Description
A Study on Black Box In Strategic HRM Research From Resource-Based And Social Exchange Perspectives, Social exchange theory is a social psychological and sociological perspective that explains social change and stability as a process of negotiated exchanges between parties. Social exchange theory posits that all human relationships are formed by the use of a subjective cost-benefit analysis and the comparison of alternatives. The theory has roots in economics, psychology and sociology. Social exchange theory features many of the main assumptions found in rational choice theory and structuralism.
A STUDY ON BLACK BOX IN STRATEGIC HRM RESEARCH FROM RESOURCE-BASED AND SOCIAL EXCHANGE PERSPECTIVES
Abstract:In this dissertation, I examine the specific mediating mechanisms through which High Performance Work Systems influence overall unit performance. In particular, I draw mainly on two theoretical perspectives, the resource-based view of the firm and behavioral perspective, to propose and test the mediated model of strategic human resource management. The data were collected from 322 managers and 526 employees for a sample of 76 units of Japanese companies in various industries. The data were aggregated to the
unit-level of analysis. On one hand, the data provided support for many of the hypotheses advanced in the dissertation. The results of the hierarchical regression analyses revealed that High Performance Work Systems was positively related to both the level of human capital that the unit possesses and the collective normative contract that the employees working in the unit share. In addition, the level of human capital and collective normative contract were significantly related to most of the HR-related outcomes, which, in turn, were significantly related to overall unit performance. Moreover, as hypothesized, the level of human capital acted as mediators of the relationship between High Performance Work Systems and HR-related outcomes while HR-related outcomes played the role of mediator on the relationships between the level of human capital or collective normative contract and overall unit performance. On the other hand, the mediating hypothesis for collective normative contract as well as interaction hypothesis for the level of human capital* collective normative contract were not supported. The implications of the findings and future research directions are also discussed.
TABLE OF CONTENTS List of Tables List of Figures vi vii
Chapter 1: PROBLEM STATEMENT 1-7 Research Questions??????????????????????7 Chapter 2: THEORETICAL REVIEWS 8-24 Resource-Based View of the Firm????????????????..8 Behavioral Perspective????????????????????...12 Understanding employee attitudes and behaviors: Social exchange perspective???????????????...14 Integrating resource-based view and behavioral perspective???.18 Conceptualization of performance?????????????..20 Review of Past Research Related to Framework???????????22 Chapter Summary??????????????????????...24 Chapter 3: MODEL DEVELOPMENT AND HYPOTHESES 25-46 Scope of the Research?????????????????????.25 Level of analysis????????????????????.25 Unit performance indicators????????????????26 Employee groups????????????????????27 Research Model???.????????????????????.29 High Performance Work System, Level of Human Capital, and Performance????????????????????.30 High Performance Work System, Collective Normative Contract, and Performance????????????????????.34 High Performance Work System and collective normative contract?????????????...35 Collective normative contract and HR-related outcomes????...39 Moderating Effects of Collective Normative Contract????????...43 Chapter 4: RESEARCH METHODS 47-64 Sample???????????????????????????47 Data Collection Procedure..????.??????????????...49 Survey Translation Procedure????????????..?????..51 Aggregation Issues??????????????????????.52 Measures?????????????????????????? 55 High Performance Work Systems?????????????..56 The level of human capital????????????????..58 Collective normative contract???????????????.58 HR-Related Outcomes?????????????????????58 Perceived labor productivity???????????????..58 Unitlevel in-role or task performance??..????????.?59
iv
Unit-level overall organizational citizenship behaviors..????.?60 Unit Performance??????????????????..???.?...60 Analytic Procedures??????????????..????.???...62 Chapter 5: RESULTS 65-84
Chapter 6: DISCUSSION 85-100 Limitations??????????????????????????92 Implications for Research and Practice???????????????.96 Conclusion??????????????????????????.99 APPENDIX A APPENDIX B APPENDIX C REFERENCES 101-107 108-112 113-115 116-146
v
LIST OF TABLES
1. Changes in the Strategic HRM Perspective from a "Traditional" HRM Perspective 2. Summary of Dissertation Hypotheses 3. Response Patterns 4. Factor Loadings for 21-Item High Performance Work System Scale 5. Results of Exploratory Factor Analysis on the HR-Related Outcomes and Unit Performance 6. Descriptives of All the Variables 7. Regression Results for Industry Effect 8. Regression Results for Hypothesis 1 and 2 9. Regression Results for Hypothesis 4 10. Regression Results for Hypothesis 5 and 6 11. Regression Results for Hypothesis 7 12. Regression Result for Hypothesis 8 13. Regression Result for Hypothesis 9 14. Path Analytic Regression Analyses Results 15. Summary Results for the Hypotheses 62 67 68 70 73 74 75 77 79 80 83-84 3 45-46 47-48 57
vi
LIST OF FIGURES
1. The Black Box in Strategic HRM Research 2. Resource-Based View of the Firm in Strategic HRM 3. Behavioral Perspective in Strategic HRM 4. Social Exchange Theory for Strategic HRM 5. Integration of the Resource-Based View of the Firm and Behavioral Perspective on Strategic HRM 6. Overall Theoretical Framework 7. Proposed Model of High Performance Work System 8. Path Analytic Results
4 12 13 15
19 22 30 82
vii
Chapter 1 Problem Statements The introduction of strategic HRM has advanced our understanding of the relationships among strategy, human capital, human resource management (HRM), and firm performance by shifting the focus from "traditional" HRM in several ways. First, the level of analysis has shifted from a traditional micro focus on examining individual HRM functions (i.e., selection/recruitment, training and development, compensation, performance appraisal, job design, etc.) to a firm level of analysis by adopting a system approach to HRM practices. This involves examining the entire system of HRM practices as a whole, rather than examining individual subfunctional practices in isolation. As a result, the notion of synergy or complementarity of HRM practices became particularly important (Baird & Meshoulam, 1988; Becker & Huselid, 1998; Huselid, 1995; Wright & Snell, 1998). For example, MacDuffie (1995: 198) noted "an HR bundle or system must be integrated with complementary bundles of practices from core business functions" and examined the interaction between flexible production and human resource capabilities that included HRM practices such as recruitment and hiring, contingent compensation, status differentiation, and training. This notion of internal fit or alignment of different HRM practices is less salient in the traditional, subfunctional perspective to HRM. Second, strategic HRM focuses on higher-level contingency variables. Underlying this perspective is the argument that the effectiveness of HRM systems depends on how they are aligned with variables external to the HRM system such as strategy, technology, environmental conditions, and the like (Baird & Meshoulam,
1
1988; Wright & Snell, 1998). As an example, Youndt, Snell, Dean, and Lepak (1996) examined the relationship between manufacturing strategy, human resource management practices, and operational performance (employee productivity, machine efficiency, and customer alignment), and found a moderating effect of manufacturing strategy on the relationship between the use of a human-capital-enhancing HR system and operational performance. As exemplified by Youndt et al. (1996), strategy is considered as one of the most important contingent factors that HRM system needs to be aligned with (e.g., Arthur, 1994; McMahan, Virick, & Wright, 1999; Wright & McMahan, 1992). In traditional HRM approaches, the notion of fit tends to focus on individual employees and their fit with the internal surroundings (e.g., person-job fit or person-organization fit) rather than broader environmental factors. Third, strategic HRM research places emphasis on firm-level performance as the dependent variable rather than focusing solely on individual performance (cf. Rogers & Wright, 1998; Wright, 1998). For instance, Delery and Doty (1996) used return on asset (ROA) and return on equity (ROE) and Huselid (1995) used gross rate of return (GRATE) and a variant of Tobin's Q (i.e., firm market value/book value) as their dependent variables. In contrast, traditional HRM research generally focused on individual outcome behaviors such as task performance (cf., Locke & Latham, 1990), absenteeism (cf., Harrison & Martocchio, 1998), withdrawal behaviors (cf. Harrison & Martocchio, 1998; Hulin, 1991), and/or turnover (cf., Griffeth, Hom, & Gaertner, 2000). In general, the basic distinction between traditional HRM and strategic HRM can be summarized as the difference in the focus on micro- vs. macro-perspectives. Table 1 highlights these differences.
2
Table 1. Changes in the Strategic HRM Perspective from a "Traditional" HRM Perspective Level of Analysis "Traditional" HRM Subfunctions of HRM (Subfunctional-level: staffing, training & development, compensation & incentives, performance appraisal, etc.) Individual HRM practices (e.g., staffing practices [Terpstra & Rozell, 1993], training practices [Russell, Terborg, & Powers, 1985], and compensation practices [Gerhart & Milkovich, 1990]) Strategic HRM HRM as a whole (Organizational-level)
Focus
Contingencies Person-Environment Fit (e.g., Dawis, 2000; Edwards & Van Harrison, 1993), PersonOrganization Fit (e.g., Kristof, 1996; Van Vianen, 2000), Person-Job Fit (e.g. KristofBrown, 2000) Performance Individual employee performance
System or Configuration of HRM practices (e.g., "commitment" [Arthur, 1994], "human capital enhancing" [Youndt et al., 1996], "High Performance Work System" [Becker & Huselid, 1998; Huselid, 1995]) External Fit (Strategy-HRM), Internal Fit (HRM-HRM)
Organizational performance
Microperspective
Macroperspective
Despite this progress, one of the critical issues that remain unclear is an understanding of the mediating mechanism or process through which HRM practices influence firm performance. Interestingly, there is no empirically based research that has examined critical employee attitudes and behaviors in strategic HRM and linked them to other outcomes such as turnover rate, labor productivity, and so forth. Moreover, only a limited number of researchers have articulated potential mediators
3
only to a limited extent (Applebaum, Bailey, Berg, & Kalleberg, 2000; Delery, 1998). As Figure 1 illustrates, there is a "black box" in empirical examinations of HRperformance relationship where the mediating mechanism is typically implied but not measured to date. It is understandable that previous examinations of strategic HRM research have focused on investigating the overall linkages between HRM and performance, given the relative infancy of the field. However, advances in the strategic HRM area are impeded because linkages between HR practices and performance are tenuous at best without an understanding of the mediating processes though which HRM practices affect performance. In addition, the utility of strategic HRM research for practitioners is limited without explicating the processes that occur within this black box. FIGURE 1. The Black Box in Strategic HRM Research Firm strategy
A system or a bundle of HRM practices
Firm performance
Mediating mechanism or process With regard to this black box, one of the dominant, theoretical perspectives that have been used in the strategic HRM literature is the behavioral perspective, advanced by Schuler and Jackson (1987; Jackson et al., 1989). "A behavioral perspective assumes that employers use personnel practices as a means for eliciting and controlling
4
employee attitudes and behaviors" (Jackson et al., 1989: 728). The primary logic for this is that certain attitudes and behaviors are needed for certain strategies. An example of the behavioral perspective can be found in Miles and Snow's (1984) description of the employee behaviors needed for different strategic business types: defender, prospector, analyzer, and reactor. An important component of the behavioral perspective is an emphasis on the employee attitudes and behaviors that contribute to effective implementation of strategic objectives and organizational functioning. While conceptual arguments abound, however, none of these studies have actually examined employee attitudes and behaviors espoused to be critical to this process. Thus, previous empirical examinations of strategic HRM have not tested these theoretical assertions that employee attitudes and behaviors are critical in the effective implementation of different strategies. In addition, the behavioral perspective has not explicitly articulated the types of critical employee behaviors that lead to higher firm performance. In short, a key mediator of the strategic HRM research is neither well understood nor tested. In addition to the behavioral perspective, some strategic HRM researchers focused more on the competency of employees. Deriving from the resource-based view of the firm (e.g., Barney, 1991), these researchers suggest that firms' internal resources (human capital) directly influence firm performance (Wright & McMahan, 1992). Barney (1991) asserts that human capital may be one of the few firm resources that have the potential for sustainable competitive advantage because it may satisfy the four characteristics (i.e., value, rarity, inimitability, and nonsubstitutability) that resources must possess to provide firms with sustainable competitive advantage. While there is
5
empirical evidence that illustrates the positive influence of human capital on firm performance, this line of research typically de-emphasizes the role of HRM activity in obtaining human capital or eliciting the necessary behaviors. As noted by Austin, Villanova, Kane, and Bernardin (1991), performance is a function of ability (or in an aggregate form, human capital) and motivation (or effort) and both are needed for higher performance. As such, the sole use of resource-based view in strategic HRM might have been based on a faulty assumption that ability necessarily translates into performance without considering the other side of the equation - motivation as shown by the employees exhibiting desired attitudes and behaviors. Despite different conceptualization of the functional form of individual performance (as an additive model where ability and motivation influences performance independently or interactive model where ability and motivation influences performance in a multiplicative manner) (Austin et al., 1991), both ability and effort are instrumental in achieving higher performance. In other words, human capital may provide the potential for higher firm performance while employee attitudes and behaviors may be instrumental in realizing that potential. Taken together, both resourcebased view of the firm and behavioral perspective suggest that the attributes (human capital) and efforts (as exemplified by aggregate employee attitudes and behaviors) are critical components in understanding firm performance. However, previous research has not integrated these two perspectives to examine the relationships among HRM, human capital, employee attitudes and behaviors, and firm performance. Hence, the primary research questions in this dissertation focus on addressing these issues.
6
Research Question 1: What are the roles of human capital and employee attitudes and behaviors on the relationship between human resource management practices and unit performance. Research Question 2: What are the processes/mechanisms through which a system of human resources management impact unit performance?
7
Chapter 2 Theoretical Reviews In this chapter I will draw from resource-based view of the firm (Barney, 1991; Wernerfelt, 1984), the behavioral perspective (Schuler & Jackson, 1987), and social exchange theory (Blau, 1964; Gouldner, 1960; Homans, 1958, 1961) to develop a parsimonious research framework for this dissertation. Resource-Based View of the Firm (Barney, 1991; Wernerfelt, 1984) One of the major theoretical perspectives that strategic HRM researchers have adopted is the resource-based view of the firm. "The resource-based view of competitive advantage differs from the traditional strategy paradigm in that the emphasis of the resource-based view of competitive advantage is on the link between strategy and the internal resources of the firm" (Wright & McMahan, 1992: 300). The resource-based view of competitive advantage makes a couple of fundamental assumptions: 1) resource heterogeneity, i.e., not all firms have access to or possess the same resources, and 2) resource immobility, i.e. some resources are harder to transfer or purchase from the market. According to the resource-based view of the firm (Barney, 1991; Wernerfelt, 1984), "a firm is said to have a competitive advantage when it is implementing a value creating strategy not simultaneously being implemented by any current or potential competitors. A firm is said to have a sustained competitive advantage when it is implementing a value creating strategy not simultaneously being implemented by any current or potential competitors and when these other firms are unable to duplicate the benefits of this strategy" (Barney, 1991: 102).
8
Firm resources can be defined as "all assets, capabilities, organizational processes, firm attributes, information, knowledge, etc. controlled by a firm that enable firms to conceive of and implement strategies that improve its efficiency and effectiveness (Daft, 1983)" (Barney, 1991: 101). Moreover, firm resources can be broadly classified into four categories: physical capital, organizational capital, social, and human capital. Physical capital refers to assets that are tangible and include the physical technology used in a firm, a firm's plant and equipment, its geographical location, and its access to raw materials. Organizational capital is those intangible assets, including a firm's reporting structure, its formal and informal planning, controlling, and coordinating systems, as well as informal relations among groups within a firm and between a firm and those in its environment whereas social capital refers to the specific component of intangible asset that are based on ability and capability to build and maintain good relationships between the firm and other constituents such as stockholders, customers, other organizations, and employees (Sullivan, 2002). Human capital includes the training, experience, judgment, intelligence, relationships, and insights of individual managers and workers in a firm (Barney, 1991). While all four resources have the potential for competitive advantage, firm resources must possess four attributes, 1) value, 2) rarity, 3) inimitability, and 4) nonsubstitutability, to realize a sustained competitive advantage. A firm resource adds value when it exploits opportunities and/or neutralizes threats in a firm's environment (Lepak & Snell, 1999). It is rare when only a small number of current and potential competitors have it. A firm resource must be imperfectly imitable where other firms who do not possess these valuable and rare
9
resources cannot obtain them easily. Finally, a firm resource needs to be imperfectly substitutable where other firms cannot use strategically equivalent resources to conceive of and implement certain strategies. As noted by Sullivan (2002), intellectual capital that includes organizational, social, and human capital may provide sources of competitive advantage. In addition, Barney (1991) pointed out, in particular, that human capital as a set has the potential for creating and maintaining competitive advantage because it is likely to depend on unique historical conditions, social complexity, and causally ambiguity, which are three conditions that make resources relatively more inimitable. In the context of human capital, the level of human capital has an influence on firm performance (e.g., Becker, 1964; Hitt et al., 2001; Mincer, 1974). For instance, Pennings et al. (1998: 426) noted that "professionals endowed with a high level of human capital are more likely to deliver consistent and high-quality services" and the contribution of human capital investment to firm survival is critical. Similarly, Snell and Dean (1992) noted that human capital adds value to the firm because of enhanced potential for productivity provided by higher knowledge and skills. In other words, the higher the level of knowledge, skills, and abilities of employees, the more potential human capital has for impacting firm performance. The main emphasis of the resource-based view of the firm (Barney, 1991) is how the level of intellectual capital that firms possess or acquire can generate above average rent in terms of improved firm performances. Researchers adopting this perspective suggest that the level of intellectual capital, in general, and human capital, in particular, is directly influenced by HRM practices that are aimed toward
10
selecting/recruiting and training/developing employees (McMahan et al., 1999; Wright & McMahan, 1992; Wright & Snell, 1991). While the resource-based view of the firm has noted the importance of intellectual capital, the level of human capital is critical when applied to strategic HRM. In addition, the resource-based view of the firm emphasized the role of shared experiences and learning that facilitate the accumulation of individual employees' tacit knowledge and accretion of these individual knowledge into a collective such as units' or firms' organizational capital (e.g., Berman, Down, & Hill, 2002; Guillen, 2000; Lado & Wilson, 1994). As MacDuffie (1995: 199) noted, "Innovative human resource practices are likely to contribute to improved economic performance only when three conditions are met: when employees possess knowledge and skills that managers lack; when employees are motivated to apply this skill and knowledge through discretionary effort; and when the firm's business or production strategy can only be achieved when employees contribute such discretionary effort (Levine and Tyson 1990; Bailey 1992)." While this is not an issue with physical and, to a lesser extent, organizational capital, the actions or behaviors of employees are a critical issue for human (Coff, 1997) and, to a lesser extent, social capital because human capital by itself cannot influence firm performance unless employees are induced to perform or actually use their human capital to improve performance. Although social capital may initially emanate from the employees, through organizational learning, the firm can acquire social capital by utilizing tacit knowledge of the employees. Therefore, in terms of strategic HRM, the most critical component of intellectual capital that a firm can impact may be the level of human capital that resides
11
in the employees. Figure 2 depicts this emphasis of the resource-based view of the firm on human capital as a source of competitive advantage. Therefore, there are two ways, directly and indirectly, that human capital of the firm can influence firm performance. FIGURE 2. Resource-Based View of the Firm in Strategic HRM
Human Capital for the Firm Employee Behavior Firm Performance
The emphasis on human capital characteristics that arise from the resourcebased view of the firm is important, especially given that only a handful of research has specifically examined the linkages between human capital and firm performance (Hitt et al., 2001; Wright et al., 1995). Moreover, despite the growing use of the resource-based view of the firm as an underlying theoretical logic for strategic HRM, much of the empirical research has focused on the relationship between HRM practices and firm performance while ignoring human capital, thereby increasing the importance of examining this component. Delery (1998: 290) explicitly acknowledged that, "a firm does not gain a competitive advantage from HRM practices, per se, but from the human resources that the firm attracts and retains." Consequently, understanding the linkage between an HRM system and performance must explicitly consider the relationships among a bundle of HRM policies, level of human capital, and performance. Behavioral Perspective The behavioral perspective in strategic HRM (Jackson et al., 1989; Schuler & Jackson, 1987) posits that different role behaviors are required for different types of 12
strategies that firms pursue. "The theory focuses on employee behavior as the mediator between strategy and firm performance" (Wright & McMahan, 1992: 303) or "between HR practices and sustainable competitive advantage" (McMahan et al., 1999: 106). Examples of employee behaviors may be being flexible to perform in-role or extra-role tasks or being efficient in performing required tasks. While useful in conceptualizing the role of employee attitudes/behaviors, its sole emphasis on employee behaviors may be too simplistic as the behavioral perspective typically ignores the level of human capital. It is simplistic in that "it assumes that the [sole] purpose of various employment practices is to elicit and control employee attitudes and behaviors" (Wright & McMahan, 1992: 303) and does not take into account the effect of the level of human capital that firms acquire and/or develop. To confound issues more, HRM practices are often used as proxies for employee behaviors, which are not a good proxy measure for the concept of employee behaviors and assume that an implementation of a policy automatically results in the desired employee attitudes and/or behaviors. There are many factors that might influence behaviors that go beyond HRM practices. Without actually measuring these factors, one cannot be certain why HRM practices lead to certain employee behaviors. Figure 3 illustrates the behavioral perspective. FIGURE 3. Behavioral Perspective in Strategic HRM Human Capital for the firm Firm Performance Employee Behaviors
13
Understanding employee attitudes and behaviors: Social exchange theory. When examining employees' attitudes and behaviors and their effects on firm performance at the firm level of analysis, social exchange theory is a critical perspective that needs to be integrated into theorizing and empirical research that explicate the mediating mechanisms. Social exchange theory focuses on the motivational component of employee-organization relationships and provides insights regarding the implications of the fit between the expected inducements and contributions provided in an employeeemployer exchange (Tsui, Pearce, Porter, & Tripoli, 1997). Specifically, social exchange theorists (e.g., Gouldner, 1960) examine the exchanges that occur between employers and employees regarding perceptions of reciprocity at an individual level of analysis. The essence of the social exchange theory is the notion of norm of reciprocity that develops, which makes employees feel obligated to respond equitably to treatments from others (including one's employer). As Wayne, Shore, and Liden (1997: 83) noted, "employees seek a balance in their exchange relationships with organizations by having attitudes and behaviors commensurate with the degree of employer commitment to them as individuals." Though the exact focus of social exchange theorists varies, a common theme is that the perceived balance between organizational inducement and employee contributions has performance implications. For instance, Wayne et al. (1997) found that the use of HRM practices that were developmental in nature was positively related to perceived organizational support. Perceived organizational support, in turn, has been found to be positively associated with affective, organizational commitment (Eisenberger, Fasolo, & Davis-LaMastro, 1990), constructive suggestions (Eisenberger
14
et al., 1990), and citizenship behaviors (Wayne et al., 1997). In addition, perceived organizational support is negatively associated with absenteeism (Eisenberger et al., 1990) and turnover intentions (Guzzo, Noonan, & Elron, 1994). Several variables including perceived organizational support, leader-member exchange (e.g., Liden, Sparrowe, & Wayne, 1997; Wayne et al., 1997), organizational commitment (e.g., Shore, Berksdale, & Shore, 1995; Shore & Wayne, 1993), and organizational citizenship behaviors (e.g., Masterson, Lewis, Goldman, & Taylor, 2000), for example, have been used to capture the notion of social exchange. Hence, social exchange theory provides insights regarding the specific mediating factors that likely account for the relationship between HRM and performance. Figure 4 depicts social exchange perspective as an important mediating mechanism for the HRM practices-firm performance relationship. FIGURE 4. Social Exchange Theory for Strategic HRM
Human Resource Management Practices
Aggregate Employee Attitudes & Behaviors e.g.,
• Leader-Member-Exchange • Perceived Organizational Support • Organizational Commitment • Organizational Citizenship Behaviors
Firm Performance
Within an organizational setting, the employer and the employees can be considered parties to the social exchange relationships. First, assuming that the organization first acts in such a way to provide employees with something of social value such as providing job security or access to training programs, the employees,
15
then, perceive that the organization has provided something of social value to them. Only then, the employees will feel indebted to reciprocate to the organization with something of equal or greater value. Although Schuler and Jackson (1987; Jackson et al., 1989) and other scholars in strategic HRM have not articulated the specific variables embedded within the behavioral perspective, those that have been used in previous studies in organizational behavior research as representing the notion of social exchange include leader-memberexchange (e.g., Masterson et al., 2000; Settoon, Bennett, & Liden, 1996; Wayne et al., 1997), organizational justice (e.g., Konovsky & Pugh, 1994; Masterson et al., 2000; Moorman, 1991; Organ & Konovsky, 1989), perceived organizational support (e.g., Armeli, Eisenberger, Fasolo, & Lynch, 1998; Eisenberger et al., 1986; Eisenberger et al., 1990; Masterson et al., 2000; Settoon et al., 1996), and psychological contract (e.g., Rousseau, 1989, 1995). At the same time, organizational commitment (e.g., Eisenberger et al., 1986; Eisenberger et al., 1990; Shore & Wayne, 1993), organizational citizenship or extra-role behaviors (e.g., Konovsky & Pugh, 1994; Lambert, 2000; Moorman, 1991; Organ & Konovsky, 1989; Van Dyne & Ang, 1998), and task or in-role performance (e.g., Austin et al., 1991) have been used as exemplar of employees' affective reactions to organizational actions. Of these social exchange variables used in the field, psychological contract, which is defined as "individual beliefs, shaped by the organization, regarding terms of an exchange agreement between individuals and their organization" (Rousseau, 1995: 9), represent an aspect of social exchange that is based on obligation that arises from the feeling of indebtedness for the actions provided by the company and the notion of
16
reciprocity. Moreover, while psychological contract refers to the individual employee's beliefs regarding the obligatory aspects of social exchange relationship between the individual and the organization, Rousseau (1995) also discussed the concept of normative contract, which refers to individual employee's perception of the obligations that the organization has toward all of the employees working for the firm. According to Chan (1998), five different types of compositions models that describe the functional relations among constructs in the same content domain at a higher level of analysis exist: 1) additive, 2) direct consensus, 3) referent-shift consensus, 4) dispersion, and 5) process composition. Of particular importance for the present dissertation is the referent-shift consensus model, which creates a distinct concept when individual responses are aggregated, although the concept originally is derived from individual-level unit when examining unit-level attitudes. He provides self-efficacy, collective efficacy, and team efficacy as an example of individual-level unit (self-efficacy, defined as an individual's belief and confidence in mobilizing one's resources to achieve successful task performance [Bandura, 1977]) having a different meaning at the higher-level of analysis (collective efficacy is defined as the individual team member's belief and confidence that the team can mobilize its resources to achieve successful task performance, which is then used to create team efficacy (grouplevel construct), which refers to the team members' overall belief and confidence that the team as a collective can mobilize its resources to achieve successful task performance. A similar shift is expected for psychological contract to normative contract to collective normative contract.
17
Collective normative contract is defined here as representing employees' collective beliefs regarding terms of social exchanges between the employees and their organization regarding the obligations that the employer and employees have toward each other. When comparing different operationalization of social exchange relationships, collective normative contract appears to convey the basic components of any social exchange relationships that may be necessary for further reciprocation by the employees. Furthermore, other variables that have been used to represent individual's social exchange relationship such as perceived organizational support, leader-memberexchange, and organizational commitment have not been conceptualized at and do not possess parallel concept at higher level of analysis. Hence, collective normative contract appears more appropriate than other operationalizations of social exchange relationships. Integrating resource-based view and behavioral perspectives. As the above discussion suggests, there appears to be a disconnect in our understanding of the relationship among human capital, HRM, and performance or employee behavior, HRM, and performance. According to the resource-based view of the firm, human resources have the potential to provide firms with sustainable competitive advantage (Barney, 1991). While Hitt et al. (2001) examined the relationship between human capital (quality of school and total experience in a firm) and firm performance, they did not examine the influence of attitudes and behaviors and a system of HRM practices on these relationships. One of the reasons that this is the case may be due to the relative emphasis placed by resource-based view of the firm on human capital and the emphasis placed on employee attitudes and behaviors by the behavioral perspective without
18
considering these perspectives simultaneously, for instance. By integrating both perspectives and simultaneously looking at human capital and employee attitudes and behaviors such as collective normative contract of the employees as important components to firm performance, we can gain a better understanding of how HRM practices lead to firm performance. However, having highly knowledgeable, skilled, and capable workforce is a necessary but not a maximally effective condition toward improving firm performance. In order for human capital to actually contribute to firm performance, employee behaviors that are necessary need to be exhibited. In other words, having competent human capital provides firms with the potential for improving firm performance whereas employee behaviors are a necessary catalyst for converting this potential into reality. HRM practices are the means by which these human capital and employee behaviors are obtained. Therefore, it is necessary to examine the relationships among human capital, employee attitudes and behaviors, HRM, and performance from a holistic perspective, incorporating all four concepts in a study. Figure 5 shows the inter-relationships among these variables. To do so, however, we must also gain a better understanding how we conceptualize HRM policies. FIGURE 5. Integration of the Resource-Based View of the Firm and Behavioral Perspective on Strategic HRM Human Capital of the Firm • Knowledge, skills, & abilities Social Exchange Relationships • Normative contract
HRM Activities
Firm Performance
19
Conceptualization of performance. Another issue that is critical to fully understand the HRM-firm performance relationship pertains to the conceptualization of performance (Becker & Gerhart, 1996; Rogers & Wright, 1998). As noted by Wright (1998), there seems to be a consensus within the realm of strategic HRM that maximizing organizational performance, particularly financial performance, is the major goal to be achieved. This preference for financial measures of performance has been shown by Rogers and Wright (1998) who reviewed the literature and noted that out of 80 dependent variables included in the strategic HRM research, accounting measures such as return on asset, return on equity, profits, and sales and market measures such as stock price and Tobin's Q were used in more than half of the research. However, the "appropriate dependent variable will vary with the level of analysis" (Becker & Gerhart, 1996: 791). At the business unit-level of analysis, perhaps the productivity of research and development personnel or their turnover rate may be more important for firms pursuing a differentiation strategy whereas the productivity of a firm's production staff may be more critical for firms following cost leadership strategy. Further, "the focus should be on variables that have inherent meaning for a particular context" (Becker & Gerhart, 1996: 791). For example, efficiency-based financial measures such as return on assets or return on equity may be more appropriate for firms pursuing cost leadership strategies in most of their business units whereas sales growth (ratio) or revenue growth (ratio) may be more appropriate for firms pursuing product differentiation strategies for the majority of their business units. Thus, it may be more appropriate for strategic HRM research to include multiple indicators of firm performance and make differential predictions based on them.
20
Dyer and Reeves (1995) noted different types of performance measures that are most appropriate for strategic HRM research. They proposed that four effectiveness measures are (1) human resource outcomes such as absenteeism, turnover, job satisfaction, and individual or group performance, (2) organizational outcomes such as productivity, quality, and service, (3) financial or accounting outcomes such as profitability, return on assets, and return on invested capital, and (4) stock market performance (stock value or shareholder return). Again, pointing to the fact that determination of which performance measures to include may be affected by the specific context of the research. For instance, turnover or retention rate may be more important for information technology companies that are knowledge intensive than for manufacturing companies that produce standardized products. Or, turnover or retention rate may be more critical for firms that utilize team-based production. Given different conceptualizations of performance and an existence of multiple types of performance indicators, it is important to clearly differentiate these outcomes and investigate the impact of human capital, employee attitudes and behaviors, and HRM for multiple outcomes if we are to fully understand the HRM-firm performance relationship. Perhaps, the outcome measures can be ordered from proximal to distal with employees as an anchor. Hence, HR outcomes would be the most proximal, which leads to organizational outcomes. Organizational outcomes, in turn, may lead to financial or accounting outcomes and ultimately, market measures (cf. Becker & Huselid, 1998). The relationship between HRM activities and firm performance, as noted before, has been relatively well established. However, the intermediate linkages between HRM
21
activities and firm performance are not yet well understood. Hence, it is important to be able to explicate these intermediate linkages, which is the primary objective of this dissertation. The integration of resource-based view and behavioral perspective (and social exchange theory) makes it imperative that HR-related outcomes such as labor productivity and turnover rate, for example, be examine as the more proximal outcomes to the level of human capital and perceptions and reactions to social exchange relationship. The logical next step is to link these HR-related outcomes to organizational outcomes (such as sales growth and operating profits). Figure 6 illustrates the overall framework for this dissertation. FIGURE 6. Overall Theoretical Framework
Level of Human Capital High Performance Work System HRM Practices HR Outcomes Collective Normative Contract
Unit Performance
Review of Past Research Related to the Framework Although there is no empirical research in the area of strategic HRM that has examined these specific linkages from High Performance Work System to the level of human capital and collective normative contract, or the level of human capital and collective normative contract to HR-related outcomes, previous empirical research has examined the relationships between High Performance Work System and some of the
22
HR-related outcomes such as turnover and labor productivity, as well as HR-related outcomes to firm/unit performance. For example, Huselid (1995) found that turnover rate and labor productivity mediated the relationship between High Performance Work System and two indicators of firm performance, i.e., logarithm of Tobin's Q (market value of the firm / replacement costs of its asset) and gross rate of return on capital for the manufacturing companies. Way (2002) and Batt (2003) found significant relationships between High Performance Work System and labor productivity, and High Performance Work System and turnover rate for small companies and service companies, respectively. Similarly, Ichniowski, Shaw, and Prennushi (1997) found a positive relationship between innovative work practices and labor productivity for steelfinishing lines. However, none of the studies examined the level of human capital or collective normative contract as additional mediating mechanisms between High Performance Work System and (firm) performance. In addition, in the strategy literature, Carpenter, Sanders, and Gregersen (2001) found a relationship between CEO's international assignment experiences (as representing the level of human capital) and multinational companies' return on asset. Hitt et al. (2001) found a relationship between human capital and performance of law firms (net income to total firm revenue). However, these studies did not examine the relationship between High Performance Work System and the level of human capital or the level of human capital and HR-related outcomes. Thus, it is unclear if the level of human capital indeed acts as a mediator of the relationship between High Performance Work System and HR-related outcomes. Therefore, one of the contributions of the dissertation is explicating this mediating mechanism through the resource-based lens. In
23
addition, investigating the mediating influences of collective normative contract and HR-related outcomes are other contributions that I intend to make in this dissertation. Chapter Summary In summary, in this chapter I have identified several theoretical perspectives that are relevant and important in examining the research questions identified in the first chapter and provided an integrated framework that outlines two general mediating mechanisms between a system of HRM practices and firm performance relationships. Given the objective of the dissertation, focusing on both the resource-based view and the behavioral perspective provide a framework for understanding the potential firm performance implications of a firm's human capital as well as aggregate attitudes and behaviors of employees. In addition, understanding the role of a bundle of HRM practices in these relationships is critical as the system of HRM practices is the primary ways to influence the level of human capital that organizations possess and aggregate employee attitudes and behaviors that are exhibited for the organizations.
24
Chapter 3 Model Development and Hypotheses Scope of the Research For the current dissertation, the critical issue is to examine the mediating mechanisms through which a system of HRM activities leads to performance via level of human capital and employee attitudes and behaviors. Hence, the main emphasis of the dissertation is on explicating the relationships among HRM system, human capital, employee attitudes and behaviors, and outcomes. Given an existence of different employee groups within a firm (e.g., Lepak & Snell, 1999; Osterman, 1987), associated differences in expected attitudes and behaviors that contribute to firm effectiveness, and HRM practices used to elicit such attitudes and behaviors, the boundary conditions for the dissertation is placed around three issues including level of analysis, performance indicators, and type of employee groups. Level of analysis. Previous research on strategic HRM has been conducted at both the corporate-level (e.g., Arthur, 1992, 1994; Becker & Huselid, 1998; Delaney & Huselid, 1996; Guthrie, 2001; Huselid, 1995; Jackson et al., 1989; Terpstra & Rozell, 1993; Whitener, 2001) and business-unit level (Delery & Doty, 1996; Ichniowski et al., 1997; Koch & McGrath, 1996; McDuffie, 1995; Russell et al., 1985; Snell & Dean, 1992, 1994; Youndt et al., 1996). Although there are certain advantages as well as disadvantages to selecting one level vs. the other, the most appropriate level of analysis for the current dissertation is primarily at the business unit level because social exchange relationships between a corporation and employees is likely to be confounded with variety of factors external to the social exchange relationships that exist between
25
the employer and the employees. In other word, at the corporate level, the relationships among the variables of interest may differ from those of the business units. Also, corporations may have multiple business units with different objectives for one business unit vs. the other. Aggregating these units may mask the important relationships that exist among the variables included in the dissertation. Unit performance indicators. It is widely acknowledged that firm performance is a multi-dimensional rather than a uni-dimensional construct (Dyer & Reeves, 1995; Rogers & Wright, 1998; Wright & Sherman, 1999). To reflect this multidimensionality, I will examine multiple outcomes in this dissertation. Specifically, I examine outcomes that are more intermediate in terms of High Performance Work System' impact on employees' behavioral outcomes, which I refer to as HR-related outcomes, and include such aggregate behaviors as labor productivity, level of extra-role behaviors exhibited by the employees in the unit, and quality of task performance as well as more distal establishment performance outcomes such as business unit performance. As Dyer and Reeves (1995) and Becker and Huselid (1998) noted, the dependent variables (performance) can be ordered from more proximal such as labor productivity to more distal variables such as return on asset and market-to-book ratio when examined from the strategic HRM perspective. In addition, Gupta (1987) noted the difficulty of obtaining financial measures of business units across multiple industries, although more objective measures of unit performance such as market share and delinquency rates exist when examining bank branches (e.g., Ryan, Schmit, & Johnson, 1996) or sales growth ratio and return on assets for business units in the U.S. pulp and paper industry (Davis, Robinson, Pearce,
26
& Park, 1992). However, more perceptual measures such as the level of customer satisfaction (e.g., Gelade & Ivery, 2003; Ryan et al., 1996) and relative firm performance (e.g., Delaney & Huselid, 1996; Powell, 1992) that have been used to assess performance may be more appropriate. Dollinger and Golden (1992) and Powell (1992), for example, found measures of perceived organizational performance to correlate positively (with medium to high correlations) with objective measures of firm performance. Furthermore, the use of perceptual measures allows an analysis of forprofit and not-for-profit organizations to be conducted because objective performance data are typically not available for the non-profit organizations (Delaney & Huselid, 1996). Given the difficulty in obtaining objective, financial measures for the business units (Gupta, 1987) and the lack of comparability of these objective measures across multiple industries, more perceptual measure of unit performance is used. Employees groups. Finally, as several scholars have noted (Jackson et al., 1989; Lepak & Snell, 1999; Osterman, 1987; Rousseau, 1995; Tsui et al., 1995), there are some indications that different HRM systems are used for different groups of employees depending on their employment relationships. For example, Lepak and Snell (2002) found different types of HR configurations for different types of employee groups. One approach that others have used to deal with this is to focus their attention on only those workers deemed most critical to a firm's success, i.e., core employees (e.g., Arthur, 1992; Delery & Doty, 1996; MacDuffie, 1995), or to aggregate two different groups, core and support, to a firm level (Huselid, 1995). As noted by Lepak and Snell (1999), there are some limitations associated with this approach of combining core and support employees to focus on permanent
27
employees because different HR policies may be used for different employee groups, which may confound the findings. Here, permanent employee is defined as those nonmanagerial employees who are directly involved in the production of products or selling of products/services (core employees) and those employees who provide administrative and supporting services (support employees) on a long-term, full employment basis (e.g., Barnett & Miner, 1992; Davis-Blake & Uzzi, 1993; Lepak, Taylor, Tekleab, Marrone, & Cohen, 2002). However, there are several substantive and methodological tradeoffs that need to be taken into account when considering the use of different employee groups, as compared to the permanent employee group. First, one of the complication introduced by using different employee groups may be related to aggregation and generalizability issues. For instance, inconsistency in the types of jobs classified as one type of employment group vs. the other by the respondents in different companies might introduces aggregation as well as generalizability problems. If the respondents from each firm do not agree on the types of jobs classified as core vs. support, for example, there may not be a concept that can be used at the unit-level across units or between firms. In addition, depending on the specific sample of units included, the results may not be generalizable to other settings beyond this dissertation. Second, not all units have both employment groups and given that the main purpose of the dissertation is to examine the mediating mechanisms through which High Performance Work System influence unit performance, multiple employee responses are necessarily for each unit and collecting enough employee responses for each of the employment group for each unit and having enough sample size at the end is beyond the
28
capabilities of any scholars, unless multiple units from a single company or industry (e.g., Gelade & Ivery, 2003; Ryan et al., 1996), or large-scale survey data are made available to scholars (e.g., Way, 2002). Given the objective of the dissertation and feasibility considerations, therefore, I focus on permanent (i.e., core and support) employees as the primary target of this dissertation. Research Model Thus far, I have described the boundary conditions of this dissertation. The following sections will discuss specific hypotheses that are incorporated in the overall framework described in Chapter 2. The following sections are broken down into three sections. The first section describes the influence of High Performance Work System on the level of human capital present in the unit as well as its indirect effect on HR-related outcomes and unit performance. The second section describes the impact of the High Performance Work System on collective normative contract as well as its indirect effect on HR-related outcomes and unit performance. Finally, the third section discusses the interaction between level of human capital and collective normative contract on HRrelated outcomes. With the first two sections, I hypothesize that these variables act as two broad mediating mechanisms where a system of HRM practices affects unit performance. The first mediating mechanism is though human capital where High Performance Work System influences HR-related outcomes, which, in turn, acts as a mediator to influence performance. I also anticipate a mediating process through social exchange where High Performance Work System affects employees' collective perceptions of exchange relationships with the unit (collective normative contracts) that has impact on HR-related outcomes, which, in turn, lead to unit-valued outcomes.
29
Figure 7 illustrates the proposed research model for this dissertation. I describe these relationships in more detail below. Figure 7. Proposed Model of High Performance Work System Level of Human Capital High Performance Work System
• labor
HR Outcomes Unit Performance
productivity • task performance • OCBs Social Exchange • Collective Normative Contract
High Performance Work System, Level of Human Capital, and Performance High Performance Work System (e.g. Huselid, 1995) refers to a collection of HRM practices that are aimed at obtaining and insuring high level of competency or human capital for a firm's workforce as one of the primary objectives, on the one hand, and coordinating and eliciting desired employee behavior over time (Wright & Snell, 1991) that adds value to the organization, on the other. Huselid (1995) also examined High Performance Work System and implied two different influences of these HR practices included in High Performance Work System when he factor analyzed these HR practices into two and labeled those: 1) HRM practices involving selection and training as "employee skills and organization structures," emphasizing human capital; and 2) HRM practices involving compensation and performance appraisal as "motivation" factors, focusing on employee behavior.
30
The relationship between High Performance Work System and the level of human capital is rather straightforward (cf. Wright & Snell, 1991). For example, researches in the selection and staffing literature have long recognized that selection/staffing practices have an effect on the characteristics of the employees/managers selected for the job position (e.g., Guthrie & Olian, 1991). Similarly, Delaney and Huselid (1996) noted that organizations could adopt different HRM practices that place emphasis on improving the quality of the individual hired, or on raising the skills and abilities of current employee, or both. The more rigorous and comprehensive recruitment and selection procedures are, the higher the level of knowledge, skills, and abilities for the recruited employees likely be than when these HRM policies are not utilized. In addition, comprehensive, on-going training and development programs can augment and improve employee knowledge, skills, and abilities. Therefore, there is likely to be a positive relationship between High Performance Work System and the level of human capital (cf. Hitt & Barr, 1989). In addition, High Performance Work System usually entails compensation policies that lead the market as well as performance appraisal policies that are developmental, which may attract potential applicants with higher talents. Hence, I expect the following: Hypothesis 1: High Performance Work System is positively related to the level of human capital for the unit. Higher levels of human capital have the potential to provide a sustainable competitive advantage to firms because employees differ in their amount of knowledge, skills, and abilities that they possess. For example, Hitt et al. (2001) found a positive, curvilinear relationship between human capital and firm performance for a professional
31
service firms. More specifically, they found articulable knowledge, which is a type of knowledge that can be codified and transferable (Lane & Lubatkin, 1998), and tacit knowledge, which is another type of knowledge that is difficult to codify and transfer (Lane & Lubatkin, 1998), to have an inverted U-shaped relationship with return on sales. Similarly, Pennings, Lee, and Van Witteloostuijn (1998) found human capital characteristics of accountants to be negatively related to firm dissolution (i.e. positively related to firm survival). These empirical studies have shown the relationship between human capital and firm performance to be positive. Moreover, Bettencourt, Gwinner, and Meuter (2001) found employee's knowledge to be a significant predictor of serviceoriented organizational citizenship behaviors. In particular, they argue that (procedural) knowledge enhances the repertoire that employees have developed in terms of useful ways of interacting with specific customer types. An analogy may be drawn for interacting and dealing with other coworkers. Therefore, I expect the following: Hypothesis 2: The level of human capital for the unit is related to HR-related outcomes such as labor productivity, unit-level task performance, and unit-level organizational citizenship behaviors. A growing number of studies (Arthur, 1994; Becker & Huselid, 1998; Guthrie, 2001; Huselid, 1995; Ichniowski et al., 1997) have also shown that a system of HRM practices to be directly and positively related to HR-related outcomes such as labor productivity and turnover rates as well as establishment level performance (Guthrie, 2001; Huselid, 1995; Way, 2002). However, a system of HRM practices is not likely to influence these HR-related outcomes directly per se (cf. Becker & Gerhart, 1996). It is through increases in labor productivity stemming from an increase in the level of human
32
capital that enables firms to reap the benefits of competence-enhancing HRM practices embedded within High Performance Work System (cf. Snell & Dean, 1992). It is only through its influence on employees' level of knowledge, skills, and abilities that the potential for increased performance gains can be realized. Thus, the level of human capital is likely to mediate the relationship between High Performance Work System and HR-related outcomes. Hypothesis 3: The level of human capital mediates the relationship between High Performance Work System and HR-related outcomes such as labor productivity, unit-level task performance, and unit-level organizational citizenship behaviors. In turn, HR-related outcomes are likely to mediate the relationship between the level of human capital and unit level performance (cf. McMahan et al., 1999). For example, Hitt et al. (2001) found a positive relationship between the level of human capital and indicators of law firm performance. However, even when the firm initially obtains employees who possess high knowledge, skills, and abilities (i.e., human capital), if many of them exit the company shortly afterward, it is not likely to influence firm performance in a positive manner. In addition, if the employees are not motivated to use their skills, the level of task performance may only be minimally adequate, which does not justify the above-average market rate paid for these highly skilled employees. Thus, again, the performance for the unit may not improve simply by recruiting highly knowledgeable, skilled, and capable individuals. In other words, HR-related outcome such as an increase in labor productivity, aggregate task performance level, and
33
aggregate organizational citizenship behaviors are expected to mediate the relationship between the level of human capital and unit performance. Hypothesis 4: HR-related outcomes such as labor productivity, unit-level task performance, and unit-level organizational citizenship behaviors mediate the relationship between the level of human capital for the establishment and unit performance. High Performance Work System, Collective Normative Contract, and Performance While the behavioral perspective provides the general foundation to expect that employee attitudes and behaviors mediate the relationship between HRM policies and unit performance, incorporating a social exchange perspective (e.g., Blau, 1964; Gouldner, 1960) provides a compelling theoretical logic to further explicate the mediating mechanisms outlined by the behavioral perspective. Social exchange theory posits that two individuals or parties often decide to enter exchange relationships when both individuals perceive that it is beneficial to do so (Blau, 1964; Gouldner, 1960; Homans, 1958, 1961). Initially, parties to an exchange are cautious and conscious about the currency of exchange (be it monetary incentives or approval) that is being exchanged between each party. This monitoring may decrease over time when each party finds the relationship satisfying and trust develops between the parties, thereby increasing the value of contributions both in terms of acceptable alternatives and time between reciprocation. Typically, both parties strive for a fair exchange with equivalent contributions judged through time (Tsui et al., 1997). When one party to the relationship goes beyond required obligations (be it initial payment to or repayment of valued contribution by the other party), the other party subsequently
34
feels indebted to contribute back to tilt the balance in the other direction. This component of the exchange is known as the norm of reciprocity (Homans, 1961). In an organizational setting, one of the possible pairs of partners to social exchanges is the employees and the employing organization. Within this potential partnership, employees as well as an employer can initiate the social exchange cycle. However, it is typically assumed that the organization initiates the action by providing employees with something that the employees' value - e.g., providing job security, access to training programs - that are represented by the HR policies that the organization utilizes. If the employees perceive that the organization has provided something of value to them, the employees are likely to feel obligated to reciprocate by providing the organization with something of equal or greater value. Within this social exchange process, it may be important to distinguish two components to understand the processes in detail: 1) the way employees perceive and interpret the organization's actions, and 2) the employees' reactions to them. The first component relates to the employees' collective perceptions and attributions of the organizational actions. When employees share perceptions of HRM policies as valuable and attribute the intention of the organization in a favorable manner, they collectively should possess the notion that the employees in the organization need to repay the organization, i.e., collective normative contract, and they are likely to feel obligated to reciprocate with behaviors that the organization values. Hence, the second component of the social exchange is the reciprocation provided by the employees. High Performance Work System and Collective Normative Contract. When examining the variables that are based on the notion of social exchange as perceived by
35
the employee, collective normative contract seems quite relevant. Although different variables have been examined in the literature previously, scholars, in general, have suggested that the nature and strength of obligations incurred through a social exchange depends on the quality of the relationship between the partners to the exchange (Eisenberger et al., 1986; Lambert, 2000; Van Dyne, Graham, & Dienisch, 1994). Konovsky and Pugh (1994) identified a set of mediating variables that provides the foundation for the social exchange by setting the tone for the exchange relationship as "macromotives," which may include perceived organizational support and psychological contract (cf., Lambert, 2000), among others. However, given the notion of organizational climate, which is defined as a set of shared perceptions of the policies, practices, and procedures that is developed through interactions (James et al., 1988; Schneider & Reichers, 1983), it is likely that shared perceptions with regard to the terms of the contract exist within organizational units. In fact, Marcoulides and Heck's (1993) study examined the collective perception of the quality of leader-member interactions as an aspect of organizational climate and collective commitment of employees as shared attitudes; Naumann and Bennette's (2000) and Colquitt, Noe, and Jackson's (2002) study indicated the existence of climate for procedural justice; and Rogg, Schmidt, Shull, and Schmitt's (2001) study on strategic HRM examined the relationship between HRM philosophy, organizational climate (commitment as an aspect of organizational climate), and performance (customer satisfaction). Of particular relevance to the current study is the study conducted by Marcoulides and Heck (1993) that found a positive relationship between organizational climate and shared employee attitudes.
36
It has been noted that organizational climate develops though three different, yet interrelated, means (Schneider & Reichers, 1983): 1) from social interactions of employees and through shared meanings (symbolic interactionist approach); 2) from attraction/selection/attrition (ASA) resulting in homogeneity (ASA approach); and 3) from mere exposure to the same policies, practices, and procedures (the structuralist approach). The first approach to organizational climate emphasizes interaction process among the employees in which employees try to make sense of the external events and assign meanings to these. The second approach (attraction-selection-attrition) focuses on the notion of person-organization fit with the organizational goals (Schneider, Goldstein, & Smith, 1995). Organizational goals, and the processes, structures, and culture that emerge to facilitate achievement of the organizational goals originally articulated by the founder are thought to determine the kinds of people who are attracted to, are selected by, and retained by the organization (Schneider et al., 1995). Finally, the third approach to climate draws attention to the structural characteristics of the organization. By being exposed to certain types of situational conditions such as leadership (e.g., Bass, 1990), job design (e.g., Hackman & Oldham, 1980), groups or teams (e.g., Guzzo & Shea, 1992), pay systems (e.g., Lawler, 1981), dynamics of the external environment (e.g., Joyce & Slocum, 1990), employees are likely to develop similar perceptions. As noted above, these approaches, however, are not mutually exclusive and they seem to differ on the emphasis placed. Perhaps, the most relevant approaches when examining the formation of HRM policies and their effects on employee perceptions
37
may be the second and the third approach to organizational climate. High Performance Work System may affect the organization in two ways: by attracting particular types of applicants and enabling the organization to select highly talented recruits who are likely to stay with the company and the other by providing common understanding to employees' experiences in terms of shared meaning by socializing them into interpreting experiences in particular ways. For instance, High Performance Work System is typically used to attract and select employees with high potential, which is likely to impact employee perceptions. For instance, providing rigorous training may lead to an increase in perceived obligations that the employees feel toward the organization in that the employees may interpret receiving this training as an indication of the goodwill from the organization. Similarly, extensive recruitment and selection procedures may signal to employees that their organization places great value on them and consider them as crucial in the survival and success of the organization. While there are only few empirical studies that had examined the effects of HRM policies on social exchange variables, Wayne et al. (1997) found developmental experiences and number of promotions to be positively related to perceived organizational support. Developmental experiences in their study referred to those job assignments that are challenging and developmental. Similarly, Whitener's (2001) study indicated that performance appraisal and internal rewards that are part of "high commitment" HR practices were correlated with perceived organizational support. Moreover, Taylor et al. (1995) examined due process in performance appraisal where due process performance appraisal practices include participative decision-making, self-
38
evaluation, and feedback. They found that these appraisal practices were positively related to procedural fairness perception of the employees. In addition, they found these practices to be positively related to employees' evaluation of the supervisor (or the leader). While these studies focused on the individual-level of analysis, using different variables to represent social exchanges, I can infer from the results that High Performance Work System will be associated with employees' shared perceptions of the terms of contracts that they are expected to receive and return. As Kozlowski and Klein (2000:30) noted these aggregate employee attitudes and behaviors can be considered a shared construct, which "describes the characteristics that are common to - that is, shared by - the members of a unit." They also note that organizational climate, collective efficacy, and group norms are examples of shared unit-level properties and these shared unit properties emerge as consensual, collective aspects of the unit as a whole. Therefore, I expect the following: Hypothesis 5: High Performance Work System has a positive relationship with collective normative contract. More specifically, High Performance Work System is positively related to employees' shared perceptions regarding obligations (collective normative contract) toward the organization. Collective normative contract and HR-related outcomes. Unlike the relationship between HRM policies and aggregate employee attitudes and behaviors, there are more studies that examined the impact of employees' reactions to organization's social exchange relationship, especially with regard to behaviors that contribute indirectly to the organization by maintaining its social and psychological environment. Several terms
39
have been used to describe such behaviors in the past, including organizational citizenship behavior (OCB: e.g., Organ, 1988), prosocial organizational behavior (Brief & Motowidlo, 1986; George, 1990, 1991; George & Bettenhausen, 1990; O'Reilly & Chatman, 1986), organizational spontaneity (George & Brief, 1992; George & Jones, 1997), extra-role behavior (Van Dyne & LePine, 1998; Wright & George, 1993), and contextual performance (Borman & Motowidlo, 1993, 1997; Borman, White, & Dorsey, 1995; Motowidlo & Van Scotter, 1994; Van Scotter & Motowidlo, 1996). More recently, empirical research has accumulated that suggests that extra-role behaviors lead to greater overall organizational effectiveness (MacKenzie, Podsakoff, & Paine, 1999). Podsakoff and MacKenzie (1994) examined the impact of three forms of OCBs (helping, sportsmanship, and civic virtue) on the performance of life insurance agencies. They found that all three forms of OCBs had significant effects on unit-level performance and together accounted for approximately 17% of the variance in this criterion variable. Walz and Niehoff (1996) found similar results by examining the relationships between OCB dimensions and a multitude of performance measures in limited menu restaurants. In particular, they found that the overall OCB accounted for an average of about 29% of the variance in the six objective criterion variables. Similarly, George and Bettenhausen (1990) examined the relationship between employees' prosocial behaviors for the retail store and sales performance and found a positive correlation of .33. Particularly relevant is the study conducted by George and Bettenhausen (1990) who examined the relationship between group-level prosocial behavior and turnover rate. Their results indicated that group-level prosocial or organizational citizenship
40
behavior was negatively correlated with turnover rate for retail stores. In addition, they also found a positive relationship between group-level prosocial behavior and labor productivity, which they called sales performance (total store sales / number of sales associates working in the store). Extrapolating from this finding, I expect that employees' shared perception of normative contract, which is the result of social exchange relationship that the employees have with the organization, to be positively related to labor productivity, task performance, and citizenship behaviors exhibited in the unit. Therefore, I hypothesize the following: Hypothesis 6: Collective normative contract is positively related to HR-related outcomes such as labor productivity, unit-level task performance, and unit level organizational citizenship behaviors. The relationships above describe a positive relationship between High Performance Work System and collective normative contract, and collective normative contract and HR-related outcomes. In addition to these direct relationships, I expect mediating mechanisms that occur through these relationships. First, I expect that aggregate employee perceptions of normative contract to mediate the relationship between High Performance Work System and HR-related outcomes. There is an emerging body of evidence that suggests the linkages from HRM practices to aggregate employee attitudes and behaviors to HR-related outcomes and, eventually, to unit performance. Lambert (2000), for example, found that HRM practices related to worklife benefits were positively related to perceived organizational support, which was also positively related to helping behaviors (one aspect of OCB). Similarly, Marcoulides and
41
Heck (1993) found organizational climate had both a direct and an indirect (through shared employee attitudes) effect on organizational performance. In addition, empirical research suggests that aggregate employee behaviors, extra-role behaviors in particular, lead to greater overall organizational effectiveness (MacKenzie, Podsakoff, & Paine, 1999). Podsakoff and MacKenzie (1994) examined the impact of three forms of OCBs (helping, sportsmanship, and civic virtue) on the performance of life insurance agencies. They found that all three forms of OCBs aggregated to the agency level had significant effects on unit-level performance and together accounted for approximately 17% of the variance in this criterion variable. Similarly, Walz and Niehoff (1996) found that aggregate helping behavior was positively related to overall operating efficiency, customer satisfaction, revenue to fulltime equivalent, and quality of performance and negatively related to food cost percentage. They also found that overall OCBs accounted for an average of about 29% of the variance in the six objective criterion variables. Similarly, George and Bettenhausen (1990) examined the relationship between aggregate employees' prosocial behaviors for retail stores and sales performance, and found a positive correlation of .33. These results suggest that aggregate employee attitudes and behaviors mediate the relationship between HR practices and outcomes. Combining findings from these empirical studies, I expect the following: Hypothesis 7: Employees' shared perceptions of normative contract mediate the relationship between High Performance Work System and HR-related outcomes. Finally, Koys (2001) examined the relationship between HR outcome (turnover rate) and business unit effectiveness in a panel design, longitudinal study and found that
42
HR outcome precedes business unit effectiveness. Combining this finding with the other studies that found relationship between aggregate employees' reactions and unit effectiveness (George & Bettenhausen, 2000; Podsakoff & MacKenzie, 1994; Walz & Niehoff, 1996), it is expected that HR-related outcomes such as labor productivity and turnover rate act as a mediator between aggregate employees' reactions and unit performance. Hypothesis 8: HR-related outcomes such as labor productivity, unit-level task performance, and unit-level organizational citizenship behaviors mediate the relationship between collective normative contract and unit performance. Moderating Effect of Collective Normative Contract Finally, I expect employees' perceptions of their social exchange relationship with the organization to interact with the level of human capital to have an influence on HR-related outcomes. Given that these shared perceptions of social exchange can be considered "macromotives" (cf. Lambert, 2000), those units with higher amount of collective normative contract are more likely to have employees who are more motivated and willing to exert more effort on behalf of the unit. Based on the notion of reciprocity norm, when employees perceive the organization as a valuable exchange partner and trust that the organization will reciprocate, employees are more likely to exert extra effort that may be substantially more than their fair share that brings them up to par with what the organization has already given them. In a sense, the employees proactively engage in social exchange to further develop the relationship with the organization.
43
Given that employees with higher qualifications (i.e., who possess higher amount of knowledge, skills, and abilities) can generally perform at a higher level than those who do not (e.g., LePine & Van Dyne, 2001; Mohammed, Mathieu, & Bartlett, 2002; Vance, Coovert, MacCallum, & Hedge, 1989), if these employees with higher knowledge, skills, and abilities are motivated to exert extra effort that is over and above their expected contribution, the increases in labor productivity that originate from the increase in the aggregate level of task performance may be substantially larger than those with lower qualifications. In addition, when these highly qualified employees are motivated to help each other and/or voice innovative ways of organizing and executing the tasks (e.g., LePine & Van Dyne, 2001), the cost saving associated with reduced training needs or new process may be larger. Furthermore, George and Bettenhausen's (1990) study found a significantly negative correlation between group's norm on organizational citizenship behavior and turnover rate for the store. Therefore, it can be expected that when employees share a particular norm with regard to the obligations they have toward the organization, they are more likely to exhibit more organizational citizenship behaviors. When employees with high qualifications exhibit the organizational citizenship behavior, the benefits that other employees receive are likely to higher than when employees with low qualifications provide those behaviors. Hence, a higher level of human capital is likely to lead to positive synergy when it is coupled with a higher level of collective normative contract, which can be considered to represent a component of employee motivation. The interactive effect of ability and motivation on individual performance is relatively well established in the performance
44
literature (e.g., Austin et al., 1991; Kanfer & Ackerman, 1989). I expect a similar relationship at the unit-level of analysis. Thus, I expect the following: Hypothesis 9: Employees' collective normative contract with the organization moderates the relationship between the level of human capital and HR-related outcomes such that units with employees who have more favorable perceptions of the organization have higher levels of HR-related outcomes than those units with employees who do not perceive higher collective normative contract. In summary, the present dissertation proposes two sets of hypotheses that are associated with the level of human capital, i.e., Hypotheses 1 through 3, and collective normative contract, i.e., Hypotheses 5 through 7, along with mediating role of HRrelated outcomes on the relationships between the level of human capital and unit performance (Hypothesis 4), and collective normative contract and unit performance (Hypothesis 8). Finally, Hypothesis 9 posits the interactive effect of the level of human capital (as representing higher level construct of ability) and collective normative contract (as representing higher level construct of a component of motivation) on HRrelated outcomes. Table 2 summarizes all of the hypotheses advanced for the present dissertation. Table 2. Summary of Dissertation Hypotheses Hypothesis 1. High Performance Work System is positively related to the level of human capital. Hypothesis 2. The level of human capital is significantly related to HR-related outcomes. Hypothesis 3. The level of human capital mediates the relationships between High
45
Performance Work System and HR-related outcomes. Hypothesis 4. HR-related outcomes mediate the relationship between the level of human capital and unit performance. Hypothesis 5. High Performance Work System is positively related to collective normative contract. Hypothesis 6. Collective normative contract is significantly related to HR-related outcomes. Hypothesis 7. Collective normative contract mediates the relationships between High Performance Work System and HR-related outcomes. Hypothesis 8. HR-related outcomes mediate the relationships between collective normative contract and unit performance. Hypothesis 9. The level of human capital and collective normative contract interact to influence HR-related outcomes.
46
Chapter 4 Research Method Sample The sample for this dissertation includes 76 business units from 56 different companies located in Japan, ranging widely in terms of industries and geographical regions of Japan that the companies serve. Initially, 120 company contacts received a letter from a Japanese faculty member who provided assistance during the data collection procedure, inquiring about potential companies that may be willing to participate in the dissertation. Seventy-six companies were identified as having high likelihood of participating, given company contacts' relationships with these companies. Out of 76 companies, 56 companies provided at least one managerial and two employee responses that are usable for a company response rate of 73.68%. The names of the companies that were contacted for the dissertation are listed in Appendix B. In general, all but one (agriculture, forestry, and fishing) primary (SIC single-code) industry were represented. There were 23 units where only one managerial response was obtained. Table 3 summarizes the response distribution obtained for these seventysix units. Table 3. Response Patterns Number of Units 14 7 2 1 1 1 4 1 Managerial Responses 1 1 1 1 1 1 2 2 Employee Responses 234 578 23
47
3 1
2 2 2 3 3 3 3 3 3 4 4 4 4 5 5 5 6 6 7 7 7 7 8 8 10 12 14 15 19 26
4 61 161 32 42 74 81 111 121 41 61 81 91 31 61 181 61 111 61 71 161 261 91 201 41 61 101 181 31 48
The managerial sample consisted of 324 senior-level managers working in the establishments. The average number of managerial responses received for each establishment was 4.26 (ranged from 1 to 26). The managers, on average, had 19.02 years of work experience at the establishment (SD = 11.63), 10.16 years of job tenure (SD = 10.44), were 47.73 years old on average (SD = 7.91), were predominantly male (96%) and were at the level of middle management.
48
For the employee sample, the number of respondents was 525 with an average of 6.89 responses per establishment (ranged between 2 and 48). The employees had 8.05 years of average work experience at the establishment (SD = 8.12), 6.14 years of average job tenure (SD = 6.46), were 35.90 years old on average (SD = 8.43), were predominantly male (81%) and employed full-time (90%). Data Collection Procedure First, I obtained consent from the top management team members of the participating companies through the contacts that the Japanese faculty member had. Then, the human resources managers for the participating firms were contacted mostly through direct visit to the companies, followed up by either a phone call or e-mail to solicit their assistance in gathering the information necessary to send out the surveys. Data collection procedure involved surveying managers and employees at the establishment level. Given the nature of the data collection procedure and unwillingness of many of the companies to divulge employee attitudes information, managers heading the unit requested the number of managerial and employee surveys that they are willing and/or able to distribute and the Japanese faculty member reproduced all the necessary forms to create a packet that included a Japanese version of the survey (either managerial or employee version: listed in Appendix A), a cover letter, and a pre-paid, university-addressed envelope for the requested number. The self-addressed envelope was addressed to the faculty member's university address in Kunitachi-shi, Tokyo, Japan and all the surveys (i.e., both managerial and employee) were directly returned to this address. Therefore, the managers and employees who responded to the surveys were convenient samples selected by the managers. This might raise a concern that the
49
responses are biased upward because the manager might have selected only those managers and employees who would respond favorably to the questions. However, this concern is mitigated because the mean of the responses across the units were generally around the mid-point of the scale anchors (ranged from 3.42 to 5.15 out of 1 to 7 Likerttype anchors) and the standard deviations ranged between .52 and .83. Therefore, the responses do not seem overly skewed in a positive way (i.e., all favorable). In addition, units from Izumigou had lower ratings for High Performance Work System and the level of human capital than those obtained from the other units, which indicates that not all respondents chosen rated the items in a favorable manner. Both the managers and the employees who received the survey were told in the cover letter the purpose of the project. First, the importance of employees as a valuable asset to the organization was noted as the reason for the emergence of strategic human resource management as a field, which was used as a primer for the study objective. They were told that the main purpose of the project was to examine the relationship between a set of human resource management practices and firm performance as well as the mediating mechanism through which this occurs. They were also told that this project is also for completion of my dissertation. Furthermore, in the cover letter, the anonymity of their responses were guaranteed as the managerial and employee surveys distributed for the unit only had the code that identifies the unit, i.e., every survey for a particular unit had the same identification code, and voluntary nature of the responses were emphasized such that some managers and employees chose not to fill in demographic information if they felt uncomfortable in doing so.
50
Given the level of analysis involved in the dissertation, it was critical that high response rate be obtained from the managers as well as the employees to represent the unit (i.e., establishments or work locations). To ensure a higher response rate, I followed up with the managers in charge through phone calls and/or e-mails two weeks after the initial survey distribution. A second round of follow-up was conducted by mail with another sets of surveys enclosed. This was followed up again by phone calls and/or e-mails for all potential units. A final round of follow-up was conducted by mail with another survey packet. However, given that the managers requested the number of packets that he/she was willing to distribute, the representativeness of the respondents cannot be ascertained. Survey Translation Procedure The surveys were translated into Japanese through an iterative process. Several steps were taken to ensure the effectiveness of the translation process. First, a Japanese tenured full professor in Human Resource Management at the Tokyo Economics University in Kunitachi, Japan, who is also proficient in English, provided his assistance in translating the English version of the surveys into Japanese. This faculty member specializes in cross-cultural research on compensation between the U.S. and Japan and has also taught at the University of North Carolina at Chapel Hill and North Carolina State University for several years during his academic career. The translated versions of the surveys were sent back to me, the principal investigator. As a Japanese native, I checked the integrity and validity of the items and noted any concerns regarding the translation. These comments were sent back to the
51
Japanese faculty member along with the Japanese surveys. He then addressed these concerns and sent back the revised version of the Japanese surveys back to me, the Ph.D. candidate. This process was done several times until no further concern emerged. To validate the survey translation, two Japanese employees in no way affiliated with this study were asked to read through the Japanese surveys for readability and ease of comprehension. Any concerns were noted and sent back to the Japanese faculty member who further revised the surveys to address these concerns until no further concern surfaced. As a final check, a third Japanese native translated the survey back into English, and the Japanese and English versions were compared for any discrepancies; none was detected. In general, I followed the procedures recommended by Brislin (1981) for survey translations across different languages and found these surveys to be equivalent in meaning. Aggregation Issues The survey items were reworded to reflect the unit-level of analysis by changing the focus of items to the establishment. For instance, items for measuring psychological contract was reworded as the following: "The employees make personal sacrifices for this unit" with an instruction to answer the items for the average employees in the unit. This approach is consistent with the guidelines by multilevel scholars (House, Rousseau, & Thomas-Hunt, 1995; Klein, Dansereau, & Hall, 1994) to specify and explicate the level of constructs in the study. The interrater agreement was assessed via Rwg (James, Demaree, & Wolf, 1984, 1993) for all the variables included in the dissertation. The average Rwg for High Performance Work System was .96 (Median = .97) for the 30- item scale; for human capital, it was .97 (Median = .97); for collective
52
normative contract, it was .81 (Median = .88); for labor productivity, it was .90 (Median = .92); for unit-level task performance, it was .95 (Median = .96); for unit-level organizational citizenship behaviors, it was .97 (Median = .98); and for overall unit performance, it was .96 (Median = .97). These results provide some evidence that aggregation of the variables is justifiable because the Rwgs exceed the recommended value of .60 (James, 1982). More specifically, Schneider and Bowen (1985: 426) noted, "the appropriate test for withinsetting agreement would be a measure of homogeneity rather than an index like analysis of variance (ANOVA) or the intra-class correlation coefficient (ICC) that depend upon between setting differences for significance." In addition, Lindell and Brandt (1999) have shown Rwg's superiority as an index of inter-rater agreement when compared to other indices such as Content Validity Index (CVI: Lawshe, 1975) or T (Tinsley & Weiss, 1975). Furthermore, Rwg can be considered a measure of interchangeability among raters (Ryan et al., 1996), which is an important characteristic of this measure for the current study, given that I expected some establishments to have a single managerial response for the variables of interest. Given that the level of agreement among all of the raters is well above the recommended value of .60 by James (1982), and substantially exceeds more commonly accepted value of .70 (cf., Klein et al., 2000), there is some empirical evidence suggesting that one rater can reasonably measure the aggregated variables in an accurate manner. In addition, however, I examined intraclass correlation coefficients (ICC) to assess the variability in managerial and employee responses. ICC(1) compares the variance between units of analysis (establishments in this case) to the variance within
53
units of analysis using the individual ratings of each respondent (Schneider, White, & Paul, 1998). On the other hand, ICC(2) compares the between unit variability to that of within unit variability, using the average ratings of respondents within each unit (Bartko, 1976). ICC(1) and ICC(2) were calculated by using Bartko's (1976) formula with a one-way random-effects analysis of variance (ANOVA). The ICCs were calculated for all the measures used in the present dissertation. ICC(1) for High Performance Work System, human capital, collective normative contract, labor productivity, unit-level task performance, unit-level organizational citizenship behaviors, and overall unit performance was .23, .22, .16, .25, .20, .26, and .25, respectively. ICC(2) for these same set of variables was .68, .55, .56, .59, .52, .60, and .59, respectively. "Although there are no strict standards of acceptability for either ICC(1) or ICC(2) values, James (1982) reported a median ICC(1) value of .12 in the organizational literature, and Glick (1985) recommended an ICC(2) cutoff of .60" (Schneider et al., 1998: 155) that can be used as a rule-of-thumb to compare the ICC values obtained for the current dissertation. All of the ICC(1) values obtained for the dissertation exceeds the median value of .12 reported by James (1982) while ICC(2) values either approach or equal (in the cases of human capital, collective normative contract, and labor productivity, unit-level task performance, unit-level organizational citizenship behaviors, and overall unit performance), or exceed (in the cases of High Performance Work System) this value. Thus, all of the indicators appear to justify aggregation of the variables to the unit-level of analysis. Therefore, responses from all the managers in a unit, except where only one managerial response was obtained, were
54
averaged to create the unit-level variables for the level of human capital, perceived labor productivity, unit-level task performance, unit-level overall organizational citizenship behaviors, and unit performance. All responses from the employees for a unit were averaged to create the unit-level variables of High Performance Work System and collective normative contract. Justifications for the rating sources are provided below. Measures To maximize the content validity of the questionnaire, all of the items were obtained from an existing measures, except two items for labor productivity developed for the present study. Each scale contains multiple-items scales with 7-point Likert-type anchors ranging from strongly disagree (1) to strongly agree (7), unless otherwise noted. All of the items are affixed in Appendix C. To reduce potential for common method bias, managerial and employee responses were used to evaluate different scales. In general, managers are considered more reliable and valid source of performance data, i.e., unit performance (e.g., Gupta, 1987; Gupta & Govindarajan, 1986), as well as HR-related outcomes such as labor productivity (Way, 2002). In addition, supervisory ratings are frequently used to measure individual employee's task performance and organizational citizenship behaviors (e.g., LePine, Erez, & Johnson, 2002; Organ & Ryan, 1995). Thus, managerial ratings were used for these measures. In addition, managers were deemed more knowledgeable about the unit's level of human capital as compared to the employees. Therefore, managerial ratings of the level of human capital were also used. For collective normative contract, aggregate employee responses are more appropriate,
55
given the psychological origin of the concept. Given that managerial ratings were used for many of the variables included in the dissertation, employee responses of High Performance Work System were used to reduce potential for common method biases. In addition, the employees' perceptions of HR practices in use are more important in influencing their attitudes and behaviors than what the managers say. Therefore, aggregate employee ratings were used for High Performance Work System. High Performance Work System. Due to space limitation, Japanese faculty in Human Resource Management, initially, selected thirty items from Lepak and Snell's (2002) sixty-item HR practices scale that are deemed most appropriate for the Japanese companies. The Japanese faculty selected these items that would be considered a part of High Performance Work System in the Japanese context for which the employees will be familiar with. Thus, for example, items such as "employees are required to participate in cross-functional teams and networks" and "the selection process emphasizes their ability to collaborate and work in teams were not selected because these are typically used in Japanese companies. A system of HR practices used by the establishment, i.e., High Performance Work System, was obtained from the employees and aggregated to the unit level. To examine the factor structure of these items, exploratory factor analysis with principal component extraction was initially performed. Using a combination of KaiserGuttman rule of Eigen values greater than one (Guttman, 1954; Kaiser, 1970) and a Scree test (Chattell, 1966) of Eigen values plotted against factors indicated a singlefactor solution. More specifically, first factor had an Eigen value of 8.14, which explained 27.15% of variance. This was followed by Eigen values of 3.01, 1.98, 1.82,
56
1.56, 1.45, 1.34, 1.19, and 1.13. Given that I expected all of the items to compose High Performance Work System, I also conducted principal axis factoring where singlefactor solution is imposed. The result of the factor analysis was comparable to that of principal component analysis. However, the factor loadings of nine items were below .35. Therefore, I dropped those items that did not load above .35 from the analysis and run another principal axis factoring. The remaining twenty-one items loaded at or above .35 on the single factor and this factor explained 35.82% of variance in the items and had an Eigen value of 7.88, followed by additional Eigenvalues above 1 of 1.93, 1.62, 1.26, and 1.16. The factor loadings of these twenty-one items are depicted on Table 4. Items selected for High Performance Work System is described in Appendix C with asterisks (*). Table 4. Factor Loadings for 21-Item High Performance Work System Scale HPWS Items Job Design 2 Job Design 3 Job Design 5 Job Design 6 Selection 1 Selection 2 Selection 3 Selection 4 Selection 5 Training & Development 1 Training & Development 3 Training & Development 4 Training & Development 5 Performance Appraisal 1 Performance Appraisal 2 Performance Appraisal 3 Performance Appraisal 4 Compensation 1 Compensation 2 Compensation 3 Compensation 7 57 Factor Loading .37 .41 .62 .55 .65 .60 .35 .44 .60 .67 .58 .41 .68 .65 .68 .75 .75 .44 .68 .64 .50
The resulting 21-item scale had a reliability of .90. This alpha is comparable to the one that Lepak and Snell (2002) obtained for their commitment HR scale (? = .89). In addition, the mean Rwg of .96 for this scale was highly comparable to that of .97 obtained by Lepak and Snell (2002). Therefore, these results provide some evidence for the validity of the translation procedure and the resultant scale. Human capital. Managers assessed the level of human capital that each unit possesses, using a four-item scale from Youndt and Snell's (2001) intellectual capital scale. A sample item states, "Our employees working in the unit are highly skilled." Cronbach's alpha for this four-item human capital scale was .92. Collective normative contract. Five items taken from Shore, Tetrick, and Berksdale's (1999, as cited in Rupp & Cropanzano, 2002) psychological contract/social exchange scale is reworded to capture the notion of normative contract at the unit level of analysis (cf., Rousseau, 1995). Rupp and Cropanzano (2002) reworded the items to assess relational aspect of social exchange relationships for supervisor and organization and found alphas of .89 and .91. For this variable, employee responses were used, as employee response was deemed most appropriate because of the psychological nature of concept (cf., Rousseau, 1995). A reworded, sample item states, "Our relationship with the establishment continues to evolve and develop." The collective normative contract scale had a reliability of .86. HR-Related Outcomes Labor productivity. Three-item scale of perceived labor productivity was used to measure the labor productivity of the establishments, rather than an objective measure
58
of labor productivity, i.e., the logarithm of sales per employee, which is a widely used measure of productivity, due to the difficulty of collecting such objective measure at the unit-level (e.g., Gupta & Govindarajan, 1987). I supplemented an item from Way (2002) ("The employees' productivity is higher than those of my major competitors") with two additional items developed for the study ("The employees in this establishment produce outputs in an efficient manner," and "The employees' productive power is higher than those of the competitors"). The resulting three-item scale had a reliability of .91. Unit-level task performance. Five items used by Williams and Anderson (1991) to measure in-role performance were used to rate aggregate, unit level performance of the establishment. This measure of task/in-role performance has been used in several studies at the individual-level of analysis (e.g., Eisenberger, Armeli, Rexwinkel, Lynch, & Rhoades, 2001; Hui, Law, & Chen, 1999; Lynch, Eisenberger, & Armeli, 1999; Turnley, Bolino, Lester, & Bloodgood, 2003). The coefficient alphas for these studies ranged from .75 to .93. Managers rated the level of employee task performance for the unit as a whole. A sample item stated, "The employees in our unit perform tasks that are expected of them." The reliability of this scale was .82. Given that Hui et al.'s (1999) study of Chinese employees had the lowest reliabilities of .75 reported in previous studies for this scale, a reliability of .82 obtained at the unit-level of analysis for this dissertation compares favorably to that by Hui et al. (1999), which provides an additional evidence that the translation procedure was done properly and the resultant scale is valid.
59
Unit-level overall organizational citizenship behaviors. An abbreviated, ten-item version of the Williams and Anderson's (1991) organizational citizenship behavior scale was used to evaluate the level of organizational citizenship behaviors exhibited by the employees in the unit. The original items were created to tap into two different aspects of organizational citizenship behaviors that are targeted toward the individuals (OCBI) and the organization (OCBO). Williams and Anderson (1991) reported a reliability of .88 and .75 for OCBI and OCBO, respectively. Similarly, Turnley et al. (2003) reported a reliability of .83 and .88 for OCBI and OCBO, respectively. Three items from the organizational citizenship behavior targeted toward the organization was initially eliminated because of an anticipated confound between inrole performance and organizational citizenship behaviors. Managers rated the remaining eleven items for the entire unit. Given recent evidence that OCBI and OCBO do not have distinct antecedents or consequences (LePine et al., 2002) and relatively high correlation between OCBI and OCBO for the present dissertation (r = .66, p< .001), all the items, except one, were combined to create unit-level, overall organizational citizenship behaviors. One item had a cross-loading with other performance indicators and, hence, dropped from the scale. The result of this factor analysis is described after the unit performance measure below. The reliability of this ten-item scale was .87. Unit Performance Delaney and Huselid's (1996) eight-item perceived organizational performance measure was adapted to assess unit's level of performance compared to other units similar to the focal unit. An example item asks the manager to rate the establishment's
60
performance for the past three years as compared to that of similar establishments (i.e., "How would you compare the establishment's [performance] over the past 3 years to those of the other establishments that do the same kind of work?" Delaney and Huselid (1996) found a reliability of .85 for perceived organizational performance (for 590 firms) from a nationally representative sample of U.S. work establishments, because they obtained their data from the National Organization Survey, which was conducted in 1991 with support from the National Science Foundation. Gupta (1987) also used a similar comparative subjective measure of performance. Three items from this overall unit performance measure were dropped due to low loadings and/or high cross-loadings with other performance indicators, as described below. This five-item measure of unit performance had a reliability of .92. Given that the reliability of this scale is comparable to the ones found for the representative sample of U.S. establishments by Delaney and Huselid (1996), this provides some evidence that the survey translation was satisfactorily performed and the resulting scale is valid. Given that all the HR-related outcomes and unit performance were obtained from the managers, I conducted exploratory factor analysis with principal component extraction with oblique rotation to assess the factor structure and discriminant validity of the measures. I expected four-factor structures for unit-level task performance, unitlevel overall organizational citizenship behaviors, perceived labor productivity, and overall unit performance. Thus, four-factor structure was initially imposed. The result of the factor analysis is shown in Table 5. The four factors collectively explained 64.49% of variance with Eigenvalues of 10.81, 3.89, 1.85, and 1.52 for the first four factors. The items generally loaded significantly onto the expected construct with few exceptions.
61
Table 5. Results of Exploratory Factor Analysis on the HR-Related Outcomes and Unit Performance Factor 1 .01 .30 .03 .28 .26 .18 .00 -.03 -.02 .82 .59 .47 .71 .66 .63 .44 .59 .55 .36 .48 -.34 -.14 -.10 .09 .17 .18 .50 .47 10.81 38.60 Factor 2 -.09 -.26 -.38 .18 -.04 -.05 -.01 -.03 -.34 .00 -.06 .05 -.19 .04 -.13 -.47 .12 .16 .24 -.10 -.45 -.63 -.65 -.81 -.87 -.72 -.43 -.46 3.89 13.88 Factor 3 -.06 -.02 .15 .62 .38 .60 .64 .76 .58 -.18 .15 .12 -.06 .19 -.04 .15 .25 .37 .08 .07 .15 .17 .11 .02 .11 -.01 -.01 .05 1.85 6.60 Factor 4 -.90 -.61 -.64 -.39 -.42 -.16 -.10 .14 .14 -.19 -.04 .20 .07 -.09 -.00 .03 -.07 .03 -.14 -.22 -.39 -.29 -.23 -.07 .06 -.23 -.09 -.08 1.52 5.42
Labor Productivity 1 Labor Productivity 2 Labor Productivity 3 Task Performance 1 Task Performance 2 Task Performance 3 Task Performance 4 Task Performance 5 Task Performance 6 OCBI 1 OCBI 2 OCBI 3 OCBI 4 OCBI 5 OCBI 6 OCBI 7 OCBO 1 OCBO 2 OCBO 3 OCBO 4 Overall Unit Performance 1 Overall Unit Performance 2 Overall Unit Performance 3 Overall Unit Performance 4 Overall Unit Performance 5 Overall Unit Performance 6 Overall Unit Performance 7 Overall Unit Performance 8 Eigen Value % variance explained
Note: Principal axis factoring analysis with oblique rotation, 4-factor solution Using a factor loading at or above .35 and differences of at least .10 between crossloadings, an item was deleted from task performance and organizational citizenship behavior - interpersonal (OCBI) each, as well as three items from overall unit
62
performance. The resulting factor loadings were cleaner that explained 66.18% variance in the items with Eigenvalues of 8.61, 3.53, 1.80, and 1.29. The items retained for further analyses are listed in Appendix C with asterisks (*). Analytic Procedures A combination of hierarchical regression analysis and path analytic procedure in regression was used to examine individual hypotheses as well as the overall model. Although it would have been better to be able to conduct structural equation modeling analysis, the sample size of 76 units does not allow this type of analysis. Even if I use single indicators with reliability corrections (cf., Jöreskog & Sörbom, 1989) by fixing the loadings by the square root of the reliability of the scale (?r) and the measurement errors by the product of the unreliability by the variance, i.e., (1 - r)*SD2, it would still require, at least, 26 parameter estimates (18 structural path estimates, 1 factor variance, and 7 disturbance terms), not to mention possible covariances among the disturbance terms (for HR-related outcomes). The sample size to parameter estimate ratio of 5 to 1 is recommended as a minimum requirement for the structural equation modeling to derive accurate parameter estimates (Bentler & Chou, 1987) and the ratio for the dissertation does not reach this minimum (ratio = 2.92). Therefore, regression analyses were used to examine the hypotheses, and path analysis in regression is used to evaluate the overall model. By using the path analytic procedure, I do not account for potential unreliability of the measures, which may result in some inaccuracy with regard to the beta coefficients estimated. On the other hand, using a structural equation modeling without having enough sample size also may result in inaccurate parameter estimates. However,
63
given relatively high reliabilities for the variables used in the dissertation noted above, this may be less of an issue in this particular case. Hence, I chose to use path analytic procedure in regression to be able to model the interrelationships among the variables included in the model. In addition, following Cohen and Cohen (1983), I standardized the level of human capital and collective normative contract before creating interaction terms to reduce multicollinearity problems inherent in higher order terms.
64
Chapter 5 Results Table 6 shows the means, standard deviations, and correlations among the variables of interest. Unstandardized means and standard deviations of the variables for the level of human capital and collective normative contract are listed for informational purpose only because I standardized these variable with a z-score transformation for the analyses to create a product term for the moderated regression analyses. The first three variables are dummy coded, control variables for those companies who supplied multiple units. The first dummy coded variable (control 1) is for Mitsui Kinzoku (1 = units from Mitsui Kinzoku, 0 = all the other) with thirteen units. The second dummy coded variable (control 2) is for Izumigou (with four units) and the last dummy coded variable (control 3) is for Cannon (with five units). In addition, to examine the effect of the industry, I created six dummy coded variables (dummy 1: 1 = Construction, 0 = all the other; dummy 2: 1 = Manufacturing, 0 = all the other; dummy 3: 1 = Transportation, Communication, Electric, Gas, & Sanitary Services, 0 = all the other; dummy 4: 1 = Wholesale Trade, 0 = all the other; dummy 5: 1 = Retail Trade, 0 = all the other; and dummy 6: 1 = Finance, Insurance, & Real Estate, 0 = all the other) and entered these variables in step 1 of the regression analyses for perceived labor productivity, unit-level task performance, unit-level overall organizational citizenship behaviors, and overall unit performance. None of the beta coefficients for these six dummy coded variables achieved significance at .05 level. The results of this regression analyses are shown in Table 7.
65
Table 6. Descriptives of All the Variablesa Variables 1. Control 1 Mitsui Kinzoku 2. Control 2 Izumigou 3. Control 3 Cannon 4. High Performance Work System 5. The Level of Human Capitalb b 6. Collective Normative Contract 7. Perceived Labor Productivity 9. Unit-Level Task Performance 8. Unit-Level Organizational Citizenship Behavior 9. Overall Unit Performance M 0.17 0.05 0.05 4.15 4.60 4.50 4.59 1.94 SD 0.38 0.22 0.22 0.55 0.77 0.70 0.83 0.61 1 -0.11 -0.11 0.15 -0.09 0.14 -0.11 0.01 -0.06** -0.29* -0.24 -0.12 -0.02 -0.20 0.30** 0.29** 0.20 0.19** 0.38 2 3 4 5 6 7 8 9
0.25* 0.54 * 0.04** 0.16** 0.71 ** 0.22* 0.47 0.52 0.28 0.42 **
5.17 0.52 -0.11 -0.01** 0.17** 0.33** 0.35** 0.28* 0.39** 0.55** 4.68 0.72 -0.13 -0.37 0.30 0.26* 0.52** 0.23 * 0.67** 0.44** 0.28*
Note: a N = 76; b means and SDs are provided for informational purpose only
*
p?? .05; ** p < .01 (two-tailed)
67
Table 7. Regression Results for Industry Effectsa Dependent Variables Unit-Level Task Unit-Level Performance OCBs 1.85*** 5.16*** .02 .20 -.16 .11 -.13 .07 0.91 .07 1.46 .11 .08 .00 -.04 .02 .03 .06 0.15 .01
Variable Constant Step 1: Industry Controls Industry Dummy 1: Construction Industry Dummy 2: Manufacturing Industry Dummy 3: Transportation, Communication, Electric, Gas, & Sanitary Services Industry Dummy 4: Wholesale Trade Industry Dummy 5: Retail Trade Industry Dummy 6: Finance, Insurance, & Real Estate Overall F R2
Perceived Labor Productivity 4.70*** -.05 -.14 -.11 .17 .02 -.08
Overall Unit Performance 4.44*** .01 .17 .06 .25 .22 .10 0.91 .07
Note: a n = 76. Values are standardized estimates. None of steps or the betas was significant at * p?? .05 (two-tailed).
68
The directions of the correlations among the variables of interest were in the expected direction. For the dummy coded company variables, units from Izumigou Kabushiki Gaisha had negative correlations with the variables included in the dissertation while units from Cannon had positive correlations, in general. To examine the hypotheses, I ran several regressions. For Hypotheses 1, 2, 5, and 6, the proposed predictor was entered one at a time to assess the independent effect of the predictor. For Hypotheses 3, 4, 7, and 8, which posit mediating effects of certain variables, I followed Baron and Kenny's (1986) four-step procedure where 1) the direct effect of independent variable on the dependent variable is examined, which shows that there is an effect that may be mediated; 2) the direct effect of independent variable on the "proposed" mediator is examined to illustrate that the independent variable is related to the mediator; 3) the direct effect of mediator on the dependent variable is evaluated while the effect of independent variable is taken into account; and 4) the reduction in the beta coefficient associated with the independent variable when the effect of mediator is accounted for. The steps 3 and 4 essentially are tested in the same regression analysis. The procedure used to test mediating effect and associated results are explained and discussed in more detail in the section describing the appropriate hypotheses. All of the beta coefficients reported here are standardized ones. As depicted in Table 8, Hypothesis 1, which posited a positive relationship between High Performance Work System and the level of human capital that the unit possesses, was not supported (? = .14, p> .1). High Performance Work System explained only 2% of variance in the level of human capital for the unit (F = 1.22, p> .1) above and beyond the effect of controls in step 1 (R2 = .14, F = 4.01, p< .01). For
69
the level of human capital, two control variables had significant betas, -.24 (p < .05) for Izumigou and .27 (p < .05) for Cannon. Given that Izumigou is in the hospitality industry (i.e., hotels) and Cannon in the electronics industry, these results seem indicative of the level of human capital necessary for the lower-level employees' jobs for these companies. Table 8. Regression Results for Hypotheses 1 and 2a Dependent Variable Perceived Unit-Level Labor Task Productivity Performance -.09 .04 -.03 -.18 .18 .38***
Variable 1. Control 1 Control 2 Control 3 2a. High Performance Work Systems 2b. Level of Human Capital Overall F R2 ?F ?R2
Level of Human Capital -.08 -.24* .27**
UnitLevel OCB -.10 -.01 .15
.14
.75*** 3.35** .16 1.22 .02 19.33*** .52 70.67*** .48 .16
.44*** 9.25*** .34 17.55*** .10
.34** 2.78* .14 8.11**
Note: a n = 76. Values are standardized estimates.
b
Beta coefficients shown are standardized betas from each step. p?? .10;
*
+
p?? .05;
**
p?? .01; *** p?? .001 (two-tailed)
For Hypothesis 2, results indicated strong support for all three HR-related outcome variables considered. Particularly, the level of human capital was significantly,
70
positively related to perceived labor productivity of the unit (? = .75, p< .001;??F = 70.67, p< .001,??R2 = .48), unit-level task performance (? = .44, p< .001;??F = 17.55, p< .001,??R2 = .16), and unit-level organizational citizenship behaviors (? = .34, p< .01;??F = 8.11, p < .01,??R2 = .10), after the effect of company dummies are accounted for. For perceived labor productivity and unit-level organizational citizenship behaviors, control variables were not significantly related. For unit-level task performance, Cannon (control 3) had significantly, positive beta (.38, p< .001). For Hypothesis 3, the condition that the independent variable, i.e. High Performance Work System, is related to the mediator (second condition put forth by Baron and Kenny [1986] was not satisfied (see Hypothesis 1). Therefore, there is nothing that the High Performance Work System can mediate. Hence, Hypothesis 3 is not supported. To examine the mediating effect of HR-related outcomes on the relationship between the level of human capital and overall unit performance (Hypothesis 4), I followed the same procedure by Baron and Kenny (1986) noted above. Specifically, I ran separate regression analyses where I entered the level of human capital in the second step, following the three company dummies and, then, entering only one of the HR-related outcomes in the third step by itself. The results of the analyses are illustrated in Table 9. As shown in Table 9, Model 2, the level of human capital was significantly, positively related to overall unit performance, which satisfies the first condition that the independent variable is related to the dependent variable (? = .39, p< .001). The second condition that the independent variable be related to the mediator is satisfied by supporting evidence for Hypothesis 2, i.e., the level of human capital was significantly,
71
positively related to all three HR-related outcomes. The third condition that the mediator is related to the dependent variable, i.e., overall unit performance, was only satisfied for perceived labor productivity. Therefore, only perceived labor productivity has the potential to act as a mediator. When examining Model 3a, the standardized beta coefficient associated with the level of human capital became non-significant (? = -.15, p > .1) from that in Model 2 (?? = .39, p < .001). Therefore, perceived labor productivity acted as a full mediator of the relationship between the level of human capital and overall unit performance while the other two HR-related outcomes did not act as a mediator. Therefore, there results provide some support for Hypothesis 4. Hypotheses 5 through 8 dealt with the mediating mechanism from High Performance Work System to overall unit performance through collective normative contract and HR-related outcomes. For Hypothesis 5, the results depicted in Table 10 provide strong support for the hypothesis that posited a positive relationship between High Performance Work System and collective normative contract (? = .52, p< .001). High Performance Work System explained 22% additional variance (?F = 20.69, p< .001) over and above that accounted by the firm control variables (R2 = .07, F = 1.97, p > .1)
72
Table 9. Results for Hypothesis 4a Overall Unit Performanceb Model 2 Model 3a Model 3b -.11 -.08 -.12 -.28** -.39*** -.27** .16 .17 .11 .39*** -.15 .31** .73*** .18 10.38*** .37 14.92*** .13 22.92*** .62 46.50*** .25 8.99*** .39 2.55 .02
Variable Step 1. Control 1 Control 2 Control 3 Step 2. Level of Human Capital Step 3a. Perceived Labor Productivity Step 3b. Unit-Level Task Performance Step 3c. Unit-Level Organizational Citizenship Behaviors Overall F R2 Change in F 2 Change in R Note: a n = 76. Values are standardized estimates.
b
Model 1 -.14 -.37*** .26**
Model 3c -.10 -.29 .15 .35 .12 8.65*** .38 1.48 .01
7.43*** .24
Beta coefficients shown in Model 2 are standardized betas from step 1 and 2.
+
p?? .10;
*
p?? .05;
**
p?? .01; *** p?? .001 (two-tailed)
73
Hypothesis 6 posited that collective normative contract is positively related to HR-related outcomes. The results of the regression analyses provided moderate but consistent support for this hypothesis. Specifically, collective normative contract was marginally, significantly related to perceived labor productivity (? = .21, p< .1;??F = 3.08, p< .1,??R2 = .03) and unit-level task performance (? = .19, p< .1;??F = 3.03, p< .01,??R2 = .03) but significantly, positively related to unit-level overall organizational citizenship behaviors (? = .28, p< .05;??F = 5.88, p< .05,??R2 = .07). Table 10. Results for Hypotheses 5 and 6 Dependent Variables Perceived Unit-Level Labor Task Productivity Performance -.09 -.03 .18 .04 -.18 .38***
Variable Step 1. Control 1 Control 2 Control 3 Step 2a. High Performance Work System Step 2b. Collective Normative Contract Overall F R2 Change in F 2 Change in R
Collective Normative Contract .15 -.09 .21+
Unit-Level OCB -.10 -.01 .15
.52***
.21+ 7.40*** .29 21.97*** .22 1.64 .08 3.08+ .03
.19+ 4.82** .21 3.03+ .03
.28* 2.15+ .11 5.68* .07
Note: a n = 76. Values are standardized estimates.
b
Beta coefficients shown are standardized betas from each step.
+
p?? .10;
*
p?? .05;
**
p?? .01; *** p?? .001 (two-tailed)
74
Table 11. Results for Hypothesis 7 Perceived Labor Productivity Unit-Level Task Performance Model Model Model Model Model Model 1 2 3a 1 2 3b -.09 -.12 -.13 .04 -.02 -.02 -.03 .01 .01 -.18 -.08 -.08 *** .18 .13 .12 .38 .26 * .26 * .15 .05 .19 1.12 .04 1.17 .07 1.28 .02 1.32 .09 1.86 .02 5.27*** .18 7.30*** .29*** 11.17 .11 .37*** .36** .03 5.78*** .29 0.06 .00 0.92 .04 2.95* .14*** 8.76 .10 Unit-Level Overall OCB Model Model Model 1 2 3c -.10 -.16 -.17 -.01 .09 .08 .15 .04 .04 .36** .29* .15 2.63* .16 1.29 .02
Variable Step 1. Control 1 Control 2 Control 3 Step 2. High Performance Work System Step 3. Collective Normative Contract Overall F R2 Change in F 2 Change in R
Note: a n = 76. Values are standardized estimates.
b
Beta coefficients shown are standardized betas from the last steps.
+
p?? .10;
*
p?? .05;
**
p?? .01; *** p?? .001 (two-tailed)
75
Hypothesis 7, which posited the mediating role of collective normative contract, did not receive support. Table 11 shows the standardized betas associated with each step. Given that High Performance Work System is not significantly related to perceived labor productivity (? = .15, p> .1) when entered in the second step, collective normative contract cannot mediate the relationship between High Performance Work System and perceived labor productivity because this does not satisfy Baron and Kenny's (1986) first condition. Similarly, High Performance Work System was not significantly related to unit-level task performance (? = .03, p> .1) or unit-level organizational citizenship behaviors (? = .15, p> .1) when entered in the third step, which does not satisfy the third condition that the mediator is related to the dependent variable (Baron & Kenny, 1986). Thus, for unit-level task performance and unit-level organizational citizenship behaviors, there is no mediation. Hence, the results of these analyses do not support the hypothesis that collective normative contract acts as a mediator of the relationship between High Performance Work System and HR-related outcomes. Rather, collective normative contract appears to have indirect relationships to High Performance Works System and HR-related outcomes.
76
Table 12. Results for Hypothesis 8 Overall Unit Performanceb Model 2 Model 3a Model 3b -.17 -.09 -.17+ -.36*** -.36*** -.31** .23* .15 .13 .17 .04 .11 .62*** .30** 6.31*** .26 2.50 .03 22.16*** .61 63.38*** .35 7.03*** .33 7.58** .07
Variable Step 1. Control 1 Control 2 Control 3 Step 2. Collective Normative Contract Step 3a. Perceived Labor Productivity Step 3b. Unit-Level Task Performance Step 3c. Unit-Level Organizational Citizenship Behaviors Overall F R2 Change in F 2 Change in R Note: a n = 76. Values are standardized estimates.
b
Model 1 -.14 -.37*** .26**
Model 3c -.14 -.36*** .21* .11
7.427*** .23
.20+ 5.93*** .30 3.52+ .04
Beta coefficients shown in Model 2 are standardized betas from step 1 and 2.
+
p?? .10;
*
p?? .05;
**
p?? .01; *** p?? .001 (two-tailed)
77
To examine Hypothesis 8 that proposed the mediating role of HR-related outcomes on the relationship between collective normative contract and overall unit performance, I regressed overall unit performance onto collective normative contract to examine if the independent variable (i.e., collective normative contract) is related to the dependent variable (condition 1). The result of the regression analysis indicated that collective normative contract was not significantly related to overall unit performance (? = .17, p> .1). Therefore, HR-related outcomes do not have a relationship to mediate. Hence, there is no support for the mediation. Table 12 depicts these results. Therefore, Hypothesis 8 was not supported. Finally, the interaction hypothesis for the level of human capital and collective normative contract (Hypothesis 9) was tested by creating an interaction term, using a zscore transformed variables for the level of human capital and collective normative contract. The results of the moderated regression analyses where the level of human capital and collective normative contract was entered in Step 2 after controlling for the effect of company, followed by the introduction of the interaction term in Step 3, did not provide much support for this hypothesis. The beta coefficients associated with the interaction term was .10 (p > .1) when perceived labor productivity was the dependent variable, -.09 (p > .1) when unit-level task performance was the dependent variable, and -.21 (p < .1) when unit-level organizational citizenship behavior was the dependent variable. Therefore, Hypothesis 9 did not receive any support. Table 13 depicts these findings.
78
Table 13. Results for Hypothesis 9 Dependent Variablesb Perceived Labor Unit-Level Task Unit-Level Productivity Performance OCB -.09 .03 .18 .036 -.176 .377 -.10 .01 .15
Variable Step 1. Control 1 Control 2 Control 3 Step 2. the Level of Human Capital Collective Normative Contract Step 3. The Level of Human Capital* Collective Normative Contract Overall F R2 Change in F 2 Change in R
.75*** .23**
.44*** .21*
.35** .29**
-.10 15.81*** .58 1.34 .01
-.09 7.29*** .39 0.77 .01
-.21+ 3.75** .25 3.03+ .03
Note: a n = 76. Values are standardized estimates.
b
Beta coefficients shown are standardized betas from the last steps.
+
p?? .10;
*
p?? .05;
**
p?? .01; *** p?? .001 (two-tailed)
To examine the proposed model more in its entirety, path analytic procedure in regression was used. The results of this path analytic, regression analyses are shown in Table 14. The standardized beta coefficient values are shown from the last step of the analysis for each dependent variable along with the incremental variance (?R2) accounted for the last step.
79
Table 14. Path Analytic Regression Analyses Resultsa Dependent Variablesb Variablec 2. High Performance Work Systems 3. Level of Human Capital Collective Normative Contract 4. Level of Human Capital * Normative Contract 5. Perceived Labor Productivity Unit-Level Task Performance Unit-Level OCBs Overall F R2 Change in F 2 Change in R Level of Human Capital .14 Collective Normative Contract .52*** Perceived Labor Productivity -.11 .77 *** .28** -.14 .82*** .33*** -.13 Unit-Level Task Performance .28* .40*** .08 .26* .42*** .09 -.04 Unit-Level OCBs .22 .32** .19 Overall Unit Performance
-.02 .18 .38*** -.14 .03 .25 + -.08 -.17 .70*** .14 -.05 11.30*** .64 12.63*** .21
3.35** .16 1.33 .02
7.40*** .29 21.97*** .22
15.74*** 14.01*** .58 .59 42.19*** .52 2.09 .01
8.54*** .43 8.10*** .14
7.25*** .43 0.15 .00
3.66** .24 4.50* .10
3.47** .26 2.00 .02
Note: ab n = 76. Values are standardized estimates Beta coefficients shown are standardized betas from the last steps. c Beta coefficients for the controls are not shown for clarity. + p?? .10; * p?? .05; ** p?? .01; *** p?? .001 (two-tailed)
81
Figure 8. Path Analytic Results ? = .77, p < .001 Level of Human Capital (Managerial Rating) ? = .40, p < .001 Perceived Labor Productivity for the Unit (Managerial rating) ? = .70, p < .001
High Performance Work System (Employee Rating)
? = .28, p < .05
Unit-Level Task Performance (Managerial rating)
Overall Unit Performance (Managerial Rating)
? = .52, p < .001
? = .28, p < .01 Normative Contract (Employee Rating) ? = .32, p < .001 Unit-Level OCBs (Managerial rating)
Note: non-significant paths are not shown
82
Figure 8 illustrates the results of the path analysis in a schematic form. To reduce cluttering, Figure 8 contains only those standardized beta coefficients that are significantly related (p < .1 and below) to other variables. In addition, it includes the rating sources (either employee response aggregates or managerial response aggregates) for each variable. In Figure 8, the dotted lines illustrates the main effects of High Performance Work System on the HR-related outcomes after the effects of the level of human capital and collective normative contract are accounted for, i.e., beta coefficients from the third step as shown in Table 14. In general, the results depicted in Table 14 and Figure 8 are similar to the ones found when each variable was treated separately. Table 15 summarizes the results for all of the hypotheses advanced in this dissertation. In general, some of the hypotheses received moderate to strong support while others did not receive support Table 15. Summary Results for the Hypotheses Hypothesis 1. Hypothesis 2. Hypothesis 3. Hypothesis 4. High Performance Work System is positively related to the level of human capital - not supported The level of human capital is significantly related to HR-related outcomes - strongly supported (3 out of 3 HR-related outcomes) The level of human capital mediates the relationships between High Performance Work System and HR-related outcomes - not supported HR-related outcomes mediate the relationship between the level of human capital and unit performance - some support (1 out of 3 HRrelated outcomes, perceived labor productivity acted as a mediator) Hypothesis High Performance Work System is positively related to collective
83
5. Hypothesis 6.
normative contract - strongly supported Collective normative contract is significantly related to HR-related outcomes - moderately supported (significant for 1of the 3 HR-related outcomes and marginally significant for the other 2)
Hypothesis 7. Hypothesis 8. Hypothesis 9.
Collective normative contract mediates the relationships between High Performance Work System and HR-related outcomes - not supported HR-related outcomes mediate the relationships between collective normative contract and unit performance - not supported The level of human capital and collective normative contract interact to influence HR-related outcomes - very limited support was provided.
84
Chapter 6 Discussion The main objectives of the dissertation were to uncover the underlying mechanisms through which High Performance Work System affects unit performance by integrating two main theoretical perspectives, i.e., resource-based view of the firm and the behavioral perspective, that often have been employed and discussed but not directly tested. Although previous studies often implied that employees' human capital and behaviors are key to improving firm performance (e.g., Arthur, 1992, 1994; MacDuffie, 1995; Tsui et al., 1995; Wright et al., 1995), empirical examinations of the mediating mechanism, or what has been termed a "black box," through which a system of human resources management practices affects firm performance have only begun more recently (e.g., Batt, 2002; cf. AOM symposium: Unlocking the black box of strategic human resource management research; Burtons & O'Reilly, 2000; Chadwick, 2000; Wright, Gardner, Moynihan, & Park, 2000). Batt (2002: 587) echoed a similar concern more recently that "prior research is theoretically undeveloped and has not specified the mediating employee behaviors that explain the relationship between HR practices and performance." Applebaum et al. (2000) also noted that our understanding on the High Performance Work System's effects on the workers has been a major gap in the literature. The dissertation contributes to the strategic human resources management literature by explicating these mediating mechanisms. More specifically, the findings provide empirical support for the assertions of the resource-based view of the firm that human capital can be considered one resource that may have positive impact on (unit) performance (e.g., Barney, 1991) and of the
85
behavioral perspectives that employees' attitudes and behaviors are the critical mediating mechanisms through which High Performance Work System impact (unit) performance (e.g., Jackson et al., 1989; Schuler & Jackson, 1987). Similarly, it is one of the first studies in the area of strategic HRM to demonstrate that the previous use of the resource-based view of the firm and the behavioral perspectives (McMahan, Virick, & Wright, 1999; Wright & McMahan, 1992) is not unfounded, as it provides an empirical evidence that explicitly test the mediating role of the resources (the level of human capital) and employee attitudes and behaviors (collective normative contract) as well as HR-related outcomes. Although the level of human capital did not act as a mediator, it had significant, positive relationships to all of the HR-related outcomes. I found that there are, at least, one mediating mechanism where HR-related outcomes acted as mediators of the relationships between the level of human capital and overall unit performance and collective normative contract and overall unit performance. In addition, High Performance Work System was positively related to collective normative contract. Thus, High Performance Work System appears to have a positive influence on the shared perceptions regarding the terms of obligations or social exchange between establishment and the employees as a collective that have positive influence on perceived labor productivity of the unit. It may be the case that the increase in shared perceptions is accomplished through rigorous selection, higher incentives, better training and development, and appropriate performance appraisal and developmental feedback, among others, that attract and retain similar types of employees to the firm. Similarly, through better job design, better incentive structure, long-term focus on training and development and performance appraisal, it may be the
86
case that High Performance Work System improves upon the long-term relational aspects of employees' contract with the unit so that employees in general feel higher level of obligations to reciprocate. In fact, Wright and Snell (1991) noted that HRM practices contribute to both competence and behavior managements. They referred to competence management as those HRM practices such as aptitude tests and skill training programs that are aimed at ensuring high level of competency or human capital for a firm's workforce. Behavior management focuses on those HRM practices intended for controlling and coordinating employee behavior, which include those practices such as incentives and performance management. For instance, selection and recruitment practices can influence the level of knowledge, skills, and abilities new recruits possess whereas training and development affect the level of knowledge, skills, and abilities or competence existing employees possess. At the same time, compensation / reward practices and performance management practices typically influence the behaviors exhibited by the employees within the firm. Similarly, Huselid (1995) found two factors, skills and motivation, for the HR policy items he used. Although I did not find a relationship between High Performance Work System and the level of human capital, the influence of High Performance Work System on the level of human capital and employee attitudes seems logical and a larger sample size might have been needed to uncover such effect. The level of human capital and collective normative contract, in turn, have positive influence on HR-related outcomes so that the labor productivity of the unit increased, performance level for the unit improved, and overall organizational citizenship behaviors enriched.
87
Furthermore, perceived labor productivity acted as mediators of the relationships between the level of human capital and unit performance, and collective normative contract with unit performance. Although perceived labor productivity and turnover rate have been shown to act as mediators on the relationship between High Performance Work System and firm performance by Huselid (1995), this is one of the first studies that illustrate the mediating role of HR-related outcomes on the relationship between the level of human capital and/or collective normative contract, the central variables from the resource-based view of the firm and the behavioral perspective. These results extend the previous research by uncovering the underlying mechanisms that more accurately depict the processes that occur between High Performance Work System and unit performance. However, there are some findings (or non-findings) that indicate that the process is more complicated than those depicted in Figure 8, the proposed model, and the results shown in Table 14, the path analytic result. Perhaps, as Becker and Huselid (1998) proposed, there are additional mediating mechanisms between High Performance Work System and unit performance that are not captured in this dissertation. For example, the level of human capital and collective normative contract did not mediate the relationship between High Performance Work System and any of the HR-related outcomes, despite collective normative contract having relationships with High Performance Work System and perceived labor productivity. The level of human capital, on the other hand, was not related to High Performance Work System in the current dissertation. It may also be the case that some of these HR-related outcomes, although considered as a set in this dissertation, have interrelationships among each
88
other such that one variable may be an antecedent to the other. Due to the crosssectional nature of the dataset, it is not possible to discern these relationships empirically in more detail. However, if the findings from individual-level studies extend to the unit-level of analysis, unit-level task performance and organizational citizenship behaviors may behave as antecedents to unit-turnover rate or perceived labor productivity. For instance, task performance and organizational citizenship behaviors are typically considered antecedents to voluntary turnover of the employees (e.g., Van Scotter, 2000; Williams & Livingstone, 1994). For example, Van Scotter (2000) analyzed the data longitudinally and found task performance and organizational citizenship behaviors to predict turnover decisions. If these relationships hold at the unit-level of analysis, it might be beneficial to further differentiate HR-related outcomes further, although longitudinal data collection at the unit-level of analysis from multiple sources will be a significant endeavor. Related, there may be contingent factors that have impact on these relationships found for the dissertation (cf., Youndt et al., 1996; Wright & Sherman, 1999; Wright & Snell, 1998). For example, the relationship between the level of human capital and perceived labor productivity may be enhanced for firms following differentiation strategy, whereas for firms pursuing cost leadership strategy, the relationship between the level of human capital and perceived labor productivity may be non-significant. The difficulty of finding support for the contingency or fit perspective has been noted before (Wright & Snell, 1998; see also Delery & Doty, 1996) and part of the reasons for not finding this contingency or the moderating effect of strategy may stem from the fact that the relationships associated with the mediating variables were not considered, i.e.,
89
strategy does not moderate the relationship between High Performance Work System and labor productivity per se but it may moderate the relationship between the level of human capital and labor productivity. This possibility also highlights the importance of a mediation approach to strategic HRM. Therefore, future research should extend the current findings by collecting additional variables that act as mediators and/or moderators to explicate the relationships between High Performance Work System and (unit) performance further. Another non-significant finding relates to the proposed interaction between the level of human capital and collective normative contract (Hypothesis 9), which received limited support (one marginally significant interaction term out of three possible). It appears to be the case that the levels of human capital and collective normative contract have additive rather than an interactive effect on the HR-related outcomes. Although this interaction effect between ability and motivation is relatively more established at the individual-level of analysis in the individual task performance literature (e.g., Austin et al., 1991), perhaps, collective normative contract does not capture the "action" aspects of motivation (Kuhl, 1985). Kuhl distinguished between two motivational aspects: choice and action. "Choice motivation determines an individual's decision to engage in a specific behavior, whereas action, or control motivation, determines the individual's maintenance of effort and persistence" (Kirk & Brown, 2003: 41). Kanfer (1992) also noted a similar distinction between distal and proximal motivation processes where a notion of proximity to action is central to her framework. Proximal constructs, such as self-efficacy, have been seen to affect processes close to actual behavior such as translating intentions into behavior (Kirk & Brown, 2003), which is
90
similar to the notion of action motivation. On the other hand, distal constructs, such as need for achievement, have been argued to influence processes distant from actual behavior, such as intentions to behave (Kirk & Brown, 2003), which is similar to the notion of choice motivation. Therefore, team efficacy (referent-shift consensus model of self-efficacy: cf., Chan, 1998), defined as team members' aggregate belief and confidence in team's ability to mobilize resources for successful task completion (Chan, 1998), may be more proximal than collective normative contract. It might be the case that collective normative contract behaves in a more distal way such that team efficacy, for example, acts as more proximal construct that mediates the relationship between collective normative contract and HR-related outcomes. In addition, team efficacy may interact with the level of human capital to enhance the impact of ability on HR-related outcomes such that units with higher human capital level will obtain higher labor productivity or unit-level task performance level when team efficacy is also high. It may also the case that non-significant findings regarding the interaction effect of the level of human capital and collective normative contract is methodological, rather than substantive. As the difficulty of finding interaction effects with moderated regression analyses has been well recognized due, for example, to unreliability of measures and small sample size (e.g., Jaccard, Turrisi, & Wan, 1990; Jaccard & Wan, 1995). As noted in Aiken and West (1991), Busemeyer and Jones (1983) provided a formula to calculate the reliability of the cross-product terms. According to the formula, the reliability of the interaction terms used for the level of human capital and collective normative contract is .80. Hence, it is not the unreliability of the interaction terms that contributed to the lack of findings but it might have due to the relatively small sample
91
size obtained for the dissertation, although the sample size appears to be large enough for main effects with medium to strong effect sizes. Assuming that the effect size of the interaction is relatively small (.1 to .3), as typically appears to be the case, I had .14 to .76 power to detect significance at .05 significance level for a sample size of 76 (cf., Cohen & Cohen, 1983). The correlation between the interaction term and the HR-related outcomes ranged from .10 to .12. Therefore, for the interaction effects, it appears likely that I did not have enough power, given the sample size of 76 units collected for the dissertation. Therefore, the present results need to be tested and replicated in future studies that utilize larger sample sizes for the unit. Limitations The results of the dissertation should be interpreted in light of its limitations. I focus on permanent employees rather than on the core employees for the dissertation due to this permanent (vs. temporary) distinction being more salient and accepted in the Japanese business world. In addition, it is still consistent with some of the prior research conducted in the US (e.g., Jackson et al., 1989) that used this distinction. Thus, the results of the dissertation should be generalizable to other non-Japanese studies that utilize permanent employees as their sample. However, the findings could have been stronger if only core employees were utilized for aggregation. Henceforth, these findings may be considered more conservative estimates of the relationships investigated in the current dissertation. Second, the data collection procedure that was necessitated by the use of Japanese sample where it was extremely difficult to obtain employee attitudinal data,
92
the number of surveys distributed to the employees were dictated by the managers heading up the unit. It may be the case that the managers only chose those employees who would be most favorably predisposed to answer the questions in a positive manner. However, the mean of the responses obtained from the employees were 4.5 for collective normative contract (corresponding to neither agree nor disagree to somewhat agree) with a standard deviation of .70. The responses also had a range of 3.84 (minimum = 2.56 and maximum = 6.40) with no skewness in the distribution (skewness statistic = -.25). Therefore, it does not appear to be the case that employees who were selected responded to the questions in a positively biased manner. Therefore, when considering the responses as a whole, it appears more likely to be representative of the Japanese employees working in these units. However, future research should replicate the findings and assess if the responses for this dissertation are biased. Third, the results of the dissertation might have been even stronger if there was more variability among the units. Although ICC(1) values found in the dissertation (.16 to .61) exceeded the median values found in the organizational literature of .12 by James (1982), many of the ICC(2) values just approached the recommended value of .60 by Glick (1985 as cited in Schneider et al., 1998). Recall that ICC(1) compares the variance between units of analysis (establishments in this case) to the variance within units of analysis using the individual ratings of each respondent (Schneider, White, & Paul, 1998) while ICC(2) compares the between unit variability to that of within unit variability, using the average ratings of respondents within each unit (Bartko, 1976). Therefore, it appears that a range restriction might have been a problem in some cases that constrained the size of the correlation possible, which might have contributed to
93
some of the null findings. Therefore, future studies may design the research to include more units that are not performing well. Moreover, I cannot establish the causal relationship among the variables of interest due to the cross-sectional nature of the dissertation, although the sequences of variables depicted is consistent with Becker and Huselid (1998) and others. It may be the case that higher performing establishments are more able to institute High Performance Work System due perhaps to more slack resources that the establishments possess or more financial support that the establishments receive from the parent companies. However, this concern for reverse causality is mitigated somewhat because I found two sets of mediating effect. However, a longitudinal study that incorporates multi-wave design would be beneficial in explicating the causal relationships fully. Fourth and related, the lower variability of variables across units may be due to the focus on only the permanent employees, which included both core and support employee groups. It may be the case that more standardized High Performance Work System is used for the permanent employees across industries and companies that might contributed to a reduction in the overall variability obtained for the present dissertation, thereby constraining the size of the correlations found. Thus, another improvement for future research may be to collect data from both permanent and temporary employees from multiple units because High Performance Work System are less likely to be used for temporary employees and examine the variability of a system of HR practices as well as other variables of interest and how it affects the results. Finally, there may be some additional, methodological concerns with regard to the dissertation. First, all of the variables are perceptual in nature. Although the
94
difficulty of obtaining objective measure of performance at the unit-level of analysis has been well recognized (e.g., Gupta, 1987), it would have been ideal if an objective measure of unit performance were obtained. However, exploratory factor analysis clearly distinguished all the four outcome variables taken from the managers. Therefore, the relationships found between HR-related outcomes and unit performance does not appear to be affected by percept-percept bias. Related, potential for common method biases may be a concern as many of the variables are obtained from the managers that may contribute to inflating the relationships among variables. However, this concern is minimized by several factors. First, I use ratings from two different sources (managers and employees) and obtained multiple responses so that these responses can be aggregated to the unit level of analysis. Second, the pattern of the significant relationships found does not indicate any significant biases associated with one particular source. More specifically, the ratings for High Performance Work System were obtained from the employees, which were significantly related to both the level of human capital (managerial ratings) and collective normative contract (employee ratings). Similarly, both the level of human capital and collective normative contract were related to HR-related outcomes (managerial ratings) in a differential manner. Finally, mediation analyses reduce the concern further as support was found for both the level of human capital (manager rated) and collective normative contract (employee rated) for the relationship between High Performance Work System (employee rated) and HR-related outcomes (manager rated). Therefore, overall evidence appears to suggest that the common method bias did not have substantial influences on the results for this dissertation.
95
Implications for Research and Practice These limitations aside, the results of this dissertation have important implications primarily for scholars and for managers. First, it is one of the first studies that examined the linkages between High Performance Work System and (unit) performance. Finding these intermediate linkages helps scholars understand the processes further, which might be helpful in discovering additional mediators, moderators, or consequences at the different level of analysis. For example, by adopting a cross-level or multi-level perspective (Klein et al., 1994; Klein & Kozlowski, 2000; Ostroff & Bowen, 2000), the current results can be extended to examining individual consequences of High Performance Work System via the level of human capital and collective normative contract, or firm-level consequences of High Performance Work System through unit performance. For the cross-level, "disaggregation" model where the influences of High Performance Work System at the unit on individual employees are explored, collective normative contract may be one of the critical linkages to increasing normative commitment (Meyer & Allen, 1991), which "refers to a perceived obligation to stay with the organization" (Irving, Coleman, & Cooper, 1997: 444), which is important because previous empirical studies have had difficulty in finding unique antecedents that predict normative commitment (cf., Irving et al., 1997; Ko, Price, & Mueller, 1997; Meyer, Allen, & Smith, 1993). Therefore, cross-level linkages between High Performance Work System, its mediators, and individual-level attitudes and behaviors can be examined in the future that may reveal unexpected or unexplored relationships among these variables.
96
For the cross-level, "aggregation" model where the effects of High Performance Work System at the unit on financial indicators of firm performance are investigated, HR-related outcomes such as unit-level organizational citizenship behaviors, or unit innovative behaviors (not examined) may spill over to other units within the organization to facilitate effective functioning of the firm as a whole. Thus, as Katz and Kahn (1964: 133) noted, "the resources of people in innovation, in spontaneous cooperation, in protective and creative behavior are thus vital to organizational survival and effectiveness." In addition, given that HR-related outcomes focused on aggregate behaviors at the unit-level of analysis, there may be additional variables that act as mediators on the relationships between the level of human capital and collective normative contract, and HR-related outcomes, in addition to the ones already noted above. For example, borrowing from the micro organizational behavior literature, organizational commitment and job satisfaction have been found to relate to turnover, performance, and organizational citizenship behaviors (e.g., Cohen, 1993; Organ & Ryan, 1995; Tett & Meyer, 1993), which may act as additional aggregate variables that mediate these relationships. Thus, there are additional opportunities at the unit level of analysis to examine these processes further. One of the difficulties that previous empirical research has encountered in findings the moderating effects of strategy on the relationship between High Performance Work System and firm performance might be attributed to the existence of these mediating processes. It may be the case that strategy moderates the relationship between High Performance Work System and the level of human capital (for "buy"
97
strategy) or High Performance Work System and normative commitment (for "make" strategy: Miles & Snow, 1984) but not the other. Given the existence of these mediating processes, it is less likely to find significant moderating effect of strategy on the relationship between High Performance Work System to (unit) performance, as there was no direct relationship between these two when mediating variables are taken into account (see Figure 12). Therefore, the influence of organizational characteristics such as strategy may be better understood when these mediating processes are considered. For managers, the results of the dissertation may be helpful in facilitating organizational interventions to increase the benefit of High Performance Work System. For instance, if the organizational decision makers have not felt that the firm is reaping the benefit of the High Performance Work System to the maximum extent possible, the source of the problem may be diagnosed by examining the level of human capital and/or collective normative contract for the specific units in the organization. If it is the case that the unit does not have sufficient number of employees with high knowledge, skills, and abilities, components of High Performance Work System such as selection and recruitment or training and development may be revised to enhance the level of human capital for the units. On the other hand, if it is the employee attitudes that were identified as a source of problem, some components of High Performance Work System such as performance appraisal and compensation may be modified to improve employees' attitudes and consequent behaviors. In addition to devising the components of High Performance Work System, the results from the dissertation highlights the importance of both the level of human capital and collective normative contract on labor productivity. Therefore, managers need to be
98
aware of these linkages, at minimum, and be ready to provide support to strengthen these linkages. For example, given that the level of human capital is positively related to all three HR-related outcomes, managers may be able to utilize a combination of management by objectives and performance incentives to underscore the linkages between high performance and reward to entice employees with higher knowledge, skills, and abilities to work harder and smarter so that their contribution to the unit will be higher. In a similar vein, employees' collective normative contract may be enhanced if the organization can show that they value and respect employee contributions and that they care (i.e., perceived organizational support: Eisenberger et al., 1986; Eisenberger et al., 1990). Once the employees' perceive that the organization has provided them something that they value, they are more likely to reciprocate by exerting more effort on the task as well as providing more extra-role behaviors that are beneficial to the organization as a whole. Finally, given the strong relationship between labor productivity and overall unit performance, the highest return for the managers may reside in retaining highly skilled employees who also provide suggestions and helping to others to improve the efficiency and effectiveness of unit function. These employees are central in maintaining and improving overall unit performance. Thus, the benefits that accrue from retaining one such employee are likely to be substantially higher than retaining several mediocre employees. Conclusion In conclusion, the results of the dissertation contributes to the literature on strategic human resources management by opening up the black box to explicate the
99
mediating processes associated with High Performance Work System and performance with important implications for managers and scholars.
100
Appendix A. Japanese Versions of the Surveys
101
102
103
104
105
106
107
Appendix B. Industrial Affiliations of the Companies for Primary (One-Digit) SIC Code Agriculture, Forestry, & Fishing NONE Minin g Sumitomo Kinzoku Kouzan Construction (group coding 1) Kurogane Kousakujo Maeda Kensetsu Nihon Seitai Taihei Seisakujo Manufacturing (group coding 2) Ajinomoto Takara Corporation Aoinetsu Kougyou Asahi Kasei Amidasu Canon Daiku Denki Seisakujo Eiwa Kasei Kougyou Fuji Denki Fuji Kasei Gaikin Plant Honshuu Kagaku Ishikawajima Harima Jyuukougyou Itogumi Mokuzai Kanematsu Kao Koiwai Nyuugou Kyokuyou Suisan Kyouwa Hakkou - Iyakuhin Mazda Mitsubishi Gas Kagaku Gouseijushi Mitsubishi Gashin Kagaku Mitsubishi Jushi Mitsui Kagaku - Haisii Mitsui Kagaku Sanshi
108
Mitsui Kinzoku Mitsui Kinzoku Kougyou - Atsuen Kakou Jigyoubu Mitsui Kinzoku Kougyou - Ceramic Jigyousho Mitsui Kinzoku Kougyou - Douhaku Jigyou Honbu Mitsui Kinzoku Kougyou - Hachinohe Seirenjo Mitsui Kinzoku Kougyou - Hikoshima Seirenjo Mitsui Kinzoku Kougyou - Himaku Zairyou Jigyoubu Mitsui Kinzoku Kougyou - Kamioka Kougyoujo Mitsui Kinzoku Kougyou - MCS Mitsui Kinzoku Kougyou - Mitsuike Jimusho Mitsui Kinzoku Kougyou - Nirezaki Jigyousho Mitsui Kinzoku Kougyou - Paato Raito Jigyoubu Mitsui Kinzoku Kougyou - Sougou Kenkyuujo Mitsui Kinzoku Kougyou - Takehara Seitetsujo Mizakura Kyougyou Nihon Denchi Nihon Piranii Kougyou Nikkei Kinzoku Kogaisha Okinawa Dennou Okinawa Shokuryo Okinawa Yakult Riken Seiko Epson Shouwa Yakuhin Kakou Sony ST Kagaku Tagagi Kougyou Tama Chuuou Sangyou Tonbo Enpitsu Touei Foods Yaezan Okishoku Yamato Kagaku Kougyou Youkouzoku Kagakuhin Co. Yuuki Gousei Yakuhin Kougyou Transportation, Communication, Electric, Gas, & Sanitary Services (group coding 3)
109
Hokkaido Kuukou Hokusou Sekiyu Idemitsu Kousan JPower Systems Kokusai Cleaning Kousan Kaiyou Mitubishi Souko Nagase Sekiyu Nagoya Terebi Jigyou Nihon Contena Terminal Teikoku Tsuushin Kougyou Terebi Asahi Wholesale Trade (group coding 4) Canon Hanbai GE Yokokawa Medical Hikarigaoka Kousan Itohira Kousan Jetronix Kyouwa Medix Maruichi Sanshou Marunin Shouji Mitsubishi Shouji Miyako Okishoku Nakamura Bussan Nihon Chemical Shouji Nihon Kousan Okishoku Shouji Sekkou Shouji Tokuho Shouji Retail Trade (group coding 5) DIX Hitachi Taga Technology Hogi Medical KMS Korudia Nihon Acuroid
110
SE Labo SG Shinsei Shizuoka Gas Shuei Trading Takeo - Fukuoka Branch Touyou Kougei Finance, Insurance, & Real Estate (group coding 6) C.P.U. Global Insurance Juutaku Anshin Hoshou Mito Shouken Oioi Songai Hoken Ryuugin Hoshou Ryuukyuu D.C. Ryuukyuu Ginkou Ryuukyuu Lease Sanin Goudou Ginko Shinkou Shouken - Ootsu Branch Touden Kougyou Insurance Center Tougin Lease Services (group coding 7) A. Human Net Amurasu PTY Ltd. Aoi Promotion Asahi Beer ASM - Japan Big Ability Central Jyouho Center Consultanto Daiichi Lease (Service - Leasing) Diya-Tec E-Venture Support Forum Fuji Xerox Sougou Kyouiku Kenkyuujo Izumigou Appointment Center
111
Izumigou Plaza Hotel Toba Izumigou Yatsugatake Branch Izumigou Yokohama Branch Jinzai Haken Center Okinawa Kokubu Kokusai Jinji Kenkyuujo Leading Edge Inc. Mercer HR Consulting Moon Beach Resort Hotel Nihon Bousei Kougyou Nihon Gijutsu Iten Nihon Kyuushoku Service Nikkan Kougyou Koukokusha Nikkan Kougyou Service Center Nishii Associates Nomura Sougou Kenkyuujo Point Ryuukyu Biru Management Touwa Food Service Trans-Tec Ueno Trans-Tec US Visvalley Yatsugatake Izumigou Honsha YK Service Public Administration Shakai Hoken Roumu Jimusho Songai Hokenryouritsu Sanshutsu Kikou
112
Appendix C High Performance Work System: "To what extent do you agree with the following statements? When answering this section, please think about those HR practices used for the average employees in the establishment!" Job Design 1. jobs include a wide variety of tasks. 2. employees are involved in job rotation.* 3. employees are empowered to make decisions.* 4. employees have a high degree of job security. 5. employee are encourage to take up challenging assignments.* 6. employees' request for job assignments is taken into account in job rotations decisions.* Selection 7. selection is comprehensive (uses interviews, tests, etc.).* 8. selection involves screening many job candidates.* 9. selection focuses on selecting the best all around candidate, regardless of the specific job.* 10. selection emphasizes promotion from within.* 11. selection places priority on their potential to learn (e.g., aptitude).* 12. selection emphasizes their capacity to perform well right away. Training & Development 13. training is continuous.* 14. training uses outside instruction (seminars, conferences, etc.). 15. training programs are comprehensive.* 16. training programs strive to develop firm-specific skills/knowledge.* 17. the training programs emphasize on-the-job experiences.* Performance Management 18. performance is based on objective, quantifiable results.* 19. performance appraisals include management by objective with mutual goal setting.* 20. performance appraisal include developmental feedback.* 21. performance appraisal emphasize development of abilities/skills.* 22. performance appraisal emphasizes the output of the team. 23. performance appraisal is geared toward those who put in longer hours. Compensation & Incentives 24. compensation packages include an extensive benefits package.* 25. our compensations include high wages.* 26. the incentive system is tied to skill-based pays.* 27. the incentive system has a group-based component (gainsharing, etc.). 28. our incentive system values seniority. 29. our incentives include annual increases in salary. 30. are contingent on performance.*
113
Human Capital: "Our employees working in the unit?" 1. are highly skilled. 2. are widely considered the best in our industry. 3. are creative and bright. 4. are experts in their particular jobs and functions. Collective Normative Contact 1. The things employees do on the job today will benefit their standing in the long run. 2. Our relationship with the establishment continues to evolve and develop 3. We have significant opportunities to take on assignments that enhance our value 4. We don't mind working hard today—We know we will eventually be rewarded by our establishment 5. We try to look out for the best interest of the establishment because we can rely on our establishment to take care of us. HR-Related Outcomes Perceived Labor Productivity 1. The employees' productivity is higher than those of my major competitors.* 2. The employees in this establishment produce outputs in an efficient manner.* 3. The employees' productive power is higher than those of the competitors.* Unit-Level Task Performance: "The employees?" 1. adequately complete assigned duties.* 2. fulfill responsibilities specified in job description. 3. perform tasks that are expected of them.* 4. engage in activities that will directly affect their performance evaluation.* 5. neglect aspects of the job they are obligated to perform (reverse-coded).* 6. fail to perform essential duties (reverse-coded).* Unit-Level Organizational Citizenship Behaviors: "The employees in our unit?" 1. help others who have heavy work loads.* 2. help others who have been absent.* 3. assist supervisors with their work (when not asked).* 4. take time to listen to co-workers' problems and worries.* 5. go out of way to help new employees.* 6. take personal interest in other employees.* 7. pass along information to co-workers. 8. have attendance at work that is above the norm.* 9. give advance notice when unable to come to work.* 10. adhere to informal rules devised to maintain order.* 11. conserve and protect organizational property.* Subjective Unit Performance: "How would you compare the establishment's [???.] over the past 3 years to that of the other establishments that do the same kind of work?" 1. performance 2. quality of products, services, or program* 3. development of new products, services, or programs* 4. ability to attract essential employees*
114
5. 6. 7. 8.
ability to retain essential employees* satisfaction of customers or clients* relations between management and other employees relations among employees in general
115
REFERENCES Adams, J. S. 1965. Inequity in social exchange. In L. Berkowitz (Ed.), Advances in experimental social psychology, 2: 267-299. New York: Academic Press. Aiken, L. S., & West, S. G. 1991. Multiple regression: Testing and interpreting interactions. Newbury Park, CA: Sage. Allen, N. J., & Meyer, J. P. 1990. Organizational socialization tactics: A longitudinal analysis of links to newcomers' commitment and role orientation. Academy of Management Journal, 33: 847-858. Angle, H. L., & Perry, J. L. 1981. An empirical assessment of organizational commitment and organizational effectiveness. Administrative Science Quarterly, 26: 113. Applebaum, E., Bailey, T., Berg, P., & Kalleberg, A. L. 2000. Manufacturing advantage: Why high-performance work systems pay off. Ithaca, New York: Cornell University Press. Armeli, S., Eisenberger, R., Fasolo, P., & Lynch, P. 1998. Perceived organizational support and police performance: The moderating influence of socioemotional needs. Journal of Applied Psychology, 83: 288-297. Arthur, J. B. 1994. Effects of human resource systems on manufacturing performance and turnover. Academy of Management Journal, 37: 670-687. Arthur, J. B. 1992. The link between business strategy and industrial relations systems in American steel minimills. Industrial and Labor Relations Review, 45: 488506.
116
Austin, J. T., Villanova, P., Kane, J. S., & Bernardin, H. J. 1991. Construct validation of performance measures: Definitional issues, development, and evaluation of indicators. In G. R. Ferris (Ed.), Research in Personnel and Human Resources Management, 9: 159-233. Greenwich, CT: JAI Press. Baird, L., & Meshoulam, I. 1988. Managing two fits of strategic human resource management. Academy of Management Review, 13: 116-128. Balkin, D. B., & Gomez-Mejia, L. R. 1984. Determinants of R and D compensation strategies in the high tech industry. Personnel Psychology, 37: 635-650. Barnett, W. P., & Miner, A. S. 1992. Standing on the shoulders of others: Career interdependence in job mobility. Administrative Science Quarterly, 37: 262-281. Barney, J. 1991. Firm resources and sustained competitive advantage. Journal of Management, 17: 99-120. Baron, R. M. & Kenny, D. A. 1986. The moderator-mediator distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51: 1173-1182. Bartko, J. J. 1976. On various intraclass correlation reliability coefficients. Psychological Bulletin, 83, 762-765. Bass, B. 1990. Bass and Stogdill's handbook of leadership: Theory, research, and managerial applications (3rd ed.). New York: Free Press. Becker, B., & Gerhart, B. 1996. The impact of human resource management on organizational performance: Progress and prospects. Academy of Management Journal, 39: 779-801.
117
Becker, B. E., & Huselid, M. A. 1998. High Performance Work System and firm performance: A synthesis of research and managerial implications. In G. R. Ferris (Ed.), Research in personnel and human resources management, 16: 53-101. Greenwich, CT: JAI Press. Becker, G. S. 1964. Human capital. New York: Columbia University Press. Bentler, P. M. 1998. EQS for Windows 5.7b. Encino, CA: Multivariate Software, Inc. Bentler, P. M., & Chou, C. H. 1987. Practical issues in structural modeling. Sociological Methods & Research, 16: 78-117. Berman, S. L., Down, J., & Hill, C. W. L. 2002. Tacit knowledge as a source of competitive advantage in the National Basketball Association. Academy of Management Journal, 45: 13-31. Bettencourt, L. A., Gwinner, K. P., & Meuter, M. L. 2001. A comparison of attitude, personality, and knowledge predictors of service-oriented organizational citizenship behaviors. Journal of Applied Psychology, 86: 29-41. Bishop, J. W., Scott, K. D., & Burroughs, S. M. 2000. Support, commitment, and employee outcomes in a team environment. Journal of Management, 26: 11131132. Blau, P. 1964. Exchange and power in social life. New York: Wiley. Borman, W. C., & Motowidlo, S. J. 1997. Task performance and contextual performance: The meaning for personnel selection research. Human Performance, 10: 99-109.
118
Borman, W. C., & Motowidlo, S. J. 1993. Expanding the criterion domain to include elements of contextual performance. In N. Schmitt & W. C. Borman (Eds.), Personnel selection in organizations: 71-98. San Francisco: Jossey-Bass. Borman, W. C., White, L. A., & Dorsey, D. W. 1995. Effects of ratee task performance and interpersonal factors on supervisor and peer performance ratings. Journal of Applied Psychology, 80: 168-177. Brief, A. P., & Motowidlo, S. J. 1986. Prosocial organizational behaviors. Academy of Management Review, 11: 710-725. Brislin, R. W. 1981. Cross-cultural encounters: Face-to-face interaction. New York: Pergamon Brockner, J. 2002. Making sense of procedural fairness: How high procedural fairness can reduce or heighten the influence of outcome favorability. Academy of Management Review, 27: 58-76. Brockner, J., Tyler, T. R., & Cooper-Schneider, R. 1992. The influence of prior commitment on reactions to perceived unfairness: The higher they are, the harder they fall. Administrative Science Quarterly, 37: 241-261. Burton, M. D., & O'Reilly, C. A., III. 2000. The impact of high commitment work systems on IPOs: Additive or multiplicative effects? Paper presented at the 60th annual meeting of the Academy of Management, Toronto, Canada. Busemeyer, J. R., & Jones, L. E. 1983. Analyses of multiplicative combination rules when the causal variables are measured with error. Psychological Bulletin, 93: 549-562.
119
Carpenter, M. A., Sanders, W. G., & Gregersen, H. B. 2001. Bundling human capital with organizational context: The impact of international assignment experience on multinational firm performance and CEO pay. Academy of Management Journal, 44: 493-511. Cattell, R.B. 1966. The scree test for the number of factors. Multivariate Behavioral Research, 1: 245-276. Chadwick, C. D. 2000. Empirical insights on the origins of synergies in strategic human resource management. Paper presented at the 60th annual meeting of the Academy of Management, Toronto, Canada. Chadwick, C., & Cappelli, P. 1999. Alternatives to generic strategy typologies in strategic human resource management. In P. M. Wright, L. Dyer, J. Boudreau, & G. Milkovich (Eds.), Research in personnel and human resources management, Supplement 4: 1-29. Greenwich, CT: JAI Press. Chan, D. 1998. Functional relations among constructs in the same content domain at different levels of analysis: A typology of composition models. Journal of Applied Psychology, 83: 234-246. Chen, X.-P., Hui, C., & Sego, D. J. 1998. The role of organizational citizenship behavior in turnover: Conceptualization and preliminary tests of key hypotheses. Journal of Applied Psychology, 83: 922-931. Colquitt, J. A. 2001. On the dimensionality of organizational justice: A construct validation of a measure. Journal of Applied Psychology, 86: 386-400.
120
Colquitt, J. A., Conlon, D. E., Wesson, M. J., Porter, C. O. L. H., & Ng, K. Y. 2001. Justice at the millennium: A meta-analytic review of 25 years of organizational justice research. Journal of Applied Psychology, 86: 425-445. Colquitt, J. A., Noe, R. A., & Jackson, C. L. 2002. Justice in teams: Antecedents and consequences of procedural justice climate. Personnel Psychology, 55: 83-109. Cohen, A. 1993. Organizational commitment and turnover: A meta-analysis. Academy of Management Journal, 36: 1140-1157. Cohen, J., & Cohen, P. 1983. Applied multiple regression/correlation analysis for the behavioral science (2nd ed.). Hillsdale, New Jersey: Lawrence Erlbaum Associates, Publishers. Coff, R. W. 1997. Human assets and management dilemmas: Coping with hazards on the road to resource-based theory. Academy of Management Review, 23: 374-402. Cropanzano, R. 2001. Justice in the workplace: From theory to practice, Vol. 2. Mahwah, NJ: Lawrence Erlbaum. Cropanzano, R. 1993. Justice in the workplace: Approaching fairness in human resource management. Hillsdale, NJ: Lawrence Erlbaum. Cropanzano, R., & Greenberg, J. 1997. Progress in organizational justice: Tunneling through the maze. In C. L. Cooper & I. T. Robertson (Eds.), International Review of Industrial and Organizational Psychology, 12: 317-372. New York: Wiley. Cropanzano, R., & Kacmar, M. 1995. Organizational politics, justice, and support: Managing the social climate of the workplace. Westport, CT: Quoram Books.
121
Davis-Blake, A., & Uzzi, B. 1993. Determinants of employment externalization: A study of temporary workers and independent contractors. Administrative Science Quarterly, 38: 195-223. Davis, P. S., Robinson, R. B., Jr., Pearce, J. A., II, & Park, S. H. 1992. Business unit relatedness and performance: A look at the pulp and paper industry. Strategic Management Journal, 5: 349-361. Dawis, R. V. 2000. P-E fit as paradigm: Comments on Tinsley (2000). Journal of Vocational Behavior, 56: 180-183. Delaney, J. T., & Huselid, M. A. 1996. The impact of human resource management practices on perceptions of organizational performance. Academy of Management Journal, 39: 949-969. Delery, J. E. 1998. Issues of fit in strategic human resource management: Implications for research. Human Resource Management Review, 8: 289-309. Delery, J. E., & Doty, D. H. 1996. Modes of theorizing in strategic human resource management: Tests of universalistic, contingency, and configurational performance predictions. Academy of Management Journal, 39: 802-835. Dollinger, M. J., & Golden, P. A. 1992. Interorganizational and collective strategies in small firms: Environmental effects and performance. Journal of Management, 18: 695-715. Doty, D. H., Gulick, W. H., & Huber, G. P. 1993. Fit, equifinality, and organizational effectiveness: A test of two configurational theories. Academy of Management Journal, 36: 1196-1250.
122
Drexler, J. A. 1977. Organizational climate: Its homogeneity within organizations. Journal of Applied Psychology, 62: 38-42. Dunnette, M. D., Lawler, E. E., & Weick, K. E. 1967. The role of financial compensation in managerial motivation. Organizational Behavior & Human Decision Processes, 2: 175-216. Dyer, L., & Reeves, T. 1995, May. Human resource strategies and firm performance: What do we know and where do we need to go? Paper presented at the 10th World Congress of the International Industrial Relations Association, Washington, D.C. Edwards, J. R., & Van Harrison, R. 1993. Job demands and worker health: Three-dimensional reexamination of the relationship between person-environment fit and strain. Journal of Applied Psychology, 78: 628-648. Eisenberger, R., Armeli, S., Rexwinkel, B., Lynch, P. D., & Rhoades, L. 2001. Reciprocation of perceived organizational support. Journal of Applied Psychology, 86: 42-51. Eisenberger, R., Cummings, J., Armeli, S., & Lynch, P. 1997. Perceived organizational support, discretionary treatment, and job satisfaction. Journal of Applied Psychology, 82: 812-820. Eisenberger, R., Fasolo, P., & Davis-LaMastro, V. 1990. Perceived organizational support and employee diligence, commitment, and innovation. Journal of Applied Psychology, 75: 51-59. Eisenberger, R., Huntington, R., Hutchison, S., & Sowa, D. 1986. Perceived organizational support. Journal of Applied Psychology, 71: 500-507.
123
Folger, R., & Cropanzano, R. 1998. Organizational justice and human resource management. Thousand Oaks, CA: Sage. Gattiker, U. E. 1995. Firm and taxpayer returns from training of semiskilled employees. Academy of Management Journal, 38: 1152-1173. George, J. 1991. State or trait: Effects of positive mood on prosocial behavior at work. Journal of Applied Psychology, 76: 299-307. George, J. M. 1990. Personality, affect, and behavior in groups. Journal of Applied Psychology, 75: 107-116. George, J. M., & Bettenhausen, K. 1990. Understanding prosocial behavior, sales performance, and turnover: A group-level analysis in a service context. Journal of Applied Psychology, 75: 698-709. George, J. M., & Brief, A. P. 1992. Feeling good - doing good: A conceptual analysis of the mood at work - organizational spontaneity relationship. Psychological Bulletin, 112: 310-329. George, J. M., & Jones, G. R. 1997. Organizational spontaneity in context. Human Performance, 10: 153-170. Gerhart, B., & Milkovich, G. T. 1990. Organizational differences in managerial compensation and financial performance. Academy of Management Journal, 33: 663691. Gerstner, C. R., & Day, D. V. 1997. Meta-analytic review of leader-member exchange theory: Correlates and construct issues. Journal of Applied Psychology, 82: 827-844.
124
Gilliland, S., Steiner, D., & Skarlicki, D. 2001. Theoretical and cultural perspectives on organizational justice. Greenwich, CT: Information Age Publishing. Glick, W. H. 1985. Conceptualizing and measuring organizational and psychological climate: Pitfalls in multilevel research. Academy of Management Review, 10: 601-616. Graen, G. 1976. Role-making processes within complex organizations. In M. D. Dunnette (Ed.), Handbook of industrial and organizational psychology: 1201-1245. Chicago: Rand McNally. Greenberg, J. 1993. Stealing in the name of justice: Informational and interpersonal moderators of theft reactions to underpayment inequity. Organizational Behavior and Human Decision Processes, 54: 81-103. Greenberg, J. 1990. Organizational justice: Yesterday, today, and tomorrow. Journal of Management, 16: 399-432. Greenberg, J. 1987. Reactions to procedural justice in payment distributions: Do the means justify the ends? Journal of Applied Psychology, 72: 55-61. Greenberg, J., & Cropanzano, R. 2001. Advances in organizational justice. Stanford, CA: Stanford Business Books. Griffeth, R. W., Hom, P. W., & Gaertner, S. 2000. A meta-analysis of antecedents and correlates of employee turnover: Update, moderator tests, and research implications for the next millennium. Journal of Management, 26: 463-488. Gouldner, A. W. 1960. The norm of reciprocity. American Sociological Review, 25: 161-178.
125
Guillen, M. F. 2000. Business groups in emerging economies: A resource-based view. Academy of Management Journal, 43: 362-380. Guion, R. M. 1965. Personnel testing. New York: McGraw-Hill. Gupta, A. K. 1987. SBU strategies, corporate-SBU relations, and SBU effectiveness in strategy implementation. Academy of Management Journal, 30: 477500. Gupta, A. K., & Govindarajan, V. 1986. Resource sharing among SBUs: Strategic antecedents and administrative implications. Academy of Management Journal, 29: 695-714. Guthrie, J. P. 2001. High-involvement work practices, turnover, and productivity: Evidence from New Zealand. Academy of Management Journal, 44: 180192. Guthrie, J. P., & Olian, J. D. 1991. Does context affect staffing decisions? The case of general managers. Personnel Psychology, 44: 263-292. Guttman, L. 1954. Some necessary conditions for common factor analysis. Psychometrika, 19: 149-161. Guzzo, R. A., Noonan, K. A., & Elron, E. 1994. Expatriate managers and the psychological contract. Journal of Applied Psychology, 79: 617-626. Guzzo, R. A., & Shea, G. P. 1992. Group performance and intergroup relations in organizations. In M. D. Dunnette & L. M. Hough (Eds.), Handbook of industrial and organizational psychology (2nd Ed.), 3: 1-45. Palo Alto, CA: Consulting Psychologists Press.
126
Hackman, J. R., & Oldham, G. R. 1980. Work redesign. Reading, MA: AddisonWesley. Harrison, D. A., & Martocchio, J. J. 1998. Time for absenteeism: A 20-year review of origins, offshoots, and outcomes. Journal of Management, 24: 305-350. Hitt, M. A., & Barr, S. H. 1989. Managerial selection decision models: Examination of configural cue processing. Journal of Applied Psychology, 74: 53-61. Hitt, M. A., Bierman, L., Shimizu, K., & Kochhar, R. 2001. Direct and moderating effects of human capital on strategy and performance in professional service firms: A resource-based perspective. Academy of Management Journal, 44: 13-28. Hom, P. W., Caranikas-Walker, F., Prussia, G. E., & Griffeth, R. W. 1992. A meta-analytical structural equations analysis of a model of employee turnover. Journal of Applied Psychology, 77: 890-909. Homans, G. C. 1958. Social behaviors as exchange. American Journal of Sociology, 63: 597-606. Homans, G. C. 1961. Social behavior. New York: Harcourt, Brace, and World. House, R. Rousseau, D. M., & Thomas-Hunt, M. 1995. The meso paradigm: A framework for the integration of micro and macro organizational behavior. In L. L. Cummings & B. M. Staw (Eds.), Research in organizational behavior, 17: 71-114. Greenwich, CT: JAI Press. Hui, C., Law, K. S., & Chen, Z. X. 1999. A structural equation model of the effects of negative affectivity, leader-member exchange, and perceived job mobility on in-role and extra-role performance: A Chinese case. Organizational Behavior & Human Decision Processes, 77: 3-21.
127
Hulin, C. 1991. Adaptation, persistence, and commitment in organizations. In M. D. Dunnette & L. M. Hough (Eds.), Handbook of Industrial and Organizational Psychology (2nd ed.), Vol. 2: 445-505. Palo Alto, CA: Consulting Psychological Press. Huselid, M. A. 1995. The impact of human resource management practices on turnover, productivity, and corporate financial performance. Academy of Management Journal, 38: 635-672. Iaffaldano, M. T., & Muchinsky, P. M. 1986. Job satisfaction and job performance: A meta-analysis. Psychological Bulletin, 97: 251-273. Ichniowski, C., Shaw, K., & Prennushi, G. 1997. The effects of human resource management practices on productivity: A study of steel finishing lines. The American Economic Review, 87: 291-313. Igalens, J., & Roussel, P. 1999. A study of the relationships between compensation package, work motivation and job satisfaction. Journal of Organizational Behavior, 20: 1003-1025. Irving, P. G., Coleman, D. F., & Cooper, C. L. 1997. Further assessments of a three-component model of occupational commitment: Generalizability and differences across occupations. Journal of Applied Psychology, 82: 444-452. Jaccard, J., Turrisi, R., & Wan, C. K. 1990. Interaction effects in multiple regression analysis. Newbury Park, CA: Sage. Jaccard, J., & Wan, C. K. 1995. Measurement error in the analysis of interaction effects between continuous predictors using multiple regression: Multiple indicator and structural equation approaches. Psychological Bulletin, 117: 348-357.
128
Jackson, S. E., Schuler, R. S., & Rivero, J. C. 1989. Organizational characteristics as predictors of personnel practices. Personnel Psychology, 42: 727-786. James, L. R. 1982. Aggregation bias in estimates of perceptual agreement. Journal of Applied Psychology, 67: 219-229. James, L. R., Demaree, R. G., & Wolf, G. 1993. Rwg: An assessment of withingroup agreement. Journal of Applied Psychology, 78: 306-309. James, L. R., Demaree, R. G., & Wolf, G. 1984. Estimating within-group interrater reliability with and without response bias. Journal of Applied Psychology, 69: 85-98. James, L. R., Joyce, W. F., & Slocum, J. W., Jr. 1988. Comment: Organizations do not cognize. Academy of Management Review, 13: 129-132. Johns, G., & Xie, J. L. 1998. Perceptions of absence from work: People's Republic of China versus Canada. Journal of Applied Psychology, 83: 515-530. Jöreskog, K. G., & Sörbom, D. 1989. LISREL VII: User's reference guide. Mooresville, IN: Scientific Software. Joyce, W. F., & Slocum, J. W., Jr. 1990. Strategic context and organizational climate. In B. Schneider (Ed.), Organizational climate and culture: 130-150. San Francisco: Jossey-Bass. Kaiser, H.F. 1970. Second generation Little Jiffy. Psychometrika, 35: 401-417. Kanfer, R. 1992. Work motivation: New directions in theory and research. In C. L. Cooper & I. T. Robertson (Eds.), International review of industrial and organizational psychology, 7: 1-53. Chichester, England: Wiley.
129
Kanfer, R., & Ackerman, P. L. 1989. Motivation and cognitive abilities: An integrative/aptitude-treatment interaction approach. Journal of Applied Psychology, 74: 657-690. Kanfer, R., Crosby, J. V., & Brandt, D. M. 1988. Investigating behavioral antecedents of turnover at three job tenure levels. Journal of Applied Psychology, 73: 331-335. Katz, D., & Kahn, R. L. 1964. The motivational basis of organizational behavior. Behavioral Science, 9: 131-146. Kirk, A. K, & Brown, D. F. 2003. Latent constructs of proximal and distal motivation predicting performance under maximum test conditions. Journal of Applied Psychology, 88: 40-49. Klein, K. J., Bliese, P. D., Kozlowski, S. W. J., Dansereau, F. Gavin, M. B., Griffin, M. A., Hofmann, D. A., James, L. R., Yammarino, F. J., & Bligh, M. C. 2000. Multilevel analytical techniques: Commonalities, differences, and continuing questions. In K. J. Klein & S. W. J. Kozlowski (Eds.), Multilevel theory, research, and methods in organizations: Foundations, extensions, and new directions: 512-553. San Francisco: Jossey-Bass. Klein, K. J., Dansereau, F., & Hall, R. J. 1994. Levels issues in theory development, data collection, and analysis. Academy of Management Review, 19: 195229. Klein, K. J., & Kozlowski, S. W. J. 2000. From micro to meso: Critical steps in conceptualizing and conducting multilevel research. Organizational Research Method, 3: 211-236.
130
Kleinbeck, U., & Fuhrmann, H. 2000. Effects of a psychologically based management system on work motivation and productivity. Applied Psychology: An International Review, 49: 596-610. Ko, J.-K., Price, J. L., & Mueller, C. W. 1997. Assessment of Meyer and Allen's three-component model of organizational commitment in South Korea. Journal of Applied Psychology, 82: 961-973. Koch, M. J., & McGrath, R. G. 1996. Improving labor productivity: Human resource management policies do matter. Strategic Management Journal, 17: 335-354. Konovsky, M. A., & Pugh, S. D. 1994. Citizenship behavior and social exchange. Academy of Management Journal, 37: 656-669. Koys, D. J. 2001. The effects of employee satisfaction, organizational citizenship behavior, and turnover on organizational effectiveness: A unit-level, longitudinal study. Personnel Psychology, 54: 101-114. Kozlowski, S. W. J., & Hattrup, K. 1992. A disagreement about within-group agreement: Disentangling issues of consistency versus consensus. Journal of Applied Psychology, 77: 161-167. Kozlowski, S. W. J., & Klein, K. J. 2000. A multilevel approach to theory and research in organizations: Contextual, temporal, and emergent processes. In K. J. Klein & S. W. Kozlowski (Eds.), Multilevel theory, research, and methods in organizations: Foundations, extensions, and new directions: 3-90. San Francisco: Jossey-Bass. Kristof, A. L. 1996. Person-organization fit: An integrative review of its conceptualizations, measurement, and implications. Personnel Psychology, 49: 1-49.
131
Kristof-Brown, A. L. 2000. Perceived applicant fit: Distinguishing between recruiters' perceptions of person-job and person-organization fit. Personnel Psychology, 53: 643-671. Lado, A. A., & Wilson, M. C. 1994. Human resource systems and sustained competitive advantage: A competency-based perspective. Academy of Management Review, 19: 699-727. Lambert, S. J. 2000. Added benefits: The link between work-life benefits and organizational citizenship behavior. Academy of Management Journal, 43: 801-815. Lane, P. J., & Lubatkin, M. 1998. Relative absorptive capacity and interorganizational learning. Strategic Management Journal, 19: 461-477. Lawler, E. E. III. 1981. Pay and organizational development. Reading, MA: Addison-Wesley. Lawshe, C. H. 1975. A quantitative approach to content validity. Personnel Psychology, 28: 563-575. Leifer, R., O'Connor, G. C., & Rice, M. 2001. Implementing radical innovation in mature firms: The role of hubs. Academy of Management Executive, 15: 102-113. Lepak, D. P. 2001, August. A conceptual examination of the link between strategy and human resource practices: The mediating effects of employee group and human resource philosophy. Paper presented at the 61st Annual meeting of the Academy of Management, Washington D.C. Lepak, D. P., & Snell, S. A. 2002. Examining the human resource architecture: The relationships among human capital, employment, and human resource configurations. Journal of Management, 28: 517-543.
132
Lepak, D. P., & Snell, S. A. 1999. The human resource architecture: Toward a theory of human capital allocation and development. Academy of Management Review, 24: 31-48. Lepak, D. P., Taylor, M. S., Tekleab, A. G., Marrone, J. A., & Cohen, D. J. 2002, April. Understanding the use and performance implications of high investment HR systems for different employee groups. Paper presented at the Strategic HRM conference at the University of Bath, Bath, England. LePine, J. A., Erez, A., & Johnson, D. E. 2002. The nature and dimensionality of organizational citizenship behavior: A critical review and meta-analysis. Journal of Applied Psychology, 87: 52-65. LePine, J. A., & Van Dyne, Linn. 2001. Voice and cooperative behavior as contrasting forms of contextual performance: Evidence of differential relationships with Big Five personality characteristics and cognitive ability. Journal of Applied Psychology, 86: 326-336. Leventhal, G. S., Karuza, J., & Fry, W. R. 1980. Beyond fairness: A theory of allocation preferences. In G. Miikula (Ed.), Justice and Social Interaction: 167-218. New York: Springer-Verlag. Liden, R. C., Sparrowe, R. T., & Wayne, S. J. 1997. Leader-member exchange theory: The past and potential for the future. In G. R. Ferris (Ed.), Research in Personnel and Human Resource Management, 15: 47-119. Greenwich, CT: JAI Press. Lind, E. A., & Tyler, T. R. 1988. The social psychology of procedural justice. New York: Plenum Press.
133
Lindell, M. K., & Brandt, C. J. 1999. Assessing interrater agreement on the job relevance of a test: A comparison of the CVI, T, rwg(j), and r*wg(j) indexes. Journal of Applied Psychology, 84: 640-647. Locke, E. A., & Latham, G. P. 1990. A theory of goal setting & task performance. Englewood Cliffs, NJ: Prentice Hall. Lynch, P. D., Eisenberger, R. & Armeli, S. 1999. Perceived organizational support: Inferior-versus-superior performance by wary employees. Journal of Applied Psychology, 84: 467-483. MacDuffie, J. P. 1995. Human resource bundles and manufacturing performance: Organizational logic and flexible production systems in the world auto industry. Industrial and Labor Relations Review, 48: 197-221. MacKenzie, S. B., Podsakoff, P. M., & Paine, J. B. 1999. Do citizenship behaviors matter more for managers than for salespeople? Journal of the Academy of Marketing Science, 27: 396-410. McMahan, G. C., Virick, M., & Wright, P. M. 1999. Alternative theoretical perspectives for strategic human resource management revisited: Progress, problems, and prospects. In P. M. Wright & L. Dyer & J. Boudreau & G. Milkovich (Eds.), Research in personnel and human resources management, Supplement 4: 99-122. Greenwich, CT: JAI Press. Marcoulides, G. A., & Heck, R. H. 1993. Organizational culture and performance: Proposing and testing a model. Organizational Science, 4: 209-225.
134
Masterson, S. S. 2001. A trickle-down model of organizational justice: Relating employees' and customers' perceptions of and reactions to fairness. Journal of Applied Psychology, 86: 594-604. Masterson, S. S., Lewis, K., Goldman, B. M., & Taylor, M. S. 2000. Integrating justice and social exchange: The differing effects of fair procedures and treatment on work relationships. Academy of Management Journal, 43: 738-748. Mathieu, J. E., & Zajac, D. M. 1990. A review and meta-analysis of the antecedents, correlates, and consequences of organizational commitment. Psychological Bulletin, 108: 171-194. McMahan, G. C., Virick, M., & Wright, P. M. 1999. Alternative theoretical perspectives for strategic human resource management revisited: Progress, problems, and prospects. In P. M. Wright, L. Dyer, J. Boudreau, & G. Milkovich (Eds.), Research in Personnel and Human Resources Management: Supplement 4 (pp. 99-122). Greenwich, CT: JAI Press. Meyer, J. P., & Allen, N. J. 1991. A three-component conceptualization of organizational commitment. Human Resource Management Review, 1: 61-89. Meyer, J. P., Allen, N. J., & Smith, C. A. 1993. Commitment to organizations and occupations: Extension and test of a three-component conceptualization. Journal of Applied Psychology, 78: 538-551. Miles, R. E., & Snow, C. C. 1984. Designing strategic human resources systems. Organizational Dynamics, summer: 36-52. Mincer, J. 1974. Schooling, experience, and earnings. New York: Columbia University Press.
135
Mohammed, S., Mathieu, J. E., & Bartlett, A. L. 2002. Technical-administrative task performance, leadership task performance, and contextual performance: Considering the influence of team- and task-related composition variables. Journal of Organizational Behavior, 23: 795-814. Moorman, R. H. 1991. Relationship between organizational justice and organizational citizenship behaviors: Do fairness perceptions influence employee citizenship? Journal of Applied Psychology, 76: 845-855. Morrison, E. W., & Phelps, C. C. 1999. Taking charge at work: Extrarole efforts to initiate workplace change. Academy of Management Journal, 42: 403-419. Morrison, E. W., & Robinson, S. L. 1997. When employees feel betrayed: A model of how psychological contract violation develops. Academy of Management Review, 22: 226-256. Motowidlo, S. J., & Van Scotter, J. R. 1994. Evidence that task performance should be distinguished from contextual performance. Journal of Applied Psychology, 79: 475-480. Murphy, K. R., & Cleveland, J. N. 1995. Understanding performance appraisal: Social, organizational, and goal-based perspectives. Thousand Oaks, CA: Sage. Naumann, S. E., & Bennett, N. 2000. A case for procedural justice climate: Development and test of a multilevel model. Academy of Management Journal, 43: 881-889. Nicholson, N., & Johns, G. 1985. Absence culture and the psychological contract - Who is in control of absence? Academy of Management Review, 10: 397407.
136
O'Reilly, C., III, & Chatman, J. 1986. Organizational commitment and psychological attachment: The effects of compliance, identification, and internalization on prosocial behavior. Journal of Applied Psychology, 71: 492-499. Organ, D. W. 1997. Organizational citizenship behavior: It's construct clean-up time. Human Performance, 10: 85-97. Organ, D. W. 1988. Organizational citizenship behavior: The good soldier syndrome. Lexington, MA: Lexington Books/D. C. Heath and Company. Organ, D. W., & Konovsky, M. 1989. Cognitive versus affective determinants of organizational citizenship behavior. Journal of Applied Psychology, 74: 157-164. Organ, D. W., & Ryan, K. 1995. A meta-analytic review of attitudinal and dispositional predictors of organizational citizenship behaviors. Personnel Psychology, 48: 775-802. Osterman, P. 1987. Choice of employment systems in internal labor markets. Industrial Relations, 26: 46-67. Ostroff, C., & Bowen, D. E. 2000. Moving HR to a higher level: HR practices and organizational effectiveness. In K. J. Klein & S. W. Kozlowski (Eds.), Multilevel theory, research, and methods in organizations: Foundations, extensions, and new directions (p. 211-266). San Francisco: Jossey-Bass. Peck, S. R. 1994. Exploring the link between organizational strategy and the employment relationship: The role of human resource policies. Journal of Management Studies, 31: 715-736. Pennings, J. M., Lee, K., & Van Witteloostuijn, A. 1998. Human capital, social capital, and firm dissolution. Academy of Management Journal, 41: 425-440.
137
Pfeffer, J. 1998. The human equation: Building profits by putting people first. Boston, MA: Harvard Business School Press. Pfeffer, J. 1994. Competitive advantage through people: Unleashing the power of the work force. Boston: Harvard Business School Press. Pinfield, L. T., & Berner, M. F. 1994. Employment systems: Toward a coherent conceptualization of internal labor markets. In G. Ferris (Ed.), Research in Personnel and Human Resources Management, 12: 41-78. Greenwich, CT: JAI Press. Podsakoff, P. M., & MacKenzie, S. B. 1994. Organizational citizenship behaviors and sales unit effectiveness. Journal of Marketing Research, 3: 351-363. Powell, T. C. 1992. Organizational alignment as competitive advantage. Strategic Management Journal, 13: 119-134. Quinn, R. E. 1988. Beyond rational management: Mastering the paradoxes and competing demands of high performance. San Francisco: Jossey-Bass. Quinn, R. E., & Rohrbaugh, J. 1983. A spatial model of effectiveness criteria: Towards a competing values approach to organizational analysis. Management Science, 29: 363-377. Reichers, A. E. 1985. A review and reconceptualization of organizational commitment. Academy of Management Review, 10: 465-476. Rigdon, E. E., Schumacker, R. E., & Marcoulides, G. A. 1998. A comparative review of interaction and nonlinear modeling. In R. E. Schumacker & G. A. Marcoulides (Eds.), Interaction and nonlinear effects in structural equation modeling: 116. Mahwah, NJ: Lawrence Erlbaum Associates.
138
Roberson, Q. M., Moye, N. A., & Locke, E. A. 1999. Identifying a missing link between participation and satisfaction: The mediating role of procedural justice perceptions. Journal of Applied Psychology, 84: 585-593. Rogers, E. W., & Wright, P. M. 1998. Measuring organizational performance in strategic human resource management: Problems, prospects, and performance information markets. Human Resource Management Reviews, 8: 311-331. Roggs, K. L., Schmidt, D. B., Shull, C., & Schmitt, N. 2001. Human resource practices, organizational climate, and customer satisfaction. Journal of Management, 27: 431-449. Rousseau, D. M. 2000. Psychological contract report inventory technical report.http://www.andrew.cmu.edu/user/rousseau/0_reports/PCI.pdf. Rousseau, D. M. 1995. Psychological contracts in organizations: Understanding written and unwritten agreements. Thousand Oaks, CA: Sage Publications. Russell, J. S., Terborg, J. R., & Powers, M. L. 1985. Organizational performance and organizational level training and support. Personnel Psychology, 38: 849-863. Rupp, D. E., & Cropanzano, R. 2002. The mediating effects of social exchange relationships in predicting workplace outcomes from multifoci organizational justice. Organizational Behavior and Human Decision Processes, 89: 925-946. Ryan, A. M., Schmit, M. J., & Johnson, R. 1996. Attitudes and effectiveness: Examining relations at an organizational level. Personnel Psychology, 49: 853-882. Scandura, T. A., & Graen, G. B. 1984. Moderating effects of initial leadermember exchange status on the effects of a leadership intervention. Journal of Applied Psychology, 71: 579-584.
139
Schneider, B. 1975. Organizational climate: Individual preferences and organizational realities revisited. Journal of Applied Psychology, 60: 459-465 Schneider, B., & Bowen, D. E. 1985. Employee and customer perceptions of service in banks: Replication and extension. Journal of Applied Psychology, 70: 423433. Schneider, B., Goldstein, H. W., & Smith, D. B. 1995. The ASA framework: An update. Personnel Psychology, 48: 747-773. Schneider, B., & Reichers, A. 1983. On the etiology of climates. Personnel Psychology, 36: 19-40. Schneider, B., Salvaggio, A. N., & Subirats, M. 2002. Climate strength: A new direction for climate research. Journal of Applied Psychology, 87: 220-229. Schneider, B., White, S. S., & Paul, M. C. 1998. Linking service climate and customer perceptions of service quality: Test of a causal model. Journal of Applied Psychology, 83: 150-163. Schuler, R. S., & Jackson, S. E. 1987. Linking competitive strategies with human resource management practices. Academy of Management Executive, 1: 207219. Settoon, R. P., Bennett, N., & Liden, R.C. 1996. Social exchange in organizations: Perceived organizational support, leader-member exchange, and employee reciprocity. Journal of Applied Psychology, 81: 219-227. Sheppard, B. H., Lewicki, R. J., & Minton, J. W. 1992. Organizational justice: The search for fairness in the workplace. New York: Lexington Books/Macmillan.
140
Shore, L. M., Barksdale, K., & Shore, T. H. 1995. Managerial perceptions of employee commitment to the organization. Academy of Management Journal, 38: 15931615. Shore, L. M., & Wayne, S. J. 1993. Commitment and employee behavior: Comparison of affective commitment and continuance commitment with perceived organizational support. Journal of Applied Psychology, 78: 774-780. Snell, S. A., & Dean, J. W., Jr. 1994. Strategic compensation for integrated manufacturing: The moderating effects of jobs and organizational inertia. Academy of Management Journal, 37: 1109-1140. Snell, S. A., & Dean, J. W., Jr. 1992. Integrated manufacturing and human resource management: A human capital perspective. Academy of Management Journal, 35: 467-504. Sparrowe, R. T., & Liden, R. C. 1997. Process and structure in leader-memberexchange. Academy of Management Review, 22: 522-552. Sullivan, P. H. 2000. Value-driven intellectual capital: How to convert intangible corporate assets into market value. New York: John Wiley & Sons. Taylor, M. S., Masterson, S. S., Renard, M. K., & Tracy, K. B. 1998. Managers' reactions to procedurally just performance management systems. Academy of Management Journal, 41: 568-579. Taylor, M. S., Tracy, K. B., Renard, M. K., Harrison, J. K., & Carroll, S. 1995. Due process in performance appraisals: A quasi-experiment in procedural justice. Administrative Science Quarterly, 40: 495-523.
141
Terpstra, D. E., & Rozell, E. J. 1993. The relationship of staffing practices to organizational level measures of performance. Personnel Psychology, 46: 27-48. Tett, R. P., & Meyer, J. P. 1993. Job satisfaction, organizational commitment, turnover intention, and turnover: Path analyses based on meta-analytic findings. Personnel Psychology, 46: 259-293. Thomas, J. B., Clark, S. M., & Gioia, D. A 1993. Strategic sensemaking and organizational performance: Linkages among scanning, interpretation, action, and outcomes. Academy of Management Journal, 36: 239-270. Tinsley, H. E. A., & Weiss, D. J. 1975. Interrater reliability and agreement of subjective judgments. Journal of Counseling Psychology, 22: 358-376. Tsui, A. S., Pearce, J. L., Porter, L. W., & Hite, J. P. 1995. Choice of employeeorganization relationship: Influence of external and internal organizational factors. In G. R. Ferris (Ed.), Research in Personnel and Human Resources Management: Vol. 13 (p. 117-151). Greenwich, CT: JAI Press. Tsui, A. S., Pearce, J. L., Porter, L. W., & Tripoli, A. A. 1997. Alternative approaches to the employee-organization relationship: Does investment in employee payoff? Academy of Management Journal, 40: 1089-1121. Tyler, T. R., & Lind, E. A. 1992. A relational model of authority in groups. In M. Zanna (Ed.), Advances in Experimental Social Psychology, 25: 115-192. New York: Academic Press. Tziner, A., & Kopelman, R. 1988. Effects of rating format on goal-setting dimensions: A field experiment. Journal of Applied Psychology, 73: 323-326.
142
Van Dyne, L., & Ang, S. 1998. Organizational citizenship behavior of contingent workers in Singapore. Academy of Management Journal, 41: 692-703. Van Dyne, L., Graham, J. W., & Dienesch, R. M. 1994. Organizational citizenship behavior: Construct redefinition, measurement, and validation. Academy of Management Journal, 37: 765-802. Van Dyne, L., & LePine, J. A. 1998. Helping and voice extra-role behaviors: Evidence of construct and predictive validity. Academy of Management Journal, 41: 108-119. Vance, R. J., Coovert, M. D., MacCallum, R. C., & Hedge, J. W. 1989. Construct models of task performance. Journal of Applied Psychology, 74: 447-455. Van Scotter, J. R. 2000. Relationships of task performance and contextual performance with turnover, job satisfaction, and affective commitment. Human Resource Management Review, 10: 79-95. Van Scotter, J. R., & Motowidlo, S. J. 1996. Evidence for two factors of contextual performance: Job dedication and interpersonal facilitation. Journal of Applied Psychology, 81: 525-531. Van Vianen, A. E. M. 2000. Person-organization fit: The match between newcomers' and recruiters' preferences for organizational cultures. Personnel Psychology, 53: 113-149. Walton, E. J., & Dawson, S. 2001. Managers' perceptions of criteria of organizational effectiveness. Journal of Management Studies, 38: 173-199.
143
Walz, S. M., & Niehoff, B. P. 1996. Organizational citizenship behaviors and their effect on organizational effectiveness in limited-menu restaurants. Academy of Management Best Paper Proceedings, 56th Annual Meeting, Cincinnati, OH, 307-311. Wanous, J. P. 1980. Organizational entry: Recruitment, selection, and socialization of newcomers. Reading, MA: Addison-Wesley. Wanous, J. P., & Hudy, M. J. 2001. Single-item reliability: A replication and extension. Organizational Research Methods, 4: 361-375. Wanous, J. P., Reichers, A. E., & Hudy, M. J. 1997. Overall job satisfaction: How good are single-item measures? Journal of Applied Psychology, 82: 247-252. Way, S. A. 2002. High Performance Work System and intermediate indicators of firm performance within the US small business sector. Journal of Management, 28: 765-785. Wayne, S. J., Shore, L. M., & Liden, R. C. 1997. Perceived organizational support and leader-member exchange: A social exchange perspective. Academy of Management Journal, 40: 82-111. Weick, K. 1979. The social psychology of organizing. Reading, MA: AddisonWesley. Wernerfelt, B. 1984. A resource-based view of the firm. Strategic Management Journal, 5: 171-180. Whitener, E. M. 2001. Do "high commitment" human resource practices affect employee commitment? A cross-level analysis using hierarchical linear modeling. Journal of Management, 27: 515-535.
144
Williams, L. J., & Anderson, S. E. 1991. Job satisfaction and organizational commitment as predictors of organizational citizenship and in-role behaviors. Journal of Management, 17: 601-617. Williams, C. R., & Livingstone, L. P. 1994. Another look at the relationship between performance and voluntary turnover. Academy of Management Journal, 37: 269-298. Wright, P. M. 1998. Introduction: Strategic human resource management research in the 21st century. Human Resource Management Review, 8: 187-191. Wright, P. M., Gardner, T., Moynihan, L. M., & Park, H. J. 2000. Unlocking the black box: Examining the processes through which human resource management practices impact business performance. Paper presented at the 60th annual meeting of the Academy of Management, Toronto, Canada. Wright, P. M., & George, J. M. 1993. Productivity and extra-role behavior: The effects of goals and incentives on spontaneous helping. Journal of Applied Psychology, 78: 374-381. Wright, P. M., & McMahan, G. C. 1992. Theoretical perspectives for strategic human resource management. Journal of Management, 18: 295-320. Wright, P. M., & Sherman, W. S. 1999. Failing to find fit in strategic human resource management: Theoretical and empirical problems. In P. M. Wright, L. Dyer, J. Boudreau, & G. Milkovich (Eds.), Research in Personnel and Human Resources Management, Supplement 4: 53-74. Greenwich, CT: JAI Press.
145
Wright, P. M., Smart, D. L., & McMahan, G. C. 1995. Matches between human resources and strategy among NCAA basketball teams. Academy of Management Journal, 38: 1052-1074. Wright, P. M., & Snell, S. A. 1991. Toward an integrative view of strategic human resource management. Human Resource Management Review, 1: 203-225. Wright, P. M., & Snell, S. A. 1998. Toward a unifying framework for exploring fit and flexibility in strategic human resource management. Academy of Management Review, 23: 756-772. Youndt, M. A., Snell, S. A., Dean, J. W., Jr., & Lepak, D. P. 1996. Human resource management, manufacturing strategy, and firm performance. Academy of Management Journal, 39: 836-866.
146
doc_431442459.docx
A Study on Black Box In Strategic HRM Research From Resource-Based And Social Exchange Perspectives, Social exchange theory is a social psychological and sociological perspective that explains social change and stability as a process of negotiated exchanges between parties. Social exchange theory posits that all human relationships are formed by the use of a subjective cost-benefit analysis and the comparison of alternatives. The theory has roots in economics, psychology and sociology. Social exchange theory features many of the main assumptions found in rational choice theory and structuralism.
A STUDY ON BLACK BOX IN STRATEGIC HRM RESEARCH FROM RESOURCE-BASED AND SOCIAL EXCHANGE PERSPECTIVES
Abstract:In this dissertation, I examine the specific mediating mechanisms through which High Performance Work Systems influence overall unit performance. In particular, I draw mainly on two theoretical perspectives, the resource-based view of the firm and behavioral perspective, to propose and test the mediated model of strategic human resource management. The data were collected from 322 managers and 526 employees for a sample of 76 units of Japanese companies in various industries. The data were aggregated to the
unit-level of analysis. On one hand, the data provided support for many of the hypotheses advanced in the dissertation. The results of the hierarchical regression analyses revealed that High Performance Work Systems was positively related to both the level of human capital that the unit possesses and the collective normative contract that the employees working in the unit share. In addition, the level of human capital and collective normative contract were significantly related to most of the HR-related outcomes, which, in turn, were significantly related to overall unit performance. Moreover, as hypothesized, the level of human capital acted as mediators of the relationship between High Performance Work Systems and HR-related outcomes while HR-related outcomes played the role of mediator on the relationships between the level of human capital or collective normative contract and overall unit performance. On the other hand, the mediating hypothesis for collective normative contract as well as interaction hypothesis for the level of human capital* collective normative contract were not supported. The implications of the findings and future research directions are also discussed.
TABLE OF CONTENTS List of Tables List of Figures vi vii
Chapter 1: PROBLEM STATEMENT 1-7 Research Questions??????????????????????7 Chapter 2: THEORETICAL REVIEWS 8-24 Resource-Based View of the Firm????????????????..8 Behavioral Perspective????????????????????...12 Understanding employee attitudes and behaviors: Social exchange perspective???????????????...14 Integrating resource-based view and behavioral perspective???.18 Conceptualization of performance?????????????..20 Review of Past Research Related to Framework???????????22 Chapter Summary??????????????????????...24 Chapter 3: MODEL DEVELOPMENT AND HYPOTHESES 25-46 Scope of the Research?????????????????????.25 Level of analysis????????????????????.25 Unit performance indicators????????????????26 Employee groups????????????????????27 Research Model???.????????????????????.29 High Performance Work System, Level of Human Capital, and Performance????????????????????.30 High Performance Work System, Collective Normative Contract, and Performance????????????????????.34 High Performance Work System and collective normative contract?????????????...35 Collective normative contract and HR-related outcomes????...39 Moderating Effects of Collective Normative Contract????????...43 Chapter 4: RESEARCH METHODS 47-64 Sample???????????????????????????47 Data Collection Procedure..????.??????????????...49 Survey Translation Procedure????????????..?????..51 Aggregation Issues??????????????????????.52 Measures?????????????????????????? 55 High Performance Work Systems?????????????..56 The level of human capital????????????????..58 Collective normative contract???????????????.58 HR-Related Outcomes?????????????????????58 Perceived labor productivity???????????????..58 Unitlevel in-role or task performance??..????????.?59
iv
Unit-level overall organizational citizenship behaviors..????.?60 Unit Performance??????????????????..???.?...60 Analytic Procedures??????????????..????.???...62 Chapter 5: RESULTS 65-84
Chapter 6: DISCUSSION 85-100 Limitations??????????????????????????92 Implications for Research and Practice???????????????.96 Conclusion??????????????????????????.99 APPENDIX A APPENDIX B APPENDIX C REFERENCES 101-107 108-112 113-115 116-146
v
LIST OF TABLES
1. Changes in the Strategic HRM Perspective from a "Traditional" HRM Perspective 2. Summary of Dissertation Hypotheses 3. Response Patterns 4. Factor Loadings for 21-Item High Performance Work System Scale 5. Results of Exploratory Factor Analysis on the HR-Related Outcomes and Unit Performance 6. Descriptives of All the Variables 7. Regression Results for Industry Effect 8. Regression Results for Hypothesis 1 and 2 9. Regression Results for Hypothesis 4 10. Regression Results for Hypothesis 5 and 6 11. Regression Results for Hypothesis 7 12. Regression Result for Hypothesis 8 13. Regression Result for Hypothesis 9 14. Path Analytic Regression Analyses Results 15. Summary Results for the Hypotheses 62 67 68 70 73 74 75 77 79 80 83-84 3 45-46 47-48 57
vi
LIST OF FIGURES
1. The Black Box in Strategic HRM Research 2. Resource-Based View of the Firm in Strategic HRM 3. Behavioral Perspective in Strategic HRM 4. Social Exchange Theory for Strategic HRM 5. Integration of the Resource-Based View of the Firm and Behavioral Perspective on Strategic HRM 6. Overall Theoretical Framework 7. Proposed Model of High Performance Work System 8. Path Analytic Results
4 12 13 15
19 22 30 82
vii
Chapter 1 Problem Statements The introduction of strategic HRM has advanced our understanding of the relationships among strategy, human capital, human resource management (HRM), and firm performance by shifting the focus from "traditional" HRM in several ways. First, the level of analysis has shifted from a traditional micro focus on examining individual HRM functions (i.e., selection/recruitment, training and development, compensation, performance appraisal, job design, etc.) to a firm level of analysis by adopting a system approach to HRM practices. This involves examining the entire system of HRM practices as a whole, rather than examining individual subfunctional practices in isolation. As a result, the notion of synergy or complementarity of HRM practices became particularly important (Baird & Meshoulam, 1988; Becker & Huselid, 1998; Huselid, 1995; Wright & Snell, 1998). For example, MacDuffie (1995: 198) noted "an HR bundle or system must be integrated with complementary bundles of practices from core business functions" and examined the interaction between flexible production and human resource capabilities that included HRM practices such as recruitment and hiring, contingent compensation, status differentiation, and training. This notion of internal fit or alignment of different HRM practices is less salient in the traditional, subfunctional perspective to HRM. Second, strategic HRM focuses on higher-level contingency variables. Underlying this perspective is the argument that the effectiveness of HRM systems depends on how they are aligned with variables external to the HRM system such as strategy, technology, environmental conditions, and the like (Baird & Meshoulam,
1
1988; Wright & Snell, 1998). As an example, Youndt, Snell, Dean, and Lepak (1996) examined the relationship between manufacturing strategy, human resource management practices, and operational performance (employee productivity, machine efficiency, and customer alignment), and found a moderating effect of manufacturing strategy on the relationship between the use of a human-capital-enhancing HR system and operational performance. As exemplified by Youndt et al. (1996), strategy is considered as one of the most important contingent factors that HRM system needs to be aligned with (e.g., Arthur, 1994; McMahan, Virick, & Wright, 1999; Wright & McMahan, 1992). In traditional HRM approaches, the notion of fit tends to focus on individual employees and their fit with the internal surroundings (e.g., person-job fit or person-organization fit) rather than broader environmental factors. Third, strategic HRM research places emphasis on firm-level performance as the dependent variable rather than focusing solely on individual performance (cf. Rogers & Wright, 1998; Wright, 1998). For instance, Delery and Doty (1996) used return on asset (ROA) and return on equity (ROE) and Huselid (1995) used gross rate of return (GRATE) and a variant of Tobin's Q (i.e., firm market value/book value) as their dependent variables. In contrast, traditional HRM research generally focused on individual outcome behaviors such as task performance (cf., Locke & Latham, 1990), absenteeism (cf., Harrison & Martocchio, 1998), withdrawal behaviors (cf. Harrison & Martocchio, 1998; Hulin, 1991), and/or turnover (cf., Griffeth, Hom, & Gaertner, 2000). In general, the basic distinction between traditional HRM and strategic HRM can be summarized as the difference in the focus on micro- vs. macro-perspectives. Table 1 highlights these differences.
2
Table 1. Changes in the Strategic HRM Perspective from a "Traditional" HRM Perspective Level of Analysis "Traditional" HRM Subfunctions of HRM (Subfunctional-level: staffing, training & development, compensation & incentives, performance appraisal, etc.) Individual HRM practices (e.g., staffing practices [Terpstra & Rozell, 1993], training practices [Russell, Terborg, & Powers, 1985], and compensation practices [Gerhart & Milkovich, 1990]) Strategic HRM HRM as a whole (Organizational-level)
Focus
Contingencies Person-Environment Fit (e.g., Dawis, 2000; Edwards & Van Harrison, 1993), PersonOrganization Fit (e.g., Kristof, 1996; Van Vianen, 2000), Person-Job Fit (e.g. KristofBrown, 2000) Performance Individual employee performance
System or Configuration of HRM practices (e.g., "commitment" [Arthur, 1994], "human capital enhancing" [Youndt et al., 1996], "High Performance Work System" [Becker & Huselid, 1998; Huselid, 1995]) External Fit (Strategy-HRM), Internal Fit (HRM-HRM)
Organizational performance
Microperspective
Macroperspective
Despite this progress, one of the critical issues that remain unclear is an understanding of the mediating mechanism or process through which HRM practices influence firm performance. Interestingly, there is no empirically based research that has examined critical employee attitudes and behaviors in strategic HRM and linked them to other outcomes such as turnover rate, labor productivity, and so forth. Moreover, only a limited number of researchers have articulated potential mediators
3
only to a limited extent (Applebaum, Bailey, Berg, & Kalleberg, 2000; Delery, 1998). As Figure 1 illustrates, there is a "black box" in empirical examinations of HRperformance relationship where the mediating mechanism is typically implied but not measured to date. It is understandable that previous examinations of strategic HRM research have focused on investigating the overall linkages between HRM and performance, given the relative infancy of the field. However, advances in the strategic HRM area are impeded because linkages between HR practices and performance are tenuous at best without an understanding of the mediating processes though which HRM practices affect performance. In addition, the utility of strategic HRM research for practitioners is limited without explicating the processes that occur within this black box. FIGURE 1. The Black Box in Strategic HRM Research Firm strategy
A system or a bundle of HRM practices
Firm performance
Mediating mechanism or process With regard to this black box, one of the dominant, theoretical perspectives that have been used in the strategic HRM literature is the behavioral perspective, advanced by Schuler and Jackson (1987; Jackson et al., 1989). "A behavioral perspective assumes that employers use personnel practices as a means for eliciting and controlling
4
employee attitudes and behaviors" (Jackson et al., 1989: 728). The primary logic for this is that certain attitudes and behaviors are needed for certain strategies. An example of the behavioral perspective can be found in Miles and Snow's (1984) description of the employee behaviors needed for different strategic business types: defender, prospector, analyzer, and reactor. An important component of the behavioral perspective is an emphasis on the employee attitudes and behaviors that contribute to effective implementation of strategic objectives and organizational functioning. While conceptual arguments abound, however, none of these studies have actually examined employee attitudes and behaviors espoused to be critical to this process. Thus, previous empirical examinations of strategic HRM have not tested these theoretical assertions that employee attitudes and behaviors are critical in the effective implementation of different strategies. In addition, the behavioral perspective has not explicitly articulated the types of critical employee behaviors that lead to higher firm performance. In short, a key mediator of the strategic HRM research is neither well understood nor tested. In addition to the behavioral perspective, some strategic HRM researchers focused more on the competency of employees. Deriving from the resource-based view of the firm (e.g., Barney, 1991), these researchers suggest that firms' internal resources (human capital) directly influence firm performance (Wright & McMahan, 1992). Barney (1991) asserts that human capital may be one of the few firm resources that have the potential for sustainable competitive advantage because it may satisfy the four characteristics (i.e., value, rarity, inimitability, and nonsubstitutability) that resources must possess to provide firms with sustainable competitive advantage. While there is
5
empirical evidence that illustrates the positive influence of human capital on firm performance, this line of research typically de-emphasizes the role of HRM activity in obtaining human capital or eliciting the necessary behaviors. As noted by Austin, Villanova, Kane, and Bernardin (1991), performance is a function of ability (or in an aggregate form, human capital) and motivation (or effort) and both are needed for higher performance. As such, the sole use of resource-based view in strategic HRM might have been based on a faulty assumption that ability necessarily translates into performance without considering the other side of the equation - motivation as shown by the employees exhibiting desired attitudes and behaviors. Despite different conceptualization of the functional form of individual performance (as an additive model where ability and motivation influences performance independently or interactive model where ability and motivation influences performance in a multiplicative manner) (Austin et al., 1991), both ability and effort are instrumental in achieving higher performance. In other words, human capital may provide the potential for higher firm performance while employee attitudes and behaviors may be instrumental in realizing that potential. Taken together, both resourcebased view of the firm and behavioral perspective suggest that the attributes (human capital) and efforts (as exemplified by aggregate employee attitudes and behaviors) are critical components in understanding firm performance. However, previous research has not integrated these two perspectives to examine the relationships among HRM, human capital, employee attitudes and behaviors, and firm performance. Hence, the primary research questions in this dissertation focus on addressing these issues.
6
Research Question 1: What are the roles of human capital and employee attitudes and behaviors on the relationship between human resource management practices and unit performance. Research Question 2: What are the processes/mechanisms through which a system of human resources management impact unit performance?
7
Chapter 2 Theoretical Reviews In this chapter I will draw from resource-based view of the firm (Barney, 1991; Wernerfelt, 1984), the behavioral perspective (Schuler & Jackson, 1987), and social exchange theory (Blau, 1964; Gouldner, 1960; Homans, 1958, 1961) to develop a parsimonious research framework for this dissertation. Resource-Based View of the Firm (Barney, 1991; Wernerfelt, 1984) One of the major theoretical perspectives that strategic HRM researchers have adopted is the resource-based view of the firm. "The resource-based view of competitive advantage differs from the traditional strategy paradigm in that the emphasis of the resource-based view of competitive advantage is on the link between strategy and the internal resources of the firm" (Wright & McMahan, 1992: 300). The resource-based view of competitive advantage makes a couple of fundamental assumptions: 1) resource heterogeneity, i.e., not all firms have access to or possess the same resources, and 2) resource immobility, i.e. some resources are harder to transfer or purchase from the market. According to the resource-based view of the firm (Barney, 1991; Wernerfelt, 1984), "a firm is said to have a competitive advantage when it is implementing a value creating strategy not simultaneously being implemented by any current or potential competitors. A firm is said to have a sustained competitive advantage when it is implementing a value creating strategy not simultaneously being implemented by any current or potential competitors and when these other firms are unable to duplicate the benefits of this strategy" (Barney, 1991: 102).
8
Firm resources can be defined as "all assets, capabilities, organizational processes, firm attributes, information, knowledge, etc. controlled by a firm that enable firms to conceive of and implement strategies that improve its efficiency and effectiveness (Daft, 1983)" (Barney, 1991: 101). Moreover, firm resources can be broadly classified into four categories: physical capital, organizational capital, social, and human capital. Physical capital refers to assets that are tangible and include the physical technology used in a firm, a firm's plant and equipment, its geographical location, and its access to raw materials. Organizational capital is those intangible assets, including a firm's reporting structure, its formal and informal planning, controlling, and coordinating systems, as well as informal relations among groups within a firm and between a firm and those in its environment whereas social capital refers to the specific component of intangible asset that are based on ability and capability to build and maintain good relationships between the firm and other constituents such as stockholders, customers, other organizations, and employees (Sullivan, 2002). Human capital includes the training, experience, judgment, intelligence, relationships, and insights of individual managers and workers in a firm (Barney, 1991). While all four resources have the potential for competitive advantage, firm resources must possess four attributes, 1) value, 2) rarity, 3) inimitability, and 4) nonsubstitutability, to realize a sustained competitive advantage. A firm resource adds value when it exploits opportunities and/or neutralizes threats in a firm's environment (Lepak & Snell, 1999). It is rare when only a small number of current and potential competitors have it. A firm resource must be imperfectly imitable where other firms who do not possess these valuable and rare
9
resources cannot obtain them easily. Finally, a firm resource needs to be imperfectly substitutable where other firms cannot use strategically equivalent resources to conceive of and implement certain strategies. As noted by Sullivan (2002), intellectual capital that includes organizational, social, and human capital may provide sources of competitive advantage. In addition, Barney (1991) pointed out, in particular, that human capital as a set has the potential for creating and maintaining competitive advantage because it is likely to depend on unique historical conditions, social complexity, and causally ambiguity, which are three conditions that make resources relatively more inimitable. In the context of human capital, the level of human capital has an influence on firm performance (e.g., Becker, 1964; Hitt et al., 2001; Mincer, 1974). For instance, Pennings et al. (1998: 426) noted that "professionals endowed with a high level of human capital are more likely to deliver consistent and high-quality services" and the contribution of human capital investment to firm survival is critical. Similarly, Snell and Dean (1992) noted that human capital adds value to the firm because of enhanced potential for productivity provided by higher knowledge and skills. In other words, the higher the level of knowledge, skills, and abilities of employees, the more potential human capital has for impacting firm performance. The main emphasis of the resource-based view of the firm (Barney, 1991) is how the level of intellectual capital that firms possess or acquire can generate above average rent in terms of improved firm performances. Researchers adopting this perspective suggest that the level of intellectual capital, in general, and human capital, in particular, is directly influenced by HRM practices that are aimed toward
10
selecting/recruiting and training/developing employees (McMahan et al., 1999; Wright & McMahan, 1992; Wright & Snell, 1991). While the resource-based view of the firm has noted the importance of intellectual capital, the level of human capital is critical when applied to strategic HRM. In addition, the resource-based view of the firm emphasized the role of shared experiences and learning that facilitate the accumulation of individual employees' tacit knowledge and accretion of these individual knowledge into a collective such as units' or firms' organizational capital (e.g., Berman, Down, & Hill, 2002; Guillen, 2000; Lado & Wilson, 1994). As MacDuffie (1995: 199) noted, "Innovative human resource practices are likely to contribute to improved economic performance only when three conditions are met: when employees possess knowledge and skills that managers lack; when employees are motivated to apply this skill and knowledge through discretionary effort; and when the firm's business or production strategy can only be achieved when employees contribute such discretionary effort (Levine and Tyson 1990; Bailey 1992)." While this is not an issue with physical and, to a lesser extent, organizational capital, the actions or behaviors of employees are a critical issue for human (Coff, 1997) and, to a lesser extent, social capital because human capital by itself cannot influence firm performance unless employees are induced to perform or actually use their human capital to improve performance. Although social capital may initially emanate from the employees, through organizational learning, the firm can acquire social capital by utilizing tacit knowledge of the employees. Therefore, in terms of strategic HRM, the most critical component of intellectual capital that a firm can impact may be the level of human capital that resides
11
in the employees. Figure 2 depicts this emphasis of the resource-based view of the firm on human capital as a source of competitive advantage. Therefore, there are two ways, directly and indirectly, that human capital of the firm can influence firm performance. FIGURE 2. Resource-Based View of the Firm in Strategic HRM
Human Capital for the Firm Employee Behavior Firm Performance
The emphasis on human capital characteristics that arise from the resourcebased view of the firm is important, especially given that only a handful of research has specifically examined the linkages between human capital and firm performance (Hitt et al., 2001; Wright et al., 1995). Moreover, despite the growing use of the resource-based view of the firm as an underlying theoretical logic for strategic HRM, much of the empirical research has focused on the relationship between HRM practices and firm performance while ignoring human capital, thereby increasing the importance of examining this component. Delery (1998: 290) explicitly acknowledged that, "a firm does not gain a competitive advantage from HRM practices, per se, but from the human resources that the firm attracts and retains." Consequently, understanding the linkage between an HRM system and performance must explicitly consider the relationships among a bundle of HRM policies, level of human capital, and performance. Behavioral Perspective The behavioral perspective in strategic HRM (Jackson et al., 1989; Schuler & Jackson, 1987) posits that different role behaviors are required for different types of 12
strategies that firms pursue. "The theory focuses on employee behavior as the mediator between strategy and firm performance" (Wright & McMahan, 1992: 303) or "between HR practices and sustainable competitive advantage" (McMahan et al., 1999: 106). Examples of employee behaviors may be being flexible to perform in-role or extra-role tasks or being efficient in performing required tasks. While useful in conceptualizing the role of employee attitudes/behaviors, its sole emphasis on employee behaviors may be too simplistic as the behavioral perspective typically ignores the level of human capital. It is simplistic in that "it assumes that the [sole] purpose of various employment practices is to elicit and control employee attitudes and behaviors" (Wright & McMahan, 1992: 303) and does not take into account the effect of the level of human capital that firms acquire and/or develop. To confound issues more, HRM practices are often used as proxies for employee behaviors, which are not a good proxy measure for the concept of employee behaviors and assume that an implementation of a policy automatically results in the desired employee attitudes and/or behaviors. There are many factors that might influence behaviors that go beyond HRM practices. Without actually measuring these factors, one cannot be certain why HRM practices lead to certain employee behaviors. Figure 3 illustrates the behavioral perspective. FIGURE 3. Behavioral Perspective in Strategic HRM Human Capital for the firm Firm Performance Employee Behaviors
13
Understanding employee attitudes and behaviors: Social exchange theory. When examining employees' attitudes and behaviors and their effects on firm performance at the firm level of analysis, social exchange theory is a critical perspective that needs to be integrated into theorizing and empirical research that explicate the mediating mechanisms. Social exchange theory focuses on the motivational component of employee-organization relationships and provides insights regarding the implications of the fit between the expected inducements and contributions provided in an employeeemployer exchange (Tsui, Pearce, Porter, & Tripoli, 1997). Specifically, social exchange theorists (e.g., Gouldner, 1960) examine the exchanges that occur between employers and employees regarding perceptions of reciprocity at an individual level of analysis. The essence of the social exchange theory is the notion of norm of reciprocity that develops, which makes employees feel obligated to respond equitably to treatments from others (including one's employer). As Wayne, Shore, and Liden (1997: 83) noted, "employees seek a balance in their exchange relationships with organizations by having attitudes and behaviors commensurate with the degree of employer commitment to them as individuals." Though the exact focus of social exchange theorists varies, a common theme is that the perceived balance between organizational inducement and employee contributions has performance implications. For instance, Wayne et al. (1997) found that the use of HRM practices that were developmental in nature was positively related to perceived organizational support. Perceived organizational support, in turn, has been found to be positively associated with affective, organizational commitment (Eisenberger, Fasolo, & Davis-LaMastro, 1990), constructive suggestions (Eisenberger
14
et al., 1990), and citizenship behaviors (Wayne et al., 1997). In addition, perceived organizational support is negatively associated with absenteeism (Eisenberger et al., 1990) and turnover intentions (Guzzo, Noonan, & Elron, 1994). Several variables including perceived organizational support, leader-member exchange (e.g., Liden, Sparrowe, & Wayne, 1997; Wayne et al., 1997), organizational commitment (e.g., Shore, Berksdale, & Shore, 1995; Shore & Wayne, 1993), and organizational citizenship behaviors (e.g., Masterson, Lewis, Goldman, & Taylor, 2000), for example, have been used to capture the notion of social exchange. Hence, social exchange theory provides insights regarding the specific mediating factors that likely account for the relationship between HRM and performance. Figure 4 depicts social exchange perspective as an important mediating mechanism for the HRM practices-firm performance relationship. FIGURE 4. Social Exchange Theory for Strategic HRM
Human Resource Management Practices
Aggregate Employee Attitudes & Behaviors e.g.,
• Leader-Member-Exchange • Perceived Organizational Support • Organizational Commitment • Organizational Citizenship Behaviors
Firm Performance
Within an organizational setting, the employer and the employees can be considered parties to the social exchange relationships. First, assuming that the organization first acts in such a way to provide employees with something of social value such as providing job security or access to training programs, the employees,
15
then, perceive that the organization has provided something of social value to them. Only then, the employees will feel indebted to reciprocate to the organization with something of equal or greater value. Although Schuler and Jackson (1987; Jackson et al., 1989) and other scholars in strategic HRM have not articulated the specific variables embedded within the behavioral perspective, those that have been used in previous studies in organizational behavior research as representing the notion of social exchange include leader-memberexchange (e.g., Masterson et al., 2000; Settoon, Bennett, & Liden, 1996; Wayne et al., 1997), organizational justice (e.g., Konovsky & Pugh, 1994; Masterson et al., 2000; Moorman, 1991; Organ & Konovsky, 1989), perceived organizational support (e.g., Armeli, Eisenberger, Fasolo, & Lynch, 1998; Eisenberger et al., 1986; Eisenberger et al., 1990; Masterson et al., 2000; Settoon et al., 1996), and psychological contract (e.g., Rousseau, 1989, 1995). At the same time, organizational commitment (e.g., Eisenberger et al., 1986; Eisenberger et al., 1990; Shore & Wayne, 1993), organizational citizenship or extra-role behaviors (e.g., Konovsky & Pugh, 1994; Lambert, 2000; Moorman, 1991; Organ & Konovsky, 1989; Van Dyne & Ang, 1998), and task or in-role performance (e.g., Austin et al., 1991) have been used as exemplar of employees' affective reactions to organizational actions. Of these social exchange variables used in the field, psychological contract, which is defined as "individual beliefs, shaped by the organization, regarding terms of an exchange agreement between individuals and their organization" (Rousseau, 1995: 9), represent an aspect of social exchange that is based on obligation that arises from the feeling of indebtedness for the actions provided by the company and the notion of
16
reciprocity. Moreover, while psychological contract refers to the individual employee's beliefs regarding the obligatory aspects of social exchange relationship between the individual and the organization, Rousseau (1995) also discussed the concept of normative contract, which refers to individual employee's perception of the obligations that the organization has toward all of the employees working for the firm. According to Chan (1998), five different types of compositions models that describe the functional relations among constructs in the same content domain at a higher level of analysis exist: 1) additive, 2) direct consensus, 3) referent-shift consensus, 4) dispersion, and 5) process composition. Of particular importance for the present dissertation is the referent-shift consensus model, which creates a distinct concept when individual responses are aggregated, although the concept originally is derived from individual-level unit when examining unit-level attitudes. He provides self-efficacy, collective efficacy, and team efficacy as an example of individual-level unit (self-efficacy, defined as an individual's belief and confidence in mobilizing one's resources to achieve successful task performance [Bandura, 1977]) having a different meaning at the higher-level of analysis (collective efficacy is defined as the individual team member's belief and confidence that the team can mobilize its resources to achieve successful task performance, which is then used to create team efficacy (grouplevel construct), which refers to the team members' overall belief and confidence that the team as a collective can mobilize its resources to achieve successful task performance. A similar shift is expected for psychological contract to normative contract to collective normative contract.
17
Collective normative contract is defined here as representing employees' collective beliefs regarding terms of social exchanges between the employees and their organization regarding the obligations that the employer and employees have toward each other. When comparing different operationalization of social exchange relationships, collective normative contract appears to convey the basic components of any social exchange relationships that may be necessary for further reciprocation by the employees. Furthermore, other variables that have been used to represent individual's social exchange relationship such as perceived organizational support, leader-memberexchange, and organizational commitment have not been conceptualized at and do not possess parallel concept at higher level of analysis. Hence, collective normative contract appears more appropriate than other operationalizations of social exchange relationships. Integrating resource-based view and behavioral perspectives. As the above discussion suggests, there appears to be a disconnect in our understanding of the relationship among human capital, HRM, and performance or employee behavior, HRM, and performance. According to the resource-based view of the firm, human resources have the potential to provide firms with sustainable competitive advantage (Barney, 1991). While Hitt et al. (2001) examined the relationship between human capital (quality of school and total experience in a firm) and firm performance, they did not examine the influence of attitudes and behaviors and a system of HRM practices on these relationships. One of the reasons that this is the case may be due to the relative emphasis placed by resource-based view of the firm on human capital and the emphasis placed on employee attitudes and behaviors by the behavioral perspective without
18
considering these perspectives simultaneously, for instance. By integrating both perspectives and simultaneously looking at human capital and employee attitudes and behaviors such as collective normative contract of the employees as important components to firm performance, we can gain a better understanding of how HRM practices lead to firm performance. However, having highly knowledgeable, skilled, and capable workforce is a necessary but not a maximally effective condition toward improving firm performance. In order for human capital to actually contribute to firm performance, employee behaviors that are necessary need to be exhibited. In other words, having competent human capital provides firms with the potential for improving firm performance whereas employee behaviors are a necessary catalyst for converting this potential into reality. HRM practices are the means by which these human capital and employee behaviors are obtained. Therefore, it is necessary to examine the relationships among human capital, employee attitudes and behaviors, HRM, and performance from a holistic perspective, incorporating all four concepts in a study. Figure 5 shows the inter-relationships among these variables. To do so, however, we must also gain a better understanding how we conceptualize HRM policies. FIGURE 5. Integration of the Resource-Based View of the Firm and Behavioral Perspective on Strategic HRM Human Capital of the Firm • Knowledge, skills, & abilities Social Exchange Relationships • Normative contract
HRM Activities
Firm Performance
19
Conceptualization of performance. Another issue that is critical to fully understand the HRM-firm performance relationship pertains to the conceptualization of performance (Becker & Gerhart, 1996; Rogers & Wright, 1998). As noted by Wright (1998), there seems to be a consensus within the realm of strategic HRM that maximizing organizational performance, particularly financial performance, is the major goal to be achieved. This preference for financial measures of performance has been shown by Rogers and Wright (1998) who reviewed the literature and noted that out of 80 dependent variables included in the strategic HRM research, accounting measures such as return on asset, return on equity, profits, and sales and market measures such as stock price and Tobin's Q were used in more than half of the research. However, the "appropriate dependent variable will vary with the level of analysis" (Becker & Gerhart, 1996: 791). At the business unit-level of analysis, perhaps the productivity of research and development personnel or their turnover rate may be more important for firms pursuing a differentiation strategy whereas the productivity of a firm's production staff may be more critical for firms following cost leadership strategy. Further, "the focus should be on variables that have inherent meaning for a particular context" (Becker & Gerhart, 1996: 791). For example, efficiency-based financial measures such as return on assets or return on equity may be more appropriate for firms pursuing cost leadership strategies in most of their business units whereas sales growth (ratio) or revenue growth (ratio) may be more appropriate for firms pursuing product differentiation strategies for the majority of their business units. Thus, it may be more appropriate for strategic HRM research to include multiple indicators of firm performance and make differential predictions based on them.
20
Dyer and Reeves (1995) noted different types of performance measures that are most appropriate for strategic HRM research. They proposed that four effectiveness measures are (1) human resource outcomes such as absenteeism, turnover, job satisfaction, and individual or group performance, (2) organizational outcomes such as productivity, quality, and service, (3) financial or accounting outcomes such as profitability, return on assets, and return on invested capital, and (4) stock market performance (stock value or shareholder return). Again, pointing to the fact that determination of which performance measures to include may be affected by the specific context of the research. For instance, turnover or retention rate may be more important for information technology companies that are knowledge intensive than for manufacturing companies that produce standardized products. Or, turnover or retention rate may be more critical for firms that utilize team-based production. Given different conceptualizations of performance and an existence of multiple types of performance indicators, it is important to clearly differentiate these outcomes and investigate the impact of human capital, employee attitudes and behaviors, and HRM for multiple outcomes if we are to fully understand the HRM-firm performance relationship. Perhaps, the outcome measures can be ordered from proximal to distal with employees as an anchor. Hence, HR outcomes would be the most proximal, which leads to organizational outcomes. Organizational outcomes, in turn, may lead to financial or accounting outcomes and ultimately, market measures (cf. Becker & Huselid, 1998). The relationship between HRM activities and firm performance, as noted before, has been relatively well established. However, the intermediate linkages between HRM
21
activities and firm performance are not yet well understood. Hence, it is important to be able to explicate these intermediate linkages, which is the primary objective of this dissertation. The integration of resource-based view and behavioral perspective (and social exchange theory) makes it imperative that HR-related outcomes such as labor productivity and turnover rate, for example, be examine as the more proximal outcomes to the level of human capital and perceptions and reactions to social exchange relationship. The logical next step is to link these HR-related outcomes to organizational outcomes (such as sales growth and operating profits). Figure 6 illustrates the overall framework for this dissertation. FIGURE 6. Overall Theoretical Framework
Level of Human Capital High Performance Work System HRM Practices HR Outcomes Collective Normative Contract
Unit Performance
Review of Past Research Related to the Framework Although there is no empirical research in the area of strategic HRM that has examined these specific linkages from High Performance Work System to the level of human capital and collective normative contract, or the level of human capital and collective normative contract to HR-related outcomes, previous empirical research has examined the relationships between High Performance Work System and some of the
22
HR-related outcomes such as turnover and labor productivity, as well as HR-related outcomes to firm/unit performance. For example, Huselid (1995) found that turnover rate and labor productivity mediated the relationship between High Performance Work System and two indicators of firm performance, i.e., logarithm of Tobin's Q (market value of the firm / replacement costs of its asset) and gross rate of return on capital for the manufacturing companies. Way (2002) and Batt (2003) found significant relationships between High Performance Work System and labor productivity, and High Performance Work System and turnover rate for small companies and service companies, respectively. Similarly, Ichniowski, Shaw, and Prennushi (1997) found a positive relationship between innovative work practices and labor productivity for steelfinishing lines. However, none of the studies examined the level of human capital or collective normative contract as additional mediating mechanisms between High Performance Work System and (firm) performance. In addition, in the strategy literature, Carpenter, Sanders, and Gregersen (2001) found a relationship between CEO's international assignment experiences (as representing the level of human capital) and multinational companies' return on asset. Hitt et al. (2001) found a relationship between human capital and performance of law firms (net income to total firm revenue). However, these studies did not examine the relationship between High Performance Work System and the level of human capital or the level of human capital and HR-related outcomes. Thus, it is unclear if the level of human capital indeed acts as a mediator of the relationship between High Performance Work System and HR-related outcomes. Therefore, one of the contributions of the dissertation is explicating this mediating mechanism through the resource-based lens. In
23
addition, investigating the mediating influences of collective normative contract and HR-related outcomes are other contributions that I intend to make in this dissertation. Chapter Summary In summary, in this chapter I have identified several theoretical perspectives that are relevant and important in examining the research questions identified in the first chapter and provided an integrated framework that outlines two general mediating mechanisms between a system of HRM practices and firm performance relationships. Given the objective of the dissertation, focusing on both the resource-based view and the behavioral perspective provide a framework for understanding the potential firm performance implications of a firm's human capital as well as aggregate attitudes and behaviors of employees. In addition, understanding the role of a bundle of HRM practices in these relationships is critical as the system of HRM practices is the primary ways to influence the level of human capital that organizations possess and aggregate employee attitudes and behaviors that are exhibited for the organizations.
24
Chapter 3 Model Development and Hypotheses Scope of the Research For the current dissertation, the critical issue is to examine the mediating mechanisms through which a system of HRM activities leads to performance via level of human capital and employee attitudes and behaviors. Hence, the main emphasis of the dissertation is on explicating the relationships among HRM system, human capital, employee attitudes and behaviors, and outcomes. Given an existence of different employee groups within a firm (e.g., Lepak & Snell, 1999; Osterman, 1987), associated differences in expected attitudes and behaviors that contribute to firm effectiveness, and HRM practices used to elicit such attitudes and behaviors, the boundary conditions for the dissertation is placed around three issues including level of analysis, performance indicators, and type of employee groups. Level of analysis. Previous research on strategic HRM has been conducted at both the corporate-level (e.g., Arthur, 1992, 1994; Becker & Huselid, 1998; Delaney & Huselid, 1996; Guthrie, 2001; Huselid, 1995; Jackson et al., 1989; Terpstra & Rozell, 1993; Whitener, 2001) and business-unit level (Delery & Doty, 1996; Ichniowski et al., 1997; Koch & McGrath, 1996; McDuffie, 1995; Russell et al., 1985; Snell & Dean, 1992, 1994; Youndt et al., 1996). Although there are certain advantages as well as disadvantages to selecting one level vs. the other, the most appropriate level of analysis for the current dissertation is primarily at the business unit level because social exchange relationships between a corporation and employees is likely to be confounded with variety of factors external to the social exchange relationships that exist between
25
the employer and the employees. In other word, at the corporate level, the relationships among the variables of interest may differ from those of the business units. Also, corporations may have multiple business units with different objectives for one business unit vs. the other. Aggregating these units may mask the important relationships that exist among the variables included in the dissertation. Unit performance indicators. It is widely acknowledged that firm performance is a multi-dimensional rather than a uni-dimensional construct (Dyer & Reeves, 1995; Rogers & Wright, 1998; Wright & Sherman, 1999). To reflect this multidimensionality, I will examine multiple outcomes in this dissertation. Specifically, I examine outcomes that are more intermediate in terms of High Performance Work System' impact on employees' behavioral outcomes, which I refer to as HR-related outcomes, and include such aggregate behaviors as labor productivity, level of extra-role behaviors exhibited by the employees in the unit, and quality of task performance as well as more distal establishment performance outcomes such as business unit performance. As Dyer and Reeves (1995) and Becker and Huselid (1998) noted, the dependent variables (performance) can be ordered from more proximal such as labor productivity to more distal variables such as return on asset and market-to-book ratio when examined from the strategic HRM perspective. In addition, Gupta (1987) noted the difficulty of obtaining financial measures of business units across multiple industries, although more objective measures of unit performance such as market share and delinquency rates exist when examining bank branches (e.g., Ryan, Schmit, & Johnson, 1996) or sales growth ratio and return on assets for business units in the U.S. pulp and paper industry (Davis, Robinson, Pearce,
26
& Park, 1992). However, more perceptual measures such as the level of customer satisfaction (e.g., Gelade & Ivery, 2003; Ryan et al., 1996) and relative firm performance (e.g., Delaney & Huselid, 1996; Powell, 1992) that have been used to assess performance may be more appropriate. Dollinger and Golden (1992) and Powell (1992), for example, found measures of perceived organizational performance to correlate positively (with medium to high correlations) with objective measures of firm performance. Furthermore, the use of perceptual measures allows an analysis of forprofit and not-for-profit organizations to be conducted because objective performance data are typically not available for the non-profit organizations (Delaney & Huselid, 1996). Given the difficulty in obtaining objective, financial measures for the business units (Gupta, 1987) and the lack of comparability of these objective measures across multiple industries, more perceptual measure of unit performance is used. Employees groups. Finally, as several scholars have noted (Jackson et al., 1989; Lepak & Snell, 1999; Osterman, 1987; Rousseau, 1995; Tsui et al., 1995), there are some indications that different HRM systems are used for different groups of employees depending on their employment relationships. For example, Lepak and Snell (2002) found different types of HR configurations for different types of employee groups. One approach that others have used to deal with this is to focus their attention on only those workers deemed most critical to a firm's success, i.e., core employees (e.g., Arthur, 1992; Delery & Doty, 1996; MacDuffie, 1995), or to aggregate two different groups, core and support, to a firm level (Huselid, 1995). As noted by Lepak and Snell (1999), there are some limitations associated with this approach of combining core and support employees to focus on permanent
27
employees because different HR policies may be used for different employee groups, which may confound the findings. Here, permanent employee is defined as those nonmanagerial employees who are directly involved in the production of products or selling of products/services (core employees) and those employees who provide administrative and supporting services (support employees) on a long-term, full employment basis (e.g., Barnett & Miner, 1992; Davis-Blake & Uzzi, 1993; Lepak, Taylor, Tekleab, Marrone, & Cohen, 2002). However, there are several substantive and methodological tradeoffs that need to be taken into account when considering the use of different employee groups, as compared to the permanent employee group. First, one of the complication introduced by using different employee groups may be related to aggregation and generalizability issues. For instance, inconsistency in the types of jobs classified as one type of employment group vs. the other by the respondents in different companies might introduces aggregation as well as generalizability problems. If the respondents from each firm do not agree on the types of jobs classified as core vs. support, for example, there may not be a concept that can be used at the unit-level across units or between firms. In addition, depending on the specific sample of units included, the results may not be generalizable to other settings beyond this dissertation. Second, not all units have both employment groups and given that the main purpose of the dissertation is to examine the mediating mechanisms through which High Performance Work System influence unit performance, multiple employee responses are necessarily for each unit and collecting enough employee responses for each of the employment group for each unit and having enough sample size at the end is beyond the
28
capabilities of any scholars, unless multiple units from a single company or industry (e.g., Gelade & Ivery, 2003; Ryan et al., 1996), or large-scale survey data are made available to scholars (e.g., Way, 2002). Given the objective of the dissertation and feasibility considerations, therefore, I focus on permanent (i.e., core and support) employees as the primary target of this dissertation. Research Model Thus far, I have described the boundary conditions of this dissertation. The following sections will discuss specific hypotheses that are incorporated in the overall framework described in Chapter 2. The following sections are broken down into three sections. The first section describes the influence of High Performance Work System on the level of human capital present in the unit as well as its indirect effect on HR-related outcomes and unit performance. The second section describes the impact of the High Performance Work System on collective normative contract as well as its indirect effect on HR-related outcomes and unit performance. Finally, the third section discusses the interaction between level of human capital and collective normative contract on HRrelated outcomes. With the first two sections, I hypothesize that these variables act as two broad mediating mechanisms where a system of HRM practices affects unit performance. The first mediating mechanism is though human capital where High Performance Work System influences HR-related outcomes, which, in turn, acts as a mediator to influence performance. I also anticipate a mediating process through social exchange where High Performance Work System affects employees' collective perceptions of exchange relationships with the unit (collective normative contracts) that has impact on HR-related outcomes, which, in turn, lead to unit-valued outcomes.
29
Figure 7 illustrates the proposed research model for this dissertation. I describe these relationships in more detail below. Figure 7. Proposed Model of High Performance Work System Level of Human Capital High Performance Work System
• labor
HR Outcomes Unit Performance
productivity • task performance • OCBs Social Exchange • Collective Normative Contract
High Performance Work System, Level of Human Capital, and Performance High Performance Work System (e.g. Huselid, 1995) refers to a collection of HRM practices that are aimed at obtaining and insuring high level of competency or human capital for a firm's workforce as one of the primary objectives, on the one hand, and coordinating and eliciting desired employee behavior over time (Wright & Snell, 1991) that adds value to the organization, on the other. Huselid (1995) also examined High Performance Work System and implied two different influences of these HR practices included in High Performance Work System when he factor analyzed these HR practices into two and labeled those: 1) HRM practices involving selection and training as "employee skills and organization structures," emphasizing human capital; and 2) HRM practices involving compensation and performance appraisal as "motivation" factors, focusing on employee behavior.
30
The relationship between High Performance Work System and the level of human capital is rather straightforward (cf. Wright & Snell, 1991). For example, researches in the selection and staffing literature have long recognized that selection/staffing practices have an effect on the characteristics of the employees/managers selected for the job position (e.g., Guthrie & Olian, 1991). Similarly, Delaney and Huselid (1996) noted that organizations could adopt different HRM practices that place emphasis on improving the quality of the individual hired, or on raising the skills and abilities of current employee, or both. The more rigorous and comprehensive recruitment and selection procedures are, the higher the level of knowledge, skills, and abilities for the recruited employees likely be than when these HRM policies are not utilized. In addition, comprehensive, on-going training and development programs can augment and improve employee knowledge, skills, and abilities. Therefore, there is likely to be a positive relationship between High Performance Work System and the level of human capital (cf. Hitt & Barr, 1989). In addition, High Performance Work System usually entails compensation policies that lead the market as well as performance appraisal policies that are developmental, which may attract potential applicants with higher talents. Hence, I expect the following: Hypothesis 1: High Performance Work System is positively related to the level of human capital for the unit. Higher levels of human capital have the potential to provide a sustainable competitive advantage to firms because employees differ in their amount of knowledge, skills, and abilities that they possess. For example, Hitt et al. (2001) found a positive, curvilinear relationship between human capital and firm performance for a professional
31
service firms. More specifically, they found articulable knowledge, which is a type of knowledge that can be codified and transferable (Lane & Lubatkin, 1998), and tacit knowledge, which is another type of knowledge that is difficult to codify and transfer (Lane & Lubatkin, 1998), to have an inverted U-shaped relationship with return on sales. Similarly, Pennings, Lee, and Van Witteloostuijn (1998) found human capital characteristics of accountants to be negatively related to firm dissolution (i.e. positively related to firm survival). These empirical studies have shown the relationship between human capital and firm performance to be positive. Moreover, Bettencourt, Gwinner, and Meuter (2001) found employee's knowledge to be a significant predictor of serviceoriented organizational citizenship behaviors. In particular, they argue that (procedural) knowledge enhances the repertoire that employees have developed in terms of useful ways of interacting with specific customer types. An analogy may be drawn for interacting and dealing with other coworkers. Therefore, I expect the following: Hypothesis 2: The level of human capital for the unit is related to HR-related outcomes such as labor productivity, unit-level task performance, and unit-level organizational citizenship behaviors. A growing number of studies (Arthur, 1994; Becker & Huselid, 1998; Guthrie, 2001; Huselid, 1995; Ichniowski et al., 1997) have also shown that a system of HRM practices to be directly and positively related to HR-related outcomes such as labor productivity and turnover rates as well as establishment level performance (Guthrie, 2001; Huselid, 1995; Way, 2002). However, a system of HRM practices is not likely to influence these HR-related outcomes directly per se (cf. Becker & Gerhart, 1996). It is through increases in labor productivity stemming from an increase in the level of human
32
capital that enables firms to reap the benefits of competence-enhancing HRM practices embedded within High Performance Work System (cf. Snell & Dean, 1992). It is only through its influence on employees' level of knowledge, skills, and abilities that the potential for increased performance gains can be realized. Thus, the level of human capital is likely to mediate the relationship between High Performance Work System and HR-related outcomes. Hypothesis 3: The level of human capital mediates the relationship between High Performance Work System and HR-related outcomes such as labor productivity, unit-level task performance, and unit-level organizational citizenship behaviors. In turn, HR-related outcomes are likely to mediate the relationship between the level of human capital and unit level performance (cf. McMahan et al., 1999). For example, Hitt et al. (2001) found a positive relationship between the level of human capital and indicators of law firm performance. However, even when the firm initially obtains employees who possess high knowledge, skills, and abilities (i.e., human capital), if many of them exit the company shortly afterward, it is not likely to influence firm performance in a positive manner. In addition, if the employees are not motivated to use their skills, the level of task performance may only be minimally adequate, which does not justify the above-average market rate paid for these highly skilled employees. Thus, again, the performance for the unit may not improve simply by recruiting highly knowledgeable, skilled, and capable individuals. In other words, HR-related outcome such as an increase in labor productivity, aggregate task performance level, and
33
aggregate organizational citizenship behaviors are expected to mediate the relationship between the level of human capital and unit performance. Hypothesis 4: HR-related outcomes such as labor productivity, unit-level task performance, and unit-level organizational citizenship behaviors mediate the relationship between the level of human capital for the establishment and unit performance. High Performance Work System, Collective Normative Contract, and Performance While the behavioral perspective provides the general foundation to expect that employee attitudes and behaviors mediate the relationship between HRM policies and unit performance, incorporating a social exchange perspective (e.g., Blau, 1964; Gouldner, 1960) provides a compelling theoretical logic to further explicate the mediating mechanisms outlined by the behavioral perspective. Social exchange theory posits that two individuals or parties often decide to enter exchange relationships when both individuals perceive that it is beneficial to do so (Blau, 1964; Gouldner, 1960; Homans, 1958, 1961). Initially, parties to an exchange are cautious and conscious about the currency of exchange (be it monetary incentives or approval) that is being exchanged between each party. This monitoring may decrease over time when each party finds the relationship satisfying and trust develops between the parties, thereby increasing the value of contributions both in terms of acceptable alternatives and time between reciprocation. Typically, both parties strive for a fair exchange with equivalent contributions judged through time (Tsui et al., 1997). When one party to the relationship goes beyond required obligations (be it initial payment to or repayment of valued contribution by the other party), the other party subsequently
34
feels indebted to contribute back to tilt the balance in the other direction. This component of the exchange is known as the norm of reciprocity (Homans, 1961). In an organizational setting, one of the possible pairs of partners to social exchanges is the employees and the employing organization. Within this potential partnership, employees as well as an employer can initiate the social exchange cycle. However, it is typically assumed that the organization initiates the action by providing employees with something that the employees' value - e.g., providing job security, access to training programs - that are represented by the HR policies that the organization utilizes. If the employees perceive that the organization has provided something of value to them, the employees are likely to feel obligated to reciprocate by providing the organization with something of equal or greater value. Within this social exchange process, it may be important to distinguish two components to understand the processes in detail: 1) the way employees perceive and interpret the organization's actions, and 2) the employees' reactions to them. The first component relates to the employees' collective perceptions and attributions of the organizational actions. When employees share perceptions of HRM policies as valuable and attribute the intention of the organization in a favorable manner, they collectively should possess the notion that the employees in the organization need to repay the organization, i.e., collective normative contract, and they are likely to feel obligated to reciprocate with behaviors that the organization values. Hence, the second component of the social exchange is the reciprocation provided by the employees. High Performance Work System and Collective Normative Contract. When examining the variables that are based on the notion of social exchange as perceived by
35
the employee, collective normative contract seems quite relevant. Although different variables have been examined in the literature previously, scholars, in general, have suggested that the nature and strength of obligations incurred through a social exchange depends on the quality of the relationship between the partners to the exchange (Eisenberger et al., 1986; Lambert, 2000; Van Dyne, Graham, & Dienisch, 1994). Konovsky and Pugh (1994) identified a set of mediating variables that provides the foundation for the social exchange by setting the tone for the exchange relationship as "macromotives," which may include perceived organizational support and psychological contract (cf., Lambert, 2000), among others. However, given the notion of organizational climate, which is defined as a set of shared perceptions of the policies, practices, and procedures that is developed through interactions (James et al., 1988; Schneider & Reichers, 1983), it is likely that shared perceptions with regard to the terms of the contract exist within organizational units. In fact, Marcoulides and Heck's (1993) study examined the collective perception of the quality of leader-member interactions as an aspect of organizational climate and collective commitment of employees as shared attitudes; Naumann and Bennette's (2000) and Colquitt, Noe, and Jackson's (2002) study indicated the existence of climate for procedural justice; and Rogg, Schmidt, Shull, and Schmitt's (2001) study on strategic HRM examined the relationship between HRM philosophy, organizational climate (commitment as an aspect of organizational climate), and performance (customer satisfaction). Of particular relevance to the current study is the study conducted by Marcoulides and Heck (1993) that found a positive relationship between organizational climate and shared employee attitudes.
36
It has been noted that organizational climate develops though three different, yet interrelated, means (Schneider & Reichers, 1983): 1) from social interactions of employees and through shared meanings (symbolic interactionist approach); 2) from attraction/selection/attrition (ASA) resulting in homogeneity (ASA approach); and 3) from mere exposure to the same policies, practices, and procedures (the structuralist approach). The first approach to organizational climate emphasizes interaction process among the employees in which employees try to make sense of the external events and assign meanings to these. The second approach (attraction-selection-attrition) focuses on the notion of person-organization fit with the organizational goals (Schneider, Goldstein, & Smith, 1995). Organizational goals, and the processes, structures, and culture that emerge to facilitate achievement of the organizational goals originally articulated by the founder are thought to determine the kinds of people who are attracted to, are selected by, and retained by the organization (Schneider et al., 1995). Finally, the third approach to climate draws attention to the structural characteristics of the organization. By being exposed to certain types of situational conditions such as leadership (e.g., Bass, 1990), job design (e.g., Hackman & Oldham, 1980), groups or teams (e.g., Guzzo & Shea, 1992), pay systems (e.g., Lawler, 1981), dynamics of the external environment (e.g., Joyce & Slocum, 1990), employees are likely to develop similar perceptions. As noted above, these approaches, however, are not mutually exclusive and they seem to differ on the emphasis placed. Perhaps, the most relevant approaches when examining the formation of HRM policies and their effects on employee perceptions
37
may be the second and the third approach to organizational climate. High Performance Work System may affect the organization in two ways: by attracting particular types of applicants and enabling the organization to select highly talented recruits who are likely to stay with the company and the other by providing common understanding to employees' experiences in terms of shared meaning by socializing them into interpreting experiences in particular ways. For instance, High Performance Work System is typically used to attract and select employees with high potential, which is likely to impact employee perceptions. For instance, providing rigorous training may lead to an increase in perceived obligations that the employees feel toward the organization in that the employees may interpret receiving this training as an indication of the goodwill from the organization. Similarly, extensive recruitment and selection procedures may signal to employees that their organization places great value on them and consider them as crucial in the survival and success of the organization. While there are only few empirical studies that had examined the effects of HRM policies on social exchange variables, Wayne et al. (1997) found developmental experiences and number of promotions to be positively related to perceived organizational support. Developmental experiences in their study referred to those job assignments that are challenging and developmental. Similarly, Whitener's (2001) study indicated that performance appraisal and internal rewards that are part of "high commitment" HR practices were correlated with perceived organizational support. Moreover, Taylor et al. (1995) examined due process in performance appraisal where due process performance appraisal practices include participative decision-making, self-
38
evaluation, and feedback. They found that these appraisal practices were positively related to procedural fairness perception of the employees. In addition, they found these practices to be positively related to employees' evaluation of the supervisor (or the leader). While these studies focused on the individual-level of analysis, using different variables to represent social exchanges, I can infer from the results that High Performance Work System will be associated with employees' shared perceptions of the terms of contracts that they are expected to receive and return. As Kozlowski and Klein (2000:30) noted these aggregate employee attitudes and behaviors can be considered a shared construct, which "describes the characteristics that are common to - that is, shared by - the members of a unit." They also note that organizational climate, collective efficacy, and group norms are examples of shared unit-level properties and these shared unit properties emerge as consensual, collective aspects of the unit as a whole. Therefore, I expect the following: Hypothesis 5: High Performance Work System has a positive relationship with collective normative contract. More specifically, High Performance Work System is positively related to employees' shared perceptions regarding obligations (collective normative contract) toward the organization. Collective normative contract and HR-related outcomes. Unlike the relationship between HRM policies and aggregate employee attitudes and behaviors, there are more studies that examined the impact of employees' reactions to organization's social exchange relationship, especially with regard to behaviors that contribute indirectly to the organization by maintaining its social and psychological environment. Several terms
39
have been used to describe such behaviors in the past, including organizational citizenship behavior (OCB: e.g., Organ, 1988), prosocial organizational behavior (Brief & Motowidlo, 1986; George, 1990, 1991; George & Bettenhausen, 1990; O'Reilly & Chatman, 1986), organizational spontaneity (George & Brief, 1992; George & Jones, 1997), extra-role behavior (Van Dyne & LePine, 1998; Wright & George, 1993), and contextual performance (Borman & Motowidlo, 1993, 1997; Borman, White, & Dorsey, 1995; Motowidlo & Van Scotter, 1994; Van Scotter & Motowidlo, 1996). More recently, empirical research has accumulated that suggests that extra-role behaviors lead to greater overall organizational effectiveness (MacKenzie, Podsakoff, & Paine, 1999). Podsakoff and MacKenzie (1994) examined the impact of three forms of OCBs (helping, sportsmanship, and civic virtue) on the performance of life insurance agencies. They found that all three forms of OCBs had significant effects on unit-level performance and together accounted for approximately 17% of the variance in this criterion variable. Walz and Niehoff (1996) found similar results by examining the relationships between OCB dimensions and a multitude of performance measures in limited menu restaurants. In particular, they found that the overall OCB accounted for an average of about 29% of the variance in the six objective criterion variables. Similarly, George and Bettenhausen (1990) examined the relationship between employees' prosocial behaviors for the retail store and sales performance and found a positive correlation of .33. Particularly relevant is the study conducted by George and Bettenhausen (1990) who examined the relationship between group-level prosocial behavior and turnover rate. Their results indicated that group-level prosocial or organizational citizenship
40
behavior was negatively correlated with turnover rate for retail stores. In addition, they also found a positive relationship between group-level prosocial behavior and labor productivity, which they called sales performance (total store sales / number of sales associates working in the store). Extrapolating from this finding, I expect that employees' shared perception of normative contract, which is the result of social exchange relationship that the employees have with the organization, to be positively related to labor productivity, task performance, and citizenship behaviors exhibited in the unit. Therefore, I hypothesize the following: Hypothesis 6: Collective normative contract is positively related to HR-related outcomes such as labor productivity, unit-level task performance, and unit level organizational citizenship behaviors. The relationships above describe a positive relationship between High Performance Work System and collective normative contract, and collective normative contract and HR-related outcomes. In addition to these direct relationships, I expect mediating mechanisms that occur through these relationships. First, I expect that aggregate employee perceptions of normative contract to mediate the relationship between High Performance Work System and HR-related outcomes. There is an emerging body of evidence that suggests the linkages from HRM practices to aggregate employee attitudes and behaviors to HR-related outcomes and, eventually, to unit performance. Lambert (2000), for example, found that HRM practices related to worklife benefits were positively related to perceived organizational support, which was also positively related to helping behaviors (one aspect of OCB). Similarly, Marcoulides and
41
Heck (1993) found organizational climate had both a direct and an indirect (through shared employee attitudes) effect on organizational performance. In addition, empirical research suggests that aggregate employee behaviors, extra-role behaviors in particular, lead to greater overall organizational effectiveness (MacKenzie, Podsakoff, & Paine, 1999). Podsakoff and MacKenzie (1994) examined the impact of three forms of OCBs (helping, sportsmanship, and civic virtue) on the performance of life insurance agencies. They found that all three forms of OCBs aggregated to the agency level had significant effects on unit-level performance and together accounted for approximately 17% of the variance in this criterion variable. Similarly, Walz and Niehoff (1996) found that aggregate helping behavior was positively related to overall operating efficiency, customer satisfaction, revenue to fulltime equivalent, and quality of performance and negatively related to food cost percentage. They also found that overall OCBs accounted for an average of about 29% of the variance in the six objective criterion variables. Similarly, George and Bettenhausen (1990) examined the relationship between aggregate employees' prosocial behaviors for retail stores and sales performance, and found a positive correlation of .33. These results suggest that aggregate employee attitudes and behaviors mediate the relationship between HR practices and outcomes. Combining findings from these empirical studies, I expect the following: Hypothesis 7: Employees' shared perceptions of normative contract mediate the relationship between High Performance Work System and HR-related outcomes. Finally, Koys (2001) examined the relationship between HR outcome (turnover rate) and business unit effectiveness in a panel design, longitudinal study and found that
42
HR outcome precedes business unit effectiveness. Combining this finding with the other studies that found relationship between aggregate employees' reactions and unit effectiveness (George & Bettenhausen, 2000; Podsakoff & MacKenzie, 1994; Walz & Niehoff, 1996), it is expected that HR-related outcomes such as labor productivity and turnover rate act as a mediator between aggregate employees' reactions and unit performance. Hypothesis 8: HR-related outcomes such as labor productivity, unit-level task performance, and unit-level organizational citizenship behaviors mediate the relationship between collective normative contract and unit performance. Moderating Effect of Collective Normative Contract Finally, I expect employees' perceptions of their social exchange relationship with the organization to interact with the level of human capital to have an influence on HR-related outcomes. Given that these shared perceptions of social exchange can be considered "macromotives" (cf. Lambert, 2000), those units with higher amount of collective normative contract are more likely to have employees who are more motivated and willing to exert more effort on behalf of the unit. Based on the notion of reciprocity norm, when employees perceive the organization as a valuable exchange partner and trust that the organization will reciprocate, employees are more likely to exert extra effort that may be substantially more than their fair share that brings them up to par with what the organization has already given them. In a sense, the employees proactively engage in social exchange to further develop the relationship with the organization.
43
Given that employees with higher qualifications (i.e., who possess higher amount of knowledge, skills, and abilities) can generally perform at a higher level than those who do not (e.g., LePine & Van Dyne, 2001; Mohammed, Mathieu, & Bartlett, 2002; Vance, Coovert, MacCallum, & Hedge, 1989), if these employees with higher knowledge, skills, and abilities are motivated to exert extra effort that is over and above their expected contribution, the increases in labor productivity that originate from the increase in the aggregate level of task performance may be substantially larger than those with lower qualifications. In addition, when these highly qualified employees are motivated to help each other and/or voice innovative ways of organizing and executing the tasks (e.g., LePine & Van Dyne, 2001), the cost saving associated with reduced training needs or new process may be larger. Furthermore, George and Bettenhausen's (1990) study found a significantly negative correlation between group's norm on organizational citizenship behavior and turnover rate for the store. Therefore, it can be expected that when employees share a particular norm with regard to the obligations they have toward the organization, they are more likely to exhibit more organizational citizenship behaviors. When employees with high qualifications exhibit the organizational citizenship behavior, the benefits that other employees receive are likely to higher than when employees with low qualifications provide those behaviors. Hence, a higher level of human capital is likely to lead to positive synergy when it is coupled with a higher level of collective normative contract, which can be considered to represent a component of employee motivation. The interactive effect of ability and motivation on individual performance is relatively well established in the performance
44
literature (e.g., Austin et al., 1991; Kanfer & Ackerman, 1989). I expect a similar relationship at the unit-level of analysis. Thus, I expect the following: Hypothesis 9: Employees' collective normative contract with the organization moderates the relationship between the level of human capital and HR-related outcomes such that units with employees who have more favorable perceptions of the organization have higher levels of HR-related outcomes than those units with employees who do not perceive higher collective normative contract. In summary, the present dissertation proposes two sets of hypotheses that are associated with the level of human capital, i.e., Hypotheses 1 through 3, and collective normative contract, i.e., Hypotheses 5 through 7, along with mediating role of HRrelated outcomes on the relationships between the level of human capital and unit performance (Hypothesis 4), and collective normative contract and unit performance (Hypothesis 8). Finally, Hypothesis 9 posits the interactive effect of the level of human capital (as representing higher level construct of ability) and collective normative contract (as representing higher level construct of a component of motivation) on HRrelated outcomes. Table 2 summarizes all of the hypotheses advanced for the present dissertation. Table 2. Summary of Dissertation Hypotheses Hypothesis 1. High Performance Work System is positively related to the level of human capital. Hypothesis 2. The level of human capital is significantly related to HR-related outcomes. Hypothesis 3. The level of human capital mediates the relationships between High
45
Performance Work System and HR-related outcomes. Hypothesis 4. HR-related outcomes mediate the relationship between the level of human capital and unit performance. Hypothesis 5. High Performance Work System is positively related to collective normative contract. Hypothesis 6. Collective normative contract is significantly related to HR-related outcomes. Hypothesis 7. Collective normative contract mediates the relationships between High Performance Work System and HR-related outcomes. Hypothesis 8. HR-related outcomes mediate the relationships between collective normative contract and unit performance. Hypothesis 9. The level of human capital and collective normative contract interact to influence HR-related outcomes.
46
Chapter 4 Research Method Sample The sample for this dissertation includes 76 business units from 56 different companies located in Japan, ranging widely in terms of industries and geographical regions of Japan that the companies serve. Initially, 120 company contacts received a letter from a Japanese faculty member who provided assistance during the data collection procedure, inquiring about potential companies that may be willing to participate in the dissertation. Seventy-six companies were identified as having high likelihood of participating, given company contacts' relationships with these companies. Out of 76 companies, 56 companies provided at least one managerial and two employee responses that are usable for a company response rate of 73.68%. The names of the companies that were contacted for the dissertation are listed in Appendix B. In general, all but one (agriculture, forestry, and fishing) primary (SIC single-code) industry were represented. There were 23 units where only one managerial response was obtained. Table 3 summarizes the response distribution obtained for these seventysix units. Table 3. Response Patterns Number of Units 14 7 2 1 1 1 4 1 Managerial Responses 1 1 1 1 1 1 2 2 Employee Responses 234 578 23
47
3 1
2 2 2 3 3 3 3 3 3 4 4 4 4 5 5 5 6 6 7 7 7 7 8 8 10 12 14 15 19 26
4 61 161 32 42 74 81 111 121 41 61 81 91 31 61 181 61 111 61 71 161 261 91 201 41 61 101 181 31 48
The managerial sample consisted of 324 senior-level managers working in the establishments. The average number of managerial responses received for each establishment was 4.26 (ranged from 1 to 26). The managers, on average, had 19.02 years of work experience at the establishment (SD = 11.63), 10.16 years of job tenure (SD = 10.44), were 47.73 years old on average (SD = 7.91), were predominantly male (96%) and were at the level of middle management.
48
For the employee sample, the number of respondents was 525 with an average of 6.89 responses per establishment (ranged between 2 and 48). The employees had 8.05 years of average work experience at the establishment (SD = 8.12), 6.14 years of average job tenure (SD = 6.46), were 35.90 years old on average (SD = 8.43), were predominantly male (81%) and employed full-time (90%). Data Collection Procedure First, I obtained consent from the top management team members of the participating companies through the contacts that the Japanese faculty member had. Then, the human resources managers for the participating firms were contacted mostly through direct visit to the companies, followed up by either a phone call or e-mail to solicit their assistance in gathering the information necessary to send out the surveys. Data collection procedure involved surveying managers and employees at the establishment level. Given the nature of the data collection procedure and unwillingness of many of the companies to divulge employee attitudes information, managers heading the unit requested the number of managerial and employee surveys that they are willing and/or able to distribute and the Japanese faculty member reproduced all the necessary forms to create a packet that included a Japanese version of the survey (either managerial or employee version: listed in Appendix A), a cover letter, and a pre-paid, university-addressed envelope for the requested number. The self-addressed envelope was addressed to the faculty member's university address in Kunitachi-shi, Tokyo, Japan and all the surveys (i.e., both managerial and employee) were directly returned to this address. Therefore, the managers and employees who responded to the surveys were convenient samples selected by the managers. This might raise a concern that the
49
responses are biased upward because the manager might have selected only those managers and employees who would respond favorably to the questions. However, this concern is mitigated because the mean of the responses across the units were generally around the mid-point of the scale anchors (ranged from 3.42 to 5.15 out of 1 to 7 Likerttype anchors) and the standard deviations ranged between .52 and .83. Therefore, the responses do not seem overly skewed in a positive way (i.e., all favorable). In addition, units from Izumigou had lower ratings for High Performance Work System and the level of human capital than those obtained from the other units, which indicates that not all respondents chosen rated the items in a favorable manner. Both the managers and the employees who received the survey were told in the cover letter the purpose of the project. First, the importance of employees as a valuable asset to the organization was noted as the reason for the emergence of strategic human resource management as a field, which was used as a primer for the study objective. They were told that the main purpose of the project was to examine the relationship between a set of human resource management practices and firm performance as well as the mediating mechanism through which this occurs. They were also told that this project is also for completion of my dissertation. Furthermore, in the cover letter, the anonymity of their responses were guaranteed as the managerial and employee surveys distributed for the unit only had the code that identifies the unit, i.e., every survey for a particular unit had the same identification code, and voluntary nature of the responses were emphasized such that some managers and employees chose not to fill in demographic information if they felt uncomfortable in doing so.
50
Given the level of analysis involved in the dissertation, it was critical that high response rate be obtained from the managers as well as the employees to represent the unit (i.e., establishments or work locations). To ensure a higher response rate, I followed up with the managers in charge through phone calls and/or e-mails two weeks after the initial survey distribution. A second round of follow-up was conducted by mail with another sets of surveys enclosed. This was followed up again by phone calls and/or e-mails for all potential units. A final round of follow-up was conducted by mail with another survey packet. However, given that the managers requested the number of packets that he/she was willing to distribute, the representativeness of the respondents cannot be ascertained. Survey Translation Procedure The surveys were translated into Japanese through an iterative process. Several steps were taken to ensure the effectiveness of the translation process. First, a Japanese tenured full professor in Human Resource Management at the Tokyo Economics University in Kunitachi, Japan, who is also proficient in English, provided his assistance in translating the English version of the surveys into Japanese. This faculty member specializes in cross-cultural research on compensation between the U.S. and Japan and has also taught at the University of North Carolina at Chapel Hill and North Carolina State University for several years during his academic career. The translated versions of the surveys were sent back to me, the principal investigator. As a Japanese native, I checked the integrity and validity of the items and noted any concerns regarding the translation. These comments were sent back to the
51
Japanese faculty member along with the Japanese surveys. He then addressed these concerns and sent back the revised version of the Japanese surveys back to me, the Ph.D. candidate. This process was done several times until no further concern emerged. To validate the survey translation, two Japanese employees in no way affiliated with this study were asked to read through the Japanese surveys for readability and ease of comprehension. Any concerns were noted and sent back to the Japanese faculty member who further revised the surveys to address these concerns until no further concern surfaced. As a final check, a third Japanese native translated the survey back into English, and the Japanese and English versions were compared for any discrepancies; none was detected. In general, I followed the procedures recommended by Brislin (1981) for survey translations across different languages and found these surveys to be equivalent in meaning. Aggregation Issues The survey items were reworded to reflect the unit-level of analysis by changing the focus of items to the establishment. For instance, items for measuring psychological contract was reworded as the following: "The employees make personal sacrifices for this unit" with an instruction to answer the items for the average employees in the unit. This approach is consistent with the guidelines by multilevel scholars (House, Rousseau, & Thomas-Hunt, 1995; Klein, Dansereau, & Hall, 1994) to specify and explicate the level of constructs in the study. The interrater agreement was assessed via Rwg (James, Demaree, & Wolf, 1984, 1993) for all the variables included in the dissertation. The average Rwg for High Performance Work System was .96 (Median = .97) for the 30- item scale; for human capital, it was .97 (Median = .97); for collective
52
normative contract, it was .81 (Median = .88); for labor productivity, it was .90 (Median = .92); for unit-level task performance, it was .95 (Median = .96); for unit-level organizational citizenship behaviors, it was .97 (Median = .98); and for overall unit performance, it was .96 (Median = .97). These results provide some evidence that aggregation of the variables is justifiable because the Rwgs exceed the recommended value of .60 (James, 1982). More specifically, Schneider and Bowen (1985: 426) noted, "the appropriate test for withinsetting agreement would be a measure of homogeneity rather than an index like analysis of variance (ANOVA) or the intra-class correlation coefficient (ICC) that depend upon between setting differences for significance." In addition, Lindell and Brandt (1999) have shown Rwg's superiority as an index of inter-rater agreement when compared to other indices such as Content Validity Index (CVI: Lawshe, 1975) or T (Tinsley & Weiss, 1975). Furthermore, Rwg can be considered a measure of interchangeability among raters (Ryan et al., 1996), which is an important characteristic of this measure for the current study, given that I expected some establishments to have a single managerial response for the variables of interest. Given that the level of agreement among all of the raters is well above the recommended value of .60 by James (1982), and substantially exceeds more commonly accepted value of .70 (cf., Klein et al., 2000), there is some empirical evidence suggesting that one rater can reasonably measure the aggregated variables in an accurate manner. In addition, however, I examined intraclass correlation coefficients (ICC) to assess the variability in managerial and employee responses. ICC(1) compares the variance between units of analysis (establishments in this case) to the variance within
53
units of analysis using the individual ratings of each respondent (Schneider, White, & Paul, 1998). On the other hand, ICC(2) compares the between unit variability to that of within unit variability, using the average ratings of respondents within each unit (Bartko, 1976). ICC(1) and ICC(2) were calculated by using Bartko's (1976) formula with a one-way random-effects analysis of variance (ANOVA). The ICCs were calculated for all the measures used in the present dissertation. ICC(1) for High Performance Work System, human capital, collective normative contract, labor productivity, unit-level task performance, unit-level organizational citizenship behaviors, and overall unit performance was .23, .22, .16, .25, .20, .26, and .25, respectively. ICC(2) for these same set of variables was .68, .55, .56, .59, .52, .60, and .59, respectively. "Although there are no strict standards of acceptability for either ICC(1) or ICC(2) values, James (1982) reported a median ICC(1) value of .12 in the organizational literature, and Glick (1985) recommended an ICC(2) cutoff of .60" (Schneider et al., 1998: 155) that can be used as a rule-of-thumb to compare the ICC values obtained for the current dissertation. All of the ICC(1) values obtained for the dissertation exceeds the median value of .12 reported by James (1982) while ICC(2) values either approach or equal (in the cases of human capital, collective normative contract, and labor productivity, unit-level task performance, unit-level organizational citizenship behaviors, and overall unit performance), or exceed (in the cases of High Performance Work System) this value. Thus, all of the indicators appear to justify aggregation of the variables to the unit-level of analysis. Therefore, responses from all the managers in a unit, except where only one managerial response was obtained, were
54
averaged to create the unit-level variables for the level of human capital, perceived labor productivity, unit-level task performance, unit-level overall organizational citizenship behaviors, and unit performance. All responses from the employees for a unit were averaged to create the unit-level variables of High Performance Work System and collective normative contract. Justifications for the rating sources are provided below. Measures To maximize the content validity of the questionnaire, all of the items were obtained from an existing measures, except two items for labor productivity developed for the present study. Each scale contains multiple-items scales with 7-point Likert-type anchors ranging from strongly disagree (1) to strongly agree (7), unless otherwise noted. All of the items are affixed in Appendix C. To reduce potential for common method bias, managerial and employee responses were used to evaluate different scales. In general, managers are considered more reliable and valid source of performance data, i.e., unit performance (e.g., Gupta, 1987; Gupta & Govindarajan, 1986), as well as HR-related outcomes such as labor productivity (Way, 2002). In addition, supervisory ratings are frequently used to measure individual employee's task performance and organizational citizenship behaviors (e.g., LePine, Erez, & Johnson, 2002; Organ & Ryan, 1995). Thus, managerial ratings were used for these measures. In addition, managers were deemed more knowledgeable about the unit's level of human capital as compared to the employees. Therefore, managerial ratings of the level of human capital were also used. For collective normative contract, aggregate employee responses are more appropriate,
55
given the psychological origin of the concept. Given that managerial ratings were used for many of the variables included in the dissertation, employee responses of High Performance Work System were used to reduce potential for common method biases. In addition, the employees' perceptions of HR practices in use are more important in influencing their attitudes and behaviors than what the managers say. Therefore, aggregate employee ratings were used for High Performance Work System. High Performance Work System. Due to space limitation, Japanese faculty in Human Resource Management, initially, selected thirty items from Lepak and Snell's (2002) sixty-item HR practices scale that are deemed most appropriate for the Japanese companies. The Japanese faculty selected these items that would be considered a part of High Performance Work System in the Japanese context for which the employees will be familiar with. Thus, for example, items such as "employees are required to participate in cross-functional teams and networks" and "the selection process emphasizes their ability to collaborate and work in teams were not selected because these are typically used in Japanese companies. A system of HR practices used by the establishment, i.e., High Performance Work System, was obtained from the employees and aggregated to the unit level. To examine the factor structure of these items, exploratory factor analysis with principal component extraction was initially performed. Using a combination of KaiserGuttman rule of Eigen values greater than one (Guttman, 1954; Kaiser, 1970) and a Scree test (Chattell, 1966) of Eigen values plotted against factors indicated a singlefactor solution. More specifically, first factor had an Eigen value of 8.14, which explained 27.15% of variance. This was followed by Eigen values of 3.01, 1.98, 1.82,
56
1.56, 1.45, 1.34, 1.19, and 1.13. Given that I expected all of the items to compose High Performance Work System, I also conducted principal axis factoring where singlefactor solution is imposed. The result of the factor analysis was comparable to that of principal component analysis. However, the factor loadings of nine items were below .35. Therefore, I dropped those items that did not load above .35 from the analysis and run another principal axis factoring. The remaining twenty-one items loaded at or above .35 on the single factor and this factor explained 35.82% of variance in the items and had an Eigen value of 7.88, followed by additional Eigenvalues above 1 of 1.93, 1.62, 1.26, and 1.16. The factor loadings of these twenty-one items are depicted on Table 4. Items selected for High Performance Work System is described in Appendix C with asterisks (*). Table 4. Factor Loadings for 21-Item High Performance Work System Scale HPWS Items Job Design 2 Job Design 3 Job Design 5 Job Design 6 Selection 1 Selection 2 Selection 3 Selection 4 Selection 5 Training & Development 1 Training & Development 3 Training & Development 4 Training & Development 5 Performance Appraisal 1 Performance Appraisal 2 Performance Appraisal 3 Performance Appraisal 4 Compensation 1 Compensation 2 Compensation 3 Compensation 7 57 Factor Loading .37 .41 .62 .55 .65 .60 .35 .44 .60 .67 .58 .41 .68 .65 .68 .75 .75 .44 .68 .64 .50
The resulting 21-item scale had a reliability of .90. This alpha is comparable to the one that Lepak and Snell (2002) obtained for their commitment HR scale (? = .89). In addition, the mean Rwg of .96 for this scale was highly comparable to that of .97 obtained by Lepak and Snell (2002). Therefore, these results provide some evidence for the validity of the translation procedure and the resultant scale. Human capital. Managers assessed the level of human capital that each unit possesses, using a four-item scale from Youndt and Snell's (2001) intellectual capital scale. A sample item states, "Our employees working in the unit are highly skilled." Cronbach's alpha for this four-item human capital scale was .92. Collective normative contract. Five items taken from Shore, Tetrick, and Berksdale's (1999, as cited in Rupp & Cropanzano, 2002) psychological contract/social exchange scale is reworded to capture the notion of normative contract at the unit level of analysis (cf., Rousseau, 1995). Rupp and Cropanzano (2002) reworded the items to assess relational aspect of social exchange relationships for supervisor and organization and found alphas of .89 and .91. For this variable, employee responses were used, as employee response was deemed most appropriate because of the psychological nature of concept (cf., Rousseau, 1995). A reworded, sample item states, "Our relationship with the establishment continues to evolve and develop." The collective normative contract scale had a reliability of .86. HR-Related Outcomes Labor productivity. Three-item scale of perceived labor productivity was used to measure the labor productivity of the establishments, rather than an objective measure
58
of labor productivity, i.e., the logarithm of sales per employee, which is a widely used measure of productivity, due to the difficulty of collecting such objective measure at the unit-level (e.g., Gupta & Govindarajan, 1987). I supplemented an item from Way (2002) ("The employees' productivity is higher than those of my major competitors") with two additional items developed for the study ("The employees in this establishment produce outputs in an efficient manner," and "The employees' productive power is higher than those of the competitors"). The resulting three-item scale had a reliability of .91. Unit-level task performance. Five items used by Williams and Anderson (1991) to measure in-role performance were used to rate aggregate, unit level performance of the establishment. This measure of task/in-role performance has been used in several studies at the individual-level of analysis (e.g., Eisenberger, Armeli, Rexwinkel, Lynch, & Rhoades, 2001; Hui, Law, & Chen, 1999; Lynch, Eisenberger, & Armeli, 1999; Turnley, Bolino, Lester, & Bloodgood, 2003). The coefficient alphas for these studies ranged from .75 to .93. Managers rated the level of employee task performance for the unit as a whole. A sample item stated, "The employees in our unit perform tasks that are expected of them." The reliability of this scale was .82. Given that Hui et al.'s (1999) study of Chinese employees had the lowest reliabilities of .75 reported in previous studies for this scale, a reliability of .82 obtained at the unit-level of analysis for this dissertation compares favorably to that by Hui et al. (1999), which provides an additional evidence that the translation procedure was done properly and the resultant scale is valid.
59
Unit-level overall organizational citizenship behaviors. An abbreviated, ten-item version of the Williams and Anderson's (1991) organizational citizenship behavior scale was used to evaluate the level of organizational citizenship behaviors exhibited by the employees in the unit. The original items were created to tap into two different aspects of organizational citizenship behaviors that are targeted toward the individuals (OCBI) and the organization (OCBO). Williams and Anderson (1991) reported a reliability of .88 and .75 for OCBI and OCBO, respectively. Similarly, Turnley et al. (2003) reported a reliability of .83 and .88 for OCBI and OCBO, respectively. Three items from the organizational citizenship behavior targeted toward the organization was initially eliminated because of an anticipated confound between inrole performance and organizational citizenship behaviors. Managers rated the remaining eleven items for the entire unit. Given recent evidence that OCBI and OCBO do not have distinct antecedents or consequences (LePine et al., 2002) and relatively high correlation between OCBI and OCBO for the present dissertation (r = .66, p< .001), all the items, except one, were combined to create unit-level, overall organizational citizenship behaviors. One item had a cross-loading with other performance indicators and, hence, dropped from the scale. The result of this factor analysis is described after the unit performance measure below. The reliability of this ten-item scale was .87. Unit Performance Delaney and Huselid's (1996) eight-item perceived organizational performance measure was adapted to assess unit's level of performance compared to other units similar to the focal unit. An example item asks the manager to rate the establishment's
60
performance for the past three years as compared to that of similar establishments (i.e., "How would you compare the establishment's [performance] over the past 3 years to those of the other establishments that do the same kind of work?" Delaney and Huselid (1996) found a reliability of .85 for perceived organizational performance (for 590 firms) from a nationally representative sample of U.S. work establishments, because they obtained their data from the National Organization Survey, which was conducted in 1991 with support from the National Science Foundation. Gupta (1987) also used a similar comparative subjective measure of performance. Three items from this overall unit performance measure were dropped due to low loadings and/or high cross-loadings with other performance indicators, as described below. This five-item measure of unit performance had a reliability of .92. Given that the reliability of this scale is comparable to the ones found for the representative sample of U.S. establishments by Delaney and Huselid (1996), this provides some evidence that the survey translation was satisfactorily performed and the resulting scale is valid. Given that all the HR-related outcomes and unit performance were obtained from the managers, I conducted exploratory factor analysis with principal component extraction with oblique rotation to assess the factor structure and discriminant validity of the measures. I expected four-factor structures for unit-level task performance, unitlevel overall organizational citizenship behaviors, perceived labor productivity, and overall unit performance. Thus, four-factor structure was initially imposed. The result of the factor analysis is shown in Table 5. The four factors collectively explained 64.49% of variance with Eigenvalues of 10.81, 3.89, 1.85, and 1.52 for the first four factors. The items generally loaded significantly onto the expected construct with few exceptions.
61
Table 5. Results of Exploratory Factor Analysis on the HR-Related Outcomes and Unit Performance Factor 1 .01 .30 .03 .28 .26 .18 .00 -.03 -.02 .82 .59 .47 .71 .66 .63 .44 .59 .55 .36 .48 -.34 -.14 -.10 .09 .17 .18 .50 .47 10.81 38.60 Factor 2 -.09 -.26 -.38 .18 -.04 -.05 -.01 -.03 -.34 .00 -.06 .05 -.19 .04 -.13 -.47 .12 .16 .24 -.10 -.45 -.63 -.65 -.81 -.87 -.72 -.43 -.46 3.89 13.88 Factor 3 -.06 -.02 .15 .62 .38 .60 .64 .76 .58 -.18 .15 .12 -.06 .19 -.04 .15 .25 .37 .08 .07 .15 .17 .11 .02 .11 -.01 -.01 .05 1.85 6.60 Factor 4 -.90 -.61 -.64 -.39 -.42 -.16 -.10 .14 .14 -.19 -.04 .20 .07 -.09 -.00 .03 -.07 .03 -.14 -.22 -.39 -.29 -.23 -.07 .06 -.23 -.09 -.08 1.52 5.42
Labor Productivity 1 Labor Productivity 2 Labor Productivity 3 Task Performance 1 Task Performance 2 Task Performance 3 Task Performance 4 Task Performance 5 Task Performance 6 OCBI 1 OCBI 2 OCBI 3 OCBI 4 OCBI 5 OCBI 6 OCBI 7 OCBO 1 OCBO 2 OCBO 3 OCBO 4 Overall Unit Performance 1 Overall Unit Performance 2 Overall Unit Performance 3 Overall Unit Performance 4 Overall Unit Performance 5 Overall Unit Performance 6 Overall Unit Performance 7 Overall Unit Performance 8 Eigen Value % variance explained
Note: Principal axis factoring analysis with oblique rotation, 4-factor solution Using a factor loading at or above .35 and differences of at least .10 between crossloadings, an item was deleted from task performance and organizational citizenship behavior - interpersonal (OCBI) each, as well as three items from overall unit
62
performance. The resulting factor loadings were cleaner that explained 66.18% variance in the items with Eigenvalues of 8.61, 3.53, 1.80, and 1.29. The items retained for further analyses are listed in Appendix C with asterisks (*). Analytic Procedures A combination of hierarchical regression analysis and path analytic procedure in regression was used to examine individual hypotheses as well as the overall model. Although it would have been better to be able to conduct structural equation modeling analysis, the sample size of 76 units does not allow this type of analysis. Even if I use single indicators with reliability corrections (cf., Jöreskog & Sörbom, 1989) by fixing the loadings by the square root of the reliability of the scale (?r) and the measurement errors by the product of the unreliability by the variance, i.e., (1 - r)*SD2, it would still require, at least, 26 parameter estimates (18 structural path estimates, 1 factor variance, and 7 disturbance terms), not to mention possible covariances among the disturbance terms (for HR-related outcomes). The sample size to parameter estimate ratio of 5 to 1 is recommended as a minimum requirement for the structural equation modeling to derive accurate parameter estimates (Bentler & Chou, 1987) and the ratio for the dissertation does not reach this minimum (ratio = 2.92). Therefore, regression analyses were used to examine the hypotheses, and path analysis in regression is used to evaluate the overall model. By using the path analytic procedure, I do not account for potential unreliability of the measures, which may result in some inaccuracy with regard to the beta coefficients estimated. On the other hand, using a structural equation modeling without having enough sample size also may result in inaccurate parameter estimates. However,
63
given relatively high reliabilities for the variables used in the dissertation noted above, this may be less of an issue in this particular case. Hence, I chose to use path analytic procedure in regression to be able to model the interrelationships among the variables included in the model. In addition, following Cohen and Cohen (1983), I standardized the level of human capital and collective normative contract before creating interaction terms to reduce multicollinearity problems inherent in higher order terms.
64
Chapter 5 Results Table 6 shows the means, standard deviations, and correlations among the variables of interest. Unstandardized means and standard deviations of the variables for the level of human capital and collective normative contract are listed for informational purpose only because I standardized these variable with a z-score transformation for the analyses to create a product term for the moderated regression analyses. The first three variables are dummy coded, control variables for those companies who supplied multiple units. The first dummy coded variable (control 1) is for Mitsui Kinzoku (1 = units from Mitsui Kinzoku, 0 = all the other) with thirteen units. The second dummy coded variable (control 2) is for Izumigou (with four units) and the last dummy coded variable (control 3) is for Cannon (with five units). In addition, to examine the effect of the industry, I created six dummy coded variables (dummy 1: 1 = Construction, 0 = all the other; dummy 2: 1 = Manufacturing, 0 = all the other; dummy 3: 1 = Transportation, Communication, Electric, Gas, & Sanitary Services, 0 = all the other; dummy 4: 1 = Wholesale Trade, 0 = all the other; dummy 5: 1 = Retail Trade, 0 = all the other; and dummy 6: 1 = Finance, Insurance, & Real Estate, 0 = all the other) and entered these variables in step 1 of the regression analyses for perceived labor productivity, unit-level task performance, unit-level overall organizational citizenship behaviors, and overall unit performance. None of the beta coefficients for these six dummy coded variables achieved significance at .05 level. The results of this regression analyses are shown in Table 7.
65
Table 6. Descriptives of All the Variablesa Variables 1. Control 1 Mitsui Kinzoku 2. Control 2 Izumigou 3. Control 3 Cannon 4. High Performance Work System 5. The Level of Human Capitalb b 6. Collective Normative Contract 7. Perceived Labor Productivity 9. Unit-Level Task Performance 8. Unit-Level Organizational Citizenship Behavior 9. Overall Unit Performance M 0.17 0.05 0.05 4.15 4.60 4.50 4.59 1.94 SD 0.38 0.22 0.22 0.55 0.77 0.70 0.83 0.61 1 -0.11 -0.11 0.15 -0.09 0.14 -0.11 0.01 -0.06** -0.29* -0.24 -0.12 -0.02 -0.20 0.30** 0.29** 0.20 0.19** 0.38 2 3 4 5 6 7 8 9
0.25* 0.54 * 0.04** 0.16** 0.71 ** 0.22* 0.47 0.52 0.28 0.42 **
5.17 0.52 -0.11 -0.01** 0.17** 0.33** 0.35** 0.28* 0.39** 0.55** 4.68 0.72 -0.13 -0.37 0.30 0.26* 0.52** 0.23 * 0.67** 0.44** 0.28*
Note: a N = 76; b means and SDs are provided for informational purpose only
*
p?? .05; ** p < .01 (two-tailed)
67
Table 7. Regression Results for Industry Effectsa Dependent Variables Unit-Level Task Unit-Level Performance OCBs 1.85*** 5.16*** .02 .20 -.16 .11 -.13 .07 0.91 .07 1.46 .11 .08 .00 -.04 .02 .03 .06 0.15 .01
Variable Constant Step 1: Industry Controls Industry Dummy 1: Construction Industry Dummy 2: Manufacturing Industry Dummy 3: Transportation, Communication, Electric, Gas, & Sanitary Services Industry Dummy 4: Wholesale Trade Industry Dummy 5: Retail Trade Industry Dummy 6: Finance, Insurance, & Real Estate Overall F R2
Perceived Labor Productivity 4.70*** -.05 -.14 -.11 .17 .02 -.08
Overall Unit Performance 4.44*** .01 .17 .06 .25 .22 .10 0.91 .07
Note: a n = 76. Values are standardized estimates. None of steps or the betas was significant at * p?? .05 (two-tailed).
68
The directions of the correlations among the variables of interest were in the expected direction. For the dummy coded company variables, units from Izumigou Kabushiki Gaisha had negative correlations with the variables included in the dissertation while units from Cannon had positive correlations, in general. To examine the hypotheses, I ran several regressions. For Hypotheses 1, 2, 5, and 6, the proposed predictor was entered one at a time to assess the independent effect of the predictor. For Hypotheses 3, 4, 7, and 8, which posit mediating effects of certain variables, I followed Baron and Kenny's (1986) four-step procedure where 1) the direct effect of independent variable on the dependent variable is examined, which shows that there is an effect that may be mediated; 2) the direct effect of independent variable on the "proposed" mediator is examined to illustrate that the independent variable is related to the mediator; 3) the direct effect of mediator on the dependent variable is evaluated while the effect of independent variable is taken into account; and 4) the reduction in the beta coefficient associated with the independent variable when the effect of mediator is accounted for. The steps 3 and 4 essentially are tested in the same regression analysis. The procedure used to test mediating effect and associated results are explained and discussed in more detail in the section describing the appropriate hypotheses. All of the beta coefficients reported here are standardized ones. As depicted in Table 8, Hypothesis 1, which posited a positive relationship between High Performance Work System and the level of human capital that the unit possesses, was not supported (? = .14, p> .1). High Performance Work System explained only 2% of variance in the level of human capital for the unit (F = 1.22, p> .1) above and beyond the effect of controls in step 1 (R2 = .14, F = 4.01, p< .01). For
69
the level of human capital, two control variables had significant betas, -.24 (p < .05) for Izumigou and .27 (p < .05) for Cannon. Given that Izumigou is in the hospitality industry (i.e., hotels) and Cannon in the electronics industry, these results seem indicative of the level of human capital necessary for the lower-level employees' jobs for these companies. Table 8. Regression Results for Hypotheses 1 and 2a Dependent Variable Perceived Unit-Level Labor Task Productivity Performance -.09 .04 -.03 -.18 .18 .38***
Variable 1. Control 1 Control 2 Control 3 2a. High Performance Work Systems 2b. Level of Human Capital Overall F R2 ?F ?R2
Level of Human Capital -.08 -.24* .27**
UnitLevel OCB -.10 -.01 .15
.14
.75*** 3.35** .16 1.22 .02 19.33*** .52 70.67*** .48 .16
.44*** 9.25*** .34 17.55*** .10
.34** 2.78* .14 8.11**
Note: a n = 76. Values are standardized estimates.
b
Beta coefficients shown are standardized betas from each step. p?? .10;
*
+
p?? .05;
**
p?? .01; *** p?? .001 (two-tailed)
For Hypothesis 2, results indicated strong support for all three HR-related outcome variables considered. Particularly, the level of human capital was significantly,
70
positively related to perceived labor productivity of the unit (? = .75, p< .001;??F = 70.67, p< .001,??R2 = .48), unit-level task performance (? = .44, p< .001;??F = 17.55, p< .001,??R2 = .16), and unit-level organizational citizenship behaviors (? = .34, p< .01;??F = 8.11, p < .01,??R2 = .10), after the effect of company dummies are accounted for. For perceived labor productivity and unit-level organizational citizenship behaviors, control variables were not significantly related. For unit-level task performance, Cannon (control 3) had significantly, positive beta (.38, p< .001). For Hypothesis 3, the condition that the independent variable, i.e. High Performance Work System, is related to the mediator (second condition put forth by Baron and Kenny [1986] was not satisfied (see Hypothesis 1). Therefore, there is nothing that the High Performance Work System can mediate. Hence, Hypothesis 3 is not supported. To examine the mediating effect of HR-related outcomes on the relationship between the level of human capital and overall unit performance (Hypothesis 4), I followed the same procedure by Baron and Kenny (1986) noted above. Specifically, I ran separate regression analyses where I entered the level of human capital in the second step, following the three company dummies and, then, entering only one of the HR-related outcomes in the third step by itself. The results of the analyses are illustrated in Table 9. As shown in Table 9, Model 2, the level of human capital was significantly, positively related to overall unit performance, which satisfies the first condition that the independent variable is related to the dependent variable (? = .39, p< .001). The second condition that the independent variable be related to the mediator is satisfied by supporting evidence for Hypothesis 2, i.e., the level of human capital was significantly,
71
positively related to all three HR-related outcomes. The third condition that the mediator is related to the dependent variable, i.e., overall unit performance, was only satisfied for perceived labor productivity. Therefore, only perceived labor productivity has the potential to act as a mediator. When examining Model 3a, the standardized beta coefficient associated with the level of human capital became non-significant (? = -.15, p > .1) from that in Model 2 (?? = .39, p < .001). Therefore, perceived labor productivity acted as a full mediator of the relationship between the level of human capital and overall unit performance while the other two HR-related outcomes did not act as a mediator. Therefore, there results provide some support for Hypothesis 4. Hypotheses 5 through 8 dealt with the mediating mechanism from High Performance Work System to overall unit performance through collective normative contract and HR-related outcomes. For Hypothesis 5, the results depicted in Table 10 provide strong support for the hypothesis that posited a positive relationship between High Performance Work System and collective normative contract (? = .52, p< .001). High Performance Work System explained 22% additional variance (?F = 20.69, p< .001) over and above that accounted by the firm control variables (R2 = .07, F = 1.97, p > .1)
72
Table 9. Results for Hypothesis 4a Overall Unit Performanceb Model 2 Model 3a Model 3b -.11 -.08 -.12 -.28** -.39*** -.27** .16 .17 .11 .39*** -.15 .31** .73*** .18 10.38*** .37 14.92*** .13 22.92*** .62 46.50*** .25 8.99*** .39 2.55 .02
Variable Step 1. Control 1 Control 2 Control 3 Step 2. Level of Human Capital Step 3a. Perceived Labor Productivity Step 3b. Unit-Level Task Performance Step 3c. Unit-Level Organizational Citizenship Behaviors Overall F R2 Change in F 2 Change in R Note: a n = 76. Values are standardized estimates.
b
Model 1 -.14 -.37*** .26**
Model 3c -.10 -.29 .15 .35 .12 8.65*** .38 1.48 .01
7.43*** .24
Beta coefficients shown in Model 2 are standardized betas from step 1 and 2.
+
p?? .10;
*
p?? .05;
**
p?? .01; *** p?? .001 (two-tailed)
73
Hypothesis 6 posited that collective normative contract is positively related to HR-related outcomes. The results of the regression analyses provided moderate but consistent support for this hypothesis. Specifically, collective normative contract was marginally, significantly related to perceived labor productivity (? = .21, p< .1;??F = 3.08, p< .1,??R2 = .03) and unit-level task performance (? = .19, p< .1;??F = 3.03, p< .01,??R2 = .03) but significantly, positively related to unit-level overall organizational citizenship behaviors (? = .28, p< .05;??F = 5.88, p< .05,??R2 = .07). Table 10. Results for Hypotheses 5 and 6 Dependent Variables Perceived Unit-Level Labor Task Productivity Performance -.09 -.03 .18 .04 -.18 .38***
Variable Step 1. Control 1 Control 2 Control 3 Step 2a. High Performance Work System Step 2b. Collective Normative Contract Overall F R2 Change in F 2 Change in R
Collective Normative Contract .15 -.09 .21+
Unit-Level OCB -.10 -.01 .15
.52***
.21+ 7.40*** .29 21.97*** .22 1.64 .08 3.08+ .03
.19+ 4.82** .21 3.03+ .03
.28* 2.15+ .11 5.68* .07
Note: a n = 76. Values are standardized estimates.
b
Beta coefficients shown are standardized betas from each step.
+
p?? .10;
*
p?? .05;
**
p?? .01; *** p?? .001 (two-tailed)
74
Table 11. Results for Hypothesis 7 Perceived Labor Productivity Unit-Level Task Performance Model Model Model Model Model Model 1 2 3a 1 2 3b -.09 -.12 -.13 .04 -.02 -.02 -.03 .01 .01 -.18 -.08 -.08 *** .18 .13 .12 .38 .26 * .26 * .15 .05 .19 1.12 .04 1.17 .07 1.28 .02 1.32 .09 1.86 .02 5.27*** .18 7.30*** .29*** 11.17 .11 .37*** .36** .03 5.78*** .29 0.06 .00 0.92 .04 2.95* .14*** 8.76 .10 Unit-Level Overall OCB Model Model Model 1 2 3c -.10 -.16 -.17 -.01 .09 .08 .15 .04 .04 .36** .29* .15 2.63* .16 1.29 .02
Variable Step 1. Control 1 Control 2 Control 3 Step 2. High Performance Work System Step 3. Collective Normative Contract Overall F R2 Change in F 2 Change in R
Note: a n = 76. Values are standardized estimates.
b
Beta coefficients shown are standardized betas from the last steps.
+
p?? .10;
*
p?? .05;
**
p?? .01; *** p?? .001 (two-tailed)
75
Hypothesis 7, which posited the mediating role of collective normative contract, did not receive support. Table 11 shows the standardized betas associated with each step. Given that High Performance Work System is not significantly related to perceived labor productivity (? = .15, p> .1) when entered in the second step, collective normative contract cannot mediate the relationship between High Performance Work System and perceived labor productivity because this does not satisfy Baron and Kenny's (1986) first condition. Similarly, High Performance Work System was not significantly related to unit-level task performance (? = .03, p> .1) or unit-level organizational citizenship behaviors (? = .15, p> .1) when entered in the third step, which does not satisfy the third condition that the mediator is related to the dependent variable (Baron & Kenny, 1986). Thus, for unit-level task performance and unit-level organizational citizenship behaviors, there is no mediation. Hence, the results of these analyses do not support the hypothesis that collective normative contract acts as a mediator of the relationship between High Performance Work System and HR-related outcomes. Rather, collective normative contract appears to have indirect relationships to High Performance Works System and HR-related outcomes.
76
Table 12. Results for Hypothesis 8 Overall Unit Performanceb Model 2 Model 3a Model 3b -.17 -.09 -.17+ -.36*** -.36*** -.31** .23* .15 .13 .17 .04 .11 .62*** .30** 6.31*** .26 2.50 .03 22.16*** .61 63.38*** .35 7.03*** .33 7.58** .07
Variable Step 1. Control 1 Control 2 Control 3 Step 2. Collective Normative Contract Step 3a. Perceived Labor Productivity Step 3b. Unit-Level Task Performance Step 3c. Unit-Level Organizational Citizenship Behaviors Overall F R2 Change in F 2 Change in R Note: a n = 76. Values are standardized estimates.
b
Model 1 -.14 -.37*** .26**
Model 3c -.14 -.36*** .21* .11
7.427*** .23
.20+ 5.93*** .30 3.52+ .04
Beta coefficients shown in Model 2 are standardized betas from step 1 and 2.
+
p?? .10;
*
p?? .05;
**
p?? .01; *** p?? .001 (two-tailed)
77
To examine Hypothesis 8 that proposed the mediating role of HR-related outcomes on the relationship between collective normative contract and overall unit performance, I regressed overall unit performance onto collective normative contract to examine if the independent variable (i.e., collective normative contract) is related to the dependent variable (condition 1). The result of the regression analysis indicated that collective normative contract was not significantly related to overall unit performance (? = .17, p> .1). Therefore, HR-related outcomes do not have a relationship to mediate. Hence, there is no support for the mediation. Table 12 depicts these results. Therefore, Hypothesis 8 was not supported. Finally, the interaction hypothesis for the level of human capital and collective normative contract (Hypothesis 9) was tested by creating an interaction term, using a zscore transformed variables for the level of human capital and collective normative contract. The results of the moderated regression analyses where the level of human capital and collective normative contract was entered in Step 2 after controlling for the effect of company, followed by the introduction of the interaction term in Step 3, did not provide much support for this hypothesis. The beta coefficients associated with the interaction term was .10 (p > .1) when perceived labor productivity was the dependent variable, -.09 (p > .1) when unit-level task performance was the dependent variable, and -.21 (p < .1) when unit-level organizational citizenship behavior was the dependent variable. Therefore, Hypothesis 9 did not receive any support. Table 13 depicts these findings.
78
Table 13. Results for Hypothesis 9 Dependent Variablesb Perceived Labor Unit-Level Task Unit-Level Productivity Performance OCB -.09 .03 .18 .036 -.176 .377 -.10 .01 .15
Variable Step 1. Control 1 Control 2 Control 3 Step 2. the Level of Human Capital Collective Normative Contract Step 3. The Level of Human Capital* Collective Normative Contract Overall F R2 Change in F 2 Change in R
.75*** .23**
.44*** .21*
.35** .29**
-.10 15.81*** .58 1.34 .01
-.09 7.29*** .39 0.77 .01
-.21+ 3.75** .25 3.03+ .03
Note: a n = 76. Values are standardized estimates.
b
Beta coefficients shown are standardized betas from the last steps.
+
p?? .10;
*
p?? .05;
**
p?? .01; *** p?? .001 (two-tailed)
To examine the proposed model more in its entirety, path analytic procedure in regression was used. The results of this path analytic, regression analyses are shown in Table 14. The standardized beta coefficient values are shown from the last step of the analysis for each dependent variable along with the incremental variance (?R2) accounted for the last step.
79
Table 14. Path Analytic Regression Analyses Resultsa Dependent Variablesb Variablec 2. High Performance Work Systems 3. Level of Human Capital Collective Normative Contract 4. Level of Human Capital * Normative Contract 5. Perceived Labor Productivity Unit-Level Task Performance Unit-Level OCBs Overall F R2 Change in F 2 Change in R Level of Human Capital .14 Collective Normative Contract .52*** Perceived Labor Productivity -.11 .77 *** .28** -.14 .82*** .33*** -.13 Unit-Level Task Performance .28* .40*** .08 .26* .42*** .09 -.04 Unit-Level OCBs .22 .32** .19 Overall Unit Performance
-.02 .18 .38*** -.14 .03 .25 + -.08 -.17 .70*** .14 -.05 11.30*** .64 12.63*** .21
3.35** .16 1.33 .02
7.40*** .29 21.97*** .22
15.74*** 14.01*** .58 .59 42.19*** .52 2.09 .01
8.54*** .43 8.10*** .14
7.25*** .43 0.15 .00
3.66** .24 4.50* .10
3.47** .26 2.00 .02
Note: ab n = 76. Values are standardized estimates Beta coefficients shown are standardized betas from the last steps. c Beta coefficients for the controls are not shown for clarity. + p?? .10; * p?? .05; ** p?? .01; *** p?? .001 (two-tailed)
81
Figure 8. Path Analytic Results ? = .77, p < .001 Level of Human Capital (Managerial Rating) ? = .40, p < .001 Perceived Labor Productivity for the Unit (Managerial rating) ? = .70, p < .001
High Performance Work System (Employee Rating)
? = .28, p < .05
Unit-Level Task Performance (Managerial rating)
Overall Unit Performance (Managerial Rating)
? = .52, p < .001
? = .28, p < .01 Normative Contract (Employee Rating) ? = .32, p < .001 Unit-Level OCBs (Managerial rating)
Note: non-significant paths are not shown
82
Figure 8 illustrates the results of the path analysis in a schematic form. To reduce cluttering, Figure 8 contains only those standardized beta coefficients that are significantly related (p < .1 and below) to other variables. In addition, it includes the rating sources (either employee response aggregates or managerial response aggregates) for each variable. In Figure 8, the dotted lines illustrates the main effects of High Performance Work System on the HR-related outcomes after the effects of the level of human capital and collective normative contract are accounted for, i.e., beta coefficients from the third step as shown in Table 14. In general, the results depicted in Table 14 and Figure 8 are similar to the ones found when each variable was treated separately. Table 15 summarizes the results for all of the hypotheses advanced in this dissertation. In general, some of the hypotheses received moderate to strong support while others did not receive support Table 15. Summary Results for the Hypotheses Hypothesis 1. Hypothesis 2. Hypothesis 3. Hypothesis 4. High Performance Work System is positively related to the level of human capital - not supported The level of human capital is significantly related to HR-related outcomes - strongly supported (3 out of 3 HR-related outcomes) The level of human capital mediates the relationships between High Performance Work System and HR-related outcomes - not supported HR-related outcomes mediate the relationship between the level of human capital and unit performance - some support (1 out of 3 HRrelated outcomes, perceived labor productivity acted as a mediator) Hypothesis High Performance Work System is positively related to collective
83
5. Hypothesis 6.
normative contract - strongly supported Collective normative contract is significantly related to HR-related outcomes - moderately supported (significant for 1of the 3 HR-related outcomes and marginally significant for the other 2)
Hypothesis 7. Hypothesis 8. Hypothesis 9.
Collective normative contract mediates the relationships between High Performance Work System and HR-related outcomes - not supported HR-related outcomes mediate the relationships between collective normative contract and unit performance - not supported The level of human capital and collective normative contract interact to influence HR-related outcomes - very limited support was provided.
84
Chapter 6 Discussion The main objectives of the dissertation were to uncover the underlying mechanisms through which High Performance Work System affects unit performance by integrating two main theoretical perspectives, i.e., resource-based view of the firm and the behavioral perspective, that often have been employed and discussed but not directly tested. Although previous studies often implied that employees' human capital and behaviors are key to improving firm performance (e.g., Arthur, 1992, 1994; MacDuffie, 1995; Tsui et al., 1995; Wright et al., 1995), empirical examinations of the mediating mechanism, or what has been termed a "black box," through which a system of human resources management practices affects firm performance have only begun more recently (e.g., Batt, 2002; cf. AOM symposium: Unlocking the black box of strategic human resource management research; Burtons & O'Reilly, 2000; Chadwick, 2000; Wright, Gardner, Moynihan, & Park, 2000). Batt (2002: 587) echoed a similar concern more recently that "prior research is theoretically undeveloped and has not specified the mediating employee behaviors that explain the relationship between HR practices and performance." Applebaum et al. (2000) also noted that our understanding on the High Performance Work System's effects on the workers has been a major gap in the literature. The dissertation contributes to the strategic human resources management literature by explicating these mediating mechanisms. More specifically, the findings provide empirical support for the assertions of the resource-based view of the firm that human capital can be considered one resource that may have positive impact on (unit) performance (e.g., Barney, 1991) and of the
85
behavioral perspectives that employees' attitudes and behaviors are the critical mediating mechanisms through which High Performance Work System impact (unit) performance (e.g., Jackson et al., 1989; Schuler & Jackson, 1987). Similarly, it is one of the first studies in the area of strategic HRM to demonstrate that the previous use of the resource-based view of the firm and the behavioral perspectives (McMahan, Virick, & Wright, 1999; Wright & McMahan, 1992) is not unfounded, as it provides an empirical evidence that explicitly test the mediating role of the resources (the level of human capital) and employee attitudes and behaviors (collective normative contract) as well as HR-related outcomes. Although the level of human capital did not act as a mediator, it had significant, positive relationships to all of the HR-related outcomes. I found that there are, at least, one mediating mechanism where HR-related outcomes acted as mediators of the relationships between the level of human capital and overall unit performance and collective normative contract and overall unit performance. In addition, High Performance Work System was positively related to collective normative contract. Thus, High Performance Work System appears to have a positive influence on the shared perceptions regarding the terms of obligations or social exchange between establishment and the employees as a collective that have positive influence on perceived labor productivity of the unit. It may be the case that the increase in shared perceptions is accomplished through rigorous selection, higher incentives, better training and development, and appropriate performance appraisal and developmental feedback, among others, that attract and retain similar types of employees to the firm. Similarly, through better job design, better incentive structure, long-term focus on training and development and performance appraisal, it may be the
86
case that High Performance Work System improves upon the long-term relational aspects of employees' contract with the unit so that employees in general feel higher level of obligations to reciprocate. In fact, Wright and Snell (1991) noted that HRM practices contribute to both competence and behavior managements. They referred to competence management as those HRM practices such as aptitude tests and skill training programs that are aimed at ensuring high level of competency or human capital for a firm's workforce. Behavior management focuses on those HRM practices intended for controlling and coordinating employee behavior, which include those practices such as incentives and performance management. For instance, selection and recruitment practices can influence the level of knowledge, skills, and abilities new recruits possess whereas training and development affect the level of knowledge, skills, and abilities or competence existing employees possess. At the same time, compensation / reward practices and performance management practices typically influence the behaviors exhibited by the employees within the firm. Similarly, Huselid (1995) found two factors, skills and motivation, for the HR policy items he used. Although I did not find a relationship between High Performance Work System and the level of human capital, the influence of High Performance Work System on the level of human capital and employee attitudes seems logical and a larger sample size might have been needed to uncover such effect. The level of human capital and collective normative contract, in turn, have positive influence on HR-related outcomes so that the labor productivity of the unit increased, performance level for the unit improved, and overall organizational citizenship behaviors enriched.
87
Furthermore, perceived labor productivity acted as mediators of the relationships between the level of human capital and unit performance, and collective normative contract with unit performance. Although perceived labor productivity and turnover rate have been shown to act as mediators on the relationship between High Performance Work System and firm performance by Huselid (1995), this is one of the first studies that illustrate the mediating role of HR-related outcomes on the relationship between the level of human capital and/or collective normative contract, the central variables from the resource-based view of the firm and the behavioral perspective. These results extend the previous research by uncovering the underlying mechanisms that more accurately depict the processes that occur between High Performance Work System and unit performance. However, there are some findings (or non-findings) that indicate that the process is more complicated than those depicted in Figure 8, the proposed model, and the results shown in Table 14, the path analytic result. Perhaps, as Becker and Huselid (1998) proposed, there are additional mediating mechanisms between High Performance Work System and unit performance that are not captured in this dissertation. For example, the level of human capital and collective normative contract did not mediate the relationship between High Performance Work System and any of the HR-related outcomes, despite collective normative contract having relationships with High Performance Work System and perceived labor productivity. The level of human capital, on the other hand, was not related to High Performance Work System in the current dissertation. It may also be the case that some of these HR-related outcomes, although considered as a set in this dissertation, have interrelationships among each
88
other such that one variable may be an antecedent to the other. Due to the crosssectional nature of the dataset, it is not possible to discern these relationships empirically in more detail. However, if the findings from individual-level studies extend to the unit-level of analysis, unit-level task performance and organizational citizenship behaviors may behave as antecedents to unit-turnover rate or perceived labor productivity. For instance, task performance and organizational citizenship behaviors are typically considered antecedents to voluntary turnover of the employees (e.g., Van Scotter, 2000; Williams & Livingstone, 1994). For example, Van Scotter (2000) analyzed the data longitudinally and found task performance and organizational citizenship behaviors to predict turnover decisions. If these relationships hold at the unit-level of analysis, it might be beneficial to further differentiate HR-related outcomes further, although longitudinal data collection at the unit-level of analysis from multiple sources will be a significant endeavor. Related, there may be contingent factors that have impact on these relationships found for the dissertation (cf., Youndt et al., 1996; Wright & Sherman, 1999; Wright & Snell, 1998). For example, the relationship between the level of human capital and perceived labor productivity may be enhanced for firms following differentiation strategy, whereas for firms pursuing cost leadership strategy, the relationship between the level of human capital and perceived labor productivity may be non-significant. The difficulty of finding support for the contingency or fit perspective has been noted before (Wright & Snell, 1998; see also Delery & Doty, 1996) and part of the reasons for not finding this contingency or the moderating effect of strategy may stem from the fact that the relationships associated with the mediating variables were not considered, i.e.,
89
strategy does not moderate the relationship between High Performance Work System and labor productivity per se but it may moderate the relationship between the level of human capital and labor productivity. This possibility also highlights the importance of a mediation approach to strategic HRM. Therefore, future research should extend the current findings by collecting additional variables that act as mediators and/or moderators to explicate the relationships between High Performance Work System and (unit) performance further. Another non-significant finding relates to the proposed interaction between the level of human capital and collective normative contract (Hypothesis 9), which received limited support (one marginally significant interaction term out of three possible). It appears to be the case that the levels of human capital and collective normative contract have additive rather than an interactive effect on the HR-related outcomes. Although this interaction effect between ability and motivation is relatively more established at the individual-level of analysis in the individual task performance literature (e.g., Austin et al., 1991), perhaps, collective normative contract does not capture the "action" aspects of motivation (Kuhl, 1985). Kuhl distinguished between two motivational aspects: choice and action. "Choice motivation determines an individual's decision to engage in a specific behavior, whereas action, or control motivation, determines the individual's maintenance of effort and persistence" (Kirk & Brown, 2003: 41). Kanfer (1992) also noted a similar distinction between distal and proximal motivation processes where a notion of proximity to action is central to her framework. Proximal constructs, such as self-efficacy, have been seen to affect processes close to actual behavior such as translating intentions into behavior (Kirk & Brown, 2003), which is
90
similar to the notion of action motivation. On the other hand, distal constructs, such as need for achievement, have been argued to influence processes distant from actual behavior, such as intentions to behave (Kirk & Brown, 2003), which is similar to the notion of choice motivation. Therefore, team efficacy (referent-shift consensus model of self-efficacy: cf., Chan, 1998), defined as team members' aggregate belief and confidence in team's ability to mobilize resources for successful task completion (Chan, 1998), may be more proximal than collective normative contract. It might be the case that collective normative contract behaves in a more distal way such that team efficacy, for example, acts as more proximal construct that mediates the relationship between collective normative contract and HR-related outcomes. In addition, team efficacy may interact with the level of human capital to enhance the impact of ability on HR-related outcomes such that units with higher human capital level will obtain higher labor productivity or unit-level task performance level when team efficacy is also high. It may also the case that non-significant findings regarding the interaction effect of the level of human capital and collective normative contract is methodological, rather than substantive. As the difficulty of finding interaction effects with moderated regression analyses has been well recognized due, for example, to unreliability of measures and small sample size (e.g., Jaccard, Turrisi, & Wan, 1990; Jaccard & Wan, 1995). As noted in Aiken and West (1991), Busemeyer and Jones (1983) provided a formula to calculate the reliability of the cross-product terms. According to the formula, the reliability of the interaction terms used for the level of human capital and collective normative contract is .80. Hence, it is not the unreliability of the interaction terms that contributed to the lack of findings but it might have due to the relatively small sample
91
size obtained for the dissertation, although the sample size appears to be large enough for main effects with medium to strong effect sizes. Assuming that the effect size of the interaction is relatively small (.1 to .3), as typically appears to be the case, I had .14 to .76 power to detect significance at .05 significance level for a sample size of 76 (cf., Cohen & Cohen, 1983). The correlation between the interaction term and the HR-related outcomes ranged from .10 to .12. Therefore, for the interaction effects, it appears likely that I did not have enough power, given the sample size of 76 units collected for the dissertation. Therefore, the present results need to be tested and replicated in future studies that utilize larger sample sizes for the unit. Limitations The results of the dissertation should be interpreted in light of its limitations. I focus on permanent employees rather than on the core employees for the dissertation due to this permanent (vs. temporary) distinction being more salient and accepted in the Japanese business world. In addition, it is still consistent with some of the prior research conducted in the US (e.g., Jackson et al., 1989) that used this distinction. Thus, the results of the dissertation should be generalizable to other non-Japanese studies that utilize permanent employees as their sample. However, the findings could have been stronger if only core employees were utilized for aggregation. Henceforth, these findings may be considered more conservative estimates of the relationships investigated in the current dissertation. Second, the data collection procedure that was necessitated by the use of Japanese sample where it was extremely difficult to obtain employee attitudinal data,
92
the number of surveys distributed to the employees were dictated by the managers heading up the unit. It may be the case that the managers only chose those employees who would be most favorably predisposed to answer the questions in a positive manner. However, the mean of the responses obtained from the employees were 4.5 for collective normative contract (corresponding to neither agree nor disagree to somewhat agree) with a standard deviation of .70. The responses also had a range of 3.84 (minimum = 2.56 and maximum = 6.40) with no skewness in the distribution (skewness statistic = -.25). Therefore, it does not appear to be the case that employees who were selected responded to the questions in a positively biased manner. Therefore, when considering the responses as a whole, it appears more likely to be representative of the Japanese employees working in these units. However, future research should replicate the findings and assess if the responses for this dissertation are biased. Third, the results of the dissertation might have been even stronger if there was more variability among the units. Although ICC(1) values found in the dissertation (.16 to .61) exceeded the median values found in the organizational literature of .12 by James (1982), many of the ICC(2) values just approached the recommended value of .60 by Glick (1985 as cited in Schneider et al., 1998). Recall that ICC(1) compares the variance between units of analysis (establishments in this case) to the variance within units of analysis using the individual ratings of each respondent (Schneider, White, & Paul, 1998) while ICC(2) compares the between unit variability to that of within unit variability, using the average ratings of respondents within each unit (Bartko, 1976). Therefore, it appears that a range restriction might have been a problem in some cases that constrained the size of the correlation possible, which might have contributed to
93
some of the null findings. Therefore, future studies may design the research to include more units that are not performing well. Moreover, I cannot establish the causal relationship among the variables of interest due to the cross-sectional nature of the dissertation, although the sequences of variables depicted is consistent with Becker and Huselid (1998) and others. It may be the case that higher performing establishments are more able to institute High Performance Work System due perhaps to more slack resources that the establishments possess or more financial support that the establishments receive from the parent companies. However, this concern for reverse causality is mitigated somewhat because I found two sets of mediating effect. However, a longitudinal study that incorporates multi-wave design would be beneficial in explicating the causal relationships fully. Fourth and related, the lower variability of variables across units may be due to the focus on only the permanent employees, which included both core and support employee groups. It may be the case that more standardized High Performance Work System is used for the permanent employees across industries and companies that might contributed to a reduction in the overall variability obtained for the present dissertation, thereby constraining the size of the correlations found. Thus, another improvement for future research may be to collect data from both permanent and temporary employees from multiple units because High Performance Work System are less likely to be used for temporary employees and examine the variability of a system of HR practices as well as other variables of interest and how it affects the results. Finally, there may be some additional, methodological concerns with regard to the dissertation. First, all of the variables are perceptual in nature. Although the
94
difficulty of obtaining objective measure of performance at the unit-level of analysis has been well recognized (e.g., Gupta, 1987), it would have been ideal if an objective measure of unit performance were obtained. However, exploratory factor analysis clearly distinguished all the four outcome variables taken from the managers. Therefore, the relationships found between HR-related outcomes and unit performance does not appear to be affected by percept-percept bias. Related, potential for common method biases may be a concern as many of the variables are obtained from the managers that may contribute to inflating the relationships among variables. However, this concern is minimized by several factors. First, I use ratings from two different sources (managers and employees) and obtained multiple responses so that these responses can be aggregated to the unit level of analysis. Second, the pattern of the significant relationships found does not indicate any significant biases associated with one particular source. More specifically, the ratings for High Performance Work System were obtained from the employees, which were significantly related to both the level of human capital (managerial ratings) and collective normative contract (employee ratings). Similarly, both the level of human capital and collective normative contract were related to HR-related outcomes (managerial ratings) in a differential manner. Finally, mediation analyses reduce the concern further as support was found for both the level of human capital (manager rated) and collective normative contract (employee rated) for the relationship between High Performance Work System (employee rated) and HR-related outcomes (manager rated). Therefore, overall evidence appears to suggest that the common method bias did not have substantial influences on the results for this dissertation.
95
Implications for Research and Practice These limitations aside, the results of this dissertation have important implications primarily for scholars and for managers. First, it is one of the first studies that examined the linkages between High Performance Work System and (unit) performance. Finding these intermediate linkages helps scholars understand the processes further, which might be helpful in discovering additional mediators, moderators, or consequences at the different level of analysis. For example, by adopting a cross-level or multi-level perspective (Klein et al., 1994; Klein & Kozlowski, 2000; Ostroff & Bowen, 2000), the current results can be extended to examining individual consequences of High Performance Work System via the level of human capital and collective normative contract, or firm-level consequences of High Performance Work System through unit performance. For the cross-level, "disaggregation" model where the influences of High Performance Work System at the unit on individual employees are explored, collective normative contract may be one of the critical linkages to increasing normative commitment (Meyer & Allen, 1991), which "refers to a perceived obligation to stay with the organization" (Irving, Coleman, & Cooper, 1997: 444), which is important because previous empirical studies have had difficulty in finding unique antecedents that predict normative commitment (cf., Irving et al., 1997; Ko, Price, & Mueller, 1997; Meyer, Allen, & Smith, 1993). Therefore, cross-level linkages between High Performance Work System, its mediators, and individual-level attitudes and behaviors can be examined in the future that may reveal unexpected or unexplored relationships among these variables.
96
For the cross-level, "aggregation" model where the effects of High Performance Work System at the unit on financial indicators of firm performance are investigated, HR-related outcomes such as unit-level organizational citizenship behaviors, or unit innovative behaviors (not examined) may spill over to other units within the organization to facilitate effective functioning of the firm as a whole. Thus, as Katz and Kahn (1964: 133) noted, "the resources of people in innovation, in spontaneous cooperation, in protective and creative behavior are thus vital to organizational survival and effectiveness." In addition, given that HR-related outcomes focused on aggregate behaviors at the unit-level of analysis, there may be additional variables that act as mediators on the relationships between the level of human capital and collective normative contract, and HR-related outcomes, in addition to the ones already noted above. For example, borrowing from the micro organizational behavior literature, organizational commitment and job satisfaction have been found to relate to turnover, performance, and organizational citizenship behaviors (e.g., Cohen, 1993; Organ & Ryan, 1995; Tett & Meyer, 1993), which may act as additional aggregate variables that mediate these relationships. Thus, there are additional opportunities at the unit level of analysis to examine these processes further. One of the difficulties that previous empirical research has encountered in findings the moderating effects of strategy on the relationship between High Performance Work System and firm performance might be attributed to the existence of these mediating processes. It may be the case that strategy moderates the relationship between High Performance Work System and the level of human capital (for "buy"
97
strategy) or High Performance Work System and normative commitment (for "make" strategy: Miles & Snow, 1984) but not the other. Given the existence of these mediating processes, it is less likely to find significant moderating effect of strategy on the relationship between High Performance Work System to (unit) performance, as there was no direct relationship between these two when mediating variables are taken into account (see Figure 12). Therefore, the influence of organizational characteristics such as strategy may be better understood when these mediating processes are considered. For managers, the results of the dissertation may be helpful in facilitating organizational interventions to increase the benefit of High Performance Work System. For instance, if the organizational decision makers have not felt that the firm is reaping the benefit of the High Performance Work System to the maximum extent possible, the source of the problem may be diagnosed by examining the level of human capital and/or collective normative contract for the specific units in the organization. If it is the case that the unit does not have sufficient number of employees with high knowledge, skills, and abilities, components of High Performance Work System such as selection and recruitment or training and development may be revised to enhance the level of human capital for the units. On the other hand, if it is the employee attitudes that were identified as a source of problem, some components of High Performance Work System such as performance appraisal and compensation may be modified to improve employees' attitudes and consequent behaviors. In addition to devising the components of High Performance Work System, the results from the dissertation highlights the importance of both the level of human capital and collective normative contract on labor productivity. Therefore, managers need to be
98
aware of these linkages, at minimum, and be ready to provide support to strengthen these linkages. For example, given that the level of human capital is positively related to all three HR-related outcomes, managers may be able to utilize a combination of management by objectives and performance incentives to underscore the linkages between high performance and reward to entice employees with higher knowledge, skills, and abilities to work harder and smarter so that their contribution to the unit will be higher. In a similar vein, employees' collective normative contract may be enhanced if the organization can show that they value and respect employee contributions and that they care (i.e., perceived organizational support: Eisenberger et al., 1986; Eisenberger et al., 1990). Once the employees' perceive that the organization has provided them something that they value, they are more likely to reciprocate by exerting more effort on the task as well as providing more extra-role behaviors that are beneficial to the organization as a whole. Finally, given the strong relationship between labor productivity and overall unit performance, the highest return for the managers may reside in retaining highly skilled employees who also provide suggestions and helping to others to improve the efficiency and effectiveness of unit function. These employees are central in maintaining and improving overall unit performance. Thus, the benefits that accrue from retaining one such employee are likely to be substantially higher than retaining several mediocre employees. Conclusion In conclusion, the results of the dissertation contributes to the literature on strategic human resources management by opening up the black box to explicate the
99
mediating processes associated with High Performance Work System and performance with important implications for managers and scholars.
100
Appendix A. Japanese Versions of the Surveys
101
102
103
104
105
106
107
Appendix B. Industrial Affiliations of the Companies for Primary (One-Digit) SIC Code Agriculture, Forestry, & Fishing NONE Minin g Sumitomo Kinzoku Kouzan Construction (group coding 1) Kurogane Kousakujo Maeda Kensetsu Nihon Seitai Taihei Seisakujo Manufacturing (group coding 2) Ajinomoto Takara Corporation Aoinetsu Kougyou Asahi Kasei Amidasu Canon Daiku Denki Seisakujo Eiwa Kasei Kougyou Fuji Denki Fuji Kasei Gaikin Plant Honshuu Kagaku Ishikawajima Harima Jyuukougyou Itogumi Mokuzai Kanematsu Kao Koiwai Nyuugou Kyokuyou Suisan Kyouwa Hakkou - Iyakuhin Mazda Mitsubishi Gas Kagaku Gouseijushi Mitsubishi Gashin Kagaku Mitsubishi Jushi Mitsui Kagaku - Haisii Mitsui Kagaku Sanshi
108
Mitsui Kinzoku Mitsui Kinzoku Kougyou - Atsuen Kakou Jigyoubu Mitsui Kinzoku Kougyou - Ceramic Jigyousho Mitsui Kinzoku Kougyou - Douhaku Jigyou Honbu Mitsui Kinzoku Kougyou - Hachinohe Seirenjo Mitsui Kinzoku Kougyou - Hikoshima Seirenjo Mitsui Kinzoku Kougyou - Himaku Zairyou Jigyoubu Mitsui Kinzoku Kougyou - Kamioka Kougyoujo Mitsui Kinzoku Kougyou - MCS Mitsui Kinzoku Kougyou - Mitsuike Jimusho Mitsui Kinzoku Kougyou - Nirezaki Jigyousho Mitsui Kinzoku Kougyou - Paato Raito Jigyoubu Mitsui Kinzoku Kougyou - Sougou Kenkyuujo Mitsui Kinzoku Kougyou - Takehara Seitetsujo Mizakura Kyougyou Nihon Denchi Nihon Piranii Kougyou Nikkei Kinzoku Kogaisha Okinawa Dennou Okinawa Shokuryo Okinawa Yakult Riken Seiko Epson Shouwa Yakuhin Kakou Sony ST Kagaku Tagagi Kougyou Tama Chuuou Sangyou Tonbo Enpitsu Touei Foods Yaezan Okishoku Yamato Kagaku Kougyou Youkouzoku Kagakuhin Co. Yuuki Gousei Yakuhin Kougyou Transportation, Communication, Electric, Gas, & Sanitary Services (group coding 3)
109
Hokkaido Kuukou Hokusou Sekiyu Idemitsu Kousan JPower Systems Kokusai Cleaning Kousan Kaiyou Mitubishi Souko Nagase Sekiyu Nagoya Terebi Jigyou Nihon Contena Terminal Teikoku Tsuushin Kougyou Terebi Asahi Wholesale Trade (group coding 4) Canon Hanbai GE Yokokawa Medical Hikarigaoka Kousan Itohira Kousan Jetronix Kyouwa Medix Maruichi Sanshou Marunin Shouji Mitsubishi Shouji Miyako Okishoku Nakamura Bussan Nihon Chemical Shouji Nihon Kousan Okishoku Shouji Sekkou Shouji Tokuho Shouji Retail Trade (group coding 5) DIX Hitachi Taga Technology Hogi Medical KMS Korudia Nihon Acuroid
110
SE Labo SG Shinsei Shizuoka Gas Shuei Trading Takeo - Fukuoka Branch Touyou Kougei Finance, Insurance, & Real Estate (group coding 6) C.P.U. Global Insurance Juutaku Anshin Hoshou Mito Shouken Oioi Songai Hoken Ryuugin Hoshou Ryuukyuu D.C. Ryuukyuu Ginkou Ryuukyuu Lease Sanin Goudou Ginko Shinkou Shouken - Ootsu Branch Touden Kougyou Insurance Center Tougin Lease Services (group coding 7) A. Human Net Amurasu PTY Ltd. Aoi Promotion Asahi Beer ASM - Japan Big Ability Central Jyouho Center Consultanto Daiichi Lease (Service - Leasing) Diya-Tec E-Venture Support Forum Fuji Xerox Sougou Kyouiku Kenkyuujo Izumigou Appointment Center
111
Izumigou Plaza Hotel Toba Izumigou Yatsugatake Branch Izumigou Yokohama Branch Jinzai Haken Center Okinawa Kokubu Kokusai Jinji Kenkyuujo Leading Edge Inc. Mercer HR Consulting Moon Beach Resort Hotel Nihon Bousei Kougyou Nihon Gijutsu Iten Nihon Kyuushoku Service Nikkan Kougyou Koukokusha Nikkan Kougyou Service Center Nishii Associates Nomura Sougou Kenkyuujo Point Ryuukyu Biru Management Touwa Food Service Trans-Tec Ueno Trans-Tec US Visvalley Yatsugatake Izumigou Honsha YK Service Public Administration Shakai Hoken Roumu Jimusho Songai Hokenryouritsu Sanshutsu Kikou
112
Appendix C High Performance Work System: "To what extent do you agree with the following statements? When answering this section, please think about those HR practices used for the average employees in the establishment!" Job Design 1. jobs include a wide variety of tasks. 2. employees are involved in job rotation.* 3. employees are empowered to make decisions.* 4. employees have a high degree of job security. 5. employee are encourage to take up challenging assignments.* 6. employees' request for job assignments is taken into account in job rotations decisions.* Selection 7. selection is comprehensive (uses interviews, tests, etc.).* 8. selection involves screening many job candidates.* 9. selection focuses on selecting the best all around candidate, regardless of the specific job.* 10. selection emphasizes promotion from within.* 11. selection places priority on their potential to learn (e.g., aptitude).* 12. selection emphasizes their capacity to perform well right away. Training & Development 13. training is continuous.* 14. training uses outside instruction (seminars, conferences, etc.). 15. training programs are comprehensive.* 16. training programs strive to develop firm-specific skills/knowledge.* 17. the training programs emphasize on-the-job experiences.* Performance Management 18. performance is based on objective, quantifiable results.* 19. performance appraisals include management by objective with mutual goal setting.* 20. performance appraisal include developmental feedback.* 21. performance appraisal emphasize development of abilities/skills.* 22. performance appraisal emphasizes the output of the team. 23. performance appraisal is geared toward those who put in longer hours. Compensation & Incentives 24. compensation packages include an extensive benefits package.* 25. our compensations include high wages.* 26. the incentive system is tied to skill-based pays.* 27. the incentive system has a group-based component (gainsharing, etc.). 28. our incentive system values seniority. 29. our incentives include annual increases in salary. 30. are contingent on performance.*
113
Human Capital: "Our employees working in the unit?" 1. are highly skilled. 2. are widely considered the best in our industry. 3. are creative and bright. 4. are experts in their particular jobs and functions. Collective Normative Contact 1. The things employees do on the job today will benefit their standing in the long run. 2. Our relationship with the establishment continues to evolve and develop 3. We have significant opportunities to take on assignments that enhance our value 4. We don't mind working hard today—We know we will eventually be rewarded by our establishment 5. We try to look out for the best interest of the establishment because we can rely on our establishment to take care of us. HR-Related Outcomes Perceived Labor Productivity 1. The employees' productivity is higher than those of my major competitors.* 2. The employees in this establishment produce outputs in an efficient manner.* 3. The employees' productive power is higher than those of the competitors.* Unit-Level Task Performance: "The employees?" 1. adequately complete assigned duties.* 2. fulfill responsibilities specified in job description. 3. perform tasks that are expected of them.* 4. engage in activities that will directly affect their performance evaluation.* 5. neglect aspects of the job they are obligated to perform (reverse-coded).* 6. fail to perform essential duties (reverse-coded).* Unit-Level Organizational Citizenship Behaviors: "The employees in our unit?" 1. help others who have heavy work loads.* 2. help others who have been absent.* 3. assist supervisors with their work (when not asked).* 4. take time to listen to co-workers' problems and worries.* 5. go out of way to help new employees.* 6. take personal interest in other employees.* 7. pass along information to co-workers. 8. have attendance at work that is above the norm.* 9. give advance notice when unable to come to work.* 10. adhere to informal rules devised to maintain order.* 11. conserve and protect organizational property.* Subjective Unit Performance: "How would you compare the establishment's [???.] over the past 3 years to that of the other establishments that do the same kind of work?" 1. performance 2. quality of products, services, or program* 3. development of new products, services, or programs* 4. ability to attract essential employees*
114
5. 6. 7. 8.
ability to retain essential employees* satisfaction of customers or clients* relations between management and other employees relations among employees in general
115
REFERENCES Adams, J. S. 1965. Inequity in social exchange. In L. Berkowitz (Ed.), Advances in experimental social psychology, 2: 267-299. New York: Academic Press. Aiken, L. S., & West, S. G. 1991. Multiple regression: Testing and interpreting interactions. Newbury Park, CA: Sage. Allen, N. J., & Meyer, J. P. 1990. Organizational socialization tactics: A longitudinal analysis of links to newcomers' commitment and role orientation. Academy of Management Journal, 33: 847-858. Angle, H. L., & Perry, J. L. 1981. An empirical assessment of organizational commitment and organizational effectiveness. Administrative Science Quarterly, 26: 113. Applebaum, E., Bailey, T., Berg, P., & Kalleberg, A. L. 2000. Manufacturing advantage: Why high-performance work systems pay off. Ithaca, New York: Cornell University Press. Armeli, S., Eisenberger, R., Fasolo, P., & Lynch, P. 1998. Perceived organizational support and police performance: The moderating influence of socioemotional needs. Journal of Applied Psychology, 83: 288-297. Arthur, J. B. 1994. Effects of human resource systems on manufacturing performance and turnover. Academy of Management Journal, 37: 670-687. Arthur, J. B. 1992. The link between business strategy and industrial relations systems in American steel minimills. Industrial and Labor Relations Review, 45: 488506.
116
Austin, J. T., Villanova, P., Kane, J. S., & Bernardin, H. J. 1991. Construct validation of performance measures: Definitional issues, development, and evaluation of indicators. In G. R. Ferris (Ed.), Research in Personnel and Human Resources Management, 9: 159-233. Greenwich, CT: JAI Press. Baird, L., & Meshoulam, I. 1988. Managing two fits of strategic human resource management. Academy of Management Review, 13: 116-128. Balkin, D. B., & Gomez-Mejia, L. R. 1984. Determinants of R and D compensation strategies in the high tech industry. Personnel Psychology, 37: 635-650. Barnett, W. P., & Miner, A. S. 1992. Standing on the shoulders of others: Career interdependence in job mobility. Administrative Science Quarterly, 37: 262-281. Barney, J. 1991. Firm resources and sustained competitive advantage. Journal of Management, 17: 99-120. Baron, R. M. & Kenny, D. A. 1986. The moderator-mediator distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51: 1173-1182. Bartko, J. J. 1976. On various intraclass correlation reliability coefficients. Psychological Bulletin, 83, 762-765. Bass, B. 1990. Bass and Stogdill's handbook of leadership: Theory, research, and managerial applications (3rd ed.). New York: Free Press. Becker, B., & Gerhart, B. 1996. The impact of human resource management on organizational performance: Progress and prospects. Academy of Management Journal, 39: 779-801.
117
Becker, B. E., & Huselid, M. A. 1998. High Performance Work System and firm performance: A synthesis of research and managerial implications. In G. R. Ferris (Ed.), Research in personnel and human resources management, 16: 53-101. Greenwich, CT: JAI Press. Becker, G. S. 1964. Human capital. New York: Columbia University Press. Bentler, P. M. 1998. EQS for Windows 5.7b. Encino, CA: Multivariate Software, Inc. Bentler, P. M., & Chou, C. H. 1987. Practical issues in structural modeling. Sociological Methods & Research, 16: 78-117. Berman, S. L., Down, J., & Hill, C. W. L. 2002. Tacit knowledge as a source of competitive advantage in the National Basketball Association. Academy of Management Journal, 45: 13-31. Bettencourt, L. A., Gwinner, K. P., & Meuter, M. L. 2001. A comparison of attitude, personality, and knowledge predictors of service-oriented organizational citizenship behaviors. Journal of Applied Psychology, 86: 29-41. Bishop, J. W., Scott, K. D., & Burroughs, S. M. 2000. Support, commitment, and employee outcomes in a team environment. Journal of Management, 26: 11131132. Blau, P. 1964. Exchange and power in social life. New York: Wiley. Borman, W. C., & Motowidlo, S. J. 1997. Task performance and contextual performance: The meaning for personnel selection research. Human Performance, 10: 99-109.
118
Borman, W. C., & Motowidlo, S. J. 1993. Expanding the criterion domain to include elements of contextual performance. In N. Schmitt & W. C. Borman (Eds.), Personnel selection in organizations: 71-98. San Francisco: Jossey-Bass. Borman, W. C., White, L. A., & Dorsey, D. W. 1995. Effects of ratee task performance and interpersonal factors on supervisor and peer performance ratings. Journal of Applied Psychology, 80: 168-177. Brief, A. P., & Motowidlo, S. J. 1986. Prosocial organizational behaviors. Academy of Management Review, 11: 710-725. Brislin, R. W. 1981. Cross-cultural encounters: Face-to-face interaction. New York: Pergamon Brockner, J. 2002. Making sense of procedural fairness: How high procedural fairness can reduce or heighten the influence of outcome favorability. Academy of Management Review, 27: 58-76. Brockner, J., Tyler, T. R., & Cooper-Schneider, R. 1992. The influence of prior commitment on reactions to perceived unfairness: The higher they are, the harder they fall. Administrative Science Quarterly, 37: 241-261. Burton, M. D., & O'Reilly, C. A., III. 2000. The impact of high commitment work systems on IPOs: Additive or multiplicative effects? Paper presented at the 60th annual meeting of the Academy of Management, Toronto, Canada. Busemeyer, J. R., & Jones, L. E. 1983. Analyses of multiplicative combination rules when the causal variables are measured with error. Psychological Bulletin, 93: 549-562.
119
Carpenter, M. A., Sanders, W. G., & Gregersen, H. B. 2001. Bundling human capital with organizational context: The impact of international assignment experience on multinational firm performance and CEO pay. Academy of Management Journal, 44: 493-511. Cattell, R.B. 1966. The scree test for the number of factors. Multivariate Behavioral Research, 1: 245-276. Chadwick, C. D. 2000. Empirical insights on the origins of synergies in strategic human resource management. Paper presented at the 60th annual meeting of the Academy of Management, Toronto, Canada. Chadwick, C., & Cappelli, P. 1999. Alternatives to generic strategy typologies in strategic human resource management. In P. M. Wright, L. Dyer, J. Boudreau, & G. Milkovich (Eds.), Research in personnel and human resources management, Supplement 4: 1-29. Greenwich, CT: JAI Press. Chan, D. 1998. Functional relations among constructs in the same content domain at different levels of analysis: A typology of composition models. Journal of Applied Psychology, 83: 234-246. Chen, X.-P., Hui, C., & Sego, D. J. 1998. The role of organizational citizenship behavior in turnover: Conceptualization and preliminary tests of key hypotheses. Journal of Applied Psychology, 83: 922-931. Colquitt, J. A. 2001. On the dimensionality of organizational justice: A construct validation of a measure. Journal of Applied Psychology, 86: 386-400.
120
Colquitt, J. A., Conlon, D. E., Wesson, M. J., Porter, C. O. L. H., & Ng, K. Y. 2001. Justice at the millennium: A meta-analytic review of 25 years of organizational justice research. Journal of Applied Psychology, 86: 425-445. Colquitt, J. A., Noe, R. A., & Jackson, C. L. 2002. Justice in teams: Antecedents and consequences of procedural justice climate. Personnel Psychology, 55: 83-109. Cohen, A. 1993. Organizational commitment and turnover: A meta-analysis. Academy of Management Journal, 36: 1140-1157. Cohen, J., & Cohen, P. 1983. Applied multiple regression/correlation analysis for the behavioral science (2nd ed.). Hillsdale, New Jersey: Lawrence Erlbaum Associates, Publishers. Coff, R. W. 1997. Human assets and management dilemmas: Coping with hazards on the road to resource-based theory. Academy of Management Review, 23: 374-402. Cropanzano, R. 2001. Justice in the workplace: From theory to practice, Vol. 2. Mahwah, NJ: Lawrence Erlbaum. Cropanzano, R. 1993. Justice in the workplace: Approaching fairness in human resource management. Hillsdale, NJ: Lawrence Erlbaum. Cropanzano, R., & Greenberg, J. 1997. Progress in organizational justice: Tunneling through the maze. In C. L. Cooper & I. T. Robertson (Eds.), International Review of Industrial and Organizational Psychology, 12: 317-372. New York: Wiley. Cropanzano, R., & Kacmar, M. 1995. Organizational politics, justice, and support: Managing the social climate of the workplace. Westport, CT: Quoram Books.
121
Davis-Blake, A., & Uzzi, B. 1993. Determinants of employment externalization: A study of temporary workers and independent contractors. Administrative Science Quarterly, 38: 195-223. Davis, P. S., Robinson, R. B., Jr., Pearce, J. A., II, & Park, S. H. 1992. Business unit relatedness and performance: A look at the pulp and paper industry. Strategic Management Journal, 5: 349-361. Dawis, R. V. 2000. P-E fit as paradigm: Comments on Tinsley (2000). Journal of Vocational Behavior, 56: 180-183. Delaney, J. T., & Huselid, M. A. 1996. The impact of human resource management practices on perceptions of organizational performance. Academy of Management Journal, 39: 949-969. Delery, J. E. 1998. Issues of fit in strategic human resource management: Implications for research. Human Resource Management Review, 8: 289-309. Delery, J. E., & Doty, D. H. 1996. Modes of theorizing in strategic human resource management: Tests of universalistic, contingency, and configurational performance predictions. Academy of Management Journal, 39: 802-835. Dollinger, M. J., & Golden, P. A. 1992. Interorganizational and collective strategies in small firms: Environmental effects and performance. Journal of Management, 18: 695-715. Doty, D. H., Gulick, W. H., & Huber, G. P. 1993. Fit, equifinality, and organizational effectiveness: A test of two configurational theories. Academy of Management Journal, 36: 1196-1250.
122
Drexler, J. A. 1977. Organizational climate: Its homogeneity within organizations. Journal of Applied Psychology, 62: 38-42. Dunnette, M. D., Lawler, E. E., & Weick, K. E. 1967. The role of financial compensation in managerial motivation. Organizational Behavior & Human Decision Processes, 2: 175-216. Dyer, L., & Reeves, T. 1995, May. Human resource strategies and firm performance: What do we know and where do we need to go? Paper presented at the 10th World Congress of the International Industrial Relations Association, Washington, D.C. Edwards, J. R., & Van Harrison, R. 1993. Job demands and worker health: Three-dimensional reexamination of the relationship between person-environment fit and strain. Journal of Applied Psychology, 78: 628-648. Eisenberger, R., Armeli, S., Rexwinkel, B., Lynch, P. D., & Rhoades, L. 2001. Reciprocation of perceived organizational support. Journal of Applied Psychology, 86: 42-51. Eisenberger, R., Cummings, J., Armeli, S., & Lynch, P. 1997. Perceived organizational support, discretionary treatment, and job satisfaction. Journal of Applied Psychology, 82: 812-820. Eisenberger, R., Fasolo, P., & Davis-LaMastro, V. 1990. Perceived organizational support and employee diligence, commitment, and innovation. Journal of Applied Psychology, 75: 51-59. Eisenberger, R., Huntington, R., Hutchison, S., & Sowa, D. 1986. Perceived organizational support. Journal of Applied Psychology, 71: 500-507.
123
Folger, R., & Cropanzano, R. 1998. Organizational justice and human resource management. Thousand Oaks, CA: Sage. Gattiker, U. E. 1995. Firm and taxpayer returns from training of semiskilled employees. Academy of Management Journal, 38: 1152-1173. George, J. 1991. State or trait: Effects of positive mood on prosocial behavior at work. Journal of Applied Psychology, 76: 299-307. George, J. M. 1990. Personality, affect, and behavior in groups. Journal of Applied Psychology, 75: 107-116. George, J. M., & Bettenhausen, K. 1990. Understanding prosocial behavior, sales performance, and turnover: A group-level analysis in a service context. Journal of Applied Psychology, 75: 698-709. George, J. M., & Brief, A. P. 1992. Feeling good - doing good: A conceptual analysis of the mood at work - organizational spontaneity relationship. Psychological Bulletin, 112: 310-329. George, J. M., & Jones, G. R. 1997. Organizational spontaneity in context. Human Performance, 10: 153-170. Gerhart, B., & Milkovich, G. T. 1990. Organizational differences in managerial compensation and financial performance. Academy of Management Journal, 33: 663691. Gerstner, C. R., & Day, D. V. 1997. Meta-analytic review of leader-member exchange theory: Correlates and construct issues. Journal of Applied Psychology, 82: 827-844.
124
Gilliland, S., Steiner, D., & Skarlicki, D. 2001. Theoretical and cultural perspectives on organizational justice. Greenwich, CT: Information Age Publishing. Glick, W. H. 1985. Conceptualizing and measuring organizational and psychological climate: Pitfalls in multilevel research. Academy of Management Review, 10: 601-616. Graen, G. 1976. Role-making processes within complex organizations. In M. D. Dunnette (Ed.), Handbook of industrial and organizational psychology: 1201-1245. Chicago: Rand McNally. Greenberg, J. 1993. Stealing in the name of justice: Informational and interpersonal moderators of theft reactions to underpayment inequity. Organizational Behavior and Human Decision Processes, 54: 81-103. Greenberg, J. 1990. Organizational justice: Yesterday, today, and tomorrow. Journal of Management, 16: 399-432. Greenberg, J. 1987. Reactions to procedural justice in payment distributions: Do the means justify the ends? Journal of Applied Psychology, 72: 55-61. Greenberg, J., & Cropanzano, R. 2001. Advances in organizational justice. Stanford, CA: Stanford Business Books. Griffeth, R. W., Hom, P. W., & Gaertner, S. 2000. A meta-analysis of antecedents and correlates of employee turnover: Update, moderator tests, and research implications for the next millennium. Journal of Management, 26: 463-488. Gouldner, A. W. 1960. The norm of reciprocity. American Sociological Review, 25: 161-178.
125
Guillen, M. F. 2000. Business groups in emerging economies: A resource-based view. Academy of Management Journal, 43: 362-380. Guion, R. M. 1965. Personnel testing. New York: McGraw-Hill. Gupta, A. K. 1987. SBU strategies, corporate-SBU relations, and SBU effectiveness in strategy implementation. Academy of Management Journal, 30: 477500. Gupta, A. K., & Govindarajan, V. 1986. Resource sharing among SBUs: Strategic antecedents and administrative implications. Academy of Management Journal, 29: 695-714. Guthrie, J. P. 2001. High-involvement work practices, turnover, and productivity: Evidence from New Zealand. Academy of Management Journal, 44: 180192. Guthrie, J. P., & Olian, J. D. 1991. Does context affect staffing decisions? The case of general managers. Personnel Psychology, 44: 263-292. Guttman, L. 1954. Some necessary conditions for common factor analysis. Psychometrika, 19: 149-161. Guzzo, R. A., Noonan, K. A., & Elron, E. 1994. Expatriate managers and the psychological contract. Journal of Applied Psychology, 79: 617-626. Guzzo, R. A., & Shea, G. P. 1992. Group performance and intergroup relations in organizations. In M. D. Dunnette & L. M. Hough (Eds.), Handbook of industrial and organizational psychology (2nd Ed.), 3: 1-45. Palo Alto, CA: Consulting Psychologists Press.
126
Hackman, J. R., & Oldham, G. R. 1980. Work redesign. Reading, MA: AddisonWesley. Harrison, D. A., & Martocchio, J. J. 1998. Time for absenteeism: A 20-year review of origins, offshoots, and outcomes. Journal of Management, 24: 305-350. Hitt, M. A., & Barr, S. H. 1989. Managerial selection decision models: Examination of configural cue processing. Journal of Applied Psychology, 74: 53-61. Hitt, M. A., Bierman, L., Shimizu, K., & Kochhar, R. 2001. Direct and moderating effects of human capital on strategy and performance in professional service firms: A resource-based perspective. Academy of Management Journal, 44: 13-28. Hom, P. W., Caranikas-Walker, F., Prussia, G. E., & Griffeth, R. W. 1992. A meta-analytical structural equations analysis of a model of employee turnover. Journal of Applied Psychology, 77: 890-909. Homans, G. C. 1958. Social behaviors as exchange. American Journal of Sociology, 63: 597-606. Homans, G. C. 1961. Social behavior. New York: Harcourt, Brace, and World. House, R. Rousseau, D. M., & Thomas-Hunt, M. 1995. The meso paradigm: A framework for the integration of micro and macro organizational behavior. In L. L. Cummings & B. M. Staw (Eds.), Research in organizational behavior, 17: 71-114. Greenwich, CT: JAI Press. Hui, C., Law, K. S., & Chen, Z. X. 1999. A structural equation model of the effects of negative affectivity, leader-member exchange, and perceived job mobility on in-role and extra-role performance: A Chinese case. Organizational Behavior & Human Decision Processes, 77: 3-21.
127
Hulin, C. 1991. Adaptation, persistence, and commitment in organizations. In M. D. Dunnette & L. M. Hough (Eds.), Handbook of Industrial and Organizational Psychology (2nd ed.), Vol. 2: 445-505. Palo Alto, CA: Consulting Psychological Press. Huselid, M. A. 1995. The impact of human resource management practices on turnover, productivity, and corporate financial performance. Academy of Management Journal, 38: 635-672. Iaffaldano, M. T., & Muchinsky, P. M. 1986. Job satisfaction and job performance: A meta-analysis. Psychological Bulletin, 97: 251-273. Ichniowski, C., Shaw, K., & Prennushi, G. 1997. The effects of human resource management practices on productivity: A study of steel finishing lines. The American Economic Review, 87: 291-313. Igalens, J., & Roussel, P. 1999. A study of the relationships between compensation package, work motivation and job satisfaction. Journal of Organizational Behavior, 20: 1003-1025. Irving, P. G., Coleman, D. F., & Cooper, C. L. 1997. Further assessments of a three-component model of occupational commitment: Generalizability and differences across occupations. Journal of Applied Psychology, 82: 444-452. Jaccard, J., Turrisi, R., & Wan, C. K. 1990. Interaction effects in multiple regression analysis. Newbury Park, CA: Sage. Jaccard, J., & Wan, C. K. 1995. Measurement error in the analysis of interaction effects between continuous predictors using multiple regression: Multiple indicator and structural equation approaches. Psychological Bulletin, 117: 348-357.
128
Jackson, S. E., Schuler, R. S., & Rivero, J. C. 1989. Organizational characteristics as predictors of personnel practices. Personnel Psychology, 42: 727-786. James, L. R. 1982. Aggregation bias in estimates of perceptual agreement. Journal of Applied Psychology, 67: 219-229. James, L. R., Demaree, R. G., & Wolf, G. 1993. Rwg: An assessment of withingroup agreement. Journal of Applied Psychology, 78: 306-309. James, L. R., Demaree, R. G., & Wolf, G. 1984. Estimating within-group interrater reliability with and without response bias. Journal of Applied Psychology, 69: 85-98. James, L. R., Joyce, W. F., & Slocum, J. W., Jr. 1988. Comment: Organizations do not cognize. Academy of Management Review, 13: 129-132. Johns, G., & Xie, J. L. 1998. Perceptions of absence from work: People's Republic of China versus Canada. Journal of Applied Psychology, 83: 515-530. Jöreskog, K. G., & Sörbom, D. 1989. LISREL VII: User's reference guide. Mooresville, IN: Scientific Software. Joyce, W. F., & Slocum, J. W., Jr. 1990. Strategic context and organizational climate. In B. Schneider (Ed.), Organizational climate and culture: 130-150. San Francisco: Jossey-Bass. Kaiser, H.F. 1970. Second generation Little Jiffy. Psychometrika, 35: 401-417. Kanfer, R. 1992. Work motivation: New directions in theory and research. In C. L. Cooper & I. T. Robertson (Eds.), International review of industrial and organizational psychology, 7: 1-53. Chichester, England: Wiley.
129
Kanfer, R., & Ackerman, P. L. 1989. Motivation and cognitive abilities: An integrative/aptitude-treatment interaction approach. Journal of Applied Psychology, 74: 657-690. Kanfer, R., Crosby, J. V., & Brandt, D. M. 1988. Investigating behavioral antecedents of turnover at three job tenure levels. Journal of Applied Psychology, 73: 331-335. Katz, D., & Kahn, R. L. 1964. The motivational basis of organizational behavior. Behavioral Science, 9: 131-146. Kirk, A. K, & Brown, D. F. 2003. Latent constructs of proximal and distal motivation predicting performance under maximum test conditions. Journal of Applied Psychology, 88: 40-49. Klein, K. J., Bliese, P. D., Kozlowski, S. W. J., Dansereau, F. Gavin, M. B., Griffin, M. A., Hofmann, D. A., James, L. R., Yammarino, F. J., & Bligh, M. C. 2000. Multilevel analytical techniques: Commonalities, differences, and continuing questions. In K. J. Klein & S. W. J. Kozlowski (Eds.), Multilevel theory, research, and methods in organizations: Foundations, extensions, and new directions: 512-553. San Francisco: Jossey-Bass. Klein, K. J., Dansereau, F., & Hall, R. J. 1994. Levels issues in theory development, data collection, and analysis. Academy of Management Review, 19: 195229. Klein, K. J., & Kozlowski, S. W. J. 2000. From micro to meso: Critical steps in conceptualizing and conducting multilevel research. Organizational Research Method, 3: 211-236.
130
Kleinbeck, U., & Fuhrmann, H. 2000. Effects of a psychologically based management system on work motivation and productivity. Applied Psychology: An International Review, 49: 596-610. Ko, J.-K., Price, J. L., & Mueller, C. W. 1997. Assessment of Meyer and Allen's three-component model of organizational commitment in South Korea. Journal of Applied Psychology, 82: 961-973. Koch, M. J., & McGrath, R. G. 1996. Improving labor productivity: Human resource management policies do matter. Strategic Management Journal, 17: 335-354. Konovsky, M. A., & Pugh, S. D. 1994. Citizenship behavior and social exchange. Academy of Management Journal, 37: 656-669. Koys, D. J. 2001. The effects of employee satisfaction, organizational citizenship behavior, and turnover on organizational effectiveness: A unit-level, longitudinal study. Personnel Psychology, 54: 101-114. Kozlowski, S. W. J., & Hattrup, K. 1992. A disagreement about within-group agreement: Disentangling issues of consistency versus consensus. Journal of Applied Psychology, 77: 161-167. Kozlowski, S. W. J., & Klein, K. J. 2000. A multilevel approach to theory and research in organizations: Contextual, temporal, and emergent processes. In K. J. Klein & S. W. Kozlowski (Eds.), Multilevel theory, research, and methods in organizations: Foundations, extensions, and new directions: 3-90. San Francisco: Jossey-Bass. Kristof, A. L. 1996. Person-organization fit: An integrative review of its conceptualizations, measurement, and implications. Personnel Psychology, 49: 1-49.
131
Kristof-Brown, A. L. 2000. Perceived applicant fit: Distinguishing between recruiters' perceptions of person-job and person-organization fit. Personnel Psychology, 53: 643-671. Lado, A. A., & Wilson, M. C. 1994. Human resource systems and sustained competitive advantage: A competency-based perspective. Academy of Management Review, 19: 699-727. Lambert, S. J. 2000. Added benefits: The link between work-life benefits and organizational citizenship behavior. Academy of Management Journal, 43: 801-815. Lane, P. J., & Lubatkin, M. 1998. Relative absorptive capacity and interorganizational learning. Strategic Management Journal, 19: 461-477. Lawler, E. E. III. 1981. Pay and organizational development. Reading, MA: Addison-Wesley. Lawshe, C. H. 1975. A quantitative approach to content validity. Personnel Psychology, 28: 563-575. Leifer, R., O'Connor, G. C., & Rice, M. 2001. Implementing radical innovation in mature firms: The role of hubs. Academy of Management Executive, 15: 102-113. Lepak, D. P. 2001, August. A conceptual examination of the link between strategy and human resource practices: The mediating effects of employee group and human resource philosophy. Paper presented at the 61st Annual meeting of the Academy of Management, Washington D.C. Lepak, D. P., & Snell, S. A. 2002. Examining the human resource architecture: The relationships among human capital, employment, and human resource configurations. Journal of Management, 28: 517-543.
132
Lepak, D. P., & Snell, S. A. 1999. The human resource architecture: Toward a theory of human capital allocation and development. Academy of Management Review, 24: 31-48. Lepak, D. P., Taylor, M. S., Tekleab, A. G., Marrone, J. A., & Cohen, D. J. 2002, April. Understanding the use and performance implications of high investment HR systems for different employee groups. Paper presented at the Strategic HRM conference at the University of Bath, Bath, England. LePine, J. A., Erez, A., & Johnson, D. E. 2002. The nature and dimensionality of organizational citizenship behavior: A critical review and meta-analysis. Journal of Applied Psychology, 87: 52-65. LePine, J. A., & Van Dyne, Linn. 2001. Voice and cooperative behavior as contrasting forms of contextual performance: Evidence of differential relationships with Big Five personality characteristics and cognitive ability. Journal of Applied Psychology, 86: 326-336. Leventhal, G. S., Karuza, J., & Fry, W. R. 1980. Beyond fairness: A theory of allocation preferences. In G. Miikula (Ed.), Justice and Social Interaction: 167-218. New York: Springer-Verlag. Liden, R. C., Sparrowe, R. T., & Wayne, S. J. 1997. Leader-member exchange theory: The past and potential for the future. In G. R. Ferris (Ed.), Research in Personnel and Human Resource Management, 15: 47-119. Greenwich, CT: JAI Press. Lind, E. A., & Tyler, T. R. 1988. The social psychology of procedural justice. New York: Plenum Press.
133
Lindell, M. K., & Brandt, C. J. 1999. Assessing interrater agreement on the job relevance of a test: A comparison of the CVI, T, rwg(j), and r*wg(j) indexes. Journal of Applied Psychology, 84: 640-647. Locke, E. A., & Latham, G. P. 1990. A theory of goal setting & task performance. Englewood Cliffs, NJ: Prentice Hall. Lynch, P. D., Eisenberger, R. & Armeli, S. 1999. Perceived organizational support: Inferior-versus-superior performance by wary employees. Journal of Applied Psychology, 84: 467-483. MacDuffie, J. P. 1995. Human resource bundles and manufacturing performance: Organizational logic and flexible production systems in the world auto industry. Industrial and Labor Relations Review, 48: 197-221. MacKenzie, S. B., Podsakoff, P. M., & Paine, J. B. 1999. Do citizenship behaviors matter more for managers than for salespeople? Journal of the Academy of Marketing Science, 27: 396-410. McMahan, G. C., Virick, M., & Wright, P. M. 1999. Alternative theoretical perspectives for strategic human resource management revisited: Progress, problems, and prospects. In P. M. Wright & L. Dyer & J. Boudreau & G. Milkovich (Eds.), Research in personnel and human resources management, Supplement 4: 99-122. Greenwich, CT: JAI Press. Marcoulides, G. A., & Heck, R. H. 1993. Organizational culture and performance: Proposing and testing a model. Organizational Science, 4: 209-225.
134
Masterson, S. S. 2001. A trickle-down model of organizational justice: Relating employees' and customers' perceptions of and reactions to fairness. Journal of Applied Psychology, 86: 594-604. Masterson, S. S., Lewis, K., Goldman, B. M., & Taylor, M. S. 2000. Integrating justice and social exchange: The differing effects of fair procedures and treatment on work relationships. Academy of Management Journal, 43: 738-748. Mathieu, J. E., & Zajac, D. M. 1990. A review and meta-analysis of the antecedents, correlates, and consequences of organizational commitment. Psychological Bulletin, 108: 171-194. McMahan, G. C., Virick, M., & Wright, P. M. 1999. Alternative theoretical perspectives for strategic human resource management revisited: Progress, problems, and prospects. In P. M. Wright, L. Dyer, J. Boudreau, & G. Milkovich (Eds.), Research in Personnel and Human Resources Management: Supplement 4 (pp. 99-122). Greenwich, CT: JAI Press. Meyer, J. P., & Allen, N. J. 1991. A three-component conceptualization of organizational commitment. Human Resource Management Review, 1: 61-89. Meyer, J. P., Allen, N. J., & Smith, C. A. 1993. Commitment to organizations and occupations: Extension and test of a three-component conceptualization. Journal of Applied Psychology, 78: 538-551. Miles, R. E., & Snow, C. C. 1984. Designing strategic human resources systems. Organizational Dynamics, summer: 36-52. Mincer, J. 1974. Schooling, experience, and earnings. New York: Columbia University Press.
135
Mohammed, S., Mathieu, J. E., & Bartlett, A. L. 2002. Technical-administrative task performance, leadership task performance, and contextual performance: Considering the influence of team- and task-related composition variables. Journal of Organizational Behavior, 23: 795-814. Moorman, R. H. 1991. Relationship between organizational justice and organizational citizenship behaviors: Do fairness perceptions influence employee citizenship? Journal of Applied Psychology, 76: 845-855. Morrison, E. W., & Phelps, C. C. 1999. Taking charge at work: Extrarole efforts to initiate workplace change. Academy of Management Journal, 42: 403-419. Morrison, E. W., & Robinson, S. L. 1997. When employees feel betrayed: A model of how psychological contract violation develops. Academy of Management Review, 22: 226-256. Motowidlo, S. J., & Van Scotter, J. R. 1994. Evidence that task performance should be distinguished from contextual performance. Journal of Applied Psychology, 79: 475-480. Murphy, K. R., & Cleveland, J. N. 1995. Understanding performance appraisal: Social, organizational, and goal-based perspectives. Thousand Oaks, CA: Sage. Naumann, S. E., & Bennett, N. 2000. A case for procedural justice climate: Development and test of a multilevel model. Academy of Management Journal, 43: 881-889. Nicholson, N., & Johns, G. 1985. Absence culture and the psychological contract - Who is in control of absence? Academy of Management Review, 10: 397407.
136
O'Reilly, C., III, & Chatman, J. 1986. Organizational commitment and psychological attachment: The effects of compliance, identification, and internalization on prosocial behavior. Journal of Applied Psychology, 71: 492-499. Organ, D. W. 1997. Organizational citizenship behavior: It's construct clean-up time. Human Performance, 10: 85-97. Organ, D. W. 1988. Organizational citizenship behavior: The good soldier syndrome. Lexington, MA: Lexington Books/D. C. Heath and Company. Organ, D. W., & Konovsky, M. 1989. Cognitive versus affective determinants of organizational citizenship behavior. Journal of Applied Psychology, 74: 157-164. Organ, D. W., & Ryan, K. 1995. A meta-analytic review of attitudinal and dispositional predictors of organizational citizenship behaviors. Personnel Psychology, 48: 775-802. Osterman, P. 1987. Choice of employment systems in internal labor markets. Industrial Relations, 26: 46-67. Ostroff, C., & Bowen, D. E. 2000. Moving HR to a higher level: HR practices and organizational effectiveness. In K. J. Klein & S. W. Kozlowski (Eds.), Multilevel theory, research, and methods in organizations: Foundations, extensions, and new directions (p. 211-266). San Francisco: Jossey-Bass. Peck, S. R. 1994. Exploring the link between organizational strategy and the employment relationship: The role of human resource policies. Journal of Management Studies, 31: 715-736. Pennings, J. M., Lee, K., & Van Witteloostuijn, A. 1998. Human capital, social capital, and firm dissolution. Academy of Management Journal, 41: 425-440.
137
Pfeffer, J. 1998. The human equation: Building profits by putting people first. Boston, MA: Harvard Business School Press. Pfeffer, J. 1994. Competitive advantage through people: Unleashing the power of the work force. Boston: Harvard Business School Press. Pinfield, L. T., & Berner, M. F. 1994. Employment systems: Toward a coherent conceptualization of internal labor markets. In G. Ferris (Ed.), Research in Personnel and Human Resources Management, 12: 41-78. Greenwich, CT: JAI Press. Podsakoff, P. M., & MacKenzie, S. B. 1994. Organizational citizenship behaviors and sales unit effectiveness. Journal of Marketing Research, 3: 351-363. Powell, T. C. 1992. Organizational alignment as competitive advantage. Strategic Management Journal, 13: 119-134. Quinn, R. E. 1988. Beyond rational management: Mastering the paradoxes and competing demands of high performance. San Francisco: Jossey-Bass. Quinn, R. E., & Rohrbaugh, J. 1983. A spatial model of effectiveness criteria: Towards a competing values approach to organizational analysis. Management Science, 29: 363-377. Reichers, A. E. 1985. A review and reconceptualization of organizational commitment. Academy of Management Review, 10: 465-476. Rigdon, E. E., Schumacker, R. E., & Marcoulides, G. A. 1998. A comparative review of interaction and nonlinear modeling. In R. E. Schumacker & G. A. Marcoulides (Eds.), Interaction and nonlinear effects in structural equation modeling: 116. Mahwah, NJ: Lawrence Erlbaum Associates.
138
Roberson, Q. M., Moye, N. A., & Locke, E. A. 1999. Identifying a missing link between participation and satisfaction: The mediating role of procedural justice perceptions. Journal of Applied Psychology, 84: 585-593. Rogers, E. W., & Wright, P. M. 1998. Measuring organizational performance in strategic human resource management: Problems, prospects, and performance information markets. Human Resource Management Reviews, 8: 311-331. Roggs, K. L., Schmidt, D. B., Shull, C., & Schmitt, N. 2001. Human resource practices, organizational climate, and customer satisfaction. Journal of Management, 27: 431-449. Rousseau, D. M. 2000. Psychological contract report inventory technical report.http://www.andrew.cmu.edu/user/rousseau/0_reports/PCI.pdf. Rousseau, D. M. 1995. Psychological contracts in organizations: Understanding written and unwritten agreements. Thousand Oaks, CA: Sage Publications. Russell, J. S., Terborg, J. R., & Powers, M. L. 1985. Organizational performance and organizational level training and support. Personnel Psychology, 38: 849-863. Rupp, D. E., & Cropanzano, R. 2002. The mediating effects of social exchange relationships in predicting workplace outcomes from multifoci organizational justice. Organizational Behavior and Human Decision Processes, 89: 925-946. Ryan, A. M., Schmit, M. J., & Johnson, R. 1996. Attitudes and effectiveness: Examining relations at an organizational level. Personnel Psychology, 49: 853-882. Scandura, T. A., & Graen, G. B. 1984. Moderating effects of initial leadermember exchange status on the effects of a leadership intervention. Journal of Applied Psychology, 71: 579-584.
139
Schneider, B. 1975. Organizational climate: Individual preferences and organizational realities revisited. Journal of Applied Psychology, 60: 459-465 Schneider, B., & Bowen, D. E. 1985. Employee and customer perceptions of service in banks: Replication and extension. Journal of Applied Psychology, 70: 423433. Schneider, B., Goldstein, H. W., & Smith, D. B. 1995. The ASA framework: An update. Personnel Psychology, 48: 747-773. Schneider, B., & Reichers, A. 1983. On the etiology of climates. Personnel Psychology, 36: 19-40. Schneider, B., Salvaggio, A. N., & Subirats, M. 2002. Climate strength: A new direction for climate research. Journal of Applied Psychology, 87: 220-229. Schneider, B., White, S. S., & Paul, M. C. 1998. Linking service climate and customer perceptions of service quality: Test of a causal model. Journal of Applied Psychology, 83: 150-163. Schuler, R. S., & Jackson, S. E. 1987. Linking competitive strategies with human resource management practices. Academy of Management Executive, 1: 207219. Settoon, R. P., Bennett, N., & Liden, R.C. 1996. Social exchange in organizations: Perceived organizational support, leader-member exchange, and employee reciprocity. Journal of Applied Psychology, 81: 219-227. Sheppard, B. H., Lewicki, R. J., & Minton, J. W. 1992. Organizational justice: The search for fairness in the workplace. New York: Lexington Books/Macmillan.
140
Shore, L. M., Barksdale, K., & Shore, T. H. 1995. Managerial perceptions of employee commitment to the organization. Academy of Management Journal, 38: 15931615. Shore, L. M., & Wayne, S. J. 1993. Commitment and employee behavior: Comparison of affective commitment and continuance commitment with perceived organizational support. Journal of Applied Psychology, 78: 774-780. Snell, S. A., & Dean, J. W., Jr. 1994. Strategic compensation for integrated manufacturing: The moderating effects of jobs and organizational inertia. Academy of Management Journal, 37: 1109-1140. Snell, S. A., & Dean, J. W., Jr. 1992. Integrated manufacturing and human resource management: A human capital perspective. Academy of Management Journal, 35: 467-504. Sparrowe, R. T., & Liden, R. C. 1997. Process and structure in leader-memberexchange. Academy of Management Review, 22: 522-552. Sullivan, P. H. 2000. Value-driven intellectual capital: How to convert intangible corporate assets into market value. New York: John Wiley & Sons. Taylor, M. S., Masterson, S. S., Renard, M. K., & Tracy, K. B. 1998. Managers' reactions to procedurally just performance management systems. Academy of Management Journal, 41: 568-579. Taylor, M. S., Tracy, K. B., Renard, M. K., Harrison, J. K., & Carroll, S. 1995. Due process in performance appraisals: A quasi-experiment in procedural justice. Administrative Science Quarterly, 40: 495-523.
141
Terpstra, D. E., & Rozell, E. J. 1993. The relationship of staffing practices to organizational level measures of performance. Personnel Psychology, 46: 27-48. Tett, R. P., & Meyer, J. P. 1993. Job satisfaction, organizational commitment, turnover intention, and turnover: Path analyses based on meta-analytic findings. Personnel Psychology, 46: 259-293. Thomas, J. B., Clark, S. M., & Gioia, D. A 1993. Strategic sensemaking and organizational performance: Linkages among scanning, interpretation, action, and outcomes. Academy of Management Journal, 36: 239-270. Tinsley, H. E. A., & Weiss, D. J. 1975. Interrater reliability and agreement of subjective judgments. Journal of Counseling Psychology, 22: 358-376. Tsui, A. S., Pearce, J. L., Porter, L. W., & Hite, J. P. 1995. Choice of employeeorganization relationship: Influence of external and internal organizational factors. In G. R. Ferris (Ed.), Research in Personnel and Human Resources Management: Vol. 13 (p. 117-151). Greenwich, CT: JAI Press. Tsui, A. S., Pearce, J. L., Porter, L. W., & Tripoli, A. A. 1997. Alternative approaches to the employee-organization relationship: Does investment in employee payoff? Academy of Management Journal, 40: 1089-1121. Tyler, T. R., & Lind, E. A. 1992. A relational model of authority in groups. In M. Zanna (Ed.), Advances in Experimental Social Psychology, 25: 115-192. New York: Academic Press. Tziner, A., & Kopelman, R. 1988. Effects of rating format on goal-setting dimensions: A field experiment. Journal of Applied Psychology, 73: 323-326.
142
Van Dyne, L., & Ang, S. 1998. Organizational citizenship behavior of contingent workers in Singapore. Academy of Management Journal, 41: 692-703. Van Dyne, L., Graham, J. W., & Dienesch, R. M. 1994. Organizational citizenship behavior: Construct redefinition, measurement, and validation. Academy of Management Journal, 37: 765-802. Van Dyne, L., & LePine, J. A. 1998. Helping and voice extra-role behaviors: Evidence of construct and predictive validity. Academy of Management Journal, 41: 108-119. Vance, R. J., Coovert, M. D., MacCallum, R. C., & Hedge, J. W. 1989. Construct models of task performance. Journal of Applied Psychology, 74: 447-455. Van Scotter, J. R. 2000. Relationships of task performance and contextual performance with turnover, job satisfaction, and affective commitment. Human Resource Management Review, 10: 79-95. Van Scotter, J. R., & Motowidlo, S. J. 1996. Evidence for two factors of contextual performance: Job dedication and interpersonal facilitation. Journal of Applied Psychology, 81: 525-531. Van Vianen, A. E. M. 2000. Person-organization fit: The match between newcomers' and recruiters' preferences for organizational cultures. Personnel Psychology, 53: 113-149. Walton, E. J., & Dawson, S. 2001. Managers' perceptions of criteria of organizational effectiveness. Journal of Management Studies, 38: 173-199.
143
Walz, S. M., & Niehoff, B. P. 1996. Organizational citizenship behaviors and their effect on organizational effectiveness in limited-menu restaurants. Academy of Management Best Paper Proceedings, 56th Annual Meeting, Cincinnati, OH, 307-311. Wanous, J. P. 1980. Organizational entry: Recruitment, selection, and socialization of newcomers. Reading, MA: Addison-Wesley. Wanous, J. P., & Hudy, M. J. 2001. Single-item reliability: A replication and extension. Organizational Research Methods, 4: 361-375. Wanous, J. P., Reichers, A. E., & Hudy, M. J. 1997. Overall job satisfaction: How good are single-item measures? Journal of Applied Psychology, 82: 247-252. Way, S. A. 2002. High Performance Work System and intermediate indicators of firm performance within the US small business sector. Journal of Management, 28: 765-785. Wayne, S. J., Shore, L. M., & Liden, R. C. 1997. Perceived organizational support and leader-member exchange: A social exchange perspective. Academy of Management Journal, 40: 82-111. Weick, K. 1979. The social psychology of organizing. Reading, MA: AddisonWesley. Wernerfelt, B. 1984. A resource-based view of the firm. Strategic Management Journal, 5: 171-180. Whitener, E. M. 2001. Do "high commitment" human resource practices affect employee commitment? A cross-level analysis using hierarchical linear modeling. Journal of Management, 27: 515-535.
144
Williams, L. J., & Anderson, S. E. 1991. Job satisfaction and organizational commitment as predictors of organizational citizenship and in-role behaviors. Journal of Management, 17: 601-617. Williams, C. R., & Livingstone, L. P. 1994. Another look at the relationship between performance and voluntary turnover. Academy of Management Journal, 37: 269-298. Wright, P. M. 1998. Introduction: Strategic human resource management research in the 21st century. Human Resource Management Review, 8: 187-191. Wright, P. M., Gardner, T., Moynihan, L. M., & Park, H. J. 2000. Unlocking the black box: Examining the processes through which human resource management practices impact business performance. Paper presented at the 60th annual meeting of the Academy of Management, Toronto, Canada. Wright, P. M., & George, J. M. 1993. Productivity and extra-role behavior: The effects of goals and incentives on spontaneous helping. Journal of Applied Psychology, 78: 374-381. Wright, P. M., & McMahan, G. C. 1992. Theoretical perspectives for strategic human resource management. Journal of Management, 18: 295-320. Wright, P. M., & Sherman, W. S. 1999. Failing to find fit in strategic human resource management: Theoretical and empirical problems. In P. M. Wright, L. Dyer, J. Boudreau, & G. Milkovich (Eds.), Research in Personnel and Human Resources Management, Supplement 4: 53-74. Greenwich, CT: JAI Press.
145
Wright, P. M., Smart, D. L., & McMahan, G. C. 1995. Matches between human resources and strategy among NCAA basketball teams. Academy of Management Journal, 38: 1052-1074. Wright, P. M., & Snell, S. A. 1991. Toward an integrative view of strategic human resource management. Human Resource Management Review, 1: 203-225. Wright, P. M., & Snell, S. A. 1998. Toward a unifying framework for exploring fit and flexibility in strategic human resource management. Academy of Management Review, 23: 756-772. Youndt, M. A., Snell, S. A., Dean, J. W., Jr., & Lepak, D. P. 1996. Human resource management, manufacturing strategy, and firm performance. Academy of Management Journal, 39: 836-866.
146
doc_431442459.docx