Mediating effect of IT-enabled capabilities on ...

J. Eng. Technol. Manage. 36 (2015) 1–23

Contents lists available at ScienceDirect

Journal of Engineering and Technology Management journal homepage: www.elsevier.com/locate/jengtecman

Mediating effect of IT-enabled capabilities on competitive performance outcomes: An empirical investigation of ERP implementation David Hwang a, Ma Ga (Mark) Yang b, Paul Hong c,* a

Department of Finance and Supply Chain Management, John L. Grove School of Business, Shippensburg University of Pennsylvania, Shippensburg, PA 17257, USA b Management Department, College of Business and Public Affairs, West Chester University of Pennsylvania, West Chester, PA 19383, USA c Department of Information, Operations and Technology Management, College of Business and Innovation, The University of Toledo, Toledo, OH 43606, USA

A R T I C L E I N F O

A B S T R A C T

JEL classification: M15(IT Management)

This paper presents an integrated research model related to ERP system implementation and empirically investigates its impact of this implementation on performance outcomes. We argue that effective ERP implementation requires better front-end planning to enable timely coordination (i.e., configurable), continuous interactions (i.e., adaptable), and organization-wide application (i.e., integrative). The hypothesized relationships are tested with data collected from 205 Korean manufacturing firms. Our empirical findings indicate that an ERP implementation strategy alone has no direct effect on performance but does have an indirect effect on performance through IT-enabled combinative capabilities, suggesting that IT-enabled combinative capabilities mediate performance outcomes. ß 2015 Elsevier B.V. All rights reserved.

Keywords: ERP implementation IT-enabled combinative capabilities Manufacturing firms Competitive advantage Empirical investigation

Introduction In today’s highly competitive global business environment, firms use information systems to improve customer service, shorten cycle times, and reduce cost (Hong and Kim, 2002; Jacobs and Weston, 2007; Soja, 2006). Enterprise Resource Planning (ERP) systems are considered to be strategic

* Corresponding author. Tel.: +1 419 530 2054; fax: +1 419 530 2290. E-mail addresses: [email protected] (D. Hwang), [email protected] (M.G.(. Yang), [email protected] (P. Hong). http://dx.doi.org/10.1016/j.jengtecman.2015.03.001 0923-4748/ß 2015 Elsevier B.V. All rights reserved.

2

D. Hwang et al. / Journal of Engineering and Technology Management 36 (2015) 1–23

tools in achieving competitive capabilities (Wier et al., 2007; Xue et al., 2005; You et al., 2012). ERP systems synchronize procedures, applications, and metrics that span intra- and inter-firm boundaries; provide accurate, timely, and integrated information; and improve organizational decision making (Su and Yang, 2010; Trott and Hoecht, 2004). ERP systems are associated with improving firm performance in the form of redesigned business processes, integrated managerial functions, accelerated reporting cycles, and expanded information capabilities (Mabert et al., 2003). In spite of the potential benefits of ERP systems, the research findings are often conflicting. Hayes et al. (2001) point out the significantly higher stock returns that occur upon announcing the implementation of an ERP system. Likewise, Hunton et al. (2003) call attention to the greater longterm return on assets for ERP adopters relative to non-adopters. In addition, Wier et al. (2007) report a positive relationship between ERP adoption and non-financial performance – a relationship that further impacts current and long-term return on assets. In this respect, IT investments might be directly related to performance measures such as cost, quality, time to market, and product variety (Krause et al., 2007; Nahm et al., 2003). On the other hand, Poston and Grabski (2001) suggest that ERP systems do not necessarily improve business profitability. Hitt et al. (2002) likewise suggest that although there is an improved financial performance during the implementation stage, there is no sustainable long-term impact. In addition, Mabert et al. (2003) suggest that there is a lack of evidence showing a clear link between ERP systems and direct operational costs reduction. In light of these mixed results, further research is needed to reconcile the conflicting research related to ERP implementation and its firm-level performance outcomes (Bendoly et al., 2009; Lee and Myers, 2004; You et al., 2012). In terms of identifying a direct relationship between ERP implementation and performance outcomes, careful examination of ERP system implementation requires developing empirically validated constructs (You et al., 2012). In view of such research needs, this paper aims to examine the relationship between ERP implementation strategy and competitive performance outcomes by investigating the mediating effect of IT-enabled combinative capabilities. Based on an extensive literature review, we present an integrated research model that defines important relationships related to ERP implementation. We used a carefully developed survey instrument to measure the dynamics of the relationships among ERP implementation strategy, ITenabled combinative capabilities, and competitive performance outcomes. The contributions of this research are threefold: (1) identifying and developing the theoretically rooted ERP implementation strategy construct in a way that considers dynamically changing business environments, (2) empirically validating the ways in which firms positively influence performance outcomes through ITenabled combinative capabilities, and (3) examining how dynamics within ERP implementation and its impact on a firm’s performance differ in diverse circumstances (e.g., small firms vs. large firms, process integration, and product complexity). The remaining section of this paper is organized as follows: Conceptual development section provides the theoretical base for this research argument and presents an overview of the relevant literature that defines the variables within the research model. Hypothesis development section develops hypothesized relationships among these variables. Research method section details the research methodology and describes the data collection procedures, and Research results section reports the research results. The final section, presents theoretical and managerial implications along with concluding remarks. Conceptual development In studying the role of ERP, researchers may take a variety of theoretical perspectives: (1) the dynamic capability perspective emphasizes organizational capabilities under changing environments (Sinkovics and Roath, 2004; Teece et al., 1997); (2) the resource-based view examines the enhancement provided by technological and human resources (Lengnick-Hall et al., 2004; Wernerfelt, 1984; Barney, 1991); and (3) the technology implementation view focuses on implementing a specific set of technology infrastructures and people processes (Bradley, 2008; Hong and Kim, 2002). Fig. 1 illustrates a research model that presents the ERP implementation strategy, IT-enabled combinative capabilities, and competitive performance outcomes. The subsequent section is devoted to explaining each variable. Table 1 summarizes the definitions of the variables in this research model.


3

Fig. 1. A research model.

ERP implementation strategy Over the years, the scope of ERP systems has expanded. Once perceived only as an operational tool, it has now evolved into a more comprehensive approach to developing a strategic infrastructure (i.e., adapting, reconfiguring, and integrating the flow of information and business processes (Davenport, 2000; Piccoli and Ives, 2005; Koh et al., 2011). ERP system implementation is both technical (e.g., it requires a change in technology systems) and behavioral (e.g., it requires a change in work concepts, habits, and routines) (Bradley, 2008; Finney and Corbett, 2007). Deriving from the dynamic capability perspective, ERP implementation strategy is thus defined as a firm’s organization-wide long-term initiative to enhance information flows and business processes in terms of adaptation, configuration, integration, and user training. Adaptation is essential to custom software as well as off-the-shelf software packages (Markus and Tanis, 2000). Adaptation processes iteratively adjust organizational structures and ERP systems through continuous engagement (Hong and Kim, 2002; Sambamurthy et al., 2003; Zhu et al., 2010). Configuration refers to preparing ERP systems to meet specific business requirements (Davenport, 2000). In addition, reengineering existing business processes to conform with ERP systems requires corresponding changes in other organizational components, such as the organizational structure and the organizational culture, and as a result, ‘‘organizational fit’’ is an important measure for configuration success (Soh et al., 2000; Hong and Kim, 2002). No ERP implementation may perfectly and flawlessly match users’ requirements to external contexts. Organizations with higher levels of integration are likely to achieve higher performance outcomes (Vonderembse et al., 1997). The benefits of ERP systems are enhanced as they are seamlessly connected with other information systems. Challenges in ERP integration include (a) seamlessly combining various functional ERP modules, (b) connecting with other business software applications, and (c) upgrading legacy systems with the relevant IT systems of other suppliers. User training equips people to work well with different business processes during and after ERP implementation (Soja, 2006). IT-enabled combinative capabilities Combinative capabilities within manufacturing strategies emphasize synergistic interdependence rather than uniform standardization (Rosenzweig et al., 2003; Flynn and Flynn, 2004; Kristal et al., 2010). Combinative capabilities refer to ‘‘a manufacturer’s ability to excel simultaneously on quality, delivery, flexibility, and low cost’’ (Kristal et al., 2010, p. 417). The ‘‘combinative’’ nature of


4

Table 1 Definition of variables and literature base. 2nd order construct

Variables

Definition

Literature base

ERP implementation strategy

Adaptation

The degree to which a firm accepts and adjusts new technology and systems in response to changing requirements of the external environments The degree to which a firm matches the software application packages to organizational processes The degree to which a firm achieves unity in organizational subsystems by synchronizing different departments, modules, software, and legacy systems The degree to which a firm develops and prepares its IT workforce and relevant personnel to reap the potential benefits of ERP system The degree to which a firm achieves goal congruence and consistent action across all departments and work functions The degree to which a firm enhances existing programs and procedures within its organization The degree to which a firm adapts to dynamic and unpredictable business environment in terms of business opportunities and competitive threats The degree to which a firm supports its manufacturing processes through fast data gathering, retrieving, and processing The degree to which a firm produces and delivers its products and services at competitive prices The degree to which a firm offers products/services that are dependable, consistent and conformant to high standards The degree to which a firm introduces new products fast as planned or ahead of competitors The degree to which a firm introduces new products and services with additional features, improved performance, and diverse offerings

Hong and Kim (2002), Sambamurthy et al. (2003), Zhu et al. (2010)

Configuration

Integration

User training

IT-enabled combinative capabilities

Cross-functional coordination

Process improvement Agility

Information access

Competitive performance outcomes

Cost performance Quality

Time to market

Product variety

Davenport (2000), Klein (2007)

Vonderembse et al. (1997), King and Flor (2008)

Doll et al. (1994), Soja (2006), Finney and Corbett (2007)

Eng (2006), Zhang (2005), Carr et al. (2008), Roh et al. (2011), Nevo and Wade (2010) Ravichandran and Rai (2000), Peng et al. (2008), Browning and Heath (2009) Zhang (2005), Lee et al. (2007), Sambamurthy et al. (2003), Piccoli and Ives (2005)

Premkumar et al. (2005), Klein (2007), Moorman and Miner (1997) Nicolaou et al. (2003), Krause et al. (2007) Shin et al. (2000), Ahire and Dreyfus (2000), Krause et al. (2007), Kim et al. (2012), Oliva and Watson (2011) Nahm et al. (2003), Griffin (1997), Hong et al. (2011), Doll et al. (2010) Da Silveira (1998), Lee and Tang (1997), Doll et al. (2010), Patel and Jayaram (2014)

organizational capabilities helps to meet complex competitive requirements simultaneously (Kettinger et al., 1994; Hayes et al., 2005). This also helps to overcome the ‘‘trade-off’’ between efficiency and responsiveness (Rossin, 2007). For example, customer requirements include efficiency (e.g., cross-functional coordination and process improvement) and responsiveness (e.g., agility and information access). IT-enabled combinative capabilities refer to a firm’s unique set of abilities to achieve complex customer requirements simultaneously through IT-based enhancement of organizational resources in the form of cross-functional coordination, process improvement, agility and information access (Peng et al., 2008; Nevo and Wade, 2010).


5

Cross-functional coordination indicates the extent to which intra- and inter-organizational work collaboration is required to achieve complex performance outcomes that particular functions alone may be unable to achieve – particularly in uncertain and turbulent business environments (Carr et al., 2008; Doll et al., 2010; Park and Hong, 2012). Generally, firms expect that new ERP-based system environments facilitate waste reductions in terms of removing delays, paperwork, and time-wasters (Browning and Heath, 2009; Youn et al., 2013). Process improvement covers all vital business processes, including product development, sourcing, logistics, production scheduling, and customer service (Zacharia et al., 2011; Yang et al., 2011). Organizational agility refers to the capabilities of firms to detect environmental changes and respond to them by utilizing available resources (Sambamurthy et al., 2003; Piccoli and Ives, 2005). Information access refers to the ways in which firms with effective ERP systems offer up-to-date data and information required to make decisions and provide managerial control. Such information access facilitates knowledge sharing, the redesign of business processes, workflow improvements, and innovative problem solving – all of which lead to desirable performance results (Bharadwaj, 2000; Hong et al., 2011). Competitive performance outcomes Competitive performance outcomes refer to key organizational process results measured in terms of cost, quality, time-to-market, and values (Doll et al., 2010; Hong et al., 2011). IT investment is expected to improve the performance of firms in the areas of cost, quality, delivery, product variety, and time to market (Nahm et al., 2003; Krause et al., 2007; Youn et al., 2014). By strategically deploying ERP systems, firms intend to leverage their integrative features and functionality to improve overall operational effectiveness – from product development, sourcing and procurement all the way through manufacturing, quality testing, and delivery. Firms use ERP systems to achieve cost performance in the areas of marketing, manufacturing, purchasing, inventory, and customer service through improved process efficiencies and waste reductions (Nicolaou et al., 2003; Krause et al., 2007). Firms utilize ERP systems to achieve quality requirements – a process that prevents product defects and identifies customer service issues in advance (Shin et al., 2000; Krause et al., 2007). ERP systems identify actual product defects and service failures, enabling firms to provide excellent quality management. Improving business processes, including marketing plans, product development decisions, and operational executions, is crucial if firms are to achieve desirable time-to-market performance (Doll et al., 2010; Hong et al., 2011). In view of dynamic and complex customer requirements, it is essential for firms to offer a high level of product variety, which can be accomplished by introducing and delivering products and services that feature sufficient novelty, choices, and customer value (Lee and Tang, 1997; Patel and Jayaram, 2014). Hypothesis development Three main hypotheses are based on the above-mentioned three constructs (i.e., ERP implementation strategy, IT-enabled combinative capabilities, and competitive performance outcomes). Linking ERP implementation strategy to IT-enabled combinative capabilities ERP implementation strategies facilitate cross-functional coordination (Howcroft and Truex, 2001; Hedman and Borell, 2003). ERP system adaptation enables firms to manage inventory and administrative costs while also improving customer responsiveness (Lee and Tang, 1997; Horvath, 2001). ERP systems configure information flows in the supply chain and reduce the ‘‘bullwhip effect’’ between suppliers and customers (Lee and Tang, 1997). Shang and Seddon (2002) suggest that the integration aspects of ERP systems can be helpful in achieving business goals across strategic, tactical, and operational levels. ERP systems integrate essential business processes while simultaneously enhancing functional performance in the areas of human resources, marketing, accounting, and operations. With appropriate user training, the effective use of ERP systems has a positive influence on sales and customer service (Mentzer et al., 2000).

6


ERP implementation strategies enhance process improvement. ERP system adaptation promotes information sharing with customers, suppliers, and other business partners (Sambamurthy et al., 2003). ERP system configuration allows firms to modularize their business processes (Park et al., 2012). Integrating sequential information flows through ERP systems (e.g., market opportunities scanning, product concept generation, and project definition) improves the entire new-product innovation processes. User training allows firms to eliminate redundant procedures and thus achieve process improvement (Doll et al., 1994; Harkness et al., 1996). ERP implementation strategies enhance organizational agility. ERP system configuration allows firms to sense opportunities for new competitive action in their marketplaces and seek the necessary knowledge and assets to seize those opportunities. Because ERP systems integrate diverse functional contexts, firms are able to flexibly assemble requisite assets, knowledge, and business relationships (Sambamurthy et al., 2003). This adaptable ERP process, along with adequate user training, allows firms to detect environmental changes (Sambamurthy et al., 2003). ERP implementation strategies nurture greater information access. ERP adaptation allows firms to assemble necessary information (Sambamurthy et al., 2003). ERP system configuration facilitates timely information access for functional specialists. ERP system integration allows rich information access from various information portals (Park et al., 2013; Youn et al., 2014). Employees with adequate user training have greater accessibility to the resources, ideas, and insights of other employees from different departments. Thus, successfully implemented ERP systems enhance IT-enabled combinative capabilities. Therefore, it is hypothesized that: H1. An ERP implementation strategy has a significant positive influence on IT-enabled combinative capabilities. Linking IT-enabled combinative capabilities to competitive performance outcomes Wernerfelt (1984) and Barney (1991) suggest that firms develop organizational resources and capabilities to manage their environments and enhance performance. A firm’s resources and capabilities include tangible and intangible factors, such as physical assets, human capital, and organizational routines and procedures. In this section, for the purpose of this study, the impact of four essential IT-enabled combinative capabilities (i.e., cross-functional coordination, process improvement, agility, and information access) on competitive performance outcomes (i.e., cost, quality, time to market, and product variety) is examined. IT-enabled combinative capabilities improve firms’ cost performance. Cross-functional coordination provides firms with the ability to use IT resources to produce more with less (Sambamurthy et al., 2003; Zhang, 2005). Process improvements allow firms to save costs by improving specific processes, such as logistics and production scheduling (Piccoli and Ives, 2005; Doll et al., 2010). Agility provides firms with the ability to respond and to recover from business disruptions, thus saving processing costs (Park et al., 2013). Information access allows organizational members to eliminate unnecessary waiting time (Park et al., 2012; Youn et al., 2014). IT-enabled combinative capabilities affect quality performance. Cross-functional coordination allows firms to meet multiple quality requirements through excellent product concept and design services as well as diverse service offerings (Kim et al., 2012; Oliva and Watson, 2011). Process improvement facilitates design management and operational execution and, as a result, influences quality (Ahire and Dreyfus, 2000; Bala, 2013). Agility allows firms to enhance organizational responsiveness in times of environmental turbulence to the satisfaction of customers (Kristal et al., 2010; Inman et al., 2011). IT-enabled combinative capabilities are related to time to market. Cross-functional coordination encourages collaboration and cooperation among diverse product development team members (Anderson and Narus, 1990). Process improvement helps overcome glitches and facilitates early introduction of new products (Carr et al., 2008; Rauniar et al., 2008). Highly agile firms can rapidly seize opportunities for new competitive action in their marketplaces by assembling organizational resources (Sambamurthy et al., 2003), allowing these firms to make changes more rapidly. Information access stimulates new ideas, and such access helps remove unnecessary steps in existing processes (Moorman and Miner, 1997).


7

IT-enabled combinative capabilities are useful for product variety. Cross-functional coordination helps design engineers, marketing managers, and operational managers to achieve product variety, which leads to broad customer satisfaction. Process improvement contributes to the ability of firms to achieve product variety requirements more effectively (Swink et al., 2006). Agility refers to the unique qualities of firms that contribute to their ability to deliver product variety even in very disruptive environments (Park et al., 2012; Patel and Jayaram, 2014). Information access allows firms to quickly identify customer product variety requirements (Thatte et al., 2008; Youn et al., 2014). Based on the above arguments, the following assertion is therefore hypothesized: H2. IT-enabled combinative capabilities have a significant positive influence on competitive performance outcomes. The mediating effect of IT-enabled combinative capabilities Based on two hypotheses (H1 and H2), it is argued that IT-enabled combinative capabilities mediate the relationship between ERP implementation strategies and competitive performance outcomes. In the management literature, the resource-based view argues that firms compete on the basis of ‘‘unique’’ corporate resources that are valuable, rare, difficult to imitate, and non-substitutable by competitors (Barney, 1991). These unique resources, such as ERP systems, enable firms to achieve competitive advantages and superior long-term performance. While we understand this logic, ERP implementation strategies can hardly contribute to performance outcomes directly because strategic initiatives influence performance outcomes only to the extent that they are translated into organizational-level capabilities (Hong et al., 2011; Youn et al., 2014). In order for competitive performance outcomes to be realized, therefore, unique resources, such as ERP implementation strategies, must be integrated with efficient and responsive organizational IT capabilities. Only when IT-enabled combinative capabilities are in place can firms receive tangible competitive performance benefits, such as cost reduction, quality improvement, and a wide variety of new products (Madapusi and D’Souza, 2012). Therefore, the following is hypothesized: H3. IT-enabled combinative capabilities mediate the relationship between an ERP implementation strategy and competitive performance outcomes.

Research method The current research was implemented in three phases to test the above hypotheses: the first phase was to use a pre-pilot method; the second phase was to conduct and examine a pilot study; and the last phase was to conduct a large-scale survey. In the pre-pilot phase, potential survey items were generated through a comprehensive literature review and theory development. In the pilot phase, we conducted structured interviews with two academicians and four practitioners from different manufacturing firms who have had extensive experience implementing ERP systems. Based on their comments and feedback, redundant or ambiguous items were either modified or eliminated. We used the Q-sort method to observe how the items were sorted into various sub-construct categories. Items placed in a common pool were subjected to two or three Q-sorting rounds by two independent judges per round. Based on the results, we modified or deleted inappropriate or ambiguous items. In the last stage, a large-scale survey was conducted via mail questionnaires primarily targeting Korean industries comprised of manufacturers and their suppliers. Using the data obtained from this survey, we employed the AMOS software to test the validity of the proposed SEM. Specific details of the current research methodology are described below. Sample characteristics A total of 62 questionnaire items were distributed to academic reviewers, who reviewed all the items and indicated whether they wanted to keep, delete, or modify them. The purpose of this analysis


8

was to assess (a) whether the items were thought to accurately measure the proposed sub-constructs according to the definitions provided and (b) whether any additional domains needed to be covered. After deleting and purifying a number of items based on the feedback from the reviewers, 40 items were used in the large-scale questionnaire. Via e-mail, a questionnaire containing these 40 items was distributed to 593 publicly traded Korean manufacturing firms listed on two major Koran stock markets, KOSPI (Korean Composite Stock Price Index) and KOSDAQ (Korea Securities Dealers Automated Quotation) (see http://finance.daum.net/). The typical respondent to the questionnaire held the title of Manager/Director of Operations or Manager/Director of Information Technology. More than 88% of the respondents actively used ERP systems in their firms at the time the questionnaire was distributed. We targeted Korean manufacturing firms in eight different sectors in order to increase variability in the data and generalizability of the survey results (see Table 2). Based on a careful screening of their SIC Code (i.e., industry type), we focused on potential respondents who are familiar with ERP systems. For this reason, any services or logistics sectors were purposely excluded from the list. Data collection occurred over the three months (August through October) in 2010. Of the 593 questionnaires that were distributed to Korean manufacturing firms in which ERP systems are implemented, 205 valid responses were received. These responses produced a total response rate of 34.6%, which surpasses the targeted overall response rate of over 20% in order to conduct a valid assessment (Malhotra and Grover, 1998). More than two thirds (67%) of the responding firms employed more than 250 employees (see Table 2). More than half (51%) of the respondents reported Table 2 Description of sample. (1) Respondents by SIC code SIC code

Name

20 26 28 32 34 35 36 37

Food and kindred products Paper and allied products Chemicals and allied products Stone, clay, glass, and concrete products Fabricated metal products, except machinery and transportation Industrial and commercial machinery and computer equipment Electronic and other electrical equipment and components, Transportation equipment

Frequency

Total

Percent

16 12 41 14 37 29 29 27

8% 6% 20% 7% 18% 14% 14% 13%

205

100%

(2) Respondents by position Position Directors General manager Deputy general manager Managers Assistant manager Staff Total

Frequency

Percent

2 20 32 94 54 3

1% 10% 16% 46% 26% 1%

205

100%

(3) Firms by size Number of employees Less than 100 100–249 250–499 500–999 1000–2499 2500 and over Total

Frequency

Percent

16 51 53 41 21 23

8% 25% 26% 20% 10% 11%

205

100%


9

that the level of complexity of the products in their dominant product line was above average (‘‘high’’ – 38% or ‘‘very high’’ – 15%). More than one third (38%) of the respondents represented manufacturing firms with moderate product complexity. The number of product variants that a firm produces reflects product complexity. Since the degree of product complexity is tied to the complexity of working environments, the firms with a high level of product complexity are likely to utilize ERP systems. Thus, these respondents provide an adequate basis upon which to examine the research hypotheses. Non-response bias test Considering the potential non-response error associated with questionnaires, we conducted a chisquare test of homogeneity for non-response bias by comparing the SIC group distribution for the sample population and total responses (Armstrong and Overton, 1977). As summarized in Table 3, there were no statistically significant differences among the group means for the eight different industry samples at a = 0.05 on any of the item responses described earlier. Therefore, non-response bias did not appear to be a concern. Common method bias test Given that single respondents were used for collecting data, the potential for common method bias to influence the results needs to be evaluated. Harman’s single-factor test using confirmatory factor analysis was conducted to test the hypothesis that a single factor accounts for all the variance in the data. The model fit indicates that a single factor model does not represent the data well (x2 = 3976.83, df = 740, GFI = 0.398, CFI = 0.520, IFI = 0.523, NNFI = 0.494, RMSEA = 0.146, SRMR = 0.122). Further, the average variance extracted by a single factor is 38.5%, indicating that a very small proportion of the variance in the data is accounted for by a single factor. While this test does not help preclude the possibility of method bias, it helps mitigate concerns that common method bias may be driving our results. Control variables Three control variables (i.e., firm size, product complexity, and process integration) are included in our analyses. Number of employees was used to measure firm size. Product complexity was defined as ‘‘the degree of product complexity.’’ Process integration was measured through the type of operations. The first control variable is manufacturing business unit size. It is identified as one of several important organizational characteristics that influence competitive performance outcomes (Chen and Hambrick, 1995). Company size matters in ERP system adoption (Laukkanen et al., 2007). ERP benefits are different according to firm size (Mabert et al., 2003). Larger firms experience more ERP benefits

Table 3 A Chi-square test of non-response biases. SIC

Total sample distribution

20 26 28 32 34 35 36 37

0.061 0.040 0.216 0.052 0.185 0.196 0.103 0.147

Response 17 13 41 14 36 28 29 27

Total

1.000

205

Expected frequency 12 8 44 11 38 40 21 30 205.00

Chi-square 1.67 2.67 0.24 1.01 0.11 3.65 2.97 0.31 12.62

Note. Sample population and response group are homogeneous in SIC code distribution at degree of freedom (df) = 7, a = 0.05 (Chi-square critical value = 14.067).

10


(Laukkanen et al., 2007). In light of its potential impact on organizations, it is appropriate to include manufacturing business unit size as a control variable. The second control variable is process integration. Managers at leading companies identify satisfying the real needs of customers as the key to competitive success. For higher levels of customer satisfaction, firms develop their core competencies. Process integration is essential for improving organizational capabilities and increasing competitive performance outcomes (Scheer and Habermann, 2000). Firms with integrated business processes can implement ERP systems more successfully (Ross and Vitale, 2000). Because of its potential impact on organizations, we introduced another control variable. This control variable measures the level of production process integration (job shop, projects, batch processing, manufacturing cells, assembly line, flexible manufacturing, highvolume discrete part production, and continuous flow process). The third control variable is product complexity. In a manufacturing firm, product complexity increases with the number of product components and their sub components. Increased product complexity requires higher inventory levels and costs. Increased product complexity increases inventory and decreases service levels (Lee and Tang, 1997). It has an impact on organizational capabilities, such as process improvement and agility. An increased number of products require additional time for ERP system implementation as well. Thus, product complexity is introduced as a control variable. Level of complexity was measured with 5-point Likert scale item with the following response options: very low, low, moderate, high, and very high.

Research results To examine causal relationships among variables, we tested the three proposed hypotheses with valid and reliable scales that measured some critical dimensions of ERP system implementation strategies (ERPIS), IT-enabled combinative capabilities (ITCC), and competitive performance outcomes (CPO). A structural equation modeling (SEM) framework was used to explore possible relationships among the constructs and to test the hypotheses (Bollen, 1993). In general, the SEM is composed of two elements: (1) the measurement model and (2) the structural model. The measurement model in SEM is used to measure and assess the reliability and validity of latent variables, whereas the structural model is applied to investigate the complex interrelations among latent variables (Jo¨reskog and So¨rbom, 1989). Since the reliability and validity of each construct were checked earlier through rigorous analysis, the SEM analysis focused on the structural model. To explore relationships among the ERPIS, ITCC, and CPO, AMOS software was used. Since it would be better to use several indicators of a construct than a single indicator, we used composite measures as multiple indicators for each construct (Hair et al., 1995). Composite measures were calculated by dividing the sum of individual scores of items in each sub-construct by the number of items. These composite measures were used as observable indicators of the exogenous latent construct (ERPIS) and endogenous latent constructs (ITCC and CPO). Measurement model, validity, and reliability In the first step of the analysis, we tested the measurement model and established the validity and reliability of the items using confirmatory factor analysis. ERPIS, ITCC, and CPO are hypothesized as a second-order construct. Therefore, we conducted this first step analysis in three stages. In stage 1, we conducted confirmatory factor analysis at the first-order level for all the constructs in our model – this measurement model is referred to as the first-order measurement model. In stage 2, we validated the second-order specification for the sub-dimensions of the ERPIS, ITCC and CPO. In the third and final stage, we conducted confirmatory factor analysis of all the items representing ERPIS, ITCC, and CPO as second-order constructs – the measurement models are referred to as the second-order measurement models. Table 4a presents the fit statistics for the first- and second-order measurement models. Table 4b presents the fit statistics for the ERPIS, ITCC, and CPO measurement models to help validate the second-order specifications of all three second-order constructs. Item loadings (both the first- and second-order models) are presented in Tables 5a and 5b. Tables 6a and 6b present the inter-factor


11

Table 4a Fit statistics for validating the measurement model (n = 205). Fit statistic

Measurement model (first-order)

Measurement model (second-order)

x2

1093.034 758 0.047 0.816 0.951 0.941 0.952 0.045

1206.103 836 0.047 0.800 0.946 0.941 0.946 0.0600

df RMSEA GFI CFI NNFI IFI SRMR a b c

Recommended values

0.08 marginal fitc; 0.05 good fita,b >0.8 marginal fit; >0.9 good fitb

0.8 marginal fit; >0.9 good fitb SRMR: