postgraduate
students articulated their limitations on retrieving current information on a particular
area of research and enhancing their knowledge and research skills in using the online
research databases. In a related study, Islam (2014) found that students expressed their
constraints vis-à-vis slow wireless internet speeds compared to the LAN as well as
difficulty in getting wireless internet access from anywhere and anytime in the campus.
Moreover, the convenience of access, in particular, has been frequently cited as a
major advantage by users of electronic journals in multiple studies (Maughan, 1999;
Tenner & Yang, 1999; Hiller, 2002). In Liew et al., (2000), the benefits of electronic
resources were found to be the facility to link to additional information, the ability to
browse, and the currency of materials. Other benefits and advantages included robust
search capabilities, hyperlinks to outside content, efficient acquisition of information
(Dillon & Hahn, 2002; Ray & Day, 1998), time saving factors, and print and save
capabilities (Togia & Tsigilis, 2009). These benefits and advantages were realized all
the more by students when their learning and academic pursuits became greatly
enhanced.
189
The influence of lecturers’ perceived ease of use on intention to use, actual use
and gratification in using ICT facilities was exhibited in terms of their easy access to
the university research databases, the minimal effort required in interacting with the
research databases system, and it being a stimulating interaction. It was also evident
that lecturers found it easy to select articles/journals of different categories using
university research databases. However, Islam (2011a) demonstrated that some
barriers still persisted. These were manifested by the postgraduate students’
inadequacy in choosing articles or journals from the online research databases,
requiring a lot of mental effort to download research materials, and the difficulty in
searching for research materials from the online research databases. In addition, it was
not easy to become skillful in using online research databases according to the
students. Subsequently, the findings of the TSM (Islam, 2014) indicated that students
showed that some obstacles still lingered; these were noticeable in the inadequate
speed experienced by the students in downloading the materials from the wireless
internet. Moreover, it was not easy to connect and register for wireless facilities as
perceived by the students.
The significant effect of intention to use on actual use and gratification in
using ICT facilities was captured in terms of their intention to carry out research
activities, use computer applications to present lecture materials in class, use
Microsoft office applications to write research papers, use ICT facilities for teaching,
learning and research purposes frequently, and use the university research databases to
update themselves on the research areas they are pursuing. Nonetheless, prior study
indicated that postgraduate students’ intention to use online research databases was
limited to their wiliness to carry out research activities and write theses and journals
(Islam, 2011a). Thus, the results of TAG model confirmed lecturers’ adoption and
190
overall gratification in using ICT facilities for their teaching, learning and research
purposes in higher education. However, the actual use was found to be underutilized,
especially by the lecturers of the University of Malaya in Malaysia and Jiaxing
University in China.
Additionally, the findings of this study also revealed that the TAG model was
able to measure lecturers’ adoption and gratification in using ICT facilities at two
comprehensive public universities in Malaysia and China. However, the TAG model
works differently in cross-cultural settings. This meant that cross-culture did interact
with the exogenous and mediating variables to influence the lecturers’ adoption and
gratification in using ICT facilities in higher education. It is, therefore, reasonable to
conclude that cross-culture was a moderating variable for the TAG model. This is a
research area worth looking into by the university authorities, researchers,
academicians and administrations in higher education.
Finally, the findings have several important implications for ICT providers in
universities worldwide. The rapid uptake of the technology has made ICT facilities
indispensable tools for research, teaching and learning, and considering their
paramount importance in the contemporary context of university education, it would
do the providers and universities well to address issues of ICT usefulness, ease of use,
efficacy, use and gratification. The benefits gained from ICT use would certainly
increase lecturers’ gratification, therefore resulting in their continued adoption of the
facility. The present study applies systematic research approaches of model
development and validation by using a Rasch model and a structural equation
modeling so that researchers from other countries might benefit from applying these
techniques in their future studies. Researchers could also benefit from this study by
knowing the constructs of the TAG model, namely, computer self-efficacy, perceived
191
usefulness, perceived ease of use, intention to use, actual use and gratification of ICT
facilities, which could be applied to their further studies. The findings of this study
confirmed the cross-cultural validation of the TAG model in assessing technology
adoption and gratification in higher education, whereas existing models such as TAM
and Diffusion and Innovation theory are only allowing researchers to measure either
technology acceptance or adoption. As a result, the TAG model could be a unique
contribution to researchers from other countries.
Albeit the research methodology appears valid, reliable and eliciting, only two
higher educational institutions with a small sample size are used. It is therefore, the
research may lack in a variation of respondents. A wider research using more
organizations with more samples may help to validate the (TAG) model with a greater
acceptability. Critics may generate the questions on its validity showing less number
of samples as the excuse. However, the usage of ICT and its users hold a global
pattern; thus, this model has a great potential to work globally regardless of
geographical, economical, social and cultural patterns in a particular country or region.
Further research in this area should examine the true utility of all the services provided
in a university before money is invested in their purchase and installation.
Furthermore, future research could investigate additional constructs, namely, teaching
effectiveness, and knowledge and skill acquisition in the classroom to included into
the TAG model which may foster a better understanding of it. Additional studies
could also be done upon this study by applying the TAG model in cross-cultures to
measure new areas in detail such as digital library services, online education, online
shopping, mobile learning, social networks, and any other innovative technological
services.
192
However, further research on cross-validation of the TAG model could be
done to explain the moderating effects of various demographic attributes on lecturers’
or students’ adoption and gratification. The gender, faculties, and nationalities of the
teachers and students might be explored in this regard.
193
REFERENCES
Adamson, I., & Shine, J. (2003). Extending the new technology acceptance model to
measure the end user information systems satisfaction in a mandatory
environment: A bank's treasury. Technology Analysis & Strategic Management,
15(4): 441-455.
Afshari, M., Bakar, K.A., Luan, W.S., & Siraj, S. (2012). Factors affecting the
transformational leadership role of principals in implementing ict in schools.
The Turkish Online Journal of Educational Technology, 11 (4): 164-176.
Agarwal, R., & Karahanna, E. (2000). Time flies when you’re having fun: cognitive
absorption and beliefs about information technology uses. MIS Quarterly,
24(4): 665-694.
Agarwal, R., & Prasad, J. (1998). A conceptual and operational definition of personal
innovativeness in the domain of information technology. Information Systems
Research, 9(2): 204-215.
Agarwal, S., Erramilli, M., & Chekitan, S. (2003). Market orientation and
performance in service firms: role of innovation. Journal of Service Marketing,
17(1): 68-82.
Ahmad, T.B.T., Basha, K.M., Marzuki, A.M., Hisham, N.A., & Sahari, M. (2010).
Faculty’s acceptance of computer-based technology: cross-validation of an
extended model. Australasian Journal of Educational Technology, 26(2): 268-
279.
Al-Jabri, I.M., & Roztocki, N. (2010). Adoption and Use of Information Technology
in Mandatory Settings: Preliminary Insights from Saudi Arabia. Proceedings of
the Sixteenth Americas Conference on Information Systems, Lima, Peru,
August 12th
-15th
. Available at:
http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1965269
Aiken, R.L. (1993). Personality, theories, research, and applications. NJ: Prentice-
Hall Inc.
194
Alenezi, A.R., Karim, A.M.A., & Veloo, A. (2010). An empirical investigation into
the role of enjoyment, computer anxiety, computer self-efficacy and internet
experience in influencing the students' intention to use e-Learning: A case study
from Saudi Arabian governmental universities. The Turkish Online Journal of
Educational Technology, 9(4): 22-34.
Al-Gahtani, S.S., Hubona G.S., & Wang J. (2007). Information technology (IT) in
Saudi Arabia: Culture and the acceptance and use of IT. Information and
Management, 44(8): 681–691.
Al-Ruz, J.A., & Khasawneh, S. (2011). Jordanian pre-service teachers' and technology
integration: A human resource development approach. Educational Technology
& Society, 14(4): 77–87.
Aljuaid, N.M.F., Alzahrani, M.A.R., & Islam, A.Y.M.A. (2014). Assessing mobile
learning readiness in Saudi Arabia higher education: An empirical study. The
Malaysian Online Journal of Educational Technology, 2(2): 1-14.
Amin, H. (2007). Internet banking adoption among young intellectuals. Journal of
Internet Banking and Commerce, 12(3): 1-13.
Anderson E.W., Fornell, C., & Lehmann, D.R. (1994). Customer satisfaction, market
share and profitability. Journal of Marketing, 58(3): 53-66.
Azizan, F.Z. (2010). Blended Learning in Higher Education Institute in Malaysia.
Paper presented at the Regional Conference on Knowledge Integration in ICT,
Selangor, Malaysia. Available at:
http://library.oum.edu.my/oumlib/sites/default/files/file_attachments/odl-
resources/4334/blended-learning.pdf
Babbie, E. (2001). The practice of social research, Belmont, CA: Wadsworth
Thomson Learning.
Ball, D., & Levy, Y. (2008). Emerging educational technology: assessing the factors
that influence instructors’ acceptance in information systems and other
classrooms. Journal of Information Systems Education, 19(4): 431-443.
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change.
Psychological Review, 84: 191-125.
195
Bandura, A. (1986). Social foundations of thought and action: A social cognitive
theory. Englewood Cliffs, New Jersey: Prentice Hall.
Bandura, A., & Schunk, D.H. (1981). Cultivating competence, self-efficacy and
intrinsic interest through self-motivation. Journal of Personality and Social
Psychology, 41: 586-598.
Barkhi, R., Belanger, F., & Hicks, J. (2008). A model of the determinants of
purchasing from virtual stores. Journal of Organizational Computing and
Electronic Commerce, 18: 177–196. DOI: 10.1080/10919390802198840
Bate, F. (2010). A bridge too far? Explaining beginning teachers’ use of ICT in
Australian schools. Australasian Journal of Educational Technology, 26(7):
1042-1061.
Blackwell, C.K., Lauricella, A.R., Wartella, E., Robb. M., & Schomburg, R. (2013).
Adoption and use of technology in early education the interplay of extrinsic
barriers and teacher attitudes. Computers & Education, 69: 310-319.
Bond, T.G., & Fox, C.M. (2001). Applying the Rasch model: Fundamental
measurement in the human science. Lawrence Erlbaum, Mahwah, NJ.
Bringula, R.P., & Basa, R.S. (2011). Factors affecting faculty web portal usability.
Educational Technology & Society, 14 (4): 253–265.
Buabeng-Andoh, C. (2012). Factors influencing teachers’ adoption and integration of
information and communication technology into teaching: A review of the
literature. International Journal of Education and Development using
Information and Communication Technology, 8(1): 136-155.
Byrne, B.M. (2000). Structural equation modeling with Amos: Basic concepts,
applications and programming. Mahwah, NJ: Lawrence Erlbaum.
Clemons, D.S., & Woodruff, R.B. (1992). Broadening the view of consumers (dis)
satisfaction: A proposed means–end disconfirmation model. In Marketing theory
and applications, Allen, C. T., Madden, T. J., Shimp, T. A., Howell, R. D.,
Zinkhan, G. M., Heisley, D. D., 1992. Ed. Chicago: American Marketing
Association (pp 209-216).
196
Chang, C.C., Yan, C.F., & Tseng, J.S. (2012). Perceived convenience in an extended
technology acceptance model: Mobile technology and english learning for
college students. Australasian Journal of Educational Technology, 28(5): 809-
826. http://www.ascilite.org.au/ajet/ajet28/chang-cc.html
Chang, S., & Tung, F.C. (2008). An empirical investigation of students’ behavioural
intentions to use the online learning course websites. British Journal of
Educational Technology, 39(1): 71-83. doi:10.1111/j.1467-8535.2007.00742.x
Chau, P.Y.K. (1996). An empirical assessment of a modified technology acceptance
model. Journal of Management Information Systems, 13(2): 185-204.
Charalambous, K., & Papaioannou, F. (2010). The public primary school principals’
self-perceived competence and use of ICT for personal, teaching and
administrative purposes. In A. Jimoyiannis (ed.), Proceedings of the 7th Pan-
Hellenic Conference with International Participation: ICT in Education, 1, 177-
184, University of Peloponnese, Korinthos, Greece, 23-26 September. Available
at: http://www.etpe.gr/custom/pdf/etpe1582.pdf
Chitungo, S.K., & Munongo, S. (2013). Extending the technology acceptance model
to mobile banking adoption in rural Zimbabwe. Journal of Business
Administration and Education, 3(1): 51-79.
Chong, A.Y., Chan, F.T.S., & Ooi, K. (2012). Predicting consumer decisions to adopt
mobile commerce: Cross country empirical examination between China and
Malaysia. Decision Support Systems, 53(1): 34–43.
doi:10.1016/j.dss.2011.12.001
Churchill, G. A., & Suprenant, C. (1982). An investigation into the determinants of
customer satisfaction. Journal of Marketing Research, 19: 491-504.
Compeau, R.D., & Higgins, A.C. (1995). Computer Self-efficacy: Development of a
Measure and Initial Test. MIS Quarterly, 19(2): 189-211.
Compeau, R.D., Higgins, A.C., & Huff, S. (1999). Social Cognitive Theory and
Individual Reactions to Computing Technology: A longitudinal Study. MIS
Quarterly, 23(2): 145-158.
197
Dabholkar, P.A., & Bogazzi, R. (2002). An attitudinal model of technology-based
self-service: Moderating effect of consumer traits and situational factors.
Journal of Academy of Marketing Science, 30(3): 184–202.
Davis, F.D. (1989). Perceived usefulness, perceived ease of use and user acceptance
of information technology. MIS Quarterly, 13(3): 319-340.
Davis, F.D. (1993). User acceptance of information technology: system characteristics,
user perceptions and behavioral impacts. International Journal of Man-
Machine Studies, 38(3): 475-487.
Davis, F. Bagozzi, R.P., & Warshaw, P.R. (1989). User acceptance of computer-
technology: a comparison of two theoretical models. Management Science,
38(8): 982-1003.
Delcourt, M.A.B., & Kinzie, M.B. (1993). Computer technologies in teacher
education: the measurement of attitude and self-efficacy. Journal of Research
and Development in Education, 27(1): 34-41.
Devaraj, S., Fan, M., & Kohli, R. (2002). Antecedents of b2c channel satisfaction and
preference: Validating e-commerce metrics. Information Systems Research
13(3):316-333. http://dx.doi.org/10.1287/isre.13.3.316.77
Dillon, A., & Morris, G.M. (1996). User acceptance of information technology
theories and models. In M. Williams (ed) Annual Review of Information
Science and Technology, 31, Medford, NJ: Information Today, pp. 3-32.
Dillon, I.F., & Hahn, K. (2002). Are researchers ready for the electronic only journal
collection? Libraries and the Academy, 2: 375-390.
Du Plessis, A., & Webb, P. (2012). Digital immigrant teacher perceptions of an
extended Cyberhunt strategy. Australasian Journal of Educational Technology,
28(2): 341-363. http://www.ascilite.org.au/ajet/ajet28/duplessis.html
Eastin, M.S., & LaRose, R. (2000). Internet self-efficacy and the psychology of the
digital divide. Journal of Computer-Mediated Communication, 6(1).
doi: 10.1111/j.1083-6101.2000.tb00110.x
198
Elen, J., Clarebout, G., Sarfo, F.K., Louw, L.P.F., Pöysä-Tarhonen, J., & Stassens, N.
(2010). Computer and information and communication technology: Students'
culture specific interpretations. Educational Technology & Society, 13 (4): 227–
239.
Ellen, P.S., Bearden, W.O., & Sharma, S. (1991). Resistance for technological
innovations: An examination of the role of self-efficacy and performance
satisfaction. Journal of Academic of Marketing Science, 19(4): 297-307.
Eyal, L. (2012). Digital assessment literacy—the core role of the teacher in a digital
environment. Educational Technology &Society, 15 (2): 37–49.
Ferdousi, B. J. (2009). A Study of Factors that Affect Instructors’ Intention to Use E-
Learning Systems in Two-Year Colleges. PhD Dessertation. Graduate School of
Computer and Information Sciences, Nova Southeastern University.
Fishbein, M. (2000). The role of theory in HIV prevention. AIDS Care, 12(3), 273–
278.
Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention and behavior: an
introduction to theory and research. MA: Addison-Wesley.
Fu, J.S. (2013). ICT in education: A critical literature review and its implications.
International Journal of Education and Development using Information and
Communication Technology, 9(1): 112-125.
Gibson, S.G., Harris, M.L., & Colaric, S.M. (2008). Technology acceptance in an
academic context: Faculty acceptance of online education, Journal of Education
for Business, 83(6): 355-359. http://dx.doi.org/10.3200/JOEB.83.6.355-359
Gibson, P.A., Stringer, K., Cotton, S.R., Simoni, Z., O’Neal, L.J., & Howell-Moroney,
M. (2014). Changing teachers, changing students? The impact of a teacher-
focused intervention on students’ computer usage, attitudes, and anxiety.
Computers & Education, 71: 165-174.
Goktas, Y., Gedik, N., & Baydas, O. (2013). Enablers and barriers to the use of ICT in
primary schools in Turkey: A comparative study of 2005–2011. Computers &
Education, 68: 211-222. http://dx.doi.org/10.1016/j.compedu.2013.05.002
199
Gong, M., Xu, Y., & Yu, Y. (2004). An enhanced technology acceptance model for
web-based learning. Journal of Information Systems Education, 15(4): 365-373.
Groff, J., & Mouza, C. (2008). A framework for addressing challenges to classroom
technology use. Association for the Advancement of Computing in Education
Journal, 16(1): 21-46.
Gultekin, K. (2011). Technology acceptance and the effect of gender in the Turkish
national police: The Case of the polnet system. Turkish Journal of Police
Studies, 13(3): 61-80.
Guo, Z., & Stevens, K.J. (2011). Factors influencing perceived usefulness of wikis for
group collaborative learning by first year students. Australasian Journal of
Educational Technology, 27(2): 221-242.
http://www.ascilite.org.au/ajet/ajet27/guo.html
Guriting, P., & Ndubisi, N. (2006). Borneo online banking: evaluating customer
perceptions and behavioral intention. Management Research News, 29(1/2): 6-
15.
Hanafizadeh, P., Behboudi, M., Koshksaray, A.A., & Tabar, M.J.S. (2014). Mobile-
banking adoption by Iranian bank clients. Telematics and Informatics, 31: 62–78.
http://dx.doi.org/10.1016/j.tele.2012.11.001
Havelka, D. (2003). Predicting software self-efficacy among business students: A
preliminary assessment. Journal of Information Systems Education, 14(2): 145-
152.
Hair, Jr. J.F., Black, W.C., Babin, B.J., & Anderson, R.E. (2010). Multivariate data
analysis a global perception: Pearson Education, Inc., Upper Saddle River,
New Jersey.
Hiller, S. (2002). How different are they? A comparison by academic area of library
use, priorities, and information needs at the University of Washington. Issues in
Science and Technology Librarianship. Available at: http://istl.org/02-
winter/article1.html
200
Hong, K.S., & Songan, P. (2011). ICT in the changing landscape of higher education
in Southeast Asia. In Hong, K. S. & Lai, K. W. (Eds), ICT for accessible,
effective and efficient higher education: Experiences of Southeast Asia.
Australasian Journal of Educational Technology, 27(Special issue, 8): 1276-
1290. http://www.ascilite.org.au/ajet/ajet27/hong.html
Hu, L.T., & Bentler, P.M. (1999). Cutoff criteria for fit indexes in covariance structure
analysis: Conventional criteria versus new alternatives. Structural Equation
Modeling, 6: 1-55.
Hu, P.J., Clark, T.H.K., & Ma, W.W. (2003). Examining technology acceptance by
school teachers: A longitudinal study. Information and Management, 41(2):
227-241.
Hu, P.J., Chau, P.Y.K., Sheng, O.R.L., & Tam, K.Y. (1999). Examining the
technology acceptance model using physician acceptance of telemedicine
technology. Journal of Management Information Systems, 16(2): 91-112.
Huang, E. (2008). Use and gratification in e-consumers. Internet Research, 18(4):
405-426.
Huang, C.J., Liu, M.C., Chang, K.E., Sung, Y.T., Huang, T.H., Chen, C.H., Shen,
H.Y., Huang, K.L., Liao, J.J., Hu, K.W., Luo, Y.C., & Chang, T.Y. (2010). A
learning assistance tool for enhancing ICT literacy of elementary school
students. Educational Technology & Society, 13 (3): 126–138.
Huang, L., Ming-te, L., & Wong, B.K. (2003). Testing of the cross-cultural
applicability of technology acceptance model: Evidence from the PRC,
information technology and organizations. Hershey, New Jersey, USA, Idea
Group Publishing. DOI: 10.4018/978-1-59140-066-0.ch094
Islam, A.Y.M.A. (2011a). Online database adoption and satisfaction model. Germany:
Lambert Academic Publishing.
Islam, A.Y.M.A. (2011b). Viability of the extended technology acceptance model: an
empirical study. Journal of Information and Communication Technology, 10:
85-98.
201
Islam, A.Y.M.A., Ahmad, T.B.T., Madziah, A.Z., & Sahari, M. (2011). Determinants
of Postgraduate Students’ Adoption of an Online Database and Their
Satisfaction in Using it, The New Intellectual Horizon in Education. Malaysia,
KL: International Islamic University Malaysia Press.
Islam, A.Y.M.A. (2014). Validation of the technology satisfaction model (TSM)
developed in Higher Education: The Application of Structural Equation
Modeling. International Journal of Technology and Human Interaction, 10(3):
44-57. doi:10.4018/ijthi.2014070104
Islam, A.Y.M.A., Leng, C.H., & Singh, D. (2015). Efficacy of the technology
satisfaction model (TSM): An empirical study. International Journal of
Technology and Human Interaction, 11(2): 45-60. DOI:
10.4018/ijthi.2014070104
Jang, S.J., & Tsai, M.F. (2013). Exploring the TPACK of Taiwanese secondary school
science teachers using a new contextualized TPACK model. Australasian
Journal of Educational Technology, 29(4): 566-580.
Jackson, C.M., Chow, R.A., & Leitch, R.A. (1997). Toward an understanding of the
behavioral intention to use an information system. Decision Science, 28(2):
357-389.
Kaba, B, & Osei-Bryson, K.M. (2013). Examining influence of national culture on
individuals’ attitude and use of information and communication technology:
Assessment of moderating effect of culture through cross countries study.
International Journal of Information Management, 33(3): 441–452
Karahanna, E., & Straub, D. (1999). The psychological origins of perceived
usefulness and ease of use. Information and Management, 35: 237-250.
Keengwe, J., Onchwari, G., & Wachira, P. (2008). Computer technology integration
and student learning: barriers and promise. Journal of Science of Education
Technology, 17: 560–565.
Kenzie, M., Delecourt, M., & Powers, S. (1994). Computer technologies; Attitudes
and self-efficacy across undergraduates disciplines. Research in Higher
Education, 35(6): 745-768.
202
Kim, H., Choi, H., Han, J. & So, H.J. (2012). Enhancing teachers’ ICT capacity for
the 21st century learning environment: Three cases of teacher education in
Korea. In C. P. Lim & C. S. Chai (Eds), Building the ICT capacity of the next
generation of teachers in Asia. Australasian Journal of Educational Technology,
28(Special issue, 6): 965-982. http://www.ascilite.org.au/ajet/ajet28/kim.html
Kregor, G., Breslin, M., & Fountain, W. (2012). Experience and beliefs of technology
users at an Australian university: Keys to maximising e-learning potential.
Australasian Journal of Educational Technology, 28(8): 1382-1404.
http://www.ascilite.org.au/ajet/ajet28/kregor.html
Kreijns, K., Vermeulen, M., Kirschner, P.A., Buuren, H.V., & Acker, F.V. (2013).
Adopting the integrative model of behaviour prediction to explain teachers’
willingness to use ICT: A perspective for research on teachers’ ict usage in
pedagogical practices. Technology, Pedagogy and Education, 22(1): 55-71. DOI:
10.1080/1475939X.2012.754371
Kumar, R.G., & Ravindran, S. (2012). An empirical study on service quality
perceptions and continuance intention in mobile banking context in India.
Journal of Internet Banking and Commerce, 17(1): 1-22.
Kyriakidou, V., Michalakelis, C., & Sphicopoulos, T. (2013). Assessment of
information and communications technology maturity level.
Telecommunications Policy, 37(1): 48–62. doi:10.1016/j.telpol.2012.08.001
Lallmahamood, M. (2007). An examination of individual’s perceived security and
privacy of the internet in Malaysia and the influence of this on their intention to
use e-commerce: Using an extension of the technology acceptance model.
Journal of Internet Banking and Commerce, 12(3): 1-26.
Lai, K.W. (2011). Digital technology and the culture of teaching and learning in
higher education. In Hong, K. S. & Lai, K. W. (Eds), ICT for accessible,
effective and efficient higher education: Experiences of Southeast Asia.
Australasian Journal of Educational Technology, 27(Special issue, 8): 1263-
1275. http://www.ascilite.org.au/ajet/ajet27/lai.html
Lai, H.M., & Chen, C.P. (2011). Factors influencing secondary school teachers’
adoption of teaching blogs. Computers & Education, 56: 948-960.
doi:10.1016/j.compedu.2010.11.010
203
Langeard, E., Bateson, J.E.G., Lovelock, C.H., & Eiglier, P. (1981). Marketing of
service: new insights from customers and managers. Cambridge, MA:
Marketing Science Institute.
Lau, W.W.F., & Yuen, A.H.K. (2013). Educational technology training workshops for
mathematics teachers: An exploration of perception changes. Australasian
Journal of Educational Technology, 29(4): 595-611.
Legris, P., Ingham, J., & Collerette, P. (2003). Why do people use information
technology? A critical review of the technology acceptance model. Information
& Management, 40: 191-204.
Lee, D.Y., & Lehto, M.R. (2013). User acceptance of youtube for procedural learning:
An extension of the technology acceptance model. Computers & Education, 61:
193-208. http://dx.doi.org/10.1016/j.compedu.2012.10.001
Lee, T.M., & Park, C. (2008). Mobile technology usage and b2b market performance
under mandatory adoption. Industrial Marketing Management, 37(7): 833-840.
Lee, H., Fiore, A. M., & Kim, J. (2006). The role of the technology acceptance model
in explaining effects of image interactivity technology on consumer responses.
International Journal of Retail & Distribution Management, 34(8): 621-644.
Lee, Y.H., Hsieh, Y.C., & Hsu, C.N. (2011). Adding innovation diffusion theory to
the technology acceptance model: Supporting employees' intentions to use e-
learning systems. Educational Technology & Society, 14 (4): 124–137.
Lee, Y-H., Hsiao, C., & Purnomo, S.H. (2014). An empirical examination of
individual and system characteristics on enhancing e-learning acceptance.
Australasian Journal of Educational Technology, 30(5), 561-579.
Levin, T., & Wadmany, R. (2008). Teachers' views on factors affecting effective
integration of information technology in the classroom: Developmental scenery.
Journal of Technology and Teacher Education, 16(2): 233-263.
Liew, C.L., Foo, S., & Chennupati, K.R. (2000). A study of graduate student end-
users’ use and perception of electronic journals. Online Information Review,
24(4): 302-315.
204
Lim, C.P., & Pannen, P. (2012). Building the capacity of Indonesian education
universities for ICT in pre-service teacher education: A case study of a strategic
planning exercise. In C. P. Lim & C. S. Chai (Eds), Building the ICT capacity of
the next generation of teachers in Asia. Australasian Journal of Educational
Technology, 28(Special issue, 6): 1061-1067.
http://www.ascilite.org.au/ajet/ajet28/lim-cp.html
Liu, S.H. (2012). A multivariate model of factors influencing technology use by pre-
service teachers during practice teaching. Educational Technology & Society, 15
(4): 137–149.
Liu, S., Liao, H., & Pratt, J. (2009). Impact of media richness and flow on e-learning
technology acceptance. Computers & Education, 52(3): 599-607.
Ma, Q., & Liu, L. (2004). Technology acceptance model: a meta-analysis of empirical
findings. Journal of Organizational End User Computing, 60: 59-72.
Mac Callum, K., & Jeffrey, L. (2013). The influence of students' ICT skills and their
adoption of mobile learning. Australasian Journal of Educational Technology,
29(3): 303-314.
Marsh, H.W., Hau, K.T., & Wen, Z. (2004). In search of golden rules: Comment on
hypothesis testing approaches to setting cutoff values for fit indices and dangers
in overgeneralizing findings. Structural Equation Modeling, 11: 320-341.
Martínez-Torres, M.R., Marín, S.L.T., García, F.B., Vázquez, G., Oliva, M.A., &
Torres, T. (2008). A technological acceptance of e-learning tools used in
practical and laboratory teaching, according to the European higher education
area. Behaviour & Information Technology, 27(6): 495–505. DOI:
10.1080/01449290600958965
Mathieson, K. (1991). Predicting user intention: comparing the TAM with the theory
of planned behavior. Information Systems research, 2: 173-191.
Mathieson, M., Peacock, E., & Chin, W.W. (2001). Extending the technology
acceptance model: the influence of perceived user resources. Data Base for
Advance in Information Systems, 32(3): 86-112.
205
Mathwick, C., Rigdon, E., & Malhotra, N. (2001). Experiential value
conceptualization measurement and application in the catalog and internet
shopping environment. Journal of Retailing, 77(1): 39-53.
Maughan, P.D. (1999). Library resources and services: a cross-disciplinary survey of
faculty and graduate student use and satisfaction. Journal of Academic
Librarianship, 25(5): 354-366.
McDougall, G.H.G., & Levesque, T. (2000). Customer satisfaction with services:
putting perceived value into the equation. Journal of Service Marketing, 14(5):
392-401.
Moses, P., Wong, S.L., Bakar, K.A. & Rosnaini, M. (2013). Perceived usefulness and
perceived ease of use: Antecedents of attitude towards laptop use among
Science and Mathematics teachers in Malaysia. The Asia-Pacific Education
Researcher, 22(3), 293-299.
Moos, D.C., & Azevedo, R. (2009). Learning with computer-based learning
environments: a literature review of computer self-efficacy. Review of
Educational Research, 79(2): 576-600.
Niu, H. (2014). Shopping in cyberspace: Adolescent technology acceptance attitude
with decision-making styles. International Journal of Technology and Human
Interaction, 10(3), 1-18. doi:10.4018/ijthi.2014070101
Oliver, R., & Towers, S. (2000). Up time: Information Communication Technology:
Literacy and Access for Tertiary students in Australia. Canberra: Department of
Education, Training and Youth Affairs.
Pajares, F. (2002). Self-efficacy beliefs in academic contexts: an outline, Retreived
December 1, 2011 from http://www.uky.edu/~eushe2/Pajares/efftalk.html
Pynoo, B., Tondeur, J., Braak, J.V., Duyck, W., Sijnave, B., & Duyck, P. (2012).
Teachers’ acceptance and use of an educational portal. Computers & Education,
58, 1308–1317. doi:10.1016/j.compedu.2011.12.026
Ramayah, T., Ignatius, J., & Aafaqi, B. (2005). PC usage among students in a private
institution of higher learning: The moderating role of prior experience. Journal
of Educators and Education, 20: 131-152.
206
Ramayah, T., & Lo, M.C. (2007). Impact of shared beliefs on perceived usefulness
and ease of use in the implementation of an entrerprise resource planning
system. Management Research News, 30(6): 420-431.
Rasimah, C.M.Y., Ahmad, A., & Zaman, H.B. (2011). Evaluation of user acceptance
of mixed reality technology. In Hong, K. S. & Lai, K. W. (Eds), ICT for
accessible, effective and efficient higher education: Experiences of Southeast
Asia. Australasian Journal of Educational Technology, 27(Special issue, 8):
1369-1387. http://www.ascilite.org.au/ajet/ajet27/rasimah.html
Ray, K., & Day, J. (1998). Student attitudes towards electronic information resources.
Information Research, 4(2). Available at http://informationr.net/ir/4-
2/paper54.html
Roblyer, D.M. (2006). Integrating Educational Technology into Teaching (4th ed.).
NJ: Pearson Prentice Hall.
Rogers, E. M. (1995). Diffusion of innovations (4th ed.). New York: The Free Press.
Saleh, A.S. (2009). Information Technology (IT) in the Sultan Qaboos University;
Sultanate of Oman: A Study on the IT Culture and the Acceptance and the Use
of IT. Unpublished Doctoral Dissertation, International Islamic University
Malaysia.
Schepers, J., & Wetzels, M. (2007). A meta-analysis of the technology acceptance
model: investigating subjective norm and moderation effects. Information &
Management, 44: 90-103.
Schunk, D.H. (1991). Learning Theories: an Educational Perspective. New York:
McMillan Publishing Company.
Shamdasani, P., Mukherjee, A., & Malhotra, N. (2008). Antecedents and
consequences of service quality in consumer evaluation of selt-service internet
technology. The Service Industry Journal, 28(1): 117-138.
Shin, D.H. (2012). Factors that influence pre-service teachers’ ICT usage in education.
Information Technology & People, 25 (1): 55-80. DOI
10.1108/09593841211204344
207
Shittu, A.T., Basha, K.M., Rahman, N.S.N.A., & Ahmad, T.B.T. (2011). Investigating
students’ attitude and intention to use social software in higher institution of
learning in Malaysia. Multicultural Education & Technology Journal, 5(3): 194-
208. http://dx.doi.org/10.1108/17504971111166929
Shittu, A.T., Basha, K.M., Rahman, N.S.N.A., & Ahmad, T.B.T. (2013).
Determinants of social networking software acceptance: A multi-theoretical
approach. The Malaysian Online Journal of Educational Technology, 1(1): 27-
43.
Shroff, R.H., Deneen, C.D., & Ng, E.M.W. (2011). Analysis of the technology
acceptance model in examining students’ behavioural intention to use an e-
portfolio system. Australasian Journal of Educational Technology, 27(4): 600-
618. http://www.ascilite.org.au/ajet/ajet27/shroff.html
Shipps, B., & Phillips, B. (2013). Social networks, interactivity and satisfaction:
Assessing socio-technical behavioral factors as an extension to technology
acceptance. Journal of Theoretical and Applied Electronic Commerce Research,
8(1): 35-52. DOI: 10.4067/S0718-18762013000100004
Singh, N., Fassott, G., Chao, M.C.H., and Hoffmann, J.A. (2006). Understanding
international web site usage: A cross-national study of German, Brazilian, and
Taiwanese online consumers. International Marketing Review, 23(1): 83-97.
DOI 10.1108/02651330610646304
Sobel, M.E. (1982). Asymptotic confidence intervals for indirect effects in structural
equation models. Sociological methodology, 13: 290-312.
Soffer, T., Nachmias, R., & Ram, J. (2010). Diffusion of web supported instruction in
higher education - The case of Tel-Aviv University. Educational Technology &
Society, 13 (3): 212–223.
Spector, J.M. (2013). Emerging educational technologies and research directions.
Educational Technology & Society, 16 (2): 21–30.
Stone, R. W., & Baker-Eveleth, L.J. (2013). Students’ intentions to purchase
electronic textbooks. Journal of Computing in Higher Education, 25(1): 27-47.
doi: 10.1007/s12528-013-9065-7.
208
Straub, D., Keil, M., & Brenner, W. (1997). Testing the technology acceptance model
across cultures: A three country study. Information and Management, 33(1): 1-
11.
Suki, N.M., Ramayah, T., Nee, A.S., & Suki, N.M. (2014). Consumer intention to use
anti-spyware software: An application of structural equation modeling.
International Journal of Technology and Human Interaction, 10(3), 19-31.
doi:10.4018/ijthi.2014070102
Sun, H., & Zhang, P. (2006). The role of moderating factors in user technology
acceptance. International Journal of human Computer Studies, 64(2): 53-78.
Sutherland, R., Robertson, S., & John, P. (2009). Improving classroom learning with
ICT. London: Routledge.
Szymanski, D.M., & Hise, R.T. (2000). E-satisfation: an initial examination. Journal
of Retailing, 76(3): 309-322.
Taylor, S., & Todd, P. (1995). Understanding information technology usage: a test of
competing models. Information Systems Research, 6(2): 144-176.
Tenopir, C. (2003). Use and users of electronic library resources: an overview and
analysis of recent research studies. Washington, DC: Council on Library and
Information Resources. Available at:
http://www.clir.org/pubs/reports/pub120/pub120.pdf.
Tenner, E., & Yang, Z.Y. (1999). End-user acceptance of electronic journals: a case
study from a major academic research library. Technical Services Quarterly,
17(2): 1-11.
Teo, T. (2011). Factors influencing teachers’ intention to use technology: Model
development and test. Computers & Education, 57: 2432–2440.
doi:10.1016/j.compedu.2011.06.008
Teo, T., Luan, W.S., & Sing, C.C. (2008). A cross-cultural examination of the
intention to use technology between Singaporean and Malaysian pre-service
teachers: An application of the Technology Acceptance Model (TAM).
Educational Technology and Society, 11(4): 265–280.
209
Terzis, V., Moridis, C.N., Economides, A.A., & Rebolledo-Mendez, G. (2013).
Computer based assessment acceptance: A cross-cultural study in Greece and
Mexico. Educational Technology & Society, 16 (3): 411–424.
Tezci, E. (2011). Factors that influence pre-service teachers’ ICT usage in education.
European Journal of Teacher Education, 34(4): 483-499.
http://dx.doi.org/10.1080/02619768.2011.587116
Thompson, R., Compeau, R.D., & Higgins, C. (2006). Intentions to use information
technologies: an integrative model. Journal of organizational and End User
Computing, 18(3): 25-47.
Togia, A., & Tsigilis, N. (2009). Awareness and use of electronic information
resources by education graduate students: Preliminary results from the
Aristotle University of Thessaloniki. In Proceedings of QQML2009:
Qualitative and Quantitative Methods in Libraries, International Conference,
Chania Crete Greece, 26-29 May.
Usluel, Y.K., Askar, P., & Bas, T. (2008). A structural equation model for ICT usage
in higher education. Educational Technology & Society, 11 (2): 262-273.
Venkatesh, V., & Davis, F.D. (1996). A model of the antecedents of perceived ease of
use: development and test. Decision Science, 27(3): 451-482.
Venkatesh, V. (2000). Determinants of perceived ease of use: integrating control,
intrinsic motivation, and emotion into the technology acceptance model.
Information Systems Research, 4(4): 342-365.
Venkatesh, V., and Morris, G.M. (2000). Why don’t men ever stop to ask for
directions? Gender, social influence and their role in technology acceptance
and usage behavior. MIS Quarterly, 24(1) : 115-139.
Venkatesh, V., Morris, G.M., Davis, B.G., & Davis, D.F. (2003). User acceptance of
information technology: toward a unified view. MIS Quarterly, 27(3): 425-478.
Wang, Y.S., Lin, H.H., & Luarn, P. (2006). Predicting consumer intention to use
mobile service. Information Systems Journal, 16: 157–179.
210
Wong, K.T., Teo, T., & Russo, S. (2012). Influence of gender and computer teaching
efficacy on computer acceptance among Malaysian student teachers: An
extended technology acceptance model. Australasian Journal of Educational
Technology, 28(7): 1190-1207. http://www.ascilite.org.au/ajet/ajet28/wong-
kt.html
Wozney, L., Venkatesh, V., & Abrami, C.P. (2006). Implementing computer
technologies: teachers' perceptions and practices. Journal of Technology and
Teacher Education, 14(1): 173- 207.
Wu, W., Chang, H., & Guo, C. (2008). An empirical assessment of science teacher’s
intention towards technology integration. Journal of Computers in
Mathematics and Science Teaching, 27(4): 499-520.
Yousafzai, S. Y., Foxall, G.R., & Pallister, J. G. (2007). Technology acceptance: a
meta-analysis of TAM (Part I). Journal of Modeling in Management, 2(3): 251-
280.
Yuena, A.H.K., & Ma, W.W.K. (2008). Exploring teacher acceptance of e-learning
technology. Asia-Pacific Journal of Teacher Education, 36(3): 229–243. DOI:
10.1080/13598660802232779
Zejno, B., & Islam, A. (2012). Development and validation of library ICT sage scale
for the IIUM postgraduate students. OIDA International Journal of Sustainable
Development, 3(10): 11-18.
Zhao, Y., Pugh, K., Sheldon, S., & Byers, J.L. (2002). Conditions for classroom
technology innovations. Teachers College Record, 104(3): 482–515.
Zhou, G., Zhang, Z. & Li, Y. (2011). Are secondary preservice teachers well prepared
to teach with technology? A case study from China. Australasian Journal of
Educational Technology, 27(6): 943-960.
http://www.ascilite.org.au/ajet/ajet27/zhou.html
211
APPENDIX A The original Model of Computer Self-efficacy (CSE) and Outputs
212
Assessment of normality (Group number 1)
Variable min max skew c.r. kurtosis c.r.
cse15 1.000 7.000 -1.064 -8.644 1.448 5.883
cse14 1.000 7.000 -.898 -7.298 .876 3.557
cse13 1.000 7.000 -.927 -7.534 1.094 4.445
cse12 2.000 7.000 -.882 -7.164 .964 3.917
cse11 2.000 7.000 -.980 -7.959 .587 2.383
cse10 2.000 7.000 -.758 -6.155 .302 1.227
cse9 2.000 7.000 -.759 -6.162 .363 1.474
cse8 2.000 7.000 -.708 -5.749 .111 .450
cse7 2.000 7.000 -1.025 -8.331 .865 3.514
cse6 1.000 7.000 -1.545 -12.552 3.292 13.374
cse5 2.000 7.000 -1.154 -9.378 1.290 5.239
cse4 1.000 7.000 -1.021 -8.296 1.496 6.075
cse3 2.000 7.000 -.815 -6.624 .462 1.877
cse2 2.000 7.000 -.698 -5.669 .189 .768
cse1 1.000 7.000 -1.029 -8.358 1.306 5.305
Multivariate 248.386 109.436
Regression Weights: (Group number 1 - Default model)
Estimate S.E. C.R. P Label
cse1 <--- CSE 1.000
cse2 <--- CSE 1.001 .055 18.189 *** par_1
cse3 <--- CSE .975 .053 18.261 *** par_2
cse4 <--- CSE .868 .053 16.219 *** par_3
cse5 <--- CSE .922 .054 17.110 *** par_4
cse6 <--- CSE .835 .063 13.223 *** par_5
cse7 <--- CSE .978 .052 18.796 *** par_6
cse8 <--- CSE .974 .059 16.380 *** par_7
cse9 <--- CSE 1.039 .056 18.599 *** par_8
cse10 <--- CSE 1.021 .053 19.283 *** par_9
cse11 <--- CSE .980 .054 18.032 *** par_10
cse12 <--- CSE .947 .050 18.805 *** par_11
cse13 <--- CSE .867 .061 14.226 *** par_12
cse14 <--- CSE .844 .070 12.104 *** par_13
cse15 <--- CSE .908 .056 16.206 *** par_14
Standardized Regression Weights: (Group number 1 - Default model)
Estimate
cse1 <--- CSE .765
cse2 <--- CSE .827
cse3 <--- CSE .833
cse4 <--- CSE .759
213
Estimate
cse5 <--- CSE .797
cse6 <--- CSE .640
cse7 <--- CSE .859
cse8 <--- CSE .771
cse9 <--- CSE .854
cse10 <--- CSE .874
cse11 <--- CSE .834
cse12 <--- CSE .857
cse13 <--- CSE .682
cse14 <--- CSE .591
cse15 <--- CSE .763
Variances: (Group number 1 - Default model)
Estimate S.E. C.R. P Label
CSE .746 .084 8.918 *** par_15
e1 .529 .040 13.229 *** par_16
e2 .345 .027 12.805 *** par_17
e3 .314 .025 12.732 *** par_18
e4 .412 .031 13.305 *** par_19
e5 .364 .028 13.077 *** par_20
e6 .749 .055 13.678 *** par_21
e7 .253 .020 12.578 *** par_22
e8 .483 .036 13.233 *** par_23
e9 .298 .024 12.578 *** par_24
e10 .239 .019 12.331 *** par_25
e11 .314 .025 12.825 *** par_26
e12 .241 .019 12.598 *** par_27
e13 .645 .047 13.579 *** par_28
e14 .990 .072 13.760 *** par_29
e15 .442 .033 13.334 *** par_30
Squared Multiple Correlations: (Group number 1 - Default model)
Estimate
cse15 .581
cse14 .350
cse13 .465
cse12 .735
cse11 .695
cse10 .765
cse9 .730
cse8 .594
cse7 .738
214
Estimate
cse6 .410
cse5 .636
cse4 .577
cse3 .693
cse2 .684
cse1 .585
Total Effects (Group number 1 - Default model)
CSE
cse15 .908
cse14 .844
cse13 .867
cse12 .947
cse11 .980
cse10 1.021
cse9 1.039
cse8 .974
cse7 .978
cse6 .835
cse5 .922
cse4 .868
cse3 .975
cse2 1.001
cse1 1.000
Standardized Total Effects (Group number 1 - Default model)
CSE
cse15 .763
cse14 .591
cse13 .682
cse12 .857
cse11 .834
cse10 .874
cse9 .854
cse8 .771
cse7 .859
cse6 .640
cse5 .797
cse4 .759
cse3 .833
cse2 .827
cse1 .765
215
Direct Effects (Group number 1 - Default model)
CSE
cse15 .908
cse14 .844
cse13 .867
cse12 .947
cse11 .980
cse10 1.021
cse9 1.039
cse8 .974
cse7 .978
cse6 .835
cse5 .922
cse4 .868
cse3 .975
cse2 1.001
cse1 1.000
Standardized Direct Effects (Group number 1 - Default model)
CSE
cse15 .763
cse14 .591
cse13 .682
cse12 .857
cse11 .834
cse10 .874
cse9 .854
cse8 .771
cse7 .859
cse6 .640
cse5 .797
cse4 .759
cse3 .833
cse2 .827
cse1 .765
Modification Indices (Group number 1 - Default model)
Covariances: (Group number 1 - Default model)
M.I. Par Change
e14 <--> e15 18.019 .146
e13 <--> e15 5.832 .068
e13 <--> e14 109.539 .431
216
M.I. Par Change
e12 <--> e14 6.215 .065
e11 <--> e13 9.395 .073
e10 <--> e12 8.036 .038
e9 <--> e10 14.658 .057
e8 <--> e11 16.854 .086
e8 <--> e10 9.674 .058
e8 <--> e9 44.292 .137
e7 <--> e14 7.503 -.073
e7 <--> e13 7.850 -.061
e7 <--> e10 6.090 -.034
e6 <--> e12 24.981 -.114
e6 <--> e10 4.856 -.050
e6 <--> e7 11.624 .080
e5 <--> e12 23.425 -.078
e5 <--> e11 17.730 .077
e5 <--> e10 15.511 -.064
e5 <--> e8 11.478 -.076
e5 <--> e7 23.725 .081
e5 <--> e6 36.863 .166
e4 <--> e14 6.105 -.082
e4 <--> e11 4.453 -.041
e4 <--> e10 13.067 -.062
e4 <--> e8 4.910 -.052
e4 <--> e5 23.677 .100
e3 <--> e14 12.370 -.104
e3 <--> e13 16.722 -.098
e3 <--> e11 19.933 -.076
e3 <--> e9 23.173 -.081
e3 <--> e8 26.268 -.107
e3 <--> e4 56.250 .145
e2 <--> e14 4.138 -.063
e2 <--> e13 13.164 -.091
e2 <--> e11 15.185 -.070
e2 <--> e9 13.621 -.065
e2 <--> e8 10.050 -.069
e2 <--> e5 7.979 -.054
e2 <--> e3 56.002 .134
e1 <--> e15 10.752 -.084
e1 <--> e14 8.571 -.110
e1 <--> e11 11.842 -.075
e1 <--> e9 11.853 -.074
e1 <--> e8 14.635 -.102
e1 <--> e6 15.802 -.131
217
M.I. Par Change
e1 <--> e3 44.476 .146
e1 <--> e2 111.191 .241
Regression Weights: (Group number 1 - Default model)
M.I. Par Change
cse15 <--- cse14 11.510 .094
cse15 <--- cse1 4.254 -.063
cse14 <--- cse15 7.184 .132
cse14 <--- cse13 56.822 .347
cse13 <--- cse14 69.943 .278
cse13 <--- cse3 4.742 -.088
cse12 <--- cse6 14.418 -.088
cse12 <--- cse5 8.074 -.074
cse11 <--- cse13 4.884 .059
cse11 <--- cse8 6.516 .069
cse11 <--- cse5 6.103 .073
cse11 <--- cse3 5.678 -.069
cse11 <--- cse2 4.465 -.059
cse11 <--- cse1 4.694 -.056
cse10 <--- cse5 5.353 -.061
cse10 <--- cse4 5.304 -.061
cse9 <--- cse8 17.140 .109
cse9 <--- cse3 6.611 -.073
cse9 <--- cse2 4.011 -.055
cse9 <--- cse1 4.703 -.055
cse8 <--- cse11 4.757 .077
cse8 <--- cse9 10.888 .113
cse8 <--- cse3 7.463 -.097
cse8 <--- cse1 5.791 -.076
cse7 <--- cse14 4.798 -.047
cse7 <--- cse13 4.083 -.049
cse7 <--- cse6 6.709 .061
cse7 <--- cse5 8.178 .076
cse6 <--- cse12 6.004 -.113
cse6 <--- cse5 12.638 .157
cse6 <--- cse1 6.244 -.098
cse5 <--- cse12 5.654 -.078
cse5 <--- cse11 5.009 .069
cse5 <--- cse8 4.432 -.060
cse5 <--- cse7 5.644 .076
cse5 <--- cse6 21.254 .128
cse5 <--- cse4 9.580 .098
cse4 <--- cse5 8.131 .094
218
M.I. Par Change
cse4 <--- cse3 15.975 .131
cse3 <--- cse14 7.907 -.067
cse3 <--- cse13 8.692 -.079
cse3 <--- cse11 5.641 -.069
cse3 <--- cse9 5.715 -.067
cse3 <--- cse8 10.155 -.086
cse3 <--- cse4 22.783 .142
cse3 <--- cse2 16.466 .114
cse3 <--- cse1 17.629 .109
cse2 <--- cse13 6.842 -.073
cse2 <--- cse11 4.296 -.063
cse2 <--- cse3 15.947 .121
cse2 <--- cse1 44.064 .181
cse1 <--- cse15 4.294 -.076
cse1 <--- cse14 5.475 -.071
cse1 <--- cse8 5.647 -.082
cse1 <--- cse6 9.108 -.101
cse1 <--- cse3 12.633 .132
cse1 <--- cse2 32.603 .205
Model Fit Summary
CMIN
Model NPAR CMIN DF P CMIN/DF
Default model 30 834.732 90 .000 9.275
Saturated model 120 .000 0
Independence model 15 5453.298 105 .000 51.936
RMR, GFI
Model RMR GFI AGFI PGFI
Default model .067 .759 .679 .569
Saturated model .000 1.000
Independence model .629 .157 .037 .138
Baseline Comparisons
Model NFI
Delta1
RFI
rho1
IFI
Delta2
TLI
rho2 CFI
Default model .847 .821 .861 .838 .861
Saturated model 1.000 1.000 1.000
Independence model .000 .000 .000 .000 .000
219
Parsimony-Adjusted Measures
Model PRATIO PNFI PCFI
Default model .857 .726 .738
Saturated model .000 .000 .000
Independence model 1.000 .000 .000
NCP
Model NCP LO 90 HI 90
Default model 744.732 655.974 840.944
Saturated model .000 .000 .000
Independence model 5348.298 5109.707 5593.206
FMIN
Model FMIN F0 LO 90 HI 90
Default model 2.113 1.885 1.661 2.129
Saturated model .000 .000 .000 .000
Independence model 13.806 13.540 12.936 14.160
RMSEA
Model RMSEA LO 90 HI 90 PCLOSE
Default model .145 .136 .154 .000
Independence model .359 .351 .367 .000
AIC
Model AIC BCC BIC CAIC
Default model 894.732 897.265 1014.175 1044.175
Saturated model 240.000 250.132 717.770 837.770
Independence model 5483.298 5484.564 5543.019 5558.019
ECVI
Model ECVI LO 90 HI 90 MECVI
Default model 2.265 2.040 2.509 2.272
Saturated model .608 .608 .608 .633
Independence model 13.882 13.278 14.502 13.885
HOELTER
Model HOELTER
.05
HOELTER
.01
Default model 54 59
Independence model 10 11
220
APPENDIX B The original Model of Perceived Usefulness (PU) and Outputs
221
Assessment of normality (Group number 1)
Variable min max skew c.r. kurtosis c.r.
pu16 1.000 7.000 -1.216 -9.881 2.085 8.470
pu15 1.000 7.000 -.910 -7.390 .666 2.704
pu14 1.000 7.000 -1.216 -9.877 1.330 5.402
pu13 1.000 7.000 -.814 -6.614 .505 2.050
pu12 1.000 7.000 -.694 -5.636 -.207 -.842
pu11 1.000 7.000 -.872 -7.081 .671 2.725
pu10 2.000 7.000 -.943 -7.658 .445 1.806
pu9 1.000 7.000 -1.064 -8.643 .743 3.017
pu8 1.000 7.000 -.859 -6.981 .531 2.155
pu7 1.000 7.000 -.664 -5.397 .055 .223
pu6 1.000 7.000 -.844 -6.857 .421 1.712
pu5 1.000 7.000 -.494 -4.010 -.274 -1.112
pu4 1.000 7.000 -.636 -5.165 .040 .163
pu3 1.000 7.000 -.527 -4.284 -.084 -.343
pu2 1.000 7.000 -.814 -6.613 .222 .901
pu1 1.000 7.000 -.883 -7.176 .417 1.692
Multivariate 168.404 69.817
Regression Weights: (Group number 1 - Default model)
Estimate S.E. C.R. P Label
pu1 <--- PU 1.000
pu2 <--- PU 1.077 .051 20.921 *** par_1
pu3 <--- PU 1.119 .055 20.495 *** par_2
pu4 <--- PU 1.091 .053 20.556 *** par_3
pu5 <--- PU 1.083 .053 20.345 *** par_4
pu6 <--- PU 1.102 .053 20.634 *** par_5
pu7 <--- PU 1.095 .059 18.630 *** par_6
pu8 <--- PU .820 .052 15.843 *** par_7
pu9 <--- PU .659 .055 11.927 *** par_8
pu10 <--- PU .707 .050 14.189 *** par_9
pu11 <--- PU .921 .053 17.257 *** par_10
pu12 <--- PU .881 .077 11.389 *** par_11
pu13 <--- PU .852 .055 15.448 *** par_12
pu14 <--- PU .685 .053 12.931 *** par_13
pu15 <--- PU .658 .055 11.968 *** par_14
pu16 <--- PU .723 .050 14.469 *** par_15
Standardized Regression Weights: (Group number 1 - Default model)
Estimate
pu1 <--- PU .817
pu2 <--- PU .852
222
Estimate
pu3 <--- PU .850
pu4 <--- PU .850
pu5 <--- PU .847
pu6 <--- PU .849
pu7 <--- PU .798
pu8 <--- PU .709
pu9 <--- PU .565
pu10 <--- PU .651
pu11 <--- PU .755
pu12 <--- PU .544
pu13 <--- PU .696
pu14 <--- PU .605
pu15 <--- PU .568
pu16 <--- PU .663
Variances: (Group number 1 - Default model)
Estimate S.E. C.R. P Label
PU 1.009 .103 9.813 *** par_16
e1 .503 .039 12.748 *** par_17
e2 .441 .036 12.356 *** par_18
e3 .487 .040 12.210 *** par_19
e4 .460 .037 12.270 *** par_20
e5 .465 .038 12.263 *** par_21
e6 .472 .038 12.403 *** par_22
e7 .689 .053 12.875 *** par_23
e8 .672 .050 13.359 *** par_24
e9 .936 .068 13.680 *** par_25
e10 .687 .051 13.476 *** par_26
e11 .645 .049 13.192 *** par_27
e12 1.861 .135 13.785 *** par_28
e13 .780 .058 13.386 *** par_29
e14 .820 .060 13.592 *** par_30
e15 .915 .067 13.695 *** par_31
e16 .674 .050 13.455 *** par_32
Squared Multiple Correlations: (Group number 1 - Default model)
Estimate
pu16 .439
pu15 .323
pu14 .366
pu13 .484
pu12 .296
223
Estimate
pu11 .571
pu10 .424
pu9 .319
pu8 .503
pu7 .637
pu6 .722
pu5 .718
pu4 .723
pu3 .722
pu2 .726
pu1 .667
Total Effects (Group number 1 - Default model)
PU
pu16 .723
pu15 .658
pu14 .685
pu13 .852
pu12 .881
pu11 .921
pu10 .707
pu9 .659
pu8 .820
pu7 1.095
pu6 1.102
pu5 1.083
pu4 1.091
pu3 1.119
pu2 1.077
pu1 1.000
Standardized Total Effects (Group number 1 - Default model)
PU
pu16 .663
pu15 .568
pu14 .605
pu13 .696
pu12 .544
pu11 .755
pu10 .651
pu9 .565
pu8 .709
224
PU
pu7 .798
pu6 .849
pu5 .847
pu4 .850
pu3 .850
pu2 .852
pu1 .817
Direct Effects (Group number 1 - Default model)
PU
pu16 .723
pu15 .658
pu14 .685
pu13 .852
pu12 .881
pu11 .921
pu10 .707
pu9 .659
pu8 .820
pu7 1.095
pu6 1.102
pu5 1.083
pu4 1.091
pu3 1.119
pu2 1.077
pu1 1.000
Standardized Direct Effects (Group number 1 - Default model)
PU
pu16 .663
pu15 .568
pu14 .605
pu13 .696
pu12 .544
pu11 .755
pu10 .651
pu9 .565
pu8 .709
pu7 .798
pu6 .849
pu5 .847
pu4 .850
225
PU
pu3 .850
pu2 .852
pu1 .817
Modification Indices (Group number 1 - Default model)
Covariances: (Group number 1 - Default model)
M.I. Par Change
e15 <--> e16 75.620 .352
e14 <--> e16 80.538 .345
e14 <--> e15 106.810 .460
e13 <--> e16 30.720 .209
e13 <--> e15 70.903 .368
e13 <--> e14 52.205 .300
e12 <--> e13 22.088 .293
e11 <--> e16 8.070 .098
e11 <--> e14 6.551 .097
e11 <--> e12 15.655 .226
e10 <--> e16 48.446 .246
e10 <--> e15 13.105 .148
e10 <--> e14 32.229 .220
e10 <--> e13 20.453 .172
e10 <--> e11 20.549 .158
e9 <--> e16 31.894 .232
e9 <--> e15 17.520 .199
e9 <--> e14 67.782 .371
e9 <--> e13 13.246 .161
e9 <--> e11 9.809 .127
e9 <--> e10 123.618 .460
e8 <--> e16 20.822 .160
e8 <--> e15 14.373 .154
e8 <--> e14 9.092 .116
e8 <--> e13 5.514 .089
e8 <--> e11 6.200 .086
e8 <--> e10 41.578 .228
e8 <--> e9 46.613 .281
e7 <--> e16 8.160 -.103
e7 <--> e15 6.625 -.108
e7 <--> e14 14.917 -.153
e7 <--> e13 7.924 -.110
e7 <--> e12 15.831 .237
e7 <--> e10 30.420 -.201
e7 <--> e9 32.239 -.240
226
M.I. Par Change
e7 <--> e8 6.588 -.093
e6 <--> e16 5.589 -.072
e6 <--> e15 23.824 -.172
e6 <--> e14 4.446 -.071
e6 <--> e13 18.444 -.141
e6 <--> e12 8.189 -.144
e6 <--> e10 14.800 -.118
e6 <--> e9 9.737 -.111
e6 <--> e8 4.402 -.064
e6 <--> e7 20.834 .143
e5 <--> e16 32.329 -.172
e5 <--> e15 7.484 -.096
e5 <--> e14 34.761 -.195
e5 <--> e13 13.661 -.120
e5 <--> e10 29.641 -.166
e5 <--> e9 33.304 -.204
e5 <--> e7 6.546 .079
e4 <--> e16 9.624 -.093
e4 <--> e15 8.576 -.102
e4 <--> e14 24.924 -.165
e4 <--> e11 15.466 -.117
e4 <--> e10 9.445 -.093
e4 <--> e9 28.315 -.187
e4 <--> e8 16.656 -.123
e4 <--> e7 8.339 .089
e4 <--> e5 50.429 .183
e3 <--> e16 21.953 -.145
e3 <--> e15 13.264 -.130
e3 <--> e14 42.345 -.221
e3 <--> e13 12.532 -.118
e3 <--> e11 8.520 -.089
e3 <--> e10 38.488 -.194
e3 <--> e9 29.551 -.197
e3 <--> e8 19.927 -.138
e3 <--> e7 31.300 .178
e3 <--> e6 6.255 .067
e3 <--> e5 42.953 .174
e3 <--> e4 31.849 .149
e2 <--> e16 17.065 -.122
e2 <--> e15 18.347 -.146
e2 <--> e14 4.885 -.072
e2 <--> e13 6.319 -.080
e2 <--> e12 8.232 -.139
227
M.I. Par Change
e2 <--> e5 5.094 .057
e1 <--> e15 9.125 -.109
e1 <--> e12 20.662 -.233
e1 <--> e11 4.143 -.062
e1 <--> e7 7.498 -.088
e1 <--> e6 7.647 .074
e1 <--> e5 10.546 -.087
e1 <--> e2 69.127 .217
Regression Weights: (Group number 1 - Default model)
M.I. Par Change
pu16 <--- pu15 50.216 .256
pu16 <--- pu14 49.900 .261
pu16 <--- pu13 15.250 .133
pu16 <--- pu10 27.113 .200
pu16 <--- pu9 21.317 .165
pu16 <--- pu8 9.943 .114
pu16 <--- pu5 8.202 -.094
pu16 <--- pu3 5.481 -.074
pu16 <--- pu2 4.180 -.068
pu15 <--- pu16 41.073 .284
pu15 <--- pu14 66.156 .348
pu15 <--- pu13 35.180 .234
pu15 <--- pu10 7.331 .121
pu15 <--- pu9 11.707 .142
pu15 <--- pu8 6.860 .110
pu15 <--- pu6 5.946 -.091
pu15 <--- pu2 4.487 -.081
pu14 <--- pu16 43.750 .278
pu14 <--- pu15 70.915 .334
pu14 <--- pu13 25.907 .191
pu14 <--- pu10 18.032 .179
pu14 <--- pu9 45.296 .265
pu14 <--- pu8 4.340 .083
pu14 <--- pu7 5.032 -.075
pu14 <--- pu5 8.810 -.107
pu14 <--- pu4 6.191 -.089
pu14 <--- pu3 10.562 -.113
pu13 <--- pu16 16.696 .169
pu13 <--- pu15 47.090 .267
pu13 <--- pu14 32.350 .227
pu13 <--- pu12 15.288 .109
pu13 <--- pu10 11.449 .140
228
M.I. Par Change
pu13 <--- pu9 8.855 .115
pu13 <--- pu6 4.614 -.075
pu12 <--- pu13 10.958 .186
pu12 <--- pu11 6.364 .143
pu12 <--- pu7 5.338 .116
pu12 <--- pu1 6.323 -.142
pu11 <--- pu16 4.389 .079
pu11 <--- pu14 4.061 .074
pu11 <--- pu12 10.838 .084
pu11 <--- pu10 11.509 .129
pu11 <--- pu9 6.559 .091
pu10 <--- pu16 26.323 .198
pu10 <--- pu15 8.702 .107
pu10 <--- pu14 19.968 .166
pu10 <--- pu13 10.153 .110
pu10 <--- pu11 8.361 .100
pu10 <--- pu9 82.620 .328
pu10 <--- pu8 19.853 .162
pu10 <--- pu7 10.268 -.098
pu10 <--- pu5 7.518 -.090
pu10 <--- pu3 9.608 -.099
pu9 <--- pu16 17.323 .187
pu9 <--- pu15 11.631 .144
pu9 <--- pu14 41.983 .280
pu9 <--- pu13 6.572 .102
pu9 <--- pu10 69.154 .374
pu9 <--- pu8 22.246 .200
pu9 <--- pu7 10.872 -.118
pu9 <--- pu5 8.436 -.111
pu9 <--- pu4 7.029 -.101
pu9 <--- pu3 7.367 -.101
pu8 <--- pu16 11.318 .129
pu8 <--- pu15 9.546 .112
pu8 <--- pu14 5.634 .088
pu8 <--- pu10 23.276 .186
pu8 <--- pu9 31.161 .200
pu8 <--- pu4 4.146 -.067
pu8 <--- pu3 4.981 -.071
pu7 <--- pu16 4.440 -.083
pu7 <--- pu15 4.403 -.078
pu7 <--- pu14 9.252 -.116
pu7 <--- pu12 10.965 .088
pu7 <--- pu10 17.048 -.164
229
M.I. Par Change
pu7 <--- pu9 21.566 -.172
pu7 <--- pu6 5.235 .076
pu7 <--- pu3 7.856 .092
pu6 <--- pu15 15.847 -.125
pu6 <--- pu13 9.187 -.090
pu6 <--- pu12 5.676 -.054
pu6 <--- pu10 8.305 -.097
pu6 <--- pu9 6.519 -.080
pu6 <--- pu7 7.080 .071
pu5 <--- pu16 17.615 -.139
pu5 <--- pu15 4.978 -.070
pu5 <--- pu14 21.580 -.148
pu5 <--- pu13 6.804 -.077
pu5 <--- pu10 16.631 -.135
pu5 <--- pu9 22.295 -.146
pu5 <--- pu4 12.663 .100
pu5 <--- pu3 10.830 .090
pu4 <--- pu16 5.245 -.075
pu4 <--- pu15 5.704 -.074
pu4 <--- pu14 15.475 -.125
pu4 <--- pu11 6.324 -.074
pu4 <--- pu10 5.300 -.076
pu4 <--- pu9 18.957 -.134
pu4 <--- pu8 7.983 -.088
pu4 <--- pu5 12.921 .101
pu4 <--- pu3 8.033 .077
pu3 <--- pu16 11.962 -.117
pu3 <--- pu15 8.823 -.095
pu3 <--- pu14 26.290 -.167
pu3 <--- pu13 6.243 -.075
pu3 <--- pu10 21.596 -.158
pu3 <--- pu9 19.784 -.141
pu3 <--- pu8 9.551 -.099
pu3 <--- pu7 10.636 .088
pu3 <--- pu5 11.005 .096
pu3 <--- pu4 8.000 .082
pu2 <--- pu16 9.300 -.098
pu2 <--- pu15 12.205 -.106
pu2 <--- pu12 5.705 -.052
pu2 <--- pu1 21.350 .133
pu1 <--- pu15 6.066 -.079
pu1 <--- pu12 14.313 -.087
pu1 <--- pu2 17.049 .121
230
Model Fit Summary
CMIN
Model NPAR CMIN DF P CMIN/DF
Default model 32 1217.231 104 .000 11.704
Saturated model 136 .000 0
Independence model 16 5260.401 120 .000 43.837
RMR, GFI
Model RMR GFI AGFI PGFI
Default model .137 .623 .507 .476
Saturated model .000 1.000
Independence model .816 .178 .068 .157
Baseline Comparisons
Model NFI
Delta1
RFI
rho1
IFI
Delta2
TLI
rho2 CFI
Default model .769 .733 .784 .750 .783
Saturated model 1.000 1.000 1.000
Independence model .000 .000 .000 .000 .000
Parsimony-Adjusted Measures
Model PRATIO PNFI PCFI
Default model .867 .666 .679
Saturated model .000 .000 .000
Independence model 1.000 .000 .000
NCP
Model NCP LO 90 HI 90
Default model 1113.231 1004.610 1229.278
Saturated model .000 .000 .000
Independence model 5140.401 4906.343 5380.785
FMIN
Model FMIN F0 LO 90 HI 90
Default model 3.082 2.818 2.543 3.112
Saturated model .000 .000 .000 .000
Independence model 13.317 13.014 12.421 13.622
231
RMSEA
Model RMSEA LO 90 HI 90 PCLOSE
Default model .165 .156 .173 .000
Independence model .329 .322 .337 .000
AIC
Model AIC BCC BIC CAIC
Default model 1281.231 1284.109 1408.636 1440.636
Saturated model 272.000 284.233 813.472 949.472
Independence model 5292.401 5293.840 5356.104 5372.104
ECVI
Model ECVI LO 90 HI 90 MECVI
Default model 3.244 2.969 3.537 3.251
Saturated model .689 .689 .689 .720
Independence model 13.398 12.806 14.007 13.402
HOELTER
Model HOELTER
.05
HOELTER
.01
Default model 42 46
Independence model 12 12
232
Appendix C The original Model of Perceived Ease of Use (PEU) and Outputs
233
Assessment of normality (Group number 1)
Variable min max skew c.r. kurtosis c.r.
peu15 2.000 7.000 -.850 -6.907 .309 1.255
peu14 1.000 7.000 -.753 -6.120 .602 2.445
peu13 1.000 7.000 -.864 -7.019 .327 1.328
peu12 1.000 7.000 -.437 -3.553 .207 .843
peu11 1.000 7.000 -.898 -7.296 .466 1.893
peu10 1.000 7.000 -.838 -6.808 .377 1.533
peu9 1.000 7.000 -.880 -7.153 .394 1.602
peu8 1.000 7.000 -.709 -5.757 .093 .378
peu7 1.000 7.000 -.574 -4.661 -.003 -.014
peu6 1.000 7.000 -.371 -3.012 -.189 -.769
peu5 1.000 7.000 -.610 -4.957 .123 .498
peu4 1.000 7.000 -.545 -4.427 .056 .226
peu3 1.000 7.000 -.649 -5.269 .050 .202
peu2 1.000 7.000 -.808 -6.564 .308 1.250
peu1 1.000 7.000 -.952 -7.730 .769 3.122
Multivariate 142.575 62.817
Regression Weights: (Group number 1 - Default model)
Estimate S.E. C.R. P Label
peu1 <--- PEU 1.000
peu2 <--- PEU 1.176 .089 13.265 *** par_1
peu3 <--- PEU 1.095 .085 12.950 *** par_2
peu4 <--- PEU 1.132 .093 12.156 *** par_3
peu5 <--- PEU 1.195 .091 13.190 *** par_4
peu6 <--- PEU 1.257 .096 13.116 *** par_5
peu7 <--- PEU 1.065 .085 12.511 *** par_6
peu8 <--- PEU 1.286 .105 12.191 *** par_7
peu9 <--- PEU 1.198 .102 11.725 *** par_8
peu10 <--- PEU 1.175 .102 11.482 *** par_9
peu11 <--- PEU 1.001 .094 10.649 *** par_10
peu12 <--- PEU 1.031 .082 12.594 *** par_11
peu13 <--- PEU 1.184 .104 11.366 *** par_12
peu14 <--- PEU 1.162 .092 12.686 *** par_13
peu15 <--- PEU .777 .074 10.428 *** par_14
Standardized Regression Weights: (Group number 1 - Default model)
Estimate
peu1 <--- PEU .656
peu2 <--- PEU .743
peu3 <--- PEU .732
peu4 <--- PEU .687
234
Estimate
peu5 <--- PEU .762
peu6 <--- PEU .751
peu7 <--- PEU .709
peu8 <--- PEU .700
peu9 <--- PEU .668
peu10 <--- PEU .653
peu11 <--- PEU .584
peu12 <--- PEU .710
peu13 <--- PEU .645
peu14 <--- PEU .724
peu15 <--- PEU .574
Variances: (Group number 1 - Default model)
Estimate S.E. C.R. P Label
PEU .687 .096 7.134 *** par_15
e1 .910 .069 13.223 *** par_16
e2 .770 .061 12.672 *** par_17
e3 .714 .056 12.641 *** par_18
e4 .982 .075 13.015 *** par_19
e5 .708 .057 12.428 *** par_20
e6 .839 .066 12.633 *** par_21
e7 .770 .059 12.957 *** par_22
e8 1.180 .093 12.735 *** par_23
e9 1.220 .095 12.881 *** par_24
e10 1.279 .099 12.976 *** par_25
e11 1.331 .099 13.488 *** par_26
e12 .718 .055 12.950 *** par_27
e13 1.353 .103 13.128 *** par_28
e14 .841 .065 12.849 *** par_29
e15 .844 .062 13.537 *** par_30
Squared Multiple Correlations: (Group number 1 - Default model)
Estimate
peu15 .329
peu14 .524
peu13 .416
peu12 .504
peu11 .341
peu10 .426
peu9 .447
peu8 .490
peu7 .503
235
Estimate
peu6 .564
peu5 .581
peu4 .472
peu3 .536
peu2 .552
peu1 .430
PEU
peu15 .777
peu14 1.162
peu13 1.184
peu12 1.031
peu11 1.001
peu10 1.175
peu9 1.198
peu8 1.286
peu7 1.065
peu6 1.257
peu5 1.195
peu4 1.132
peu3 1.095
peu2 1.176
peu1 1.000
Standardized Total Effects (Group number 1 - Default model)
PEU
peu15 .574
peu14 .724
peu13 .645
peu12 .710
peu11 .584
peu10 .653
peu9 .668
peu8 .700
peu7 .709
peu6 .751
peu5 .762
peu4 .687
peu3 .732
peu2 .743
peu1 .656
236
Direct Effects (Group number 1 - Default model)
PEU
peu15 .777
peu14 1.162
peu13 1.184
peu12 1.031
peu11 1.001
peu10 1.175
peu9 1.198
peu8 1.286
peu7 1.065
peu6 1.257
peu5 1.195
peu4 1.132
peu3 1.095
peu2 1.176
peu1 1.000
Standardized Direct Effects (Group number 1 - Default model)
PEU
peu15 .574
peu14 .724
peu13 .645
peu12 .710
peu11 .584
peu10 .653
peu9 .668
peu8 .700
peu7 .709
peu6 .751
peu5 .762
peu4 .687
peu3 .732
peu2 .743
peu1 .656
Modification Indices (Group number 1 - Default model)
Covariances: (Group number 1 - Default model)
M.I. Par Change
e14 <--> e15 17.239 .186
e13 <--> e14 54.686 .422
e12 <--> e15 10.367 .133
237
M.I. Par Change
e12 <--> e14 19.688 .186
e11 <--> e15 9.040 .166
e11 <--> e14 4.495 .119
e11 <--> e12 52.356 .375
e10 <--> e13 59.690 .537
e9 <--> e15 5.964 -.130
e9 <--> e13 34.826 .401
e9 <--> e10 172.831 .870
e8 <--> e15 14.611 -.202
e8 <--> e13 40.201 .426
e8 <--> e12 17.601 -.207
e8 <--> e11 4.684 -.143
e8 <--> e10 132.798 .754
e8 <--> e9 179.338 .857
e7 <--> e14 13.342 -.159
e7 <--> e13 21.557 -.252
e7 <--> e10 8.174 -.151
e6 <--> e15 7.001 -.119
e6 <--> e13 11.630 -.195
e6 <--> e11 4.100 -.114
e6 <--> e10 8.718 -.165
e6 <--> e9 10.412 -.176
e6 <--> e7 4.107 .088
e5 <--> e12 18.259 -.166
e5 <--> e11 14.163 -.196
e5 <--> e10 18.363 -.220
e5 <--> e9 19.848 -.224
e5 <--> e8 7.290 -.134
e5 <--> e7 32.773 .230
e5 <--> e6 23.850 .207
e4 <--> e14 6.059 -.120
e4 <--> e13 9.901 -.192
e4 <--> e11 12.868 -.216
e4 <--> e10 13.295 -.217
e4 <--> e9 19.790 -.259
e4 <--> e8 8.977 -.173
e4 <--> e6 16.924 .202
e4 <--> e5 50.626 .322
e3 <--> e14 5.247 -.096
e3 <--> e13 34.399 -.309
e3 <--> e10 35.765 -.306
e3 <--> e9 25.610 -.254
e3 <--> e8 24.438 -.245
238
M.I. Par Change
e3 <--> e6 14.093 .158
e3 <--> e5 14.955 .150
e3 <--> e4 20.269 .203
e2 <--> e13 25.768 -.278
e2 <--> e10 12.880 -.191
e2 <--> e9 12.307 -.183
e2 <--> e3 40.798 .258
e1 <--> e13 8.909 -.175
e1 <--> e11 11.359 .195
e1 <--> e8 10.827 -.182
e1 <--> e5 10.242 -.139
e1 <--> e3 9.645 .134
e1 <--> e2 57.356 .341
Regression Weights: (Group number 1 - Default model)
M.I. Par Change
peu15 <--- peu14 7.623 .098
peu15 <--- peu12 4.808 .086
peu15 <--- peu11 5.761 .079
peu15 <--- peu8 6.987 -.082
peu14 <--- peu15 11.220 .143
peu14 <--- peu13 30.570 .174
peu14 <--- peu12 9.161 .121
peu14 <--- peu7 6.221 -.096
peu13 <--- peu14 24.212 .222
peu13 <--- peu10 32.673 .230
peu13 <--- peu9 18.289 .173
peu13 <--- peu8 19.246 .173
peu13 <--- peu7 10.030 -.153
peu13 <--- peu6 4.660 -.093
peu13 <--- peu4 4.930 -.098
peu13 <--- peu3 14.815 -.186
peu13 <--- peu2 10.640 -.149
peu13 <--- peu1 4.837 -.104
peu12 <--- peu15 6.746 .102
peu12 <--- peu14 8.733 .098
peu12 <--- peu11 33.415 .180
peu12 <--- peu8 8.439 -.084
peu12 <--- peu5 6.999 -.090
peu11 <--- peu15 5.875 .128
peu11 <--- peu12 24.286 .242
peu11 <--- peu5 5.409 -.106
peu11 <--- peu4 6.402 -.110
239
M.I. Par Change
peu11 <--- peu1 6.163 .116
peu10 <--- peu13 33.323 .221
peu10 <--- peu9 90.773 .374
peu10 <--- peu8 63.588 .306
peu10 <--- peu5 7.025 -.119
peu10 <--- peu4 6.621 -.110
peu10 <--- peu3 15.406 -.185
peu10 <--- peu2 5.319 -.103
peu9 <--- peu13 19.447 .166
peu9 <--- peu10 94.638 .372
peu9 <--- peu8 85.901 .348
peu9 <--- peu6 4.175 -.084
peu9 <--- peu5 7.596 -.121
peu9 <--- peu4 9.859 -.131
peu9 <--- peu3 11.036 -.153
peu9 <--- peu2 5.085 -.098
peu8 <--- peu15 9.506 -.155
peu8 <--- peu13 22.461 .176
peu8 <--- peu12 8.183 -.134
peu8 <--- peu10 72.760 .323
peu8 <--- peu9 94.278 .369
peu8 <--- peu4 4.475 -.088
peu8 <--- peu3 10.541 -.148
peu8 <--- peu1 5.885 -.108
peu7 <--- peu14 5.918 -.084
peu7 <--- peu13 12.046 -.104
peu7 <--- peu10 4.479 -.065
peu7 <--- peu5 12.563 .125
peu6 <--- peu15 4.559 -.092
peu6 <--- peu13 6.506 -.081
peu6 <--- peu10 4.783 -.071
peu6 <--- peu9 5.481 -.076
peu6 <--- peu5 9.162 .112
peu6 <--- peu4 8.451 .103
peu6 <--- peu3 6.092 .096
peu5 <--- peu12 8.510 -.108
peu5 <--- peu11 9.048 -.094
peu5 <--- peu10 10.078 -.094
peu5 <--- peu9 10.454 -.097
peu5 <--- peu7 15.306 .140
peu5 <--- peu6 9.602 .099
peu5 <--- peu4 25.291 .164
peu5 <--- peu3 6.468 .091
240
M.I. Par Change
peu5 <--- peu1 5.576 -.083
peu4 <--- peu13 5.530 -.079
peu4 <--- peu11 8.210 -.104
peu4 <--- peu10 7.282 -.093
peu4 <--- peu9 10.401 -.112
peu4 <--- peu8 4.302 -.070
peu4 <--- peu6 6.790 .097
peu4 <--- peu5 19.389 .174
peu4 <--- peu3 8.740 .123
peu3 <--- peu13 19.233 -.128
peu3 <--- peu10 19.610 -.131
peu3 <--- peu9 13.474 -.109
peu3 <--- peu8 11.726 -.100
peu3 <--- peu6 5.664 .076
peu3 <--- peu5 5.739 .082
peu3 <--- peu4 10.114 .103
peu3 <--- peu2 16.896 .139
peu3 <--- peu1 5.246 .080
peu2 <--- peu13 14.412 -.115
peu2 <--- peu10 7.064 -.082
peu2 <--- peu9 6.478 -.079
peu2 <--- peu3 17.629 .156
peu2 <--- peu1 31.208 .204
peu1 <--- peu13 4.974 -.072
peu1 <--- peu11 7.245 .093
peu1 <--- peu8 5.185 -.074
peu1 <--- peu3 4.155 .081
peu1 <--- peu2 23.689 .183
Model Fit Summary
CMIN
Model NPAR CMIN DF P CMIN/DF
Default model 30 1152.997 90 .000 12.811
Saturated model 120 .000 0
Independence model 15 3947.317 105 .000 37.593
RMR, GFI
Model RMR GFI AGFI PGFI
Default model .185 .674 .565 .506
Saturated model .000 1.000
Independence model .833 .233 .124 .204
241
Baseline Comparisons
Model NFI
Delta1
RFI
rho1
IFI
Delta2
TLI
rho2 CFI
Default model .708 .659 .724 .677 .723
Saturated model 1.000 1.000 1.000
Independence model .000 .000 .000 .000 .000
Parsimony-Adjusted Measures
Model PRATIO PNFI PCFI
Default model .857 .607 .620
Saturated model .000 .000 .000
Independence model 1.000 .000 .000
NCP
Model NCP LO 90 HI 90
Default model 1062.997 957.164 1176.250
Saturated model .000 .000 .000
Independence model 3842.317 3640.567 4051.339
FMIN
Model FMIN F0 LO 90 HI 90
Default model 2.919 2.691 2.423 2.978
Saturated model .000 .000 .000 .000
Independence model 9.993 9.727 9.217 10.257
RMSEA
Model RMSEA LO 90 HI 90 PCLOSE
Default model .173 .164 .182 .000
Independence model .304 .296 .313 .000
AIC
Model AIC BCC BIC CAIC
Default model 1212.997 1215.530 1332.439 1362.439
Saturated model 240.000 250.132 717.770 837.770
Independence model 3977.317 3978.584 4037.038 4052.038
242
Appendix D The original Model of Intention to Use (INT) and Outputs
243
Assessment of normality (Group number 1)
Variable min max skew c.r. kurtosis c.r.
int10 1.000 7.000 -1.194 -9.700 1.697 6.895
int9 1.000 7.000 -1.102 -8.955 1.431 5.812
int8 2.000 77.000 17.451 141.776 330.280 1341.605
int7 2.000 7.000 -.843 -6.849 .407 1.653
int6 2.000 7.000 -.750 -6.095 -.174 -.707
int5 2.000 7.000 -.962 -7.819 .566 2.298
int4 1.000 7.000 -.573 -4.659 -.241 -.979
int3 1.000 7.000 -1.866 -15.158 3.929 15.958
int2 1.000 7.000 -1.644 -13.353 3.022 12.276
int1 2.000 7.000 -1.065 -8.650 .605 2.458
Multivariate 447.968 287.713
Regression Weights: (Group number 1 - Default model)
Estimate S.E. C.R. P Label
int1 <--- INT 1.000
int2 <--- INT .803 .062 12.897 *** par_1
int3 <--- INT .949 .072 13.178 *** par_2
int4 <--- INT .839 .107 7.829 *** par_3
int5 <--- INT 1.074 .068 15.914 *** par_4
int6 <--- INT 1.195 .070 17.187 *** par_5
int7 <--- INT 1.206 .068 17.651 *** par_6
int8 <--- INT 1.491 .244 6.107 *** par_7
int9 <--- INT 1.232 .075 16.518 *** par_8
int10 <--- INT 1.144 .076 15.049 *** par_9
Standardized Regression Weights: (Group number 1 - Default model)
Estimate
int1 <--- INT .758
int2 <--- INT .632
int3 <--- INT .646
int4 <--- INT .401
int5 <--- INT .765
int6 <--- INT .845
int7 <--- INT .869
int8 <--- INT .315
int9 <--- INT .810
int10 <--- INT .738
244
Variances: (Group number 1 - Default model)
Estimate S.E. C.R. P Label
INT .618 .072 8.624 *** par_10
e1 .458 .037 12.233 *** par_11
e2 .601 .046 13.156 *** par_12
e3 .779 .059 13.140 *** par_13
e4 2.270 .164 13.837 *** par_14
e5 .504 .041 12.365 *** par_15
e6 .352 .032 10.952 *** par_16
e7 .292 .029 10.183 *** par_17
e8 12.459 .894 13.935 *** par_18
e9 .494 .042 11.855 *** par_19
e10 .678 .053 12.684 *** par_20
Squared Multiple Correlations: (Group number 1 - Default model)
Estimate
int10 .544
int9 .655
int8 .099
int7 .755
int6 .715
int5 .586
int4 .161
int3 .417
int2 .399
int1 .575
Total Effects (Group number 1 - Default model)
INT
int10 1.144
int9 1.232
int8 1.491
int7 1.206
int6 1.195
int5 1.074
int4 .839
int3 .949
int2 .803
int1 1.000
245
Standardized Total Effects (Group number 1 - Default model)
INT
int10 .738
int9 .810
int8 .315
int7 .869
int6 .845
int5 .765
int4 .401
int3 .646
int2 .632
int1 .758
Direct Effects (Group number 1 - Default model)
INT
int10 1.144
int9 1.232
int8 1.491
int7 1.206
int6 1.195
int5 1.074
int4 .839
int3 .949
int2 .803
int1 1.000
Standardized Direct Effects (Group number 1 - Default model)
INT
int10 .738
int9 .810
int8 .315
int7 .869
int6 .845
int5 .765
int4 .401
int3 .646
int2 .632
int1 .758
246
Appendix E The original Model of Actual Use (USE) and Outputs
247
Assessment of normality (Group number 1)
Variable min max skew c.r. kurtosis c.r.
use16 1.000 7.000 -.908 -7.378 -.236 -.960
use15 1.000 7.000 -1.681 -13.658 3.123 12.685
use14 1.000 7.000 -1.670 -13.565 2.957 12.013
use13 1.000 7.000 -.691 -5.613 -.382 -1.550
use12 1.000 7.000 .015 .119 -1.459 -5.926
use11 1.000 7.000 -1.457 -11.838 1.968 7.994
use10 1.000 7.000 -2.093 -17.001 5.058 20.545
use9 1.000 7.000 -.510 -4.147 -.850 -3.453
use8 1.000 7.000 -1.077 -8.752 .854 3.469
use7 1.000 7.000 -.746 -6.059 -.187 -.759
use6 2.000 7.000 -1.880 -15.274 4.500 18.278
use5 2.000 7.000 -1.990 -16.164 4.948 20.100
use4 1.000 7.000 -2.024 -16.447 4.592 18.653
use3 1.000 7.000 -1.967 -15.979 4.347 17.656
use2 1.000 7.000 -2.338 -18.993 7.340 29.814
use1 1.000 7.000 -.287 -2.334 -.956 -3.881
Multivariate 152.659 63.289
Regression Weights: (Group number 1 - Default model)
Estimate S.E. C.R. P Label
use1 <--- USE 1.000
use2 <--- USE .799 .107 7.462 *** par_1
use3 <--- USE .619 .115 5.389 *** par_2
use4 <--- USE 1.010 .134 7.540 *** par_3
use5 <--- USE .942 .120 7.857 *** par_4
use6 <--- USE .837 .112 7.476 *** par_5
use7 <--- USE .905 .156 5.796 *** par_6
use8 <--- USE 1.114 .157 7.100 *** par_7
use9 <--- USE .911 .168 5.413 *** par_8
use10 <--- USE .947 .127 7.477 *** par_9
use11 <--- USE 1.096 .155 7.060 *** par_10
use12 <--- USE .336 .150 2.244 .025 par_11
use13 <--- USE .515 .136 3.787 *** par_12
use14 <--- USE 1.043 .148 7.031 *** par_13
use15 <--- USE 1.005 .144 6.976 *** par_14
use16 <--- USE 1.014 .175 5.784 *** par_15
Standardized Regression Weights: (Group number 1 - Default model)
Estimate
use1 <--- USE .415
use2 <--- USE .663
248
Estimate
use3 <--- USE .355
use4 <--- USE .681
use5 <--- USE .773
use6 <--- USE .664
use7 <--- USE .386
use8 <--- USE .560
use9 <--- USE .353
use10 <--- USE .672
use11 <--- USE .564
use12 <--- USE .125
use13 <--- USE .222
use14 <--- USE .582
use15 <--- USE .573
use16 <--- USE .398
Variances: (Group number 1 - Default model)
Estimate S.E. C.R. P Label
USE .547 .135 4.054 *** par_16
e1 2.629 .193 13.645 *** par_17
e2 .446 .036 12.407 *** par_18
e3 1.452 .105 13.766 *** par_19
e4 .645 .053 12.148 *** par_20
e5 .327 .031 10.703 *** par_21
e6 .486 .039 12.397 *** par_22
e7 2.560 .188 13.632 *** par_23
e8 1.488 .114 13.052 *** par_24
e9 3.187 .232 13.708 *** par_25
e10 .595 .048 12.459 *** par_26
e11 1.406 .107 13.108 *** par_27
e12 3.899 .278 14.014 *** par_28
e13 2.807 .201 13.946 *** par_29
e14 1.163 .091 12.782 *** par_30
e15 1.129 .088 12.865 *** par_31
e16 2.990 .220 13.574 *** par_32
Squared Multiple Correlations: (Group number 1 - Default model)
Estimate
use16 .158
use15 .328
use14 .338
use13 .049
use12 .016
249
Estimate
use11 .318
use10 .452
use9 .125
use8 .313
use7 .149
use6 .441
use5 .598
use4 .463
use3 .126
use2 .439
use1 .172
Total Effects (Group number 1 - Default model)
USE
use16 1.014
use15 1.005
use14 1.043
use13 .515
use12 .336
use11 1.096
use10 .947
use9 .911
use8 1.114
use7 .905
use6 .837
use5 .942
use4 1.010
use3 .619
use2 .799
use1 1.000
Standardized Total Effects (Group number 1 - Default model)
USE
use16 .398
use15 .573
use14 .582
use13 .222
use12 .125
use11 .564
use10 .672
use9 .353
use8 .560
250
USE
use7 .386
use6 .664
use5 .773
use4 .681
use3 .355
use2 .663
use1 .415
Direct Effects (Group number 1 - Default model)
USE
use16 1.014
use15 1.005
use14 1.043
use13 .515
use12 .336
use11 1.096
use10 .947
use9 .911
use8 1.114
use7 .905
use6 .837
use5 .942
use4 1.010
use3 .619
use2 .799
use1 1.000
Standardized Direct Effects (Group number 1 - Default model)
USE
use16 .398
use15 .573
use14 .582
use13 .222
use12 .125
use11 .564
use10 .672
use9 .353
use8 .560
use7 .386
use6 .664
use5 .773
use4 .681
251
USE
use3 .355
use2 .663
use1 .415
Modification Indices (Group number 1 - Default model)
Covariances: (Group number 1 - Default model)
M.I. Par Change
e15 <--> e16 47.450 .663
e14 <--> e16 20.360 .441
e14 <--> e15 153.629 .757
e13 <--> e16 8.970 .443
e12 <--> e16 36.930 1.057
e12 <--> e13 18.864 .726
e11 <--> e13 19.534 .456
e11 <--> e12 9.775 .379
e10 <--> e14 6.852 .118
e10 <--> e12 11.093 -.269
e9 <--> e16 23.303 .764
e9 <--> e13 8.648 .448
e9 <--> e12 49.139 1.255
e8 <--> e12 29.408 .676
e8 <--> e11 6.801 .200
e8 <--> e9 27.205 .592
e7 <--> e16 35.802 .851
e7 <--> e15 5.291 -.205
e7 <--> e13 8.583 .400
e7 <--> e12 35.372 .956
e7 <--> e9 67.692 1.204
e7 <--> e8 40.874 .651
e6 <--> e16 11.980 -.223
e6 <--> e15 6.978 -.106
e6 <--> e14 8.801 -.121
e6 <--> e13 4.328 -.129
e6 <--> e10 6.631 -.076
e6 <--> e7 4.354 -.124
e5 <--> e16 24.816 -.275
e5 <--> e15 13.220 -.125
e5 <--> e14 14.502 -.133
e5 <--> e13 4.511 -.113
e5 <--> e12 21.728 -.291
e5 <--> e11 5.418 -.089
e5 <--> e10 12.219 .089
252
M.I. Par Change
e5 <--> e9 16.287 -.229
e5 <--> e7 12.978 -.184
e5 <--> e6 22.937 .110
e4 <--> e16 7.007 -.197
e4 <--> e15 6.062 -.114
e4 <--> e14 24.295 -.233
e4 <--> e13 8.682 -.211
e4 <--> e12 7.987 -.238
e4 <--> e9 18.431 -.329
e4 <--> e7 7.785 -.192
e4 <--> e6 5.459 .072
e4 <--> e5 25.782 .134
e3 <--> e14 5.719 -.163
e3 <--> e8 10.936 -.253
e2 <--> e16 9.429 -.189
e2 <--> e12 17.335 -.290
e2 <--> e11 6.005 -.105
e2 <--> e9 6.792 -.165
e2 <--> e8 26.622 -.227
e2 <--> e7 6.725 -.148
e2 <--> e6 6.578 .066
e2 <--> e5 4.828 .048
e2 <--> e4 14.998 .115
e2 <--> e3 22.284 .202
e1 <--> e15 5.405 -.210
e1 <--> e12 11.910 .563
e1 <--> e10 5.820 -.161
e1 <--> e8 12.728 .369
e1 <--> e7 18.798 .579
Regression Weights: (Group number 1 - Default model)
M.I. Par Change
use16 <--- use15 30.093 .372
use16 <--- use14 12.687 .236
use16 <--- use13 8.479 .149
use16 <--- use12 36.288 .266
use16 <--- use9 20.070 .207
use16 <--- use7 29.865 .277
use16 <--- use6 6.099 -.233
use16 <--- use5 8.291 -.281
use16 <--- use2 4.814 -.216
use15 <--- use16 39.124 .183
use15 <--- use14 95.960 .407
253
M.I. Par Change
use15 <--- use7 4.417 -.067
use15 <--- use5 4.452 -.129
use15 <--- use1 4.372 -.065
use14 <--- use16 16.788 .122
use14 <--- use15 97.673 .426
use14 <--- use6 4.499 -.127
use14 <--- use5 4.886 -.137
use14 <--- use4 11.810 -.175
use14 <--- use3 4.921 -.096
use13 <--- use16 7.387 .122
use13 <--- use12 18.535 .183
use13 <--- use11 12.598 .209
use13 <--- use9 7.446 .121
use13 <--- use7 7.157 .130
use13 <--- use4 4.194 -.158
use12 <--- use16 30.410 .291
use12 <--- use13 17.830 .244
use12 <--- use11 6.303 .174
use12 <--- use10 5.500 -.224
use12 <--- use9 42.307 .339
use12 <--- use8 19.126 .295
use12 <--- use7 29.493 .311
use12 <--- use5 7.228 -.297
use12 <--- use2 8.832 -.331
use12 <--- use1 9.621 .173
use11 <--- use13 18.469 .153
use11 <--- use12 9.606 .096
use11 <--- use8 4.437 .088
use10 <--- use14 4.292 .064
use10 <--- use12 10.902 -.068
use10 <--- use5 4.153 .092
use10 <--- use1 4.714 -.050
use9 <--- use16 19.195 .211
use9 <--- use13 8.174 .151
use9 <--- use12 48.284 .316
use9 <--- use8 17.708 .259
use9 <--- use7 56.460 .392
use9 <--- use5 5.435 -.234
use9 <--- use4 8.913 -.247
use8 <--- use12 28.899 .170
use8 <--- use11 4.399 .092
use8 <--- use9 23.445 .160
use8 <--- use7 34.120 .212
254
M.I. Par Change
use8 <--- use3 9.408 -.150
use8 <--- use2 13.637 -.261
use8 <--- use1 10.296 .114
use7 <--- use16 29.494 .234
use7 <--- use13 8.113 .135
use7 <--- use12 34.757 .241
use7 <--- use9 58.300 .325
use7 <--- use8 26.611 .285
use7 <--- use5 4.334 -.188
use7 <--- use1 15.192 .178
use6 <--- use16 9.887 -.061
use6 <--- use15 4.447 -.060
use6 <--- use14 5.511 -.065
use6 <--- use13 4.093 -.043
use6 <--- use5 7.788 .114
use5 <--- use16 20.530 -.076
use5 <--- use15 8.479 -.071
use5 <--- use14 9.143 -.072
use5 <--- use13 4.269 -.038
use5 <--- use12 21.358 -.073
use5 <--- use10 6.191 .076
use5 <--- use9 14.072 -.062
use5 <--- use7 10.868 -.060
use5 <--- use6 11.897 .117
use5 <--- use4 12.736 .103
use4 <--- use16 5.784 -.054
use4 <--- use14 15.225 -.125
use4 <--- use13 8.212 -.071
use4 <--- use12 7.849 -.060
use4 <--- use9 15.898 -.089
use4 <--- use7 6.506 -.063
use4 <--- use5 8.774 .140
use4 <--- use2 7.725 .133
use3 <--- use8 7.118 -.111
use3 <--- use2 11.370 .231
use2 <--- use16 7.781 -.052
use2 <--- use12 17.037 -.073
use2 <--- use9 5.858 -.045
use2 <--- use8 17.410 -.100
use2 <--- use7 5.619 -.048
use2 <--- use4 7.321 .087
use2 <--- use3 19.184 .120
use1 <--- use12 11.703 .142
255
M.I. Par Change
use1 <--- use8 8.288 .162
use1 <--- use7 15.681 .189
Model Fit Summary
CMIN
Model NPAR CMIN DF P CMIN/DF
Default model 32 875.902 104 .000 8.422
Saturated model 136 .000 0
Independence model 16 2285.935 120 .000 19.049
RMR, GFI
Model RMR GFI AGFI PGFI
Default model .293 .738 .658 .565
Saturated model .000 1.000
Independence model .554 .422 .345 .373
Baseline Comparisons
Model NFI
Delta1
RFI
rho1
IFI
Delta2
TLI
rho2 CFI
Default model .617 .558 .646 .589 .644
Saturated model 1.000 1.000 1.000
Independence model .000 .000 .000 .000 .000
Parsimony-Adjusted Measures
Model PRATIO PNFI PCFI
Default model .867 .535 .558
Saturated model .000 .000 .000
Independence model 1.000 .000 .000
NCP
Model NCP LO 90 HI 90
Default model 771.902 681.164 870.095
Saturated model .000 .000 .000
Independence model 2165.935 2014.355 2324.875
FMIN
Model FMIN F0 LO 90 HI 90
Default model 2.217 1.954 1.724 2.203
Saturated model .000 .000 .000 .000
256
Model FMIN F0 LO 90 HI 90
Independence model 5.787 5.483 5.100 5.886
RMSEA
Model RMSEA LO 90 HI 90 PCLOSE
Default model .137 .129 .146 .000
Independence model .214 .206 .221 .000
AIC
Model AIC BCC BIC CAIC
Default model 939.902 942.780 1067.307 1099.307
Saturated model 272.000 284.233 813.472 949.472
Independence model 2317.935 2319.374 2381.637 2397.637
ECVI
Model ECVI LO 90 HI 90 MECVI
Default model 2.379 2.150 2.628 2.387
Saturated model .689 .689 .689 .720
Independence model 5.868 5.484 6.271 5.872
HOELTER
Model HOELTER
.05
HOELTER
.01
Default model 59 64
Independence model 26 28
257
Appendix F The original Model of Gratification (GRAT) and Outputs
258
Assessment of normality (Group number 1)
Variable min max skew c.r. kurtosis c.r.
grat10 1.000 7.000 -.866 -7.039 .204 .829
grat9 1.000 7.000 -.900 -7.310 .825 3.352
grat8 1.000 7.000 -.784 -6.365 .458 1.860
grat7 1.000 7.000 -.862 -6.999 .716 2.907
grat6 1.000 7.000 -.917 -7.452 .642 2.608
grat5 1.000 7.000 -1.105 -8.977 1.426 5.794
grat4 1.000 7.000 -.951 -7.725 .945 3.837
grat3 1.000 7.000 -.893 -7.259 .649 2.637
grat2 1.000 7.000 -.993 -8.067 .932 3.786
grat1 1.000 7.000 -.922 -7.487 .684 2.779
Multivariate 125.522 80.618
Regression Weights: (Group number 1 - Default model)
Estimate S.E. C.R. P Label
grat1 <--- GRAT 1.000
grat2 <--- GRAT 1.249 .064 19.414 *** par_1
grat3 <--- GRAT 1.179 .062 19.173 *** par_2
grat4 <--- GRAT 1.075 .061 17.619 *** par_3
grat5 <--- GRAT .769 .060 12.837 *** par_4
grat6 <--- GRAT 1.260 .065 19.281 *** par_5
grat7 <--- GRAT 1.202 .065 18.442 *** par_6
grat8 <--- GRAT 1.270 .063 20.218 *** par_7
grat9 <--- GRAT 1.122 .060 18.744 *** par_8
grat10 <--- GRAT 1.125 .082 13.686 *** par_9
Standardized Regression Weights: (Group number 1 - Default model)
Estimate
grat1 <--- GRAT .783
grat2 <--- GRAT .857
grat3 <--- GRAT .843
grat4 <--- GRAT .795
grat5 <--- GRAT .612
grat6 <--- GRAT .857
grat7 <--- GRAT .833
grat8 <--- GRAT .892
grat9 <--- GRAT .838
grat10 <--- GRAT .652
259
Variances: (Group number 1 - Default model)
Estimate S.E. C.R. P Label
GRAT .854 .093 9.194 *** par_10
e1 .537 .042 12.840 *** par_11
e2 .483 .040 12.104 *** par_12
e3 .483 .040 12.216 *** par_13
e4 .575 .045 12.866 *** par_14
e5 .842 .062 13.607 *** par_15
e6 .491 .040 12.193 *** par_16
e7 .544 .044 12.378 *** par_17
e8 .352 .031 11.186 *** par_18
e9 .457 .037 12.412 *** par_19
e10 1.462 .108 13.552 *** par_20
Squared Multiple Correlations: (Group number 1 - Default model)
Estimate
grat10 .425
grat9 .701
grat8 .797
grat7 .694
grat6 .734
grat5 .375
grat4 .632
grat3 .711
grat2 .734
grat1 .614
Total Effects (Group number 1 - Default model)
GRAT
grat10 1.125
grat9 1.122
grat8 1.270
grat7 1.202
grat6 1.260
grat5 .769
grat4 1.075
grat3 1.179
grat2 1.249
grat1 1.000
260
Standardized Total Effects (Group number 1 - Default model)
GRAT
grat10 .652
grat9 .838
grat8 .892
grat7 .833
grat6 .857
grat5 .612
grat4 .795
grat3 .843
grat2 .857
grat1 .783
Direct Effects (Group number 1 - Default model)
GRAT
grat10 1.125
grat9 1.122
grat8 1.270
grat7 1.202
grat6 1.260
grat5 .769
grat4 1.075
grat3 1.179
grat2 1.249
grat1 1.000
Standardized Direct Effects (Group number 1 - Default model)
GRAT
grat10 .652
grat9 .838
grat8 .892
grat7 .833
grat6 .857
grat5 .612
grat4 .795
grat3 .843
grat2 .857
grat1 .783
261
Modification Indices (Group number 1 - Default model)
Covariances: (Group number 1 - Default model)
M.I. Par Change
e9 <--> e10 4.652 .095
e7 <--> e8 53.929 .186
e5 <--> e9 13.050 .121
e5 <--> e8 19.791 -.136
e5 <--> e6 4.392 -.073
e4 <--> e5 34.429 .216
e3 <--> e9 5.297 -.061
e3 <--> e8 12.315 -.084
e3 <--> e7 12.806 -.103
e2 <--> e10 8.809 .136
e2 <--> e9 21.996 -.124
e2 <--> e5 23.175 -.167
e2 <--> e3 20.615 .124
e1 <--> e10 8.520 -.137
e1 <--> e8 21.892 -.116
e1 <--> e7 15.806 -.118
e1 <--> e5 14.906 .137
e1 <--> e4 6.984 .080
e1 <--> e3 47.652 .194
Regression Weights: (Group number 1 - Default model)
M.I. Par Change
grat9 <--- grat5 7.959 .087
grat9 <--- grat2 5.207 -.061
grat8 <--- grat7 15.171 .096
grat8 <--- grat5 12.104 -.099
grat8 <--- grat1 7.978 -.079
grat7 <--- grat8 9.257 .091
grat7 <--- grat1 5.716 -.079
grat5 <--- grat4 11.715 .128
grat5 <--- grat2 5.416 -.081
grat5 <--- grat1 5.355 .092
grat4 <--- grat5 20.979 .157
grat3 <--- grat2 4.884 .061
grat3 <--- grat1 17.247 .131
grat2 <--- grat10 4.920 .052
grat2 <--- grat9 5.969 -.074
grat2 <--- grat5 14.144 -.121
grat2 <--- grat3 5.393 .067
262
M.I. Par Change
grat1 <--- grat10 4.748 -.052
grat1 <--- grat7 4.379 -.060
grat1 <--- grat5 9.081 .099
grat1 <--- grat3 12.365 .104
Model Fit Summary
CMIN
Model NPAR CMIN DF P CMIN/DF
Default model 20 297.460 35 .000 8.499
Saturated model 55 .000 0
Independence model 10 3359.686 45 .000 74.660
RMR, GFI
Model RMR GFI AGFI PGFI
Default model .069 .863 .784 .549
Saturated model .000 1.000
Independence model .994 .212 .037 .173
Baseline Comparisons
Model NFI
Delta1
RFI
rho1
IFI
Delta2
TLI
rho2 CFI
Default model .911 .886 .921 .898 .921
Saturated model 1.000 1.000 1.000
Independence model .000 .000 .000 .000 .000
Parsimony-Adjusted Measures
Model PRATIO PNFI PCFI
Default model .778 .709 .716
Saturated model .000 .000 .000
Independence model 1.000 .000 .000
NCP
Model NCP LO 90 HI 90
Default model 262.460 211.105 321.291
Saturated model .000 .000 .000
Independence model 3314.686 3128.226 3508.442
263
FMIN
Model FMIN F0 LO 90 HI 90
Default model .753 .664 .534 .813
Saturated model .000 .000 .000 .000
Independence model 8.506 8.392 7.920 8.882
RMSEA
Model RMSEA LO 90 HI 90 PCLOSE
Default model .138 .124 .152 .000
Independence model .432 .420 .444 .000
AIC
Model AIC BCC BIC CAIC
Default model 337.460 338.606 417.089 437.089
Saturated model 110.000 113.151 328.978 383.978
Independence model 3379.686 3380.259 3419.500 3429.500
ECVI
Model ECVI LO 90 HI 90 MECVI
Default model .854 .724 1.003 .857
Saturated model .278 .278 .278 .286
Independence model 8.556 8.084 9.047 8.558
HOELTER
Model HOELTER
.05
HOELTER
.01
Default model 67 77
Independence model 8 9
264
Appendix G The Hypothesized TAG Model’s Outputs
Assessment of normality (Group number 1)
Variable min max skew c.r. kurtosis c.r.
use11 1.000 7.000 -1.457 -11.838 1.968 7.994
use6 2.000 7.000 -1.880 -15.274 4.500 18.278
use5 2.000 7.000 -1.990 -16.164 4.948 20.100
use4 1.000 7.000 -2.024 -16.447 4.592 18.653
use2 1.000 7.000 -2.338 -18.993 7.340 29.814
peu2 1.000 7.000 -.808 -6.564 .308 1.250
peu4 1.000 7.000 -.545 -4.427 .056 .226
peu6 1.000 7.000 -.371 -3.012 -.189 -.769
peu7 1.000 7.000 -.574 -4.661 -.003 -.014
peu9 1.000 7.000 -.880 -7.153 .394 1.602
peu15 2.000 7.000 -.850 -6.907 .309 1.255
pu2 1.000 7.000 -.814 -6.613 .222 .901
pu3 1.000 7.000 -.527 -4.284 -.084 -.343
pu4 1.000 7.000 -.636 -5.165 .040 .163
pu5 1.000 7.000 -.494 -4.010 -.274 -1.112
pu6 1.000 7.000 -.844 -6.857 .421 1.712
pu7 1.000 7.000 -.664 -5.397 .055 .223
pu11 1.000 7.000 -.872 -7.081 .671 2.725
cse4 1.000 7.000 -1.021 -8.296 1.496 6.075
cse7 2.000 7.000 -1.025 -8.331 .865 3.514
cse9 2.000 7.000 -.759 -6.162 .363 1.474
cse10 2.000 7.000 -.758 -6.155 .302 1.227
cse11 2.000 7.000 -.980 -7.959 .587 2.383
cse12 2.000 7.000 -.882 -7.164 .964 3.917
cse15 1.000 7.000 -1.064 -8.644 1.448 5.883
int10 1.000 7.000 -1.194 -9.700 1.697 6.895
int8 2.000 77.000 17.451 141.776 330.280 1341.605
int7 2.000 7.000 -.843 -6.849 .407 1.653
int5 2.000 7.000 -.962 -7.819 .566 2.298
int3 1.000 7.000 -1.866 -15.158 3.929 15.958
int2 1.000 7.000 -1.644 -13.353 3.022 12.276
int1 2.000 7.000 -1.065 -8.650 .605 2.458
grat10 1.000 7.000 -.866 -7.039 .204 .829
grat9 1.000 7.000 -.900 -7.310 .825 3.352
grat7 1.000 7.000 -.862 -6.999 .716 2.907
grat6 1.000 7.000 -.917 -7.452 .642 2.608
grat4 1.000 7.000 -.951 -7.725 .945 3.837
grat3 1.000 7.000 -.893 -7.259 .649 2.637
265
Variable min max skew c.r. kurtosis c.r.
Multivariate 919.030 165.848
Regression Weights: (Group number 1 - Default model)
Estimate S.E. C.R. P Label
PU <--- CSE .583 .065 9.039 *** par_33
PEU <--- CSE .487 .061 7.980 *** par_34
INT <--- PU .267 .072 3.702 *** par_35
INT <--- PEU .487 .100 4.853 *** par_38
USE <--- PU .031 .055 .569 .569 par_36
USE <--- PEU .094 .079 1.181 .238 par_39
USE <--- INT .273 .051 5.307 *** par_41
GRAT <--- PU .555 .075 7.359 *** par_37
GRAT <--- PEU .479 .105 4.578 *** par_40
GRAT <--- USE -.126 .078 -1.615 .106 par_42
GRAT <--- INT .208 .065 3.188 .001 par_43
grat3 <--- GRAT 1.000
grat4 <--- GRAT .951 .052 18.290 *** par_1
grat6 <--- GRAT 1.111 .055 20.283 *** par_2
grat7 <--- GRAT 1.042 .055 18.874 *** par_3
grat9 <--- GRAT .999 .051 19.762 *** par_4
grat10 <--- GRAT .979 .071 13.791 *** par_5
int1 <--- INT 1.000
int2 <--- INT .827 .052 15.766 *** par_6
int3 <--- INT .941 .061 15.352 *** par_7
int5 <--- INT 1.023 .057 17.955 *** par_8
int7 <--- INT .964 .058 16.563 *** par_9
int8 <--- INT 1.327 .224 5.935 *** par_10
int10 <--- INT .986 .066 14.890 *** par_11
cse15 <--- CSE 1.000
cse12 <--- CSE 1.065 .057 18.813 *** par_12
cse11 <--- CSE 1.086 .061 17.852 *** par_13
cse10 <--- CSE 1.135 .060 18.921 *** par_14
cse9 <--- CSE 1.174 .063 18.730 *** par_15
cse7 <--- CSE 1.065 .059 18.081 *** par_16
cse4 <--- CSE .910 .061 14.964 *** par_17
pu11 <--- PU 1.000
pu7 <--- PU 1.328 .083 15.998 *** par_18
pu6 <--- PU 1.293 .078 16.510 *** par_19
pu5 <--- PU 1.323 .078 17.052 *** par_20
pu4 <--- PU 1.313 .078 16.807 *** par_21
pu3 <--- PU 1.368 .080 17.051 *** par_22
pu2 <--- PU 1.226 .076 16.075 *** par_23
peu15 <--- PEU 1.000
266
Estimate S.E. C.R. P Label
peu9 <--- PEU 1.221 .137 8.942 *** par_24
peu7 <--- PEU 1.397 .125 11.148 *** par_25
peu6 <--- PEU 1.523 .143 10.680 *** par_26
peu4 <--- PEU 1.446 .136 10.653 *** par_27
peu2 <--- PEU 1.383 .130 10.673 *** par_28
use2 <--- USE 1.000
use4 <--- USE 1.324 .102 13.020 *** par_29
use5 <--- USE 1.206 .088 13.746 *** par_30
use6 <--- USE 1.110 .088 12.675 *** par_31
use11 <--- USE 1.118 .130 8.572 *** par_32
Standardized Regression Weights: (Group number 1 - Default model)
Estimate
PU <--- CSE .528
PEU <--- CSE .566
INT <--- PU .278
INT <--- PEU .394
USE <--- PU .044
USE <--- PEU .103
USE <--- INT .373
GRAT <--- PU .488
GRAT <--- PEU .328
GRAT <--- USE -.078
GRAT <--- INT .176
grat3 <--- GRAT .794
grat4 <--- GRAT .780
grat6 <--- GRAT .843
grat7 <--- GRAT .803
grat9 <--- GRAT .832
grat10 <--- GRAT .620
int1 <--- INT .822
int2 <--- INT .700
int3 <--- INT .689
int5 <--- INT .788
int7 <--- INT .750
int8 <--- INT .297
int10 <--- INT .684
cse15 <--- CSE .760
cse12 <--- CSE .873
cse11 <--- CSE .836
cse10 <--- CSE .879
cse9 <--- CSE .873
cse7 <--- CSE .845
267
Estimate
cse4 <--- CSE .721
pu11 <--- PU .705
pu7 <--- PU .832
pu6 <--- PU .857
pu5 <--- PU .890
pu4 <--- PU .880
pu3 <--- PU .893
pu2 <--- PU .834
peu15 <--- PEU .599
peu9 <--- PEU .552
peu7 <--- PEU .754
peu6 <--- PEU .738
peu4 <--- PEU .712
peu2 <--- PEU .708
use2 <--- USE .687
use4 <--- USE .740
use5 <--- USE .822
use6 <--- USE .729
use11 <--- USE .475
Variances: (Group number 1 - Default model)
Estimate S.E. C.R. P Label
CSE .610 .069 8.786 *** par_44
e39 .538 .068 7.859 *** par_45
e40 .307 .050 6.107 *** par_46
e43 .484 .053 9.093 *** par_47
e42 .294 .041 7.143 *** par_48
e41 .427 .050 8.601 *** par_49
e1 .566 .047 11.943 *** par_50
e2 .562 .046 12.116 *** par_51
e3 .484 .044 11.020 *** par_52
e4 .578 .049 11.800 *** par_53
e5 .429 .038 11.252 *** par_54
e6 1.483 .112 13.278 *** par_55
e7 .331 .032 10.377 *** par_56
e8 .490 .040 12.325 *** par_57
e9 .676 .054 12.491 *** par_58
e10 .439 .039 11.234 *** par_59
e11 .498 .043 11.666 *** par_60
e12 12.523 .901 13.892 *** par_61
e13 .764 .061 12.500 *** par_62
e14 .447 .034 12.959 *** par_63
e15 .217 .019 11.470 *** par_64
268
Estimate S.E. C.R. P Label
e16 .311 .025 12.186 *** par_65
e17 .231 .020 11.292 *** par_66
e18 .262 .023 11.450 *** par_67
e19 .276 .023 12.021 *** par_68
e20 .468 .036 13.188 *** par_69
e21 .756 .057 13.338 *** par_70
e22 .584 .047 12.404 *** par_71
e23 .450 .037 12.050 *** par_72
e24 .344 .030 11.331 *** par_73
e25 .376 .032 11.625 *** par_74
e26 .356 .032 11.257 *** par_75
e27 .490 .039 12.412 *** par_76
e28 .807 .064 12.597 *** par_77
e29 1.532 .117 13.058 *** par_78
e30 .669 .060 11.081 *** par_79
e31 .876 .079 11.081 *** par_80
e32 .918 .079 11.629 *** par_81
e33 .857 .073 11.714 *** par_82
e34 .415 .035 11.804 *** par_83
e35 .536 .049 11.017 *** par_84
e36 .259 .029 8.887 *** par_85
e37 .401 .036 11.164 *** par_86
e38 1.587 .119 13.328 *** par_87
Squared Multiple Correlations: (Group number 1 - Default model)
Estimate
PEU .321
PU .279
INT .298
USE .204
GRAT .557
use11 .226
use6 .532
use5 .675
use4 .547
use2 .472
peu2 .502
peu4 .507
peu6 .544
peu7 .568
peu9 .305
peu15 .359
pu2 .696
269
Estimate
pu3 .797
pu4 .774
pu5 .791
pu6 .735
pu7 .692
pu11 .496
cse4 .519
cse7 .715
cse9 .762
cse10 .773
cse11 .698
cse12 .761
cse15 .577
int10 .467
int8 .088
int7 .562
int5 .622
int3 .474
int2 .490
int1 .675
grat10 .384
grat9 .692
grat7 .644
grat6 .711
grat4 .608
grat3 .630
Total Effects (Group number 1 - Default model)
CSE PEU PU INT USE GRAT
PEU .487 .000 .000 .000 .000 .000
PU .583 .000 .000 .000 .000 .000
INT .393 .487 .267 .000 .000 .000
USE .171 .227 .104 .273 .000 .000
GRAT .617 .552 .597 .174 -.126 .000
use11 .191 .254 .117 .305 1.118 .000
use6 .190 .252 .116 .303 1.110 .000
use5 .207 .273 .126 .329 1.206 .000
use4 .227 .300 .138 .362 1.324 .000
use2 .171 .227 .104 .273 1.000 .000
peu2 .674 1.383 .000 .000 .000 .000
peu4 .704 1.446 .000 .000 .000 .000
peu6 .742 1.523 .000 .000 .000 .000
peu7 .680 1.397 .000 .000 .000 .000
270
CSE PEU PU INT USE GRAT
peu9 .595 1.221 .000 .000 .000 .000
peu15 .487 1.000 .000 .000 .000 .000
pu2 .715 .000 1.226 .000 .000 .000
pu3 .798 .000 1.368 .000 .000 .000
pu4 .766 .000 1.313 .000 .000 .000
pu5 .771 .000 1.323 .000 .000 .000
pu6 .754 .000 1.293 .000 .000 .000
pu7 .775 .000 1.328 .000 .000 .000
pu11 .583 .000 1.000 .000 .000 .000
cse4 .910 .000 .000 .000 .000 .000
cse7 1.065 .000 .000 .000 .000 .000
cse9 1.174 .000 .000 .000 .000 .000
cse10 1.135 .000 .000 .000 .000 .000
cse11 1.086 .000 .000 .000 .000 .000
cse12 1.065 .000 .000 .000 .000 .000
cse15 1.000 .000 .000 .000 .000 .000
int10 .387 .480 .263 .986 .000 .000
int8 .521 .647 .354 1.327 .000 .000
int7 .379 .470 .257 .964 .000 .000
int5 .402 .498 .273 1.023 .000 .000
int3 .370 .459 .251 .941 .000 .000
int2 .325 .403 .221 .827 .000 .000
int1 .393 .487 .267 1.000 .000 .000
grat10 .605 .541 .585 .170 -.123 .979
grat9 .617 .552 .597 .174 -.126 .999
grat7 .643 .575 .623 .181 -.131 1.042
grat6 .686 .613 .664 .193 -.140 1.111
grat4 .587 .525 .568 .165 -.120 .951
grat3 .617 .552 .597 .174 -.126 1.000
Standardized Total Effects (Group number 1 - Default model)
CSE PEU PU INT USE GRAT
PEU .566 .000 .000 .000 .000 .000
PU .528 .000 .000 .000 .000 .000
INT .370 .394 .278 .000 .000 .000
USE .220 .250 .148 .373 .000 .000
GRAT .491 .378 .525 .147 -.078 .000
use11 .104 .119 .070 .177 .475 .000
use6 .160 .183 .108 .272 .729 .000
use5 .181 .206 .122 .306 .822 .000
use4 .163 .185 .109 .276 .740 .000
use2 .151 .172 .102 .256 .687 .000
peu2 .401 .708 .000 .000 .000 .000
271
CSE PEU PU INT USE GRAT
peu4 .403 .712 .000 .000 .000 .000
peu6 .418 .738 .000 .000 .000 .000
peu7 .427 .754 .000 .000 .000 .000
peu9 .313 .552 .000 .000 .000 .000
peu15 .339 .599 .000 .000 .000 .000
pu2 .440 .000 .834 .000 .000 .000
pu3 .471 .000 .893 .000 .000 .000
pu4 .464 .000 .880 .000 .000 .000
pu5 .469 .000 .890 .000 .000 .000
pu6 .452 .000 .857 .000 .000 .000
pu7 .439 .000 .832 .000 .000 .000
pu11 .372 .000 .705 .000 .000 .000
cse4 .721 .000 .000 .000 .000 .000
cse7 .845 .000 .000 .000 .000 .000
cse9 .873 .000 .000 .000 .000 .000
cse10 .879 .000 .000 .000 .000 .000
cse11 .836 .000 .000 .000 .000 .000
cse12 .873 .000 .000 .000 .000 .000
cse15 .760 .000 .000 .000 .000 .000
int10 .253 .270 .190 .684 .000 .000
int8 .110 .117 .082 .297 .000 .000
int7 .277 .296 .208 .750 .000 .000
int5 .292 .311 .219 .788 .000 .000
int3 .255 .272 .191 .689 .000 .000
int2 .259 .276 .194 .700 .000 .000
int1 .304 .324 .228 .822 .000 .000
grat10 .304 .234 .326 .091 -.048 .620
grat9 .408 .314 .437 .122 -.065 .832
grat7 .394 .303 .422 .118 -.063 .803
grat6 .414 .319 .443 .124 -.066 .843
grat4 .383 .295 .409 .114 -.061 .780
grat3 .390 .300 .417 .117 -.062 .794
Direct Effects (Group number 1 - Default model)
CSE PEU PU INT USE GRAT
PEU .487 .000 .000 .000 .000 .000
PU .583 .000 .000 .000 .000 .000
INT .000 .487 .267 .000 .000 .000
USE .000 .094 .031 .273 .000 .000
GRAT .000 .479 .555 .208 -.126 .000
use11 .000 .000 .000 .000 1.118 .000
use6 .000 .000 .000 .000 1.110 .000
use5 .000 .000 .000 .000 1.206 .000
272
CSE PEU PU INT USE GRAT
use4 .000 .000 .000 .000 1.324 .000
use2 .000 .000 .000 .000 1.000 .000
peu2 .000 1.383 .000 .000 .000 .000
peu4 .000 1.446 .000 .000 .000 .000
peu6 .000 1.523 .000 .000 .000 .000
peu7 .000 1.397 .000 .000 .000 .000
peu9 .000 1.221 .000 .000 .000 .000
peu15 .000 1.000 .000 .000 .000 .000
pu2 .000 .000 1.226 .000 .000 .000
pu3 .000 .000 1.368 .000 .000 .000
pu4 .000 .000 1.313 .000 .000 .000
pu5 .000 .000 1.323 .000 .000 .000
pu6 .000 .000 1.293 .000 .000 .000
pu7 .000 .000 1.328 .000 .000 .000
pu11 .000 .000 1.000 .000 .000 .000
cse4 .910 .000 .000 .000 .000 .000
cse7 1.065 .000 .000 .000 .000 .000
cse9 1.174 .000 .000 .000 .000 .000
cse10 1.135 .000 .000 .000 .000 .000
cse11 1.086 .000 .000 .000 .000 .000
cse12 1.065 .000 .000 .000 .000 .000
cse15 1.000 .000 .000 .000 .000 .000
int10 .000 .000 .000 .986 .000 .000
int8 .000 .000 .000 1.327 .000 .000
int7 .000 .000 .000 .964 .000 .000
int5 .000 .000 .000 1.023 .000 .000
int3 .000 .000 .000 .941 .000 .000
int2 .000 .000 .000 .827 .000 .000
int1 .000 .000 .000 1.000 .000 .000
grat10 .000 .000 .000 .000 .000 .979
grat9 .000 .000 .000 .000 .000 .999
grat7 .000 .000 .000 .000 .000 1.042
grat6 .000 .000 .000 .000 .000 1.111
grat4 .000 .000 .000 .000 .000 .951
grat3 .000 .000 .000 .000 .000 1.000
Standardized Direct Effects (Group number 1 - Default model)
CSE PEU PU INT USE GRAT
PEU .566 .000 .000 .000 .000 .000
PU .528 .000 .000 .000 .000 .000
INT .000 .394 .278 .000 .000 .000
273
CSE PEU PU INT USE GRAT
USE .000 .103 .044 .373 .000 .000
GRAT .000 .328 .488 .176 -.078 .000
use11 .000 .000 .000 .000 .475 .000
use6 .000 .000 .000 .000 .729 .000
use5 .000 .000 .000 .000 .822 .000
use4 .000 .000 .000 .000 .740 .000
use2 .000 .000 .000 .000 .687 .000
peu2 .000 .708 .000 .000 .000 .000
peu4 .000 .712 .000 .000 .000 .000
peu6 .000 .738 .000 .000 .000 .000
peu7 .000 .754 .000 .000 .000 .000
peu9 .000 .552 .000 .000 .000 .000
peu15 .000 .599 .000 .000 .000 .000
pu2 .000 .000 .834 .000 .000 .000
pu3 .000 .000 .893 .000 .000 .000
pu4 .000 .000 .880 .000 .000 .000
pu5 .000 .000 .890 .000 .000 .000
pu6 .000 .000 .857 .000 .000 .000
pu7 .000 .000 .832 .000 .000 .000
pu11 .000 .000 .705 .000 .000 .000
cse4 .721 .000 .000 .000 .000 .000
cse7 .845 .000 .000 .000 .000 .000
cse9 .873 .000 .000 .000 .000 .000
cse10 .879 .000 .000 .000 .000 .000
cse11 .836 .000 .000 .000 .000 .000
cse12 .873 .000 .000 .000 .000 .000
cse15 .760 .000 .000 .000 .000 .000
int10 .000 .000 .000 .684 .000 .000
int8 .000 .000 .000 .297 .000 .000
int7 .000 .000 .000 .750 .000 .000
int5 .000 .000 .000 .788 .000 .000
int3 .000 .000 .000 .689 .000 .000
int2 .000 .000 .000 .700 .000 .000
int1 .000 .000 .000 .822 .000 .000
grat10 .000 .000 .000 .000 .000 .620
grat9 .000 .000 .000 .000 .000 .832
grat7 .000 .000 .000 .000 .000 .803
grat6 .000 .000 .000 .000 .000 .843
grat4 .000 .000 .000 .000 .000 .780
grat3 .000 .000 .000 .000 .000 .794
274
Indirect Effects (Group number 1 - Default model)
CSE PEU PU INT USE GRAT
PEU .000 .000 .000 .000 .000 .000
PU .000 .000 .000 .000 .000 .000
INT .393 .000 .000 .000 .000 .000
USE .171 .133 .073 .000 .000 .000
GRAT .617 .073 .042 -.034 .000 .000
use11 .191 .254 .117 .305 .000 .000
use6 .190 .252 .116 .303 .000 .000
use5 .207 .273 .126 .329 .000 .000
use4 .227 .300 .138 .362 .000 .000
use2 .171 .227 .104 .273 .000 .000
peu2 .674 .000 .000 .000 .000 .000
peu4 .704 .000 .000 .000 .000 .000
peu6 .742 .000 .000 .000 .000 .000
peu7 .680 .000 .000 .000 .000 .000
peu9 .595 .000 .000 .000 .000 .000
peu15 .487 .000 .000 .000 .000 .000
pu2 .715 .000 .000 .000 .000 .000
pu3 .798 .000 .000 .000 .000 .000
pu4 .766 .000 .000 .000 .000 .000
pu5 .771 .000 .000 .000 .000 .000
pu6 .754 .000 .000 .000 .000 .000
pu7 .775 .000 .000 .000 .000 .000
pu11 .583 .000 .000 .000 .000 .000
cse4 .000 .000 .000 .000 .000 .000
cse7 .000 .000 .000 .000 .000 .000
cse9 .000 .000 .000 .000 .000 .000
cse10 .000 .000 .000 .000 .000 .000
cse11 .000 .000 .000 .000 .000 .000
cse12 .000 .000 .000 .000 .000 .000
cse15 .000 .000 .000 .000 .000 .000
int10 .387 .480 .263 .000 .000 .000
int8 .521 .647 .354 .000 .000 .000
int7 .379 .470 .257 .000 .000 .000
int5 .402 .498 .273 .000 .000 .000
int3 .370 .459 .251 .000 .000 .000
int2 .325 .403 .221 .000 .000 .000
int1 .393 .487 .267 .000 .000 .000
grat10 .605 .541 .585 .170 -.123 .000
grat9 .617 .552 .597 .174 -.126 .000
grat7 .643 .575 .623 .181 -.131 .000
grat6 .686 .613 .664 .193 -.140 .000
grat4 .587 .525 .568 .165 -.120 .000
275
CSE PEU PU INT USE GRAT
grat3 .617 .552 .597 .174 -.126 .000
Standardized Indirect Effects (Group number 1 - Default model)
CSE PEU PU INT USE GRAT
PEU .000 .000 .000 .000 .000 .000
PU .000 .000 .000 .000 .000 .000
INT .370 .000 .000 .000 .000 .000
USE .220 .147 .103 .000 .000 .000
GRAT .491 .050 .037 -.029 .000 .000
use11 .104 .119 .070 .177 .000 .000
use6 .160 .183 .108 .272 .000 .000
use5 .181 .206 .122 .306 .000 .000
use4 .163 .185 .109 .276 .000 .000
use2 .151 .172 .102 .256 .000 .000
peu2 .401 .000 .000 .000 .000 .000
peu4 .403 .000 .000 .000 .000 .000
peu6 .418 .000 .000 .000 .000 .000
peu7 .427 .000 .000 .000 .000 .000
peu9 .313 .000 .000 .000 .000 .000
peu15 .339 .000 .000 .000 .000 .000
pu2 .440 .000 .000 .000 .000 .000
pu3 .471 .000 .000 .000 .000 .000
pu4 .464 .000 .000 .000 .000 .000
pu5 .469 .000 .000 .000 .000 .000
pu6 .452 .000 .000 .000 .000 .000
pu7 .439 .000 .000 .000 .000 .000
pu11 .372 .000 .000 .000 .000 .000
cse4 .000 .000 .000 .000 .000 .000
cse7 .000 .000 .000 .000 .000 .000
cse9 .000 .000 .000 .000 .000 .000
cse10 .000 .000 .000 .000 .000 .000
cse11 .000 .000 .000 .000 .000 .000
cse12 .000 .000 .000 .000 .000 .000
cse15 .000 .000 .000 .000 .000 .000
int10 .253 .270 .190 .000 .000 .000
int8 .110 .117 .082 .000 .000 .000
int7 .277 .296 .208 .000 .000 .000
int5 .292 .311 .219 .000 .000 .000
int3 .255 .272 .191 .000 .000 .000
int2 .259 .276 .194 .000 .000 .000
int1 .304 .324 .228 .000 .000 .000
grat10 .304 .234 .326 .091 -.048 .000
grat9 .408 .314 .437 .122 -.065 .000
276
CSE PEU PU INT USE GRAT
grat7 .394 .303 .422 .118 -.063 .000
grat6 .414 .319 .443 .124 -.066 .000
grat4 .383 .295 .409 .114 -.061 .000
grat3 .390 .300 .417 .117 -.062 .000
Covariances: (Group number 1 - Default model)
M.I. Par Change
e39 <--> e40 152.910 .298
e43 <--> CSE 56.460 .233
e43 <--> e40 31.918 -.136
e43 <--> e39 22.504 -.140
e42 <--> CSE 7.039 .065
e42 <--> e40 4.004 -.038
e38 <--> CSE 8.679 .153
e37 <--> e43 4.127 .054
e36 <--> e39 4.433 -.049
e33 <--> CSE 9.699 -.125
e33 <--> e40 5.317 .071
e33 <--> e39 24.481 .190
e33 <--> e43 5.832 -.093
e33 <--> e41 4.442 .078
e32 <--> e36 4.430 .067
e31 <--> CSE 6.344 -.104
e31 <--> e39 35.597 .235
e31 <--> e43 22.890 -.188
e31 <--> e37 5.472 -.083
e31 <--> e34 4.346 -.074
e31 <--> e32 16.343 .212
e30 <--> e34 7.561 .086
e29 <--> e39 7.104 .131
e29 <--> e41 6.213 .119
e29 <--> e35 5.230 .119
e28 <--> CSE 16.321 .152
e28 <--> e40 9.108 -.088
e28 <--> e43 26.731 .186
e28 <--> e37 17.553 .137
e28 <--> e36 4.353 -.060
e28 <--> e31 10.718 -.156
e27 <--> CSE 4.029 .060
e27 <--> e40 6.950 .061
e27 <--> e43 9.632 .089
277
M.I. Par Change
e27 <--> e41 15.681 -.109
e27 <--> e31 5.307 -.088
e27 <--> e30 19.439 .148
e26 <--> CSE 7.247 -.072
e26 <--> e43 6.763 -.066
e26 <--> e42 5.285 -.046
e26 <--> e41 8.294 .071
e26 <--> e31 10.110 .108
e26 <--> e30 8.264 -.086
e25 <--> e34 4.511 .049
e25 <--> e28 6.242 .079
e24 <--> e42 5.298 -.045
e24 <--> e34 6.708 -.058
e24 <--> e33 5.359 -.075
e24 <--> e32 7.365 .091
e24 <--> e31 15.146 .129
e24 <--> e25 9.070 .064
e23 <--> e42 5.899 .053
e23 <--> e34 4.520 .053
e23 <--> e33 6.709 .093
e23 <--> e28 6.830 .088
e22 <--> e40 14.691 .097
e22 <--> e43 5.540 -.073
e22 <--> e42 5.347 -.057
e22 <--> e41 28.171 .160
e22 <--> e24 5.573 -.061
e22 <--> e23 6.142 .072
e21 <--> CSE 16.407 .146
e21 <--> e39 6.519 -.087
e21 <--> e42 8.952 .081
e21 <--> e36 4.166 .056
e21 <--> e26 5.669 -.070
e21 <--> e25 7.033 -.079
e20 <--> e43 4.753 .059
e20 <--> e29 4.382 .095
e20 <--> e28 6.924 .087
e20 <--> e23 9.464 .078
e19 <--> e39 5.310 -.050
e19 <--> e27 4.077 -.042
e19 <--> e26 12.329 -.066
e19 <--> e25 4.447 .040
e18 <--> e25 7.235 -.051
e17 <--> e33 4.725 -.058
278
M.I. Par Change
e17 <--> e32 11.474 .093
e17 <--> e27 7.779 .055
e17 <--> e23 6.276 -.048
e17 <--> e18 5.594 .035
e16 <--> e43 8.441 .067
e16 <--> e36 5.405 .042
e16 <--> e32 7.987 -.087
e16 <--> e23 4.378 .045
e16 <--> e22 4.314 -.050
e15 <--> e30 4.686 -.050
e14 <--> e40 5.586 -.052
e14 <--> e39 6.910 -.070
e14 <--> e38 4.065 .091
e14 <--> e37 5.263 .056
e14 <--> e35 9.214 -.086
e14 <--> e32 5.361 -.084
e14 <--> e31 4.987 -.080
e14 <--> e28 6.968 .086
e14 <--> e23 5.217 -.057
e13 <--> e38 5.235 .136
e13 <--> e34 8.122 -.091
e13 <--> e32 8.252 .137
e13 <--> e20 5.765 -.078
e13 <--> e19 4.571 -.056
e11 <--> e40 6.325 .060
e11 <--> e43 5.436 -.067
e11 <--> e42 9.056 -.070
e11 <--> e38 6.585 .126
e11 <--> e34 6.462 -.067
e11 <--> e31 5.481 .091
e11 <--> e29 6.037 -.120
e11 <--> e25 5.683 -.061
e11 <--> e24 8.076 .070
e11 <--> e16 5.958 .056
e11 <--> e13 16.220 .140
e10 <--> CSE 4.114 .060
e10 <--> e40 4.822 -.050
e10 <--> e39 4.386 -.059
e10 <--> e29 4.388 -.098
e10 <--> e14 4.692 .055
e10 <--> e11 4.844 .060
e9 <--> e40 5.574 -.064
e9 <--> e37 8.153 .086
279
M.I. Par Change
e9 <--> e33 4.336 -.090
e9 <--> e32 8.090 -.128
e9 <--> e31 4.794 -.097
e9 <--> e28 20.939 .186
e9 <--> e20 7.626 .085
e8 <--> e40 12.138 -.081
e8 <--> e38 10.361 -.154
e8 <--> e34 4.574 .055
e8 <--> e32 6.031 -.094
e8 <--> e31 16.132 -.153
e8 <--> e25 7.439 .068
e8 <--> e20 5.619 .062
e8 <--> e19 8.045 .059
e8 <--> e14 4.386 .054
e8 <--> e13 7.491 -.093
e8 <--> e11 9.318 -.085
e8 <--> e9 11.629 .109
e7 <--> e34 4.204 .046
e7 <--> e29 11.280 .141
e7 <--> e23 5.189 .054
e7 <--> e21 4.122 -.059
e7 <--> e20 5.062 .052
e7 <--> e11 6.611 -.063
e7 <--> e8 10.629 .078
e6 <--> e43 10.993 -.160
e6 <--> e32 6.722 -.168
e6 <--> e31 11.890 .222
e6 <--> e29 56.114 .601
e6 <--> e28 4.610 -.126
e6 <--> e27 9.977 -.147
e6 <--> e24 4.484 .086
e6 <--> e23 9.160 -.137
e6 <--> e16 11.577 -.127
e6 <--> e7 4.947 -.091
e5 <--> e40 6.498 -.058
e5 <--> e43 6.155 .069
e5 <--> e31 15.580 -.147
e5 <--> e25 4.659 -.053
e5 <--> e24 4.304 -.049
e5 <--> e10 8.571 .078
e5 <--> e8 4.588 .058
e5 <--> e6 4.901 .100
e4 <--> e40 8.478 .075
280
M.I. Par Change
e4 <--> e42 4.605 -.054
e4 <--> e33 17.675 .173
e4 <--> e27 6.018 -.075
e4 <--> e22 11.648 .114
e4 <--> e19 5.017 -.052
e4 <--> e11 7.875 .088
e4 <--> e8 4.383 -.064
e3 <--> e38 5.145 -.114
e3 <--> e36 6.289 .060
e3 <--> e32 5.537 .095
e3 <--> e28 8.880 -.109
e3 <--> e27 5.092 -.065
e2 <--> e43 7.029 .082
e2 <--> e41 4.632 -.063
e2 <--> e27 13.606 .110
e2 <--> e20 4.105 .058
e1 <--> e30 4.839 -.081
e1 <--> e11 6.291 -.078
Model Fit Summary
CMIN
Model NPAR CMIN DF P CMIN/DF
Default model 87 1781.099 654 .000 2.723
Saturated model 741 .000 0
Independence model 38 11224.961 703 .000 15.967
RMR, GFI
Model RMR GFI AGFI PGFI
Default model .188 .800 .773 .706
Saturated model .000 1.000
Independence model .576 .152 .106 .144
Baseline Comparisons
Model NFI
Delta1
RFI
rho1
IFI
Delta2
TLI
rho2 CFI
Default model .841 .829 .893 .885 .893
Saturated model 1.000 1.000 1.000
Independence model .000 .000 .000 .000 .000
281
Parsimony-Adjusted Measures
Model PRATIO PNFI PCFI
Default model .930 .783 .831
Saturated model .000 .000 .000
Independence model 1.000 .000 .000
NCP
Model NCP LO 90 HI 90
Default model 1127.099 1005.428 1256.393
Saturated model .000 .000 .000
Independence model 10521.961 10182.141 10868.198
FMIN
Model FMIN F0 LO 90 HI 90
Default model 4.509 2.853 2.545 3.181
Saturated model .000 .000 .000 .000
Independence model 28.418 26.638 25.778 27.514
RMSEA
Model RMSEA LO 90 HI 90 PCLOSE
Default model .066 .062 .070 .000
Independence model .195 .191 .198 .000
AIC
Model AIC BCC BIC CAIC
Default model 1955.099 1974.161 2301.482 2388.482
Saturated model 1482.000 1644.354 4432.228 5173.228
Independence model 11300.961 11309.287 11452.255 11490.255
ECVI
Model ECVI LO 90 HI 90 MECVI
Default model 4.950 4.642 5.277 4.998
Saturated model 3.752 3.752 3.752 4.163
Independence model 28.610 27.750 29.487 28.631
HOELTER
Model HOELTER
.05
HOELTER
.01
Default model 159 165
Independence model 27 28
282
Appendix H The Revised TAG Model’s Outputs
Assessment of normality (Group number 1)
Variable min max skew c.r. kurtosis c.r.
use6 2.000 7.000 -1.880 -15.274 4.500 18.278
use5 2.000 7.000 -1.990 -16.164 4.948 20.100
use4 1.000 7.000 -2.024 -16.447 4.592 18.653
peu2 1.000 7.000 -.808 -6.564 .308 1.250
peu4 1.000 7.000 -.545 -4.427 .056 .226
peu6 1.000 7.000 -.371 -3.012 -.189 -.769
peu7 1.000 7.000 -.574 -4.661 -.003 -.014
pu2 1.000 7.000 -.814 -6.613 .222 .901
pu3 1.000 7.000 -.527 -4.284 -.084 -.343
pu4 1.000 7.000 -.636 -5.165 .040 .163
pu5 1.000 7.000 -.494 -4.010 -.274 -1.112
pu6 1.000 7.000 -.844 -6.857 .421 1.712
pu7 1.000 7.000 -.664 -5.397 .055 .223
pu11 1.000 7.000 -.872 -7.081 .671 2.725
cse4 1.000 7.000 -1.021 -8.296 1.496 6.075
cse7 2.000 7.000 -1.025 -8.331 .865 3.514
cse9 2.000 7.000 -.759 -6.162 .363 1.474
cse10 2.000 7.000 -.758 -6.155 .302 1.227
cse11 2.000 7.000 -.980 -7.959 .587 2.383
cse12 2.000 7.000 -.882 -7.164 .964 3.917
int7 2.000 7.000 -.843 -6.849 .407 1.653
int5 2.000 7.000 -.962 -7.819 .566 2.298
int3 1.000 7.000 -1.866 -15.158 3.929 15.958
int2 1.000 7.000 -1.644 -13.353 3.022 12.276
int1 2.000 7.000 -1.065 -8.650 .605 2.458
grat9 1.000 7.000 -.900 -7.310 .825 3.352
grat6 1.000 7.000 -.917 -7.452 .642 2.608
grat4 1.000 7.000 -.951 -7.725 .945 3.837
grat3 1.000 7.000 -.893 -7.259 .649 2.637
Multivariate 431.540 101.261
Regression Weights: (Group number 1 - Default model)
Estimate S.E. C.R. P Label
PU <--- CSE .557 .058 9.602 *** par_24
PEU <--- CSE .588 .066 8.864 *** par_25
INT <--- PU .367 .073 5.009 *** par_26
INT <--- PEU .225 .071 3.154 .002 par_28
GRAT <--- PU .568 .077 7.407 *** par_27
283
Estimate S.E. C.R. P Label
GRAT <--- PEU .275 .069 3.994 *** par_29
USE <--- INT .419 .055 7.562 *** par_30
GRAT <--- INT .250 .059 4.206 *** par_31
grat3 <--- GRAT 1.000
grat4 <--- GRAT .970 .053 18.442 *** par_1
grat6 <--- GRAT 1.107 .056 19.835 *** par_2
grat9 <--- GRAT .985 .052 19.036 *** par_3
int1 <--- INT 1.000
int2 <--- INT .841 .052 16.289 *** par_4
int3 <--- INT .941 .061 15.528 *** par_5
int5 <--- INT 1.008 .057 17.667 *** par_6
int7 <--- INT .916 .058 15.763 *** par_7
cse12 <--- CSE 1.000
cse11 <--- CSE 1.018 .047 21.772 *** par_8
cse10 <--- CSE 1.071 .044 24.308 *** par_9
cse9 <--- CSE 1.110 .047 23.835 *** par_10
cse7 <--- CSE 1.000 .045 22.272 *** par_11
cse4 <--- CSE .855 .050 17.073 *** par_12
pu11 <--- PU 1.000
pu7 <--- PU 1.328 .083 15.975 *** par_13
pu6 <--- PU 1.294 .078 16.495 *** par_14
pu5 <--- PU 1.324 .078 17.038 *** par_15
pu4 <--- PU 1.313 .078 16.789 *** par_16
pu3 <--- PU 1.368 .080 17.026 *** par_17
pu2 <--- PU 1.229 .076 16.088 *** par_18
peu7 <--- PEU 1.000
peu6 <--- PEU 1.150 .083 13.812 *** par_19
peu4 <--- PEU 1.086 .083 13.104 *** par_20
peu2 <--- PEU .991 .077 12.923 *** par_21
use4 <--- USE 1.000
use5 <--- USE .968 .071 13.704 *** par_22
use6 <--- USE .841 .067 12.621 *** par_23
Standardized Regression Weights: (Group number 1 - Default model)
Estimate
PU <--- CSE .536
PEU <--- CSE .525
INT <--- PU .375
INT <--- PEU .247
GRAT <--- PU .494
GRAT <--- PEU .258
USE <--- INT .447
GRAT <--- INT .213
284
Estimate
grat3 <--- GRAT .798
grat4 <--- GRAT .800
grat6 <--- GRAT .843
grat9 <--- GRAT .822
int1 <--- INT .834
int2 <--- INT .723
int3 <--- INT .700
int5 <--- INT .788
int7 <--- INT .723
cse12 <--- CSE .870
cse11 <--- CSE .832
cse10 <--- CSE .881
cse9 <--- CSE .877
cse7 <--- CSE .843
cse4 <--- CSE .719
pu11 <--- PU .704
pu7 <--- PU .831
pu6 <--- PU .857
pu5 <--- PU .890
pu4 <--- PU .879
pu3 <--- PU .892
pu2 <--- PU .836
peu7 <--- PEU .746
peu6 <--- PEU .771
peu4 <--- PEU .740
peu2 <--- PEU .702
use4 <--- USE .725
use5 <--- USE .855
use6 <--- USE .717
Variances: (Group number 1 - Default model)
Estimate S.E. C.R. P Label
CSE .688 .064 10.773 *** par_32
e39 .530 .068 7.843 *** par_33
e40 .625 .081 7.758 *** par_34
e43 .534 .057 9.387 *** par_35
e41 .432 .052 8.368 *** par_36
e42 .501 .068 7.415 *** par_37
e1 .562 .050 11.343 *** par_38
e2 .519 .046 11.210 *** par_39
e3 .489 .048 10.174 *** par_40
e5 .457 .042 10.797 *** par_41
e7 .313 .032 9.655 *** par_42
285
Estimate S.E. C.R. P Label
e8 .461 .039 11.940 *** par_43
e9 .659 .054 12.234 *** par_44
e10 .443 .041 10.814 *** par_45
e11 .550 .046 11.842 *** par_46
e15 .222 .020 11.298 *** par_47
e16 .317 .026 12.070 *** par_48
e17 .228 .021 10.994 *** par_49
e18 .255 .023 11.097 *** par_50
e19 .279 .024 11.863 *** par_51
e20 .471 .036 13.119 *** par_52
e21 .757 .057 13.346 *** par_53
e22 .586 .047 12.426 *** par_54
e23 .450 .037 12.061 *** par_55
e24 .344 .030 11.349 *** par_56
e25 .376 .032 11.637 *** par_57
e26 .358 .032 11.291 *** par_58
e27 .486 .039 12.399 *** par_59
e30 .687 .065 10.565 *** par_60
e31 .781 .078 9.986 *** par_61
e32 .843 .079 10.718 *** par_62
e33 .873 .076 11.442 *** par_63
e35 .567 .053 10.608 *** par_64
e36 .217 .036 6.031 *** par_65
e37 .420 .040 10.575 *** par_66
Squared Multiple Correlations: (Group number 1 - Default model)
Estimate
PEU .276
PU .287
INT .253
USE .200
GRAT .560
use6 .514
use5 .730
use4 .525
peu2 .493
peu4 .547
peu6 .594
peu7 .557
pu2 .698
pu3 .796
pu4 .773
pu5 .791
286
Estimate
pu6 .735
pu7 .691
pu11 .496
cse4 .516
cse7 .711
cse9 .769
cse10 .776
cse11 .692
cse12 .756
int7 .522
int5 .622
int3 .490
int2 .523
int1 .696
grat9 .676
grat6 .711
grat4 .641
grat3 .636
Total Effects (Group number 1 - Default model)
CSE PEU PU INT USE GRAT
PEU .588 .000 .000 .000 .000 .000
PU .557 .000 .000 .000 .000 .000
INT .337 .225 .367 .000 .000 .000
USE .141 .094 .154 .419 .000 .000
GRAT .563 .331 .660 .250 .000 .000
use6 .119 .079 .129 .352 .841 .000
use5 .137 .091 .149 .405 .968 .000
use4 .141 .094 .154 .419 1.000 .000
peu2 .583 .991 .000 .000 .000 .000
peu4 .639 1.086 .000 .000 .000 .000
peu6 .677 1.150 .000 .000 .000 .000
peu7 .588 1.000 .000 .000 .000 .000
pu2 .685 .000 1.229 .000 .000 .000
pu3 .762 .000 1.368 .000 .000 .000
pu4 .732 .000 1.313 .000 .000 .000
pu5 .738 .000 1.324 .000 .000 .000
pu6 .721 .000 1.294 .000 .000 .000
pu7 .740 .000 1.328 .000 .000 .000
pu11 .557 .000 1.000 .000 .000 .000
cse4 .855 .000 .000 .000 .000 .000
cse7 1.000 .000 .000 .000 .000 .000
cse9 1.110 .000 .000 .000 .000 .000
287
CSE PEU PU INT USE GRAT
cse10 1.071 .000 .000 .000 .000 .000
cse11 1.018 .000 .000 .000 .000 .000
cse12 1.000 .000 .000 .000 .000 .000
int7 .309 .206 .337 .916 .000 .000
int5 .340 .227 .370 1.008 .000 .000
int3 .317 .212 .346 .941 .000 .000
int2 .283 .189 .309 .841 .000 .000
int1 .337 .225 .367 1.000 .000 .000
grat9 .554 .326 .650 .246 .000 .985
grat6 .623 .367 .731 .277 .000 1.107
grat4 .546 .321 .640 .242 .000 .970
grat3 .563 .331 .660 .250 .000 1.000
Standardized Total Effects (Group number 1 - Default model)
CSE PEU PU INT USE GRAT
PEU .525 .000 .000 .000 .000 .000
PU .536 .000 .000 .000 .000 .000
INT .331 .247 .375 .000 .000 .000
USE .148 .111 .168 .447 .000 .000
GRAT .471 .310 .574 .213 .000 .000
use6 .106 .079 .120 .321 .717 .000
use5 .126 .094 .143 .382 .855 .000
use4 .107 .080 .121 .324 .725 .000
peu2 .369 .702 .000 .000 .000 .000
peu4 .389 .740 .000 .000 .000 .000
peu6 .405 .771 .000 .000 .000 .000
peu7 .392 .746 .000 .000 .000 .000
pu2 .448 .000 .836 .000 .000 .000
pu3 .478 .000 .892 .000 .000 .000
pu4 .471 .000 .879 .000 .000 .000
pu5 .477 .000 .890 .000 .000 .000
pu6 .459 .000 .857 .000 .000 .000
pu7 .446 .000 .831 .000 .000 .000
pu11 .377 .000 .704 .000 .000 .000
cse4 .719 .000 .000 .000 .000 .000
cse7 .843 .000 .000 .000 .000 .000
cse9 .877 .000 .000 .000 .000 .000
cse10 .881 .000 .000 .000 .000 .000
cse11 .832 .000 .000 .000 .000 .000
cse12 .870 .000 .000 .000 .000 .000
int7 .239 .179 .271 .723 .000 .000
int5 .261 .195 .295 .788 .000 .000
int3 .231 .173 .262 .700 .000 .000
288
CSE PEU PU INT USE GRAT
int2 .239 .179 .271 .723 .000 .000
int1 .276 .206 .312 .834 .000 .000
grat9 .387 .255 .472 .175 .000 .822
grat6 .397 .262 .484 .180 .000 .843
grat4 .377 .248 .460 .171 .000 .800
grat3 .375 .248 .458 .170 .000 .798
Direct Effects (Group number 1 - Default model)
CSE PEU PU INT USE GRAT
PEU .588 .000 .000 .000 .000 .000
PU .557 .000 .000 .000 .000 .000
INT .000 .225 .367 .000 .000 .000
USE .000 .000 .000 .419 .000 .000
GRAT .000 .275 .568 .250 .000 .000
use6 .000 .000 .000 .000 .841 .000
use5 .000 .000 .000 .000 .968 .000
use4 .000 .000 .000 .000 1.000 .000
peu2 .000 .991 .000 .000 .000 .000
peu4 .000 1.086 .000 .000 .000 .000
peu6 .000 1.150 .000 .000 .000 .000
peu7 .000 1.000 .000 .000 .000 .000
pu2 .000 .000 1.229 .000 .000 .000
pu3 .000 .000 1.368 .000 .000 .000
pu4 .000 .000 1.313 .000 .000 .000
pu5 .000 .000 1.324 .000 .000 .000
pu6 .000 .000 1.294 .000 .000 .000
pu7 .000 .000 1.328 .000 .000 .000
pu11 .000 .000 1.000 .000 .000 .000
cse4 .855 .000 .000 .000 .000 .000
cse7 1.000 .000 .000 .000 .000 .000
cse9 1.110 .000 .000 .000 .000 .000
cse10 1.071 .000 .000 .000 .000 .000
cse11 1.018 .000 .000 .000 .000 .000
cse12 1.000 .000 .000 .000 .000 .000
int7 .000 .000 .000 .916 .000 .000
int5 .000 .000 .000 1.008 .000 .000
int3 .000 .000 .000 .941 .000 .000
int2 .000 .000 .000 .841 .000 .000
int1 .000 .000 .000 1.000 .000 .000
grat9 .000 .000 .000 .000 .000 .985
grat6 .000 .000 .000 .000 .000 1.107
grat4 .000 .000 .000 .000 .000 .970
grat3 .000 .000 .000 .000 .000 1.000
289
Standardized Direct Effects (Group number 1 - Default model)
CSE PEU PU INT USE GRAT
PEU .525 .000 .000 .000 .000 .000
PU .536 .000 .000 .000 .000 .000
INT .000 .247 .375 .000 .000 .000
USE .000 .000 .000 .447 .000 .000
GRAT .000 .258 .494 .213 .000 .000
use6 .000 .000 .000 .000 .717 .000
use5 .000 .000 .000 .000 .855 .000
use4 .000 .000 .000 .000 .725 .000
peu2 .000 .702 .000 .000 .000 .000
peu4 .000 .740 .000 .000 .000 .000
peu6 .000 .771 .000 .000 .000 .000
peu7 .000 .746 .000 .000 .000 .000
pu2 .000 .000 .836 .000 .000 .000
pu3 .000 .000 .892 .000 .000 .000
pu4 .000 .000 .879 .000 .000 .000
pu5 .000 .000 .890 .000 .000 .000
pu6 .000 .000 .857 .000 .000 .000
pu7 .000 .000 .831 .000 .000 .000
pu11 .000 .000 .704 .000 .000 .000
cse4 .719 .000 .000 .000 .000 .000
cse7 .843 .000 .000 .000 .000 .000
cse9 .877 .000 .000 .000 .000 .000
cse10 .881 .000 .000 .000 .000 .000
cse11 .832 .000 .000 .000 .000 .000
cse12 .870 .000 .000 .000 .000 .000
int7 .000 .000 .000 .723 .000 .000
int5 .000 .000 .000 .788 .000 .000
int3 .000 .000 .000 .700 .000 .000
int2 .000 .000 .000 .723 .000 .000
int1 .000 .000 .000 .834 .000 .000
grat9 .000 .000 .000 .000 .000 .822
grat6 .000 .000 .000 .000 .000 .843
grat4 .000 .000 .000 .000 .000 .800
grat3 .000 .000 .000 .000 .000 .798
Indirect Effects (Group number 1 - Default model)
CSE PEU PU INT USE GRAT
PEU .000 .000 .000 .000 .000 .000
PU .000 .000 .000 .000 .000 .000
INT .337 .000 .000 .000 .000 .000
USE .141 .094 .154 .000 .000 .000
290
CSE PEU PU INT USE GRAT
GRAT .563 .056 .092 .000 .000 .000
use6 .119 .079 .129 .352 .000 .000
use5 .137 .091 .149 .405 .000 .000
use4 .141 .094 .154 .419 .000 .000
peu2 .583 .000 .000 .000 .000 .000
peu4 .639 .000 .000 .000 .000 .000
peu6 .677 .000 .000 .000 .000 .000
peu7 .588 .000 .000 .000 .000 .000
pu2 .685 .000 .000 .000 .000 .000
pu3 .762 .000 .000 .000 .000 .000
pu4 .732 .000 .000 .000 .000 .000
pu5 .738 .000 .000 .000 .000 .000
pu6 .721 .000 .000 .000 .000 .000
pu7 .740 .000 .000 .000 .000 .000
pu11 .557 .000 .000 .000 .000 .000
cse4 .000 .000 .000 .000 .000 .000
cse7 .000 .000 .000 .000 .000 .000
cse9 .000 .000 .000 .000 .000 .000
cse10 .000 .000 .000 .000 .000 .000
cse11 .000 .000 .000 .000 .000 .000
cse12 .000 .000 .000 .000 .000 .000
int7 .309 .206 .337 .000 .000 .000
int5 .340 .227 .370 .000 .000 .000
int3 .317 .212 .346 .000 .000 .000
int2 .283 .189 .309 .000 .000 .000
int1 .337 .225 .367 .000 .000 .000
grat9 .554 .326 .650 .246 .000 .000
grat6 .623 .367 .731 .277 .000 .000
grat4 .546 .321 .640 .242 .000 .000
grat3 .563 .331 .660 .250 .000 .000
Standardized Indirect Effects (Group number 1 - Default model)
CSE PEU PU INT USE GRAT
PEU .000 .000 .000 .000 .000 .000
PU .000 .000 .000 .000 .000 .000
INT .331 .000 .000 .000 .000 .000
USE .148 .111 .168 .000 .000 .000
GRAT .471 .053 .080 .000 .000 .000
use6 .106 .079 .120 .321 .000 .000
use5 .126 .094 .143 .382 .000 .000
use4 .107 .080 .121 .324 .000 .000
peu2 .369 .000 .000 .000 .000 .000
peu4 .389 .000 .000 .000 .000 .000
291
CSE PEU PU INT USE GRAT
peu6 .405 .000 .000 .000 .000 .000
peu7 .392 .000 .000 .000 .000 .000
pu2 .448 .000 .000 .000 .000 .000
pu3 .478 .000 .000 .000 .000 .000
pu4 .471 .000 .000 .000 .000 .000
pu5 .477 .000 .000 .000 .000 .000
pu6 .459 .000 .000 .000 .000 .000
pu7 .446 .000 .000 .000 .000 .000
pu11 .377 .000 .000 .000 .000 .000
cse4 .000 .000 .000 .000 .000 .000
cse7 .000 .000 .000 .000 .000 .000
cse9 .000 .000 .000 .000 .000 .000
cse10 .000 .000 .000 .000 .000 .000
cse11 .000 .000 .000 .000 .000 .000
cse12 .000 .000 .000 .000 .000 .000
int7 .239 .179 .271 .000 .000 .000
int5 .261 .195 .295 .000 .000 .000
int3 .231 .173 .262 .000 .000 .000
int2 .239 .179 .271 .000 .000 .000
int1 .276 .206 .312 .000 .000 .000
grat9 .387 .255 .472 .175 .000 .000
grat6 .397 .262 .484 .180 .000 .000
grat4 .377 .248 .460 .171 .000 .000
grat3 .375 .248 .458 .170 .000 .000
Modification Indices (Group number 1 - Default model)
Covariances: (Group number 1 - Default model)
M.I. Par Change
e39 <--> e40 138.247 .405
e43 <--> CSE 59.435 .266
e43 <--> e40 27.557 -.190
e43 <--> e39 24.673 -.153
e42 <--> CSE 11.188 .115
e33 <--> e39 27.440 .204
e31 <--> CSE 4.103 -.088
e31 <--> e39 38.408 .239
e31 <--> e43 18.723 -.175
e31 <--> e32 4.734 .109
e30 <--> e32 4.219 -.095
e27 <--> e40 6.720 .087
e27 <--> e43 11.115 .099
e27 <--> e41 10.739 -.093
292
M.I. Par Change
e27 <--> e31 6.742 -.097
e27 <--> e30 20.072 .154
e26 <--> CSE 7.121 -.076
e26 <--> e40 4.559 .064
e26 <--> e43 9.081 -.080
e26 <--> e41 7.428 .069
e26 <--> e31 9.890 .105
e26 <--> e30 11.014 -.102
e24 <--> e33 7.350 -.089
e24 <--> e32 5.391 .077
e24 <--> e31 14.227 .123
e24 <--> e25 9.175 .065
e23 <--> e42 4.175 .059
e23 <--> e33 9.432 .113
e22 <--> e40 14.242 .138
e22 <--> e43 4.167 -.067
e22 <--> e42 4.235 -.067
e22 <--> e41 23.213 .149
e22 <--> e27 4.144 -.061
e22 <--> e24 5.276 -.060
e22 <--> e23 6.315 .073
e21 <--> CSE 16.488 .156
e21 <--> e39 6.825 -.089
e21 <--> e42 11.634 .123
e21 <--> e36 4.064 .057
e21 <--> e26 5.311 -.068
e21 <--> e25 6.807 -.078
e20 <--> e43 7.838 .080
e20 <--> e26 4.640 -.051
e20 <--> e23 8.341 .074
e19 <--> e39 6.085 -.054
e19 <--> e41 4.479 .046
e19 <--> e27 4.321 -.044
e19 <--> e26 14.087 -.071
e19 <--> e25 5.249 .044
e18 <--> e25 6.206 -.047
e17 <--> e33 5.645 -.065
e17 <--> e32 10.280 .088
e17 <--> e31 6.073 -.067
e17 <--> e27 7.835 .056
e17 <--> e23 7.802 -.054
e16 <--> e43 8.489 .071
e16 <--> e36 4.250 .039
293
M.I. Par Change
e16 <--> e33 4.817 .068
e16 <--> e32 8.629 -.091
e16 <--> e22 4.462 -.052
e15 <--> e30 4.947 -.054
e11 <--> CSE 7.263 .092
e11 <--> e40 14.229 .135
e11 <--> e43 7.113 -.084
e11 <--> e42 4.645 -.069
e11 <--> e25 4.369 -.056
e11 <--> e24 7.167 .069
e11 <--> e19 4.004 -.046
e11 <--> e16 7.396 .065
e10 <--> CSE 5.808 .077
e10 <--> e11 12.174 .101
e9 <--> e40 8.579 -.114
e9 <--> e37 7.960 .087
e9 <--> e36 4.857 -.060
e9 <--> e20 5.681 .073
e8 <--> e40 14.499 -.125
e8 <--> e31 12.703 -.131
e8 <--> e25 8.432 .071
e8 <--> e24 4.254 -.049
e8 <--> e20 4.848 .057
e8 <--> e19 8.363 .060
e8 <--> e11 7.545 -.079
e8 <--> e9 7.027 .083
e7 <--> e23 4.898 .052
e7 <--> e21 4.368 -.061
e5 <--> e31 8.752 -.113
e5 <--> e25 4.247 -.053
e5 <--> e10 8.195 .081
e3 <--> CSE 5.332 -.080
e3 <--> e43 5.025 -.073
e3 <--> e27 11.615 -.102
e3 <--> e22 8.883 .097
e2 <--> e27 9.149 .089
e1 <--> e30 4.702 -.083
e1 <--> e11 4.345 -.068
Regression Weights: (Group number 1 - Default model)
M.I. Par Change
PEU <--- PU 94.519 .522
PU <--- PEU 91.573 .430
294
M.I. Par Change
INT <--- CSE 59.435 .387
USE <--- CSE 11.188 .168
use6 <--- int3 8.396 .091
use5 <--- cse11 4.302 .064
use5 <--- int3 4.162 -.056
use4 <--- pu6 4.561 .069
peu2 <--- PU 10.711 .198
peu2 <--- GRAT 7.821 .151
peu2 <--- pu2 10.560 .131
peu2 <--- pu3 11.211 .129
peu2 <--- pu4 5.221 .090
peu2 <--- pu6 17.764 .165
peu2 <--- pu7 13.426 .136
peu2 <--- pu11 7.955 .117
peu2 <--- cse7 5.263 -.119
peu2 <--- cse9 4.184 -.099
peu2 <--- cse10 7.318 -.137
peu2 <--- cse12 4.523 -.114
peu2 <--- grat9 8.441 .125
peu2 <--- grat4 4.935 .094
peu2 <--- grat3 7.648 .113
peu6 <--- CSE 4.103 -.127
peu6 <--- PU 16.359 .243
peu6 <--- INT 4.216 -.131
peu6 <--- pu3 23.117 .184
peu6 <--- pu4 13.890 .146
peu6 <--- pu5 25.763 .200
peu6 <--- pu6 7.725 .108
peu6 <--- pu7 15.572 .145
peu6 <--- pu11 10.194 .132
peu6 <--- cse7 5.415 -.120
peu6 <--- cse10 7.710 -.139
peu6 <--- cse11 6.906 -.131
peu6 <--- int5 4.206 -.096
peu6 <--- int3 4.804 -.097
peu6 <--- int2 13.670 -.190
peu7 <--- PU 4.486 .117
peu7 <--- pu2 16.422 .148
peu7 <--- pu4 5.045 .081
peu7 <--- pu6 4.248 .074
peu7 <--- pu7 5.645 .080
peu7 <--- cse11 4.413 .096
pu2 <--- PEU 10.086 .138
295
M.I. Par Change
pu2 <--- INT 12.650 .166
pu2 <--- USE 8.905 .153
pu2 <--- use6 7.863 .112
pu2 <--- use5 6.129 .103
pu2 <--- peu2 7.740 .079
pu2 <--- peu4 5.059 .061
pu2 <--- peu7 23.555 .145
pu2 <--- cse10 8.003 .104
pu2 <--- int7 7.175 .093
pu2 <--- int5 5.486 .080
pu2 <--- int3 11.423 .110
pu2 <--- int2 10.862 .125
pu2 <--- int1 10.436 .118
pu3 <--- CSE 7.121 -.110
pu3 <--- INT 6.056 -.103
pu3 <--- USE 7.038 -.122
pu3 <--- use5 6.211 -.092
pu3 <--- use4 5.628 -.072
pu3 <--- cse4 10.974 -.112
pu3 <--- cse7 16.424 -.137
pu3 <--- cse11 4.255 -.068
pu3 <--- cse12 6.760 -.091
pu3 <--- int3 8.442 -.085
pu3 <--- int2 7.247 -.091
pu3 <--- int1 6.805 -.086
pu4 <--- int2 4.056 .069
pu5 <--- peu4 4.813 .052
pu5 <--- peu6 8.596 .069
pu5 <--- int2 4.740 -.072
pu6 <--- USE 4.334 .104
pu6 <--- use5 4.318 .084
pu6 <--- use4 6.415 .084
pu6 <--- cse10 4.239 -.074
pu7 <--- PEU 4.261 .098
pu7 <--- USE 5.440 -.131
pu7 <--- GRAT 9.993 .137
pu7 <--- use6 6.739 -.114
pu7 <--- use5 4.649 -.098
pu7 <--- peu2 6.705 .080
pu7 <--- peu6 4.701 .064
pu7 <--- peu7 4.672 .071
pu7 <--- cse11 6.546 -.103
pu7 <--- int2 4.290 -.086
296
M.I. Par Change
pu7 <--- grat9 8.931 .102
pu7 <--- grat6 15.425 .123
pu7 <--- grat3 11.140 .109
pu11 <--- CSE 16.488 .226
pu11 <--- USE 7.778 .173
pu11 <--- use5 8.753 .148
pu11 <--- use4 5.388 .095
pu11 <--- cse4 5.591 .107
pu11 <--- cse7 17.611 .191
pu11 <--- cse9 15.610 .169
pu11 <--- cse10 10.724 .146
pu11 <--- cse11 16.042 .177
pu11 <--- cse12 14.522 .179
cse4 <--- INT 5.518 .105
cse4 <--- int3 9.935 .099
cse4 <--- int2 9.117 .110
cse4 <--- int1 6.955 .093
cse7 <--- PU 4.174 -.070
cse7 <--- peu4 4.591 -.045
cse7 <--- pu2 7.537 -.062
cse7 <--- pu3 10.981 -.072
cse7 <--- pu5 5.025 -.050
cse7 <--- int2 4.703 .063
cse11 <--- INT 4.226 .078
cse11 <--- USE 5.407 .098
cse11 <--- use5 6.833 .088
cse11 <--- peu4 4.494 -.047
cse11 <--- int7 9.954 .089
int7 <--- CSE 7.263 .134
int7 <--- PEU 20.842 .213
int7 <--- PU 8.866 .141
int7 <--- peu2 13.089 .110
int7 <--- peu4 16.157 .118
int7 <--- peu6 17.520 .120
int7 <--- peu7 13.327 .117
int7 <--- pu2 6.960 .083
int7 <--- pu3 7.912 .085
int7 <--- pu5 13.655 .115
int7 <--- pu6 4.692 .066
int7 <--- pu7 10.317 .093
int7 <--- pu11 5.239 .075
int7 <--- cse9 6.649 .098
int7 <--- cse10 4.164 .081
297
M.I. Par Change
int7 <--- cse11 12.844 .141
int7 <--- cse12 5.222 .096
int7 <--- grat4 4.193 .068
int5 <--- CSE 5.808 .111
int5 <--- cse7 5.554 .089
int5 <--- cse9 7.835 .099
int5 <--- cse10 5.993 .090
int5 <--- cse11 4.137 .074
int5 <--- cse12 4.190 .080
int5 <--- int7 5.255 .079
int5 <--- grat9 4.518 .067
int3 <--- PEU 5.380 -.118
int3 <--- peu2 5.853 -.080
int3 <--- peu4 7.335 -.086
int3 <--- peu6 5.481 -.073
int2 <--- PEU 11.988 -.148
int2 <--- GRAT 4.029 -.078
int2 <--- peu4 11.278 -.090
int2 <--- peu6 19.795 -.117
int2 <--- peu7 4.072 -.059
int2 <--- pu3 4.207 -.057
int2 <--- pu5 5.096 -.064
int2 <--- pu7 5.419 -.062
int2 <--- grat6 5.076 -.063
int2 <--- grat3 4.592 -.063
int1 <--- cse4 4.498 .070
grat9 <--- peu6 6.144 -.068
grat9 <--- int5 5.275 .081
grat6 <--- CSE 5.332 -.116
grat6 <--- INT 4.225 -.104
grat6 <--- pu2 4.987 -.071
grat6 <--- cse4 7.546 -.112
grat6 <--- cse9 4.231 -.079
grat6 <--- cse10 4.992 -.089
grat6 <--- cse11 7.795 -.111
grat6 <--- int5 4.943 -.083
grat6 <--- int2 5.236 -.094
grat4 <--- INT 4.580 .107
grat4 <--- pu2 5.759 .075
grat4 <--- int7 5.365 .086
298
Model Fit Summary
CMIN
Model NPAR CMIN DF P CMIN/DF
Default model 66 1095.131 369 .000 2.968
Saturated model 435 .000 0
Independence model 29 8860.661 406 .000 21.824
RMR, GFI
Model RMR GFI AGFI PGFI
Default model .186 .838 .809 .711
Saturated model .000 1.000
Independence model .592 .165 .105 .154
Baseline Comparisons
Model NFI
Delta1
RFI
rho1
IFI
Delta2
TLI
rho2 CFI
Default model .876 .864 .914 .906 .914
Saturated model 1.000 1.000 1.000
Independence model .000 .000 .000 .000 .000
Parsimony-Adjusted Measures
Model PRATIO PNFI PCFI
Default model .909 .797 .831
Saturated model .000 .000 .000
Independence model 1.000 .000 .000
NCP
Model NCP LO 90 HI 90
Default model 726.131 630.633 829.246
Saturated model .000 .000 .000
Independence model 8454.661 8151.764 8763.937
FMIN
Model FMIN F0 LO 90 HI 90
Default model 2.772 1.838 1.597 2.099
Saturated model .000 .000 .000 .000
Independence model 22.432 21.404 20.637 22.187
299
RMSEA
Model RMSEA LO 90 HI 90 PCLOSE
Default model .071 .066 .075 .000
Independence model .230 .225 .234 .000
AIC
Model AIC BCC BIC CAIC
Default model 1227.131 1237.980 1489.904 1555.904
Saturated model 870.000 941.507 2601.915 3036.915
Independence model 8918.661 8923.428 9034.122 9063.122
ECVI
Model ECVI LO 90 HI 90 MECVI
Default model 3.107 2.865 3.368 3.134
Saturated model 2.203 2.203 2.203 2.384
Independence model 22.579 21.812 23.362 22.591
HOELTER
Model HOELTER
.05
HOELTER
.01
Default model 150 157
Independence model 21 22
300
Appendix I The Outputs of Invariance Analyses
“The Outputs of the Revised TAG Model for Malaysian Participants”
Estimates (Group number 1 - Default model)
Scalar Estimates (Group number 1 - Default model)
Maximum Likelihood Estimates
Regression Weights: (Group number 1 - Default model)
Estimate S.E. C.R. P Label
PU <--- CSE .625 .077 8.109 *** par_24
PEU <--- CSE .715 .097 7.376 *** par_25
INT <--- PU .521 .090 5.764 *** par_26
INT <--- PEU .041 .068 .603 .547 par_28
GRAT <--- PU .503 .112 4.490 *** par_27
GRAT <--- PEU .257 .076 3.381 *** par_29
USE <--- INT .202 .052 3.883 *** par_30
GRAT <--- INT .405 .100 4.061 *** par_31
grat3 <--- GRAT 1.000
grat4 <--- GRAT .851 .066 12.968 *** par_1
grat6 <--- GRAT 1.075 .062 17.461 *** par_2
grat9 <--- GRAT .889 .060 14.870 *** par_3
int1 <--- INT 1.000
int2 <--- INT .886 .064 13.896 *** par_4
int3 <--- INT 1.001 .087 11.552 *** par_5
int5 <--- INT 1.102 .087 12.660 *** par_6
int7 <--- INT .973 .099 9.832 *** par_7
cse12 <--- CSE 1.000
cse11 <--- CSE .901 .068 13.334 *** par_8
cse10 <--- CSE 1.134 .065 17.486 *** par_9
cse9 <--- CSE 1.125 .069 16.356 *** par_10
cse7 <--- CSE .946 .064 14.770 *** par_11
cse4 <--- CSE .820 .072 11.306 *** par_12
pu11 <--- PU 1.000
pu7 <--- PU 1.288 .109 11.805 *** par_13
pu6 <--- PU 1.099 .091 12.079 *** par_14
pu5 <--- PU 1.396 .108 12.943 *** par_15
pu4 <--- PU 1.313 .103 12.771 *** par_16
pu3 <--- PU 1.390 .110 12.603 *** par_17
pu2 <--- PU 1.093 .092 11.846 *** par_18
peu7 <--- PEU 1.000
peu6 <--- PEU 1.131 .094 12.000 *** par_19
301
Estimate S.E. C.R. P Label
peu4 <--- PEU 1.064 .091 11.749 *** par_20
peu2 <--- PEU 1.018 .084 12.184 *** par_21
use4 <--- USE 1.000
use5 <--- USE 1.024 .094 10.861 *** par_22
use6 <--- USE 1.010 .109 9.273 *** par_23
Standardized Regression Weights: (Group number 1 - Default model)
Estimate
PU <--- CSE .629
PEU <--- CSE .570
INT <--- PU .563
INT <--- PEU .056
GRAT <--- PU .404
GRAT <--- PEU .260
USE <--- INT .318
GRAT <--- INT .301
grat3 <--- GRAT .860
grat4 <--- GRAT .746
grat6 <--- GRAT .894
grat9 <--- GRAT .825
int1 <--- INT .860
int2 <--- INT .808
int3 <--- INT .719
int5 <--- INT .781
int7 <--- INT .649
cse12 <--- CSE .865
cse11 <--- CSE .767
cse10 <--- CSE .898
cse9 <--- CSE .869
cse7 <--- CSE .816
cse4 <--- CSE .688
pu11 <--- PU .742
pu7 <--- PU .809
pu6 <--- PU .822
pu5 <--- PU .880
pu4 <--- PU .873
pu3 <--- PU .867
pu2 <--- PU .811
peu7 <--- PEU .800
peu6 <--- PEU .809
peu4 <--- PEU .782
302
Estimate
peu2 <--- PEU .817
use4 <--- USE .815
use5 <--- USE .848
use6 <--- USE .683
Variances: (Group number 1 - Default model)
Estimate S.E. C.R. P Label
CSE .651 .086 7.567 *** par_32
e39 .390 .067 5.838 *** par_33
e40 .692 .112 6.168 *** par_34
e43 .362 .053 6.818 *** par_35
e41 .422 .065 6.447 *** par_36
e42 .201 .032 6.240 *** par_37
e1 .351 .048 7.338 *** par_38
e2 .578 .066 8.730 *** par_39
e3 .291 .046 6.308 *** par_40
e5 .370 .047 7.944 *** par_41
e7 .194 .029 6.586 *** par_42
e8 .230 .030 7.726 *** par_43
e9 .514 .059 8.721 *** par_44
e10 .427 .053 8.000 *** par_45
e11 .716 .080 8.997 *** par_46
e15 .219 .028 7.968 *** par_47
e16 .371 .041 9.028 *** par_48
e17 .202 .028 7.211 *** par_49
e18 .268 .034 7.911 *** par_50
e19 .292 .034 8.624 *** par_51
e20 .487 .052 9.354 *** par_52
e21 .524 .057 9.250 *** par_53
e22 .565 .064 8.894 *** par_54
e23 .374 .043 8.754 *** par_55
e24 .364 .045 8.006 *** par_56
e25 .345 .043 8.080 *** par_57
e26 .412 .050 8.212 *** par_58
e27 .400 .045 8.841 *** par_59
e30 .575 .076 7.588 *** par_60
e31 .692 .093 7.414 *** par_61
e32 .735 .094 7.805 *** par_62
e33 .527 .073 7.237 *** par_63
e35 .112 .020 5.765 *** par_64
e36 .091 .019 4.784 *** par_65
303
Estimate S.E. C.R. P Label
e37 .260 .032 8.226 *** par_66
Squared Multiple Correlations: (Group number 1 - Default model)
Estimate
PEU .325
PU .395
INT .343
USE .101
GRAT .578
use6 .467
use5 .719
use4 .665
peu2 .668
peu4 .612
peu6 .654
peu7 .640
pu2 .658
pu3 .751
pu4 .763
pu5 .775
pu6 .675
pu7 .654
pu11 .551
cse4 .473
cse7 .666
cse9 .755
cse10 .806
cse11 .588
cse12 .748
int7 .421
int5 .610
int3 .518
int2 .653
int1 .740
grat9 .681
grat6 .799
grat4 .556
grat3 .740
304
Matrices (Group number 1 - Default model)
Total Effects (Group number 1 - Default model)
CSE PEU PU INT USE GRAT
PEU .715 .000 .000 .000 .000 .000
PU .625 .000 .000 .000 .000 .000
INT .355 .041 .521 .000 .000 .000
USE .072 .008 .105 .202 .000 .000
GRAT .642 .273 .714 .405 .000 .000
use6 .072 .008 .106 .204 1.010 .000
use5 .074 .008 .108 .207 1.024 .000
use4 .072 .008 .105 .202 1.000 .000
peu2 .727 1.018 .000 .000 .000 .000
peu4 .760 1.064 .000 .000 .000 .000
peu6 .809 1.131 .000 .000 .000 .000
peu7 .715 1.000 .000 .000 .000 .000
pu2 .683 .000 1.093 .000 .000 .000
pu3 .869 .000 1.390 .000 .000 .000
pu4 .821 .000 1.313 .000 .000 .000
pu5 .873 .000 1.396 .000 .000 .000
pu6 .687 .000 1.099 .000 .000 .000
pu7 .806 .000 1.288 .000 .000 .000
pu11 .625 .000 1.000 .000 .000 .000
cse4 .820 .000 .000 .000 .000 .000
cse7 .946 .000 .000 .000 .000 .000
cse9 1.125 .000 .000 .000 .000 .000
cse10 1.134 .000 .000 .000 .000 .000
cse11 .901 .000 .000 .000 .000 .000
cse12 1.000 .000 .000 .000 .000 .000
int7 .345 .040 .507 .973 .000 .000
int5 .391 .045 .574 1.102 .000 .000
int3 .355 .041 .521 1.001 .000 .000
int2 .315 .036 .462 .886 .000 .000
int1 .355 .041 .521 1.000 .000 .000
grat9 .571 .243 .635 .360 .000 .889
grat6 .690 .294 .768 .436 .000 1.075
grat4 .546 .233 .608 .345 .000 .851
grat3 .642 .273 .714 .405 .000 1.000
305
Standardized Total Effects (Group number 1 - Default model)
CSE PEU PU INT USE GRAT
PEU .570 .000 .000 .000 .000 .000
PU .629 .000 .000 .000 .000 .000
INT .386 .056 .563 .000 .000 .000
USE .123 .018 .179 .318 .000 .000
GRAT .518 .277 .573 .301 .000 .000
use6 .084 .012 .122 .217 .683 .000
use5 .104 .015 .152 .269 .848 .000
use4 .100 .014 .146 .259 .815 .000
peu2 .466 .817 .000 .000 .000 .000
peu4 .446 .782 .000 .000 .000 .000
peu6 .461 .809 .000 .000 .000 .000
peu7 .456 .800 .000 .000 .000 .000
pu2 .510 .000 .811 .000 .000 .000
pu3 .545 .000 .867 .000 .000 .000
pu4 .549 .000 .873 .000 .000 .000
pu5 .553 .000 .880 .000 .000 .000
pu6 .517 .000 .822 .000 .000 .000
pu7 .508 .000 .809 .000 .000 .000
pu11 .467 .000 .742 .000 .000 .000
cse4 .688 .000 .000 .000 .000 .000
cse7 .816 .000 .000 .000 .000 .000
cse9 .869 .000 .000 .000 .000 .000
cse10 .898 .000 .000 .000 .000 .000
cse11 .767 .000 .000 .000 .000 .000
cse12 .865 .000 .000 .000 .000 .000
int7 .251 .036 .366 .649 .000 .000
int5 .302 .044 .440 .781 .000 .000
int3 .278 .040 .405 .719 .000 .000
int2 .312 .045 .455 .808 .000 .000
int1 .332 .048 .485 .860 .000 .000
grat9 .427 .228 .473 .248 .000 .825
grat6 .463 .247 .512 .269 .000 .894
grat4 .386 .206 .427 .224 .000 .746
grat3 .445 .238 .493 .259 .000 .860
Direct Effects (Group number 1 - Default model)
CSE PEU PU INT USE GRAT
PEU .715 .000 .000 .000 .000 .000
PU .625 .000 .000 .000 .000 .000
INT .000 .041 .521 .000 .000 .000
USE .000 .000 .000 .202 .000 .000
306
CSE PEU PU INT USE GRAT
GRAT .000 .257 .503 .405 .000 .000
use6 .000 .000 .000 .000 1.010 .000
use5 .000 .000 .000 .000 1.024 .000
use4 .000 .000 .000 .000 1.000 .000
peu2 .000 1.018 .000 .000 .000 .000
peu4 .000 1.064 .000 .000 .000 .000
peu6 .000 1.131 .000 .000 .000 .000
peu7 .000 1.000 .000 .000 .000 .000
pu2 .000 .000 1.093 .000 .000 .000
pu3 .000 .000 1.390 .000 .000 .000
pu4 .000 .000 1.313 .000 .000 .000
pu5 .000 .000 1.396 .000 .000 .000
pu6 .000 .000 1.099 .000 .000 .000
pu7 .000 .000 1.288 .000 .000 .000
pu11 .000 .000 1.000 .000 .000 .000
cse4 .820 .000 .000 .000 .000 .000
cse7 .946 .000 .000 .000 .000 .000
cse9 1.125 .000 .000 .000 .000 .000
cse10 1.134 .000 .000 .000 .000 .000
cse11 .901 .000 .000 .000 .000 .000
cse12 1.000 .000 .000 .000 .000 .000
int7 .000 .000 .000 .973 .000 .000
int5 .000 .000 .000 1.102 .000 .000
int3 .000 .000 .000 1.001 .000 .000
int2 .000 .000 .000 .886 .000 .000
int1 .000 .000 .000 1.000 .000 .000
grat9 .000 .000 .000 .000 .000 .889
grat6 .000 .000 .000 .000 .000 1.075
grat4 .000 .000 .000 .000 .000 .851
grat3 .000 .000 .000 .000 .000 1.000
Standardized Direct Effects (Group number 1 - Default model)
CSE PEU PU INT USE GRAT
PEU .570 .000 .000 .000 .000 .000
PU .629 .000 .000 .000 .000 .000
INT .000 .056 .563 .000 .000 .000
USE .000 .000 .000 .318 .000 .000
GRAT .000 .260 .404 .301 .000 .000
use6 .000 .000 .000 .000 .683 .000
use5 .000 .000 .000 .000 .848 .000
use4 .000 .000 .000 .000 .815 .000
peu2 .000 .817 .000 .000 .000 .000
peu4 .000 .782 .000 .000 .000 .000
307
CSE PEU PU INT USE GRAT
peu6 .000 .809 .000 .000 .000 .000
peu7 .000 .800 .000 .000 .000 .000
pu2 .000 .000 .811 .000 .000 .000
pu3 .000 .000 .867 .000 .000 .000
pu4 .000 .000 .873 .000 .000 .000
pu5 .000 .000 .880 .000 .000 .000
pu6 .000 .000 .822 .000 .000 .000
pu7 .000 .000 .809 .000 .000 .000
pu11 .000 .000 .742 .000 .000 .000
cse4 .688 .000 .000 .000 .000 .000
cse7 .816 .000 .000 .000 .000 .000
cse9 .869 .000 .000 .000 .000 .000
cse10 .898 .000 .000 .000 .000 .000
cse11 .767 .000 .000 .000 .000 .000
cse12 .865 .000 .000 .000 .000 .000
int7 .000 .000 .000 .649 .000 .000
int5 .000 .000 .000 .781 .000 .000
int3 .000 .000 .000 .719 .000 .000
int2 .000 .000 .000 .808 .000 .000
int1 .000 .000 .000 .860 .000 .000
grat9 .000 .000 .000 .000 .000 .825
grat6 .000 .000 .000 .000 .000 .894
grat4 .000 .000 .000 .000 .000 .746
grat3 .000 .000 .000 .000 .000 .860
Indirect Effects (Group number 1 - Default model)
CSE PEU PU INT USE GRAT
PEU .000 .000 .000 .000 .000 .000
PU .000 .000 .000 .000 .000 .000
INT .355 .000 .000 .000 .000 .000
USE .072 .008 .105 .000 .000 .000
GRAT .642 .017 .211 .000 .000 .000
use6 .072 .008 .106 .204 .000 .000
use5 .074 .008 .108 .207 .000 .000
use4 .072 .008 .105 .202 .000 .000
peu2 .727 .000 .000 .000 .000 .000
peu4 .760 .000 .000 .000 .000 .000
peu6 .809 .000 .000 .000 .000 .000
peu7 .715 .000 .000 .000 .000 .000
pu2 .683 .000 .000 .000 .000 .000
pu3 .869 .000 .000 .000 .000 .000
pu4 .821 .000 .000 .000 .000 .000
pu5 .873 .000 .000 .000 .000 .000
308
CSE PEU PU INT USE GRAT
pu6 .687 .000 .000 .000 .000 .000
pu7 .806 .000 .000 .000 .000 .000
pu11 .625 .000 .000 .000 .000 .000
cse4 .000 .000 .000 .000 .000 .000
cse7 .000 .000 .000 .000 .000 .000
cse9 .000 .000 .000 .000 .000 .000
cse10 .000 .000 .000 .000 .000 .000
cse11 .000 .000 .000 .000 .000 .000
cse12 .000 .000 .000 .000 .000 .000
int7 .345 .040 .507 .000 .000 .000
int5 .391 .045 .574 .000 .000 .000
int3 .355 .041 .521 .000 .000 .000
int2 .315 .036 .462 .000 .000 .000
int1 .355 .041 .521 .000 .000 .000
grat9 .571 .243 .635 .360 .000 .000
grat6 .690 .294 .768 .436 .000 .000
grat4 .546 .233 .608 .345 .000 .000
grat3 .642 .273 .714 .405 .000 .000
Standardized Indirect Effects (Group number 1 - Default model)
CSE PEU PU INT USE GRAT
PEU .000 .000 .000 .000 .000 .000
PU .000 .000 .000 .000 .000 .000
INT .386 .000 .000 .000 .000 .000
USE .123 .018 .179 .000 .000 .000
GRAT .518 .017 .170 .000 .000 .000
use6 .084 .012 .122 .217 .000 .000
use5 .104 .015 .152 .269 .000 .000
use4 .100 .014 .146 .259 .000 .000
peu2 .466 .000 .000 .000 .000 .000
peu4 .446 .000 .000 .000 .000 .000
peu6 .461 .000 .000 .000 .000 .000
peu7 .456 .000 .000 .000 .000 .000
pu2 .510 .000 .000 .000 .000 .000
pu3 .545 .000 .000 .000 .000 .000
pu4 .549 .000 .000 .000 .000 .000
pu5 .553 .000 .000 .000 .000 .000
pu6 .517 .000 .000 .000 .000 .000
pu7 .508 .000 .000 .000 .000 .000
pu11 .467 .000 .000 .000 .000 .000
cse4 .000 .000 .000 .000 .000 .000
cse7 .000 .000 .000 .000 .000 .000
cse9 .000 .000 .000 .000 .000 .000
309
CSE PEU PU INT USE GRAT
cse10 .000 .000 .000 .000 .000 .000
cse11 .000 .000 .000 .000 .000 .000
cse12 .000 .000 .000 .000 .000 .000
int7 .251 .036 .366 .000 .000 .000
int5 .302 .044 .440 .000 .000 .000
int3 .278 .040 .405 .000 .000 .000
int2 .312 .045 .455 .000 .000 .000
int1 .332 .048 .485 .000 .000 .000
grat9 .427 .228 .473 .248 .000 .000
grat6 .463 .247 .512 .269 .000 .000
grat4 .386 .206 .427 .224 .000 .000
grat3 .445 .238 .493 .259 .000 .000
Modification Indices (Group number 1 - Default model)
Covariances: (Group number 1 - Default model)
M.I. Par Change
e39 <--> e40 42.821 .285 e43 <--> CSE 42.500 .255 e43 <--> e40 23.599 -.210 e43 <--> e39 29.546 -.169 e37 <--> e43 8.027 .073 e33 <--> e39 4.113 .079 e33 <--> e37 4.221 -.066 e31 <--> e39 21.624 .206 e31 <--> e43 6.033 -.108 e27 <--> e40 4.661 .094 e27 <--> e30 6.983 .106 e26 <--> CSE 4.962 -.092 e26 <--> e43 4.486 -.069 e25 <--> e31 5.972 .105 e25 <--> e27 12.258 -.105 e25 <--> e26 8.995 .094 e24 <--> e43 4.304 -.065 e24 <--> e31 4.392 .093 e24 <--> e27 4.089 .063 e23 <--> e40 5.673 -.101 e23 <--> e41 6.473 -.086 e23 <--> e35 5.416 .043 e23 <--> e31 7.601 -.119 e23 <--> e24 4.192 -.062 e22 <--> e40 6.173 .129 e22 <--> e41 4.288 .086 e22 <--> e32 5.750 -.127 e21 <--> e26 9.866 -.116 e21 <--> e23 8.587 .100 e20 <--> e40 6.223 -.117
310
M.I. Par Change
e20 <--> e43 10.586 .110 e19 <--> e43 7.195 .072 e19 <--> e27 6.115 -.067 e19 <--> e26 5.847 -.069 e19 <--> e21 5.472 .071 e19 <--> e20 14.171 .110 e18 <--> e42 4.117 .041 e18 <--> e32 8.341 .111 e18 <--> e25 5.458 -.061 e17 <--> e31 10.343 -.111 e17 <--> e30 6.665 .081 e17 <--> e27 9.723 .076 e17 <--> e18 5.555 .048 e16 <--> e33 9.241 .113 e16 <--> e32 4.692 -.093 e15 <--> e27 4.831 -.054 e11 <--> CSE 8.686 .153 e11 <--> e40 22.743 .274 e11 <--> e39 8.016 .117 e11 <--> e43 11.037 -.136 e11 <--> e22 5.012 .110 e11 <--> e20 5.532 -.105 e11 <--> e16 8.176 .114 e10 <--> e36 8.299 -.056 e10 <--> e35 4.009 .041 e10 <--> e21 4.265 .078 e10 <--> e20 5.651 -.086 e10 <--> e18 8.183 .083 e10 <--> e11 5.419 .103 e9 <--> e40 7.201 -.133 e9 <--> e20 18.413 .166 e9 <--> e18 5.512 -.072 e9 <--> e17 4.887 -.062 e9 <--> e15 7.733 .077 e8 <--> e40 12.693 -.124 e8 <--> e39 13.041 -.091 e8 <--> e30 4.552 -.068 e8 <--> e24 8.820 -.075 e8 <--> e20 4.248 .056 e8 <--> e19 18.575 .094 e8 <--> e11 7.747 -.092 e7 <--> e27 6.907 .065 e7 <--> e18 5.691 -.051 e7 <--> e17 15.041 .074 e5 <--> CSE 4.074 .080 e5 <--> e43 4.551 .067 e5 <--> e23 6.517 .078 e5 <--> e17 4.308 .051 e5 <--> e10 9.610 .105 e3 <--> CSE 4.504 -.082 e3 <--> e41 4.572 .070
311
M.I. Par Change
e3 <--> e27 5.814 -.075 e3 <--> e26 13.318 .118 e3 <--> e23 7.154 -.080 e2 <--> CSE 4.436 .100 e2 <--> e41 7.474 -.114 e2 <--> e30 4.535 .103 e2 <--> e26 4.403 -.083 e2 <--> e24 8.600 .111 e2 <--> e5 5.156 -.085 e1 <--> e30 4.263 -.084 e1 <--> e11 4.574 -.090 e1 <--> e5 4.317 -.064
Variances: (Group number 1 - Default model)
M.I. Par Change
Regression Weights: (Group number 1 - Default model)
M.I. Par Change
PEU <--- PU 24.090 .411 PU <--- PEU 26.532 .255 INT <--- CSE 42.500 .391 use6 <--- INT 6.776 .144 use6 <--- int5 5.667 .088 use6 <--- int3 8.973 .113 use6 <--- int1 6.571 .115 use5 <--- int5 9.783 -.082 use4 <--- peu2 4.441 .048 use4 <--- pu6 4.072 .054 peu2 <--- use6 4.708 -.182 peu4 <--- cse9 7.080 .171 peu6 <--- PU 9.286 .260 peu6 <--- pu3 12.033 .179 peu6 <--- pu4 13.517 .203 peu6 <--- pu5 12.332 .184 peu6 <--- pu7 10.680 .170 peu6 <--- pu11 5.834 .149 pu2 <--- PEU 6.142 .123 pu2 <--- peu4 6.275 .086 pu2 <--- peu7 10.782 .123 pu2 <--- cse10 5.503 .109 pu3 <--- CSE 4.962 -.141 pu3 <--- pu11 4.198 -.094 pu3 <--- cse4 8.042 -.146 pu3 <--- cse7 9.438 -.163 pu3 <--- cse12 4.749 -.116 pu4 <--- cse9 7.347 -.118 pu4 <--- cse10 6.118 -.111 pu5 <--- int2 7.628 -.160 pu6 <--- use4 4.351 .165
312
M.I. Par Change
pu6 <--- peu6 5.756 -.078 pu6 <--- grat6 5.156 -.087 pu7 <--- PEU 4.718 .128 pu7 <--- peu2 4.072 .090 pu7 <--- peu6 6.079 .098 pu7 <--- peu7 6.483 .113 pu7 <--- grat4 4.044 .099 pu7 <--- grat3 4.204 .099 cse4 <--- INT 4.189 .148 cse4 <--- peu2 4.843 -.089 cse4 <--- peu7 5.707 -.096 cse4 <--- cse7 4.189 .111 cse4 <--- int3 17.719 .208 cse4 <--- int2 7.076 .166 cse4 <--- int1 5.528 .139 cse7 <--- pu2 5.167 -.086 cse7 <--- pu3 4.285 -.066 cse7 <--- cse4 7.120 .113 cse7 <--- int2 10.998 .166 cse9 <--- peu4 4.279 .061 cse10 <--- peu6 4.776 -.057 cse10 <--- int1 7.108 .114 cse11 <--- peu2 4.463 .076 cse11 <--- int7 8.691 .120 cse12 <--- int3 5.405 .083 int7 <--- CSE 8.686 .235 int7 <--- PEU 30.037 .358 int7 <--- PU 15.738 .316 int7 <--- GRAT 4.250 .134 int7 <--- peu2 26.066 .252 int7 <--- peu4 19.199 .198 int7 <--- peu6 20.575 .200 int7 <--- peu7 26.608 .254 int7 <--- pu2 13.549 .212 int7 <--- pu3 11.677 .165 int7 <--- pu4 7.723 .144 int7 <--- pu5 14.671 .188 int7 <--- pu6 15.328 .227 int7 <--- pu7 19.128 .213 int7 <--- pu11 10.352 .185 int7 <--- cse9 10.608 .194 int7 <--- cse10 4.805 .134 int7 <--- cse11 15.562 .259 int7 <--- grat4 5.997 .134 int5 <--- cse9 6.214 .120 int5 <--- grat9 8.529 .137 int3 <--- PEU 5.201 -.129 int3 <--- peu2 7.304 -.115 int3 <--- cse4 8.050 .159 int2 <--- PEU 5.346 -.092 int2 <--- PU 4.767 -.106
313
M.I. Par Change
int2 <--- peu6 4.558 -.057 int2 <--- peu7 8.327 -.086 int2 <--- pu2 4.625 -.075 int2 <--- pu3 4.545 -.063 int2 <--- pu5 9.901 -.093 int2 <--- pu7 7.090 -.079 int2 <--- cse4 4.055 .079 int2 <--- cse7 8.981 .121 int2 <--- int7 4.230 -.070 int1 <--- cse4 4.689 .083 int1 <--- cse10 8.092 .103 grat9 <--- CSE 4.074 .122 grat9 <--- INT 4.966 .150 grat9 <--- pu6 4.687 .096 grat9 <--- cse9 5.084 .102 grat9 <--- cse10 6.491 .119 grat9 <--- int5 11.731 .155 grat9 <--- int1 4.034 .110 grat6 <--- CSE 4.504 -.126 grat6 <--- pu2 6.360 -.108 grat6 <--- pu6 6.963 -.114 grat6 <--- cse10 6.273 -.114 grat6 <--- cse11 6.443 -.124 grat6 <--- int1 4.825 -.118 grat4 <--- CSE 4.436 .153 grat4 <--- peu7 4.434 .095 grat4 <--- pu2 5.468 .123 grat4 <--- pu5 8.413 .130 grat4 <--- pu7 5.013 .100 grat4 <--- cse4 4.725 .128 grat4 <--- cse10 4.472 .118 grat4 <--- cse11 5.089 .135 grat4 <--- int7 5.006 .114 grat3 <--- int7 5.418 -.100
“The Outputs of the Revised TAG Model for Chinese Participants”
Estimates (Group number 1 - Default model)
Scalar Estimates (Group number 1 - Default model)
Maximum Likelihood Estimates
Regression Weights: (Group number 1 - Default model)
Estimate S.E. C.R. P Label
PU <--- CSE .370 .077 4.802 *** par_24
PEU <--- CSE .432 .098 4.382 *** par_25
314
Estimate S.E. C.R. P Label
INT <--- PU .061 .128 .474 .636 par_26
INT <--- PEU .500 .147 3.412 *** par_28
GRAT <--- PU .587 .125 4.690 *** par_27
GRAT <--- PEU .327 .137 2.383 .017 par_29
USE <--- INT .402 .087 4.640 *** par_30
GRAT <--- INT .161 .085 1.893 .058 par_31
grat3 <--- GRAT 1.000
grat4 <--- GRAT 1.076 .091 11.851 *** par_1
grat6 <--- GRAT 1.143 .100 11.457 *** par_2
grat9 <--- GRAT 1.092 .093 11.792 *** par_3
int1 <--- INT 1.000
int2 <--- INT .827 .086 9.610 *** par_4
int3 <--- INT .923 .093 9.885 *** par_5
int5 <--- INT 1.007 .084 11.943 *** par_6
int7 <--- INT .953 .079 12.088 *** par_7
cse12 <--- CSE 1.000
cse11 <--- CSE 1.108 .064 17.221 *** par_8
cse10 <--- CSE 1.023 .062 16.392 *** par_9
cse9 <--- CSE 1.118 .065 17.203 *** par_10
cse7 <--- CSE 1.037 .064 16.194 *** par_11
cse4 <--- CSE .844 .070 12.088 *** par_12
pu11 <--- PU 1.000
pu7 <--- PU 1.425 .150 9.497 *** par_13
pu6 <--- PU 1.448 .147 9.857 *** par_14
pu5 <--- PU 1.401 .139 10.096 *** par_15
pu4 <--- PU 1.410 .144 9.792 *** par_16
pu3 <--- PU 1.436 .142 10.093 *** par_17
pu2 <--- PU 1.359 .143 9.487 *** par_18
peu7 <--- PEU 1.000
peu6 <--- PEU 1.144 .168 6.797 *** par_19
peu4 <--- PEU 1.049 .169 6.221 *** par_20
peu2 <--- PEU .822 .135 6.091 *** par_21
use4 <--- USE 1.000
use5 <--- USE 1.060 .138 7.674 *** par_22
use6 <--- USE .874 .116 7.534 *** par_23
Standardized Regression Weights: (Group number 1 - Default model)
Estimate
PU <--- CSE .388
PEU <--- CSE .437
INT <--- PU .054
INT <--- PEU .464
GRAT <--- PU .501
315
Estimate
GRAT <--- PEU .289
USE <--- INT .422
GRAT <--- INT .153
grat3 <--- GRAT .726
grat4 <--- GRAT .830
grat6 <--- GRAT .792
grat9 <--- GRAT .816
int1 <--- INT .808
int2 <--- INT .656
int3 <--- INT .673
int5 <--- INT .805
int7 <--- INT .812
cse12 <--- CSE .865
cse11 <--- CSE .890
cse10 <--- CSE .863
cse9 <--- CSE .893
cse7 <--- CSE .862
cse4 <--- CSE .724
pu11 <--- PU .624
pu7 <--- PU .825
pu6 <--- PU .869
pu5 <--- PU .900
pu4 <--- PU .865
pu3 <--- PU .900
pu2 <--- PU .820
peu7 <--- PEU .685
peu6 <--- PEU .707
peu4 <--- PEU .655
peu2 <--- PEU .533
use4 <--- USE .642
use5 <--- USE .846
use6 <--- USE .691
Variances: (Group number 1 - Default model)
Estimate S.E. C.R. P Label
CSE .693 .092 7.509 *** par_32
e39 .533 .112 4.757 *** par_33
e40 .546 .118 4.646 *** par_34
e43 .607 .102 5.936 *** par_35
e41 .458 .085 5.402 *** par_36
e42 .586 .134 4.366 *** par_37
e1 .776 .092 8.436 *** par_38
e2 .453 .065 6.990 *** par_39
316
Estimate S.E. C.R. P Label
e3 .669 .087 7.650 *** par_40
e5 .518 .071 7.281 *** par_41
e7 .417 .058 7.233 *** par_42
e8 .710 .080 8.823 *** par_43
e9 .808 .092 8.751 *** par_44
e10 .433 .059 7.343 *** par_45
e11 .368 .051 7.214 *** par_46
e15 .234 .029 8.137 *** par_47
e16 .222 .029 7.692 *** par_48
e17 .249 .030 8.179 *** par_49
e18 .221 .029 7.628 *** par_50
e19 .257 .031 8.225 *** par_51
e20 .448 .048 9.256 *** par_52
e21 .986 .103 9.549 *** par_53
e22 .599 .069 8.715 *** par_54
e23 .426 .052 8.265 *** par_55
e24 .291 .038 7.681 *** par_56
e25 .422 .051 8.314 *** par_57
e26 .304 .040 7.679 *** par_58
e27 .565 .064 8.801 *** par_59
e30 .764 .113 6.776 *** par_60
e31 .884 .140 6.307 *** par_61
e32 .992 .139 7.120 *** par_62
e33 1.148 .132 8.722 *** par_63
e35 1.016 .127 7.993 *** par_64
e36 .318 .087 3.643 *** par_65
e37 .597 .084 7.116 *** par_66
Squared Multiple Correlations: (Group number 1 - Default model)
Estimate
PEU .191
PU .151
INT .227
USE .178
GRAT .470
use6 .477
use5 .716
use4 .412
peu2 .285
peu4 .428
peu6 .500
peu7 .469
pu2 .672
317
Estimate
pu3 .810
pu4 .747
pu5 .809
pu6 .755
pu7 .680
pu11 .389
cse4 .524
cse7 .743
cse9 .797
cse10 .744
cse11 .793
cse12 .748
int7 .659
int5 .648
int3 .453
int2 .430
int1 .653
grat9 .665
grat6 .628
grat4 .688
grat3 .526
Total Effects (Group number 1 - Default model)
CSE PEU PU INT USE GRAT
PEU .432 .000 .000 .000 .000 .000
PU .370 .000 .000 .000 .000 .000
INT .238 .500 .061 .000 .000 .000
USE .096 .201 .024 .402 .000 .000
GRAT .397 .407 .597 .161 .000 .000
use6 .084 .176 .021 .351 .874 .000
use5 .102 .213 .026 .426 1.060 .000
use4 .096 .201 .024 .402 1.000 .000
peu2 .355 .822 .000 .000 .000 .000
peu4 .453 1.049 .000 .000 .000 .000
peu6 .494 1.144 .000 .000 .000 .000
peu7 .432 1.000 .000 .000 .000 .000
pu2 .503 .000 1.359 .000 .000 .000
pu3 .531 .000 1.436 .000 .000 .000
pu4 .522 .000 1.410 .000 .000 .000
pu5 .518 .000 1.401 .000 .000 .000
pu6 .535 .000 1.448 .000 .000 .000
pu7 .527 .000 1.425 .000 .000 .000
pu11 .370 .000 1.000 .000 .000 .000
318
CSE PEU PU INT USE GRAT
cse4 .844 .000 .000 .000 .000 .000
cse7 1.037 .000 .000 .000 .000 .000
cse9 1.118 .000 .000 .000 .000 .000
cse10 1.023 .000 .000 .000 .000 .000
cse11 1.108 .000 .000 .000 .000 .000
cse12 1.000 .000 .000 .000 .000 .000
int7 .227 .477 .058 .953 .000 .000
int5 .240 .504 .061 1.007 .000 .000
int3 .220 .462 .056 .923 .000 .000
int2 .197 .414 .050 .827 .000 .000
int1 .238 .500 .061 1.000 .000 .000
grat9 .433 .445 .652 .175 .000 1.092
grat6 .453 .466 .683 .184 .000 1.143
grat4 .427 .438 .642 .173 .000 1.076
grat3 .397 .407 .597 .161 .000 1.000
Standardized Total Effects (Group number 1 - Default model)
CSE PEU PU INT USE GRAT
PEU .437 .000 .000 .000 .000 .000
PU .388 .000 .000 .000 .000 .000
INT .224 .464 .054 .000 .000 .000
USE .094 .196 .023 .422 .000 .000
GRAT .355 .360 .509 .153 .000 .000
use6 .065 .135 .016 .291 .691 .000
use5 .080 .166 .019 .357 .846 .000
use4 .061 .126 .015 .271 .642 .000
peu2 .233 .533 .000 .000 .000 .000
peu4 .286 .655 .000 .000 .000 .000
peu6 .309 .707 .000 .000 .000 .000
peu7 .299 .685 .000 .000 .000 .000
pu2 .319 .000 .820 .000 .000 .000
pu3 .350 .000 .900 .000 .000 .000
pu4 .336 .000 .865 .000 .000 .000
pu5 .349 .000 .900 .000 .000 .000
pu6 .338 .000 .869 .000 .000 .000
pu7 .320 .000 .825 .000 .000 .000
pu11 .242 .000 .624 .000 .000 .000
cse4 .724 .000 .000 .000 .000 .000
cse7 .862 .000 .000 .000 .000 .000
cse9 .893 .000 .000 .000 .000 .000
cse10 .863 .000 .000 .000 .000 .000
cse11 .890 .000 .000 .000 .000 .000
cse12 .865 .000 .000 .000 .000 .000
319
CSE PEU PU INT USE GRAT
int7 .182 .377 .044 .812 .000 .000
int5 .180 .373 .044 .805 .000 .000
int3 .151 .312 .037 .673 .000 .000
int2 .147 .304 .036 .656 .000 .000
int1 .181 .375 .044 .808 .000 .000
grat9 .290 .294 .415 .125 .000 .816
grat6 .281 .285 .404 .121 .000 .792
grat4 .295 .299 .423 .127 .000 .830
grat3 .258 .261 .370 .111 .000 .726
Direct Effects (Group number 1 - Default model)
CSE PEU PU INT USE GRAT
PEU .432 .000 .000 .000 .000 .000
PU .370 .000 .000 .000 .000 .000
INT .000 .500 .061 .000 .000 .000
USE .000 .000 .000 .402 .000 .000
GRAT .000 .327 .587 .161 .000 .000
use6 .000 .000 .000 .000 .874 .000
use5 .000 .000 .000 .000 1.060 .000
use4 .000 .000 .000 .000 1.000 .000
peu2 .000 .822 .000 .000 .000 .000
peu4 .000 1.049 .000 .000 .000 .000
peu6 .000 1.144 .000 .000 .000 .000
peu7 .000 1.000 .000 .000 .000 .000
pu2 .000 .000 1.359 .000 .000 .000
pu3 .000 .000 1.436 .000 .000 .000
pu4 .000 .000 1.410 .000 .000 .000
pu5 .000 .000 1.401 .000 .000 .000
pu6 .000 .000 1.448 .000 .000 .000
pu7 .000 .000 1.425 .000 .000 .000
pu11 .000 .000 1.000 .000 .000 .000
cse4 .844 .000 .000 .000 .000 .000
cse7 1.037 .000 .000 .000 .000 .000
cse9 1.118 .000 .000 .000 .000 .000
cse10 1.023 .000 .000 .000 .000 .000
cse11 1.108 .000 .000 .000 .000 .000
cse12 1.000 .000 .000 .000 .000 .000
int7 .000 .000 .000 .953 .000 .000
int5 .000 .000 .000 1.007 .000 .000
int3 .000 .000 .000 .923 .000 .000
int2 .000 .000 .000 .827 .000 .000
int1 .000 .000 .000 1.000 .000 .000
grat9 .000 .000 .000 .000 .000 1.092
320
CSE PEU PU INT USE GRAT
grat6 .000 .000 .000 .000 .000 1.143
grat4 .000 .000 .000 .000 .000 1.076
grat3 .000 .000 .000 .000 .000 1.000
Standardized Direct Effects (Group number 1 - Default model)
CSE PEU PU INT USE GRAT
PEU .437 .000 .000 .000 .000 .000
PU .388 .000 .000 .000 .000 .000
INT .000 .464 .054 .000 .000 .000
USE .000 .000 .000 .422 .000 .000
GRAT .000 .289 .501 .153 .000 .000
use6 .000 .000 .000 .000 .691 .000
use5 .000 .000 .000 .000 .846 .000
use4 .000 .000 .000 .000 .642 .000
peu2 .000 .533 .000 .000 .000 .000
peu4 .000 .655 .000 .000 .000 .000
peu6 .000 .707 .000 .000 .000 .000
peu7 .000 .685 .000 .000 .000 .000
pu2 .000 .000 .820 .000 .000 .000
pu3 .000 .000 .900 .000 .000 .000
pu4 .000 .000 .865 .000 .000 .000
pu5 .000 .000 .900 .000 .000 .000
pu6 .000 .000 .869 .000 .000 .000
pu7 .000 .000 .825 .000 .000 .000
pu11 .000 .000 .624 .000 .000 .000
cse4 .724 .000 .000 .000 .000 .000
cse7 .862 .000 .000 .000 .000 .000
cse9 .893 .000 .000 .000 .000 .000
cse10 .863 .000 .000 .000 .000 .000
cse11 .890 .000 .000 .000 .000 .000
cse12 .865 .000 .000 .000 .000 .000
int7 .000 .000 .000 .812 .000 .000
int5 .000 .000 .000 .805 .000 .000
int3 .000 .000 .000 .673 .000 .000
int2 .000 .000 .000 .656 .000 .000
int1 .000 .000 .000 .808 .000 .000
grat9 .000 .000 .000 .000 .000 .816
grat6 .000 .000 .000 .000 .000 .792
grat4 .000 .000 .000 .000 .000 .830
grat3 .000 .000 .000 .000 .000 .726
321
Indirect Effects (Group number 1 - Default model)
CSE PEU PU INT USE GRAT
PEU .000 .000 .000 .000 .000 .000
PU .000 .000 .000 .000 .000 .000
INT .238 .000 .000 .000 .000 .000
USE .096 .201 .024 .000 .000 .000
GRAT .397 .080 .010 .000 .000 .000
use6 .084 .176 .021 .351 .000 .000
use5 .102 .213 .026 .426 .000 .000
use4 .096 .201 .024 .402 .000 .000
peu2 .355 .000 .000 .000 .000 .000
peu4 .453 .000 .000 .000 .000 .000
peu6 .494 .000 .000 .000 .000 .000
peu7 .432 .000 .000 .000 .000 .000
pu2 .503 .000 .000 .000 .000 .000
pu3 .531 .000 .000 .000 .000 .000
pu4 .522 .000 .000 .000 .000 .000
pu5 .518 .000 .000 .000 .000 .000
pu6 .535 .000 .000 .000 .000 .000
pu7 .527 .000 .000 .000 .000 .000
pu11 .370 .000 .000 .000 .000 .000
cse4 .000 .000 .000 .000 .000 .000
cse7 .000 .000 .000 .000 .000 .000
cse9 .000 .000 .000 .000 .000 .000
cse10 .000 .000 .000 .000 .000 .000
cse11 .000 .000 .000 .000 .000 .000
cse12 .000 .000 .000 .000 .000 .000
int7 .227 .477 .058 .000 .000 .000
int5 .240 .504 .061 .000 .000 .000
int3 .220 .462 .056 .000 .000 .000
int2 .197 .414 .050 .000 .000 .000
int1 .238 .500 .061 .000 .000 .000
grat9 .433 .445 .652 .175 .000 .000
grat6 .453 .466 .683 .184 .000 .000
grat4 .427 .438 .642 .173 .000 .000
grat3 .397 .407 .597 .161 .000 .000
Standardized Indirect Effects (Group number 1 - Default model)
CSE PEU PU INT USE GRAT
PEU .000 .000 .000 .000 .000 .000
322
CSE PEU PU INT USE GRAT
PU .000 .000 .000 .000 .000 .000
INT .224 .000 .000 .000 .000 .000
USE .094 .196 .023 .000 .000 .000
GRAT .355 .071 .008 .000 .000 .000
use6 .065 .135 .016 .291 .000 .000
use5 .080 .166 .019 .357 .000 .000
use4 .061 .126 .015 .271 .000 .000
peu2 .233 .000 .000 .000 .000 .000
peu4 .286 .000 .000 .000 .000 .000
peu6 .309 .000 .000 .000 .000 .000
peu7 .299 .000 .000 .000 .000 .000
pu2 .319 .000 .000 .000 .000 .000
pu3 .350 .000 .000 .000 .000 .000
pu4 .336 .000 .000 .000 .000 .000
pu5 .349 .000 .000 .000 .000 .000
pu6 .338 .000 .000 .000 .000 .000
pu7 .320 .000 .000 .000 .000 .000
pu11 .242 .000 .000 .000 .000 .000
cse4 .000 .000 .000 .000 .000 .000
cse7 .000 .000 .000 .000 .000 .000
cse9 .000 .000 .000 .000 .000 .000
cse10 .000 .000 .000 .000 .000 .000
cse11 .000 .000 .000 .000 .000 .000
cse12 .000 .000 .000 .000 .000 .000
int7 .182 .377 .044 .000 .000 .000
int5 .180 .373 .044 .000 .000 .000
int3 .151 .312 .037 .000 .000 .000
int2 .147 .304 .036 .000 .000 .000
int1 .181 .375 .044 .000 .000 .000
grat9 .290 .294 .415 .125 .000 .000
grat6 .281 .285 .404 .121 .000 .000
grat4 .295 .299 .423 .127 .000 .000
grat3 .258 .261 .370 .111 .000 .000
Covariances: (Group number 1 - Default model)
M.I. Par Change
e39 <--> e40 87.398 .438
e43 <--> CSE 23.824 .259
e43 <--> e40 7.193 -.143
e43 <--> e39 4.288 -.097
e42 <--> CSE 17.967 .228
e36 <--> CSE 4.874 .105
323
M.I. Par Change
e33 <--> CSE 4.434 -.144
e33 <--> e39 18.386 .260
e32 <--> e37 6.248 -.165
e32 <--> e36 5.010 .135
e32 <--> e35 4.203 -.173
e31 <--> CSE 4.530 -.138
e31 <--> e39 22.699 .274
e31 <--> e43 10.895 -.216
e31 <--> e32 11.958 .279
e30 <--> e32 9.747 -.231
e27 <--> e43 12.648 .175
e27 <--> e41 4.660 -.096
e27 <--> e37 5.595 .114
e27 <--> e31 5.866 -.146
e27 <--> e30 14.043 .207
e26 <--> e41 6.021 .085
e26 <--> e31 5.159 .107
e26 <--> e30 15.658 -.170
e25 <--> CSE 4.496 .091
e25 <--> e30 7.753 .136
e24 <--> e41 6.783 -.088
e24 <--> e31 6.674 .119
e24 <--> e25 4.637 .064
e23 <--> CSE 4.069 -.087
e23 <--> e40 5.088 .099
e23 <--> e33 4.056 .114
e22 <--> CSE 7.167 -.134
e22 <--> e40 4.470 .107
e22 <--> e43 4.008 -.102
e22 <--> e41 16.658 .186
e22 <--> e27 6.926 -.121
e22 <--> e24 4.900 -.077
e22 <--> e23 8.713 .121
e21 <--> CSE 24.814 .308
e21 <--> e40 4.173 -.128
e21 <--> e39 4.458 -.115
e21 <--> e42 12.977 .229
e21 <--> e36 4.152 .114
e19 <--> e39 4.032 -.060
e19 <--> e26 9.436 -.075
e19 <--> e25 4.996 .062
e17 <--> e32 9.228 .128
e17 <--> e23 6.749 -.071
324
M.I. Par Change
e16 <--> e39 10.311 -.092
e16 <--> e43 6.052 .081
e16 <--> e32 4.345 -.086
e16 <--> e31 4.782 -.088
e15 <--> e24 6.011 .056
e15 <--> e17 6.415 .051
e11 <--> e42 4.894 -.096
e11 <--> e25 4.498 -.073
e11 <--> e24 4.964 .066
e10 <--> e11 5.098 .080
e9 <--> e37 4.740 .125
e9 <--> e27 6.821 .140
e8 <--> e40 4.116 -.111
e8 <--> e31 12.413 -.236
e8 <--> e30 4.117 .125
e8 <--> e25 5.843 .107
e8 <--> e9 4.497 .126
e5 <--> e31 4.197 -.128
e5 <--> e22 4.669 .103
e3 <--> e37 4.441 -.117
e3 <--> e36 4.361 .106
e3 <--> e32 7.340 .192
e3 <--> e27 6.966 -.136
e3 <--> e22 13.982 .199
e2 <--> e27 8.047 .125
e2 <--> e22 7.868 -.128
325
TAG Model (University of Malaya in Malaysia_Unconstrain) Configural Invariance analysis
326
TAG Model (Jiaxing University in China_Unconstrain) Configural Invariance analysis
327
TAG Model (UM_constrain) Metric Invariance analysis
328
TAG Model (JU_constrain) Metric Invariance analysis
329
Appendix J The Survey Questionnaire (English version)
General Guidelines
This questionnaire attempts to investigate “Development and Validation of the
Technology Adoption and Gratification (TAG) Model to Assess Lecturers’ ICT
Usage: An Empirical Study”.
Direction
The questionnaire has seven sections. For all multiple choice questions please indicate
your response by placing a tick (/) in the appropriate box. A seven-point Likert scale
questions, please circle the number of your choice. If you wish to comment on any
question or qualify your answer, please feel free to use the space in the margin or you
may write your comments on separate sheet of paper. All information that you provide
will be kept strictly confidential. Thanks for your gracious cooperation
Section I: Demographic Information
1. Gender: Male
Female
2. Age: 25-29 years old
30-34 years old
35-39 years old
40-44 years old
45 ears above
3. Number of years working at the university
1. 1-5 years
2. 6-10 years
3. 11-15 years
4. 16-20 years
5. 21 years above
4. Highest level of educational background
a. Undergraduate
b. Master
c. PhD
5. Designation:
a. Professor
b. Associate Professor
c. Assistant Professor/Senior Lecturer
d. Lecturer
6. Faculty/School______________________________
330
7. Nationality_________________________________
Section II: Use of ICT facilities
For questions 1-16, rate the likelihood of each using the following scale:
1= Never (N) 4= Weekly (W)
2= Once a semester (OS) 5= Sometimes (SO)
3= Monthly (M) 6= Daily (D)
7 = Always (A)
No.
How often do you use the following ICT
facilities for research and academic-related
activities?
N OS M W SO D A
1 The University online research
databases/Online library research
databases. 1 2 3 4 5 6 7
2 Laptop or Desktop computers. 1 2 3 4 5 6 7
3 Wireless Internet. 1 2 3 4 5 6 7
4 Web browser (www). 1 2 3 4 5 6 7
5 Search engine (e.g. Google, Yahoo, etc.). 1 2 3 4 5 6 7
6 Microsoft Office applications. 1 2 3 4 5 6 7
7 Research related software to analyze the
data. 1 2 3 4 5 6 7
8 Document processing (e.g. PDF). 1 2 3 4 5 6 7
9 Computer graphics. 1 2 3 4 5 6 7
10 Email. 1 2 3 4 5 6 7
11 Document editing/composing. 1 2 3 4 5 6 7
12 Computer labs. 1 2 3 4 5 6 7
13 Multimedia facilities (e.g. CD-ROM, VCD,
DVD etc.). 1 2 3 4 5 6 7
14 I use ICT for preparing the course and
lecture notes. 1 2 3 4 5 6 7
15 I use ICT facilities for making
presentations in the course. 1 2 3 4 5 6 7
16 I use ICT facilities for carrying out studies
in laboratories or workshops, and making
experiments.
1 2 3 4 5 6 7
331
Section III: Perceived Ease of Use of ICT Facilities
For questions 1-15, rate how much you agree with each statement using the following
scale:
1= Strongly Disagree (SD) 4= Neither agree nor disagree (N)
2= Moderately Disagree (MD) 5= Slightly Agree (SLA)
3= Slightly Disagree (SLD) 6= Moderately Agree (MA)
7= Strongly Agree (SA)
No.
Items SD MD SLD N SLA MA SA
1 I find the university ICT facilities
easy to use. 1 2 3 4 5 6 7
2 I find it easy to access the
university research databases. 1 2 3 4 5 6 7
3 It is easy for me to become skilful
at using university databases for
conducting research. 1 2 3 4 5 6 7
4 Interacting with the research
databases system requires minimal
effort on my part. 1 2 3 4 5 6 7
5 I find it easy to get the research
databases to help facilitate my
research. 1 2 3 4 5 6 7
6 Interacting with the university
research databases system is very
stimulating for me.
1 2 3 4 5 6 7
7 I find it easy to select
articles/journals of different
categories using research databases
(Education/Engineering/Business,
etc.).
1 2 3 4 5 6 7
8 With Wireless Internet, I find it
easy to access university databases
to do research. 1 2 3 4 5 6 7
9 Wireless Internet allows me to
access research and learning
materials from Web Browser
(WWW).
1 2 3 4 5 6 7
10 Wireless Internet is easy to use for
teaching and research. 1 2 3 4 5 6 7
11 I find it easy to use computers
provided by the university. 1 2 3 4 5 6 7
12 My interaction with university ICT
services available is clear and
understandable.
1 2 3 4 5 6 7
332
13 I find it easy to download teaching,
learning and research materials
using Wireless Internet in terms of
speed.
1 2 3 4 5 6 7
14 It is easy for me to use multimedia
facilities at the university. 1 2 3 4 5 6 7
15 I find it easy to use Microsoft office
for teaching and research purposes. 1 2 3 4 5 6 7
Section IV: Perceived Usefulness of ICT Facilities
For questions 1-16, rate how much you agree with each statement using the following
scale:
1= Strongly Disagree (SD) 4= Neither agree nor disagree (N)
2= Moderately Disagree (MD) 5= Slightly Agree (SLA)
3= Slightly Disagree (SLD) 6= Moderately Agree (MA)
7= Strongly Agree (SA)
No.
Items SD MD SLD N SLA MA SA
1 Using the ICT facilities at the
university enables me to
accomplish tasks more quickly. 1 2 3 4 5 6 7
2 Using the ICT services available at
the university increases my
research productivity. 1 2 3 4 5 6 7
3 The current university ICT system
makes work more interesting. 1 2 3 4 5 6 7
4 ICT facilities at the university
improve the quality of the work I
do. 1 2 3 4 5 6 7
5 Using the university ICT facilities
enhances my research skills. 1 2 3 4 5 6 7
6 Using the university ICT facilities
would make it easier for me to find
information.
1 2 3 4 5 6 7
7 ICT services at the university make
information always available to
users. 1 2 3 4 5 6 7
8 Using the ICT facilities helps me
write my journal articles. 1 2 3 4 5 6 7
9 My job would be difficult to
perform without the ICT facilities. 1 2 3 4 5 6 7
333
10 Using the ICT facilities saves my
time. 1 2 3 4 5 6 7
11 Using the university ICT facilities
provides me with the latest
information on specific areas of
research.
1 2 3 4 5 6 7
12 I can use ICT facilities from
anywhere, anytime at the campus. 1 2 3 4 5 6 7
13 I believe that the use of ICT in the
classroom enhances student
learning in my discipline. 1 2 3 4 5 6 7
14 I believe that email and other forms
of electronic communication are
important tools in faculty/student
communication.
1 2 3 4 5 6 7
15 I believe that web-based
instructional materials enhance
student learning. 1 2 3 4 5 6 7
16 Using the ICT services enables me
to download teaching, learning and
research materials from the internet. 1 2 3 4 5 6 7
Section V: Computer Self-Efficacy of ICT facilities
For questions 1-15, rate how much you agree with each statement using the following
scale:
1= Strongly Disagree (SD) 4= Neither agree nor disagree (N)
2= Moderately Disagree (MD) 5= Slightly Agree (SLA)
3= Slightly Disagree (SLD) 6= Moderately Agree (MA)
7= Strongly Agree (SA)
No.
Items SD MD SLD N SLA MA SA
1 I have the skills required to use
computer applications for writing
my research papers.
1 2 3 4 5 6 7
2 I have the skills and knowledge
required to use computer
applications for demonstrating
specific concepts in class.
1 2 3 4 5 6 7
3 I have the skills required to use
computer applications for
presenting lectures. 1 2 3 4 5 6 7
334
4 I have the skills required to
communicate electronically with
my colleagues and students. 1 2 3 4 5 6 7
5 I have the ability to e-mail, chat,
download teaching, learning and
research materials, and search
different websites.
1 2 3 4 5 6 7
6 I have the ability to use the
Wireless Internet service provided
by the university.
1 2 3 4 5 6 7
7 I have the skills required to use ICT
facilities to enhance the
effectiveness of my teaching,
learning and research.
1 2 3 4 5 6 7
8 I feel capable of using the
university research databases for
writing journal papers. 1 2 3 4 5 6 7
9 I have the ability to navigate my
way through the ICT facilities. 1 2 3 4 5 6 7
10 I have the skills required to use the
ICT facilities to enhance the quality
of my research works. 1 2 3 4 5 6 7
11 I have the ability to save and print
journals/articles from the research
databases.
1 2 3 4 5 6 7
12 I have the knowledge and skills
required to benefit from using the
university ICT facilities. 1 2 3 4 5 6 7
13 I am capable of accessing the
research databases from the
university website. 1 2 3 4 5 6 7
14 I am capable to download and
install research related software
using ICT facilities at the
university.
1 2 3 4 5 6 7
15 I am capable of using multimedia
facilities for delivering lecturers. 1 2 3 4 5 6 7
335
Section VI: Intention to Use of ICT Facilities
1= Very Unlikely (VU) 4= Neither like or unlike (N)
2= Moderately Unlikely (MU) 5= Slightly Likely (SL)
3= Slightly Unlikely (SU) 6= Moderately Likely (ML)
7= Very likely (VL)
No.
Items VU MU SU N SL ML VL
1 I will use the ICT facilities to carry
out my research activities. 1 2 3 4 5 6 7
2 I will use computer applications to
present lecture materials in class. 1 2 3 4 5 6 7
3 I will use Microsoft office
applications to write my research
papers. 1 2 3 4 5 6 7
4 I will teach a course that will be
delivered totally on the world wide
web. 1 2 3 4 5 6 7
5 I intend to use ICT facilities for
teaching, learning and research
purposes frequently. 1 2 3 4 5 6 7
6 I intend to download research
materials from the university
databases frequently.
1 2 3 4 5 6 7
7 I will use the university research
databases to update myself on the
research areas I am pursuing. 1 2 3 4 5 6 7
8 I intend to use the ICT facilities to
upgrade my research performance. 1 2 3 4 5 6 7
9 I will use the university research
databases to write outstanding
journal articles. 1 2 3 4 5 6 7
10 I intend to use research related
software to analyze the data. 1 2 3 4 5 6 7
336
Section VII: Gratification of ICT Facilities Usage
For questions 1-10, rate how much you agree with each statement using the following
scale:
1= Very Unsatisfied (VU) 4= Not Sure (NS)
2= Moderately Unsatisfied (MU) 5= Slightly Satisfied (SLS)
3= Slightly Unsatisfied (SLU) 6= Moderately Satisfied (MS)
7= Very Satisfied (VS)
No.
Items VU MU SLU NS SLS MS VS
1 Overall I am satisfied with the
ease of completing my task using
ICT facilities. 1 2 3 4 5 6 7
2 I am satisfied with the ICT
facilities provided by the
university. 1 2 3 4 5 6 7
3 I am satisfied with the ease of use
of the ICT facilities. 1 2 3 4 5 6 7
4 The university ICT facilities have
greatly affected the way I search
for information and conduct my
research.
1 2 3 4 5 6 7
5 Providing the ICT is an
indispensable services provided
by the university. 1 2 3 4 5 6 7
6 Overall I am satisfied with the
amount of time it takes to
complete my task.
1 2 3 4 5 6 7
7 I am satisfied with the structure of
accessible information (available
as categories of research domain,
or by date of issue—of journals in
particular, or as full-texts or
abstracts of theses and
dissertations) of the university
research databases.
1 2 3 4 5 6 7
8 I am satisfied with the support
information provided by the
university’s ICT facilities. 1 2 3 4 5 6 7
9 I am satisfied in using ICT
facilities for teaching, learning
and research. 1 2 3 4 5 6 7
10 Overall I am satisfied with the
Wireless Internet service provided
at the university. 1 2 3 4 5 6 7
337
(Chinese Version)
使用指南使用指南使用指南使用指南
该问卷旨在调查““““科技应用的接受性和满意性理论模型科技应用的接受性和满意性理论模型科技应用的接受性和满意性理论模型科技应用的接受性和满意性理论模型((((TAG))))在评估教师使用信息和通信技在评估教师使用信息和通信技在评估教师使用信息和通信技在评估教师使用信息和通信技
术这一领域的发展及有效性术这一领域的发展及有效性术这一领域的发展及有效性术这一领域的发展及有效性::::实证性研究实证性研究实证性研究实证性研究””””。。。。
说明说明说明说明
该问卷由6个小部分组成。衡量方式采用李克特七点尺度量表,请在每一道题后面的选项中用
(√)标出您的选择。如果您对任何问题有不同的看法或有不同的答案,您可以写在对应题号左右
的空白处,或写在另外纸上的空白处。您所填的个人信息将会被严格保密。谢谢您的合作。
第一部分第一部分第一部分第一部分::::个人信息个人信息个人信息个人信息
1.性别: 男 女
2.年龄: 25 – 29 岁
30 – 34 岁
35 – 39 岁
40 – 44 岁
45岁及以上
3.大学的工作经历:
a. 1 – 5 年
b. 6 – 10 年
c. 11 – 15 年
d. 16 – 20 年
e. 21年及以上
第二部分第二部分第二部分第二部分::::信息和通信设备的信息和通信设备的信息和通信设备的信息和通信设备的使用频率使用频率使用频率使用频率
问题1 – 16,用以下几种标准来评定你对每一项的认同程度, 请用(√√√√)标出您的选择。
1=从不 2=每学期 3=每月 4=每周 5=有时候 6=每天 7=经常
序序序序
号号号号
您使用下列信息与通信设施来做与研究和学术相关您使用下列信息与通信设施来做与研究和学术相关您使用下列信息与通信设施来做与研究和学术相关您使用下列信息与通信设施来做与研究和学术相关
的活动的频率的活动的频率的活动的频率的活动的频率????
从
不
每
学
期
每
月
每
周
有
时
候
每
天
经
常
1 学校的研究性数据库。 1 2 3 4 5 6 7
2 笔记本电脑或台式电脑。 1 2 3 4 5 6 7
3 无线网络。 1 2 3 4 5 6 7
4 网络浏览器(www万维网)。 1 2 3 4 5 6 7
5 搜索引擎(比如:谷歌,雅虎,百度等)。 1 2 3 4 5 6 7
6 微软办公应用(Microsoft Office:Word, Excel, PPT
等)。 1 2 3 4 5 6 7
7 与研究相关的用来分析数据的软件。 1 2 3 4 5 6 7
8 文件处理应用(比如PDF转换器) 1 2 3 4 5 6 7
9 计算机绘图应用 1 2 3 4 5 6 7
10 邮件。 1 2 3 4 5 6 7
11 文档编辑/排版应用。 1 2 3 4 5 6 7
4.本人已获得的最高学历:
a. 本科
b. 硕士
c. 博士
5.职称:
a. 教授
b. 副教授
c. 助理教授
d. 高级讲师
e. 讲师
6.所在院系:_________________________________
7.国 籍: _________________________________
338
12 计算机实验室。 1 2 3 4 5 6 7
13 多媒体设施(光驱,光盘,视频播放器等)。 1 2 3 4 5 6 7
14 我使用信息与通信设施来备课和准备讲义。 1 2 3 4 5 6 7
15 教学时,我使用信息和通信设施演示课程和讲义。 1 2 3 4 5 6 7
16 在实验室或研讨会上,我使用信息与通信设施来开
展教学研究。 1 2 3 4 5 6 7
第三部分第三部分第三部分第三部分::::使用信息和通信设备的使用信息和通信设备的使用信息和通信设备的使用信息和通信设备的易用性感知易用性感知易用性感知易用性感知
问题1 – 15,用以下几种标准来评定你对每一项的认同程度,请用(√√√√)标出您的选择。
1=强烈不赞同 2=中度不赞同 3=略不赞同 4=既不赞同也不反对
5=稍赞同 6=中度赞同 7=强烈赞同
序序序序
号号号号 陈述项目陈述项目陈述项目陈述项目
强
烈
不
赞
同
中
度
不
赞
同
略
不
赞
同
既
不
赞
同
也
不
反
对
稍
赞
同
中
度
赞
同
强
烈
赞
同
1 我觉得学校的信息和通信设备容易使用。 1 2 3 4 5 6 7
2 我觉得访问学校的研究性数据库是容易的。 1 2 3 4 5 6 7
3 对我来说,我能容易的就变得熟练地使用学校数据库来做研
究。 1 2 3 4 5 6 7
4 与研究性数据库系统互动需要我很少的努力。 1 2 3 4 5 6 7
5 我觉得使用数据库来帮助促进我的研究是容易的。 1 2 3 4 5 6 7
6 对我来说,与学校研究性数据库进行互动很令人兴奋。 1 2 3 4 5 6 7
7 我觉得使用研究性数据库来选择不同类别的文章/期刊是容
易的(比如:教育类/工程类/商务类/语言类等)。 1 2 3 4 5 6 7
8 我觉得使用无线网络访问学校数据库做研究是容易的。 1 2 3 4 5 6 7
9 无线网络允许我通过网页浏览器(WWW万维网)访问研究
和学习资源。 1 2 3 4 5 6 7
10 使用无线网络来进行教学和研究是容易的。 1 2 3 4 5 6 7
11 我觉得学校提供的计算机使用起来是容易的。 1 2 3 4 5 6 7
12 对我来说,与学校可获得的信息和通信服务的互动过程是明
确的、易懂的。 1 2 3 4 5 6 7
13 在网络方面,我觉得使用无线网络下载教学、学习和研究资
料是容易的。 1 2 3 4 5 6 7
14 对我来说,在学校使用多媒体设施是容易的或方便的。 1 2 3 4 5 6 7
15 我觉得微软办公应用软件(Microsoft Office)在以教学和研
究为目的的应用中是容易的。 1 2 3 4 5 6 7
339
第四部分第四部分第四部分第四部分::::信息和通信设备的实用性感知信息和通信设备的实用性感知信息和通信设备的实用性感知信息和通信设备的实用性感知
问题1 – 16,用以下几种标准来评定你对每一项的认同程度, 请用(√√√√)标出您的选择。
1=强烈不赞同 2=中度不赞同 3=略不赞同 4=既不赞同也不反对
5=稍赞同 6=中度赞同 7=强烈赞同
序序序序
号号号号 陈述项目陈述项目陈述项目陈述项目
强
烈
不
赞
同
中
度
不
赞
同
略
不
赞
同
既
不
赞
同
也
不
反
对
稍
赞
同
中
度
赞
同
强
烈
赞
同
1 使用学校的信息和通信设施能够使我更快地完成任务。 1 2 3 4 5 6 7
2 使用学校可获得的信息和通信服务提高我的研究生产力。 1 2 3 4 5 6 7
3 当前学校的信息与通信系统使工作变得更加有趣。 1 2 3 4 5 6 7
4 学校的信息与通信设施提高我的工作质量。 1 2 3 4 5 6 7
5 使用学校的信息与通信设施增强我的研究技能。 1 2 3 4 5 6 7
6 对我来说,使用学校的信息与通信设施寻找信息会更容
易。 1 2 3 4 5 6 7
7 学校的信息与通信服务让用户始终可获得信息。 1 2 3 4 5 6 7
8 使用信息与通信设施有助于我写期刊文章。 1 2 3 4 5 6 7
9 没有信息与通信设施,我的工作会难以执行。 1 2 3 4 5 6 7
10 使用信息与通信设施节省我的时间。 1 2 3 4 5 6 7
11 使用学校的信息与通信设施给我提供特定研究领域的最新
信息。 1 2 3 4 5 6 7
12 我可以在校园里的任意时间和任何地方使用信息与通信设
施。 1 2 3 4 5 6 7
13 在我的课上,我认为信息与通信设施的使用提高学生的学
习能力。 1 2 3 4 5 6 7
14 我认为电子邮件和其他形式的电子通讯是老师/学生交流沟
通的重要工具。 1 2 3 4 5 6 7
15 我认为基于网络的教学材料提高学生的学习能力。 1 2 3 4 5 6 7
16 使用信息与通信服务允许我从英特网上下载教学,学习和
研究资料。 1 2 3 4 5 6 7
340
第五部分第五部分第五部分第五部分::::自我效能自我效能自我效能自我效能
问题1 – 15,用以下几种标准来评定你对每一项的认同程度, 请用(√√√√)标出您的选择。
1=强烈不赞同 2=中度不赞同 3=略不赞同 4=既不赞同也不反对
5=稍赞同 6=中度赞同 7=强烈赞同
序序序序
号号号号 陈述项目陈述项目陈述项目陈述项目
强烈不
赞同
中
度
不
赞
同
略不
赞同
既不赞同
也不反对
稍
赞
同
中度
赞同
强烈
赞同
1 我拥有所需的技能使用相关计算机应用来
写我的研究论文。 1 2 3 4 5 6 7
2 我拥有所需的技能和知识使用相关计算机
应用来在课堂上展现出特定具体的概念。 1 2 3 4 5 6 7
3 我拥有所需的技能使用相关计算机应用来
授课。 1 2 3 4 5 6 7
4 我拥有所需的技能来与我的同事和学生进
行电子通讯交流。 1 2 3 4 5 6 7
5 我有能力发邮件,聊天,下载教学、学习
和研究资料,以及搜索不同的网站。 1 2 3 4 5 6 7
6 我有能力使用学校提供的无线网络服务。 1 2 3 4 5 6 7
7 我拥有所需的技能使用信息与通信设备来
提高我的教学、学习与研究效益。 1 2 3 4 5 6 7
8 我相信我有能力使用学校的研究性数据库
来写期刊论文。 1 2 3 4 5 6 7
9 我有能力通过我的方式来使用信息与通信
设施。 1 2 3 4 5 6 7
10 我拥有所需的技能使用信息与通信设施来
提高我研究工作的质量。 1 2 3 4 5 6 7
11 我有能力从研究性数据库中保存和打印期
刊/文章。 1 2 3 4 5 6 7
12 我拥有所需的知识和技能从学校信息与通
信设施的使用中获益。 1 2 3 4 5 6 7
13 我能够从学校网站上访问研究性数据库资
源。 1 2 3 4 5 6 7
14 我能够使用学校的信息与通信设施下载和
安装与研究相关的软件。 1 2 3 4 5 6 7
15 我能够使用多媒体设施来授课。 1 2 3 4 5 6 7
第六部分第六部分第六部分第六部分::::使用信息和通信设备的使用信息和通信设备的使用信息和通信设备的使用信息和通信设备的意向意向意向意向
问题1 – 10,用以下几种标准来评定你对每一项的认同程度, 请用(√√√√)标出您的选择。
1=非常不可能 2=中度不可能 3=略不可能 4=不确定 5=稍可能 6=中度可能 7=非常可能
序序序序
号号号号 陈述项目陈述项目陈述项目陈述项目
非
常
不
可
能
很不
可能
略不
可能
不
确
定
稍
可
能
很
可
能
非常
可能
1 我将使用信息与通信设施来开展我的研究活
动。 1 2 3 4 5 6 7
2 在课堂上,我将使用相关的计算机应用来展示 1 2 3 4 5 6 7
341
课件/讲义材料。
3 我将使用微软办公应用软件(Microsoft Office)
来写我的研究论文。 1 2 3 4 5 6 7
4 我将完全在万维网(WWW)上教授一门课程。
(远程教育) 1 2 3 4 5 6 7
5 我打算经常使用信息与通信设施来教学,学习
和做研究。 1 2 3 4 5 6 7
6 我打算经常从学校的数据库中下载研究资料。 1 2 3 4 5 6 7
7 我将使用学校的研究性数据库在自己所从事的
的研究领域更新自己。 1 2 3 4 5 6 7
8 我打算使用信息与通信设施来提升我的研究
(绩效,成果)表现。 1 2 3 4 5 6 7
9 我将使用学校的研究性数据库来写出色的期刊
文章。 1 2 3 4 5 6 7
10 我打算使用与研究相关的软件来分析数据。 1 2 3 4 5 6 7
第七部分第七部分第七部分第七部分::::对信息和通信设备的对信息和通信设备的对信息和通信设备的对信息和通信设备的满意度满意度满意度满意度
问题1 – 10,用以下几种标准来评定你对每一项的认同程度, 请用(√√√√)标出您的选择。
1=非常不满意 2=中度不满意 3=略不满意 4=不确定 5=稍满意 6=中度满意 7=非常满意
序号序号序号序号 陈述项目陈述项目陈述项目陈述项目
非常
不满
意
中度
不满
意
略
不
满
意
不
确
定
稍
满
意
中
度
满
意
非
常
满
意
1 总体上,我对使用信息与通信设施完成任务的简易
性表示满意。 1 2 3 4 5 6 7
2 我满意于学校提供的信息与通信设施。 1 2 3 4 5 6 7
3 我满意于使用信息与通信设施的简易性。 1 2 3 4 5 6 7
4 学校的信息与通信设施已经很大程度上影响了我搜
索信息和开展研究的方式。 1 2 3 4 5 6 7
5 信息与通信服务是学校提供的必不可少的服务。 1 2 3 4 5 6 7
6 总体上,我满意于完成任务所花的时间量。 1 2 3 4 5 6 7
7
我满意于学校数据库可访问信息的布局构造(以研
究领域分类,或以特定期刊、论文全文或摘要的发
表日期呈现)。
1 2 3 4 5 6 7
8 我满意于学校的信息与通信设施所提供的支持信
息。 1 2 3 4 5 6 7
9 我满意于使用信息与通信设施来教学,学习和做研
究。 1 2 3 4 5 6 7
10 总体上,我满意于学校所提供的无线网络服务。 1 2 3 4 5 6 7