development and validation of a measurement instrument for studying supply chain management...
TRANSCRIPT
www.elsevier.com/locate/dsw
Journal of Operations Management 23 (2005) 618–641
Development and validation of a measurement instrument for
studying supply chain management practices
Suhong Li a,*, S. Subba Rao b, T.S. Ragu-Nathan b, Bhanu Ragu-Nathan b
a Computer Information Systems Department, Bryant University, 1150 Douglas Pike, Smithfield, RI 02917-1284, USAb College of Business Administration, The University of Toledo, Toledo, OH 43606, USA
Received 1 January 2003; received in revised form 1 December 2004; accepted 28 January 2005
Available onli
ne 13 March 2005Abstract
It is widely argued that competition is no longer between organizations, but among supply chains. Effective supply chain
management (SCM) has become a potentially valuable way of securing competitive advantage and improving organizational
performance. This research conceptualizes, develops, and validates six dimensions of SCM practices (strategic supplier
partnership, customer relationship, information sharing, information quality, internal lean practices, and postponement). Data
for the study were collected from 196 organizations and the measurement scales were tested and validated using structural
equation modeling. It is hoped that this study will provide a parsimonious measurement instrument to assess the performance of
the overall supply chain.
# 2005 Elsevier B.V. All rights reserved.
Keywords: Supply chain management; Measurement; Structural equation modeling
1 Council of Logistics Management (CLM) (2000) defines SCM
1. Introduction
As competition in the 1990s intensified and
markets became global, so did the challenges associa-
ted with getting a product and service to the right place
at the right time at the lowest cost. Organizations
began to realize that it is not enough to improve
efficiencies within an organization, but their whole
supply chain has to be made competitive. It has been
pointed out that understanding and practicing supply
* Corresponding author. Tel.: +1 401 232 6503;
fax: +1 401 232 6435.
E-mail address: [email protected] (S. Li).
0272-6963/$ – see front matter # 2005 Elsevier B.V. All rights reserved
doi:10.1016/j.jom.2005.01.002
chain management (SCM) has become an essential
prerequisite to staying in the competitive global race
and to growing profitably (Power et al., 2001; Moberg
et al., 2002).
SCM has been defined1 to explicitly recognize the
strategic nature of coordination between trading
partners and to explain the dual purpose of SCM: to
improve the performance of an individual organiza-
s the systemic, strategic coordination of the traditional business
unctions and tactics across these businesses functions within a
articular organization and across businesses within the supply
hain for the purposes of improving the long-term performance
f the individual organizations and the supply chain as a whole.
a
f
p
c
o
.
S. Li et al. / Journal of Operations Management 23 (2005) 618–641 619
tion, and to improve the performance of the entire
supply chain. The goal of SCM is to create sourcing,
making and delivery processes and logistics functions
seamlessly across the supply chain as an effective
competitive weapon.
The concept of SCM has received increasing
attention from academicians, consultants, and busi-
ness managers alike (Croom et al., 2000; Tan et al.,
1998; Van Hoek, 1998). Many organizations have
begun to recognize that SCM is the key to building a
sustainable competitive edge for their products or
services in an increasingly crowded marketplace
(Jones, 1998). Despite the increased attention paid
to SCM and the expectations from SCM, the literature
does not offer much evidence of successful imple-
mentations. For example, Boddy et al. (1998) found
that more than half of the respondents to their survey
considered that their organizations had not been
successful in implementing supply chain partnering;
Spekman et al. (1998) noted that 60% of supply chain
alliances tended to fail. A recent Deloitte Consulting
survey reported that only 2% of North American
manufacturers ranked their supply chains as world-
class although 91% of these same manufacturers
ranked their SCM as important or critical to their
organization’s success (Thomas, 1999). Thus, while it
is clear that SCM is important to organizations,
effective management of the supply chain does not
appear to have been realized.
While the lack of successful SCM efforts has been
attributed to the complexity of SCM itself, research in
the area of SCM has not been able to offer much
by way of guidance to help the practice of SCM. This
has been attributed primarily to conceptual confusion
and the lack of a theoretical framework in researching
SCM. It has been pointed out that the SCM
phenomenon has not been well understood in the
literature.
Many empirical studies reflect the lack of a
theoretical framework for anchoring the results of
their studies. For example, while some studies still
tend to consider SCM as being the same as integrated
logistics management, and hence focus on inventory
reduction both within and across organizations in the
supply chain (Alvarado and Kotzab, 2001; Bechtel and
Jayaram, 1997; Romano and Vinelli, 2001; Van Hoek,
1998), there are other studies that consider SCM as
just the extension of the traditional purchasing and
supplier management activities (Banfield, 1999;
Lamming, 1996). Also, much of the current empirical
research focuses on only the internal supply chain, the
upstream or downstream side of the supply chain.
Some researches have focused on certain aspect of the
internal supply chain, such as total quality manage-
ment practices (Tan et al., 2002), internal integration
(Pagell, 2004; Braganza, 2002), agile/lean manufac-
turing (Womack and Jones, 1996; Naylor et al., 1999;
McIvor, 2001), and postponement (Beamon, 1998;
Naylor et al., 1999; Van Hoek, 1998; Van Hoek et al.,
1999). Topics such as supplier selection, supplier
involvement, and manufacturing performance (Choi
and Hartley, 1996; Vonderembse and Tracey, 1999),
the influence of supplier alliances on the organization
(Stuart, 1997), success factors in strategic supplier
alliances (Monczka et al., 1998; Narasimhan and
Jayaram, 1998; Stuart, 1997), and supplier manage-
ment orientation and supplier–buyer performance
(Shin et al., 2000), have been researched on the
supplier side. Studies such as those by Clark and Lee
(2000) and Alvarado and Kotzab (2001), focus on the
downstream linkages between manufacturers and
retailers. A few recent studies have begun to consider
both the upstream and downstream sides of the supply
chain simultaneously. Tan et al. (1998) explore the
relationships between supplier management practices,
customer relations practices and organizational per-
formance; Frohlich and Westbrook (2001) investigate
the effects of supplier–customer integration on
performance. Tan et al. (2002) study SCM and
supplier evaluation practices, Min and Mentzer (2004)
develop an instrument to measure the supply chain
orientation and SCM at conceptual levels. Cigolini
et al. (2004) develop a set of supply chain techniques
and tools for examining SCM strategies. Taken
together, these studies are representative of efforts
to address various diverse but interesting aspects of
SCM practices. However, the absence of a unifying
conceptual framework, which covers upstream, inter-
nal and downstream side of a supply chain, detracts
from the usefulness of the implications of their results.
The lack of a comprehensive view of SCM
practices and the consequent lack of a reliable
operational measure of the concept have constrained
the earlier studies from offering broad-based and
generalizable implications for guiding both the
practice of SCM and further research on the topic.
S. Li et al. / Journal of Operations Management 23 (2005) 618–641620
The purpose of this research is to develop and validate
a parsimonious measurement instrument for SCM
practices. SCM practices are defined as the set of
activities undertaken by an organization to promote
effective management of its supply chain. SCM
practice is proposed to be a multi-dimensional
concept, and hence viewed as a more comprehensive
concept than the narrower view (the supplier side, the
internal side or the customer side) taken in most prior
research. Operational measures for the constructs are
then developed and tested empirically, using data
collected from respondents to a survey. It is expected
that the current research, by offering a validated
instrument to measure SCM practices, will offer
useful guidance for SCM practices measurement and
provide a springboard for further research in the area.
The remainder of this paper is organized as follows.
The next section presents the research framework,
provides the definitions and theory underlying each
dimension of SCM practices, and two of the
performance outcomes of SCM practices (delivery
performance and time to market). The research
methodology, empirical validation, and refinement
of the scales are to be found in the sections that follow.
The last section presents the discussion of results and
directions for future work.
2. Constructs and framework
SCM practices have been defined as the set of
activities undertaken in an organization to promote
effective management of its supply chain. Donlon
(1996) describes the latest evolution of SCM practices,
which includes supplier partnership, outsourcing,
cycle time compression, and continuous process flow,
and information technology sharing. Tan et al. (1998)
use purchasing, quality, and customer relations to
represent SCM practices, in their empirical study.
Alvarado and Kotzab (2001) include in their list of
SCM practices concentration on core competencies,
use of inter-organizational systems such as EDI, and
elimination of excess inventory levels by postponing
customization toward the end of the supply chain.
Tan (2001) suggests that a well-integrated supply
chain involves coordinating the flow of materials
and information among suppliers, manufacturers, and
customers, and implementing product postponement
and mass customization. Tan et al. (2002) identify six
aspects of SCM practices through factor analysis:
supply chain integration, information sharing, supply
chain characteristics, customer service management,
geographical proximity and JIT capability. Chen and
Paulraj (2004) use supplier base reduction, long-term
relationship, communication, cross-functional teams
and supplier involvement to measure buyer–supplier
relationships. Min and Mentzer (2004) identify the
concept SCM as including agreed vision and goals,
information sharing, risk and award sharing, coopera-
tion, process integration, long-term relationship and
agreed supply chain leadership. Thus, the literature
portrays SCM practices from a variety of different
perspectives with a common goal of ultimately
improving organizational performance. In reviewing
and consolidating the literature, six distinctive
dimensions of SCM practices emerge, including
strategic supplier partnership, customer relationship,
information sharing, information quality, internal lean
practices and postponement. The six constructs cover
upstream (strategic supplier partnership) and down-
stream (customer relationship) sides of a supply chain,
information flow across a supply chain (information
sharing and information quality), and internal supply
chain processes (internal lean practices and postpone-
ment) (see Fig. 1). It should be pointed out that even
though the above dimensions capture the major
aspects of SCM practices, they cannot be considered
complete. Other factors, such as total quality manage-
ment practices (Tan et al., 2002), internal integration
(Pagell, 2004; Braganza, 2002), geographical proxi-
mity, cross-functional teams, agreed vision and goals,
and agreed supply chain leadership (Min and Mentzer,
2004) are also identified in the literature. Though these
factors are of great interest, they are not included due
to the length of the survey and the concerns regarding
the parsimony of measurement instruments.
In the following paragraphs, we identify and define
the six constructs of SCM practices, as well as the two
performance outcomes of SCM practices, delivery
dependability and time to market (see Table 1).
Detailed descriptions of the constructs are presented in
the following paragraphs.
Strategic supplier partnership is defined as the
long-term relationship between the organization and
its suppliers. It is designed to leverage the strategic and
operational capabilities of individual participating
S. Li et al. / Journal of Operations Management 23 (2005) 618–641 621
Fig. 1. Theoretical framework linking SCM practices constructs and performance (domain definitions).
organizations to help them achieve significant ongoing
benefits (Balsmeier and Voisin, 1996; Monczka et al.,
1998; Noble, 1997; Stuart, 1997). A strategic partner-
ship emphasizes direct, long-term association and
encourages mutual planning and problem solving
efforts (Gunasekaran et al., 2001). Such strategic
partnerships are entered into to promote shared
benefits among the parties and ongoing participation
in one or more key strategic areas such as technology,
products, markets, etc. (Yoshino and Rangan, 1995).
Strategic partnerships with suppliers enable organiza-
tions to work more effectively with a few important
suppliers who are willing to share responsibility for
the success of the products. Suppliers participating
early in the product-design process can offer more
cost-effective design choices, help select the best
components and technologies, and help in design
assessment (Monczka et al., 1993). Strategically-
aligned organizations can work closely together and
eliminate wasteful time and effort (Balsmeier and
Voisin, 1996). An effective supplier partnership can be
a critical component of a leading edge supply chain
(Noble, 1997).
Customer relationship comprises the entire array of
practices that are employed for the purpose of managing
customer complaints, building long-term relationships
with customers, and improving customer satisfaction
(Aggarwal, 1997; Claycomb et al., 1999; Tan et al.,
1998). Noble (1997) and Tan et al. (1998) consider
customer relationship management as an important
component of SCM practices. The growth of mass
customization and personalized service is leading to an
era in which relationship management with customers
is becoming crucial for corporate survival (Wines,
1996). Close customer relationship allows an organiza-
tion to differentiate its product from competitors,
sustain customer loyalty, and dramatically extend the
value it provides to its customers (Magretta, 1998).
Information sharing refers to the extent to which
critical and proprietary information is communicated
S. Li et al. / Journal of Operations Management 23 (2005) 618–641622
Table 1
Constructs for SCM practices, delivery dependability and time to market
Constructs Definitions Literature
Strategic supplier
partnership
Strategic Supplier Partnership is defined as the long-term relationship
between the organization and its suppliers. It is designed to
leverage the strategic and operational capabilities of individual
participating organizations to help them achieve significant
ongoing benefits
Balsmeier and Voisin (1996),
Gunasekaran et al. (2001),
Lamming (1996),
Monczka et al. (1998),
Stuart (1997)
Customer relationship The entire array of practices that are employed for the
purpose of managing customer complaints,
building long-term relationships with customers,
and improving customer satisfaction
Aggarwal (1997),
Claycomb et al. (1999),
Magretta (1998), Noble (1997),
Tan et al. (1998), Wines (1996)
Information sharing The extent to which critical and proprietary information
is communicated to one’s supply chain partner
Balsmeier and Voisin (1996),
Jones (1998), Lalonde (1998),
Mentzer et al. (2000),
Monczka et al. (1998),
Novack et al. (1995),
Stein and Sweat (1998),
Towill (1997), Yu et al. (2001)
Information quality Refers to the accuracy, timeliness, adequacy, and credibility
of information exchanged
Alvarez (1994), Berry et al. (1994),
Chizzo (1998), Holmberg (2000),
Jarrell (1998), Lee et al. (1997),
Mason-Jones and Towill (1997),
McAdam and McCormack (2001),
Metters (1997), Monczka et al. (1998)
Internal lean practices The practices of eliminating waste (cost, time, etc.) in a
manufacturing system, characterized by reduced set-up
times, small lot sizes, and pull-production
Handfield and Nichols (1999),
Mason-Jones and Towill (1997),
McIvor (2001), Taylor (1999),
Womack and Jones (1996)
Postponement The practice of moving forward one or more operations
or activities (making, sourcing and delivering) to a much
later point in the supply chain
Beamon (1998),
Lee and Billington (1995),
Naylor et al. (1999),
Van Hoek (1998),
Van Hoek et al. (1999),
Waller et al. (2000)
Delivery dependability The extent to which an organization is capable of providing
on time the type and volume of product required by customer(s)
Hall (1993), Koufteros et al. (1997),
Rondeau et al. (2000)
Time to market The extent to which an organization is capable of introducing
new products faster than major competitors
Stalk (1988), Vesey (1991),
Handfield and Pannesi (1995),
Kessler and Chakrabarti (1996)
to one’s supply chain partner (Monczka et al., 1998).
Shared information can vary from strategic to tactical
in nature and from information about logistics activities
to general market and customer information (Mentzer
et al., 2000). Many researchers have suggested that the
key to the seamless supply chain is making available
undistorted and up-to-date marketing data at every node
within the supply chain (Balsmeier and Voisin, 1996;
Towill, 1997). By taking the data available and sharing
it with other parties within the supply chain, informa-
tion can be used as a source of competitive advantage
(Jones, 1998; Novack et al., 1995).
Many researchers have emphasized the importance
of information sharing in SCM practices. Lalonde
(1998) considers sharing of information as one of five
building blocks that characterize a solid supply chain
relationship. According to Stein and Sweat (1998),
supply chain partners who exchange information
regularly are able to work as a single entity. Together,
they can understand the needs of the end customer
better and hence can respond to market change
quicker. Moreover, Yu et al. (2001) point out that the
negative impact of the bullwhip effect on a supply
chain can be reduced or eliminated by sharing
S. Li et al. / Journal of Operations Management 23 (2005) 618–641 623
information with trading partners. Tompkins and Ang
(1999) consider the effective use of relevant and
timely information by all functional elements within
the supply chain as a key competitive and distinguish-
ing factor. As an example, sharing of information with
suppliers has given Dell Company the benefits of
faster cycle times, reduced inventory, and improved
forecasts. Customers, for their part, have benefited by
getting a higher-quality product at a lower price (Stein
and Sweat, 1998).
Information quality includes such aspects as the
accuracy, timeliness, adequacy, and credibility of
information exchanged (Monczka et al., 1998). While
information sharing is important, the significance of
its impact on SCM depends on what information is
shared, when and how it is shared, and with whom
(Chizzo, 1998; Holmberg, 2000). Jarrell (1998) notes
that sharing information within the entire supply chain
can create flexibility, but this requires accurate and
timely information.
In the literature there are many examples of the
dysfunctional effects of inaccurate/delayed informa-
tion, as information moves along the supply chain
(Lee et al., 1997; Mason-Jones and Towill, 1997;
McAdam and McCormack, 2001; Metters, 1997). It
has been suggested that organizations will deliberately
distort information that can potentially reach not only
their competitors, but also their own suppliers and
customers (Mason-Jones and Towill, 1997, 1999). It
appears that there is a built-in reluctance within
organizations to give away more than minimal
information (Berry et al., 1994) since information
disclosure is perceived as a loss of power. Given these
predispositions, ensuring the quality of the shared
information becomes a critical aspect of effective
SCM. Organizations need to view their information as
a strategic asset and ensure that it flows with minimum
delay and distortion. Alvarez (1994) notes that
information shared must be as accurate as possible
in order to obtain the best SCM solution.
Internal lean practices are the practices of
eliminating waste (cost, time, etc.) in a manufacturing
system, characterized by reduced set-up times, small
lot sizes, and pull-production (Womack and Jones,
1996; McIvor, 2001; Taylor, 1999). The term ‘‘lean’’
is used to refer to a system that uses less input to
produce at a mass production speed, while offering
more variety to the end customers. Elimination of
waste is a fundamental idea within the lean system. In
‘‘Lean Thinking’’ written by Womack and Jones
(1996), five principles are identified as fundamental to
the elimination of waste. (1) Specify what does and
does not create value from the customer’s perspective;
(2) identify all the steps necessary to design, order and
produce the product across the whole value stream to
highlight non-value-adding waste; (3) make those
actions that create value flow without interruption,
detours, backflows, waiting or scrap; (4) only make
what is pulled by the customers just-in-time; (5) strive
for perfection by continually removing successive
layers of waste as they are uncovered. Following these
principles, internal lean practices may include set-up
reduction, pull production, short lead times from
suppliers, streamlining ordering, receiving and other
paperwork and continuous quality improvement.
Lean thinking and lean practices have become very
important aspects of effective SCM (Handfield and
Nichols, 1999; Mason-Jones and Towill, 1997).
Organizations that have not made the effort to drive
out unnecessary cost, time and other wastes from their
internal supply chain (so that they can deliver high
quality, best value products in a timely manner) will
run the risk of losing customers. Lean operating
practices are the dominant drivers of a highly
integrated and down-sized supply chain, promising
both cost savings and productive working partner
relationships.
Postponement is defined as the practice of moving
forward one or more operations or activities (making,
sourcing and delivering) to a much later point in the
supply chain (Beamon, 1998; Naylor et al., 1999; Van
Hoek, 1998; Van Hoek et al., 1999). In general, there
are three types of postponement: form, time, and place
postponement. ‘‘Form postponement entails delaying
activities that determine the form and function of
products in the chain until customer orders have been
received. Time postponement means delaying the
forward movement of goods until customer orders
have been received. Place postponement refers to the
positioning of inventories upstream in centralized
manufacturing or distribution operations, to postpone
the forward or downward movement of goods’’ (Van
Hoek et al., 1999). Two primary considerations in
developing a postponement strategy are: (1) determin-
ing how many steps to postpone and (2) determining
which steps to postpone (Beamon, 1998).
S. Li et al. / Journal of Operations Management 23 (2005) 618–641624
Postponement allows an organization to be flexible
in developing different versions of the product in order
to meet changing customer needs, and to differentiate
a product or to modify a demand function (Waller
et al., 2000). Keeping materials undifferentiated for
as long as possible will increase an organization’s
flexibility in responding to changes in customer
demand. In addition, an organization can reduce
supply chain cost by keeping undifferentiated inven-
tories (Lee and Billington, 1995; Van Hoek et al.,
1999).
Postponement needs to match the type of products,
market demands of a company, and structure or
constraints within the manufacturing and logistics
system (Fisher et al., 1994; Fisher, 1997; Fuller et al.,
1993; Pagh and Cooper, 1998). In general, the
adoption of postponement may be appropriate in the
following conditions: innovative products (Fisher et al.,
1994; Fisher, 1997); products with high monetary
density, high specialization and wide range; markets
characterized by long delivery time, low delivery
frequency and high demand uncertainty; manufacturing
or logistics systems with small economies of scales and
no need for special knowledge (Pagh and Cooper,
1998).
Performance outcomes: In this study, the constructs
of delivery dependability and time to market have been
included primarily to evaluate the predictive validity of
the six SCM practices constructs. Delivery depend-
ability is the ability of an organization to provide
products on time and of the type and in the volume as
required by the customers (Hall, 1993; Koufteros et al.,
1997; Rondeau et al., 2000). Time to market is the
capability of an organization to introduce new products
faster than the competitors (Stalk, 1988; Vesey, 1991;
Handfield and Pannesi, 1995; Kessler and Chakrabarti,
1996). Delivery dependability and time to market are
impacted by the SCM practices like strategic supplier
partnership, information sharing, postponement, etc.
For example, strategic supplier partnership can reduce
time to market (Ragatz et al., 1997) and increase level of
customer responsiveness and satisfaction (Power et al.,
2001). Information sharing will enable organizations to
make dependable delivery and introduce products to the
market quickly (Jarrell, 1998). Postponement not only
increased the flexibility in the supply chain, but also
balances global efficiency and customer responsiveness
(Van Hoek et al., 1999).
3. Instrument development and validation
An effective instrument should cover the content
domain of each construct (Nunnally, 1978; Churchill,
1979). The items that measure a construct should
agree (converge) with each other, and the items of one
construct should disagree (discriminate) with mea-
sures of the other constructs. Each construct should be
reliable and short and easy to use. Scale development
and refinement is a two-phase approach. In the first
phase, the definitions of the constructs as well as the
measurement items for each construct are established.
In this phase, we also provide tentative indications of
reliability and validity. This phase included item
generation, pre-pilot study, and pilot study. In the
second phase, we further refine this scale and validate
the measures using large-scale survey data collected
based on the scales developed in the first phase.
3.1. Scale development
The very basic requirement for a good measure is
content validity, which means that the measurement
items in an instrument should cover the major content
of a construct (Churchill, 1979). Content validity is
usually achieved through a comprehensive literature
review and interviews with practitioners and acade-
micians. The items for SCM practices were generated
based on previous SCM literature (Aggarwal, 1997;
Claycomb et al., 1999; Forker et al., 1999; Lee and
Kim, 1999; Monczka et al., 1998; Shin et al., 2000;
Stuart, 1997; Tan et al., 1998; Vonderembse and
Tracey, 1999; Walton, 1996). All the items were
measured on a five-point scale.
In the pre-pilot study, these items were reviewed by
six academicians and re-evaluated through structured
interviews with three practitioners who were asked to
comment on the appropriateness of the research
constructs. Based on the feedback from the academi-
cians and practitioners, redundant and ambiguous
items were either modified or eliminated. New items
were added wherever deemed necessary.
The next stage, pilot-study, in the development of
scales was the application of the Q-sort procedure for
assessing initial construct validity and reliability. The
Q-sort method, a manual factor sorting technique
(Moore and Benbasat, 1991), is an iterative process in
which the degree of agreement between judges forms
S. Li et al. / Journal of Operations Management 23 (2005) 618–641 625
the basis of assessing construct validity and improving
the reliability of the constructs. For the Q-sort method
two evaluation indices are used to measure inter-judge
agreement level: Cohen’s Kappa (Cohen, 1960) and
Moore and Benbasat’s ‘‘Hit Ratio’’ (Moore and
Benbasat, 1991). Cohen’s Kappa is a measure of
agreement that can be interpreted as the proportion of
joint judgment in which there is agreement after
chance agreement is excluded. For Kappa, no general
agreement exists with respect to required scores.
However, several studies have considered scores
greater than 0.65 to be acceptable (e.g., Vessey,
1984; Jarvenpaa, 1989). Landis and Koch (1977) have
provided a more detailed guideline to interpret Kappa
by associating different values of this index to the
degree of agreement beyond chance.
Moore and Benbasat’s (1991) method requires
analysis of how many items are placed by the panel of
judges for each round within the target construct. The
higher the percentage of items placed in the target
construct, the higher the degree of inter-judge
agreement across the panel, which must have
occurred. There are no established guidelines for
determining good levels of placement, but the matrix
of the item placement ratio can be used to highlight
any potential problem areas. Item placement ratios
were calculated by counting all the items that were
correctly sorted into the target category by each of the
judges and dividing them by twice the total number of
items.
For the Q-sort method, purchasing/production
managers were requested to act as judges and sort
the items into the six dimensions of SCM practices,
based on similarities and differences among items. To
assess the reliability of the sorting conducted by the
judges, Cohen’s Kappa, the inter-judge raw agreement
scores and item placement ratios were used.
In the first round, the inter-judge raw agreement
scores averaged 0.89 and the initial overall placement
ratio of items within the target constructs was 0.87,
which are in acceptable range. Cohen’s Kappa score
averaged 0.87. Following the guidelines of Landis and
Table 2
Inter-judge agreements
Agreement measure Round 1 Round 2 Round 3
Raw agreement (%) 89 95 95
Cohen’s Kappa (%) 87 94 94
Koch (1977) for interpreting the Kappa coefficient, the
value of 0.87 was considered an excellent level of
agreement (beyond chance) for the judges in the first
round. In order to further improve the agreement
scores and Cohen’s Kappa measure of agreement, an
examination of the off-diagonal entries in the
placement matrix was conducted. Items classified in
a construct different from their target construct were
identified and dropped or reworded. Also, feedback
from both judges was obtained on each item and
incorporated into the modification of the items.
The reworded items were then entered into a
second sorting round. In the second round, the inter-
judge raw agreement scores averaged 0.95, the initial
overall placement ratio of items within the target
constructs was 0.97, and the Cohen’s Kappa score
averaged 0.94. Since the second round achieved an
excellent overall placement ratio of items within the
target constructs (0.97), it was decided to keep all the
items for the third sorting round.
The third sorting round was used to re-validate the
constructs. The third round achieved the same inter-
judge raw agreement and Cohen’s Kappa scores as the
second round, thereby indicating an excellent level of
agreement between the judges in the third round and
consistency of results between the second and third
rounds. At this stage the statistics suggested an
excellent level of inter-judge agreement indicating a
high level of reliability and construct validity. Table 2
presents a summary of agreement scores for the three
rounds. In Table 3 we present the final round of item
placement ratios. Each of the SCM practices scale is
listed on the rows of the tables. For strategic supplier
partnership, perfect item placement ratio for this scale
would be a score of 20 (10 items � 2 judges). In this
case, 18 judge-items were classified as intended, while
2 items under N/A category. The item-placement ratio
for strategic supplier partnership thus equals 18/20 or
90%. On the other hand, information sharing,
information quality, internal lean practices, and
postponement obtained a 100% item placement ratio.
All of the scales are well above the recommended
value of 0.65 (Moore and Benbasat, 1991).
3.2. Empirical scale refinement and validation
This study sought to choose respondents who can
be expected to have the best knowledge about the
S. Li et al. / Journal of Operations Management 23 (2005) 618–641626
Table 3
Item-placement ratios (final sorting round) for SCM practices
Intended SCM
practices scales
(no. of items in scale)
Actual classifications NA Total Item
placement
ratio (%)Strategic
supplier
partnership
Customer
relationship
Information
sharing
Information
quality
Internal
lean
practices
Postponement
Strategic supplier
partnership (10)
18 2 20 90
Customer relationship (9) 17 1 18 94
Information sharing (7) 14 14 100
Information quality (5) 10 10 100
Internal lean practices (8) 16 16 100
Postponement (5) 10 10 100
Total item placements = 88, Hits = 85, Overall hit ratio (%) = 97.
operation and management of the supply chain in his/
her organization. Mailing lists were obtained from two
sources: the Society of Manufacturing Engineers
(SME) and the attendees at the Council of Logistics
Management (CLM) conference in New Orleans,
2000. The lists were limited to organizations with
more than 100 employees since organizations with
less than 100 employees are unlikely to engage in
any sophisticated SCM. Six SIC codes were covered
in the study: 25 ‘‘Furniture and Fixtures’’, 30
‘‘Rubber and Plastics’’, 34 ‘‘Fabricated Metal Pro-
ducts’’, 35 ‘‘Industrial and Commercial Machinery’’,
36 ‘‘Electronic and Other Electric Equipment’’, 37
‘‘Transportation Equipment’’.
The final version of the questionnaire was
administrated to 3137 target respondents. To ensure
a reasonable response rate, the survey was sent in three
waves. The questionnaires with a cover letter
indicating the purpose and significance of the study
were mailed to the target respondents. In the cover
letter, a web-address of the online version of the
survey was also provided in case the respondents
wished to fill it in electronically. There were 196
complete and usable responses, representing a
response rate of approximately 6.3%.
A significant problem with organizational-level
research is that senior and executive-level mangers
receive many requests to participate and have very
limited time. Because this interdisciplinary research
collects information from several functional areas, the
size and scope of the research instruments are large
and time consuming to complete. This further
contributes to the low response rate. While the
response rate was less than desired, the makeup of
respondent pool was considered excellent (see
Appendix A). Among the respondents, almost 20%
of the respondents are CEO/President/Vice President/
Director. About half of the respondents are managers,
some identified them as supply chain manager, plant
manager, logistics manager or IT manager in the
questionnaire. The areas of expertise were 30%
purchasing, 47% manufacturing production, and
30% distribution/transportation/sales. It can be seen
that respondents have covered all the functions across
a supply chain from purchasing, to manufacturing, to
distribution and transportation, and to sales. Moreover,
about 30% of the respondents are responsible for more
than one job function and 60% of respondents have
stayed at the organization for more than 10 years, and
as such they should have a broad view of SCM
practices in their organization.
A concern that is typical of such surveys is that
information collected from respondents might have a
non-response bias. This research did not investigate
non-response bias directly because the mailing list had
only name and addresses of the individuals and not any
organizational details. Hence a comparison was made
between those subjects who responded after the initial
mailing and those who responded to the second/third
wave. The later wave of surveys received was
considered to be representative of non-respondents
(Amstrong and Overton, 1977; Lambert and Harring-
ton, 1990). Similar methodology has been used in
prior empirical studies of SCM (Handfield and
Bechtel, 2002; Moberg et al., 2002; Narasimhan
and Kim, 2001). Using the x2 statistic and P < 0.05, it
S. Li et al. / Journal of Operations Management 23 (2005) 618–641 627
was found that there were no significant differences
between the two groups in terms of employment size,
sales volume, and respondent’s job title.
4. Assessment of construct validities
The measurement properties of the six dimensions
of SCM practices construct were evaluated by
assessing key components of construct validity. As
per the guidelines of Bagozzi (1980), and Bagozzi and
Phillips (1982), the following measurement properties
are considered important for assessing the measures
developed in this paper: (1) content validity, (2)
internal consistency of operationalization (unidimen-
sionality and reliability), (3) convergent validity, (4)
discriminant validity, and (5) predictive validity.
An instrument has content validity if there is a
general agreement among the subjects and researchers
that the instrument has measurement items that cover
all important aspects of the variable being measured.
Unidimensionality indicates that all of the items are
measuring a single theoretical construct. Reliability
values indicate the degree to which operational
measures are free from random error and measure
the construct in a consistent manner. Convergent
validity is about the extent to which there is
consistency in measurements across multiple oper-
ationalizations (Campbell and Fiske, 1959). Discri-
minant validity refers to the independence of the
dimensions (Bagozzi and Phillips, 1991), i.e., the
extent to which measures of the six constructs are
distinctly different from each other. Predictive validity
seeks to find support for the validity of the construct by
investigating whether it exhibits relationships with
other constructs that are in accordance with theory.
LISREL was used to check the measurement
properties of the constructs. Model-data fit was
evaluated based on multiple fit indexes. The chi-
square statistic is perhaps the most popular index to
evaluate the goodness of fit of the model. It measures
the difference between the sample covariance and the
fitted covariance. However, this index has some
disadvantages. The chi-square index is sensitive to
sample size and departures from multivariate normal-
ity. Therefore, it has been suggested that it must be
interpreted with caution in most applications (Jor-
eskog and Sorbom, 1989; Chau, 1997). Researchers
are hence turning to multiple fit criteria as suggested
by Bollen and Long (1993) to reduce any measuring
biases inherent in different measures. Some of the
other measures of overall model fit that are being used
by researchers are the Goodness of Fit Index (GFI),
which indicates the relative amount of variance and
covariance jointly explained by the model. The
Adjusted Goodness of Fit Index (AGFI) differs from
the GFI in that it adjusts for the number of degrees of
freedom in the model. GFI and AGFI values range
from 0 to 1, with higher values indicating better fit
(Bryne, 1989). GFI and AGFI scores in the 0.80–0.89
range are generally interpreted as representing
reasonable fit; scores of 0.90 and above represent
good fit (Chau, 1997). The Root Mean Square
Residual (RMR) indicates the average discrepancy
between the elements in the sample covariance matrix
and the model-generated covariance matrix. RMR
values range from 0 to 1, with smaller values
indicating better models; values below 0.05 signify
good fit (Bryne, 1989).
4.1. Content validity
Content validity depends on how well the
researchers create measurement items to cover the
domain of the variable being measured (Nunnally,
1978). The evaluation of content validity is a rational
judgmental process not open to numerical evaluation.
Usual method of ensuring content validity is an
extensive review of literature for the choice of the
items and getting inputs from the practitioners and
academic researchers on the appropriateness, com-
pleteness, etc. In addition to the above, experts
performing the manual sorting of the constructs in the
Q-sort method also contributed to the content validity
of the constructs.
4.2. Unidimensionality
The model for unidimensionality can be written,
following Joreskog’s conventions of measurement
model specifications, as:
X ¼ Ljþ d
where X is a vector of p indicators, L is p � k matrix
of factor loadings, j is a k (< p)-vector of theoretical
factors, and d is a p-vector of unique scores (i.e.,
S. Li et al. / Journal of Operations Management 23 (2005) 618–641628
random errors) (Venkatraman and Ramanujam, 1987).
Assuming that E(j) = E(d) = 0, E(jj0) = F, and
E(dd0) = C, the variance–covariance matrix S of X
can be written as
S ¼ LFL0 þ C
where F is the matrix of inter-correlations among the
factors, and C is a symmetric matrix of error variances
(ud) for the measures. As mentioned before we use
multiple fit criteria to test for unidimensionality and
reduce any measuring biases inherent in different
measures. Two goodness of fit indexes were used:
GFI and RMR. GFI values of 0.90 and higher, or RMR
values of 0.05 or lower suggest no evidence of a lack
of unidimensionality.
SCM practices construct was initially represented
by 6 dimensions and 40 items (see Appendix B). A
single factor LISREL measurement model is specified
for each dimension of SCM practices. Following Sethi
and King (1994), iterative modifications were made
for each of the constructs by observing modification
indices and coefficients to improve key model fit
statistics. Further, as recommended by Joreskog and
Sorbom (1989), only one item was altered at a time to
avoid over-modification of the model. This iterative
process continued until all model parameters and key
fit indices met recommended criteria. If the constructs
have less than four items, model fit statistics could not
be obtained. In these cases, two-factor model was
tested by adding the items of another construct. The
items of another construct are added only to provide a
Table 4
Assessment of unidimensionality and convergent validity
Construct Indicators x2
Supplier rationalization (SR)* 2 37.53
Strategic supplier partnership (SSP) 6 22.50
Customer relationship practice (CRP) 5 9.69
Information sharing (IS) 6 18.44
Information quality (IQ) 5 10.78
Internal lean practices (ILP) 5 13.38
Postponement (POS)* 3 37.01
Delivery dependability (DD)* 2 47.90
Time to market (TM) 4 4.73
Note: Constructs marked by an asterisk have either 2 or 3 items and the m
model was tested by adding the items of another construct. The items
rationalization and postponement construct, respectively, while the items
construct.
common basis for comparison and to keep items in
sufficient number so that model fit statistics could be
obtained. Appendix C presents the details of this
modification process and the final items.
After this modification, supplier rationalization
was added as an additional sub-construct of SCM
practices. Two items (SSP4 and SSP10) were removed
from strategic supplier partnership; three items (CR1,
CR3, and CR7) were removed from customer
relationship, one item (IS1) was removed from
information sharing, and two items (POS2 and
POS5) were removed from postponement. No items
were removed from information quality and internal
lean practices. The results are summarized in Table 4.
The table presents the number of items measuring
each construct, and statistics for assessing the good-
ness of fit of the measurement model indicating the
unidimensionality of all constructs. The items
removed in the final instrument are identified by an
asterisk in Appendix B.
4.3. Reliability
Traditionally, the Cronbach a coefficient (Cron-
bach, 1951) has been used to evaluate reliability. A
scale is found to be reliable if a is 0.70 or higher
(Nunnally, 1978). However, it has been noted that
Cronbach a uses restrictive assumptions regarding
equal importance of all indicators and the measure of
reliability can be biased. An alternate composite
reliability measure has been suggested (Werts et al.,
P-value GFI RMR Bentler–Bonett (D)
0.01 0.94 0.04 0.95
0.01 0.96 0.03 0.95
0.08 0.98 0.02 0.97
0.03 0.97 0.02 0.97
0.06 0.98 0.02 0.98
0.02 0.97 0.04 0.95
0.07 0.96 0.04 0.94
0.00 0.92 0.05 0.91
0.09 0.99 0.03 0.98
odel fit statistics could not be obtained. In these cases, a two-factor
of strategic supplier partnership construct were added to supplier
of time to market construct were added to delivery dependability
S. Li et al. / Journal of Operations Management 23 (2005) 618–641 629
Table 5
Assessment of reliability
Construct Indicators Reliability
(rc)
Reliability
(a)
Supplier rationalization
(SR)*
2 – 0.93
Strategic supplier
partnership (SSP)
6 0.85 0.86
Customer relationship
practice (CRP)
5 0.84 0.84
Information sharing (IS) 6 0.87 0.86
Information quality (IQ) 5 0.87 0.86
Internal lean practices
(ILP)
5 0.78 0.78
Postponement (POS)* 3 0.74 0.73
Delivery dependability
(DD)*
2 – 0.93
Time to market (TM) 4 0.77 0.76
1974). This reliability measure rc for an underlying
theoretical dimension A, can be calculated as follows:
rc ¼ðP p
i¼1 liÞ2variance A
ðP p
i¼1 liÞ2variance A þP p
i¼1 ud
where rc is the composite measure reliability, p is the
number of indicators, and li is the factor loading
which relates item I to the underlying theoretical
dimension A. When rc is greater than 0.50 it implies
that the variance captured by the factor is more than
that captured by the error components (Bagozzi,
1981). Bagozzi (1981) and Werts et al. (1974) suggest
using rc in conjunction with Cronbach alpha. The
calculated values for Cronbach’s alpha were very
similar to Werts–Linn–Jorsekog coefficient (rc),
Table 5 reported rc and Cronbach’s alpha for each
dimension of SCM practices. Note that all coefficients
are greater than 0.73, indicating good construct relia-
bility of all the constructs.
4.4. Convergent validity
To assess convergent validity of constructs we look
at each item in the scale as a different approach to
measure the construct and determine if they are
convergent. The convergent validity of each scale is
checked using the Bentler–Bonett coefficient (D)
(Bentler and Bonett, 1980), which is the ratio of the
difference between the chi-square value of the null
measurement model and the chi-square value of the
specified measurement model to the chi-square value
of the null model. A value of 0.90 and above
demonstrates strong convergent validity (Hartwick
and Barki, 1994; Segar and Grover, 1993). Bentler–
Bonett coefficient (D) is shown in Table 4 for all the
constructs. Note that all the constructs have values of
0.91 or above, demonstrating strong convergent
validity.
4.5. Discriminant validity
Discriminant validity refers to the uniqueness and
the independence of the measures, i.e., the extent to
which the measures are distinctly different from each
other. A test of discriminant validity is performed
taking two constructs at a time. The constructs are
considered to be distinct if the hypothesis that the two
constructs together form a single construct is rejected.
To test this hypothesis, a pair-wise comparison of
models was performed by comparing the model with
correlation constrained to equal one with an uncon-
strained model (see Fig. 2). A difference between the
x2 values (d.f. = 1) of the two models that is significant
at P < 0.05 level would indicate support for the
discriminant validity criterion (Joreskog, 1971).
Table 6 reports the results of the 21 pair-wise tests
of discriminant validity for SCM practices. All x2
difference are significant at the P < 0.01 level,
indicating strong support for the discriminant validity
criterion.
4.6. Predictive validity
4.6.1. Predictive validity using qualitative measure
(delivery dependability and time to market)
According to Bagozzi and Phillips (1991) and
Fornell (1982), the conceptual meaning of a construct
should be determined not only by its definition and
operationalization but also by its relationship to
antecedents and consequents. Predictive validity is
represented in the form of structural relationships in
addition to the measurement models. The structural
relationship is represented as:
h ¼ Gjþ z
where h is an endogenous theoretical construct (i.e.,
performance), G the matrix of structural coefficients
relating exogenous theoretical construct (i.e., perfor-
S. Li et al. / Journal of Operations Management 23 (2005) 618–641630
Table 6
Assessment of discriminant validity
Description Model fit indices x2 statistics Difference
Unconstrained Constrained Unconstrained model (d.f.) Constrained model (d.f.)
GFI AGFI GFI AGFI
SR with SSP 0.96 0.92 0.81 0.66 35.73 (19) 181.67 (20) 145.94
SR with CRP 0.97 0.93 0.80 0.60 21.24 (13) 171.41 (14) 150.17
SR with IS 0.97 0.94 0.82 0.67 25.01 (19) 176.12 (20) 151.11
SR with IQ 0.97 0.94 0.81 0.61 19.24 (13) 164.84 (14) 145.60
SR with ILP 0.97 0.93 0.80 0.59 23.20 (13) 175.72 (14) 152.52
SR with POS 0.99 0.98 0.77 0.30 2.52 (4) 184.46 (5) 181.94
SSP with CRP 0.94 0.91 0.74 0.62 67.97 (43) 367.75 (44) 299.78
SSP with IS 0.92 0.88 0.71 0.58 106.77 (53) 472.75 (54) 365.98
SSP with IQ 0.93 0.90 0.63 0.44 74.60 (43) 641.01 (44) 566.41
SSP with ILP 0.92 0.88 0.83 0.75 95.02 (43) 215.32 (44) 120.30
SSP with POS 0.96 0.93 0.84 0.74 35.64 (26) 164.23 (27) 128.59
CRP with IS 0.93 0.90 0.67 0.51 76.54 (43) 521.12 (44) 444.58
CRP with IQ 0.95 0.92 0.66 0.46 48.70 (34) 505.56 (35) 456.86
CRP with ILP 0.95 0.91 0.78 0.66 55.69 (34) 267.41(35) 211.72
CRP with POS 0.98 0.95 0.83 0.70 19.54 (19) 154.32 (20) 134.78
IS with IQ 0.92 0.88 0.75 0.63 92.33 (43) 351.60 (44) 259.37
IS with ILP 0.93 0.89 0.79 0.68 84.45 (43) 292.69 (44) 208.24
IS with POS 0.97 0.94 0.84 0.74 28.87 (26) 165.35 (27) 134.48
IQ with ILP 0.95 0.91 0.78 0.65 56.59 (34) 280.27 (35) 223.68
IQ with POS 0.96 0.92 0.83 0.70 34.47 (19) 154.52 (20) 120.05
ILP with POS 0.94 0.89 0.81 0.66 48.85 (19) 182.01 (20) 133.16
All x2 differences are significant (for 1 degree of freedom) at the less than 0.01 level. GFI, goodness of fit index; AGFI, adjusted goodness of fit
index; d.f., degree of freedom.
Fig. 2. Illustrative example of measurement model testing discriminant validity.
S. Li et al. / Journal of Operations Management 23 (2005) 618–641 631
Fig. 3. Illustrative example of structural equation model testing predictive validity.
mance dimension), j the vector of latent variables, and
z is the residuals of endogenous theoretical construct
(see Fig. 3).
This study uses delivery dependability and time to
market to evaluate the predictive validity of the
dimensions of SCM practices. The measures for
delivery dependability and time to market were
adopted from Zhang (2001) and are presented in
Appendix B. Results of tests of unidimensionality,
convergent validity, and reliability for these two
constructs are provided in Tables 4 and 5, respectively.
Table 7 contains the values of Gammas and their
t-values. The t-values show that each of the SCM
constructs do significantly relate to the delivery
dependability and time to market constructs except
supplier rationalization and postponement, thus
establishing the predictive validity of the constructs.
Table 7
Assessment of predictive validity with delivery dependability and
time to market
Delivery
dependability
Time to market
g t-value g t-value
Supplier rationalization – – 0.07 0.69
Strategic supplier partnership 0.14 1.86* 0.32 3.47**
Customer relationship 0.34 4.20** 0.38 4.01**
Information sharing 0.13 1.73* 0.24 2.72**
Information quality 0.20 2.05* 0.18 2.07*
Internal lean practices 0.23 2.62** 0.38 3.89**
Postponement 0.06 0.66 0.09 1.01
Note: The specified model is not converged.** P-value is significant at 0.01.* P-value is significant at 0.05.
The impact of supplier rationalization on performance
outcome may be indirect through strategic partnership
with suppliers, as the rationalization of suppliers is the
basis for building a strategic partnership. Moreover,
since the implementation of postponement is largely
dependent on type of product, its overall impact on
performance outcomes may be not significant.
4.6.2. Predictive validity using quantitative
measures (SCOR model)
The supply chain operations reference model
(SCOR) developed by Supply Chain Council
(www.supply-chain.org) measures the performance
of supply chain by using the 13 metrics (see Table 8).
Those measures can be used as hard, objective
validation of the SCM practices construct.
The respondents in our survey were asked to fill in the
actual performance of their supply chain in terms of
those metrics. To seewhether the level of SCM practices
predicts the level of supply chain performance, the
respondents were first classified into two groups based
on their mean of SCM practices and t-tests were
conducted to see whether there exists significant
difference between these two groups in terms of each
metric in SCOR model except the last four items, which
were ignored from the analysis since most of the
respondent left them blank. The results are presented in
Table 8. Significant differences between those two
groups were found in most of the metrics. Compared to
organizations with lower level of SCM practices,
organizations with high level of SCM practices have
better performance in term of delivery performance to
commit date (increased from 84%to 90%or 6% higher),
S. Li et al. / Journal of Operations Management 23 (2005) 618–641632
Table 8
Assessment of predictive validity with SCOR model
Supply chain performance Actual performance
Organizations with
higher level of
SCM practices
Organizations
with lower level
of SCM practices
Mean difference t-test
Value Percentage t-value a
Delivery performance to commit date 90.4% 83.5% – 6.9 2.32 0.022
Fill rate 94.0% 91.8% – 2.2 0.67 0.509
Perfect order fulfillment 88.9% 84.8% – 4.1 0.87 0.39
Order fulfillment lead time 15 days 19 days 4 26.7 �0.81 0.419
Supply chain response time 11 days 33 days 22 200 �2.24 0.035
Production flexibility 6 days 11 days 5 83.3 �1.45 0.159
Cash-to-cash cycle time 33 days 97 days 64 193.9 �1.77 0.092
Inventory days of supply 30 days 87 days 57 190 �4.52 0.000
Net asset turns (working capital) 11 turns 5 turns 6 120 2.79 0.008
Cost of goods sold*
Total supply chain management costs*
Value-added productivity*
Warranty/returns processing costs*
Note: Items marked by an asterisk were removed from the analysis since most of respondents left them blank.
supply chain response time (decreased from 33 days to
11 days or 200% faster), cash-to-cash cycle time
(decreased from 97 days to 33 days, or 194% faster),
inventory days of supply (decreased from 87 days to 30
days, or 190% shorter), and net asset turns (from 5 turns
to 11 turns, or 120% faster).
Table 8 also shows that organizations with a high
level of SCM practices is associated with a 2%
increase in fill rate, a 4% increase in perfect order
fulfillment, a 27% decrease in order fulfillment lead
time, and a 83% increase in product flexibility.
Overall, the implementation of SCM practices has
improved the performance of supply chain as defined
by SCOR model and thus provides the support for the
predictive validity of SCM practices construct.
5. Implications and conclusions
The major contribution of the represent study is the
development of a set of SCM practices constructs as
well as a rigorously validated measurement instrument
for collecting data in further studies. The confirmation
process is according to the typical standards of scale
development (Raghunathan et al., 1999; Sethi and
King, 1994; Anderson and Gerbing, 1988). We believe
the instrument developed in this paper is parsimonious
and will be of use to researchers for further studies of
SCM practices and their relationships with other
organizational processes and outcomes like competi-
tive advantage, SCM performance, and organizational
performance.
Many organizations still tend to consider supply
chain management as being the same as integrated
logistics management or as a synonym for supplier
management though they are not. Although some
organizations have realized the importance of SCM,
they lack an understanding of what constitutes a
comprehensive set of SCM practices. The measures of
SCM practices provided in this paper can be useful to
SCM managers in evaluating their current SCM
practices. This can help the managers to identify the
strengths and weaknesses of their SCM practices.
In conclusion this research was an attempt to
conceptualize and develop measures of SCM practices
and a parsimonious measurement instrument. The
instrument was rigorously tested for content validity,
unidimensionality, discriminant validity, predictive
validity and reliability. The development of the SCM
practices measure is expected to motivate and
facilitate further theory development and empirical
investigation in this field.
6. Limitation of the study and future research
As with most of empirical research, there are a few
limitations of the present study. This study evaluates
S. Li et al. / Journal of Operations Management 23 (2005) 618–641 633
SCM practices from the standpoint of a manufacturing
firm (the centric firm in a supply chain). Some
constructs, such as internal lean practices and
postponement, may be not appropriate for distributors
and retailers (the firms at the end of a supply chain).
For a manufacturing firm, the level of postponement
may be associated with make-to-order versus make-
to-stock production systems. The instrument thus fits
best manufacturers with a make-to-order system.
As the concept of SCM is complex and involves a
network of companies in the effort of producing and
delivering a final product, its entire domain can not be
covered in just one study. Future study can develop
additional measurement for the practices of internal
supply chain, such as total quality management (Tan
et al., 2002), cross-functional coordination (Chen and
Paulraj, 2004), and internal integration (Pagell, 2004;
Braganza, 2002). Furthermore, inter-organizational
relationships, such as trust, commitment, shared vision
(Li, 2002), risk and award sharing, and agreed supply
chain leadership (Min and Mentzer, 2004) can also be
incorporated into the SCM practices construct as they
are thefoundation forbuildinganeffective supply chain.
Future research should expand the SCM practices
construct by including the above dimensions and focus
on testing and validating the combined construct.
It should be noted that the implementation of various
SCM practices may be influenced by contextual factors,
such as firm size, a firm’s position in the supply chain,
supply chain length, and channel structure. For
example, the larger organizations may have higher
levels of SCM practices since they usually have more
complex supply chain networks necessitating the need
for more effective management of supply chain. As
pointed out before, postponement and internal lean
Appendix A. Demographic data for the respondents (s
Variables Total responsesa
Number of employees (194)
100–250 74 (38.1%)
251–500 27 (13.9%)
501–1000 19 (9.8%)
>1000 74 (38.1%)
Sales volume in millions of US$ (190)
<10 5 (2.6%)
10 to <25 37 (19.5%)
25 to <50 28 (14.7%)
practice are not appropriate for firms at the end of a
supply chain (distributors and retailers). The level of
information quality may be influenced negatively by the
length of a supply chain. Since information suffers from
delay and distortion as it travels along the supply chain,
the shorter the supply chain, the less chances it will get
distorted. The higher level of postponement may be
associated with make-to-order versus make-to-stock
production systems. Future study can study the impact
of such factors on the SCM practices.
The use of single respondent may generate some
measurement inaccuracy. Future research should
survey multiple respondents (marketing, IT and
operations managers) from a single organization
using the instrument developed in this study; the
discrepancies of SCM perception between the groups
and the impact of such discrepancies on overall
performance can thus be examined. It will also be of
interest to examine the relationships between seven
constructs of SCM practices. For example, supplier
rationalization, strategic supplier partnership and
customer relationship can be combined into external
relationship management; information sharing and
information quality can be combined into information
management; and internal lean practices and post-
ponement can be combined into internal supply chain
practices. The interactions among external relation-
ship management, information management, and
internal supply chain practices can be investigated.
Future research can also compare the SCM practices
among all the participants within a supply chain
(suppliers, manufactures, distributors, wholesales,
retailers, etc.). It is of interest to investigate how
the SCM practices differ across each participating
organization within a supply chain.
ample size 196)
First wavea Second and third wavesa
36 (38.7%) 38 (37.6%)
12 (12.9%) 15 (14.6%)
7 (7.5%) 12 (11.9%)
38 (40.9%) 36 (35.6%)
4 (4.4%) 1 (1.0%)
18 (20.0%) 19 (19.0%)
9 (10.0%) 19 (19.0%)
S. Li et al. / Journal of Operations Management 23 (2005) 618–641634
Appendix A (Continued)
Variables Total responsesa First wavea Second and third wavesa
50 to <100 26 (13.7%) 14 (15.6%) 12 (12.0%)
>100 94 (49.5%) 45 (50.0%) 49 (49.0%)
Job title (194)
CEO/President/Vice President 14 (7.2%) 10 (10.6%) 4 (4.0%)
Director 35 (18.0%) 17 (18.3%) 18 (17.8%)
Manager 121 (63.4%) 54 (58.1%) 67 (66.3%)
Other 24 (12.4%) 12 (12.9%) 12 (11.9%)
Years stayed at the organization (194)
<2 15 (7.7%) 12 (12.9%) 3 (3.0%)
2–5 29 (14.9%) 12 (12.9%) 17 (16.8%)
6–10 32 (16.5%) 15 (16.1%) 17 (16.8%)
>10 118 (60.8%) 54 (58.1%) 64 (63.4%)
a Frequency in percentage.
Appendix B. Instrument for SCM practices, delivery dependability, and time-to-market
Strategic supplier partnership (SSP)
SSP1* We rely on a few dependable suppliers
SSP2* We rely on a few high quality suppliers
SSP3 We consider quality as our number one criterion in selecting suppliers
SSP4* We strive to establish long-term relationship with our suppliers
SSP5 We regularly solve problems jointly with our suppliers
SSP6 We have helped our suppliers to improve their product quality
SSP7 We have continuous improvement programs that include our key suppliers
SSP8 We include our key suppliers in our planning and goal-setting activities
SSP9 We actively involve our key suppliers in new product development processes
SSP10* We certify our suppliers for quality
Customer relationship (CR)
CR1* We frequently evaluate the formal and informal complaints of our customers
CR2 We frequently interact with customers to set reliability, responsiveness, and other standards for us
CR3* We have frequent follow-up with our customers for quality/service feedback
CR4 We frequently measure and evaluate customer satisfaction
CR5 We frequently determine future customer expectations
CR6 We facilitate customers’ ability to seek assistance from us
CR7* We share a sense of fair play with our customers
CR8 We periodically evaluate the importance of our relationship with our customers
Information sharing (IS)
IS1* We share our business units’ proprietary information with trading partners
IS2 We inform trading partners in advance of changing needs
IS3 Our trading partners share proprietary information with us
IS4 Our trading partners keep us fully informed about issues that affect our business
IS5 Our trading partners share business knowledge of core business processes with us
IS6 We and our trading partners exchange information that helps establishment of business planning
IS7 We and our trading partners keep each other informed about events or changes that may affect the other partners
Information quality (IQ)
IQ1 Information exchange between our trading partners and us is timely
IQ2 Information exchange between our trading partners and us is accurate
IQ3 Information exchange between our trading partners and us is complete
IQ4 Information exchange between our trading partners and us is adequate
IQ5 Information exchange between our trading partners and us is reliable
S. Li et al. / Journal of Operations Management 23 (2005) 618–641 635
Appendix B. (Continued)
Internal lean practices (ILP)
ILP1 Our firm reduces set-up time
ILP2 Our firm has continuous quality improvement program
ILP3 Our firm uses a ‘‘Pull’’ production system
ILP4 Our firm pushes suppliers for shorter lead-times
ILP5 Our firm streamlines ordering, receiving and other paperwork from suppliers
Postponement (POS)
POS1 Our products are designed for modular assembly
POS2* Our production process modules can be re-arranged so that customization can be carried out later at distribution centers
POS3 We delay final product assembly activities until customer orders have actually been received
POS4 We delay final product assembly activities until the last possible position (or nearest to customers) in the supply chain
POS5* Our goods are stored at appropriate distribution points close to the customers in the supply chain
Delivery dependability: an organization is capable of providing on time the type and volume of product required by customer(s)
DD1* We deliver the kind of products needed
DD2 We deliver customer order on time
DD3 We provide dependable delivery
Time to market: an organization is capable of introducing new products faster than major competitors
TM1 We deliver product to market quickly
TM2 We are first in the market in introducing new products
TM3 We have time-to-market lower than industry average
TM4 We have fast product development
Items marked by an asterisk were removed in the final instrument.
Appendix C. Description of modification process and assessment of unidimensionality and convergent
validity of SCM constructs
Items Fit indices
SCM practices-strategic supplier partnershipInitial model SSP1, SSP2, SSP3, SSP4, SSP5,
SSP6, SSP7, SSP8, SSP9, SSP10
x2 = 224.83; P = 0.00; GFI = 0.81; AGFI = 0.71; NFI = 0.65
l Coefficients of items SSP1, SSP2, and SSP10 in the above mode were very low (0.25, 0.37, and 0.40, respectively). Item SSP1 with lowest l
coefficients was dropped for the next iteration
Iteration 1 SSP2, SSP3, SSP4, SSP5, SSP6,
SSP7, SSP8, SSP9, SSP10
x2 = 70.84; P = 0.00; GFI = 0.93; AGFI = 0.88; NFI = 0.90
l Coefficients of items SSP2 and SSP10 were very low (0.34 and 0.39, respectively). Item SSP2 with lowest l coefficients was dropped for the
next iteration
Iteration 2 SSP3, SSP4, SSP5, SSP6, SSP7,
SSP8, SSP9, SSP10
x2 = 64.51; P = 0.00; GFI = 0.92; AGFI = 0.86; NFI = 0.91;
l Coefficient of item SSP10 was still very low (0.40). Item SSP10 was removed for the next iteration
Iteration 3 SSP3, SSP4, SSP5, SSP6,
SSP7, SSP8, SSP9
x2 = 53.47; P = 0.00; GFI = 0.93; AGFI = 0.85; NFI = 0.92;
Although all l coefficient were good, the AGFI (0.85) was a little low indicating possibility of error correlation. The modification index indicated
high error correlation between SSP4 and SSP5 (20.92). It was decided to drop item SSP4 since, on an examination of the description of the two
items, it appeared that item SSP4 was too general
S. Li et al. / Journal of Operations Management 23 (2005) 618–641636
Appendix C (Continued)
Iteration 4 SSP3, SSP5, SSP6, SSP7, SSP8, SSP9 x2 = 22.50; P = 0.01; GFI = 0.96; AGFI = 0.91; NFI = 0.95
The final model had both satisfactory l coefficient and excellent model fit. Therefore, no further modifications were done. The items are listed
below
Strategic supplier partnership (SSP)
SSP1* We rely on a few dependable suppliers
SSP2* We rely on a few high quality suppliers
SSP3 We consider quality as our number one criterion in selecting suppliers
SSP4* We strive to establish long-term relationship with our suppliers
SSP5 We regularly solve problems jointly with our suppliers
SSP6 We have helped our suppliers to improve their product quality
SSP7 We have continuous improvement programs that include our key suppliers
SSP8 We include our key suppliers in our planning and goal-setting activities
SSP9 We actively involve our key suppliers in new product development processes
SSP10* We certify our suppliers for quality
Further analysis To make sure no important items were deleted from the purification process, the four removed items for strategic supplier
partnership were observed carefully. It was found that SSP1, SSP2, and SSP10 are associated with supplier rationalization,
which is an important component in the partnership with suppliers
To test whether these three items form an independent construct, two two-factor models were tested by adding the six items
of strategic supplier partnership
Iteration 1 SSP1, SSP2, SSP10 x2 = 72.80; P = 0.00; GFI = 0.92; AGFI = 0.87; NFI = 0.91
SSP3, SSP5, SSP6, SSP7, SSP8, SSP9
The model tested in iteration 1 included the 3 items of supplier rationalization and the items from strategic supplier partnership. l Coefficient of
item SSP10 in iteration 1 was low (0.02). Item SSP10 was removed for the next iteration
Iteration 2 SSP1, SSP2 x2 = 35.75; P = 0.01; GFI = 0.96; AGFI = 0.92; NFI = 0.95
SSP3, SSP5, SSP6, SSP7, SSP8, SSP9
The model resulting from iteration 2 had both satisfactory l coefficient and excellent model fit. Therefore, no further modifications were done. A
new construct, supplier rationalization, was included in the further analysis. The items are listed below
Supplier rationalization (SR)
SSP1 We rely on a few dependable suppliers
SSP2 We rely on a few high quality suppliers
SCM practices-customer relationshipInitial Model CRP1, CRP2, CRP3, CRP4, CRP5, CRP6, CRP7, CRP8 x2 = 128.77; P = 0.00; GFI = 0.86; AGFI = 0.74; NFI = 0.85
l Coefficient of item CRP7 in the above model was low (0.46). Item CRP7 was removed for the next iteration
Iteration 1 CRP1, CRP2, CRP3, CRP4, CRP5, CRP6, CRP8 x2 = 90.56; P = 0.00; GFI = 0.88; AGFI = 0.77; NFI = 0.88
Although all l coefficient were good, the; AGFI (0.77) was low indicating possibility of error correlation. The modification index indicated high
error correlation between CRP2 and CRP3 (30.12). It was decided to drop item CRP3 since, on an examination of the description of the two
items, it appeared that item CRP3 could be constructed as part of CRP2.
Iteration 2 CRP1, CRP2, CRP4, CRP5, CRP6, CRP8 x2 = 33.07; P = 0.00; GFI = 0.95; AGFI = 0.88; NFI = 0.93
The model resulting from iteration 2 showed the improvements in all fit indices. However, the modification index still indicated some moderately
error correlation between CRP1 and CRP2 (17.97). It was decided to drop CRP1, based on the examination of the description of the two items, it
appeared that CRP2 subsumed CRP1
S. Li et al. / Journal of Operations Management 23 (2005) 618–641 637
Appendix C (Continued)
Iteration 3 CRP2, CRP4, CRP5, CRP6, CRP8 x2 = 9.69; P = 0.08; GFI = 0.98; AGFI = 0.94; NFI = 0.97
The final model had both satisfactory l coefficient and excellent model fit. Therefore, no further modifications were done. The items are listed
below
Customer relationship (CR)
CR1* We frequently evaluate the formal and informal complaints of our customers
CR2 We frequently interact with customers to set reliability, responsiveness, and other standards for us
CR3* We have frequent follow-up with our customers for quality/service feedback
CR4 We frequently measure and evaluate customer satisfaction
CR5 We frequently determine future customer expectations
CR6 We facilitate customers’ ability to seek assistance from us
CR7* We share a sense of fair play with our customers
CR8 We periodically evaluate the importance of our relationship with our customers
Information sharingInitial Model IS1, IS2, IS3, IS4, IS5, IS6, IS7 x2 = 80.42; P = 0.00; GFI = 0.89; AGFI = 0.79; NFI = 0.87
Although all l coefficient were good, the AGFI (0.79) was low indicating possibility of error correlation. The modification index indicated high
error correlation between IS1 and IS3 (41.19) and between IS1 and IS2 (15.53), it was decided to drop item IS1 since this item has error
correlation with the other two
Iteration 1 IS2, IS3, IS4, IS5, IS6, IS7 x2 = 18.44; P = 0.03; GFI = 0.97; AGFI = 0.93; NFI = 0.97
The final model had both satisfactory l coefficient and excellent model fit. Therefore, no further modifications were done. The items are listed
below
Information sharing (IS)
IS1* We share our business units’ proprietary information with trading partners
IS2 We inform trading partners in advance of changing needs
IS3 Our trading partners share proprietary information with us
IS4 Our trading partners keep us fully informed about issues that affect our business
IS5 Our trading partners share business knowledge of core business processes with us
IS6 We and our trading partners exchange information that helps establishment of business planning
IS7 We and our trading partners keep each other informed about events or changes that may affect the other partners
SCM practices-information qualityInitial Model IQ1, IQ2, IQ3, IQ4, IQ5 x2 = 10.78; P = 0.06; GFI = 0.98; AGFI = 0.94; NFI = 0.98
The initial model had both satisfactory l coefficients and excellent model fit. Therefore, no modifications were done. The items are listed below
Information quality (IQ)
IQ1 Information exchange between our trading partners and us is timely
IQ2 Information exchange between our trading partners and us is accurate
IQ3 Information exchange between our trading partners and us is complete
IQ4 Information exchange between our trading partners and us is adequate
IQ5 Information exchange between our trading partners and us is reliable
SCM practices-postponementInitial Model (POS1, POS2, POS3, POS4, POS5) x2 = 39.71; P = 0.00; GFI = 0.92; AGFI = 0.77; NFI = 0.83
l Coefficient of item POS1 in the above mode was very low (0.11). Item POS1 was removed for the next iteration
Iteration 1 POS1, POS2, POS3, POS4 x2 = 20.99; P = 0.00; GFI = 0.95; AGFI = 0.74; NFI = 0.89
Although all l coefficient were good, the AGFI (0.74) was very low indicating possibility of error correlation. A model run with either one of
these four items removed would not have yielded model fit statistics since only three items remained resulting in the degrees of freedom being
zero. Two two-factor models were tested by adding the items of strategic supplier partnership
S. Li et al. / Journal of Operations Management 23 (2005) 618–641638
Appendix C (Continued)
Iteration 2 POS1, POS2, POS3, POS4 x2 = 74.62; P = 0.00; GFI = 0.93; AGFI = 0.89; NFI = 0.92
SSP3, SSP5, SSP6, SSP7, SSP8, SSP9
Iteration 3 POS1, POS3, POS4 x2 = 37.01; P = 0.07; GFI = 0.96; AGFI = 0.93; NFI = 0.94
SSP3, SSP5, SSP6, SSP7, SSP8, SSP9
The model tested in iteration 2 included the items from strategic supplier partnership and the items from the iteration 1 of postponement. The
model in iteration 3 also had all the items from strategic supplier partnership and all except item POS2 from postponement. Given the indicated
error correction between items POS1 and POS2 (19.16), it was decided to drop item POS2. Based on the examination of the description of the two
items, it appeared that POS1 subsumed POS2. The model resulting from iteration 3 showed improvement. No further modifications were done.
The items are listed below
Postponement (POS)
POS1 Our products are designed for modular assembly
POS2* Our production process modules can be re-arranged so that customization can be carried out later at distribution
centers
POS3 We delay final product assembly activities until customer orders have actually been received
POS4 We delay final product assembly activities until the last possible position (or nearest to customers) in the supply
chain
POS5* Our goods are stored at appropriate distribution points close to the customers in the supply chain
Time to marketInitial Model TM1, TM2, TM3, TM4 x2 = 4.73; P = 0.09; GFI = 0.99; AGFI = 0.94; NFI = 0.98
The initial model had both satisfactory l coefficients and excellent model fit. Therefore, no modifications were done. The items are listed below
Time to market (TM)
TM1 We deliver product to market quickly
TM2 We are first in the market in introducing new products
TM3 We have time-to-market lower than industry average
TM4 We have fast product development
Delivery dependabilityInitial Model DD1, DD2, DD3
Iteration 1 DD1, DD2, DD3 x2 = 82.89; P = 0.00; GFI = 0.89; AGFI = 0.77; NFI = 0.84
TM1, TM2, TM3, TM4
Since a model run with three items would not have yielded model fit statistics. A two-factor model was tested by adding the items of time to
market. l Coefficient of item DD1 was very low (0.11). Item DD1 was removed for the next iteration
Iteration 2 DD2, DD3 x2 = 47.90; P = 0.00; GFI = 0.92; AGFI = 0.80; NFI = 0.91
TM1, TM2, TM3, TM4
The model resulting from iteration 2 showed significant improvement. No further modification can be done. The items are listed below
Delivery dependability (DD)
DD1* We deliver the kind of products needed
DD2 We deliver customer order on time
DD3 We provide dependable delivery
S. Li et al. / Journal of Operations Management 23 (2005) 618–641 639
References
Aggarwal, S., 1997. Flexibility management: the ultimate strategy.
Industrial Management 39 (1), 26–31.
Alvarado, U.Y., Kotzab, H., 2001. Supply chain management: the
integration of logistics in marketing. Industrial Marketing Man-
agement 30 (2), 183–198.
Alvarez, D., 1994. Solving the puzzle of industry’s rubic cube-
effective supply chain management. Logistics Focus 2 (4), 2–4.
Amstrong, J.S., Overton, T.S., 1977. Estimating Nonresponse Bias
in Mail Surveys. Journal of Marketing Research 14, 396–402.
Anderson, J.C., Gerbing, D.W., 1988. Structural equation modeling
in practice: a review and recommended two step approach.
Psychological Bulletin 103, 411–423.
Bagozzi, R.P., 1980. Causal Models in Marketing. Wiley, New York.
Bagozzi, R.P., 1981. An examination of the validity of two models of
attitude. Multivariate Behavioral Research 18, 21–37.
Bagozzi, R.P., Phillips, L.W., 1982. Representing and testing orga-
nizational theories: a holistic construct. Administrative Science
Quarterly 27, 459–489.
Bagozzi, R.P., Phillips, L.W., 1991. Assessing construct validity in
organizational research. Administrative Science Quarterly 36,
421–458.
Balsmeier, P.W., Voisin, W., 1996. Supply chain management: a
time-based strategy. Industrial Management 38 (5), 24–27.
Banfield, E., 1999. Harnessing Value in the Supply Chain. Wiley,
New York, NY.
Beamon, B.M., 1998. Supply chain design and analysis: models and
methods. International Journal of Production Economics 55 (3),
281–294.
Bechtel, C., Jayaram, J., 1997. Supply chain management: a stra-
tegic perspective. International Journal of Logistics Manage-
ment 8 (1), 15–34.
Bentler, P.M., Bonett, D.G., 1980. Significant test and goodness of fit
in the analysis of covariance structures. Psychological Bulletin
88, 588–606.
Berry, D., Towill, D.R., Wadsley, N., 1994. Supply chain manage-
ment in the electronics products industry. International Journal
of Physical Distribution and Logistics Management 24 (10),
20–32.
Boddy, D., Cahill, D., Charles, M., Fraser-Kraus, H., Macbeth, D.,
1998. Success and failure in implementing partnering. European
Journal of Purchasing and Supply Management 4 (2,3), 143–
151.
Bollen, K.A., Long, J.S., 1993. Testing Structural Equation Models.
Sage publications, NewBury Park, CA.
Braganza, A., 2002. Enterprise integration: creating competitive
capabilities. Integrated Manufacturing Systems 13 (8), 562–572.
Bryne, B.M., 1989. A Primer of LISREL: Basic Applications and
Programming for Confirmatory Factor Analytic Model.
Springer-Verlag, New York.
Campbell, D.T., Fiske, D.W., 1959. Convergent and discriminant
validation by the multitrait-multimethod matrix. Psychological
Bulletin 56, 81–105.
Chau, P.Y.K., 1997. Reexamining a model for evaluating informa-
tion center success using a structural equation modeling
approach. Decision Sciences 28, 309–334.
Chen, I.J., Paulraj, A., 2004. Towards a theory of supply chain
management: the constructs and measurements. Journal of
Operations Management 22 (2), 119–150.
Chizzo, S.A., 1998. Supply chain strategies: solutions for
the customer-driven enterprise. In: Software Magazine.
Supply Chain Management Directions Supplement (January),
pp. 4–9.
Choi, T.Y., Hartley, J.L., 1996. An exploration of supplier selection
practices across the supply chain. Journal of Operations
Management 14 (4), 333–343.
Churchill, G.A., 1979. A paradigm for developing better measures
of marketing constructs. Journal of Marketing Studies 16,
12–27.
Cigolini, R., Cozzi, M., Perona, M., 2004. A new framework for
supply chain management: conceptual model and empirical test.
International Journal of Operations and Production Management
24 (1), 7–14.
Clark, T.H., Lee, H.G., 2000. Performance, interdependence and
coordination in business-to-business electronic commerce and
supply chain management. Information Technology and Man-
agement 1 (1–2), 85–105.
Claycomb, C., Droge, C., Germain, R., 1999. The effect of
just-in-time with customers on organizational design and per-
formance. International Journal of Logistics Management 10
(1), 37–58.
Cohen, J., 1960. A coefficient of agreement for nominal scales.
Educational and Psychological Measurement 20 (Spring),
37–46.
Council of Logistics Management (CLM), 2000. What It’s All
About. Council of Logistics Management, Oak Brook, IL.
Cronbach, L., 1951. Coefficient alpha and the internal structure of
tests. Psychometrica 16, 297–334.
Croom, S., Romano, P., Giannakis, M., 2000. Supply chain manage-
ment: an analytical framework for critical literature review.
European Journal of Purchasing and Supply Management 6
(March (1)), 67–83.
Donlon, J.P., 1996. Maximizing value in the supply chain. Chief
Executive 117 (October), 54–63.
Fisher, M.L., 1997. What is the right supply chain for your product?
Harvard Business Review 75 (2), 105–116.
Fisher, M.L., Hammond, J.H., Obermeyer, W.R., Raman, A., 1994.
Making supply meet demand in an uncertain world. Harvard
Business Review 72 (3), 83–93.
Forker, L.B., Ruch, W.A., Hershauer, J.C., 1999. Examining sup-
plier improvement efforts from both sides. Journal of Supply
Chain Management 35 (3), 40–50.
Fornell, C., 1982. A second generation multivariate analysis.
Methods, vol. 1. Praeger, New York.
Frohlich, M.T., Westbrook, R., 2001. Arcs of integration: an inter-
national study of supply chain strategies. Journal of Operations
Management 19 (2), 185–200.
Fuller, J.B., O’Conor, J., Rawlinson, R., 1993. Tailored logistics: the
next advantage. Harvard Business Review 71 (3), 87–98.
Gunasekaran, A., Patel, C., Tirtiroglu, E., 2001. Performance mea-
sures and metrics in a supply chain environment. International
Journal of Operations and Production Management 21 (1–2),
71–87.
S. Li et al. / Journal of Operations Management 23 (2005) 618–641640
Hall, R.W., 1993. A framework for linking intangible resources and
capabilities to sustainable competitive advantage. Strategy Man-
agement Journal 14, 607–618.
Handfield, R.B., Pannesi, R.T., 1995. Antecedents of lead-time
competitiveness in make-to-order manufacturing firms. Interna-
tional Journal of Production Research 33 (2), 511–537.
Handfield, R.B., Bechtel, C., 2002. The role of trust and relationship
structure in improving supply chain responsiveness. Industrial
Marketing Management 31 (4), 367–382.
Handfield, R.B., Nichols Jr., E.L., 1999. Introduction to Supply
Chain Management. Prentice Hall, Upper Saddler River, New
Jersey.
Hartwick, J., Barki, H., 1994. Explaining the role of user participa-
tion in information systems use. Management Science 40, 440–
465.
Holmberg, S., 2000. A systems perspective on supply chain mea-
surements. International Journal of Physical Distribution and
Logistics Management 30 (10), 847–868.
Jarrell, J.L., 1998. Supply chain economics. World Trade 11 (11),
58–61.
Jarvenpaa, S., 1989. The effect of task demands and graphical
format on information process strategies. Management Science
35 (March (3)), 285–303.
Jones, C., 1998. Moving beyond ERP: making the missing link.
Logistics Focus 6 (7), 2–7.
Joreskog, K.G., 1971. Simultaneous Factor analysis in several
populations. Psychometrika 57, 409–426.
Joreskog, K.G., Sorbom, D., 1989. LISREL 7 Users’ Reference
Guide. Scientific Software Inc., Chicago, IL.
Kessler, E., Chakrabarti, A., 1996. Innovation speed: a conceptual
mode of context, antecedents, and outcomes. The Academy of
Management Review 21 (4), 1143–1191.
Koufteros, X.A., Vonderembse, M.A., Doll, W.J., 1997. Competitive
capabilities: measurement and relationships. Proceedings Deci-
sion Science Institute 3, 1067–1068.
Lalonde, B.J., 1998. Building a supply chain relationship. Supply
Chain Management Review 2 (2), 7–8.
Lambert, D.M., Harrington, T.C., 1990. Measuring nonresponse bias
in mail surveys. Journal of Business Logistics 11, 5–25.
Lamming, R.C., 1996. Squaring lean supply with supply chain
management. International Journal of Operations and Produc-
tion Management 16 (2), 183–196.
Landis, J.R., Koch, C.G., 1977. The measurement of observer
agreement for categorical data. Biometrics 33 (March), 159–
174.
Lee, H.L., Billington, C., 1995. The evolution of supply chain
management models and practices at Hewlett Packard. Interface
25 (5), 42–63.
Lee, H.L., Padmanabhan, V., Whang, S., 1997. Information distor-
tion in a supply chain: the bullwhip effect. Management Science
43 (4), 546–558.
Lee, J., Kim, Y., 1999. Effect of partnership quality on IS out-
sourcing: conceptual framework and empirical validation. Jour-
nal of Management Information Systems 15 (4), 26–61.
Li, S.H., 2002. An Integrated Model for Supply Chain Management
Practice, Performance and Competitive Advantage. Doctoral
Dissertation, University of Toledo, Toledo, OH.
Magretta, J., 1998. The power of virtual integration: an interview
with Dell computers’ Michael Dell. Harvard Business Review
76 (2), 72–84.
Mason-Jones, R., Towill, D.R., 1997. Information enrichment:
designing the supply chain for competitive advantage. Supply
Chain Management 2 (4), 137–148.
Mason-Jones, R., Towill, D.R., 1999. Total cycle time compression
and the agile supply chain. International Journal of Production
Economics 62 (1–2), 61–73.
McAdam, R., McCormack, D., 2001. Integrating business processes
for global alignment and supply chain management. Business
Process Management Journal 7 (2), 113–130.
McIvor, R., 2001. Lean supply: the design and cost reduction
dimensions. European Journal of Purchasing and Supply Chain
Management 7 (4), 227–242.
Mentzer, J.T., Min, S., Zacharia, Z.G., 2000. The nature of inter-firm
partnering in supply chain management. Journal of Retailing 76
(4), 549–568.
Metters, R., 1997. Quantifying the bullwhip effect in supply chains.
Journal of Operations Management 15 (2), 89–100.
Min, S., Mentzer, J.T., 2004. Developing and measuring supply
chain concepts. Journal of Business Logistics 25 (1), 63–99.
Moberg, C.R., Cutler, B.D., Gross, A., Speh, T.W., 2002. Identifying
antecedents of information exchange within supply chains.
International Journal of Physical Distribution & Logistics Man-
agement 32 (9), 755–770.
Monczka, R.M., Trent, R.J., Callahan, T.J., 1993. Supply base
strategies to maximize supplier performance. International Jour-
nal of Physical Distribution and Logistics 24 (1), 42–54.
Monczka, R.M., Petersen, K.J., Handfield, R.B., Ragatz, G.L., 1998.
Success factors in strategic supplier alliances: the buying com-
pany perspective. Decision Science 29 (3), 5553–5577.
Moore, G.C., Benbasat, I., 1991. Development of an instrument to
measure the perceptions of adopting an information technology
innovation. Information Systems Research 2 (2), 192–222.
Narasimhan, R., Jayaram, J., 1998. Causal linkage in supply chain
management: an exploratory study of North American manu-
facturing firms. Decision Science 29 (3), 579–605.
Narasimhan, R., Kim, S.O., 2001. Information system utilization
strategy from supply chain integration. Journal of Business
Logistics 22 (2), 51–76.
Naylor, J.B., Naim, M.M., Berry, D., 1999. Legality: integrating the
lean and agile manufacturing paradigms in the total supply
chain. International Journal of Production Economics 62
(1,2), 107–118.
Noble, D., 1997. Purchasing and supplier management as a future
competitive edge. Logistics Focus 5 (5), 23–27.
Novack, R.A., Langley Jr., C.J., Rinehart, L.M., 1995. Creating
Logistics Value: Themes for the Future. Council of Logistics
Management, Oak Brook, IL.
Nunnally, J., 1978. Psychometric Theory. McGraw-Hill, New York.
Pagell, M., 2004. Understanding the factors that enable and inhibit
the integration of operations, purchasing and logistics. Journal of
Operations Management 22 (5), 459–487.
Pagh, J.D., Cooper, M.C., 1998. Supply chain postponement and
speculation strategies: how to choose the right strategy. Journal
of Logistics Management 19 (2), 13–33.
S. Li et al. / Journal of Operations Management 23 (2005) 618–641 641
Power, D.J., Sohal, A., Rahman, S.U., 2001. Critical success factors
in agile supply chain management: an empirical study. Interna-
tional Journal of Physical Distribution and Logistics Manage-
ment 31 (4), 247–265.
Ragatz, G.L., Handfield, R.B., Scannell, T.V., 1997. Success factors
for integrating suppliers into new product development. Journal
of Product Innovation Management 14 (3), 190–202.
Raghunathan, B., Raghunathan, T.S., Tu, Q., 1999. Dimensionality
of the strategic grid framework: the construct and its measure-
ment. Information System Research 10 (4), 343–355.
Romano, P., Vinelli, A., 2001. Quality management in a supply
chain perspective: strategic and operative choices in a textile-
apparel network. International Journal of Operations and Pro-
duction Management 21 (4), 446–460.
Rondeau, P.J., Vonderembse, M.A., Ragu-Nathan, T.S., 2000.
Exploring work system practices for time-based manufacturers:
their impact on competitive advantage. Journal of Operations
Management 18 (5), 509–529.
Segar, A.H., Grover, V., 1993. Re-examining perceived ease of use
and usefulness: a confirmatory factor analysis. MIS Quarterly
17, 517–525.
Sethi, V., King, W.R., 1994. Development of measures to assess the
extent to which an information technology application provides
competitive advantage. Management Science 40 (12), 1601–
1627.
Shin, H., Collier, D.A., Wilson, D.D., 2000. Supply management
orientation and supplier/buyer performance. Journal of Opera-
tions Management 18 (3), 317–333.
Spekman, R.E., Kamauff Jr., J.W., Myhr, N., 1998. An empirical
investigation into supply chain management: a perspective on
partnerships. Supply Chain Management 3 (2), 53–67.
Stalk, G., 1988. Time- the next source of competitive advantage.
Harvard Business Review 66 (4), 41–51.
Stein, T., Sweat, J., 1998. Killer supply chains. Informationweek 708
(9), 36–46.
Stuart, F.I., 1997. Supply-chain strategy: organizational influence
through supplier alliances. British Academy of Management 8
(3), 223–236.
Tan, K.C., 2001. A framework of supply chain management litera-
ture. European Journal of Purchasing and Supply Management 7
(1), 39–48.
Tan, K.C., Kannan, V.R., Handfield, R.B., 1998. Supply chain
management: supplier performance and firm performance. Inter-
national Journal of Purchasing and Materials Management 34
(3), 2–9.
Tan, K.C., Lyman, S.B., Wisner, J.D., 2002. Supply chain manage-
ment: a strategic perspective. International Journal of Operations
and Production Management 22 (6), 614–631.
Taylor, D.H., 1999. Supply chain improvement: the lean approach.
Logistics Focus 7 (January–February (1)), 14–20.
Thomas, J., 1999. Why your supply chain doesn’t work. Logistics
Management and Distribution Report 38 (6), 42–44.
Tompkins, J., Ang, D., 1999. What are your greatest challenges
related to supply chain performance measurement? IIE Solu-
tions 31 (6), 66.
Towill, D.R., 1997. The seamless chain- the predator’s strategic
advantage. International Journal of Technology Management 13
(1), 37–56.
Van Hoek, R.I., 1998. Measuring the unmeasurable-measuring and
improving performance in the supply chain. Supply Chain
Management 3 (4), 187–192.
Van Hoek, R.I., Voss, R.I., Commandeur, H.R., 1999. Restructuring
European supply chain by implementing postponement strate-
gies. Long Range Planning 32 (5), 505–518.
Venkatraman, N., Ramanujam, V., 1987. Planning system success: a
conceptualization and an operational model. Management
Science 35, 687–705.
Vessey, I., 1984. An Investigation of the Psychological Processes
Underlying the Debugging of Computer Programs. Unpublished
Ph.D. Dissertation, Department of Commerce, The University of
Queensland, Queensland, Australia.
Vesey, J.T., 1991. The new competitors: they think in terms of speed-
to-market. Academy of Management Executive 5 (2), 23–33.
Vonderembse, M.A., Tracey, M., 1999. The impact of supplier
selection criteria and supplier involvement on manufacturing
performance. Journal of Supply Chain Management 35 (3), 33–
39.
Waller, M.A., Dabholkar, P.A., Gentry, J.J., 2000. Postponement,
product customization, and market-oriented supply chain man-
agement. Journal of Business Logistics 21 (2), 133–159.
Walton, L.W., 1996. Partnership satisfaction: using the underlying
dimensions of supply chain partnership to measure current and
expected levels of satisfaction. Journal of Business Logistics 17
(2), 57–75.
Werts, C.E., Linn, R.L., Joreskog, K.G., 1974. Interclass reliability
estimates: testing structural assumptions. Educational and Psy-
chological Measurement 34, 25–33.
Wines, L., 1996. High order strategy for manufacturing. The Journal
of Business Strategy 17 (4), 32–33.
Womack, J., Jones, D., 1996. Lean Thinking. Simon and Schuster,
New York.
Yoshino, M., Rangan, S., 1995. Strategic Alliances: An Entrepre-
neurial Approach to Globalization. Harvard Business School
Press, Boston, MA.
Yu, Z.X., Yan, H., Cheng, T.C.E., 2001. Benefits of information
sharing with supply chain partnerships. Industrial Management
and Data Systems 101 (3), 114–119.
Zhang, Q.Y., 2001. Technology Infusion Enabled Value Chain
Flexibility: A Learning and Capability-Based Perspective. Doc-
toral Dissertation, University of Toledo, Toledo, OH.