Test Assesment

Journal of Management Vol. 44 No. 7, September 2018 2690 –2715

DOI: 10.1177/0149206316646829 © The Author(s) 2016

Article reuse guidelines: sagepub.com/journals-permissions

2690

Team Manager’s Implementation, High Performance Work Systems Intensity, and Performance: A Multilevel Investigation

Jongwook Pak Seongsu Kim

Seoul National University

Recently, capturing within-organization variability during the implementation of high perfor- mance work systems (HPWS) has received considerable attention; however, the source of such variability has rarely been considered. If the utilization of HPWS is positively related to perfor- mance outcomes, examining factors contributing to an effective implementation may yield signifi- cant theoretical and practical implications. For this purpose, this study extends the extant HPWS literature in two ways. First, we attempt to conceptualize team-level HPWS intensity and identify antecedents of variance across teams. Specifically, we regard the visible role of team managers in the process of HPWS implementation as a primary interpretive filter that makes team members perceive differences in HPWS intensity, which in turn affects team performance. Second, we posit that if human resources (HR) policies are viewed as an exchange agreement between the organiza- tion and its employees, then a team manager more actively enforcing espoused HR practices may positively influence the sense of human resource management (HRM)–induced psychological contract fulfillment of team members, which in turn influences individual in-role performance and organizational citizenship behavior (OCB). Our hypotheses are tested with data from 183 matched responses from 51 teams, and the results generally support both the team-level and multilevel hypotheses. We discuss the theoretical and managerial implications of our study.

Keywords: team manager’s implementation; high performance work systems intensity; HRM- induced psychological contract fulfillment; performance

Acknowledgments: This article was accepted under the editorship of Patrick M. Wright. This work was supported by the Ministry of Education of the Republic of Korea and the National Research Foundation of Korea (NRF- 2015S1A5A2A03047915). We thank the editor and two anonymous reviewers for their constructive comments, which helped us to improve the article.

Corresponding author: Seongsu Kim, Professor of HRM, Graduate School of Business, Seoul National University, 1 Gwanak-ro, Gwanak-gu, Seoul, Korea.

E-mail: [email protected]

646829 JOMXXX10.1177/0149206316646829Journal of ManagementPak, Kim / HPWS Implementation research-article2016

Pak, Kim / HPWS Implementation 2691

Lately, interest in the implementation effectiveness of high performance work systems (HPWS) within organizations has increased (Guest & Bos-Nehles, 2013; Jiang, Takeuchi, & Lepak, 2013; Wright & Nishii, 2013). Studies in this vein reflect responses to recent concerns about the HPWS research tradition. That is, HPWS research utilizing firm-level data implicitly assumes that HPWS is implemented throughout the organization in an intended manner, with human resources (HR) professionals, management, and individuals attaching similar meanings to HR experiences (Nishii & Wright, 2007). Although this increased scrutiny of underlying assumptions has spurred a new research stream, these assumptions are only rational for studies measuring within-organization HPWS (e.g., on the individual or branch level). In reality, the issue of why different patterns of implemen- tation are observed even within organizations in which the same set of HR policies is espoused is rarely considered. In research to date, HPWS has been treated as an indepen- dent variable (i.e., a matter of adoption), and the mechanism through which it affects per- formance indicators has been the primary focus (e.g., Boon, Den Hartog, Boselie, & Paauwe, 2011; Boxall, Ang, & Bartram, 2011; Zacharatos, Barling, & Iverson, 2005). If intense utilization of HPWS can indeed facilitate effective functioning of organizations (Combs, Liu, Hall, & Ketchen, 2006; Jiang, Lepak, Hu, & Baer, 2012; Subramony, 2009), exploring antecedents that determine the intensity of implementation of HPWS may yield valuable theoretical and practical implications.

With the shift toward HPWS as a matter of implementation, the role of the team manager has increasingly been recognized. Although team managers may not be the sole deliverers of HR practices, they function as agents for organizations in enforcing HR policies in their work groups (Purcell & Hutchinson, 2007). Clearly, variability in implementation patterns can result because team managers have different management skills and levels of motivation (Cunningham & Hyman, 1999; Delery & Shaw, 2001). In addition, team managers usually have a certain amount of discretion in implementation of these practices, as HR policies can- not encompass every contingency that may arise (Nishii & Wright, 2007). Thus, we posit that distinct HPWS implementation patterns are likely to emerge at the team level since continu- ous interaction among team members can result in a shared understanding of their HR experi- ences (Dragoni, 2005). It is suggested that a visible supervisor who enforces HR policies (i.e., those related to HPWS) in a prescribed manner creates a strong psychological climate among team members concerning the work environment (Bowen & Ostroff, 2004).

The results of our study confirm that the degree to which team managers deliver unam- biguous HR messages to employees during implementation is significantly linked to the intensity of HPWS at the team level and, subsequently, to team performance. Furthermore, we identify a positive association between intense implementation of HPWS and individual- level psychological contract fulfillment, which in turn relates to individual performance. In this study, we view the relationship between team-level HPWS and employee outcomes as a function of the degree to which promises in the human resource management (HRM) domain are fulfilled. Organizations acquaint individuals with HR strategies and specific HR policies to which they are subject through new employee training programs upon employment. In the early stage, employees learn the terms of their contracts: what they are required to do and what the organization offers them in return. Over time, the efforts exerted by team managers to meet the terms of these contracts equitably can affect employees considerably in terms of psychological contract fulfillment (Rousseau, 1990; Shore & Tetrick, 1994).

2692 Journal of Management / September 2018

In the following section, we discuss theories and the findings of studies relevant to the development of our conceptual framework. We subsequently test our team-level and multi- level hypotheses based on data gathered from a large Korean company. After presenting the results, we discuss the theoretical and practical implications of our research. Limitations and suggestions for future research directions are provided accordingly.

Literature Review

Since Huselid’s (1995) seminal research, a number of studies have examined the relation- ship between HPWS and performance. Although the link between HPWS and performance is generally recognized (Combs et al., 2006; Jiang et al., 2012; Subramony, 2009), some skepti- cism still lingers about how effective HPWS truly is in enhancing performance (Paauwe, Wright, & Guest, 2013). At this juncture, implementation of HR practices has emerged as a salient research line. This research addresses the inherent problems of prior approaches toward firm-level HPWS research (Jiang et al., 2013). Studies utilizing firm-level data often base their findings upon an implicit assumption that espoused HR policies are implemented in a consistent, intentional manner across organizations (Wright & Nishii, 2013). For managerial reporting of HPWS, the problem resides in that the reality of HPWS implementation, or the actual modus operandi, cannot adequately be captured because managers or HR professionals who report on HPWS may represent espoused HR policies, and their reports may not neces- sarily reflect realized HR practices in the organization (Nishii & Wright, 2007). Indeed, many researchers have attempted to capture the variance in intensity of implementation of HPWS within organizations by measuring at the unit or individual level (e.g., Boon et al., 2011). Rarely considered is why variability in implementation patterns is observed within organiza- tions in the context of the HRM-performance link. Therefore, in this article, we explore poten- tial sources of variation in intensity of HPWS implementation within organizations.

Team Managers, HPWS Intensity, and Team Performance

Figure 1 shows the conceptual model of our study. We begin our discussion by suggesting that the distinct and consistent role of team managers in enforcing HR policies in their work groups may be critical for enhancing team performance. By its very nature, HRM concerns the implementation of strategy (Mathis & Jackson, 1985). Once strategy is formulated, HR practices are designed in a way that enables the organization to realize its strategic objectives (Schuler & MacMillan, 1984). HRM contributes to improved performance by clearly defin- ing the roles and responsibilities of individuals and motivating them to work toward perfor- mance goals during strategy implementation (Schuler & Jackson, 1987). Thus, performance may be explicated in terms of the extent to which intended HR practices are enforced within the organization.

Recently, team managers have increasingly been recognized as agents within organiza- tions who implement HR policies in their work groups (Purcell & Hutchinson, 2007; Wright & Nishii, 2013). Thus, team managers actively engaging in HRM are integral to firm strategies (Anderson, Cooper, & Zhu, 2007), as they have immediate impact on non- managerial employees (Poole & Jenkins, 1997). Therefore, business objectives are more likely to be achieved when team managers enforce HR policies as intended in their work

Pak, Kim / HPWS Implementation 2693

groups. However, variability may be evident within the organization because team manag- ers not only have responsibility to manage their members’ efforts and administer HR poli- cies, but they also have a certain amount of discretion in determining the modes through which HR policies are implemented in their work groups since HR policies cannot encom- pass every contingency that might arise (Nishii & Wright, 2007). Furthermore, team man- agers may have different levels of motivation or competence in complying with and enforcing HR policies effectively (Cunningham & Hyman, 1999; Delery & Shaw, 2001). For instance, team managers may believe that performance-oriented HR practices are not compatible with social norms governing society, or that a certain HR practice (e.g., train- ing) will not improve overall team performance if the contents of its programs do not reflect the nature of the team’s daily routines (Pak & Chung, 2013). As a consequence, a decoupling between intended and actual HR practices may arise. The wider the gap between intended and actual HR practices in teams, the harder it is to attain expected outcomes. Therefore, we contend that team performance may depend on the quality of HR policy enforcement by team managers.

Our study suggests that the relationship between team managers’ implementation behavior and team performance is mediated by HPWS intensity. In the two-step process proposed by McGuire (1972), employees should be initially aware of and comprehend messages to ensure effective communication. They should then agree on and remember the message content. Bowen and Ostroff (2004) described a set of meta-features that constitute the strength of HRM systems. They suggested that a strong HRM system emerges when the psychological climate of employees concerning their HR experiences is distinctive, consistent, and consensual. According to Bowen and Ostroff (2004), distinctiveness refers to the extent to which HR prac- tices are clearly recognized and understood by employees. Consistency denotes the degree of internal fit among adopted HR practices and unambiguous messages delivered to employees. Consensus involves agreement among organizational members concerning the company’s vision and values with regard to HRM and the equitable implementation of HR policies. They proposed that the stronger the HRM system, the more likely it is that employees will perceive

Figure 1 Proposed Model

2694 Journal of Management / September 2018

the same shared psychological climate. When variability in these perceptions of the climate in the workplace is minimized, desirable performance levels may be achieved. Therefore, it is crucial that HR messages should be clearly and consistently delivered to employees in the workplace (Chaiken, Wood, & Eagley, 1996).

We propose that team managers’ commitment to implementation of HPWS is a distinct interpretive filter that employees may utilize to make sense of their work environment. In other words, HR messages may not be delivered to employees directly from the HRM sys- tem per se. Employees may instead interpret the effectiveness of HPWS through proximate filters. According to the situational perspective (Schneider, 1990), people are inclined to search for and extract relevant contextual cues to derive specific meanings from their expe- riences; what they actually see is likely to shape their perceptions of the surrounding envi- ronment (Ross & Nisbett, 1991). This implies that team managers enforcing HR practices in their work groups may act as adjacent or readily accessible contextual filters that may affect employees’ understanding of the qualities of the HRM systems in their organiza- tions. On this matter, Bowen and Ostroff (2004: 215-216) noted that “supervisors can serve as interpretive filters of HRM practices, and when they are visible in implementing prac- tices or promote high-quality exchanges with employees, they can introduce a common interpretation among unit members.” From this perspective, we see that team managers play a critical role in shaping shared understandings of realized HR experiences as well as conveying intended information among employees.

Due to the importance of team managers during HPWS implementation (e.g., Bowen & Ostroff, 2004; Purcell & Hutchinson, 2007), two fruitful research avenues may be identified. First, scholars examining the role of team managers have created an independent research stream (e.g., Gilbert, De Winnea, & Selsa, 2011; Sterling & Boxall, 2013) separate from the HPWS-performance link. Our study suggests that strategic HRM (SHRM) research that mea- sures HPWS within organizations (i.e., a matter of implementation) has now made it possible to investigate the source of variability. In doing so, future researchers can directly determine what factors contribute to the intensity of HPWS implementation and its relation to perfor- mance outcomes. Second, we discovered that little attention has yet been given to conceptu- alizing specific behaviors related to implementation of espoused HR practices despite their increasing significance in the SHRM literature. For example, Ryu and Kim (2013) found that involvement of first-line managers (FLM) in HR is positively related to the effectiveness of HRM. Interestingly, this positive relationship turns negative when institutionally emerging HR practices (i.e., HPWS) are considered together, and this negative moderating relationship is alleviated when HR knowledge is transferred to FLMs. The results of their study imply that despite the importance of FLMs’ involvement in HRM, it does not guarantee that FLMs will act in accordance with prescribed procedures. That is, heavily involved FLMs may still enforce HR practices in a bureaucratic, seniority-oriented way even when performance-ori- ented HR practices are espoused in the organization. In this study, we argue that to create an environment in which employees experience HR practices as intended by the organization, team managers’ efforts to enforce intended HR practices is essential. Here, we define the construct team manager’s implementation of espoused HR practices (TIHR) to complement those constructs found in the extant HR literature, as the extent to which a team manager acts as a deliverer and an advocate of HR policies in the work group, following HR procedures and enforcing HR practices in a prescribed manner.

Pak, Kim / HPWS Implementation 2695

We posit that variability in HPWS intensity may be captured meaningfully at the team level. The social interaction perspective (Weick, 1995) suggests that employees working in a team are likely to influence one another in shaping understanding of their HR experi- ences. As individuals in a work group interact frequently and experience a similar work context, social interaction among team members is likely to produce a shared perception of the environment (Dragoni, 2005; Kozlowski & Klein, 2000). Over time, a collective experience can emerge that includes making sense of the visible role of managers in the work group. In this study, HPWS refers to a set of HR practices aimed at enhancing “employees’ skills, commitment, and productivity in such a way that employees become a source of sustainable competitive advantage” (Datta, Guthrie, & Wright, 2005: 136), and HPWS intensity (HPWSI) is defined as the extent to which HPWS is actually imple- mented as intended in work groups.

Although HPWS at the team level is a worthwhile subject for investigation, only a few empirical studies have examined its effect on work group outcomes (e.g., Lee, Pak, & Kim, 2014; Pak, Kim, & Li, 2015). However, several scholars have asserted that HPWSI at the team level may be positively related to team performance (TP). For instance, a recent study by Fu, Flood, Bosak, Morris, and O’Regan (2013) suggests that HPWS positively influ- ences team formation. Individuals placed on the right team cooperate better with clients, which improves organizational performance. Messersmith, Patel, Lepak, and Gould- Williams (2011) reported that HPWS is positively associated with collective job satisfac- tion, commitment, and organizational citizenship behaviors (OCB). In addition, Takeuchi, Lepak, Wang, and Takeuchi (2007) confirmed that the HPWS-performance relationship is mediated by enhancement of collective human capital and social exchange. Thus, we also posit that the intensity of implementation of HPWS at the team level is positively associated with TP. Considering all of these findings, we propose that team managers have a distinct influence on work climates in which HPWS are implemented and that different performance outcomes may be observed as a result (Bartel, 2004). Therefore, we present the following hypothesis:

Hypothesis 1: The positive relationship between TIHR and TP is mediated by HPWSI.

TIHR and HRM-Induced Psychological Contract Fulfillment (HPCF)

In this article, we regard HPWS as a means for employers to communicate to employees what is required of them in their jobs and what they can expect to receive in return for their efforts (Guzzo & Noonan, 1994). From this perspective, HR policies are agreements regard- ing the terms of exchange between organizations and their employees (Guest & Conway, 2000; Sonnenberg, 2006). Individuals are often exposed to HR policies in the early stage of their employment through such initiatives as new employee programs offered by the employer. Throughout the process of becoming familiar with their jobs, employees’ expecta- tions about what is promised regarding both their membership in and contribution to their organizations take shape.

As previously discussed, team managers, as agents of their organizations, are responsible for enforcing HR policies in their work groups. Thus, the extent to which team managers assume a visible role in the implementation of HPWS may affect the degree to which psycho- logical contracts are fulfilled (Den Hartog, Boselie, & Paauwe, 2004). The psychological

2696 Journal of Management / September 2018

contract is the target of an organization’s efforts to meet its obligations equitably (Rousseau, 1990; Shore & Tetrick, 1994), and prior studies have recognized team managers as important representatives in the employee-organization relationship (Guest & Conway, 2004; Rousseau, 1995). However, few empirical studies have been conducted of team managers’ roles in ful- filling individual psychological contracts. In this study, we suggest that TIHR may be viewed as a major means of shaping employees’ beliefs about the exchange relationship with their organization and enhancing perceived organizational support. Here, we propose the construct of HPCF, or employees’ perceptions of the degree to which organizational promises are met in the HRM domain, as the first individual-level outcome influenced by the implementation behaviors of the team manager.

From the social exchange perspective (Blau, 1964), HPCF is reciprocal in nature. When employees’ contributions are properly recognized with commensurate rewards, and when organizations provide appropriate means through which they can do their jobs (e.g., training, participation), HR practices espoused by the employer are more likely to be evi- dent (Bowen & Ostroff, 2004), and their positive impact on HPCF is more likely to be visible (McDermott, Conway, Rousseau, & Flood, 2013). Lambert (2011) suggested that what is actually delivered, and not what is promised, more accurately predicts individual attitudes. Thus, we suggest that individual perception regarding delivery of promises dur- ing the implementation phase may also be a function of intensive utilization of intended HR practices (Rousseau, 1990; Shore & Tetrick, 1994). Therefore, we propose the follow- ing hypothesis:

Hypothesis 2: The positive relationship between TIHR and HPCF is mediated by HPWSI.

HPWSI and Individual Outcomes

We suggest that implementing HPWS at greater intensity may improve employee atti- tudes and behaviors. The relationship between HPWS and individual performance outcomes can be elucidated by the ability-motivation-opportunity (AMO) framework of previous HRM research (Appelbaum, Bailey, Berg, & Kalleberg, 2000; Delery & Shaw, 2001; Gardner, Wright, & Moynihan, 2011). HPWS practices contribute to improving employee performance by enhancing the skills, competences, and motivation of employees, as well as increasing their opportunities to contribute. Studies from the behavioral perspective (Schuler, 1989; Schuler & Jackson, 1987) suggest that the appropriate role behaviors of skilled employees combined together enhance firm performance. Generally, human capital is enhanced not only via selective recruitment and staffing, but also through extensive train- ing and development programs, which are part of HPWS (Takeuchi et al., 2007). Tools for rigorous, objective performance evaluation, performance-based compensation, and advance- ment schemes boost employee morale and motivation (Boxall et al., 2011). In addition, empowerment and participative practices, which are often included in HPWS, positively affect the commitment of individuals to their jobs and their organizations (Ehrnrooth & Bjorkman, 2012).

We focus on individual in-role performance (IP) and OCB among several possible perfor- mance indicators because performance of the core job (i.e., roles and responsibilities speci- fied in the job description) appropriately represents the effectiveness of individuals at their

Pak, Kim / HPWS Implementation 2697

tasks. OCBs, including discretionary behaviors such as helping others in need and being conscientious, collectively promote the effective functioning of organizations (Organ, 1988). A previous study confirmed that HPWS is significantly related to both service performance and helping behavior, and that workplace relationships are mediated by a climate of concern (Chuang & Liao, 2010). Moreover, HPWS is associated with person-job fit, which is, in turn, related to intent to demonstrate OCB. Snape and Redman (2010) similarly provided empiri- cal evidence that HPWS positively affects individual IP and OCB.

From the reciprocity perspective (Gouldner, 1960), the HPWS-performance link is enhanced by a psychological climate characterized by fulfillment of the exchange agreement. We contend that the positive effects of HPWS on employee attitudes and behaviors are real- ized only to the extent that individuals perceive that what they receive and how they are treated are commensurate with their contributions to the organization. Hence, the more psy- chological contracts are perceived to be fulfilled, the more favorable employee attitudes will be toward coworkers and the more committed employees will be to their jobs (Ehrnrooth & Bjorkman, 2012; Podsakoff, MacKenzie, Moorman, & Fetter, 1990). Prior studies confirmed that individuals experiencing contract breach tend to reduce their commitment to and engage- ment in OCB (Coyle-Shapiro & Kessler, 2000) and that fulfillment of the psychological contract positively influences employee performance (Conway & Coyle-Shapiro, 2012). Our discussion thus far provides a rationale for the following hypothesis:

Hypothesis 3: The positive relationship between HPWSI and employee performance (i.e., IP, OCB) is mediated by HPCF.

Research Method

In this section, we first describe the research setting and introduce the measures utilized in our research model. We also provide our rationale for including supportive manager behavior (SMB) as a control variable. Then, we justify aggregation of individual responses to the team level. Finally, we outline the analytical approaches adopted in this study and detail the results of testing for discriminant and convergent validity of the study variables.

Data and Sample

Surveys were conducted among employees of a large Korean company. Before conduct- ing the surveys, the authors visited the company’s HR department three times to disclose the purpose of our study and discuss survey items as needed. During this period, we conducted several interviews to ensure that the HR practices of the company were designed according to the principles of HPWS. We also reviewed HR policies and various HR-related programs with the employee from the HR department who examined the elements and policies of the organization related to HPWS. At the end of the process, the authors and personnel from the HR department agreed that the company’s HR practices were generally compatible with the design principles of HPWS.

Each team in the organization selected for this study assumes a distinct role, and teams are generally intended to function semiautonomously. They are often assigned complex, knowl- edge-intensive tasks that require collaboration in a rapidly changing business environment.

2698 Journal of Management / September 2018

Employees are evaluated based on both individual and team performance; the company offers pay-for-performance and group-based incentive schemes.

For security purposes, web-based surveys were administered through the company intranet system to a total of 1,722 employees from 181 teams. To avoid common rater effects, the outcome variables of our study (i.e., IP, OCB) were measured by team managers for each team member. Team managers also rated TP as a unit. A total of 261 employees responded to our survey (15.2%), but we dropped 78 responses that did not match the team manager responses. Therefore, our final sample consisted of 183 matched responses (10.6%) and 51 team manager responses (28.2%). The average size of the teams in our final sample is 7.82 employees (SD = 4.05). Female respondents accounted for 39% of the sample. The average age of the respondents is 30.67 years (SD = 4.17). Most respondents (i.e., approximately 88%) hold an undergraduate degree.

We compared the composition of our sample and the company workforce and found no considerable difference between them in terms of gender and job group. In our final sample, female respondents accounted for 39%, whereas females compose 33% of the company workforce. Furthermore, our final sample, exclusive of manufacturing, consisted of person- nel from management/administration (58.8%), sales (21.4%), and R&D (19.8%), which make up 63%, 23%, and 14% of the departments in the company, respectively. Additionally, we conducted a one-sample z-test for proportions to detect significant differences between the composition of our sample and that of the company workforce. Consistent with our pre- diction, the results showed that the proportions of males (z = 1.34, p = .18) and females (z = 1.12, p = .26) were not significantly different. As for job group, the statistics were as follows: management/administration (z = 1.09, p = .28), sales (z = 1.80, p = .07), and R&D (z = 1.28, p = .20). As p-values are all above the cutoff value of .05, we identified only minimal differ- ences in terms of gender and job group between our sample and the company workforce. Therefore, we did not find evidence of nonresponse bias.

Measures

Unless otherwise stated, responses were measured on a 5-point Likert scale ranging from completely disagree (1) to completely agree (5).

TIHR. No measure has been established in previous studies to explicate the phenom- enon of interest. Although a visible supervisor was suggested by Bowen and Ostroff (2004), this variable is still conceptual despite its potential contribution to the HR implementation research. Recently, a few studies have investigated the role of line managers (e.g., Purcell & Hutchinson, 2007; Ryu & Kim, 2013). However, no construct specifically measuring team manager behaviors related to implementation of espoused HR practices is evident in the HRM literature. Therefore, we developed a construct for this purpose. To do so, we initially drew on a measure from previous literature, supervisor visibility in demonstrating procedural jus- tice (Naumann & Bennett, 2000), which is comparable in several respects. We subsequently conducted semistructured interviews with 16 MBA/EMBA students enrolled in a graduate school of business to capture those factors related to team managers who enforce HR policies in a prescribed manner in their work groups. Then, we extracted meaningful themes from our interview transcriptions and adapted them to the established construct. Finally, we developed

Pak, Kim / HPWS Implementation 2699

a five-item measure for TIHR. A sample item included, “Many times I have witnessed that my team manager puts a strong emphasis on our participation in HR programs (e.g., training, culture-building activities) even when we are busy working.” The results showed good reli- ability (α = .91). The appendix provides a full list of items.

HPWSI. To develop this construct, we used the theoretical framework of Sun, Aryee, and Law (2007), who focused on staffing, training, career development, compensation, and participation. Their conceptualization of HPWS is well aligned with the AMO framework (Appelbaum et al., 2000; Delery & Shaw, 2001; Gardner et al., 2011) in terms of the contri- bution of HPWS in enhancing organizational functioning by increasing human capital, moti- vation, and opportunity to contribute. We also utilized items from multiple previous studies; 14 items were from Sun et al. (2007), 5 items were from Takeuchi et al. (2007), 3 items were from Collins and Smith (2006), and 2 items were from Bae and Lawler (2000). From the original collection of 24 items, we adapted items to the team level and interviewed an HR manager from the target company to capture practices that were relevant to our research set- ting. During the course of the interview, we dropped 9 items to ensure that the selected items related to HPWS in our study accurately reflected the HRM system of the company from which the data was taken. Then, using 15 items, we conducted a principal component factor analysis using varimax rotation and determined that the items were loaded on six factors. The total eigenvalue was 13.78, and the cumulative explained variance was 74.62%. Following Sun et al. (2007), we developed a single comprehensive measure with which to analyze a set of HR practices (Becker & Huselid, 1998). We also found that intercorrelations between the six factors were relatively high, ranging from .55 to .75. The unitary index showed good reliability (α = .87). The appendix provides a complete list of HPWS-related items used in this study.

HPCF. We developed the next construct by adjusting the four-item scale measuring psy- chological contract fulfillment developed by Henderson, Wayne, Shore, Bommer, and Tet- rick (2008) to fit our research context. Sample items included, “Considering the promises my company has made to me [in relation to compensation, advancement, training, participation, etc.], the company hasn’t always lived up to its end of the bargain” (this item was reverse scored), and “Considering my HR experiences, my company has kept its promises to me.” Reliability of the scale was above the cutoff value of .70 (α = .77).

TP. We measured TP by utilizing the four-item TP scale of Stewart and Barrick (2000). The manager of each team rated team performance using this scale. TP was measured based on four categories (knowledge of tasks, quality of work, quantity of work, and overall per- formance) using a five-point behavior-anchored scale (ranging from 1 = somewhat below the requirements to 5 = consistently exceeds requirements). The Cronbach’s alpha for this construct was .79.

IP. With regard to employee IP, the scale developed by Griffin, Neal, and Parker (2007) was used in our study. The authors divided positive work-role behaviors into three levels, namely, individual task behaviors, team member behaviors, and organization member behav- iors. For the purpose of our research, we only utilized three items from the list of individual

2700 Journal of Management / September 2018

task behaviors that specifically address the IP of team members. We adapted the wording of the selected items to allow team managers to rate the individual task behaviors of each team member. Sample items included, “[The team member] carried out the core parts of his/her job well” and “[The team member] adapted well to changes in core tasks.” We found that the results showed good reliability (α = .84).

OCB. To complement the positive in-role behaviors of team members, we also measured OCB as an outcome. Following Organ (1997), we viewed OCB as a two-dimensional con- struct consisting of helping and conscientiousness. In addition, we utilized a scale proposed by Podsakoff et al. (1990) with team members as referents. Sample items included, “[The team member] is always ready to lend a helping hand to other team members,” and “[The team member] obeys rules and regulations even when no one is watching.” The Cronbach’s alpha for this construct was α = .92.

Control variables. We conducted a multilevel analysis of a single organization in this study. At the individual level, we controlled for the gender and age of employees. At the team level, control variables included job group (i.e., management/administration, sales, or R&D) and team size (Stewart & Barrick, 2000). Initially, we had a wider range of control variables (e.g., demographic characteristics of team managers, job group information of employees). Given our sample size (i.e., 183 responses embedded in 51 teams), we decided to minimize the number of control variables in our model. However, we had particular interest in allow- ing for SMB as an additional control variable. The propositions of implicit leadership theory suggest that respondents are likely to perceive their managers’ behaviors in connection with factors contributing to the group’s performance (Rush, Thomas, & Lord, 1977). In a similar vein, we consider the possibility that team members working under a supportive manager and therefore experiencing a more pleasant work environment may perceive more effort on the part of their team managers and therefore more intense implementation of HPWS. By includ- ing SMB in our model, we demonstrate that (1) TIHR practices and SMB are distinct con- cepts; and (2) after controlling for SMB, TIHR still has significant effects on the intensity of HPWS implementation, work group outcomes, and individual outcomes. In our framework, we utilize supportive leader behaviors developed by Podsakoff, Todor, and Schuler (1983). We selected five items relevant to our purpose and adapted them with team managers as the referents. Sample items included “[My team manager] helps us make working on our tasks more pleasant” and “[My team manager] helps us overcome problems which stop us from carrying out our tasks.” The Cronbach’s alpha value for this construct was α = .91.

Data Aggregation

Data collected from individuals were aggregated to the team level. To investigate the appropriateness of aggregating responses, we computed the appropriate rwg values using a uniform distribution as the null distribution (James, Demaree, & Wolf, 1984) and deter- mined the intraclass correlation coefficients (ICCs) (Bliese, 2000). The rwg value for TIHR is .93, which is above the generally acceptable level of .70 (George, 1990). The ICC(1) and ICC(2) values for TIHR are .25 and .77, respectively (F = 2.96, p < .001). The ICC(1) value exceeded the generally accepted cutoff level of .12 (Glick, 1985). Moreover, the result of

Pak, Kim / HPWS Implementation 2701

the F-test for TIHR was significant, indicating that data aggregation could be justified (Klein & Kozlowski, 2000). Similarly, we calculated rwg , ICC(1), and ICC(2) for HPWSI; the values were .96, .33, and .87, respectively (F = 3.90, p < .001). Hence, data aggregation was justified. The ICC(1) value of .33 for HPWSI indicated that 33% of the variance in HPWSI among team members can be explained by their team membership (Bliese, 2000). Therefore, our preceding discussion on the emergence of team-level HPWS was empirically supported. In addition, we calculated rwg , ICC(1), and ICC(2) for SMB; the values were .92, .30, and .85, respectively (F = 3.41, p < .001). Therefore, data aggregation at the team level was justified.

Analytical Strategy and Validity of Study Variables

We tested the team-level mediation hypothesis with a bootstrapping method suggested by Hayes (2013). It is now recognized that the PROCESS approach developed by Hayes is a method preferred over the four-step mediation procedure of Baron and Kenny (1986). Because PROCESS cannot be used in research of cross-level design (Hayes, 2013), for the remaining two mediation hypotheses, which are multilevel in nature, we utilized the RMediation web application developed by Tofighi and MacKinnon (2011). According to the authors, RMediation provides more accurate results compared with another widely used program, PRODCLIN (distribution of the PRODuct Confidence Limits for INdirect effects), developed by MacKinnon, Fritz, Williams, and Lockwood (2007). In testing our multilevel hypotheses using RMediation, we first conducted hierarchical linear model- ing (HLM). Then, we utilized the results obtained from HLM as starting values for RMediation.

Before testing our hypotheses, we attempted to confirm the convergent and discriminant validity of key study variables by conducting a confirmatory factor analysis (CFA) using AMOS 18. CFA generates several indices that help us to compare the model fit of the hypothesized factor structure with that of alternative ones (Hair, Black, Babin, & Anderson, 2009). In the alternative one-factor model, we included TIHR, HPWSI, HPCF, and SMB. The results were as follows: χ2 = 2,201.33, df = 430, χ2/df = 5.12, incremental fit index (IFI) = .46, Tucker-Lewis index (TLI) = .49, comparative fit index (CFI) = .50, and root mean square error of approximation (RMSEA) = .16. For the two-factor model that treats HPCF as another factor, the results were as follows: χ2 = 1,241.41, df = 401, χ2/df = 3.10, IFI = .61, TLI = .70, CFI = .73, and RMSEA = .11. In the three-factor model, we considered HPWSI as a separate factor. The results were as follows: χ2 = 958.81, df = 390, χ2/df = 2.46, IFI = .73, TLI = .79, CFI = .79, and RMSEA = .10. Finally, our hypothesized four-factor model demonstrated results that satisfied the requirements for cutoff values (χ2 = 580.85, df = 332, χ2/df = 1.75, IFI = .91, TLI = .90, CFI = .92, and RMSEA = .07. According to previous stud- ies, values over .90 are acceptable for IFI, TLI, and CFI, and the value for χ2/df should be below 3 (Browne & Cudeck, 1993; Hall, Snell, & Foust, 1999; Hu & Bentler, 1999; Kline, 2005). In addition, RMSEA must be less than .10 (Steiger, 1990). Hence, we provided evi- dence for discriminant validity among the key study variables, and the analysis also pro- vided evidence that team manager visibility and SMB are distinct concepts that should be treated separately in our model.

For confirmation of convergent validity, the factor loadings of each item exceeded the cutoff value of .50 (Hair et al., 2009; see appendix). Moreover, our analysis showed that the

2702 Journal of Management / September 2018

average variance extracted (AVE) values for key variables ranged from .58 to .77, well above the cutoff value of .50. Similarly, the values for composite reliability (CR), which were .91, .95, .80, and .91 for HPWSI, TIHR, SMB, and HPCF, respectively, were all above the recom- mended value of .70 (Nunnally, 1978). From these analyses, we provided evidence that con- vergent validity was present among the key study variables. Therefore, we decided to test our hypotheses using the models.

Results

In this section, we first describe the characteristics of the data. Then, we test our hypoth- eses. Specifically, we present the results from using PROCESS (Hayes, 2013) for testing of the team-level hypothesis. Then, we show the results of HLM and report the 95% confidence intervals (95% CIs) obtained from RMediation (Tofighi & MacKinnon, 2011) for testing of our cross-level hypotheses.

Descriptive Statistics

The means, standard deviations, Cronbach’s alpha values, and correlations between all variables are presented in Table 1. Consistent with our predictions, significant correlations were generally observed among the study variables. At the team level, SMB was positively related to TIHR (r = .59, p < .05), HPWSI (r = .57, p < .01), and TP (r = .41, p < .01). Also, TIHR was significantly and positively correlated with both HPWSI (r = .56, p < .01) and TP (r = .48, p < .01); HPWSI and TP were also correlated (r = .57, p < .01). In addition, cross- level correlations showed a significant positive relationship between HPCF and HPWSI (r = .22, p < .01) and HPCF and TP (r = .30, p < .01). At the individual level, HPCF was positively correlated with IP (r = .31, p < .01), and OCB (r = .38, p < .01). The relatively high correla- tions between the level 2 variables led us to observe that the variance inflation factor (VIF) fell below the cutoff value (VIF < 10). Cronbach’s alpha values, presented in the table along the diagonal, showed reliability of a generally acceptable level, ranging from .77 to .92.

Team-Level Hypothesis

Table 2 presents the results of the hierarchical regression analyses predicting team perfor- mance. Hypothesis 1 stated that TIHR is positively related to TP and that this relationship is mediated by HPWSI. Our analysis indicated a significant relationship between TIHR and TP (β = .39, p < .01). It is noticeable that the positive effect of SMB on TP (β = .21, p < .01) became nonsignificant when TIHR was included in the model. Thus, this study confirmed that TIHR is a meaningful predictor for TP in our research framework. Also, the results of testing using Model 3 indicated that HPWSI was positively associated with TP (β = .44, p < .01). In sum, our results indicate that a one standard deviation increase in TIHR yields a 5.8% improvement in team performance, while HPWSI enhances it by 6.6%. The change in the beta coefficient representing the effect of TIHR on TP indicated that approximately 33% of this relationship can be explained by HPWSI. Our team-level mediation hypothesis was tested by examining the indirect effects using PROCESS (Hayes, 2013). The results demon- strate that the size of the effect of TIHR on TP via HPWSI is .13 (i.e., a 1.9% improvement in TP for a one standard deviation increase in TIHR through HPWSI), not including zero

Pak, Kim / HPWS Implementation 2703

(95% CI = [.02, .25]; bias-corrected bootstrap confidence intervals based on 10,000 bootstrap samples). Therefore, Hypothesis 1 is supported.

Table 2

Results of Hierarchical Regression Analyses Predicting Team Performance

Variables Model 1 Model 2 Model 3 Model 4

Job group .01 (.08) .12 (.08) .08 (.08) .11 (.08) Team size .07 (.02) .10 (.02) .05 (.02) .07 (.02) SMB .21 (.06)** .11 (.06) .10 (.06) .10 (.06) TIHR .39 (.09)** .26 (.09)** HPWSI .44 (.10)** .39 (.10)** CI indirect effect (95%) [.02, .25] R2 .15 .25 .33 .38 ∆R2 .15 .10 .08 .05 ∆F 12.60** 14.68** 21.13** 9.64**

Note: N = 51. Standardized coefficients and standard errors are shown. SMB = supportive manager behavior; TIHR = team manager’s implementation of espoused HR practices; HPWSI = high-performance work systems intensity; CI = confidence interval. *p < .05. **p < .01.

Table 1

Means, Standard Deviations, Correlations, and Cronbach’s Alpha Values of Study Variables

Variables M SD 1 2 3 4 5 6 7 8 9 10 11

Team level (N = 51) 1. Job group 1.51 0.70 — 2. Team size 7.82 4.05 .00 — 3. SMB 3.60 0.69 −.16 .05 .91 4. TIHR 3.32 0.60 −.21 −.08 .59* .91 5. HPWSI 3.35 0.67 −.11 .03 .57** .56** .87 6. TP 3.81 0.57 .01 .07 .41** .48** .57** .79 Individual level (N = 183) 7. Gender 1.40 0.49 .20 −.19 −.29 −.38** −.39** −.41** — 8. Age 30.67 4.17 .29* .10* −.24 −.17 .17 .08 −.47** — 9. HPCF 3.33 0.57 .30* −.37* −.11 .15 .22** .30** .03 .06 .77 10. IP 3.67 0.64 −.33* .29* .15 .26 .24 .22 −.11 .07 .31** .84 11. OCB 3.79 0.65 −.35* .27 .24 .23 .21 .12 −.08 .14 .38** .49** .92

Note: Cronbach’s alpha values are in italics along the diagonal. SMB = supportive manager behavior; TIHR = team manager’s implementation of espoused HR practices; HPWSI = high-performance work systems intensity; TP = team performance; HPCF = HRM-induced psychological contract fulfillment; IP = in-role performance; OCB = organizational citizenship behavior. *p < .05. **p < .01.

2704 Journal of Management / September 2018

Cross-Level Hypotheses

In our study, Hypothesis 2 stated that TIHR is positively related to employee HPCF and that this relationship is mediated by HPWSI at the team level. Table 3 presents the results of the HLM analysis. Before testing the hypothesis, we initially examined the null model to test for between-group variance (Hox, 2010; Mathieu & Taylor, 2007). We found a significant chi-squared value for HPCF (χ2 = 129.36, p < .01). In addition, an ICC estimated in the null model showed that 24% of variance could potentially be explained by a level 2 predictor (Bryk & Raudenbush, 1992). The results implied that significant variance existed between groups; therefore, we controlled for this variance when testing our hypothesis using two- level HLM.

The results suggested that after controlling for the control variables, TIHR had a significant and positive relationship with HPCF (γ = .43, p < .01), as shown in Model 3. This represents a 7.4% improvement in HPCF when TIHR increases by one standard deviation. Model 3 indi- cated that HPWSI is positively associated with HPCF (γ = .27, p < .01), which is equivalent to a 4.6% enhancement for a one standard deviation increase in HPWSI. From the change in the beta coefficient representing the effect of TIHR on HPCF, we see that about 23% of the rela- tionship can be explained by HPWSI. In our RMediation analysis using Model 4 (Tofighi & MacKinnon, 2011), the 95% CI for the distribution of the product of the coefficients was [.10, .35] with random indirect effects of .24 (i.e., a 4.1% improvement in HPCF via HPWSI for a one standard deviation increase in TIHR). Therefore, Hypothesis 2 is supported.

Table 3

Results of HLM Predicting HRM-Induced Psychological Contract Fulfillment

Variables Null Model Model 1 Model 2 Model 3 Model 4

Intercept 3.33 (.05)** 3.33 (.05)** 3.30 (.04)** 3.33 (.05)** 3.33 (.04)** Job group .12 (.08) .15 (.08) .13 (.07) .13 (.02) Team size −.02 (.08) −.01 (.01) −.02 (.01) −.01 (.01) Gender .10 (.14) .10 (.14) .10 (.14) .10 (.14) Age −.01 (.01) −.01 (.01) −.01 (.01) −.01 (.01) SMB .14 (.07)* −.16 (.11) −.00 (.08) −.23 (.15) TIHR .43 (.10)** .33 (.13)* HPWSI .27 (.08)** .21 (.11)* σ2 .24 τ00 .08 CI indirect effect (95%) [.10, .35] Pseudo-R2 .20 .48 .39 .51 χ2 129.36** 102.30** 76.98** 85.13** 71.76**

Note: Parameter estimate and standard error are shown. Level 1 (N = 183) variables are group-mean centered. Level 2 (N = 51) variables are grand-mean centered. HLM = hierarchical linear modeling; HRM = human resource management; SMB = supportive manager behavior; TIHR = team manager’s implementation of espoused HR practices; HPWSI = high-performance work systems intensity; CI = confidence interval. σ2 indicates variance in Level 1 residuals. τ00 indicates variance in Level 2 residuals. Pseudo-R2 values are calculated consistent with the protocol in Kreft and De Leeuw (1998). *p < .05. **p < .01.

Pak, Kim / HPWS Implementation 2705

In our study, Hypothesis 3 stated that HPWSI is positively related to both IP and OCB and that these relationships are mediated by HPCF. Tables 4 and 5 show the results of the HLM analysis for the suggested relationships. As before, we examined the null models before test- ing the hypothesis to confirm the significance of between-group variance (Hox, 2010; Mathieu & Taylor, 2007). Our analysis revealed significant chi-squared values for both IP (χ2 = 167.57, p < .01) and OCB (χ2 = 177.24, p < .01). We also calculated the ICCs for each null model; these values were .43 and .49, respectively (Bryk & Raudenbush, 1992). To test our mediation hypotheses, we initially explored the effects of HPWSI on employee out- comes. Model 2 in Table 4 and Model 2 in Table 5 show that HPWSI is positively associated with IP (γ = .26, p < .01) and OCB (γ = .34, p < .01) in both models. That is, a one standard deviation increase in HPWSI yields a 4.5% increase in IP and a 5.8% increase in OCB. Furthermore, the results in Model 3 indicated that HPCF has positive relationships with both IP (γ = .33, p < .01) and OCB (γ = .35, p < .01), which is equivalent to a 5.8% and 6.0% improvement, respectively, for a one standard deviation increase in HPCF. From the change in the beta coefficient representing the effect of HPWSI on IP, we see that approximately 27% of the relationship is explained by HPCF, while around 21% is explained by the effect of HPWS on OCB. As for the indirect effects, Model 4 in Tables 4 and 5 revealed that the 95% CI for the distribution of the product of the coefficients was [.02, .28] with random indirect effects of .10 for IP (i.e., a 1.7% improvement in IP via HPCF for a one standard deviation increase in HPWSI), and [.01, .33] for OCB with β = .22 (i.e., a 3.8% improvement

Table 4

Results of HLM Predicting In-Role Performance

Variables Null Model Model 1 Model 2 Model 3 Model 4

Intercept 3.67 (.07)** 3.67 (.06)** 3.67 (.06)** 3.67 (.06)** 3.67 (.06)** Job group −.06 (.09) −.06 (.08) −.07 (.09) −.06 (.09) Team size .03 (.01) .03 (.01) .03 (.01)* .03 (.01) Gender .03 (.13) .03 (.13) −.00 (.13) −.00 (.13) Age .02 (.01) .02 (.01) .02 (.01)* .02 (.01) SMB .25 (.07)** .11 (.07) .11 (.07) .12 (.07) HPWSI .26 (.07)** .19 (.07)** HPCF .33 (.09)** .32 (.09)** σ2 .24 τ00 .18 CI indirect effect (95%) [.02, .28] Pseudo-R2 .26 .41 .50 .47 χ2 167.57** 147.57** 133.89** 109.13** 96.10**

Note: Parameter estimates and standard errors are shown. Level 1 (N = 183) variables are group-mean centered. Level 2 (N = 51) variables are grand-mean centered. HLM = hierarchical linear modeling; SMB = supportive manager behavior; HPWSI = high-performance work systems intensity; HPCF = human resource management (HRM)–induced psychological contract fulfillment; CI = confidence interval. σ2 indicates variance in Level 1 residuals. τ00 indicates variance in Level 2 residuals. Pseudo-R2 values are calculated consistent with the protocol in Kreft and De Leeuw (1998). *p < .05. **p < .01.

2706 Journal of Management / September 2018

in OCB via HPCF for a one standard deviation increase in HPWSI). Since both confidence intervals do not include zero, Hypothesis 3 is supported.

Additionally, we examined whether aggregated individual IP and OCB values are signifi- cantly associated with team performance. The more employees perform better in their jobs and show altruistic and community-centered behaviors in their work groups, the higher team performance will be. Consistent with our prediction, the effects of IP (β = .35, p < .05) and OCB (β = .23, p < .05) aggregated to the team level on team performance are found to be significant after controlling for gender, age, job group, and team size (R2 = .46; F = 31.68, p < .001). That is, a one standard deviation increase in IP yields a 5.2% improvement in team performance, while OCB enhances team performance by 3.4%.

Discussion

Although the SHRM literature has recognized within-organization variability over the implementation phase, commensurate attention has not yet been drawn to examining the source of it (Wright & Nishii, 2013). Studies of SHRM have traditionally focused on the mechanism through which a certain set of HR practices (i.e., HPWS) influences perfor- mance measures, treating the HRM system as a matter of adoption. In this article, we put forward that if implementation intensity varies within the organization, creating inconsis- tency in employee attitudes and subsequent performance, investigating the source of this

Table 5

Results of HLM Predicting Organizational Citizenship Behavior

Variables Null Model Model 1 Model 2 Model 3 Model 4

Intercept 3.79 (.07)** 3.79 (.06)** 3.79 (.06)** 3.79 (.06)** 3.79 (.06)** Job group −.13 (.11) −.13 (.09) −.13 (.11) −.13 (.09) Team size .04 (.01) .04 (.01)** .04 (.01)** .04 (.01)** Gender −.02 (.10) −.02 (.10) −.05 (.11) −.05 (.11) Age −.00 (.01) −.00 (.01) −.00 (.01) .00 (.01) SMB .21 (.08)** .21 (.08)* .21 (.08)* .02 (.07) HPWSI .34 (.09)** .27 (.09)** HPCF .35 (.08)** .35 (.08)** σ2 .22 τ00 .21 CI indirect effect (95%) [.01, .33] Pseudo-R2 .19 .50 .43 .56 χ2 177.24** 157.24** 136.51** 126.51** 88.19**

Note: Parameter estimates and standard errors are shown. Level 1 (N = 183) variables are group-mean centered. Level 2 (N = 51) variables are grand-mean centered. HLM = hierarchical linear modeling; SMB = supportive manager behavior; HPWSI = high-performance work systems intensity; HPCF = human resource management (HRM)–induced psychological contract fulfillment; CI = confidence interval. σ2 indicates variance in Level 1 residuals. τ00 indicates variance in Level 2 residuals. Pseudo-R2 values are calculated consistent with the protocol in Kreft and De Leeuw (1998). *p < .05. **p < .01.

Pak, Kim / HPWS Implementation 2707

variation should be a worthwhile pursuit. Therefore, we examine a potential antecedent of the HPWS-performance link.

Specifically, we discussed the important role of team managers in enforcing HR policies in their work groups, since HPWS implementation is primarily a team phenomenon (Guest & Bos-Nehles, 2013). Our multilevel framework showed that a team manager’s implementa- tion behavior affects both work group and employee outcomes by enhancing the team mem- bers’ shared understanding of actual HR experiences and their psychological climate. Our analysis demonstrated that HPWSI may be meaningful at the team level since aggregation is justified. This result supports the social interaction perspective (Weick, 1995), which sug- gests that shared HR experiences are created as team members share similar work contexts and interact frequently within their work groups. Our analysis also confirmed that HPWS intensity is significantly influenced by team managers’ enforcement of intended HR prac- tices. This result is consistent with research on the strength of the HRM system (Bowen & Ostroff, 2004) and the situational perspective (Schneider, 1990). To create a strong work environment, HR messages must be delivered to individuals in a distinctive, consistent, and consensual manner. In addition, HR experiences are influenced by adjacent filters, that is, the team manager’s actions.

We regarded HPWS as a type of exchange agreement between an organization and its employees (Guzzo & Noonan, 1994; Sonnenberg, 2006). In support of this position, we dem- onstrated that team members’ shared understanding of HPWS intensity is linked to fulfill- ment of individual psychological contracts. This result showed that team-level implementation intensity of HPWS mediates the relationship between team managers’ enforcement quality and psychological contract fulfillment in the HRM domain. In this work, we drew on the social exchange perspective (Blau, 1964) to explicate why contract fulfillment may induce positive employee attitudes and behaviors. When employee contributions are properly recog- nized with commensurate rewards, and when the organization provides appropriate means through which employees may be successful at their jobs (e.g., training), employees feel obliged to contribute both by engaging themselves more deeply in their jobs (Liao, Toya, Lepak, & Hong, 2009; Snape & Redman, 2010) and by exceeding in-role expectations (Gong, Chang, & Cheung, 2010). The results of our empirical investigation supported this perspective. In addition, we found that team-level HPWS intensity and in-role performance are positively associated. This result is consistent with the AMO framework (Appelbaum et al., 2000; Delery & Shaw, 2001; Gardner et al., 2011), which asserts that HPWS improves employee attitudes and behaviors by enhancing human capital, increasing motivation of employees, and giving them opportunity to contribute.

Our major contribution to the SHRM literature is to identify antecedents of the HPWS- performance relationship. So far, HPWS research has focused on unlocking the black box that leads to expected performance outcomes (e.g., Boxall et al., 2011; Zacharatos et al., 2005), while studies examining the role of line managers have created a separate research stream (e.g., Gilbert et al., 2011; Purcell & Hutchinson, 2007; Ryu & Kim, 2013; Sterling & Boxall, 2013). In this study, we attempt to integrate the findings of studies on line managers with those on the HRM-performance relationship, identifying a primary factor that contrib- utes to HPWS intensity within the organization and its effects during the implementation phase. In doing so, we proposed the concept: team managers’ implementation of espoused HR practices. Our examination of TIHR (1) recognizes team managers as agents of the

2708 Journal of Management / September 2018

organization with the responsibility of enforcing HR policies in their work groups (Purcell & Hutchinson, 2007) and (2) proposes that the ways in which team managers implement HR policies are primary interpretive mechanisms through which employees make sense of actual HR experiences. Several recent studies have acknowledged that the role of work group man- agers determines the effectiveness of HRM (e.g., Gilbert et al., 2011; Ryu & Kim, 2013; Sterling & Boxall, 2013). However, behaviors of the team manager, who is comparable to a visible supervisor (Bowen & Ostroff, 2004), in enforcing HR practices in the intended man- ner during the implementation phase have rarely been conceptualized in empirical settings. Our study concludes that the degree to which a team manager acts as a deliverer and an advocate of the espoused HR policies in the work group is significantly related to HPWS intensity at the team level and to subsequent work group and individual outcomes.

Practitioners can benefit from the findings of this study in several aspects. First, the fidelity of team managers in enforcing HR policies in their work groups was proposed to be a major source of differences in HPWS perception between team members. Therefore, companies should pay more careful attention to HR-related training offered to team man- agers. HR knowledge should be sufficiently transferred to team managers to ensure maxi- mum benefit. Training programs can be designed to provide team managers with directions on how to implement HR policies, and these programs should be regularly available to leaders to aid them in important HR decision making (for example, in an appraisal period). Moreover, standard procedures can be summarized in guidelines to which team managers can refer when necessary. Providing incentives to team managers to implement the espoused HR practices can complement HR-related training because knowledge and action are often decoupled during the implementation process. Performance appraisals in relation to people development and ensuring compliance with HR policies set by the organization can be helpful. In addition, the HR department can focus on monitoring and coordination, thus facilitating the development of a climate for implementation. Team managers should become effective communicators in enforcing HR policies in their work groups because individual performance is partly explained by HPCF. In so doing, they should appropri- ately justify HR decisions (e.g., promotion) in terms of either ability or failure to meet certain requirements.

Several limitations of this study provide promising avenues for future research. First, the data used in this study were gathered through a web-based survey. Despite our ensuring the anonymity of the responses, the response rate was low (15.2%). Although our use of a web- based survey may have induced self-selection bias, the response rate is comparable with that of previous studies (Becker & Huselid, 1998; Guthrie, 2001; Zacharatos et al., 2005). The cross-sectional nature of our data also limits our results in terms of causality among study variables. Thus, future studies can utilize a longitudinal design for a higher response rate with a larger sample.

Additionally, our survey data were gathered from a single Korean cosmetics company. In the course of our research, we ensured that the HRM system within the organization con- tained elements of HPWS and that our chosen variables were suitable to the organizational context not only by conducting multiple interviews, but also by reviewing the company’s internal archives. However, concerns remain about the generalizability of our findings. Therefore, similar studies should be conducted in different contexts to test the findings of the study.

Pak, Kim / HPWS Implementation 2709

Our primary focus was the factors that contribute to variance in the intensity of implemen- tation of HPWS in the workplace in terms of the team manager’s compliance with espoused HR practices. In the current study, team-level HPWS was highlighted, and we proposed that the enforcement of HR practices is mainly a work group phenomenon (Guest & Bos-Nehles, 2013). However, the current study did not explore how team-level HPWS is linked to team performance. Although several studies have examined the mechanisms underlying the HRM- performance relationship (cf. Jiang et al., 2013), we know little about the unique team pro- cesses that connect HPWS at the team level and subsequent outcomes. Therefore, future research is needed to explore this suggested area of research.

To the best of our knowledge, this study is the first attempt in the HPWS literature to explore the antecedents of team-level HPWSI, which influences team performance. In doing so, we specifically examined the HR implementation role of team managers. Although this study treats the HPWS-performance link as fundamental in extending the findings of the cur- rent literature, our results confirmed that approximately 66% of implementation behavior is not moderated by HPWS intensity in predicting team performance. Thus, we acknowledge that there may be other meaningful mediators affecting team performance over the course of the implementation phase. For instance, the enforcement of intended HR practices may strengthen overall perceptions of individuals’ roles and expectations of team outcomes (Biddle, 1979; Hill & McShane, 2006), thereby contributing to team performance. Also, it is possible that team managers’ implementation behavior may also affect such team interaction processes as exchanging information, learning, motivating, and negotiating among individu- als on the team (Tierney & Farmer, 2002), which may positively influence the effectiveness of the team. Thus, we expect that future research in this area will deepen our understanding of team managers’ HR responsibility in improving performance of teams.

Moreover, we also acknowledge that a team manager is not the sole deliverer of HR prac- tices. Exploring the roles of the HR department and commitment of top management will also deepen our understanding of the degree to which HPWS intensity varies and the effec- tiveness of enforcement of HR policies in organizations (Pak & Chung, 2013). We assert that additional factors should be examined in relation to the intensity of HPWS implementation. For example, employee-rated HPWS may be a function of HRM strength (Bowen & Ostroff, 2004). When a certain HRM system is operated in a distinctive, consistent, and consensual manner, employees are likely to perceive a high, rather than low, level of HPWS efficacy. In a similar vein, the attribution patterns of employees relative to the intentions of management regarding HRM system adoption may affect employee understanding of HPWS (Nishii, Lepak, & Schneider, 2008). Employees may report higher HPWS efficacy when they attri- bute the adoption of HRM systems to motivation enhancement, skill development, and opportunity to contribute (Appelbaum et al., 2000; Delery & Shaw, 2001; Gardner et al., 2011) instead of negative sense making (e.g., work pressure, exploitation) (Danford, Richardson, Pulignano, & Stewart, 2008).

2710 Journal of Management / September 2018

Appendix A

Validity of Study Variables

Survey Items Factor

Loadingsa AVE CR

High-performance work systems intensity 1. I think that selection emphasizes an individual’s ability to

collaborate and work in teams. .81 .58 .91

2. I think that selection is based upon overall fit with the company.

.95

3. I regularly go through training programs. .80 4. I have been provided with extensive training programs. .52 5. I have few opportunities for upward mobility. (R) .59 6. I think that promotion in this organization is based on

seniority. (R) .55

7. My performance appraisals include developmental feedback.

.73

8. I think that employee appraisals emphasize long-term and group-based achievement.

.81

9. My performance appraisals are based on objective quantifiable results.

.81

10. There is a close tie or matching of pay to individual/group performance.

.71

11. My incentives are based on team performance. .65 12. There is a wide range in pay within the same job grade

depending on performance. .52

13. I am often asked by my team manager to participate in decision-making.

.50

14. I am allowed to make decisions. .55 15. I am provided the opportunity to suggest improvements in

the way things are done. .71

Team manager’s implementation of espoused HR practices 1. I have frequently witnessed that my team manager acts as an

enthusiastic advocate of the HR policies of our company. .89 .70 .95

2. Many times I have observed that my team manager clearly communicates HR-related initiatives or changes in our work group.

.80

3. I have frequently observed that my team manager implements HR practices, strictly following corporate HR processes.

.81

4. Many times I have witnessed that my team manager puts a strong emphasis on our participating in HR programs (e.g., training, culture-building activities), even when we are busy working.

.77

5. Many times I have witnessed that my team manager emphasizes following HR processes and gives us clear guidance over the course of the project.

.90

Supportive manager behaviors .73 .80 1. My team manager makes working on our tasks more pleasant. .78 2. My team manager helps us overcome problems that stop us

from carrying out our tasks. .83

(continued)

Pak, Kim / HPWS Implementation 2711

Appendix B

Survey Items for Performance Measures

In-Role Performance

1. The team member carried out the core parts of his/her job well. 2. The team member adapted well to changes in core tasks. 3. The team member came up with ideas to improve the way in which his/her core tasks are done.

Organizational Citizenship Behavior

1. The team member is always ready to lend a helping hand to other team members. 2. The team member obeys rules and regulations even when no one is watching. 3. The team member helps others who have heavy workloads. 4. The team member is one of my most conscientious employees.

Team Performance

Please indicate your team’s performance for each item below (1 = somewhat below require- ments, 5 = consistently exceeds requirements).

1. Knowledge of tasks 2. Quality of work 3. Quantity of work 4. Overall performance

Survey Items Factor

Loadingsa AVE CR

3. My team manager does things to make it pleasant to be a member of the group.

.90

4. My team manager is willing to take initiative in the group. .69 5. My team manager keeps the group working together as a

team. .93

HRM-induced psychological contract fulfillment 1. In relation to compensation, advancement, training,

participation, etc., the company hasn’t always lived up to its end of the bargain (R).

.61 .77 .91

2. Considering my HR experiences, I would say that my company has kept its promises to me.

.84

3. Thinking of the HR practices that I have experienced, I would say that my company has often broken promises to me (R).

.70

4. In relation to HR practices, my company fulfills its obligations to me.

.88

Note: N = 183. AVE = average variance extracted; CR = composite reliability; R = reverse scored; HR = human resources; HRM = human resource management. aFactor loadings are derived from confirmatory factor analysis (CFA). In AMOS 18, standardized regression coefficients are equivalent to factor loadings.

Appendix A (continued)

2712 Journal of Management / September 2018

References Anderson, K. K., Cooper, B. K., & Zhu, C. J. 2007. The effect of SHRM practices on perceived firm financial

performance: Some initial evidence from Australia. Asia Pacific Journal of Human Resources, 45: 168-179. Appelbaum, E., Bailey, T., Berg, P., & Kalleberg, A. 2000. Manufacturing advantage: Why high-performance work

systems pay off. Ithaca, NY: ILR Press. Bae, J., & Lawler, J. J. 2000. Organizational and HRM strategies in Korea: Impact on firm performance in an emerg-

ing economy. Academy of Management Journal, 43: 502-517. Baron, R. M., & Kenny, D. A. 1986. The moderator-mediator variable distinction in social psychological research—

Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51(6): 1173-1182.

Bartel, A. P. 2004. Human resource management and organizational performance: Evidence from retail banking. Industrial & Labor Relations Review, 57(2): 181-203.

Becker, B. E., & Huselid, M. A. 1998. High performance work systems and firm performance: A synthesis of research and managerial implications. Research in Personnel and Human Resources Management, 16: 53-101.

Biddle, B. J. 1979. Role theory: Expectations, identities, and behaviors. New York: Academic Press. Blau, P. M. 1964. Exchange and power in social life. New York: John Wiley. Bliese, P. D. 2000. Within-group agreement, non-independence, and reliability: Implications for data aggregation

and analysis. In K. J. Klein & S. W. Kozlowski (Eds.), Multilevel theory, research, and methods in organiza- tions: Foundations, extensions and new directions: 349-381. San Francisco, CA: Jossey-Bass.

Boon, C., Den Hartog, D. N., Boselie, P., & Paauwe, J. 2011. The relationship between perceptions of HR practices and employee outcomes: Examining the role of person-organisation and person-job fit. International Journal of Human Resource Management, 22(1): 138-162.

Bowen, E., & Ostroff, C. 2004. Understanding HRM–firm performance linkages: The role of the “strength” of the HRM system. Academy of Management Review, 29: 203-221.

Boxall, P., Ang, S. H., & Bartram, T. 2011. Analysing the “black box” of HRM: Uncovering HR goals, mediators, and outcomes in a standardized service environment. Journal of Management Studies, 48: 1504-1532.

Browne, M. W., & Cudeck, R. 1993. Alternative ways of assessing model fit. In K. A. Bollen & J. S. Long (Eds.), Testing structural equation models: 136-162. Thousand Oaks, CA: SAGE Publications.

Bryk, A. S., & Raudenbush, S. W. 1992. Hierarchical linear models: Applications and data analysis methods. Newbury Park, CA: SAGE Publications.

Chaiken, S., Wood, W., & Eagly, A. H. 1996. Principles of persuasion. In E. T. Higgins & A. W. Kruglanski (Eds.), Social psychology: Handbook of basic principles: 702-744. New York: Guilford Press.

Chuang, C., & Liao, H. 2010. Strategic human resource management in service context: Taking care of business by taking care of employees and customers. Personnel Psychology, 63: 153-196.

Collins, C. J., & Smith, K. G. 2006. Knowledge exchange and combination: The role of human resource practices in the performance of high-technology firms. Academy of Management Journal, 49: 544-560.

Combs, J., Liu, Y., Hall, A., & Ketchen, D. 2006. How much do high performance work practices matter? A meta analysis of their effects on organizational performance. Personnel Psychology, 59: 501-528.

Conway, N., & Coyle-Shapiro, J. A. M. 2012. The reciprocal relationship between psychological contract ful- fillment and employee performance and the moderating role of perceived organizational support and tenure. Journal of Occupational and Organizational Psychology, 85(2): 277-299.

Coyle-Shapiro, J. A. M., & Kessler, I. 2000. Consequences of the psychological contract for the employment rela- tionship: A large scale survey. Journal of Management Studies, 37(7): 903-930.

Cunningham, I., & Hyman, J. 1999. Devolving human resource responsibilities to the line: Beginning of the end or a new beginning for personnel? Personnel Review, 28: 9-27.

Danford, A., Richardson, M., Pulignano, V., & Stewart, P. 2008. Lean production and quality of working life on the shop-floor: The experience of British and Italian car workers. In V. Pulignano, P. Stewart, A. Danford, & M. Richardson (Eds.), Flexibility at work: Critical developments in the international automobile industry: 45-82. Basingstoke, UK: Palgrave Macmillan.

Datta, D. K., Guthrie, J. P., & Wright, P. M. 2005. Human resource management and labor productivity: Does industry matter? Academy of Management Journal, 48: 135-145.

Delery, J. E., & Shaw, J. D. 2001. The strategic management of people in work organizations: Synthesis and exten- sion. In G. R. Ferris (Ed.), Research in personnel and human resource management: 165-197. Greenwich, CT: JAI.

Pak, Kim / HPWS Implementation 2713

Den Hartog, D. N., Boselie, P., & Paauwe, J. 2004. Performance management: A model and research agenda. Applied Psychology, 53(4): 556-569.

Dragoni, L. 2005. Understanding the emergence of state goal orientation in organizational work groups: The role of leadership and multilevel climate perceptions. Journal of Applied Psychology, 90: 1084-1095.

Ehrnrooth, M., & Bjorkman, I. 2012. An integrative HRM process theorization: Beyond signaling effects and mutual gains. Journal of Management Studies, 49: 1109-1135.

Fu, N., Flood, P. C., Bosak, J., Morris, T., & O’Regan, P. 2013. Exploring the performance effect of high perfor- mance work system on service supply chain in professional service firms. Supply Chain Management: An International Journal, 18: 292-307.

Gardner, T. M., Wright, P. M., & Moynihan, L. M. 2011. The impact of motivation, empowerment, and skill- enhancing practices on aggregate voluntary turnover: The mediating effect of collective affective commitment. Personnel Psychology, 64(2): 315-350.

George, J. M. 1990. Personality, affect and behavior in groups. Journal of Applied Psychology, 75: 107-116. Gilbert, C., De Winnea, S., & Selsa, L. 2011. The influence of line managers and HR department on employee’s

affective commitment. International Journal of Human Resource Management, 22(8): 1618-1637. Glick, W. H. 1985. Conceptualizing and measuring organizational and psychological climate: Pitfalls in multilevel

research. Academy of Management Review, 10: 601-616. Gong, Y., Chang, S., & Cheung, S. 2010. High performance work system and collective OCB: A collective social

exchange perspective. Human Resource Management Journal, 20: 119-137. Gouldner, A. W. 1960. The norm of reciprocity: A preliminary statement. American Sociological Review, 25: 161-

178. Griffin, M. A., Neal, A., & Parker, S. K. 2007. A new model of work role performance: Positive behavior in uncer-

tain and interdependent contexts. Academy of Management Journal, 5: 327- 347. Guest, D. E., & Bos-Nehles, A. C. 2013. HRM and performance: The role of effective implementation. In D.

E. Guest, J. Paauwe, & P. Wright (Eds.), HRM and performance: Achievements and challenges: 79-96. Chichester, UK: Wiley-Blackwell.

Guest, D. E., & Conway, N. 2000. The psychological contract in the public sector: The results of the 2000 CIPD survey of the employment relationship. London: CIPD Publishing.

Guest, D. E., & Conway, N. 2004. Employee well-being and the psychological contract. London: CIPD Publishing. Guthrie, J. P. 2001. High-involvement work practices, turnover, and productivity: Evidence from New Zealand.

Academy of Management Journal, 44: 180-190. Guzzo, R. A., & Noonan, K. A. 1994. Human resource practices as communications and the psychological contract.

Human Resource Management, 33: 447-462. Hair, J. F., Jr., Black, W. C., Babin, B. J., & Anderson, R. E. 2009. Multivariate data analysis, 7th ed. Englewood

Cliffs, NJ: Prentice Hall. Hall, R. L., Snell, A. F., & Foust, M. S. 1999. Item parceling strategies in SEM: Investigating the subtle effects of

unmodeled secondary constructs. Organizational Research Methods, 2(3): 233-256. Hayes, A. F. 2013. Introduction to mediation, moderation, and conditional process analysis: A regression-based

approach. New York: Guilford Press. Henderson, D., Wayne, S., Shore, L., Bommer, W., & Tetrick, L. 2008. Leader–member exchange, differentiation, and

psychological contract fulfillment: A multilevel examination. Journal of Applied Psychology, 93(6): 1208-1219. Hill, C. W. L., & McShane, S. L. 2006. Principles of management. London: McGraw-Hill/Irwin. Hox, J. J. 2010. Multilevel analysis: Techniques and applications. New York: Routledge. Hu, L., & Bentler, P. M. 1999. Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria

versus new alternatives. Structural Equation Modeling, 6(1): 1-55. 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. 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. Jiang, K., Lepak, D. P., Hu, J., & Baer, J. C. 2012. How does human resource management influence organizational

outcomes? A meta-analytic investigation of mediating mechanisms. Academy of Management Journal, 55(6): 1264-1294.

Jiang, K., Takeuchi, R., & Lepak, D. P. 2013. Where do we go from here? New perspectives on the black box in strategic human resource management research. Journal of Management Studies, 50: 1448-1480.

2714 Journal of Management / September 2018

Klein, K. J., & Kozlowski, S. W. J. 2000. From micro to meso: Critical steps in conceptualizing and conducting multilevel research. Organizational Research Methods, 3: 211-236.

Kline, R. B. 2005. Principles and practice of structural equation modeling. New York: Guilford Press. 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. J. Kozlowski (Eds.), Multilevel theory, research, and methods in organizations: Foundations, extensions, and new directions: 3-90. San Francisco, CA: Jossey-Bass.

Kreft, I., & De Leeuw, J. 1998. Introducing multilevel modeling. London: Sage. Lambert, L. S. 2011. Promised and delivered inducements and contributions: An integrated view of psychological

contract appraisal. Journal of Applied Psychology, 96: 695-712. Lee, H. W., Pak, J., & Kim, S. 2014. HPWS and team performance: The roles of team reflexivity and outcome

interdependence. Paper presented at the annual meeting of the Academy of Management, Philadelphia, PA.

Liao, H., Toya, K. T., Lepak, D. P., & Hong, Y. 2009. Do they see eye to eye? Management and employee per- spectives of high-performance work systems and influence processes on service quality. Journal of Applied Psychology, 94: 371-391.

MacKinnon, D. P., Fritz, M. S., Williams, J., & Lockwood, C. M. 2007. Distribution of the product confidence limits for the indirect effect: Program PRODCLIN. Behavior Research Methods, 39: 384-389.

Mathieu, J. E., & Taylor, S. R. 2007. A framework for testing meso-mediational relationships in organizational behavior. Journal of Organizational Behavior, 28(2): 141-172.

Mathis, R. L., & J. H. Jackson. 1985. Personnel/human resource management. New York: West Publishing. Messersmith, J. G., Patel, P. C., Lepak, D. P., & Gould-Williams, J. 2011. Unlocking the black box: Exploring the

link between high-performance work systems and performance. Journal of Applied Psychology, 96: 1105- 1118.

McDermott, A. M., Conway, E., Rousseau, D. M., & Flood, P. C. 2013. Promoting effective psychological contracts through leadership: The missing link between HR strategy and performance. Human Resource Management, 52(2): 289-310.

McGuire, W. J. 1972. Attitude change: The information processing paradigm. In C. G. McClintock (Ed.), Experimental social psychology: 108-141. New York: Holt, Rinehart & Winston.

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.

Nishii, L. H., Lepak, D. P., & Schneider, B. 2008. Employee attributions of the “why” of HR practices: Their effects on employee attitudes and behaviors, and customer satisfaction. Personnel Psychology, 61: 503-545.

Nishii, L. H., & Wright, P. M. 2007. Variability within organizations: Implications for strategic human manage- ment (CAHRS Working Paper #07−02). Ithaca, NY: Cornell University. Retrieved from http://digitalcommons .ilr.cornell.edu/cahrswp/467

Nunnally, J. C. 1978. Psychometric theory. New York: McGraw-Hill. Organ, D. W. 1988. Organizational citizenship behavior: The good soldier syndrome. Lexington, MA: Lexington

Books. Organ, D. W. 1997. Organizational citizenship behavior: It’s construct cleanup time. Human Performance, 10(2):

85-97. Paauwe, J., Wright, P., & Guest, D. 2013. HRM and performance: What do we know and where should we go? In

D. E. Guest, J. Paauwe, & P. M. Wright (Eds), HRM and performance: Achievements and challenges: 1-13. Chichester, UK: Wiley.

Pak, J., & Chung, G. H. 2013. Top management commitment to HRM, visibilities, and HRM gap: A qualitative approach. Paper presented at the annual meeting of the Academy of Management, Orlando, FL.

Pak, J., Kim, S., & Li, L. Z. 2015. The effects of an innovation-inducing HRM system on the innovation perfor- mance of teams and the mediating effects of team creative processes. Korean Journal of Management, 23(2): 131-157.

Podsakoff, P. M., MacKenzie, S. B., Moorman, R. H., & Fetter, R. 1990. Transformational leader behaviors and their effects on followers’ trust in leader, satisfaction, and organizational citizenship behavior. Leadership Quarterly, 1(2): 107-142.

Podsakoff, P. M., Todor, W. D., & Schuler, R. S. 1983. Leader expertise as a moderator of the effects of instrumental and supportive leader behaviors. Journal of Management, 9(2): 173-185.

Pak, Kim / HPWS Implementation 2715

Poole, M., & Jenkins, G. 1997. Responsibilities for human resource management practices in the modern enterprise: Evidence from Britain. Personnel Review, 26: 333-356.

Purcell, J., & Hutchinson, S. 2007. Front–line managers as agents in the HRM–performance causal chain: Theory, analysis and evidence. Human Resource Management Journal, 17: 3-20.

Ross, L., & Nisbett, R. E. 1991. The person and the situation: Perspectives of social psychology. Philadelphia, PA: Temple University Press.

Rousseau, D. M. 1990. New hire perceptions of their own and their employer’s obligations: A study of psychologi- cal contracts. Journal of Organizational Behavior, 11: 389-400.

Rousseau, D. M. 1995. Psychological contracts in organizations: Understanding written and unwritten agreements. Thousand Oaks, CA: SAGE Publications.

Rush, M. C., Thomas, J. C., & Lord, R. G. 1977. Implicit leadership theory: A potential threat to the internal validity of leader behavior questionnaires. Organizational Behavior and Human Performance, 20(1): 93-110.

Ryu, S., & Kim, S. 2013. First–line managers’ HR involvement and HR effectiveness: The case of South Korea. Human Resource Management, 52(6): 947-966.

Schneider, B. 1990. The climate for service: An application of the climate construct. In B. Schneider (Ed.), Organizational climate and culture: 383-412. San Francisco, CA: Jossey-Bass.

Schuler, R. 1989. Strategic human resource management and industrial relations. Human Relations, 42: 157-184. Schuler, R. S., & Jackson, S. E. 1987. Linking competitive strategies with human resource management practices.

Academy of Management Executive, 1: 207-219. Schuler, R. S., & MacMillan, I. C. 1984. Gaining competitive advantage through human resource management

practices. Human Resource Management, 23(2): 241-256. Shore, L. M., & Tetrick, L. E. 1994. The psychological contract as an explanatory framework in the employment

relationship. In C. L. Cooper & D. M. Rousseau (Eds.), Trends in organizational behavior, vol. 1: 91-109. New York: John Wiley and Sons.

Snape, E., & Redman, T. 2010. HRM practices, organizational citizenship behaviour, and performance: A multi- level analysis. Journal of Management Studies, 47: 1219-1247.

Sonnenberg, M. 2006. The signaling effect of HRM on psychological contracts of employees: A multi–level perspec- tive. Unpublished doctoral dissertation, Erasmus University, Rotterdam, the Netherlands.

Steiger, J. H. 1990. Structural model evaluation and modification: An interval estimation approach. Multivariate Behavioral Research, 25(2): 173-180.

Sterling, A., & Boxall, P. 2013. Lean production, employee learning and workplace outcomes: A case analysis through the ability-motivation-opportunity framework. Human Resource Management Journal, 23(3): 227-240.

Stewart, G. L., & Barrick, M. R. 2000. Team structure and performance: Assessing the mediating role of intrateam process and the moderating role of task type. Academy of Management Journal, 43: 135-148.

Subramony, M. 2009. A meta-analytic investigation of the relationship between HRM bundles and firm perfor- mance. Human Resource Management, 48: 745-768.

Sun, L., Aryee, S., & Law, K. S. 2007. High-performance human resource practices, citizenship behavior, and orga- nizational performance: A relational perspective. Academy of Management Journal, 50: 558-577.

Takeuchi, R., Lepak, D. P., Wang, H., & Takeuchi, K. 2007. An empirical examination of the mechanisms medi- ating between high-performance work systems and the performance of Japanese organizations. Journal of Applied Psychology, 92: 1069-1083.

Tierney, P., & Farmer, S. M. 2002. Creative self-efficacy: Its potential antecedents and relationship to creative performance. Academy of Management Journal, 45: 1137-1148.

Tofighi, D., & MacKinnon, D. P. 2011. RMediation: An R package for mediation analysis confidence intervals. Behavior Research Methods, 43: 692-700.

Weick, K. E. 1995. Sensemaking in organizations. Thousands Oaks, CA: SAGE Publications. Wright, P., & Nishii, L. 2013. Strategic HRM and organizational behaviour: Integrating multiple levels of analysis.

In D. E. Guest, J. Paauwe, & P. M. Wright (Eds), HRM and performance: Achievements and challenges: 97- 110. Chichester, UK: Wiley.

Zacharatos, A., Barling, J., & Iverson, R. D. 2005. High-performance work systems and occupational safety. Journal of Applied Psychology, 90: 77-93.