Int. J.

Production Economics 158 (2014) 7790

Contents lists available at ScienceDirect

Int. J. Production Economics

journal homepage: www.elsevier.com/locate/ijpe

The impact of barriers and benets of e-procurement on its adoption

decision: An empirical analysis
Peral Tokta-Palut a,n, Ecem Baylav a, Seyhan Teoman b, Mustafa Altunbey c
a
Department of Industrial Engineering, Dou University, 34722 Istanbul, Turkey
b
Logistics and Supply Chain Management Doctorate Program, Maltepe University, 34857 Istanbul, Turkey
c
Logistics and Supply Chain Management Doctorate Program, Dou University, 34722 Istanbul, Turkey

art ic l e i nf o a b s t r a c t

Article history: The aim of this study is to investigate the effects of the barriers and benets of e-procurement systems
Received 25 April 2013 on the e-procurement adoption decision. An empirical analysis is performed for a retail store chain
Accepted 25 June 2014 which operates in book and stationery sector in Turkey. Interpretive Structural Modeling (ISM) technique
Available online 23 July 2014
is applied to the selected barriers and benets to determine their contextual relationships. Inadequate IT
Keywords: infrastructure of suppliers/business partners is found to be the most important barrier in the adoption of
E-procurement e-procurement systems. On the other hand, the most signicant benet is determined as integrated
Barriers information sharing. The barriers and benets, which have high driving power and the capability to
Benets inuence the other drivers, are then integrated into the structural equation model. The results denote
Interpretive structural modeling
that barriers (benets) of e-procurement systems have negative (positive) effect on the e-procurement
Structural equation modeling
adoption decision for the company. In addition, it is found that the effect of benets on the adoption
decision is higher than that of the barriers. Thus, based on this analysis, it would be benecial for the
company to adopt the e-procurement system.
& 2014 Elsevier B.V. All rights reserved.

1. Introduction
Intensifying competition in today's business environment has
highlighted the need to optimize the management of supply chains.
Developments in the information and communication technology
are important tools to manage supply chains effectively. In recent
years, an increasing trend has been observed in the adoption
of e-procurement systems, which help in the integration of the
procurement process throughout the supply chain.
The term e-procurement refers to the integration of procure-
ment process, which includes operations such as negotiation,
ordering, receipt, and post-purchase review (Croom and
Brandon-Jones, 2007). Morris et al. (2000) dene e-procurement
as a system that utilizes Internet technologies and services to
automate and streamline an organization's processes from
requisition to payment.
E-procurement is not a recently discovered process as there
have been many attempts of using electronic systems for devel-
oping an automated procurement technology for organizations,
such as electronic workow systems and EDI (Gunasekaran and
Ngai, 2008). When the traditional procurement system is ana-
lyzed, it can be seen that the process is implemented by using
n
Corresponding author. Tel.: 90 216 544 5555; fax: 90 216 544 5535.
E-mail address: ppalut@dogus.edu.tr (P. Tokta-Palut).
phone, fax, and other communication channels, showing how the
procurement system is developed over time (Hawking et al.,
2004).
E-procurement, which has become one of the fundamental
elements of a supply chain, is still in the process of development
and several studies are performed on this subject. Gunasekaran
and Ngai (2008) state that the supply chain of a company cannot
be integrated successfully without the adoption of e-procurement
systems. The authors conduct a questionnaire based survey in
order to understand the adoption process of e-procurement in
Hong Kong. Davila et al. (2003) carry out a research on the
equilibrium point for the optimum e-procurement system that
should be used by companies which have different attitudes
towards technology. Panayiotou et al. (2004) work on a case study
about the Greek purchasing process and indicate the problems
which may occur. The authors also study e-procurement system
design. Croom and Brandon-Jones (2007) assess the validity of the
forecasts performed by earlier studies which made predictions
about the changes and improvements that will be caused by
e-procurement systems. Gunasekaran et al. (2009) analyze the
current state of e-procurement in SMEs located in the Southcoast
of Massachusetts, and they also examine the factors that affect the
e-procurement adoption. Nguyen (2013) performs an empirical
study for the transport and logistics companies in Australia, and
determines the principal components affecting the e-business
adoption decision.

http://dx.doi.org/10.1016/j.ijpe.2014.07.017
0925-5273/& 2014 Elsevier B.V. All rights reserved.
78 P. Tokta-Palut et al. / Int. J. Production Economics 158 (2014) 7790
In the light of the previous studies on e-procurement systems,
this study focuses on the barriers and benets of e-procurement
systems. Our aim is to investigate the effects of the barriers and
benets on the e-procurement adoption decision. For this purpose,
rst Interpretive Structural Modeling (ISM) is applied to the
barriers and benets of e-procurement in order to understand
the relations within these criteria.
Talib et al. (2011) dene ISM as an interpretive model, which
denes the relations between variables in complex systems by
representing them in a hierarchical conguration. ISM is widely
used to specify the interactions between the elements of a system,
such as barriers and benets. Gorvett and Liu (2006) assess the
risks in a rm by using ISM and analyze the risk prole of the
company. Singh and Kant (2008) use ISM in order to understand
the relations between the barriers of knowledge management and
specify the driving and dependent barriers. Chandramowli et al.
(2011) perform a case study about the barriers of developing
landll sites in order to realize urban planning issues. Luthra et al.
(2011) apply ISM to investigate the relations between the barriers
of green supply chain management for India's automobile
industry. The literature review shows that this is the rst study that
applies ISM to the barriers and benets of e-procurement systems.
In the second part of the study, the results of the ISM model are
validated by using Structural Equation Modeling (SEM). Another
point of interest is to analyze whether the effects of barriers or
benets are higher on the e-procurement adoption decision. For
this purpose, we perform an empirical analysis for a retail store
chain which operates in book and stationery sector in Turkey. To
the best of our knowledge, this is the rst study that integrates
ISM and SEM techniques.
SEM can be used for multivariate data as it is appropriate for
illustrating the relations between exogenous and endogenous
latent variables in one model (Kline, 1998; Aibinu and Al-Lawati,
2010). SEM is a widely used tool especially for psychology,
sociology, and econometrics (Golob, 2003) and there are several
studies applying SEM in the literature. Thus, we only mention the
studies in a related eld. Madeja and Schoder (2003) examine the
impacts of e-procurement adoption process on e-business success.
Ho et al. (2004) investigate the direct and performance impacts of
e-procurement on supply chain orientation from the buyerseller
perspective. Lee and Quaddus (2006) conduct a study in Singapore
about the impacts of buyersupplier relationship on e-purchasing
adoption decision. Vaidyanathan and Devaraj (2008) study
the impacts of the quality of the information ow process and
the quality of logistics fulllment on the satisfaction of
e-procurement performance. Konradt et al. (2012) analyze the
effects of organizational and technical antecedence on usability
and commercial transactions in B2B e-commerce. Finally, Devaraj
et al. (2012) investigate the effects of mixing exibilities and
purchase volumes on e-procurement performance.
The rest of the paper is organized as follows. The research
background and hypotheses are given in Section 2. The methodol-
ogy used in this study is explained in Section 3. Section 4 presents
the data analysis and results of the ISM model, measurement
model, and structural model. Finally, concluding remarks are given
in Section 5.
2. Theoretical background and research hypotheses
2.1. Theoretical background
Procurement is a vital issue in the retail sector and basic supply
chain activities start with procurement. Hence, procurement
directly affects the performance and efciency of the supply chain.
In this study, we consider a retail store chain which operates in
book and stationery sector in Turkey. The company has an annual
procurement volume of approximately 100 million USD, which
covers its demand for more than 20,000 different stock keeping
units provided by 185 suppliers. Currently, this company uses
traditional procurement methods, but the senior management
wants to decide whether to adopt an e-procurement system or
not. As one would expect, there are some barriers to switching to
an electronic procurement system for the company. On the other
hand, the company will experience some benets from the
e-procurement system when compared with the traditional pro-
curement methods. The literature reviews on the barriers and
benets of the e-procurement systems are given in Tables 1 and 2,
respectively. Based on the literature review, 20 barriers and 20
benets are determined. Then, these barriers and benets are
discussed with eight top managers of the company and their
numbers are decreased to 14 barriers and 15 benets. The barriers
and benets used in this study and their explanations are
given below.
2.1.1. Barriers of e-procurement systems
Although it is known that e-procurement systems have several
benets for the company, the company would also face some
barriers while switching to an e-procurement system. The barriers
of e-procurement systems taken into consideration in this study
are given below:
2.1.1.1. Bureaucratic disfunctionalities in practice. There are a number
of specic regulations and standards that have been developed for
e-procurement applications which require that a bureaucratic
procedure to be followed due to the nature of the legal institutions
involved and embraces audit, accountability and compliance
standards with national and international rules to ensure supply
competition and transparency in the awarding of contracts.
2.1.1.2. Cost/benet concern. Emerges where the expenses outweigh
the benets of moving to electronic procurement.
2.1.1.3. External (supply chain/business partners) incompatibi-
lity. Suppliers' strategies and willingness are not sufcient to take
on new technologies such as e-procurement. There is a lack of
readiness by external parties to engage in electronic interactions.
2.1.1.4. High investment cost of IT infrastructure/software. The company
cannot afford too high investment cost on a new IT infrastructure and
software which are necessary for e-procurement adoption.
2.1.1.5. Inadequate business processes to support e-procure-
ment. Specied aims, goals and context of the e-procurement
application are not aligned with other organizational and managerial
choices and not integrated with the organization's processes.
2.1.1.6. Inadequate IT infrastructure of suppliers/business partners. The
external parties of the supply chain do not have adequate IT
infrastructure compatible with the e-procurement system.
2.1.1.7. Incompliance with company culture. Cultural issues such as
absence of a clear corporate policy, lack of a widely accepted
solution, lack of leadership, lack of exible centralized control,
resistance to change, and lack of information quality, produce a
slowdown in adoption of e-procurement.
2.1.1.8. Inter-operability concerns with other systems used. It is a
technical issue related to lack of compatibility and interface with
other internal systems, mainly due to the fact that software
companies have sought to make their product unique. In doing
 P. Tokta-Palut et al. / Int. J. Production Economics 158 (2014) 7790 79

Table 1
Literature review on barriers of e-procurement systems.

Barriers of e-procurement References

Eadie et al. Gunasekaran and Ngai Eadie et al. Farzin and Nezhad Sitar Eei et al. This
(2007) (2008) (2010) (2010) (2011) (2012) study

Bureaucratic disfunctionalities in practice X X X X X

Cost/benet concern X X X X X
External (supply chain/business partners ) incompatibility X X X X X
High investment cost of IT infrastructure/software X X X X X X X
Inadequate business processes to support e-procurement X X
Inadequate IT infrastructure of suppliers/business partners X X X
Incompliance with company culture X X X
Inter-operability concerns with other systems used X X X X X
Lack of adequate technical/IT infrastructure X X X X X X X
Lack of e-procurement knowledge/skilled personnel X X X X X X X
Lack of exibility in process and documentation X X X
Lack of system integration with suppliers/business partners X X X X
Lack of top management support X X X X
Religious objections to internet X X
Resistance to change of internal/external customers in supply chain X X X X X
Security, condentiality, and authentication concerns X X X X X X X
Time needed for the implementation process X
Unsure as to the legal position of e-procurement X X X

Table 2
Literature review on benets of e-procurement systems.

Benets of e-procurement References

Morris Knudsen Panayiotou Wong Eadie Gunasekaran Eadie et al. Farzin and Eei This
et al. (2003) et al. and et al. and (2010) Nezhad et al. study
(2000) (2004) Sloan (2007) Ngai (2008) (2010) (2012)
(2004)

Better management and control of suppliers X X X X X X X

Better utilization of staff X X X X
Compliance with laws and regulations X
Cost savings in overall purchasing process X X X X X X X X X X
Decentralization of power X
Decreased bureaucracy and redundancy X X X X
Easier access to market data and enhanced X X X X X X X
intelligence
Enhanced decision making X X X
Enhanced inventory management X X X X X
Improved communication and collaboration in X X X X X X X
supply chain
Improved supply chain transparency X X X X X X
Increased customer service levels X X
Increased process quality and efciency X X X X
Integrated information sharing X X X
Minimization of process errors X X X
On-line and real-time reporting X X X X X
Quicker response to problems through real-time X X X
information
Reduced administration cost X X X
Reduced paperwork X
Reduction in processing time X X X X X
Simplied and streamlined purchasing process X X X X X X X X
Standardization of process X
Time savings in overall purchasing process X X X X X X X X X X
Wider range of suppliers X

so, they have endeavored to stop migration of data between 2.1.1.10. Lack of e-procurement knowledge/skilled personnel. Mainly
systems. related to personnel issues such as older generations that have not
kept up to the advances in IT related elds, but relying heavily on
traditional forms and means of procurement.

2.1.1.9. Lack of adequate technical/IT infrastructure. Need of

adequate IT infrastructure to carry out e-procurement processes 2.1.1.11. Lack of exibility in process and documentation. Insufcient
and/or technology to operate IT. exibility of e-procurement systems can prevent system use in
80 P. Tokta-Palut et al. / Int. J. Production Economics 158 (2014) 7790
certain circumstances and make exception handling necessary,
and as a result jeopardize the expected benets.
2.1.1.12. Lack of system integration with suppliers/business part-
ners. Difculties in integrating e-systems across rm boundaries
in supply chains if suppliers lack capability.
2.1.1.13. Resistance to change of internal/external customers in supply
chain. Resistance to change is an inevitable aspect of human
nature. Since the system users are generally unwilling to change
their way of working to which they are accustomed, that might be
a strong barrier in adopting e-procurement.
2.1.1.14. Security, condentiality, and authentication concerns. The
e-procurement platform transacts condential procurement data
and is exposed to several security threats. It would bring the
difculties with authentication and security into play.
2.1.2. Benets of e-procurement systems
E-procurement is a systematic way for increasing productivity
and decreasing expenditures for businesses. With e-procurement,
companies can gain some benets such that they improve their
business ability and make their supply chain operations more
observable. The benets of e-procurement systems considered in
this study are given below:
2.1.2.1. Better management and control of suppliers. Provides buyers
to locate the suppliers with the best prices and quality, and
streamline the negotiation and contracting processes through
enhance transparency and communication.
2.1.2.2. Cost savings in overall purchasing process. Increase in speed
and efciency in procurement process and communication,
increased level of inventory-cycle, less need of operational
personnel, savings on operating costs.
2.1.2.3. Decreased bureaucracy and redundancy. E-procurement
facilitates and speeds up internal control processes. It also eliminates
inefcient approval procedures and needless duplication of services.
2.1.2.4. Easier access to market data and enhanced intelligence. Enables
to monitor and scan external sources of data and intelligence easily
and to share information with others pro-actively.
2.1.2.5. Enhanced decision making. The decision-making process is
enhanced by e-procurement since relevant information is neatly
organized and time-stamped.
2.1.2.6. Enhanced inventory management. Automated inventory
control and standardized procedures make inventory operations
easier to be managed, adjusted, and upgraded.
2.1.2.7. Improved communication and collaboration in supply chain.
E-procurement allows sections of electronic documentation to ow
through the supply chain. As it improves the speed of returns and
makes it easy to communicate requirements in a quicker and
more accessible manner, it will result in a better understanding of
requirements and due compliance.
2.1.2.8. Improved supply chain transparency. Transparency of
product specications, prices, contract details, such as the
contractual conditions, time, terms of orders, etc., making these
visible to relevant parties both internally and externally.
2.1.2.9. Increased process quality and efciency. Transparency in
system and increased competition among suppliers through
easier communication, intelligence and information access provide
improvements in procurement process quality and efciency.
2.1.2.10. Integrated information sharing. Accelerates the ow of
important information between internal and external business
partners; also provides real-time information sharing within a
broader structure.
2.1.2.11. Minimization of process errors. Eliminating manual pro-
cesses and paperwork reduces the opportunity for human errors,
inaccuracies, and reworks.
2.1.2.12. On-line and real-time reporting. Real-time reporting
system that enables management to have a fast and reliable way
to compare the spending with budget, allowing quick reaction to
problems.
2.1.2.13. Quicker response to problems through real-time informa-
tion. Increased speed in transactions, tracking and reporting
through real-time information helps quicker problem solving
and reactive decisions.
2.1.2.14. Simplied and streamlined purchasing process. E-procurement
solutions simplify the purchasing process by bringing all suppliers
together, accessible from a single e-platform and eliminating the need
for a paper form, while providing streamlined procedures to expedite
order to payment processing.
2.1.2.15. Time savings in overall purchasing process. Simplied and
streamlined process; increase in information speed and transparency
enables shorter time for evaluation and the decision-making process.
2.2. Research hypotheses
This study consists of two main parts. In the rst part, the aim
is to analyze the interactions among the barriers and benets of
e-procurement systems, and to determine the most signicant
ones. Then, the effects of these barriers and benets on the
e-procurement adoption decision are investigated in the second
part of the study. We have two main research hypotheses as
follows:
Hypothesis 1 (H1). Barriers of e-procurement systems have nega-
tive effect on the adoption decision.
Hypothesis 2 (H2). Benets of e-procurement systems have
positive effect on the adoption decision.
In addition to the hypotheses given above, another point of
interest is to understand whether the effects of the barriers or
benets are higher on the e-procurement adoption decision.
Accordingly, we may recommend the company to adopt the
e-procurement system or not.
3. Methodology
3.1. Sampling and data collection
As stated before, we perform an empirical analysis in this study
and consider a retail store chain which operates in book and
stationery sector in Turkey. The senior management wants to
decide whether to switch from traditional procurement methods
to an e-procurement system or not. For this purpose, a detailed
literature review is performed to dene the perceived barriers and
 P. Tokta-Palut et al. / Int. J. Production Economics 158 (2014) 7790 81

benets of e-procurement systems. 20 barriers and 20 benets are
determined as a result of the literature review.
In the second step, eight top managers from the company have
rated these barriers and benets on a ve-point scale. By taking
the geometric mean (Aczl and Roberts, 1989) of their responses
and after a brainstorming session, we get the nal set of barriers
and benets that are rated above three. In the nal set, there are
14 barriers and 15 benets left, which are then entered into the
ISM model.
In the third step, to identify the contextual relationships within
the barriers and benets of e-procurement systems, another
brainstorming session is performed with eight top managers from
the company and four experts from academia. Based on the
outputs of this session, the ISM models for the barriers and
benets of e-procurement systems are developed.
According to the results of the ISM models, 3 barriers (7 benets)
of e-procurement systems are found to be the root causes of the
other barriers (benets). These barriers and benets are then entered
into the SEM model forming the fourth step of the study. In this step,
another questionnaire is prepared to develop the SEM model.
This questionnaire is also prepared on a ve-point scale. Here, the
aim is to investigate the effects of these barriers and benets on
the e-procurement adoption decision and the necessity of the
e-procurement system for the company.
The questionnaire is uploaded to the Intranet of the company.
Among 916 people working in a related eld, 277 respondents
answered the questionnaire, resulting in a response rate of 30.24%.
Among these responses, 21 cases contain missing values, giving a
percentage of 7.58%. On the other hand, regarding the variables, all
variables except one contain missing values ranging between

List the barriers and Literature review Define the research

benefits for e- hypotheses
procurement adoption
process Expert opinion

Test the measurement model using

Establish contextual confirmatory factor analysis
relationship between
variables Check model goodness-of-fit
Check convergent validity
Check discriminant validity
Examine the standardized residual
Develop structural covariances and modification indices
Develop reachability
self-interaction No
matrix No
matrix

Partition the Is the measurement

reachability matrix model validated?
into different levels

Yes
Develop the
reachability matrix
in its conical form Test the structural model

Check model goodness-of-fit

Analyze the path coefficients
Draw a directed and factor loadings
graph

Remove transitivity
from the digraph Is the structural model
validated?

Replace variable
nodes with Yes
relationship
statements
Analyze the effects of
barriers and benefits on e-
procurement adoption
Yes decision
Is there any conceptual
inconsistency?

No

Obtain the base elements

of barriers and benefits
for the e-procurement
adoption process

INTERPRETIVE STRUCTURAL STRUCTURAL EQUATION

MODELING MODELING
Fig. 1. The methodology used in this study (ISM part is adapted from Ravi and Shankar (2005)).
82 P. Tokta-Palut et al. / Int. J. Production Economics 158 (2014) 7790
0.72% and 3.61%. Since sample size is especially important for SEM,
Bayesian imputation (Schafer and Graham, 2002) is performed to
estimate the missing values. Readers are referred to Schafer and
Graham (2002) and Chen and stebro (2003) for the advantages of
Bayesian imputation over other techniques such as listwise dele-
tion and regression imputation.
3.2. Interpretive structural modeling
This study combines two techniques: ISM and SEM. We will
discuss ISM in this subsection, and SEM in the following one. ISM
is found and handled by Wareld (1973) and its roots come from
graph theory. The ISM process transforms unclear, poorly articu-
lated mental models of systems into visible, well-dened models
useful for many purposes (Sage, 1977).
Ravi and Shankar (2005) describe the signicant characteristics
of ISM as follows: (i) This methodology is interpretive as the
judgment of the group decides whether and how different
elements are related; (ii) it is structural as on the basis of the
relationships, an overall structure is extracted from the complex
set of elements; and (iii) it is a modeling technique as the specic
relationships and overall structure are portrayed in a
digraph model.
The steps involved in the ISM methodology are given below
(Ravi and Shankar, 2005; Govindan et al., 2012):
Step 1. Variables affecting the system under consideration are
listed.
Step 2. A contextual relationship is established among variables
with respect to which pairs of variables would be examined.
Step 3. A Structural Self-Interaction Matrix (SSIM) is developed
for variables, which indicates pairwise relationships among
variables of the system under consideration.
Step 4. Reachability matrix is developed from the SSIM and the
matrix is checked for transitivity. The transitivity of the con-
textual relation is a basic assumption made in ISM. It states that
if variable A is related to B, and B is related to C, then A is
necessarily related to C.
Step 5. The reachability matrix obtained in Step 4 is partitioned
into different levels.
Step 6. A digraph is drawn and the transitive links are removed
based on the relationships given in the reachability matrix.
Step 7. The resultant digraph is converted into an ISM model,
by replacing variable nodes with statements.
Step 8. The ISM model developed in Step 7 is reviewed to check
for conceptual inconsistency and necessary modications
are made.
In this study, after applying the ISM methodology to the
barriers and benets of e-procurement systems, the barriers and
benets that form the base of the hierarchies are then entered into
the structural equation model to investigate their effects on the
e-procurement adoption decision. The next subsection describes
the SEM methodology.
3.3. Structural equation modeling
SEM is a multivariate technique that enables the researcher to
simultaneously examine a series of interrelated dependence rela-
tionships among the measured variables and latent constructs as
well as between the latter ones (Hair et al., 2010). This study uses
Anderson and Gerbing's (1988) two-step approach, in which the
measurement model is estimated prior to the structural model.
The AMOS 16.0 software is used to test the measurement and
structural models based on the maximum likelihood estimation
method.
We test the measurement model using the conrmatory factor
analysis (CFA) as described in Hair et al. (2010). We investigate the
model 2 and its signicance rst. As Jreskog and Srbom (1993)
propose, we also examine the 2 with the number of degrees of
freedom, i.e., 2 =df . Additionally, we calculate other goodness-of-
t measures. Based on the recommendations of Hu and Bentler
(1998), we choose standardized root mean square (SRMR),
goodness-of-t index (GFI), adjusted goodness-of-t index (AGFI),
normed t index (NFI), comparative t index (CFI), and root mean
square error of approximation (RMSEA).
After evaluating the t of the measurement model, we assess
the convergent validity and discriminant validity of the constructs.
We test the convergent validity by examining (i) the factor
loadings; (ii) the average variance extracted (AVE) for each latent
construct; and (iii) construct reliability (CR). To test the discrimi-
nant validity, the AVE values for the constructs are compared with
the square of the correlation estimate between these two con-
structs (Fornell and Larcker, 1981). After assessing the convergent
and discriminant validities, we examine the standardized residual
covariances and modication indices to identify the problems in
the measurement model.
Once the measurement model is validated, we test our struc-
tural model where our main focus is to test the hypothesized
relationships. The goodness-of-t of the structural model is
evaluated with the same measures that we use to test the
measurement model. After evaluating the goodness-of-t mea-
sures for the structural model, we nally analyze the path
coefcients and loading estimates. The methodology used in this
study is summarized in Fig. 1.
4. Data analysis and results
4.1. Interpretive structural modeling
Recall that the rst part of this study aims at identifying the
relationships within the barriers and benets of e-procurement
systems. As explained in Section 3.1, 14 barriers and 15 benets are
determined and they are entered into the ISM model.
4.1.1. Structural self-interaction matrix (SSIM)
ISM methodology suggests using expert opinions based on
various management techniques such as brainstorming, nominal
technique, etc., in developing the contextual relationships among
the variables (Ravi and Shankar, 2005). Accordingly, to dene
the contextual relationships among the barriers and benets of
e-procurement systems, a brainstorming session is performed
with eight top managers from the company and four experts from
academia. Similar to the previous studies (e.g., Ravi and Shankar,
2005; Govindan et al., 2012), four symbols are used to denote the
direction of the relationship between barriers (benets) i and j:
V Barrier (benet) i will help to alleviate (achieve) barrier
(Benet) j;
A Barrier (benet) j will be alleviated (achieved) by barrier
(Benet) i;
X Barriers (benets) i and j will help to achieve each other;
O Barriers (benets) i and j are unrelated.
The structural self-interaction matrices for barriers and
benets of e-procurement systems are given in Tables A1 and
A2, respectively. All tables for the ISM methodology can be found
in Appendix A.
 P. Tokta-Palut et al. / Int. J. Production Economics 158 (2014) 7790 83

4.1.2. Initial reachability matrix
Once the SSIM is constructed, it is transformed into a binary
matrix named as the initial reachability matrix. The rules for
transforming SSIM into the initial reachability matrix are given in
Table 3. For instance, if the element (i, j) of the SSIM is V, then the
element (i, j) of the initial reachability matrix becomes 1 and the
element (j, i) becomes 0. The initial reachability matrices for
barriers and benets of e-procurement systems are given in
Tables A3 and A4, respectively.
4.1.3. Final reachability matrix
After developing the initial reachability matrix, this matrix is
checked for transitivity resulting in the nal reachability matrix.
Recall that transitivity is a basic assumption in ISM stating that if a
variable A is related to B, and B is related to C, then A is necessarily
related to C. The nal reachability matrices for barriers and
benets of e-procurement systems can be found in Tables A5
and A6, respectively.
4.1.4. Level partitions
The next step in ISM is to dene the reachability and ante-
cedent sets for each variable from the nal reachability matrix. The
reachability set consists of the variable itself and the other
Table 3
Rules for transforming SSIM into the initial reachability matrix.
Structural self-interaction matrix Initial reachability matrix
Element (i, j) Element (i, j)
V 1
A 0
X 1
O 0
variables which it may help achieve, whereas the antecedent set
consists of the variable itself and the other variables which may
help in achieving it. Thereafter, the intersection of these sets is
derived for all the variables. The variables for which the reach-
ability and intersection sets are the same constitute the top level
in the ISM hierarchy. When a variable is assigned to a level, it is
discarded from the variable set and the process is repeated for the
remaining variables. The process continues until all variables are
assigned to a level. The identied levels are then used to construct
the digraph and the nal ISM model. The level partitions
for barriers and benets of e-procurement systems are given in
Tables A7A12 and A13A16, respectively.
4.1.5. ISM model
After dening the levels, the structural model is constructed
from the nal reachability matrix, resulting in a digraph. Then, the
transitive links are removed based on the relationships given in
the nal reachability matrix, leading to the ISM model. The ISM
models for barriers and benets of e-procurement systems are
given in Figs. 2 and 3, respectively.
Fig. 2 denotes that inadequate IT infrastructure of suppliers/
business partners is a very signicant barrier for e-procurement
systems since it lies at the bottom level of the hierarchy. This
barrier then leads to the lack of e-procurement knowledge/skilled
personnel and lack of adequate technical/IT infrastructure. These
three barriers constitute the bottom two levels of the hierarchy,
meaning that these are strong drivers and may be treated as the
Element (j, i) root causes of the remaining barriers. Therefore, top management
should initiate a comprehensive strategic plan to improve the
0 technical/IT infrastructure of the company and the whole supply
1
chain. This strategy should also be supported by procurement of
1
0 skilled personnel and by training programs to improve the technical
expertise needed for e-procurement adoption. Overcoming these

11- Resistance to
change of internal/
external customers
in supply chain

2- Security, 4- High investment 7- Inadequate

1- Incompliance
confidentiality and cost of IT business processes 13- Cost/benefit
with company
authentication infrastructure/ to support e- concern
culture
concerns software procurement

12- Inter-operability 14- Lack of

concerns with other flexibility in process
systems used and documentation

6- External (Supply 10- Lack of system

3- Bureaucratic
chain/Business integration with
disfunctionalities in
partners) suppliers/business
practice
incompatibility partners

5- Lack of e-
8- Lack of adequate
procurement
technical/IT
knowledge/skilled
infrastructure
personnel

9- Inadequate IT
infrastructure of
suppliers/business
partners

Fig. 2. ISM model for the barriers of e-procurement systems.

84 P. Tokta-Palut et al. / Int. J. Production Economics 158 (2014) 7790

6- Cost savings in
overall purchasing
process

5- Better 9- Enhanced 11- Decreased 14- Time savings in 15- Increased

2- Minimization of 10- Enhanced
management and inventory bureaucracy and overall purchasing process quality and
process errors decision making
control of suppliers management redundancy process efficiency

3- Quicker response 13- Improved

1- Easier access to 8- Simplified and
to problems through 4- On-line and real- 7- Improved supply communication and
market data and streamlined
real-time time reporting chain transparency collaboration in
enhanced intelligence purchasing process
information supply chain

12- Integrated
information sharing

Fig. 3. ISM model for the benets of e-procurement systems.

Table 4 4.2. Measurement model results

Indicators of the latent variables in the measurement and structural models.
In this study, Anderson and Gerbing's (1988) two-step
Latent Indicator Description
variable approach is used, i.e., the measurement model is estimated prior
to the structural model. The AMOS 16.0 software is used to test the
Barriers R1 5. Lack of e-procurement knowledge/skilled measurement and structural models based on the maximum
personnel likelihood estimation method. Although the data does not t to
R2 8. Lack of adequate technical/IT infrastructure
R3 9. Inadequate IT infrastructure of suppliers/business
normal distribution very well (multivariate kurtosis51.29; criti-
partners cal ratio25.24), there are several studies in the literature (e.g., Hu
et al., 1992; Olsson et al., 2000) that shows the maximum like-
Benets N1 1. Easier access to market data and enhanced
intelligence lihood estimation even in the case of nonnormality outper-
N2 3. Quicker response to problems through real-time forms other estimation methods such as the generalized least
information squares or asymptotically distribution free methods (Schermelleh-
N3 4. On-line and real-time reporting Engel et al., 2003).
N4 7. Improved supply chain transparency
N5 8. Simplied and streamlined purchasing process
The measurement model consists of two latent variables, i.e.,
N6 12. Integrated information sharing Barriers and Benets, with three and seven indicators, respec-
N7 13. Improved communication and collaboration in tively. We test the measurement model using the conrmatory
supply chain factor analysis (CFA) method as described in Hair et al. (2010).
Here, we present the results for the nal measurement model. The
model 2 is calculated as 65.66 with 31 degrees of freedom and it
is signicant. However, because of the shortcomings of the 2 test
barriers will help to reduce the negative effects of all the other statistic, too much emphasis should not be given to this test
barriers up to internal/external resistance to change. (Schermelleh-Engel et al., 2003). Jreskog and Srbom (1993)
The ISM model regarding the benets of e-procurement sys- propose to compare the magnitude of 2 with the number of
tems indicates that the most signicant benet of e-procurement degrees of freedom, i.e., 2 =df . Our measurement model yields
systems is integrated information sharing. Therefore, for the 2 =df 2:12, which is below the threshold of 3.00. Additionally,
success of the whole project, enhancement and well-integration we calculate other goodness-of-t measures. Based on the recom-
in information sharing within the whole supply chain is the key mendations of Hu and Bentler (1998), we choose standardized root
factor, which has a boosting effect on the other benets. As one mean square (SRMR), goodness-of-t index (GFI), adjusted
can see from Fig. 3, integrated information sharing leads to easier goodness-of-t index (AGFI), normed t index (NFI), comparative
access to market data and enhanced intelligence; quicker response t index (CFI), and root mean square error of approximation
to problems through real-time information; on-line and real time (RMSEA). The recommended values for these t indices and the
reporting; improved supply chain transparency; simplied and results for the measurement model are given in Table 5, which
streamlined purchasing process; improved communication and shows that all values are within the recommended ranges indicat-
collaboration in supply chain; and nally cost savings in overall ing that the measurement model has a good t.
purchasing process that constitutes the top level of the hierarchy. After evaluating the t of the measurement model, one should
According to the results of the ISM models (see Figs. 2 and 3), also assess the construct validity that consists of two components:
3 barriers (7 benets) of e-procurement systems constitute the convergent validity and discriminant validity.
bottom two levels of the hierarchy and they are found to be the We test the convergent validity by examining the factor
root causes of the other barriers (benets). These barriers and loadings rst. All factor loadings are signicant at the 0.001 level
benets are then entered into the SEM model and they become the and except two of them all are above 0.70. Only R1 and R3 have
indicators of the latent variables as given in Table 4. factor loadings of 0.65 and 0.67, respectively; but since these are
 P. Tokta-Palut et al. / Int. J. Production Economics 158 (2014) 7790 85

Table 5 4.3. Structural model results and hypothesis testing

Goodness-of-t measures for the measurement model.
After testing the measurement model, we test our structural
Goodness-of-t measure Recommended value Result
(Schermelleh-Engel et al. (2003)) model and our main focus is to test the hypothesized relation-
ships. The structural model consists of two exogenous variables
2 =df r 3.00 2.118 (Barriers and Benets) and one endogenous variable (Adoption).
SRMR r 0.10 0.052 Recall that the exogenous variables Barriers and Benets have
GFI Z 0.90 0.957
three and seven indicators, respectively. On the other hand, the
AGFI Z 0.85 0.923
NFI Z 0.90 0.957 endogenous variable Adoption has a single indicator, which is
CFI Z 0.95 0.977 measured through the question that investigates the necessity of
RMSEA r 0.08 0.064 the e-procurement system for the company.
The goodness-of-t of the structural model is evaluated with
the same measures used to test the measurement model. The
model 2 is calculated as 71.62 with 39 degrees of freedom and its
Table 6 p value is 0.001. The normed 2 is 2 =df 1:84. The goodness-of-
Results for the measurement model. t measures for the structural model are given in Table 7. The
results denote that all the measures are within desirable ranges
Latent variable Indicator Factor loading AVE CR
and the structural model ts the data well.
Barriers 0.490 0.742
After evaluating the goodness-of-t measures for the structural
R1 0.654n model, we turn our attention to the path coefcients and loading
R2 0.774n estimates, which are given in Fig. 4. When we compare the loading
R3 0.665n estimates of the measurement and structural models, we can see
Benets 0.624 0.924 that they are nearly the same and the maximum difference is
N1 0.743n 0.007. This is an evidence of stability among the measured
N2 0.790n
indicator variables and supports the validity of the measurement
N3 0.748n
N4 0.809n
N5 0.865n
N6 0.756n Table 7
N7 0.810n Goodness-of-t measures for the structural model.
n
Signicant at the 0.001 level (two-tailed). Goodness-of-t measure Recommended value Result
(Schermelleh-Engel et al. (2003))

above 0.50, they are also acceptable. Second, the average variance 2 =df r 3.00 1.836
extracted (AVE) is calculated for each latent construct. AVE values SRMR r 0.10 0.050
are calculated as 0.49 and 0.62 for barriers and benets, respec- GFI Z 0.90 0.956
AGFI Z 0.85 0.926
tively. Since an AVE of 0.50 or higher suggests adequate conver-
NFI Z 0.90 0.955
gent validity, one can see that the AVE for barriers is just below CFI Z 0.95 0.979
this critical level. Third, we calculate the construct reliability (CR), RMSEA r 0.08 0.055
which is also an indicator of convergent validity. CR values are
calculated as 0.74 and 0.92 for barriers and benets, respectively.
Since they both exceed the threshold of 0.70, this also validates
convergent validity. The results for the measurement model can be
found in Table 6. R1 0.655
To test the discriminant validity, the AVE values for the
constructs are compared with the square of the correlation 0.775
R2 Barriers
estimate between these two constructs (Fornell and Larcker,
1981). The correlation estimate between Barriers and Benets is R3 0.663
H1: 0.154*
calculated as 0.30. Since AVE values for both constructs are greater
than the squared correlation estimate, this result provides good
evidence of discriminant validity.
N1 0.303
After assessing the convergent and discriminant validities, we
Adoption
examine the standardized residual covariances and modication
N2 0.785 0.747
indices to identify the problems in the measurement model.
Standardized residual covariances should be less than |2.50| for a
N3 0.741 H2: 0.484**
correct model. The results denote that only the standardized
residual covariance between R2 and N4 is outside this range and 0.807
N4 Benefits
it is calculated as  2.68. However, it is below another threshold of 0.869
|4.00|, where existence of values above |4.00| suggests a potentially
N5 0.755
unacceptable degree of error (Hair et al., 2010). Therefore, the
standardized residual covariances indicate no problem in the
N6 0.811
measurement model. On the other hand, we also examine the
modication indices. High modication indices suggest that the t
N7
could be improved signicantly by freeing the corresponding path
to be estimated. Since the modication indices of the measure- *
p-value = 0.023 **
p-value < 0.001
ment model (available upon request) are reasonable, this result
also provides good evidence of model validation. Fig. 4. Results of the path analysis.
86 P. Tokta-Palut et al. / Int. J. Production Economics 158 (2014) 7790

model (Hair et al., 2010). From Fig. 4, it can be seen that the path
coefcient estimate between barriers and adoption is  0.15 and it
is signicant at the 0.05 level. Thus, our rst hypothesis is
supported, i.e., barriers of e-procurement systems have negative
effect on the adoption decision. On the other hand, the path
coefcient estimate between benets and adoption is 0.48 and it is
signicant at the 0.001 level. Thus, our second hypothesis is also
supported, i.e., benets of e-procurement systems have positive
effect on the adoption decision. As stated before, besides these
hypotheses, another point of interest is to examine whether the
barriers or benets of e-procurement systems have higher impact
on the adoption decision. Since the path coefcient estimate of
benets is higher than that of the barriers, we conclude that
benets of an e-procurement system overweigh its barriers. Thus,
it would be benecial for the company to adopt the e-procurement
system. However, it is also noteworthy that the squared multiple
correlations (i.e., R2) of Adoption is calculated as 0.21. In other
words, benets and barriers explain 21% of the variance of
Adoption. Therefore, adding other latent constructs that may affect
the e-procurement adoption decision would increase the variance
explained.
5. Conclusion
This study investigates the effects of the barriers and benets of
e-procurement systems on the e-procurement adoption decision.
An empirical analysis has been performed for a retail store chain
which operates in book and stationery sector in Turkey.
The rst aim of the study is to determine the best route for the
company to follow at their transition period from traditional
methods of purchasing to e-procurement adoption. This is
achieved by measuring the driving power of each potential barrier
and benet as well as the inter-relations within them. The second
and the ultimate aim of the study is to put forth whether barriers
or benets have higher impact on the company's e-procurement
adoption decision by using an approach based on cost/benet
analysis.
The results of the rst part of the study denote that among the
barriers of e-procurement systems, inadequate IT infrastructure of
suppliers/business partners is the most important barrier in the
adoption of e-procurement systems. This barrier then leads to two
other important barriers, which are lack of adequate technical/IT
infrastructure and lack of e-procurement knowledge/skilled
personnel. Thus, in tackling the barriers of e-procurement systems,
the management of the company should focus more on
IT/technical infrastructure of the supply chain, and skills,
knowledge and capability of human resources for adopting e-
procurement. Overcoming these barriers will help to reduce the
negative effects of all other barriers up to internal/external
resistance to change.
When we turn back to the benets of e-procurement systems,
integrated information sharing forms the base of the hierarchy and
it is found as the most signicant benet. Therefore, enhancement
and well-integration in information sharing within the supply
chain is the key factor for the success of e-procurement adoption.
An improvement in integrated information sharing helps to
achieve all other benets, which are easier access to market data
and enhanced intelligence; quicker response to problems through
real-time information; on-line and real time reporting; improved
supply chain transparency; simplied and streamlined purchasing
process; improved communication and collaboration in supply
chain; and nally cost savings in overall purchasing process.
The results of the second part of the study denote that barriers
(benets) of e-procurement systems have negative (positive) effect
on the e-procurement adoption decision and the effect of benets
on the adoption decision is higher than that of the barriers, which
might be considered as indirect costs of adoption. In other words,
the potential benets of e-procurement systems have stronger
impacts than the costs that the company may put up with due to
the potential barriers. Thus, based on the results of our cost/
benet analysis, it would be benecial for the company to adopt
the e-procurement system.
It is noteworthy that the results of this study might differ for
other companies and sectors. Thus, as a further study, the models
could be enhanced with other factors and a sector-base analysis
could be performed from a larger perspective. Besides, the method
proposed in this paper could be considered as an instrument and a
guide for change management for the companies. Accordingly,
companies could apply this method to help their transition
decisions from traditional systems to e-systems.
Acknowledgments
The authors are grateful to the Editor and the referee for their
valuable comments and suggestions.
Appendix A. Tables for the ISM methodology
See Tables A1-A16.

Table A1
Structural self-interaction matrix for barriers of e-procurement systems.

Barriers 14 13 12 11 10 9 8 7 6 5 4 3 2 1

1. Incompliance with company culture O O O V O O O A O O O V O

2. Security, condentiality, and authentication concerns V O A V A O O A A O O O
3. Bureaucratic disfunctionalities in practice O O O V O O O V O O O
4. High investment cost of IT infrastructure/software O X O V O A A O A A
5. Lack of e-procurement knowledge/skilled personnel V O O V O O O V V
6. External (Supply chain/Business partners) incompatibility V O V V X A O O
7. Inadequate business processes to support e-procurement X O X V O O O
8. Lack of adequate technical/IT infrastructure V O V V V A
9. Inadequate IT infrastructure of suppliers/business partners V O V V V
10. Lack of system integration with suppliers/business partners V O A V
11. Resistance to change of internal/external customers in supply chain A A A
12. Inter-operability concerns with other systems used V O
13. Cost/benet concern O
14. Lack of exibility in process and documentation
 P. Tokta-Palut et al. / Int. J. Production Economics 158 (2014) 7790 87

Table A2
Structural self-interaction matrix for benets of e-procurement systems.

Benets 15 14 13 12 11 10 9 8 7 6 5 4 3 2 1

1. Easier access to market data and enhanced intelligence V V X A A V V A A V X A X O

2. Minimization of process errors X V A A O O V A A V O O X
3. Quicker response to problems through real-time information V V X A X V V O A V V X
4. On-line and real-time reporting V V X X X V V V X V V
5. Better management and control of suppliers X V X A A X V A X V
6. Cost savings in overall purchasing process A O A A A A A A A
7. Improved supply chain transparency V V X X V V V V
8. Simplied and streamlined purchasing process V V A O X V O
9. Enhanced inventory management X V A A A X
10. Enhanced decision making X V A A X
11. Decreased bureaucracy and redundancy X V A A
12. Integrated information sharing V V X
13. Improved communication and collaboration in supply chain V V
14. Time savings in overall purchasing process X
15. Increased process quality and efciency

Table A3 Table A5
Initial reachability matrix for barriers of e-procurement systems. Final reachability matrix for barriers of e-procurement systems.

Barriers 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Barriers 1 2 3 4 5 6 7 8 9 10 11 12 13 14

1 1 0 1 0 0 0 0 0 0 0 1 0 0 0 1 1 0 1 0 0 0 1a 0 0 0 1 0 0 0
2 0 1 0 0 0 0 0 0 0 0 1 0 0 1 2 0 1 0 0 0 0 1a 0 0 0 1 0 0 1
3 0 0 1 0 0 0 1 0 0 0 1 0 0 0 3 1a 1a 1 0 0 0 1 0 0 0 1 1a 0 1a
4 0 0 0 1 0 0 0 0 0 0 1 0 1 0 4 0 0 0 1 0 0 0 0 0 0 1 0 1 0
5 0 0 0 1 1 1 1 0 0 0 1 0 0 1 5 1a 1a 0 1 1 1 1 0 0 1a 1 1a 1a 1
6 0 1 0 1 0 1 0 0 0 1 1 1 0 1 6 0 1 0 1 0 1 1a 0 0 1 1 1 1a 1
7 1 1 0 0 0 0 1 0 0 0 1 1 0 1 7 1 1 1a 0 0 0 1 0 0 1a 1 1 0 1
8 0 0 0 1 0 0 0 1 0 1 1 1 0 1 8 0 1a 0 1 0 1a 1a 1 0 1 1 1 1a 1
9 0 0 0 1 0 1 0 1 1 1 1 1 0 1 9 0 1a 0 1 0 1 1a 1 1 1 1 1 1a 1
10 0 1 0 0 0 1 0 0 0 1 1 0 0 1 10 0 1 0 1a 0 1 1a 0 0 1 1 1a 0 1
11 0 0 0 0 0 0 0 0 0 0 1 0 0 0 11 0 0 0 0 0 0 0 0 0 0 1 0 0 0
12 0 1 0 0 0 0 1 0 0 1 1 1 0 1 12 1a 1 0 0 0 1a 1 0 0 1 1 1 0 1
13 0 0 0 1 0 0 0 0 0 0 1 0 1 0 13 0 0 0 1 0 0 0 0 0 0 1 0 1 0
14 0 0 0 0 0 0 1 0 0 0 1 0 0 1 14 1a 1a 0 0 0 0 1 0 0 0 1 1a 0 1

a
Values obtained by incorporating transitivity.

Table A4 Table A6
Initial reachability matrix for benets of e-procurement systems. Final reachability matrix for benets of e-procurement systems.

Benets 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Benets 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

a a a a a a
1 1 0 1 0 1 1 0 0 1 1 0 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
2 0 1 1 0 0 1 0 0 1 0 0 0 0 1 1 2 1a 1 1 1a 1a 1 0 0 1 1a 1a 0 1a 1 1
3 1 1 1 1 1 1 0 0 1 1 1 0 1 1 1 3 1 1 1 1 1 1 1a 1a 1 1 1 1a 1 1 1
4 1 0 1 1 1 1 1 1 1 1 1 1 1 1 1 4 1 1a 1 1 1 1 1 1 1 1 1 1 1 1 1
5 1 0 0 0 1 1 1 0 1 1 0 0 1 1 1 5 1 1a 1a 1a 1 1 1 1a 1 1 1a 1a 1 1 1
6 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 6 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0
7 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 7 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
8 1 1 0 0 1 1 0 1 0 1 1 0 0 1 1 8 1 1 1a 1a 1 1 1a 1 1a 1 1 0 1a 1 1
9 0 0 0 0 0 1 0 0 1 1 0 0 0 1 1 9 0 1a 0 0 1a 1 0 0 1 1 1a 0 0 1 1
10 0 0 0 0 1 1 0 0 1 1 1 0 0 1 1 10 1a 1a 1a 1a 1 1 1a 1a 1 1 1 0 1a 1 1
11 1 0 1 1 1 1 0 1 1 1 1 0 0 1 1 11 1 1a 1 1 1 1 1a 1 1 1 1 1a 1a 1 1
12 1 1 1 1 1 1 1 0 1 1 1 1 1 1 1 12 1 1 1 1 1 1 1 1a 1 1 1 1 1 1 1
13 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 13 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
14 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 14 0 1a 0 0 1a 1a 0 0 1a 1a 1a 0 0 1 1
15 0 1 0 0 1 1 0 0 1 1 1 0 0 1 1 15 1a 1 1a 1a 1 1 1a 1a 1 1 1 0 1a 1 1

a
Values obtained by incorporating transitivity.
88 P. Tokta-Palut et al. / Int. J. Production Economics 158 (2014) 7790

Table A7
Level partitions for barriers of e-procurement systems Iteration 1.

Barrier Reachability set Antecedent set Intersection set Level

1 1, 3, 7, 11 1, 3, 5, 7, 12, 14 1, 3, 7
2 2, 7, 11, 14 2, 3, 5, 6, 7, 8, 9, 10, 12, 14 2, 7, 14
3 1, 2, 3, 7, 11, 12, 14 1, 3, 7 1, 3, 7
4 4, 11, 13 4, 5, 6, 8, 9, 10, 13 4, 13
5 1, 2, 4, 5, 6, 7, 10, 11, 12, 13, 14 5 5
6 2, 4, 6, 7, 10, 11, 12, 13, 14 5, 6, 8, 9, 10, 12 6, 10, 12
7 1, 2, 3, 7, 10, 11, 12, 14 1, 2, 3, 5, 6, 7, 8, 9, 10, 12, 14 1, 2, 3, 7, 10, 12, 14
8 2, 4, 6, 7, 8, 10, 11, 12, 13, 14 8, 9 8
9 2, 4, 6, 7, 8, 9, 10, 11, 12, 13, 14 9 9
10 2, 4, 6, 7, 10, 11, 12, 14 5, 6, 7, 8, 9, 10, 12 6, 7, 10, 12
11 11 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14 11 I
12 1, 2, 6, 7, 10, 11, 12, 14 3, 5, 6, 7, 8, 9, 10, 12, 14 6, 7, 10, 12, 14
13 4, 11, 13 4, 5, 6, 8, 9, 13 4, 13
14 1, 2, 7, 11, 12, 14 2, 3, 5, 6, 7, 8, 9, 10, 12, 14 2, 7, 12, 14

Table A8
Level partitions for barriers of e-procurement systems Iteration 2.

Barrier Reachability set Antecedent set Intersection set Level

1 1, 3, 7 1, 3, 5, 7, 12, 14 1, 3, 7 II
2 2, 7, 14 2, 3, 5, 6, 7, 8, 9, 10, 12, 14 2, 7, 14 II
3 1, 2, 3, 7, 12, 14 1, 3, 7 1, 3, 7
4 4, 13 4, 5, 6, 8, 9, 10, 13 4, 13 II
5 1, 2, 4, 5, 6, 7, 10, 12, 13, 14 5 5
6 2, 4, 6, 7, 10, 12, 13, 14 5, 6, 8, 9, 10, 12 6, 10, 12
7 1, 2, 3, 7, 10, 12, 14 1, 2, 3, 5, 6, 7, 8, 9, 10, 12, 14 1, 2, 3, 7, 10, 12, 14 II
8 2, 4, 6, 7, 8, 10, 12, 13, 14 8, 9 8
9 2, 4, 6, 7, 8, 9, 10, 12, 13, 14 9 9
10 2, 4, 6, 7, 10, 12, 14 5, 6, 7, 8, 9, 10, 12 6, 7, 10, 12
12 1, 2, 6, 7, 10, 12, 14 3, 5, 6, 7, 8, 9, 10, 12, 14 6, 7, 10, 12, 14
13 4, 13 4, 5, 6, 8, 9, 13 4, 13 II
14 1, 2, 7, 12, 14 2, 3, 5, 6, 7, 8, 9, 10, 12, 14 2, 7, 12, 14

Table A9 Table A11

Level partitions for barriers of e-procurement systems Iteration 3. Level partitions for barriers of e-procurement systems Iteration 5.

Barrier Reachability set Antecedent set Intersection set Level Barrier Reachability set Antecedent set Intersection set Level

3 3, 12, 14 3 3 5 5 5 5 V
5 5, 6, 10, 12, 14 5 5 8 8 8, 9 8 V
6 6, 10, 12, 14 5, 6, 8, 9, 10, 12 6, 10, 12 9 8, 9 9 9
8 6, 8, 10, 12, 14 8, 9 8
9 6, 8, 9, 10, 12, 14 9 9
10 6, 10, 12, 14 5, 6, 8, 9, 10, 12 6, 10, 12
12 6, 10, 12, 14 3, 5, 6, 8, 9, 10, 12, 14 6, 10, 12, 14 III
14 12, 14 3, 5, 6, 8, 9, 10, 12, 14 12, 14 III

Table A10 Table A12

Level partitions for barriers of e-procurement systems Iteration 4. Level partitions for barriers of e-procurement systems Iteration 6.

Barrier Reachability set Antecedent set Intersection set Level Barrier Reachability set Antecedent set Intersection set Level

3 3 3 3 IV 9 9 9 9 VI
6 6, 10 5, 6, 8, 9, 10 6, 10 IV
8 6, 8, 10 8, 9 8
9 6, 8, 9, 10 9 9
10 6, 10 5, 6, 8, 9, 10 6, 10 IV
 P. Tokta-Palut et al. / Int. J. Production Economics 158 (2014) 7790 89

Table A13
Level partitions for benets of e-procurement systems Iteration 1.

Benet Reachability set Antecedent set Intersection set Level

1 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 10, 11, 12, 13, 15 1, 2, 3, 4, 5, 7, 8, 10, 11, 12, 13, 15
2 1, 2, 3, 4, 5, 6, 9, 10, 11, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 2, 3, 4, 5, 9, 10, 11, 13, 14, 15
3 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 10, 11, 12, 13, 15 1, 2, 3, 4, 5, 7, 8, 10, 11, 12, 13, 15
4 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 10, 11, 12, 13, 15 1, 2, 3, 4, 5, 7, 8, 10, 11, 12, 13, 15
5 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15
6 6 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 6 I
7 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 3, 4, 5, 7, 8, 10, 11, 12, 13, 15 1, 3, 4, 5, 7, 8, 10, 11, 12, 13, 15
8 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 14, 15 1, 3, 4, 5, 7, 8, 10, 11, 12, 13, 15 1, 3, 4, 5, 7, 8, 10, 11, 13, 15
9 2, 5, 6, 9, 10, 11, 14, 15 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15 2, 5, 9, 10, 11, 14, 15
10 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 13, 14, 15
11 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15
12 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 3, 4, 5, 7, 11, 12, 13 1, 3, 4, 5, 7, 11, 12, 13
13 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 10, 11, 12, 13, 15 1, 2, 3, 4, 5, 7, 8, 10, 11, 12, 13, 15
14 2, 5, 6, 9, 10, 11, 14, 15 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15 2, 5, 9, 10, 11, 14, 15
15 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 13, 14, 15

Table A14
Level partitions for benets of e-procurement systems Iteration 2.

Benet Reachability set Antecedent set Intersection set Level

1 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 10, 11, 12, 13, 15 1, 2, 3, 4, 5, 7, 8, 10, 11, 12, 13, 15
2 1, 2, 3, 4, 5, 9, 10, 11, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 2, 3, 4, 5, 9, 10, 11, 13, 14, 15 II
3 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 10, 11, 12, 13, 15 1, 2, 3, 4, 5, 7, 8, 10, 11, 12, 13, 15
4 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 10, 11, 12, 13, 15 1, 2, 3, 4, 5, 7, 8, 10, 11, 12, 13, 15
5 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15 II
7 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 3, 4, 5, 7, 8, 10, 11, 12, 13, 15 1, 3, 4, 5, 7, 8, 10, 11, 12, 13, 15
8 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 13, 14, 15 1, 3, 4, 5, 7, 8, 10, 11, 12, 13, 15 1, 3, 4, 5, 7, 8, 10, 11, 13, 15
9 2, 5, 9, 10, 11, 14, 15 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15 2, 5, 9, 10, 11, 14, 15 II
10 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 13, 14, 15 II
11 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15 II
12 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 3, 4, 5, 7, 11, 12, 13 1, 3, 4, 5, 7, 11, 12, 13
13 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 10, 11, 12, 13, 15 1, 2, 3, 4, 5, 7, 8, 10, 11, 12, 13, 15
14 2, 5, 9, 10, 11, 14, 15 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15 2, 5, 9, 10, 11, 14, 15 II
15 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 12, 13, 14, 15 1, 2, 3, 4, 5, 7, 8, 9, 10, 11, 13, 14, 15 II

Table A15 Chen, G., stebro, T., 2003. How to deal with missing categorical data: test of a
simple Bayesian method. Organ. Res. Methods 6 (3), 309327.
Level partitions for benets of e-procurement systems Iteration 3.
Croom, S., Brandon-Jones, A., 2007. Impact of e-procurement: experiences from
implementation in the UK public sector. J. Purch. Supply Manag. 13 (4), 294303.
Benet Reachability set Antecedent set Intersection set Level
Davila, A., Gupta, M., Palmer, R., 2003. Moving procurement systems to the
Internet: the adoption and use of e-procurement technology models. Eur.
1 1, 3, 4, 7, 8, 12, 13 1, 3, 4, 7, 8, 12, 13 1, 3, 4, 7, 8, 12, 13 III Manag. J. 21 (1), 1123.
3 1, 3, 4, 7, 8, 12, 13 1, 3, 4, 7, 8, 12, 13 1, 3, 4, 7, 8, 12, 13 III Devaraj, S., Vaidyanathan, G., Mishra, A.N., 2012. Effect of purchase volume
4 1, 3, 4, 7, 8, 12, 13 1, 3, 4, 7, 8, 12, 13 1, 3, 4, 7, 8, 12, 13 III exibility and purchase mix exibility on e-procurement performance: an
7 1, 3, 4, 7, 8, 12, 13 1, 3, 4, 7, 8, 12, 13 1, 3, 4, 7, 8, 12, 13 III analysis of two perspectives. J. Oper. Manag. 30 (78), 509520.
8 1, 3, 4, 7, 8, 13 1, 3, 4, 7, 8, 12, 13 1, 3, 4, 7, 8, 13 III Eadie, R., Perera, S., Heaney, G., 2010. Identication of e-procurement drivers and
12 1, 3, 4, 7, 8, 12, 13 1, 3, 4, 7, 12, 13 1, 3, 4, 7, 12, 13 barriers for UK construction organisations and ranking of these from the
13 1, 3, 4, 7, 8, 12, 13 1, 3, 4, 7, 8, 12, 13 1, 3, 4, 7, 8, 12, 13 III perspective of quantity surveyors. Electron. J. Inf. Technol. Constr. 15, 2343.
Eadie, R., Perera, S., Heaney, G., Carlisle, J., 2007. Drivers and barriers to public
sector e-procurement within Northern Ireland's construction industry. Elec-
tron. J. Inf. Technol. Constr. 12, 103120.
Table A16 Eei, K.S., Husain, W., Mustaffa, N., 2012. Survey on benets and barriers of
Level partitions for benets of e-procurement systems Iteration 4. e-procurement: Malaysian SMEs perspective, Int. J. Adv. Sci. Eng. Inf. Technol.
2 (6), 1419.
Farzin, S., Nezhad, H.T., 2010. E-Procurement, the golden key to optimizing the
Benet Reachability set Antecedent set Intersection set Level
supply chains system. World Acad. Sci. Eng. Technol. 66, 518524.
Fornell, C., Larcker, D.F., 1981. Evaluating structural equation models with unob-
12 12 12 12 IV
servable variables and measurement error. J. Mark. Res. 18 (1), 3950.
Golob, T.F., 2003. Structural equation modeling for travel behavior research. Transp.
Res. Part B 37 (1), 125.
Gorvett, R., Liu, N., 2006. Interpretive structural modeling of interactive risks. In:
References Proceedings of the Enterprise Risk Management Symposium. Society of
Actuaries, Chicago, IL.
Govindan, K., Palaniappan, M., Zhu, Q., Kannan, D., 2012. Analysis of third party
Aczl, J., Roberts, F.S., 1989. On the possible merging functions. Math. Soc. Sci. 17 reverse logistics provider using interpretive structural modeling. Int. J. Prod.
(3), 205243. Econ. 140 (1), 204211.
Aibinu, A.A., Al-Lawati, A.M., 2010. Using PLS-SEM technique to model construction Gunasekaran, A., McGaughey, R.E., Ngai, E.W.T., Rai, B.K., 2009. E-procurement
organizations' willingness to participate in e-bidding. Autom. Constr. 19 (6), adoption in the Southcoast SMEs. Int. J. Prod. Econ. 122 (1), 161175.
714724. Gunasekaran, A., Ngai, E.W.T., 2008. Adoption of e-procurement in Hong Kong: an
Anderson, J.C., Gerbing, D.W., 1988. Structural equation modeling in practice: a empirical research. Int. J. Prod. Econ. 113 (1), 159175.
review and recommended two-step approach. Psychol. Bull. 103 (3), 411423. Hair, J.F., Black, W.C., Babin, B.J., Anderson, R.E., 2010. Multivariate Data Analysis.
Chandramowli, S., Transue, M., Felder, F.A., 2011. Analysis of barriers to develop- Pearson Prentice Hall, New Jersey.
ment in landll communities using interpretive structural modeling. Habitat Hawking, P., Stein, A., Wyld, D.C., Foster, S., 2004. E-procurement: is the ugly
Int. 35 (2), 246253. duckling actually a swan down under? Asia Pac. J. Mark. Logist. 16 (1), 326.
90 P. Tokta-Palut et al. / Int. J. Production Economics 158 (2014) 7790
Ho, C.F., Wu, W.H., Tai, Y.M., 2004. An e-procurement impact model based on supply
chain orientation: an empirical investigation. In: Proceedings of the Fourth
Workshop on Knowledge Economy and Electronic Commerce, pp. 167179.
Hu, L., Bentler, P.M., 1998. Fit indices in covariance structure modeling: sensitivity
to unparameterized model misspecication. Psychol. Methods 3 (4), 424453.
Hu, L., Bentler, P.M., Kano, Y., 1992. Can test statistics in covariance structure
analysis be trusted? Psychol. Bull. 112 (2), 351362.
Jreskog, K.G., Srbom, D., 1993. Structural Equation Modeling with the SIMPLIS
Command Language. Scientic Software, Chicago.
Kline, R.B., 1998. Principles and Practice of Structural Equation Modeling. The
Guilford Press, New York.
Knudsen, D., 2003. Aligning corporate strategy, procurement strategy and
e-procurement tools. Int. J. Phys. Distrib. Logist. Manag. 33 (8), 720734.
Konradt, U., Lckel, L., Ellwart, T., 2012. The role of usability in business-to-business
e-commerce systems: predictors and its impact on user's strain and commer-
cial transactions. Adv. Hum.Comput. Interact. 2012, 111.
Lee, W., Quaddus, M., 2006. Buyersupplier relationship in the adoption of e-
purchasing in the small and medium printing industries in Singapore: an
empirical test using structural equation modelling. In: Proceedings of the Tenth
Pacic Asia Conference on Information Systems, pp. 337345.
Luthra, S., Kumar, V., Kumar, S., Haleem, A., 2011. Barriers to implement green
supply chain management in automobile industry using interpretive structural
modeling technique: an Indian perspective. J. Ind. Eng. Manag. 4 (2), 231257.
Madeja, N., Schoder, D., 2003. The adoption of e-procurement and its impact on
corporate success in electronic business. In: Proceedings of the Third Interna-
tional Conference on Electronic Business. Singapore.
Morris, A., Stahl, A., Herbert, R., 2000. E-Procurement: Streamlining Processes to
Maximize Effectiveness. Luminant Worldwide Corporation, USA.
Nguyen, H.O., 2013. Critical factors in e-business adoption: evidence from Austra-
lian transport and logistics companies. Int. J. Prod. Econ. 146 (1), 300312.
Olsson, U.H., Foss, T., Troye, S.V., Howell, R.D., 2000. The performance of ML, GLS,
and WLS estimation in structural equation modeling under conditions of
misspecication and nonnormality. Struct. Equ. Model. 7 (4), 557595.
Panayiotou, N.A., Gayialis, S.P., Tatsiopoulos, I.P., 2004. An e-procurement system for
governmental purchasing. Int. J. Prod. Econ. 90 (1), 79102.
Ravi, V., Shankar, R., 2005. Analysis of interactions among the barriers of reverse
logistics. Technol. Forecast. Soc. Change 72 (8), 10111029.
Sage, A.P., 1977. Interpretive Structural Modeling: Methodology for Large-Scale
Systems. McGraw-Hill, New York, pp. 91164.
Schafer, J.L., Graham, J.W., 2002. Missing data: our view of the state of the art.
Psychol. Methods 7 (2), 147177.
Schermelleh-Engel, K., Moosbrugger, H., Mller, H., 2003. Evaluating the t of
structural equation models: tests of signicance and descriptive goodness-of-t
measures. Methods Psychol. Res. Online 8 (2), 2374.
Singh, M.D., Kant, R., 2008. Knowledge management barriers: an interpretive
structural modeling approach. Int. J. Manag. Sci. Eng. Manag. 3 (2), 141150.
Sitar, C.P., 2011. The barriers of implementing e-procurement. Studia Universitatis
Vasile Goldis Arad Seria tiine Economice Partea a II-a, 12/2011, pp. 120124.
Talib, F., Rahman, Z., Qureshi, M.N., 2011. An interpretive structural modelling
approach for modelling the practices of total quality management in service
sector. Int. J. Model. Oper. Manag. 1 (3), 223250.
Vaidyanathan, G., Devaraj, S., 2008. The role of quality in e-procurement perfor-
mance: an empirical analysis. J. Oper. Manag. 26 (3), 407425.
Wareld, J.N., 1973. Binary matrices in system modeling. IEEE Trans. Syst. Man.
Cybern. 3 (5), 441449.
Wong, C.H., Sloan, B., 2004. Use of ICT for e-procurement in the UK construction
industry: a survey of SMEs readiness. In: Proceedings of the Twentieth Annual
ARCOM Conference, pp. 620628.