Decision Support Systems 62 (2014) 54–65

Contents lists available at ScienceDirect

Decision Support Systems

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

Governance–knowledge fit and strategic risk taking in supply

chain digitization
Ling Xue
Information Systems and Supply Chain Management, Bryan School of Business and Economics, University of North Carolina at Greensboro, Greensboro, NC 27402, United States

a r t i c l e i n f o a b s t r a c t

Article history: Drawing upon the theoretical lenses of strategic risk taking and governance–knowledge fit, this study examines
Received 19 June 2013 how decision rights and knowledge of the IT unit may support the firm to benefit from strategic risk taking in
Received in revised form 12 January 2014 supply chain digitization. We build a mediated-moderation model to theorize that the fit between decision rights
Accepted 11 March 2014
and knowledge of the IT unit drives the firm to strategically take risk in digitizing its supply chain, and such
Available online 17 March 2014
strategic risk taking enables the firm to realize both operational and strategic benefits. Our empirical study
Keywords:
provides support to the mediated-moderation model. Specifically, we find that when allocated with more
Strategic risk taking decision rights, the IT unit with more business knowledge and knowledge creation capability drives the firm to
Governance–knowledge fit conduct more strategic risk taking in supply chain digitization, which in turn results in operational and strategic
IT governance benefits to the firm. However, we also find that the fit between decision allocation and technical knowledge of
Supply chain digitization the IT unit does not significantly lead to strategic risk taking, suggesting that the effect of governance–knowledge
fit on strategic risk taking is contingent on knowledge type.
© 2014 Elsevier B.V. All rights reserved.

1. Introduction In this research, we specifically focus on how allocating more decision

Both practitioners and academic scholars have recognized the signif-
icant risk associated with information technology (IT) investments and
initiatives, especially in the inter-firm environment [9,22]. For example,
research has suggested that more than 70% of supply chain manage-
ment projects have failed to deliver any return on investment [20]. A
review of industry reports on CRM failure rates from 2001 to 2009 sug-
gests that over this time period, the average failure rate of CRM projects
was about 46.25% [30]. The academic studies on IT risk have recognized
the risk associated with the return of overall IT investments at the firm
level [13,56], and have identified various types of risk at different levels
of IT projects and applications [25,31,37].
Due to the risk–return association of IT, organizations are likely to
strategically take risk in adopting IT to pursue higher returns. In the
existing literature, however, many key issues such as what drives the
organization's strategic risk taking in IT and what influences the out-
comes of strategic risk taking have been under-investigated. Moreover,
since the risk decisions on IT cannot be made and executed without the
IT unit, it is critical to consider the influence of the IT unit on the deci-
sions and outcomes of strategic risk taking. The objective of this study
is to address these issues from a perspective of IT governance. A central
element of IT governance concerns the allocation of IT decision rights
between the IT unit and decision makers from other areas [49,57,65].
E-mail address: l_xue@uncg.edu.
rights to the IT unit influences the decisions and outcomes of the
organization's strategic risk taking.
In relating IT governance with strategic risk taking, our research
framework is built primarily upon two theoretical lenses. The first one
is the knowledge perspective of strategic risk taking [51,53,63], which
suggests that the knowledge and experience of decision makers lead
to strategic risk taking. When decision makers have more knowledge
about the problem domain of decision, it is more likely for them to
better understand the nature and consequences of risk [51] and more
rationally undertake risk-taking initiatives [8]. The second theoretical
lens is the perspective of governance–knowledge fit [23,57]. From this
perspective, the decision rights assigned to an IT unit (i.e., the gover-
nance configuration) should be aligned with the knowledge level of
this IT unit. Assigning appropriate decision rights allows the IT unit to
better leverage its knowledge in decision making. For some time,
these two theoretical perspectives exist, in parallel, in separate streams
of literature. However, in a very real sense, these perspectives are po-
tentially complementary to each other. When the IT unit with domain
knowledge is assigned appropriately with decision authorities, it may
also leverage such knowledge to help the organization strategically
take risk in adopting IT. The present research, therefore, constructs an
integrated framework to align these two theoretical lenses with each
other, and explicitly study how governance–knowledge fit at the IT
unit influences the organization's strategic risk taking in IT adoption
and the performance of this strategic risk taking.
To study strategic risk taking in IT, we use supply chain digitization
(refer to SC digitization hereafter) as the research context. SC digitization

http://dx.doi.org/10.1016/j.dss.2014.03.003
0167-9236/© 2014 Elsevier B.V. All rights reserved.
 L. Xue / Decision Support Systems 62 (2014) 54–65
refers to the adoption of inter-organizational systems (IOS) by business
organizations to collaborate and transact with their external partners
(e.g., key suppliers and customers) along their value/supply chains [4,
44,45]. Compared to the adoption of other intra-organizational IT sys-
tems, SC digitization is often considered as more risky [31], primarily
because the external parties and environment that organizations need
to interact with are beyond their controls [46]. In this regard, the deci-
sions on SC digitization reflect organizations' strategic risk taking in IT.
Moreover, in our empirical analysis, we focus on SC digitization by a
sample of Chinese companies. Compared to companies in developed
countries (such as the US), Chinese companies have to bear more risks
in SC digitization due to the underdeveloped macro-level network infra-
structure, the lack of IT expertise in individual companies, the turbulent
business environment of the emerging economy, and the relatively im-
mature legal and regulatory systems [15]. Therefore, the study on SC
digitization by Chinese companies provides an opportunity to better un-
derstand the strategic risk taking in IT.
In this study, we conceptualize strategic risk taking as the extent to
which organizations aggressively adopt electronic supply chain systems
to transact and interact with external partners even when they recog-
nize the associated risk. When examining governance–knowledge fit,
we consider different types of knowledge that the IT unit may possess,
including technical knowledge, business knowledge, and knowledge
generation capability. Although different types of knowledge are often
complementary and utilized together in risk taking, the IT unit is less
likely to possess the entire set of expertise [57]. Therefore, the differen-
tiation between different knowledge allows us to examine different
types of fit between the IT unit's decision rights and its knowledge,
and develop more insights about different impacts of governance–
knowledge fit.
Our analysis suggests that the allocation of more decision rights to
the IT unit is associated with more strategic risk taking in SC digitization.
In addition, with more decision rights, the IT unit's business knowledge
and knowledge generation capability also lead to more strategic risk
taking. The implication is that the fit between the IT unit's decision
rights and its business knowledge and knowledge generation capability
fosters strategic risk taking of the organization. However, our results
also suggest that the decision right allocation does not significantly
strengthen the relationship between the IT unit's technical knowledge
and strategic risk taking. Rather, higher technical knowledge is directly
associated with more strategic risk taking.
Organizations often pursue both operational benefits and strate-
gic benefits in SC digitization [28,39,45,55]. We therefore consider
these two types of benefits in examining the performance of strate-
gic risk taking. Our empirical findings suggest that strategic risk tak-
ing leads to both types of benefits, and its impact on strategic
benefits is even higher than that on operational benefits. More im-
portantly, our analysis illustrates that strategic risk taking mediates
the impact of governance–knowledge fit on the performance of SC
digitization. The results suggest that strategic risk taking mediates
the impact of fit between the IT unit's business knowledge and its de-
cision rights, and the impact of fit between the IT unit's knowledge
generation capability and its decision rights. However, risk taking
does not mediate the impact of fit between the IT unit's technical
knowledge and its decision rights. Therefore, the mediating effect
of strategic risk taking on knowledge–governance fit is dependent
on the type of knowledge.
The study contributes to multiple streams of literature. First, the
study draws upon both the risk taking literature and the IT gover-
nance literature to explain how decision allocation influences stra-
tegic risk taking and the risk–return relationship in IT initiatives.
Our theoretical perspective aligns the view from the risk taking lit-
erature (e.g., [3]) on how knowledge encourages risk taking with
the view from the IT governance literature (e.g., [57]) on how gov-
ernance knowledge fit enhances IT performance. Second, the study
adds to the literature on project-level IT risk (e.g., [25,26]) by
55
illustrating that organizations can benefit from strategic risk taking
in IT initiatives, and they can enhance their benefits through
assigning appropriate decision rights to their knowledgeable IT
units. Third, the results also help better explain the risk–return as-
sociation that prior studies have identified at the firm-level (e.g.,
[12,13]).
The rest of the paper is organized as follows. Section 2 presents the
theory and hypothesis development. Section 3 introduces the data
collection and the methodology. Section 4 presents the results.
Section 5 discusses the results and Section 6 concludes the paper.
2. Theory and hypothesis development
2.1. Theoretical basis: knowledge and risk taking
The existing theories on risk taking suggest that knowledge plays an
important role in motivating the strategic risk taking of organizational
decision makers [3]. Sitkin and Pablo [51] conceptualize that strategic
risk taking is influenced jointly by the risk perception and risk propen-
sity of organizational decision makers. Decision makers' risk perception
is dependent on their familiarity with the problem domain. The knowl-
edge that decision makers developed from their past experience makes
decision makers more familiar with the problem domain and thus helps
mitigate their risk perception. In this way, knowledge is likely to foster
strategic decisions on risk taking. This logic is consistent with the
perspective of the behavioral agency model of risk taking [33,63],
which argues that past successful experience mitigates risk perception
and fosters risk taking by reducing the uncertainty regarding the likely
outcomes of risk-taking actions.
More importantly, knowledge allows decision makers to better
understand risk and motivates them to proactively take the risk
they identify. The theory of Sitkin and Pablo [51] suggests that past
successful experience of decision makers leads to risk propensity,
because the knowledge that decision makers accumulate from the
past success is likely to make risk seem reasonable. Similarly, the
theory on reasoned risk taking [8] suggests that when decision
makers learn from their past experience and possess sufficient
knowledge about the decision domain, they are also more likely to
convert their risk-taking incentives to the actual risk actions. Based
on these views, knowledge is likely to foster strategic risk taking by
enabling decision makers to more rationally bear risk to pursue
higher returns associated with risk.
Besides the existing knowledge of decision makers, researchers have
also taken a more dynamic perspective to consider how knowledge cre-
ation capability influences risk taking (e.g., [53]). Knowledge creation
capability refers to the extent to which decision makers and knowledge
workers have access to knowledge and information from other stake-
holders and are capable of combining and exchanging information to
generate new knowledge [29,42]. With such capability, organizations
are able to develop creative ideas and solutions that are needed in han-
dling the uncertainties in implementing new technologies. The existing
literature suggests that knowledge creation capability motivates organi-
zations to take risk in innovative activities, such as new product devel-
opment [53].
2.2. Theoretical basis: governance–knowledge fit
A central issue in the governance of IT is the allocation of decision
rights [49,62], especially that between the IT unit and other functional
departments [57]. The strategic decisions of SC digitization are often
the outcomes of group decision making. In this case, the decision rights
of the IT unit mainly reflect the extent to which IT managers are includ-
ed in the strategic decision team and are involved in the decision
making processes of SC digitization.
In decision allocation, the fit between the knowledge and the
decision authority of decision makers is often considered critical to the
56 L. Xue / Decision Support Systems 62 (2014) 54–65
effectiveness of decision making. Jensen and Meckling [23] suggest that
when knowledge is valuable in decision making but is difficult to trans-
fer within the organization, co-locating decision rights and knowledge
through delegating authority increases the utilization of knowledge. Sim-
ilarly, the knowledge-view of firm [18,54] considers knowledge as a spe-
cial resource that resides in individuals, and co-locating decision rights
and knowledge is an internal coordination mechanism for the firm to le-
verage this resource. In IS research, Tiwana [57] confirms, in the context
of software development, that the fit between decision rights and the
knowledge of the IT unit and client departments improves the
effectiveness of decision making.
2.3. Governance–knowledge fit and strategic risk taking
Based on the aforementioned view of governance–knowledge fit, we
expect that the fit between the IT unit's knowledge and its decision
rights contributes to the strategic risk taking in SC digitization. Knowl-
edge–governance fit reflects the notion of moderation-based fit as in
Venkatraman [60]. That is, when more decision rights are allocated to
the IT unit, the IT unit can better contribute its knowledge in decision
making for SC digitization. The knowledge contribution by the IT unit
is therefore likely to lead to more strategic risk taking in SC digitization.
In other words, decision rights of the IT unit generate a positive moder-
ating effect on the relationship between the IT unit's knowledge and
strategic risk taking.
We consider both existing knowledge and knowledge creation
capability of the IT unit. Regarding existing knowledge, we consider
both technical knowledge and business knowledge. Technical
knowledge refers to the knowledge about the technical aspects of
digital supply chain systems, and the know-how on implementation
and operation of these systems. Business knowledge refers to the
knowledge about the internal and external business policies, rules
and processes of the organization, and the know-how on employing
IT systems to digitize the intra- and inter-organizational business
operations. Both technical knowledge and business knowledge are
needed in the strategic decision making.
When the IT unit is assigned with more decision rights, it brings
more technical knowledge into the strategic decision team. IT managers
have more chances to use their technical expertise in making decisions.
In addition, IT managers also provide other decision makers the re-
quired technical knowledge about the digital supply chain systems.
Such technical knowledge makes the whole decision team more famil-
iar with the technical aspects and potential technical challenges of SC
digitization. Decision makers are therefore more confident and more
willing to undertake risky projects to achieve SC digitization.
The case of Haier, a China-based global manufacturer of consumer
appliances, provides an illustrative example [61]. In early stages of
organizational development, the major role of Haier's IT department
was to implement various functional systems to support the operations
in different functional areas (e.g., marketing, logistics and finance). The
technical knowledge that the IT department accumulated in this stage,
however, allowed the IT department to better support the organiza-
tion's more innovative initiatives such as ERP implementation. Later,
the IT department had changed from its initial supporting function to
a strategic decision maker. The CIO also became one of the core decision
makers in the board room when the CEO proposed the “1000-day infor-
mation revolution” in 2007 [61]. The technical knowledge provided by
the IT department in strategic decision making processes enabled
Haier to take the challenge of adopting more innovative IT in the inter-
organizational environment. Eventually Haier successfully built its
“Global Value System”, an ERP-based interorganizational platform, to
link with external partners and realize its strategic goal of global
expansion. Therefore, we develop the following hypothesis regarding
how decision right allocation may moderate the relationship between
technical knowledge of the IT unit and strategic risk taking in SC
digitization.
H1a. When more decision rights are allocated to the IT unit, there is a
greater positive association between the technical knowledge of IT
unit and strategic risk taking in SC digitization.
Business knowledge is usually not the primary knowledge of the
IT unit due to the IT unit's technical specialization. However, in SC
digitization, organizations need to adapt their business processes
and corporate strategies to align with the use of IT. Therefore,
technical knowledge and business knowledge are often mutually
complementary and need to be utilized together in the strategic
decision making for SC digitization [14]. When the IT unit is
involved in the strategic decision making, it should possess a rela-
tively higher level of business knowledge to ensure the effective-
ness of decision making and resolve the knowledge-distancing
issue [23,57]. Allocating more decision rights to the IT unit with
higher business knowledge allows IT managers to better use their
business knowledge in decision making for strategic risk taking.
The possession of business knowledge makes IT managers more
familiar with the potential managerial challenges and issues caused
by SC digitization. Such familiarity makes IT managers less
concerned about handling various uncertainties in adopting new
technologies [51] and thus makes them more willing to undertake
risky initiatives.
The possession of business knowledge by IT managers also enables
IT managers to better communicate with other decision makers the de-
cision making processes. When decision makers from other functional
areas are not very familiar with the technical aspects of SC digitization,
IT managers can frame the decision problems in more informative
business terms. This way of problem framing helps other decision
makers better understand the nature of risk and the potential future re-
turn associated with the risk. It also makes the technical challenges of SC
digitization look more reasonable to other decision makers and im-
proves their willingness to take the risk [51,63]. In this regard, involving
IT managers with more business knowledge in the strategic decision
making is also likely to support and encourage risk taking by other de-
cision makers. We therefore expect that the allocation of more decision
rights to the IT unit strengthens the relationship between the IT unit's
business knowledge and the organization's strategic risk taking in SC
digitization.
H1b. When more decision rights are allocated to the IT unit, there is a
greater positive association between the business knowledge of IT unit
and strategic risk taking in SC digitization.
In addition to the existing knowledge base, knowledge creation ca-
pability of the IT unit is also important for the strategic risk taking in
SC digitization. Knowledge creation capability refers to the extent to
which the IT unit has access to other stakeholders, and is capable of ex-
changing and combing existing information and ideas into new knowl-
edge [42,53]. SC digitization is often a multistage complex process in
which the organization needs to adjust its subsequent development
and adoption based on the learning from earlier tests, pilot projects
and experiments [68]. With knowledge creation capability, the IT unit
can continually generate new knowledge and enhance the organiza-
tion's capability to handle many problems and challenges that are not
expected in advance. The organization is therefore more confident
about the outcomes of risk taking and more willing to undertake risky
initiatives.
The allocation of more decision rights to the IT unit enables the IT
unit to better leverage its knowledge creation capability. When involved
in the strategic decision making process, IT managers have more oppor-
tunities to communicate with other internal and external stakeholders,
such as other functional managers and outside partners. More critical
information about supply chain operations and collaboration is also
likely to be shared with the IT unit. The IT unit can use its knowledge
creation capability to combine more ideas and information from differ-
ent stakeholders and create more new knowledge to handle future
 L. Xue / Decision Support Systems 62 (2014) 54–65
uncertainties. Moreover, the involvement of the IT unit in decision mak-
ing also makes other decision makers more informed about the capabil-
ity of the IT unit in knowledge creation. Through more communication
and idea exchanging, other functional managers should better know
how the IT unit is capable of learning from the past and developing so-
lutions for future challenges in SC digitization. Consequently, the orga-
nization should become overall more willing to strategically take risk.
An illustrative example is the Campbell Soup Company, a 100-year-
old, $8 billion global manufacturer of packaged food [48,62]. In Project
Harmony, Campbell's initiative to standardize business processes of its
global supply chain, the company staged its SAP implementation at its
twenty-two business units. IT managers were actively involved in the
sponsoring and operating committees and effectively planned rollouts
based on learning acquired from prior implementation. The experience
was also shared with other business leaders to improve the readiness
for subsequent rollouts. For example, the rollout at the company's
world headquarters and Texas plant benefited from the experience at
the Canadian site. Such knowledge creation capability allowed the
company to overcome many challenges in deploying SAP regionally
and ultimately globally. Therefore, we develop the following hypothesis
regarding how decision right allocation may moderate the relationship
between the knowledge creation capability of the IT unit and strategic
risk taking in SC digitization.
H1c. When more decision rights are allocated to the IT unit, there is a
greater positive association between the knowledge creation capability
of the IT unit and strategic risk taking in SC digitization.
2.4. Performance of strategic risk taking in SC digitization
Organizations often pursue both operational benefits and strategic
benefits in SC digitization [39,55]. Operational benefits refer to the
improved efficiency of business processes through the use of electronic
inter-organizational systems, including reduced production and
transaction costs, lowered inventory turnover, leaner logistic steps,
automating manufacturing, and faster billing and payment settlement.
Strategic benefits refer to the improved business relationships with ex-
ternal partners (e.g., customers and suppliers) and more opportunities
for business expansion.
Organizations face various risks when pursuing operational and
strategic benefits in SC digitization. For example, since SC digitization
is a long process, the uncertainty in technological changes may drive
up the overall cost of implementation [31]. Technical risk exists also be-
cause organizations have to connect with their partners' systems which
are beyond their control [34,47]. In addition to technological risk, there
are also business risks for SC digitization, such as managerial complexity
for organizations to adjust their business processes [7,66,68] and the
transactional risk that the other parties may appropriate the benefits
of SC digitization [17].
The existing literature has argued that organizations may strategi-
cally take risk to pursue higher return of IT investment [13,56]. For ex-
ample, Dewan et al. [13] use option theory to explain how firms
realize returns from IT investment through risk taking. The idea is that
IT investment is essentially considered by decision makers as a real op-
tion. When decision makers expect that future returns exceed the risk
premium of IT investment, they will exercise the option (and make IT
investment) to take the risk so as to realize future returns. In this regard,
risk taking is a key mechanism through which firms pursue higher fu-
ture returns. This is consistent with the logic of the prospect theory
[24], which characterizes risk taking as a way to realize decision makers'
aspired-to returns.
When it comes to SC digitization, risk taking is also a key mechanism
through which firms realize the expected operational and strategic
benefits. SC digitization usually requires firms to change their existing
business processes in order to improve operational efficiency. The
57
changes on business processes are associated with risk as firms are
often uncertain about all the potential consequences. Moreover, SC dig-
itization is usually a long process that requires continuous learning [68].
Therefore, in early stages, firms are not likely to have the experience to
handle all the emerging problems. If firms never take risk, even though
they rationally expect the operational benefits of SC digitization, they
may have to forgo the opportunities of realizing these benefits. Like-
wise, in SC digitization, even though firms can identify strategic benefits
of SC digitization, they still have to cope with the technological uncer-
tainty and managerial complexity that not all other firms can handle.
If firms choose not to take risk and experiment innovative initiatives
that other competitors have not tried, they also lose the opportunities
of leveraging digitally-enabled supply chains to outperform their
competitors and reap strategic benefits of SC digitization. Therefore,
we expect that when decision makers anticipate operational and strate-
gic benefits of SC digitization, strategic risk taking is an important way
for them to realize these expected benefits. We develop the following
hypotheses regarding the performance of strategic risk taking in SC
digitization.
H2a. With the expected operational benefits of SC digitization
controlled, strategic risk-taking is positively associated with the realized
operational benefits of SC digitization;
H2b. With the expected of strategic benefits of SC digitization
controlled, strategic risk-taking is positively associated with the realized
strategic benefits of SC digitization.
2.5. The mediating role of strategic risk taking
Organizational performance often depends more on an ability to
turn knowledge into effective actions and less on knowledge itself [1].
When colocation of decision rights with knowledge allows the IT unit
to better leverage its knowledge and knowledge creation capability,
the performance of SC digitization should be enhanced. Our main inter-
est, however, is in the role that strategic risk taking plays in this process.
In previous sections, we theorize how the governance–knowledge fit of
the IT unit encourages strategic risk taking of the organization, and how
strategic risk taking enhances the realized benefits of SC digitization.
Taking these two theoretical lenses in conjecture, we consider the
potential mediating effect of strategic risk taking. With the mediating
effect of strategic risk taking, our theoretical framework is characterized
as a mediated moderation model [41]. That is, the fit (or, the complemen-
tarity) between decision right allocation and the IT unit's knowledge in
improving the benefits of SC digitization is mediated by strategic risk
taking.
The mediating role of strategic risk taking can be justified from the
perspective of bounded rationality [50]. The central idea of bounded
rationality is that although decision makers are intendedly rational,
they are so constrained by their cognitive capabilities and incomplete
information. As a consequence, decision makers cannot guarantee that
all consequences are carefully considered and have to bear risk in
their decisions [38]. Experienced decision makers often rely on their
knowledge to identify potential benefits from risk taking and justify
the opportunities of risk taking [8]. Without risk taking actions, organi-
zational decision makers have to forgo many opportunities that poten-
tially benefit the organization. This is also true in the case of IT adoption.
Because IT is inherently risky, organizations that are less comfortable
with risk taking will be less likely to use innovative technologies [35].
Consequently, these organizations will be less likely to benefit from in-
novative technologies such as SC digitization. In this regard, strategic
risk taking is a key mechanism through which the organization realizes
the benefits of SC digitization.
Governance–knowledge fit is therefore likely to enhance the perfor-
mance of SC digitization through strategic risk taking. The realization of
operational benefits from SC digitization requires changes on the
58 L. Xue / Decision Support Systems 62 (2014) 54–65
existing business processes [19]. Similarly, the realization of strategic
benefits requires the development of new technology-based solutions
[59]. Such changes and new development are often associated with un-
certainty (i.e., risk). Allocating more decision rights to the IT unit to fit its
knowledge base and knowledge creation capability allows IT managers
to better utilize their knowledge to identify the opportunities of changes
and new development. However, without the risk-taking actions to
initiate changes and new development, the organization will never con-
vert these opportunities to actual benefits. Strategic risk taking enables
the organization to implement these new initiatives and actions in the
environment with uncertainties. In this sense, strategic risk taking is
also an important way for the organization to realize the value of gover-
nance–knowledge fit in its IT unit. Therefore, we develop the following
three hypotheses regarding how strategic risk taking mediates the
impact of governance–knowledge fit on the benefits of SC digitization.
H3a. The fit between the IT unit's technological knowledge and IT-DR in
improving the operational and strategic benefits of SC digitization is
mediated by strategic risk taking;
H3b. The fit between the IT unit's business knowledge and IT-DR in
improving the operational and strategic benefits of SC digitization is
mediated by strategic risk taking;
H3c. The fit between the IT unit's knowledge generation capability and
IT-DR in improving the operational and strategic benefits of SC digitiza-
tion is mediated by strategic risk taking.
Fig. 1 presents the theoretical framework of this study.
3. Data and methodology
3.1. Data
The data used to test the theoretical framework was collected using
a large-sample survey. We developed the survey questionnaire based
on a comprehensive literature review. The questionnaire items were
reviewed by a panel of both academic scholars and practitioners. We
tailored the survey questions to fit with the context of SC digitization,
and verified content validity using a pretest of ten randomly selected
firms. The firms used in the pretest were similar to the entire sample
in terms of the average size and the average revenue. We used a Chinese
consulting company to conduct three rounds of data collection (to im-
prove response rate). Responses were collected using both telephone
interviews and online surveys. The key respondents were top execu-
tives or directors responsible for leading the projects of SC digitization.
We initially contacted a total of 5128 firms. The final sample for anal-
ysis contains 384 firms that provided completed survey response,
resulting in a response rate of 7.49%. The sample covers a variety of
Strategic Risk
Technical
Knowledge H1a
Business H1b
Knowledge
Knowledge H1c
Creation
Capability
Decision Rights
of IT Unit
Fig. 1. Theoretical framework.
industries, with 49.2% of the firms in construction, mining, and
manufacturing industries, and 50.8% in service-oriented industries.
Geographically, the sample covers 31 of the total 34 provinces in
China. 37.2% of the firms have annual revenue below 1.5 million US
dollars, 35.2% between 1.5 and 15 million, and 27.6% above 15 million.
Regarding the demographic background of these respondents (who
led SC digitization projects in organizations), 18% of them were IT
managers and the rest were senior general managers or functional
managers. 73% of the respondents were between 35 and 44 in age,
12% were under 34, 13% were between 45 and 55, and 2% were older
than 55. The average working experience is 16.4 years. We used t-tests
to assess the non-response bias. Specifically, we compared between the
firms with complete questionnaires and those without, and did not
find any statistically significant difference between the two groups in
average size, revenue and IT budget. We also compared between early
responses (the first 25% of the sample in terms of response time) and
late responses (the last 25% of the sample) [2]. Again, we did not find
any significant difference between these two groups of firms in average
size, revenue and IT budget.
3.2. Measures
All key theoretical constructs, except the moderator Strategic Risk
Taking (SRT), were measured using multi-item, 7-point Likert scales,
with “1” representing “low degree/strongly disagree” and “7”
representing “high degree/strongly agree”. As explained below, the
moderator SRT was constructed based upon the other two measures,
Supply Chain Digitization and Perceived Risk, following an approach in
the strategy literature [33]. Table 1 presents the measurement details
of the key constructs in this study. Table 2 presents the summary statistics
of and correlation between key variables used in the analysis.
3.2.1. Supply Chain Digitization (SCD)
The measure of SCD captures the extent to which firms implement
supply chain management systems to electronically connect and con-
duct transactions with their key suppliers and customers in supply
chains. We adapted three items from the literature [4] to measure
firms' electronic connection with customers using digital supply chain
systems, and another three corresponding items to measure electronic
connection at the supply-side. Prior research on electronic supply
chain management has used similar items [11,40].
3.2.2. Perceived Risk (Risk)
The measure of perceived risk reflects typical risks that organization-
al decision makers feel in SC digitization, including technological risk,
management complexity, implementation risk and transactional risk.
We adopted four measurement items used in prior research on the
adoption of inter-organizational systems [43,68]. Specifically, we
Controls
Taking
Realized Expected
H3a Operational Operational
H2a
Benefits Benefits
H3b
Employee;
H3c H2b Realized Revenue;
Strategic Years; Industry
Benefits
Expected
Controls: Expected Strategic
Benefits; IT Budget; Benefits
Propensity; Industry
 L. Xue / Decision Support Systems 62 (2014) 54–65 59

Table 1
Survey measurement items.

Construct Measurement items Literature

SC Digitization (SCD)
Perceived Risk (Risk)
IT Unit's Decision Right
(ITDR)
Risk Propensity (Propen)
Technical Knowledge (TK)
Business Knowledge (BK)
Knowledge Creation
Capability (KCC)
Expected Operational
Benefit (EOB)
Expected Strategic Benefit
(ESB)
Realized Operational Benefit
(ROB)
Realized strategic benefit
(RSB)
What are the proportions of the following items that the company transacts with/conduct through Mukhopadhyay et al. [40];
digital SC systems: (1) number of suppliers; (2) transaction volume with suppliers; (3) transaction Barua et al. [4]; Dedrick et al. [11]
activities with suppliers; (4) number of customers; (5) transaction volume with customers;
(6) transaction activities with customers. (Cronbach alpha = 0.96)
Our decision makers feel that SC digitization: (1) is a risky project; (2) may potentially generate Nicolaou and McKnight [43]
negative impact on business processes; (3) may enable our suppliers or customers to take advantage
on us; (4) is associated with technological uncertainty. (Cronbach alpha = 0.93)
The IT unit has the power to decide on the following aspects of SC digitization: (1) long-term Tiwana [57]; Tiwana and
strategies and goals; (2) system functionality; (3) budget; (4) management policies; (5) project Konsynski [58]
priorities; (6) performance evaluation; (7) time scheduling. (Cronbach alpha = 0.94)
Regarding SC digitization, our decision makers are willing to: (1) try unfamiliar new technology; Sitkin and Weingart [52];
(2) invest aggressively in emerging applications; (3) take the challenge of adapting the business/ Nicolaou and McKnight [43]
organization to the new technology implementation. (Cronbach alpha = 0.91)
Our IT unit/staff understands the following technical requirements for SC digitization: (1) hardware Tiwana [57]
requirements; (2) software requirements; (3) personnel requirements; (4) reliability and security
requirements. (Cronbach alpha = 0.90)
Our IT unit/staff understands: (1) the business operation of other departments; (2) the strategic Tiwana [57]
plan of the company; (3) the business processes of the company; (4) the business rules of the
company. (Cronbach alpha = 0.94)
Our IT unit: (1) spends time in studying the application of new IT in business; (2) is capable to quickly Smith et al. [53]
learn and employ new IT; (3) encourages IT employees to try new IT in business; (4) is familiar with
the external developers/vendors of our digital SC system; (5) encourages IT employees to communicate
with other employees. (Cronbach alpha = 0.92)
How do you evaluate the expected benefits of SC digitization in: (1) lowering logistic cost; (2) lowering Mukhopadhyay and Kekre [39];
operating cost; (3) improving inventory control; (4) increasing inventory turnover. (Cronbach Subramani [55]
alpha = 0.91)
How do you evaluate the expected benefits of SC digitization in: (1) identifying business opportunities; Mukhopadhyay and Kekre [39];
(2) expanding the market; (3) improving customer relationships; (4) improving supplier relationships; Subramani [55]
(5) establishing competitive advantage. (Cronbach alpha = 0.90)
How do you evaluate the realized benefits of SC digitization in: (the same items as those for the Mukhopadhyay and Kekre [39];
expected operational benefit). (Cronbach alpha = 0.89) Subramani [55]
How do you evaluate the realized benefits of SC digitization in: (the same items as those for the expected Mukhopadhyay and Kekre [39];
strategic benefit). (Cronbach alpha = 0.88) Subramani [55]

included one item in Nicolaou and McKnight [43] (item 1) to capture
the general perceived risk about the adoption of supply chain manage-
ment system. We also used three items in Zhu et al. [68] to capture the
perceived managerial, transactional and technological risks.
3.2.3. Strategic Risk Taking (SRT)
Using the abovementioned constructs of SCD and Risk, we construct-
ed the variable of SRT as SRT = SCD × Risk, following an approach used
in Larraza-Kintana et al. [33]. The key idea here is that if decision makers
pursue SC digitization (i.e., higher SCD) even when they perceive a
higher level of risk for SC digitization (i.e., higher Risk), it means that
they are strategically taking more risk (i.e., higher SRT). If, on the
other hand, decision makers pursue SC digitization (i.e., higher SCD)
only when they perceive a lower level of risk (i.e., lower Risk), it
means that they are not strategically taking significant risk (i.e., lower
SRT). Similarly, if decision makers do not pursue SC digitization (i.e.,
lower SCD) even when they perceive a higher level of risk for SC
digitization (i.e., higher Risk), it also means that they are not strategically
taking risk (i.e., lower SRT).
3.2.4. Expected/Realized Operational benefits (EOB/ROB)
The measure of EOB captures the extent to which organizational
decision makers expect that SC digitization will help improve the oper-
ational efficiency of the firm. Likewise, the measure of ROB captures the
extent to which organizational decision makers believe that SC digitiza-
tion actually help firms improve operational efficiency. Following the
existing research [39,55], we adopted four measurement items to
measure key aspects of operational efficiency, including logistic cost,
operating cost, inventory control and turnover. The same items were
used to measure both the expected and the realized operational
benefits.
3.2.5. Expected/Realized Strategic Benefits (ESB/RSB)
The measure of ESB captures the extent to which decision makers
expect that SC digitization will help generate long-term competitive
advantage and business expansion opportunities. Likewise, the measure
of ROB captures the extent to which organizational decision makers be-
lieve that SC digitization actually help generate competitive advantage
and business expansion opportunities [39,55]. We adopted five mea-
surement items for strategic benefits. The same items were used to
measure both the expected and the realized strategic benefits.
3.2.6. Decision right of the IT unit (ITDR)
The decision right of the IT unit is defined in this study as the degree
to which the IT unit takes responsibilities and has authorities in the key
decision making for SC digitization. We developed seven measurement
items based on the measures of decision control rights and the
measures of IT specification and implementation rights used in the
existing literature [57,58].
3.2.7. Technical knowledge of the IT unit (TK)
In this study, technical knowledge is defined as knowledge about
technical requirements on developing and maintaining electronic
supply chain systems used in SC digitization. We used four items to
capture knowledge about the technical requirements on hardware,
software, personnel, and system security. These items were adapted
from similar measurement items used in the prior studies (e.g., [57]).
3.2.8. Business knowledge of the IT unit (BK)
Following prior studies [10,57], we define the business knowledge of
the IT unit as knowledge about the overall strategic goals, business
policies, and operational processes of the organization. We adopted
four existing measurement items used in the literature [57].
60 L. Xue / Decision Support Systems 62 (2014) 54–65

Table 2
Descriptive statistics.

Variables Correlation matrix

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

1. ITDR 1.00
2. SRT 0.23 1.00
3. TK 0.39 0.31 1.00
4. BK 0.36 0.25 0.50 1.00
5. KCC 0.47 0.30 0.56 0.72 1.00
6. Propen 0.56 0.10 0.24 0.20 0.32 1.00
7. EOB 0.41 0.20 0.37 0.32 0.41 0.27 1.00
8. ESB 0.45 0.22 0.42 0.37 0.46 0.33 0.85 1.00
9. SOB 0.31 0.34 0.39 0.34 0.43 0.14 0.55 0.51 1.00
10. SSB 0.21 0.36 0.40 0.28 0.41 0.08 0.41 0.48 0.79 1.00
11. ITBgt 0.14 −0.20 −0.05 0.00 −0.05 0.25 −0.01 −0.02 −0.07 −0.05 1.00
12. EMP 0.12 0.13 0.17 0.08 0.14 0.14 0.19 0.17 0.15 0.09 0.42 1.00
13. Rev 0.04 0.16 0.17 0.10 0.11 0.03 0.15 0.10 0.15 0.07 −0.13 0.73 1.00
14. Year 0.19 0.24 0.30 0.13 0.21 0.12 0.19 0.19 0.21 0.18 −0.30 0.41 0.43 1.00
Mean 4.75 19.50 4.65 4.57 4.67 4.48 5.32 5.30 5.04 4.95 9.25 5.77 17.94 2.10
Std. 1.44 9.49 1.59 1.50 1.43 1.49 1.40 1.30 1.53 1.54 2.22 1.59 1.71 0.94

Note: Pearson correlation is reported for construct correlation; N = 384; |p| above 0.14 is significant at the 0.01 level; |p| above 0.10 is significant at the 0.05 level.

3.2.9. Knowledge generation capability of the IT unit (KG)
In this study, knowledge creation capability is defined as the extent
to which the IT unit has access to other stakeholders and is capable of
combining information and knowledge into new knowledge. We
developed four measurement items based on the idea of measuring
knowledge creation capability in the literature [53].
We also included some commonly used control variables in the
analysis. First, we included the general risk propensity (Propen) of the
strategic decision group as a control variable because risk theories [51]
suggest that the risk propensity of decision makers affects their risk-
taking decisions. We measured risk propensity using three items
adopted from the literature [52] and adapted them to fit the context
of SC digitization. Similar measurement items have also been used in
prior research [26,27] to assess the willingness to undertake risky IT ini-
tiatives. Second, in examining strategic risk taking in SC digitization, we
included the overall IT budget (nature logarithm) (ITBgt) to control for
the firm's IT scale. In examining the benefits of SC digitization, we also
included the logarithm of employee number (EMP) and the logarithm
of firm revenue (Rev) to control for the size and operational scale of
the firm. Third, in examining the benefits of SC digitization, we included
the number of years (Year) the firm has initiated SC digitization to
control for the potential difference in firm experience. Fourth, we also
included a set of dummy variables to control the industry fixed effect.
3.3. Convergent validity, discriminant validity, and common method bias
Eleven constructs in our study are measured using multiple items.
Appendix 1 shows the results of an exploratory factor analysis (EFA).
In EFA, we extract eleven factors from the response of survey items.
The eleven factors explained 76.87% of the total variance. We used
confirmatory factor analysis (CFA) to verify the convergent and discrim-
inant validity of the multi-item constructs. All items have high loadings
on the constructs they were designed to capture and low loadings
across other constructs, suggesting convergent and discriminant validi-
ty. As an overall test of convergence validity, we also used Bentler's CFI
of the measurement model to estimate the percentage of variation
explained by a proposed model relative to a model of complete inde-
pendence [6]. The CFI of our eleven-factor model is 0.921, which is
above the threshold 0.9 and suggests a good model fitting [6].
To assess the potential issue of common method bias (CMB) in the
single-contact survey, we conducted two analyses. First, we used
Harman's single-factor test to check if there existed a single factor that
accounted for a large proportion of the variance [21]. In our case, no
such single factor emerged. Second, we utilized Lindell and Whitney's
[36] approach of marker variable test. We used the dummy of company
location and the dummy of company type (i.e., 1 for state-owned; 2 for
private; and 3 for foreign-invested) as marker variables. These market
variables are theoretically uncorrelated with other constructs. The aver-
age correlation between the study's key constructs and the location
dummy (r = −0.024, T = −0.623), and the type dummy (r = 0.018,
T = 0.587) was low and non-significant, suggesting no evidence of
CMB. These results indicate that CMB is not likely to be a serious concern
in our case.
3.4. Analysis
We employed the approach of multivariate regression (MVR) to test
our hypotheses. The use of this approach allows us to follow the existing
procedures in the literature [16,41] to test the mediated moderation
relationships and address potential econometric issues (e.g., the corre-
lation between operational benefits and strategic benefits). The use of
the MVR approach on survey data has also been well adopted in the
literature (e.g., [53]). The analysis includes two main steps. In the first
step, we used a single-equation regression model to examine the impact
of the IT unit's decision rights and knowledge on strategic risk taking in
SC digitization, and estimate this model using the approach of ordinary
least-square (OLS). In the second step, we examined how strategic risk
taking influences the operational and strategic benefits from SC digitiza-
tion. We specify a system of two equations with operational benefit and
strategic benefit as the dependent variables, and estimate using the
approach of seemingly unrelated regression (SUR). SUR takes into
account the potential correlation between the error terms of the two
equations [64]. Our results, as shown below, confirm that significant
cross-equation correlation exists, which suggests the appropriateness
of using SUR instead of the traditional OLS estimation.
Regarding the first-step analysis on strategic risk taking, we specify
the following model,
SRTi ¼ β0 þ β1 ITDRi þ β2 TKi þ β3 BKi þ β4 KGi þ β5 ITDRi  TKi þ β6 ITDRi  BKi
þ β7 ITDRi  KGi þ β8 EOBi þ β9 ESBi þ ½Controls þ εi : ð1Þ
In Eq. (1), the dependent variable is strategic risk taking (SRT). Inde-
pendent variables include key explanatory variables such as decision
rights of the IT unit (ITDR), the IT unit's technical knowledge (TK), busi-
ness knowledge (BK), and knowledge generation capability (KG). Three
interaction terms are used to examine how ITDR moderates the impact
of three types of knowledge on strategic risk taking. Expected opera-
tional benefits (EOB) and strategic benefits (ESB) are included as inde-
pendent variables because expected benefit is a key driver of strategic
risk taking [24] as well as SC digitization [67]. We conducted several
 L. Xue / Decision Support Systems 62 (2014) 54–65
diagnostic checks. The potential issue of multicollinearity was checked
by examining the variance inflation factor (VIF) values of all indepen-
dent variables. All VIF values are below 5, suggesting that there is no se-
rious multicollinearity issue [3232]. The potential issue of
heteroskedasticity was checked using the Breusch–Pagan test and no
heteroskedasticity was found. We also used normality tests using the
Shapiro–Wilk test and the result suggested that regression residuals fol-
low a normal distribution.
For the second-step model on the benefits of SC digitization, we
specify a system of two equations,
ROBi ¼ α 0 þ α 1 EOBi þ α 2 RTi þ α 3 ITDRi þ α 4 TKi þ α 5 BKi þ α 6 KGi
þα 7 ITDRi  TKi þ α 8 ITDRi  BKi þ α 9 ITDRi  KGi þ ½Controls þ τi
ð2Þ
RSBi ¼ χ 0 þ χ 1 ESBi þ χ 2 RTi þ χ 3 ITDRi þ χ 4 TKi þ χ 5 BKi þ χ 6 KGi
þχ 7 ITDRi  TKi þ χ 8 ITDRi  BKi þ χ 9 ITDRi  KGi þ ½Controls þ υi :
ð3Þ
Eq. (2) (Eq. (3)) examines how realized operational benefits
(realized strategic benefits) are influenced by strategic risk taking
(RT), ITDR and knowledge. EOB and ESB are also included as controls.
RT and interaction terms are also used to test the mediated moderation
hypotheses. To take into consideration the potential cross-equation
correlation, we used seemingly unrelated regression (SUR) to estimate
this system of equations. The high cross-equation correlation indicated
in the result table (Table 4) confirmed the appropriateness of using SUR.
4. Results
The regression results of Eq. (1) are presented in Table 3. Table 3 in-
cludes the results of both benchmark models without all the interaction
terms and the full model with all interaction terms. In the benchmark
model 1, the coefficient of ITDR is positive and significant (β1 = 1.062;
T = 2.726; p b 0.01), suggesting that ITDR is positively associated with
strategic risk taking. In the full model, the coefficient of ITDR is also pos-
itive and significant (β1 = 1.055; T = 2.775; p b 0.01). These results
suggest that when allocated with more decision rights, the IT unit moti-
vates the organization to be more risk taking in SC digitization. Table 3
shows that some interaction terms are significant, which means that
ITDR moderates certain influence of the IT unit's knowledge on strategic
risk taking. Considering the full model, the coefficient of the interaction
between ITDR and business knowledge is positive and significant (β6 =
1.003; T = 4.911; p b 0.001). Similarly, the coefficient of the interaction
between ITDR and knowledge generation capability is positive and sig-
nificant (β7 = 0.605; T = 2.442; p b 0.05). These results suggest that
the positive relationship between ITDR and risk taking is stronger
when the IT unit has more business knowledge, and when the IT unit
has more knowledge generation capability. Therefore, H1b and H1c
are supported. However, the coefficient of the interaction between
ITDR and technical knowledge is not significant (β5 = − 0.012; T =
− 0.083; p N 0.1). These results suggest that the fit between ITDR and
the IT unit's technical knowledge does not significantly increase strate-
gic risk taking. Therefore, H1a is not supported. The results also indicate
that the coefficients of ESB are positive and significant, but those of EOB,
although positive, are not significant. This suggests that expected strate-
gic benefits are more important in driving strategic risk taking.
Fig. 2 illustrates how ITDR moderates the relationship between busi-
ness knowledge and strategic risk taking and the relationship between
knowledge generation capability and strategic risk taking. As Fig. 2
shows, at a high level of ITDR (one standard deviation above the
mean), business knowledge and knowledge generation capability
have higher marginal impact on strategic risk taking than when ITDR
61
is at the intermediate level (the mean) or a low level (one standard
deviation below the mean).
Table 4 presents the results of the SUR analysis. To illustrate the
mediating effect of strategic risk taking, Table 4 also includes the results
when strategic risk taking (SRT) is not included in the model. The
Model 2 in Table 4 shows that the main effect of strategic risk taking
is positive and significant on both realized operational benefit (α2 =
0.026; T = 3.604; p b 0.001) and realized strategic benefit (α2 =
0.037; T = 4.810; p b 0.001), suggesting that a positive relationship
exists both between strategic risk taking and the operational benefit
of SC digitization, and between strategic risk taking and the strategic
benefit of SC digitization. Furthermore, the impact of strategic risk
taking on strategic benefits is higher than that on operational
benefits. The positive main effects of strategic risk taking on
operational and strategic benefits also exist in Model 4 where the
ITDR–knowledge interactions are added. Therefore, H2a and H2b
are supported.
To test the mediated moderation hypotheses of H3a–H3c, we
followed the procedure suggested in Muller et al. [41]. Specifically, the
existence of mediated moderation requires that: (1) in the absence of
strategic risk taking (i.e., the mediator), the ITDR–knowledge interac-
tions have positive and significant effects on SC digitization perfor-
mance (i.e., the realized operational and strategic benefits of SC
digitization); (2) the ITDR–knowledge interactions have positive and
significant effects on strategic risk taking, and strategic risk taking has
positive and significant effects on SC digitization performance; and
(3) when strategic risk taking is included, the ITDR–knowledge
interaction effects on SC digitization performance are smaller than
when strategic risk taking is not included. The results in Table 4 indi-
cate that strategic risk taking mediates the performance influence of
the interaction between ITDR and the IT unit's business knowledge
(i.e., ITDR × BK). Specifically, in Model 3 without strategic risk taking
(i.e., the mediator), the coefficient of the ITDR × BK interaction is
positive and significant in both the equation for operational benefit
(α7 = 0.073; T = 2.746; p b 0.01) and the equation for strategic
benefit (χ7 = 0.103; T = 3.214; p b 0.01). This suggests that the fit
between ITDR and the IT unit's business knowledge improves the
operational and strategic benefits of SC digitization. However, in
Model 4 with strategic risk taking, the coefficient of ITDR × BK is
still significant but smaller in the equation for operational benefit
(α7 = 0.056; T = 2.153; p b 0.05). Also, the coefficient of ITDR
× BK is significant and smaller in the equation for strategic benefit
(χ7 = 0.082; T = 2.474; p b 0.05). The abovementioned findings
suggest that the influence of the interaction ITDR × BK on strategic
risk taking is significantly positive (see the full model in Table 3),
and the influence of strategic risk taking on both operational benefit
and strategic benefit are significantly positive (see Model 4 in
Table 4). Two Sobel tests indicate that strategic risk taking signifi-
cantly mediates the impact of ITDR × BK on both operational benefit
(T = 1.694, p b 0.1) and strategic benefit (T = 1.960, p b 0.05).
Therefore, we can conclude that strategic risk taking partially
mediates the influence of ITDR × BK on both operational benefit
and strategic benefit. H3b is thus supported.
Moreover, the results in Table 4 indicate that strategic risk taking
mediates the impact of the interaction between ITDR and the IT unit's
knowledge generation capability (i.e., ITDR × KG) on strategic benefit.
Specifically, in Model 3 without strategic risk taking, the coefficient of
the ITDR × KG interaction is positive and significant in the equation for
strategic benefit (χ8 = 0.076; T = 1.984; p b 0.05). In Model 4 with
strategic risk taking, the coefficient of ITDR × KG becomes smaller
and insignificant in the equation for strategic benefit (χ8 = 0.046;
T = 1.582; p N 0.1). A Sobel test also suggests that strategic risk tak-
ing mediates the impact of ITDR × KG on strategic benefit (T = 1.820;
p b 0.1). However, the results do not suggest that strategic risk taking
mediates the impact of ITDR × KG on operational benefit. The coeffi-
cient of ITDR × KG is insignificant in both Model 3 without strategic
62 L. Xue / Decision Support Systems 62 (2014) 54–65

Table 3
Model on strategic risk taking.

Benchmark models Full model

1 2 3 4

Intercept 6.032⁎⁎ (2.352) 6.986⁎⁎ (2.486) 10.395⁎⁎⁎⁎ (3.300) 11.188⁎⁎⁎⁎ (3.354) 11.459⁎⁎⁎⁎ (3.486)
ITDR 1.062⁎⁎⁎ (2.726) 1.082⁎⁎⁎ (2.753) 1.080⁎⁎⁎ (2.821) 1.017⁎⁎⁎ (2.599) 1.055⁎⁎⁎ (2.775)
TK 1.021⁎⁎⁎ (2.880) 1.056⁎⁎⁎ (2.983) 0.681⁎⁎ (2.100) 0.777⁎⁎ (2.351) 0.660⁎⁎ (2.044)
BK 0.072 (0.176) 0.051 (0.129) 0.056 (0.148) 0.515 (1.262) 0.252 (0.635)
KG 1.054⁎⁎ (2.089) 1.021⁎⁎ (2.048) 1.366⁎⁎ (2.973) 0.791 (1.689) 1.176⁎⁎ (2.531)
ITDR × TK 0.247 (1.530) −0.012 (−0.083)
ITDR × BK 1.347⁎⁎⁎⁎ (8.896) 1.003⁎⁎⁎⁎ (4.911)
ITDR × KG 1.403⁎⁎⁎⁎ (7.640) 0.605⁎⁎ (2.442)
EOB 0.081 (1.130) 0.116 (1.198) 0.166 (1.300) 0.169 (1.148) 0.130 (1.556)
ESB 0.349⁎⁎ (2.400) 0.353⁎⁎ (2.410) 0.318⁎⁎ (2.561) 0.487⁎⁎ (2.547) 0.236⁎ (1.783)
Risk propensity 1.040⁎⁎⁎ (2.732) 1.020⁎⁎⁎ (2.680) 0.832⁎⁎ (2.401) 0.688⁎ (1.935) 0.734⁎⁎ (2.123)
IT budget −0.269⁎⁎ (−2.177) −0.289⁎⁎ (−2.328) −0.377⁎⁎⁎⁎ (−3.330) −0.472⁎⁎⁎⁎ (−3.990) −0.436⁎⁎⁎⁎ (−3.792)

Industry dummies are also used as control variables…

Adj R2 0.169 0.179 0.317 0.287 0.330
F 5.33⁎⁎⁎⁎ 5.76⁎⁎⁎⁎ 9.29⁎⁎⁎⁎ 8.95⁎⁎⁎⁎ 10.85⁎⁎⁎⁎

Note: 1. Number of observations = 384; 2. In parentheses are T statistics.

⁎ p b 0.1.
⁎⁎ p b 0.05.
⁎⁎⁎ p b 0.01.
⁎⁎⁎⁎ p b 0.001.

risk taking (α8 = 0.044; T = 1.212; p N 0.1) and Model 4 with strate- in the equations for operational benefit, the coefficient of ITDR × TK
gic risk taking (α8 = 0.037; T = 1.023; p N 0.1). Therefore, H3c is is positive and significant both when strategic risk taking is not included
partially supported. (α6 = 0.055; T = 2.381; p b 0.05) and when it is included (α6 = 0.056;
The results in Table 4, however, shows that strategic risk taking does T = 2.451; p b 0.05), and the difference is not significant (T = 0.043;
not mediate the performance impact of the interaction between ITDR p N 0.1). In the equation for strategic benefit, the coefficient of
and the IT unit's technical knowledge (i.e., ITDR × TK). Specifically, ITDR × TK is not significant both when strategic risk taking is not in-
cluded (χ6 = − 0.013; T = − 0.535; p N 0.1) and when it is included
SRT (χ6 = − 0.010; T = − 0.464; p N 0.1), and the difference is also not
50 significant (T = 0.130; p N 0.1). In addition, Table 3 shows that the
coefficient of the interaction ITDR × TK is not significant in the model
High ITDR for strategic risk taking. These results, in conjecture, suggest that strate-
40 gic risk taking does not mediate the influence of ITDR × TK on SC digiti-
zation performance. Therefore, H3a is not support. A potential
Intermediate explanation is that when the IT unit is allocated with more decision
30
rights, its existing technical knowledge is better used in non-risky
ITDR tasks (e.g., better using mature technologies to digitize the existing
20 business processes) to improve operational performance. The existing
technical knowledge, however, is of limited value to risky tasks (e.g.,
Low ITDR
using emerging technologies to create new business processes). In addi-
10 tion, the contribution of the IT unit's existing technical knowledge to
strategic benefits is limited.
BK
0
3.07 4.57 6.07 5. Discussion

SRT In this section, we discuss the main findings of this study and their
40
High ITDR theoretical implications. The first main finding of this study is on how
the governance–knowledge fit of the IT unit fosters strategic risk taking.
Our analysis shows that allocating more strategic decision rights to the
30 IT unit motivates the organization to be more risk taking in SC digitiza-
Intermediate tion. Moreover, decision right allocation also strengthens the positive
relationship between strategic risk taking and the IT unit's business
ITDR
20 knowledge and knowledge generation capability. These observations
square with the existing theories on risk taking that decision makers'
knowledge fosters strategic risk taking [51,53,63]. In addition, these
Low ITDR
10 observations illustrate more implications of governance–knowledge fit
and complement the related literature [23,57] by showing how the
governance–knowledge fit contributes to the firm's strategic risk taking.
KG The study also generates deeper insights on how the influence of the
0
1.81 3.24 4.67 6.1 governance–knowledge fit of the IT unit on strategic risk taking is
dependent on the type of knowledge. Our findings suggest that decision
Fig. 2. The moderating effects of ITDR. ITDR — Decision Rights of IT Unit; BK — Business right allocation complements the IT unit's business knowledge and
Knowledge; KG — Knowledge Generation Capability; SRT — Strategic Risk Taking. knowledge generation capability in encouraging more strategic risk
 L. Xue / Decision Support Systems 62 (2014) 54–65 63

taking by the firm. However, decision right allocation does not com-

−0.114⁎⁎ (−2.143)

−0.087 (−1.381)

−0.010 (−0.464)

−0.001 (−0.029)
−0.034 (−0.585)
0.037 (0.464)
1.976 (1.593)

0.046 (1.582)
0.379⁎⁎⁎⁎ (7.482)
0.019⁎⁎ (2.459)

0.142⁎⁎⁎ (2.756)

0.290⁎⁎⁎⁎ (3.999)

0.082⁎⁎ (2.474)
plement the IT unit's technical knowledge in encouraging strategic
risk taking. The IT unit's technical knowledge is positively associated
with strategic risk taking, regardless of the extent to which the IT
unit is actually involved in the strategic decision group.
RSB

The second main finding of this study is the positive relationship

between strategic risk taking in SC digitization and the realization of
operational and strategic benefits in SC digitization. This finding
complements the IT risk literature [13,56] in explaining the risk–return

0.042 (0.545)
0.577 (0.486)
0.424⁎⁎⁎⁎ (9.651)

0.168⁎⁎ (2.408)
0.015⁎ (1.946)
0.004 (0.138)
0.098⁎ (1.965)
0.025 (0.406)

0.056⁎⁎ (2.451)

0.037 (1.023)
0.056⁎⁎ (2.153)

0.002 (0.049)
0.013 (0.238)
association in IT investment. While the existing literature implicitly as-
sumes strategic risk taking and used it to explain the observed positive
Model 4

relationship between the risk outcomes (i.e., uncertainty) and return,

ROB

this study explicitly captures the firm-level decisions of strategic risk

taking and shows how it is positively associated with the realized ben-
efits. In this way, this study provides direct evidence for the argument
on the risk–return association in the IT risk literature.
−0.103⁎ (−1.945)

−0.082 (−1.295)

−0.013 (−0.535)

−0.003 (−0.050)
−0.028 (−0.466)
2.089⁎ (1.674)

0.374⁎⁎⁎⁎ (7.411)

0.152⁎⁎⁎ (2.913)

0.315⁎⁎⁎⁎ (4.341)

0.103⁎⁎⁎ (3.214)
0.076⁎⁎ (2.484)

0.065 (0.794)

The third main finding of this study is on how the governance–

knowledge fit of the IT unit influences the performance of SC digiti-
zation through strategic risk taking. Our investigation suggests that
strategic risk taking is an important intervening factor that medi-
RSB

ates the impact of fit between the IT unit's decision rights and the
IT unit's business knowledge and knowledge generation capability.
In other words, when the IT unit has more business knowledge and
more capability in knowledge generation, the allocation of more
0.663 (0.555)
0.423⁎⁎⁎⁎ (9.611)

0.014 (0.287)
0.104⁎ (1.693)
0.028 (0.466)

0.055⁎⁎ (2.381)
0.073⁎⁎⁎ (2.746)

0.001 (0.024)
0.018 (0.337)
0.187⁎⁎⁎ (2.697)

0.044 (1.212)

0.062 (0.804)

strategic decision rights to the IT unit benefits by enabling the orga-

nization to more strategically pursue risk opportunities. This find-
Model 3

ing helps align the knowledge-based view of risk taking in the

ROB

strategy literature [3,51] with the governance–knowledge-fit view

in the IT governance literature [57]. The existing IT governance lit-
erature considered how governance–knowledge fit enhances IT
performance through influencing certain intermediate factors, e.g.,
−0.127⁎⁎ (−2.332)

−0.106 (−1.684)

−0.051 (−0.863)
0.256⁎⁎⁎⁎ (3.486)
0.392⁎⁎⁎⁎ (7.611)
0.036⁎⁎⁎⁎ (4.810)

decision right exercise effectiveness [57]. This study extends the

0.168⁎⁎⁎ (3.167)

governance–knowledge-fit view by showing strategic risk taking

1.898 (1.483)

0.005 (0.083)

0.006 (0.089)

as another potential mediator of performance.

The results also indicate that the mediation of strategic risk taking on
0.356
0.711
RSB

governance–knowledge fit does not apply to the IT unit's technical

knowledge. An implication is that the fit between the IT unit's decision
rights and its technical knowledge enhances SC digitization perfor-
−0.001 (−0.029)

mance mainly through non-risky ways, such as using mature technolo-

0.429⁎⁎⁎⁎ (9.781)

0.026⁎⁎ (3.604)

0.115⁎⁎ (2.280)

0.145⁎⁎ (2.090)

gies to digitize established business processes. This is also consistent

0.550 (0.466)

0.013 (0.228)

0.001 (0.029)
0.023 (0.297)
0.006 (0.119)

with the finding that the complementarity between the IT unit's

Model 2

decision rights and its technical knowledge enhances operational

0.348
0.715
ROB

benefits rather than strategic benefits of SC digitization. The evidence

of complementarity between the IT unit's decision rights and its busi-
ness knowledge, but not technical knowledge, in enhancing strategic
risk taking is also in line with the idea of Tiwana [57] that the IT unit's
−0.129⁎⁎ (−2.372)

Note: 1. Number of observations = 384; 2. In parentheses are T statistics.

−0.105 (−1.667)

−0.050 (−0.853)
0.395⁎⁎⁎⁎ (7.640)

0.165⁎⁎⁎⁎ (3.122)

0.259⁎⁎⁎⁎ (3.520)

peripheral knowledge of business, rather than its primary knowledge

of technology, is more critical to governance–knowledge fit.
1.844 (1.441)

0.003 (0.050)

0.011 (0.138)

Finally, this study complements the existing IOS literature [39,55] by

illustrating how strategic risk taking enables firms to pursue both
0.332
0.720
RSB

operational and strategic benefits through SC digitization. The consider-

Industry dummies are included as control variables…

ation of strategic risk taking in this scenario is also aligned with the
broader idea in the strategy literature (e.g., [55]) that the pursuit of
both operational and strategic benefits is a challenge for organizations
−0.001 (−0.021)
0.439⁎⁎⁎⁎ (9.841)

0.100⁎⁎ (1.974)

0.160⁎⁎ (2.272)

and requires strategic risk taking.

0.297 (0.248)

0.011 (0.218)

0.004 (0.079)
0.022 (0.357)

0.045 (0.575)
Model 1

6. Conclusion
0.318
0.715
ROB
SC digitization performance.

By developing and testing a theoretical framework of strategic

risk taking, this study addresses an important research topic
regarding how knowledge and IT governance may influence the
Cross-model corr.
Sys weighted R2

performance of SC digitization through strategic risk taking. The the-

⁎⁎⁎⁎ p b 0.001.
Firm revenue
Years of SCD

⁎⁎ p b 0.05.
⁎⁎⁎ p b 0.01.
⁎ p b 0.1.

oretical framework conceptualizes a mediated moderation model in

ITDR × KG
ITDR × BK
ITDR × TK

Employee
Intercept

which the decision right of the IT unit interacts (i.e., fits) with its
Table 4

ITDR
EOB

knowledge basis and capability to enhance strategic risk taking deci-

SRT
ESB

KG
BK
TK

sions, and strategic risk taking mediates the impact of governance–

64 L. Xue / Decision Support Systems 62 (2014) 54–65

knowledge fit on the operational benefits and strategic benefits of SC
digitization. The empirical study provides evidence on the fit be-
tween the decision right of the IT unit and its business knowledge
and knowledge creation capability in enhancing strategic risk taking,
and the mediation of strategic risk taking on the impacts of these two
types of governance–knowledge fit on the operational and strategic
benefits of SC digitization. The fit between the decision right of the
IT unit and its technical knowledge, however, appear to directly im-
prove the operational benefits of SC digitization, rather than through
strategic risk taking.
The study also has some limitations, which provide opportunities for
future research. First, this research focuses on governance–knowledge fit
in the IT unit. Future research may consider governance–knowledge fit
in other functional units of the firm. Second, this study focuses on IT gov-
ernance by considering the decision allocation to the IT unit. Future
research may consider other aspects of IT governance, such as strategic
planning and project management, to shed more light on governance–
knowledge fit. Third, in this study, we focus on the risk associated
with supply chain digitization. Future research may examine other
types of risk taking in IT adoption and implementation. Finally, future
research may also validate the findings and insights of this study using
samples of organizations in different countries and with different
cultural backgrounds. Another limitation of this study is that although
we have employed a few tests used in the existing literature to verify
the lack of common-method bias in our study, we were not able to
employ more advanced research design approaches to address this
potential issue. Future studies may consider relying on more ad-
vanced design approaches, such as the use of multiple respondents
from each organization, to more rigorously address the potential
problem of common-method bias and verify our findings.

Appendix 1. Exploratory factor analysis

Item 1 2 3 4 5 6 7 8 9 10 11

ITDR1 0.706 0.206 0.201 0.229 0.075 0.102 0.331 0.166 0.052 0.200 0.220
ITDR2 0.765 0.233 0.210 0.249 0.135 0.077 0.345 0.148 0.059 0.203 0.164
ITDR3 0.717 0.216 0.201 0.214 0.118 0.100 0.353 0.154 0.065 0.215 0.170
ITDR4 0.778 0.265 0.209 0.239 0.141 0.092 0.360 0.163 0.082 0.200 −0.108
ITDR5 0.782 0.218 0.199 0.217 0.127 0.075 0.356 0.189 0.070 0.215 −0.069
ITDR6 0.801 0.246 0.225 0.221 0.157 0.090 0.317 0.163 0.059 0.217 −0.155
ITDR7 0.788 0.253 0.195 0.245 0.167 0.115 0.303 0.190 0.029 0.225 −0.063
SCD1 0.244 0.741 0.205 −0.014 0.158 0.236 0.075 0.219 0.118 0.086 0.102
SCD2 0.231 0.815 0.218 −0.032 0.180 0.240 0.096 0.241 0.114 0.073 0.077
SCD3 0.287 0.780 0.203 −0.033 0.175 0.230 0.076 0.217 0.144 0.099 0.035
SCD4 0.314 0.818 0.197 −0.020 0.147 0.259 0.091 0.231 0.096 0.065 −0.030
SCD5 0.268 0.806 0.214 −0.033 0.158 0.224 0.061 0.233 0.127 0.057 −0.004
SCD6 0.259 0.819 0.229 0.008 0.183 0.221 0.092 0.223 0.137 0.072 −0.040
BK1 0.106 0.094 0.862 0.093 0.255 −0.028 0.055 0.253 0.036 0.175 0.311
BK2 0.134 0.081 0.868 0.107 0.267 −0.010 0.084 0.275 0.032 0.176 0.252
BK3 0.071 0.129 0.883 0.077 0.246 −0.025 0.042 0.227 0.042 0.153 0.104
BK4 0.157 0.076 0.885 0.114 0.134 −0.033 0.064 0.278 0.012 0.202 −0.057
BK5 0.206 0.122 0.842 0.116 0.129 −0.033 0.082 0.231 0.026 0.189 −0.137
KG1 0.260 0.361 0.200 0.800 0.091 0.087 0.152 −0.010 0.120 0.040 0.328
KG2 0.283 0.225 0.207 0.817 0.268 0.117 0.124 0.017 0.136 0.019 0.048
KG3 0.345 0.392 0.173 0.777 0.157 0.096 0.123 0.001 0.108 0.026 0.259
KG4 0.349 0.387 0.181 0.787 0.154 0.075 0.139 −0.020 0.122 0.069 0.233
KG5 0.245 0.414 0.207 0.788 0.227 0.106 0.146 0.003 0.106 0.028 0.170
Risk1 0.175 0.204 0.084 0.213 0.805 0.036 0.209 0.302 0.224 0.052 0.084
Risk2 0.155 0.183 0.096 0.194 0.890 0.018 0.225 0.322 0.200 0.046 0.030
Risk3 0.124 0.155 0.093 0.227 0.862 0.059 0.207 0.282 0.216 0.032 −0.039
Risk4 0.149 0.072 0.058 0.203 0.892 0.058 0.227 0.280 0.220 0.048 −0.019
TK1 0.072 0.121 0.035 0.289 0.175 0.820 0.183 0.231 0.059 0.071 0.234
TK2 0.074 0.165 0.020 0.313 0.162 0.840 0.182 0.253 0.044 0.083 0.182
TK3 0.089 0.116 0.028 0.292 0.122 0.833 0.210 0.235 0.052 0.063 0.104
TK4 0.056 0.121 0.007 0.295 0.177 0.809 0.193 0.256 0.075 0.051 −0.183
EOB1 0.074 0.059 0.329 0.062 0.330 0.215 0.842 0.202 0.311 0.108 0.014
EOB2 0.062 0.009 0.328 0.077 0.321 0.220 0.814 0.181 0.317 0.098 0.037
EOB3 0.080 −0.038 0.304 0.055 0.282 0.241 0.816 0.178 0.331 0.114 0.039
EOB4 0.062 −0.065 0.311 0.081 0.312 0.243 0.793 0.176 0.299 0.110 0.064
ESB1 0.204 0.001 0.061 0.179 0.129 0.175 0.087 0.816 0.285 0.143 0.076
ESB2 0.211 −0.005 0.060 0.186 0.116 0.146 0.086 0.824 0.265 0.116 0.073
ESB3 0.214 −0.003 0.079 0.191 0.140 0.171 0.089 0.822 0.257 0.172 0.068
ESB4 0.203 −0.010 0.082 0.190 0.120 0.174 0.116 0.833 0.267 0.148 0.070
ESB5 0.201 0.017 0.087 0.174 0.156 0.188 0.060 0.817 0.304 0.136 0.078
RSB1 0.318 0.233 0.077 0.007 0.210 0.084 0.166 0.143 0.874 0.063 0.253
RSB2 0.335 0.204 0.101 0.000 0.196 0.081 0.144 0.114 0.818 0.079 0.263
RSB3 0.318 0.208 0.049 −0.003 0.185 0.100 0.170 0.136 0.830 0.051 0.266
RSB4 0.335 0.235 0.080 −0.007 0.182 0.102 0.145 0.119 0.868 0.089 0.240
RSB5 0.328 0.257 0.079 0.000 0.195 0.089 0.148 0.134 0.858 0.035 0.247
ROB1 0.060 0.185 0.218 0.021 0.066 0.106 0.179 0.080 0.266 0.792 0.191
ROB2 0.069 0.198 0.188 0.043 0.087 0.083 0.173 0.088 0.238 0.751 0.191
ROB3 0.041 0.156 0.213 0.032 0.044 0.078 0.200 0.097 0.261 0.760 0.199
ROB4 0.040 0.207 0.246 −0.003 0.056 0.097 0.184 0.099 0.290 0.831 0.210
Propen1 0.318 0.241 0.084 0.031 0.278 0.216 0.045 0.159 0.223 0.102 0.802
Propen2 0.147 0.324 0.101 0.048 0.217 0.218 0.028 0.185 0.249 0.086 0.718
Propen3 0.030 0.285 0.090 0.033 0.221 0.226 0.061 0.177 0.217 0.081 0.766
Eigenvalue 5.597 5.010 4.557 4.224 3.752 3.464 3.227 2.854 2.589 2.113 1.772
% of Var. Explained 9.593 8.707 8.092 7.628 7.188 6.781 6.411 6.103 5.703 5.498 5.170

Bold values are indicator loadings on their corresponding factors.

 L. Xue / Decision Support Systems 62 (2014) 54–65
References
[1] M. Alavi, D.E. Leidner, Review: knowledge management and knowledge manage-
ment systems: conceptual foundations and research issues, MIS Quarterly 25 (1)
(2001) 107–136.
[2] J.S. Armstrong, T.S. Overton, Estimating nonresponse bias in mail surveys, Journal of
Market Research XIV (1977) 396–402.
[3] I.S. Baird, H. Thomas, Toward a contingency model of strategic risk taking, Academy
of Management Review (1985) 230–243.
[4] A. Barua, et al., An empirical investigation of net-enabled business value: an explor-
atory investigation, MIS Quarterly 28 (4) (2004) 585–620.
[5] M.J. Benner, M.L. Tushman, Exploitation, exploration, and process management: the
productivity dilemma revisited, Academy of Management Review 2 (2003) 238–256.
[6] P.M. Bentler, EQS structural equations program manual, BMDP Statistical Software,
Los Angeles, CA, 1995.
[7] Q. Cao, Q. Gan, M.A. Thompson, Organizational adoption of supply chain manage-
ment system: a multi-theoretic investigation, Decision Support Systems 55 (3)
(2013) 720–727.
[8] M.A. Carpenter, T.G. Pollock, M.M. Leary, Testing a model of reasoned risk-taking:
governance, the experience of principals and agents, and global strategy in high-
technology IPO firms, Strategic Management Journal 24 (9) (2003) 803–820.
[9] A.L.M. Cavaye, P.B. Cragg, Factors contributing to the success of customer oriented
interorganizational systems, Journal of Strategic Information Systems 4 (1995)
13–30.
[10] V. Choudhury, R. Sabherwal, Portfolios of control in outsourced software develop-
ment projects, Information Systems Research 14 (3) (2003) 291–314.
[11] J. Dedrick, S.X. Xu, K.X. Zhu, How does information technology shape supply-chain
structure? Evidence on the number of suppliers, Journal of Management Informa-
tion Systems 25 (2) (2008) 41–72.
[12] S. Dewan, F. Ren, Information technology and firm boundaries: impact on firm risk
and return performance, Information Systems Research 22 (2) (2009) 369–388.
[13] S. Dewan, C. Shi, V. Gurbaxani, Investigating the risk–return relationship of informa-
tion technology investment: firm-level empirical analysis, Management Science 53
(12) (2007) 1829–1842.
[14] S. Dong, S. Xu, K. Zhu, Information technology in supply chains: the value of
IT-enabled resources under competition, Information System Research 20
(1) (2009) 18–32.
[15] S. Dutta, I. Mia, The global information technology report 2010–2011, World
Economic Forum and INSEAD, 2011.
[16] J.R. Edwards, L.S. Lambert, Methods for integrating moderation and mediation: a
general analytical framework using moderated path analysis, Psychological
methods 12 (1) (2007) 1.
[17] M. Ghosh, G. John, Governance value analysis and marketing strategy, Journal of
Marketing 63 (1999) 131–145.
[18] R.M. Grant, Toward a knowledge-based theory of the firm, Strategic Management
Journal 17 (1996) 109–122.
[19] O.P. Hall, A. Scott, M. Chun, Woodsynergy Inc.: Integrating IT into the Supply Chain,
Richard Ivey School of Business Case, 2010. (9B10E013).
[20] J. Hansen, The power and promise of Web-based procurement tools, Supply Chain
Quarterly 2 (2012) 54–56.
[21] H.H. Harman, Modern Factor Analysis, University of Chicago Press, 1976.
[22] P. Hart, C. Saunders, Power and trust: critical factors in the adoption and use of
electronic data interchange, Organization Science 8 (1) (1997) 23–38.
[23] M.C. Jensen, W.H. Meckling, Specific and general knowledge and organizational
structure, in: L. Werin, H. Wijkander (Eds.), Contract Economics, Blackwell, Oxford,
1992, pp. 251–274.
[24] D. Kahneman, A. Tversky, Prospect theory: an analysis of decision under risk,
Econometrica 47 (1979) 263–291.
[25] M. Keil, et al., A framework for identifying software project risks, Communications of
the ACM 41 (11) (1998) 76–83.
[26] M. Keil, et al., A cross-cultural study on escalation of commitment behavior in
software projects, MIS Quarterly 24 (2) (2000) 299–325.
[27] M. Keil, et al., An investigation of risk perception and risk propensity on the decision
to continue a software development project, Journal of Systems and Software 53 (2)
(2000) 145–157.
[28] R. Klein, A. Rai, Interfirm strategic information flows in logistics supply chain
relationships, MIS Quarterly 33 (4) (2009) 735–762.
[29] B. Kogut, U. Zander, Knowledge of the firm, combination capabilities, and the
replication of technology, Organization Science 3 (1992) 383–397.
[30] M. Krigsman, M. Krigsman, CRM failure rates: 2001–2009. TechRepulic.com, p. Avail-
able at http://www.techrepublic.com/blog/tech-manager/crm-failure-rates-2001-
2009/1750 2009.
[31] K. Kumar, H.G. Van Dissel, Sustainable Collaboration: managing conflict and cooper-
ation in interorganizational systems, MIS Quarterly 20 (3) (1996) 279–300.
[32] M.H. Kutner, C.J. Nachesheim, J. Neter, Applied Linear Regression Models, 4th edition
McGraw-Hill Irwin, 2004. (p).
[33] M. Larraza-Kintana, et al., Disentangling compensation and employment risks using
the behavioral agency model, Strategic Management Journal 28 (2007) 1001–1019.
[34] O. Lavastrea, A. Gunasekaranb, A. Spalanzani, Supply chain risk management in
French companies, Decision Support Systems 52 (4) (2012) 828–838.
[35] D.E. Leidner, T. Kayworth, Review: a review of culture in information systems re-
search: toward a theory of information technology culture conflict, MIS Quarterly
30 (2) (2006) 357–399.
[36] M.K. Lindell, D.J. Whitney, Accounting for common method variance in cross-
sectional research designs, Journal of Applied Psychology 86 (1) (2001) 114–121.
65
[37] K. Lyytinen, L. Mathiassen, J. Ropponen, Attention shaping and software risk — a
categorical analysis of four classical risk management approaches, Information
Systems Research 9 (3) (1998) 233–255.
[38] H. Mintzberg, D. Raisinghani, A. Theoret, The structure of “unstructured” decision
processes, Administrative Science Quarterly 25 (1976) 246–275.
[39] T. Mukhopadhyay, S. Kekre, Strategic and operational benefits of electronic integration
in B2B procurement processes, Management Science 48 (10) (2002) 1301–1313.
[40] T. Mukhopadhyay, S. Kekre, S. Kalathur, Business value of information technology: a
study of electronic data interchange, MIS Quarterly 19 (2) (1995) 137–156.
[41] D. Muller, C.M. Judd, V.Y. Yzerbyt, When moderation is mediated and mediation is
moderated, Journal of Personality and Social Psychology 89 (6) (2005) 852–863.
[42] J. Nahapiet, S. Ghoshal, Social capital, intellectual capital, and the organizational
advantage, Academy of Management Review 23 (1998) 242–266.
[43] A.I. Nicolaou, D.H. McKnight, Perceived information quality in data exchanges: effects
on risk, trust, and intention to use, Information Systems Research 17 (4) (2006)
332–351.
[44] A. Rai, P. Brown, X. Tang, Organizational assimilation of electronic procurement
innovations, Journal of Management Information Systems 26 (1) (2009) 257–296.
[45] A. Rai, R. Patnayakuni, N. Seth, Firm performance impacts of digitally enabled supply
chain integration capabilities, MIS Quarterly 30 (2) (2006) 225–246.
[46] F.J. Riggins, T. Mukhopadhyay, Interdependent benefits from interorganizational
systems: opportunities for business partner reengineering, Journal of Management
Information Systems 11 (2) (1994) 37–57.
[47] F.J. Riggins, T. Mukhopadhyay, Overcoming EDI adoption and implementation risks,
International Journal of Electronic Commerce 3 (4) (1999) 103–123.
[48] W.J. Ross, C.M. Beath, Campbell Soup Company: harmonizing processes and
empowering workers, CISR Working Paper No. 374, Massachusetts Institute of
Technology, 2008.
[49] V. Sambamurthy, R.W. Zmud, Arrangements for information technology gover-
nance: a theory of multiple contingencies, MIS Quarterly 23 (2) (1999) 261–290.
[50] H.A. Simon, A behavioral model of rational choice, Quarterly Journal of Economics
69 (1) (1955) 99–118.
[51] S.B. Sitkin, A.L. Pablo, Reconceptualizing the determinants of risk behavior, Academy
of Management Review 17 (1) (1992) 9–38.
[52] S.B. Sitkin, L.R. Weingart, Determinants of risky decision-making behavior: a test of
the mediating role of risk perceptions and propensity, Academy of Management
Journal 38 (6) (1995) 1573–1592.
[53] K.G. Smith, C.J. Collins, K.D. Clark, Existing knowledge, knowledge creation capability,
and the rate of new product introduction in high-technology firms, Academy of
Management Journal 48 (2) (2005) 346–357.
[54] J.C. Spender, Making knowledge the basis of a dynamic theory of the firm, Strategic
Management Journal 17 (1996) 45–62.
[55] M. Subramani, How do suppliers benefit from information technology use in supply
chain relationships? MIS Quarterly 28 (1) (2004) 45–73.
[56] H. Tanriverdi, T. Ruefli, The role of information technology in the risk/return
relations of firms, Journal of the Association for Information Systems 5 (11–12)
(2004) 421–447.
[57] A. Tiwana, Governance–knowledge fit in systems development projects, Informa-
tion Systems Research 20 (2) (2009) 180–197.
[58] A. Tiwana, B. Konsynski, Complementarities between organizational IT architecture
and governance structure, Information Systems Research 21 (2) (2010) 288–304.
[59] Z. Ton, S.C. Wheelwright, Governance of interorganizational information systems: a re-
source dependence perspective, Harvard Business School Case, 2005. (No. 9-6-5-080).
[60] N. Venkatraman, The concept of fit in strategy research: toward verbal and statisti-
cal correspondence, Academy of Management Review 14 (3) (1989) 423–444.
[61] Z. Wang, S.-L. Pan, T. Ouyang, The Haier group: an evolving IT organizational identity,
p National University of Singapore, Asian IT Case Series, Case No. OID 001, 2011.
[62] P. Weill, J.W. Ross, IT Governance: How Top Performers Manage IT Decision Rights
for Superior Results, Harvard Business Press, 2004.
[63] R.M. Wiseman, L.R. Gomez-Mejia, A behavioral agency model of managerial risk
taking, Academy of Management Review 23 (1) (1998) 133–153.
[64] J.M. Wooldridge, Econometric Analysis of Cross Section and Panel Data, 2nd ed. The
MIT Press, Cambridge, Massachusetts, 2010.
[65] L. Xue, G. Ray, B. Gu, Environmental uncertainty and IT infrastructure governance: a
curvilinear relationship, Information Systems Research 22 (2) (2011) 301–314.
[66] Y. Yao, M. Dresner, J. Palmer, Private network EDI vs. the internet electronic
markets: a direct comparison of fulfillment performance, Management Science 55
(5) (2009) 843–852.
[67] Y. Yao, J. Palmer, M. Dresner, An interorganizational perspective on the use of
electronically-enabled supply chains, Decision Support Systems 43 (3) (2007)
884–896.
[68] K. Zhu, et al., Migration to open-standard interorganizational systems: network
effects, switching costs, and path dependency, MIS Quarterly 30 (2006) 515–539
Dr. Ling Xue is an assistant professor at the Department of Information Systems and Sup-
ply Chain Management at the Bryan School of Business and Economics, the University of
North Carolina at Greensboro. He received his Ph.D. degree in Management Science and
Information Systems from the McCombs School of Business at the University of Texas at
Austin. His research interests are in the areas of IT governance, the business value of IT,
electronic commerce and information security. His papers have been published in
Information Systems Research, MIS Quarterly, Academy of Management Journal, Journal of
Operations Management, Journal of Management Information Systems, Production and Oper-
ations Management, Decision Support Systems, International Journal of Electronic Commerce,
Journal of Global Information Management and the proceedings of International Conference
on Information Systems.