Cultures have consequences: A configural approach to leadership across two cultures

CULTURES HAVE CONSEQUENCES: A CONFIGURAL APPROACH TO LEADERSHIP ACROSS TWO CULTURES

KYOOTAI LEE Sogang University Graduate School of Management of Technology Sinsu-Dong, Mapo-Gu, Seoul, Korea Email: [email protected] TERRI A. SCANDURA University of Miami School of Business Administration Department of Management Coral Gables, FL 33124 Email: [email protected] MONICA M. SHARIF University of Miami School of Business Administration Department of Management Coral Gables, FL 33124 e-mail: [email protected]

In Press at The Leadership Quarterly

Acknowledgement: This work was supported by the Sogang University Research Grant of 2013(201310011.01). This paper was made possible by the University of Miami Center of International Business Education & Research by providing funding from the U.S. Department of Education under its Title VI, B Program for International Education.

1

CULTURES HAVE CONSEQUENCES: A CONFIGURAL APPROACH TO LEADERSHIP ACROSS TWO CULTURES

ABSTRACT This research compares the influence of country membership and cultural values (power distance and individualism/collectivism) in a model of LMX and organizational change. The results reveal cultural differences in the relationships among LMX, consultation and affective commitment to organizational change, supporting prior studies. However, there are substantial differences in the moderation of the cultural values in the relationships among the research constructs at the individual level. Our results suggest that understanding national culture and its influence on leadership may be incomplete when we focus only on mean differences at the country level to examine cross-cultural differences. To address this concern, we offer a configural approach to examine the role of culture in a leadership model across two cultures (the U.S. and Korea). Keywords: Configural Approach, Leader-Member Exchange, Consultation, Affective Commitment to Organizational Change, Cultural Values

2

INTRODUCTION The study of organizational behavior, particularly leadership, has reflected longstanding debates regarding whether leadership varies across nations or whether leadership constructs are generalizable across cultures (Avolio, Walumbwa & Weber, 2009; Den Hartog, House, Hanges, Ruiz-Quintanilla & Dorfman, 1999; Zhang et al., 2012). While emic researchers focus on understanding behaviors from the perspective of cultural insiders, those with an etic perspective seek commonalities and differences across cultures (Morris, Leung, Ames & Lickel, 1999). These assumptions have led scholars to adopt different theoretical lenses and methodologies (Morris et al., 1999). For example, etic researchers (e.g., Pillai, Scandura & Williams, 1999; Ramesh & Gelfand, 2010) adopted Hofstede’s (2001) categorization of national cultures without testing for potential differences in cultural values across more than one culture. Comparing mean scores of countries’ cultural values, these studies assumed cultural heterogeneity across nations and homogeneity within nations. In contrast, emic researchers (e.g., Ensari & Murphy, 2003; Kirkman & Shapiro, 2001; Robert, Probst, Martocchio, Drasgow & Lawler, 2000) focus on cultural values that individuals possess within a society. The emphasis from this perspective is on understanding local voices and phenomena in specific cultures. Despite the contributions of both approaches, these two streams of research have produced ambivalent results when examining the generalizability of leadership models. Consensus has not emerged regarding the uniformity or homogeneity of cultural values’ influence on attitudes and behaviors within countries. Therefore, it is necessary to address the role of cultural values in the experience and outcomes of leadership both within and between cultures. More specifically, current meta-analytic research on culture and leadership (Dulebohn, Bommer, Liden, Brouer & Ferris, 2011; Kirkman, Lowe & Gibson, 2006; Rockstuhl, Dulebohn, Ang & Shore, 2012) indicates that national culture and cultural

3

orientations are important for understanding employees’ perceptions of leadership and antecedents and outcomes of leadership. In addition, the GLOBE studies (House, Hanges, Javidan, Dorfman & Gupta, 2004) suggest that there are implicit schemas for leadership behaviors that may generalize across the 62 cultures studied. However, other research suggests that individuals’ cultural values have differential effects on attitudes and behaviors. Recent meta-analyses have shown that membership in a particular country does not moderate the relationships between leader-member exchange (hereafter LMX) and employees’ attitudes and behavior (Dulebohn et al., 2011; Rockstuhl et al., 2012). Rather, these relationships are moderated by cultural values of power distance and individualism (Dulebohn et al., 2011) as well as horizontal-individual vs. vertical-collectivistic cultures (Rockstuhl et al., 2012) that individuals possess. These studies argue that within-country (individual-level) variation in cultural values exists and can be larger than between countrylevel variation (Kirkman, Chen, Farh, Chen & Lowe, 2009; Triandis, 1995). Similar to the above arguments, prior studies (e.g., Lonner, 1980; Bass, 1997) suggested variform universals (the relationship holds but the manifestation of variables might differ across cultures), functional universals (the relationship is always found, but its magnitude differs across cultures), and variform-functional universals (a combination of the two). However, few studies have empirically identified the level of universality. More recently, Taras, Kirkman and Steel (2010) and Tsui, Nifadkar, and Ou (2007) challenged the assumption that individuals in a specific country hold shared cultural values. These scholars suggest the need to consider configural properties that individuals may differently possess within a country as well as countries’ unique cultural contexts. Configural properties refer to the asymmetry of individual characteristics within a group such as income disparity or value differences between people in contrasting regions of a country. Tsui et al. (2007) conclude:

4

“there is potential for interesting theory development by focusing on the variance of culture held by the individuals in a nation” (p. 461). To address the above limitations and to expand the understanding of configural properties of cultures, we examine the generalizability of a U.S.-based model of LMX and organizational change in another culture (Korea). We focus on LMX, defined as the exchange relationship quality between supervisors and employees (Graen & Scandura, 1987), since leaders can play an important role in shaping employees’ attitudes toward organizational change (e.g. Whelan-Berry, Gordon & Hinings, 2003). Organizational change is contextually important, as its success or failure is contingent on employees’ attitudes and reactions (Armenakis, Harris & Mossholder, 1993; Piderot, 2000) and may depend on culture (Meyer, Srinivas, Lal & Topolnytsky, 2007; Clugston, Howell & Dorfman, 2000). Despite the importance of LMX and organizational change, few studies have investigated leadership and organizational change across cultures. Most LMX studies have been conducted in countries such as Australia and the U.K. that are culturally similar to the U.S. (Dulebohn et al., 2011). The U.S. and Korea are traditionally considered as individualistic-low power distance and collectivistic-high power distance countries, respectively (Hofstede, 2001). Data from Korea should be of particular interest as “culture is not static” (Tsui et al., 2007: 465) and cultural values tend to change more in transforming economies (Fertig, 1996). For example, Korea is reported to have a mixed culture due to the westernization of its society over the past 20 years (Liden, 2012). Therefore, a comparison study of leadership during organizational change in changing cultures may be especially relevant. Overall, this paper has three purposes. First, we employ a configural approach to the study of LMX and organizational change across two cultures and make recommendations for future theory, research and practice on cultural differences in the leadership and

5

organizational change literatures. From the etic perspective, we test a model in which country is a moderator of LMX and expected outcomes (U.S. vs. Korea). Then from an emic perspective, we examine the role of cultural values (e.g., power distance and individualism/collectivism) in the relationships between the constructs in the model across the two countries. Second, we propose that generalizations using the etic approach to leadership may be incomplete since there may be unique cultural values that explain the effects of leadership on outcomes. Third, we extend research on LMX in Korea, which is relatively understudied. THEORY AND HYPOTHESES The Focal Model: LMX and Affective Commitment to Organizational Change According to LMX theory, one of the ways that supervisors and followers develop relationship quality is through a series of reciprocated exchanges (Graen & Scandura, 1987). Supervisors tend to provide more social support and share valuable information with followers in high LMX relationships (Golden & Veiga, 2008). In return, followers offer valuable resources to supervisors such as extra effort and greater dedication to the supervisors’ goals (Golden & Veiga, 2008). These exchanges provide mutual reinforcement resulting in personal obligation, gratitude, and trust (Blau, 1964). Thus, high LMX tends to shape employees’ positive attitudes toward organizations, organizational commitment (Dulebohn et al., 2011), and organizational citizenship behavior (Hofmann, Morgeson & Gerras, 2003). LMX is also positively related to positive reactions to organizational change (Van Dam, Oreg & Schyns, 2008) and mitigates negative attitudes toward change (Tierney, 1999; Furst & Cable, 2008). Affective commitment to organizational change (hereafter ACOC), defined as the desire to provide support for change based on a belief in its inherent benefits (Herscovitch & Meyer, 2002), is a critical determinant of employees’ support for organizational change and

6

reduced cynicism (Herscovitch & Meyer, 2002; Herold, Fedor & Caldwell, 2007; Bommer, Rich & Rubin, 2005; Meyer et al., 2007). Although organizational change generally increases employee uncertainty (De Cuyper, De Witte, Vander Elst & Handaja, 2010), mutual trust derived from high quality LMX can play an important role in increasing ACOC. Therefore, we propose that LMX contributes to employees’ ACOC. Consultation is defined as involvement in change when employees are “…immediately involved in decisions that affect them” (Morgan & Zeffane, 2003, p. 59). During organizational change, supervisors use various influence tactics such as sanctions, legitimation, consultation, and ingratiation (Furst & Cable, 2008). Consultation is a direct form of involvement in change (e.g., participation in decision making) (Morgan & Zeffane, 2003) in which supervisors share information about the change and allow their subordinates to voice their opinions (Yukl & Fu, 1999). Consultation can be especially effective during organizational change because employees feel a greater sense of control (Bies & Shapiro, 1988). Thus, we expect that consultation will enhance ACOC. Employees with high quality LMX are more likely to express their opinions regarding organizational change (Botero & Van Dyne, 2009). Due to mutual trust and shared goals, leaders also feel the need to consult with their most trusted subordinates (Schyns, Paul, Mohr & Blank, 2005). In addition, employees in high quality LMX have better knowledge about their supervisors’ objectives during change (Botero & Van Dyne, 2009). Since subordinate voice may disrupt relationships (Tangirala & Ramanujam, 2012), high quality LMX is an important condition for subordinates to feel psychologically safe (Detert & Burris, 2007) and to offer opinions in the context of change (Yukl & Fu, 1999). Hence, we expect LMX to relate to consultation. In addition to the aforementioned direct relationships, we propose that consultation may partially mediate the relationship between LMX and ACOC. The mediating role of

7

consultation may be salient in the context of organizational change when employees perceive higher uncertainty and ambiguity (De Cuyper et al., 2010). Supervisors may consult with all subordinates, but give more weight to the input from those with higher LMX (Tangirala & Ramanujam, 2012). Thus, supervisors’ consultative behavior toward employees sends the message that they are important to the organization and their voices matter (Tangirala & Ramanujam, 2012), which should reduce uncertainty. Hence, subordinates’ perception of LMX enhances ACOC, and consultation enhances it further. Figure 1 depicts the overall research model employed in this study. ------------------------------Insert Figure 1 about here -------------------------------The Etic Perspective: Generalizability of the LMX Model across Countries Prior scholars (e.g., Robert et al., 2000; Sagie & Aycan, 2003) have argued the need for the consideration of country membership in the understanding of organizational behavior. This perspective is based on Hofstede’s definition of culture as, “…the collective programming of the mind which distinguishes the members of one human group from another” (2001: 25). Thus, it is assumed that culture as collective programming is isomorphic with country and can distinguish one country from another. From this perspective, a country is viewed as a representative culture and differences in the causality among research constructs are considered “national cultural determinism” (McSweeney, 2002: 91). Country membership has been related to the constructs in the focal model of the current study in previous research (e.g., LMX, consultation, and ACOC). For example, Pellegrini, Scandura, and Jayaraman (2010) found that the relationship between LMX and organizational commitment was significant and positive in the U.S. but not significant in India. Parnell and

8

Hatem (1999) found that leaders’ seeking participation from subordinates was positively related to loyalty and commitment for Egyptian managers, but not for American managers. However, it should be noted that several studies have questioned national cultural determinism by showing the homogeneity of the relationships among some leadership constructs related to LMX. For example, in their meta-analysis, Rockstuhl et al. (2012) found that national culture does not influence the relationship of LMX with organizational commitment across 23 countries examined. The relationship between LMX and delegation has been found to be positive in both Eastern (Pellegrini & Scandura, 2006) and Western countries (Yukl & Fu, 1999). Similarly, the relationship between delegation and organizational commitment has been reported to be positive in both Eastern (Chen & Aryee, 2007) and Western countries (Pellegrini, 2006). Although researchers have reported inconsistent findings about whether national culture differentiates organizational behaviors, we would expect cultural determinism from an etic perspective. For the present study, based upon Hofstede (2001), the samples from the U.S. and Korea represent individualistic-low power distance and collectivistic-high power distance cultures, respectively. Thus, if national cultural determinism is assumed, these countries should have different relationships among the constructs in the focal model specified above. Thus, we will first examine the role of country as a moderator in LMX relationships with consultation and ACOC using an etic perspective. Hypothesis 1. Relationships among LMX, consultation, and ACOC will differ for the U.S. and Korea. The Emic Perspective: The Influence of Cultural Values on the LMX Model Based on prior literature (e.g., Dulebohn et al., 2011; Kirkman & Shapiro, 2001), we consider two cultural values (e.g., collectivism and power distance) at the individual level, since they can contribute to the development of interpersonal relationships and relate to

9

attachment to the group, which are the basis of organizational commitment (Erdogan & Liden, 2006; Fischer & Mansell, 2009; Kirkman & Shapiro, 2001). We hypothesize that these cultural variables will moderate the relationships between the research constructs at the individual level across countries. The above suggestion is based on the argument that individual-level variation in cultural values can be larger than country-level variation (Kirkman et al., 2009; Tsui et al., 2007). More specifically, although culture may imply a shared meaning within a country, individuals within the country may not be consistent with the shared cultural meaning overall (Fu et al., 2004; Marcus & Le, 2013). Power distance. Power distance refers to the extent to which subordinates can express disagreement with their supervisor (Hofstede, 2001). Power distance is likely to regulate individuals’ appropriate relationships between supervisors and subordinates, which are also associated with the reasons that they may perceive attachment to an organization (Fischer & Mansell, 2009). Prior studies have explained the characteristics of employees with high and low power distance, which may influence the relationship of LMX with ACOC. LMX theory assumes that employees’ attitudinal outcomes are influenced by how leaders treat them (Rockstuhl et al., 2012). In addition, the influence of LMX on attitudinal outcomes is contingent on individuals’ cultural orientations (Dickson, Den Hartog & Mitchelson, 2003; Zhang et al., 2012). For individuals with high power distance, LMX may be less important in the formation of attitudes toward the organization. Employees with high power distance tend to have unquestioning respect for authority and role-based loyalty to organizations (Chen, Friedman, Yu, Fang & Lu, 2009) and prefer to keep social distance from their superiors (Kirkman et al., 2009). Such an acceptance of power differences and the extended social distance between supervisors and subordinates reduces the influence of leaders’ behaviors on subordinates’ attitudes (Loi, Lai & Lam, 2012). On the contrary, for employees with low power distance, relative position and status in the organizations are

10

likely to be overlooked (Fischer & Mansell, 2009). These employees are more likely to have a shorter distance from their supervisors, which can improve communication quality related to organizational change (Liden, Sparrowe & Wayne, 1997). In addition, they tend to focus on the similarities with the leader and the relationship, which in turn influences their attitudes toward organizational change (Schaubroeck & Lam, 2002). Based on the above arguments, an individual’s value of high or low power distance may influence the relationship between LMX and ACOC regardless of country membership. Thus, adopting the emic perspective, the following hypothesis is suggested: Hypothesis 2. Power distance moderates the relationship between LMX and ACOC in the U.S. and Korea such that the relationship between LMX and ACOC is stronger for employees with low power distance than for those with high power distance. The effects of various forms of participation in decision making (e.g., delegation, consultation, and empowerment) on employee attitudes are contingent on cultural values. For instance, employees with high power distance are likely to undervalue participation in decision-making (Kirkman et al., 2006), delegation (Pellegrini & Scandura, 2006), and empowerment (Eylon & Au, 1999), which are similar to consultation (Morgan & Zeffane, 2003). They also expect strong direction with little communication about their jobs (Kirkman et al., 2009). Thus, although supervisors expect active voice regarding various issues during organizational change, subordinates with high power distance are not generally motivated to provide their opinions (Kirkman et al., 2009). Similarly, prior studies (e.g., Hui, Au & Fock, 2004; Robert et al., 2000; Kirkman et al., 2009) found that employees in high power distance countries would neither expect nor desire empowerment, and thus empowerment in these countries does not influence job satisfaction. Thus, for employees with high power distance, consultation may contribute less to the formation of positive attitudes toward organizational change. Conversely, employees with low power distance believe that all employees should

11

voice their opinions (Hofstede, 2001), and initiate open communications (Loi et al., 2012). Thus, they are more likely to participate in the organizational change process, which generates positive attitudes toward organizational change. We therefore propose that employees’ power distance moderates the relationship between consultation and ACOC. Hypothesis 3. Power distance moderates the relationship between consultation and ACOC in the U.S. and Korea such that the relationship between consultation and ACOC is stronger for employees with low power distance than for those with high power distance. Collectivism. Collectivism (or Individualism) is defined as the extent to which individuals see themselves as being connected (disconnected) with others (Hofstede, 2001). This cultural value tends to guide employees to accept norms and processes in their social relationships that are essential for building commitment (Fischer & Mansell, 2009). Individualistic employees are more likely to be concerned about personal relationships with supervisors and the benefits they receive from the relationship (Triandis & Gelfand, 1998). In addition, individualistic employees are more dependent on their own self-identity and autonomy (Triandis, 1995). Therefore, the influence of LMX on employees’ positive attitudinal outcomes should be more salient for employees with high individualism. Taken together, enhanced self-esteem and self-identity based on LMX may influence ACOC, which is related to acceptance of organizational goals and visions during change (Hogg & Terry, 2000). Supervisors prioritize input from high quality LMX subordinates (Tangirala & Ramanujam, 2012), which can reinforce their self-esteem and self-identity. Based on the above emic perspective, this study suggests that employees’ individualistic (collectivistic) cultural values may influence the relationship between LMX and ACOC. Hypothesis 4. Collectivism moderates the relationship between LMX and ACOC in the U.S. and Korea such that the relationship between LMX and ACOC is stronger for

12

employees with individualistic cultural values than for those with collectivistic cultural values. According to Sagie and Aycan (2003), collectivism (or individualism) helps to identify self-participation in the decision making process. More specifically, individualistic employees focus on their own interests in the decision making process, while the opposite holds for collectivistic employees (Sagie & Aycan, 2003). For instance, individualistic employees may prefer direct supervisor-subordinate interactions such as consultation in order to maximize participation opportunities for themselves; while collectivist employees may prefer to participate in the decision making process through collective representatives such as a trade union (Sagie & Aycan, 2003). In addition, Kirkman and Shapiro (2001) reported that employees with collectivism are more likely to conform to social norms. Although there are no studies that investigate the relationships among consultation (or participation in decision making), collectivism, and ACOC, some prior literature on multicultural issues hint at possible relationships. For instance, Lam, Chen, and Schaubroeck (2002), based on samples from Hong Kong, found that employees with idiocentrism (individualism at the individual level) tend to have higher individual level performance when they have high self-efficacy and opportunities to participate in decision making. However, employees with allocentrism (collectivism at the individual level) tend to have higher group level performance when they have participative decision making opportunities and group self-efficacy. Similarly, based on the comparison of three different samples from the U.S., Israel Urban, and Israel Kibbuts, Erez and Earley (1987) found that the effect of participation on job performance might be contingent on collectivism. They argued that employees from an individualistic culture favor setting their own performance goals through participation while those from a collectivistic culture tend to build their performance goals based on a collective agreement. Thus, the influence of goal setting on performance through

13

participation may be greater for employees with an individualistic culture, compared to those with a collectivistic culture. Based upon the above arguments, we expect that individualistic employees will participate more in organizational change processes through consultation to have opportunities to express their interests related to the change. They are more likely to utilize the chance to voice their interests, which in turn forms positive attitudes toward change. Conversely, collectivist employees tend to withhold their opinions and conform to organizational change practices directed by their firm. The effect of consultation on ACOC may thus be marginal for employees holding collectivist values. Therefore, Hypothesis 5. Collectivism (or individualism) moderates the relationship between consultation and ACOC in the U.S. and Korea such that the relationship between consultation and ACOC is stronger for employees with individualistic cultural values than for those with collectivistic cultural values. METHOD Samples and Procedure The data gathered in this study was part of a larger data collection originally focused on collecting international responses. Our initial call did not result in a large enough sample of respondents from different cultures. We then decided to collect data from the U.S. and from Korea in a two-country study. The U.S. sample was enlisted through a nonprofit academic service, the StudyResponse project (Stanton & Weiss, 2002). As of April 2011, 49,600 individuals, 26,000 of whom were employed, were registered with the StudyResponse service. As of February 2014, Google Scholar returned 322 search responses when using the search term “StudyResponse”. Recent studies published in top management journals have employed this service to successfully collect data (e.g., Richards & Hackett, 2012; Schaubroeck & Shao, 2012; Piccolo & Colquitt, 2006; Ng & Feldman, forthcoming). Using

14

this method, we were able to collect data from a wide variety of industries and organizations using an online survey. Participants were sent an email with a survey link and were asked to complete the survey for compensation of $5. Participants were told that the survey was completely voluntary and that the goal of the study was to “learn about the influence of organizational change on employee attitudes that may have implications for your organization’s effectiveness.” Respondents listed their StudyResponse ID number when completing the survey, which was the only identifier included with their data (the researchers did not have access to respondent identities). Five hundred and twenty-three employees were recruited to participate in this study. After receiving the email survey link, 255 completed the survey (a response rate of 49%). The link was closed within a few days because we reached our sampling goal of 250 participants (it was not necessary to send a reminder message). Therefore, the response rate may have been higher if we kept the study open for an additional amount of time and a reminder message was sent. The participants were asked if they were undergoing change and a dummy variable was created to code for those that were undergoing change versus not undergoing change. Of these respondents, one hundred forty-six indicated that their organization was undergoing change (57% of the sample) and only these respondents were included in the subsequent analyses. Responses in Korea were collected with the assistance of a regional Chamber of Commerce and Industry in a Southeastern Metropolitan city in Korea. The Chamber of Commerce and Industry sent an official participation request to member firms. The request outlined research objectives and procedures. A total of 41 companies agreed to participate in the research. The Chamber of Commerce and Industry sent a web-based survey link to the HR departments of the firms. The HR departments then sent a survey link via email to their respective employees. Employees were assured that answers would be anonymous and the

15

researchers (not firms) directly collected and handled the responses. A Total of 768 responses were initially received, yielding a response rate of 13.1%. Among them, 364 employees did not respond to all questions on the survey because they did not experience organizational change. Additionally, 105 employees completed the questionnaire but did not experience change within the last 12 months. These cases were removed from the sample for the analyses. Thus, 299 responses were used in the analyses. These individuals experienced and/or were experiencing organizational change (i.e., re-structuring, layoffs, M&A, etc.) within the past 12 months. Demographic information for both samples is shown in Table 1. ------------------------------Insert Table 1 about here -------------------------------Measures An individual who was fluent in Korean and English translated the questionnaire from English to Korean, and then another individual with proficiency in Korean and English was asked to back-translate the Korean version to English. A third expert confirmed that there were no semantic differences between the two versions (Brislin, 1980). All items were assessed on Likert-type scales in which 1 represented “strongly disagree” and 5 represented “strongly agree.” We asked the respondents to indicate the organizational change that they had experienced (or were experiencing) recently that had influenced them the most. All of the items were written with a change referent. While examining measurement equivalence between the Korean and U.S. samples, some items were deleted from the original scales. A detailed explanation concerning the measurement equivalence analysis is provided in the next section, “Measurement Equivalence”. Affective commitment to organizational change. (α = 0.85 combined, α = 0.83 for the US, and α = 0.89 for Korea). We used Herscovitch and Meyer’s (2002) 6-item measure of affective commitment to organization change. Two items were deleted for measurement 16

equivalence. A sample item from this measure is “This change is a good strategy for this organization.” LMX. (α = 0.92 combined, α = 0.92 for the US, and α = 0.91 for Korea). We measured LMX using Graen and Uhl-Bien’s (1995) LMX-7 scale. In both samples, all 7 original items were included. A sample item from this measure is “How well does your leader understand your job problems and needs?” Consultation. ( = 0.82 combined,  = 0.74 for the US,  = 0.87 for Korea). We used Morgan and Zeffane’s (2003) three-item measure of consultation. Specifically, we used three items from the involvement in change scale that the authors described as direct participation methods (e.g., supervisors’ and managers’ discussion and meetings). A sample item from this measure is “Managers at a higher level discussed changes with me.” Individualism/Collectivism. ( = 0.74 combined,  = 0.78 for the US,  = 0.70 for Korea). Following previous individual-level research, we employed Dorfman and Howell’s (1988) scale in order to measure individualism/collectivism. This measure is a fundamentally revised version of Hofstede (1980) and captures cultural dimensions at the individual level. One item was dropped due to measurement equivalence. A sample item from this measure is “Managers should encourage group loyalty even if individual goals suffer.” Power distance. ( = 0.86 combined,  = 0.83 for the US,  = 0.88 for Korea). Dorfman and Howell’s (1988) scale was adopted in order to measure power distance. One item was deleted for the improvement of measurement equivalence. A sample item from this measure is “Managers should make most decisions without consulting subordinates.” Control variables. We compared demographic characteristics between the U.S. and Korea using chi-square tests for gender and marital status and t-tests for age, education and rank. The results revealed that the samples from the two countries were significantly different

17

in terms of the four demographic variables except gender. Thus, we included gender, age, marital status, rank and education as covariates in the subsequent analyses. Measurement Equivalence This study employed a configural approach that captures differences in individual characteristics (e.g., power distance and collectivism) within countries. We followed the configural and metric equivalence tests established in the field (Schaffer & Riordan, 2003; Vandenberg & Lance, 2000). Configural equivalence signifies the equality of factor structures across sample countries (Vandenberg & Lance, 2000). Metric equivalence refers to the equality of factor-loading parameters across groups (Vandenberg & Lance, 2000). Both types of equivalence were examined by a multi-group confirmatory factor analysis using MPlus (Muthen & Muthen, 2002). The results revealed that the five-factor (e.g., ACOC, LMX, consultation, power distance, individualism/collectivism) model and factor-loadings are appropriate and equivalent across the two countries (2 = 7714.93, df = 552, RMSEA = 0.07, SRMR = 0.09, CFI = 0.91, and TLI = 0.90). Detailed results of the multi-group factor analysis including factor loadings and t-values can be provided upon request from the first author. Thus, the measures appear to have configural and metric equivalence. And, differences in the samples due to measurement and data collection methods can be ruled out as alternative explanations for our findings. Common Method Bias Since we collected responses from a single source in a cross-sectional manner, common method bias can be a potential problem that can influence overall results (Podsakoff, MacKenzie, Lee & Podsakoff, 2003). Thus, we examined the possible presence of common method variance in two different ways. First, we conducted a confirmatory factor analysis approach. A model positing that a common method factor underlies the variables is assessed

18

by linking all items to a single factor. This model did not come close to fitting the data well for the U.S. and Korea (2 = 1480.32, df = 210, RMSEA = 0.15, SRMR = 0.18, CFI = 0.54, and TLI = 0.49 for the U.S. sample; 2 = 2068.76, df = 209, RMSEA = 0.18, SRMR = 0.14, CFI = 0.50, and TLI = 0.45 for Korean sample). Second, we compared the chi-square (χ2) of the models with and without a common method latent variable for the two samples. For the U.S. sample, a model with a common method latent factor yielded χ2=439.81.91 (df=198). A model without the factor had χ2=419.81.91 (df=199). The difference was 20.01 (df=1), which is statistically significant. Similarly, for the Korean sample, the model with the common method latent factor had χ2=660.94 (df=198), and the model without the method factor yielded χ2=562.35 (df=199). The difference was 98.57 (df=1), which is statistically significant. Based on these results, we concluded that common method bias did not represent a significant problem for this study. Analytical Procedure The main purpose of this study is to identify whether countries and cultural values moderate the focal model (Figure 1). Scholars have noted several limitations of traditional moderated mediation approaches (and mediated moderation) especially regarding the unclear nature of moderation and mediation (Edwards & Lambert, 2007). Thus, this study adopted Preacher, Rucker, and Hayes’ (2007) moderated meditation method in order to test a model that combines both moderation (e.g., countries and cultural values) and mediation (e.g., consultation). The procedure includes four stages (Preacher et al., 2007). In the first two stages, the mediator variable is regressed on the independent variable (e.g., LMX), and then a multiple regression is conducted in which the dependent variable is regressed on the mediator, moderator, independent variable, and the interaction between the moderator and the mediator. The third stage examines the conditional indirect effects of the independent variable at certain values of the moderating variable. Rather than assuming normality of the 19

sample distribution, in the fourth stage, this procedure tests the non-parametric conditional effects based on a bootstrapping procedure. It should be noted that as recommended by Preacher et al. (2007) all of the variables were mean-centered before the analysis. This procedure was conducted using the SPSS PROCESS Macro provided by Hayes on the website (http://afhayes.com/spss-sas-and-mplus-macros-and-code.html). RESULTS Correlations Table 2 presents the correlations among the research constructs and demographic variables included in our analysis. Interestingly, correlations for LMX and ACOC (0.46 in Korea and 0.47 in U.S.) and for consultation and ACOC (0.50 in Korea and 0.48 in U.S.) are similar in the two countries. However, the correlations of power distance with ACOC, consultation and LMX are significantly negative in Korea, but not in the U.S. (some are even positive). These results suggest the importance of the interactions of cultural variables with focal constructs while suggesting a marginal country level influence on the relationships and the need for a configural approach to understanding cross-cultural differences. ------------------------------Insert Table 2 about here -------------------------------The Etic Perspective: Mean Differences Table 3 presents the mean differences of the focal research constructs. These results indicate that mean differences at the country level may not fully explain cross-cultural differences in organizational behavior. For instance, while LMX, consultation and power distance appear to be significantly different in the two countries, ACOC and individualism/collectivism appear to be the same in both the U.S. and Korea. This result is consistent with Fischer and Mansell’s (2009) finding that absolute differences of commitment

20

across countries are relatively small. Interestingly, although Hofstede (2001) and many other scholars have adopted the assumption that the U.S. has a high level of individualism and Korea has a high level of collectivism, the results of this study demonstrate that the two countries have similar levels of individualism/collectivism (3.48 in the U.S. vs. 3.51 in Korea, p=0.49). These results suggest that cultural values at the individual level may not be always consistent with those at the social level. ------------------------------Insert Table 3 about here -------------------------------We reviewed the literature to identify prior studies reporting main differences at the country level for the U.S. and Korea. Although there are several studies that argue high levels of collectivism and power distance in Korea, very few actually measured cultural values. Further, most recently published papers do not report means and standard deviations. In terms of power distance, similar to Hofstede, prior studies show that Koreans have power distance orientations slightly higher than average (Jung, Polyorat & Kellaris, 2009; Kim & Kim, 2010; Kwak, 2012; Kim, 2012; Kim & Leung, 2007). However, in terms of collectivism, unlike Hofstede’s study, which indicates that Koreans are extremely collectivistic, prior studies and our study show that Koreans are moderately collectivistic. Etic vs. Emic Hypothesis Testing Hypothesis 1 suggested that the focal model is moderated by country membership. Figure 2 displays the results for this hypothesis. As shown at the top of Figure 2, country membership of respondents (the U.S. and Korea) was not found to moderate the relationships in the focal model. However, when we examine the relationships among the research constructs, the relationships were significantly different. As shown in the middle section of Figure 2, LMX positively influenced ACOC and consultation (β = 0.40 and β = 0.35), and

21

consultation also positively influenced ACOC (β = 0.35) in the U.S. As shown at the bottom section of Figure 2, LMX positively influenced ACOC and consultation (β = 0.59 and β = 0.28), and consultation also positively influenced ACOC (β = 0.31) in Korea. With respect to indirect effects, consultation significantly mediated the relationship with β = 0.19 in the U.S. and with β = 0.16 in Korea. However, the differences (U.S.-Korea) in the path coefficients between the two countries were significant1 (difference = -0.19 (t = 33.03, p < 0.01) for LMX consultation; difference = 0.07 (t = 14.08, p < 0.01) for LMX ACOC; difference = 0.04 (t = 7.05, p < 0.01) for consultation  ACOC). In sum, country was found to moderate the focal model, supporting Hypothesis 1. ------------------------------Insert Figure 2 about here -------------------------------Table 4 shows the overall results of moderated mediation analyses and Figure 3 depicts key findings related to moderation effects of cultural variables. Hypothesis 2 suggested that power distance moderated the relationship between LMX and ACOC in the two countries. As shown in Figure 3 and in Table 4, the moderation of power distance is significant and negative in the U.S. (β = -0.17), but is not significant in Korea. More specifically, as a graph at the top of Figure 4 shows, the relationship between LMX and ACOC is stronger for low power distance (Slope=.56, Mean-1 S.D.) than for high power distance (Slope=.21, Mean+1 S.D.) in the U.S. Thus, Hypothesis 2 is rejected in Korea, but accepted in the U.S. ------------------------------Insert Table 4 about here -------------------------------Hypothesis 3 suggested power distance to be a moderator of the relationship between consultation and ACOC in the two countries. As shown in Figure 3 and in Table 4, while 22

moderation of power distance was not significant in the U.S., the moderation was negative and significant in Korea (β = -0.14). As the graph at the bottom of Figure 4 shows, the relationship between consultation and ACOC is stronger for employees with low power distance (Slope=.46, Mean+1 S.D.) than for those with high power distance (Slope=.17, Mean-1 S.D.) in Korea. Hence, Hypothesis 3 was accepted for Korea, but rejected for the U.S. ------------------------------Insert Figure 3 about here -------------------------------------------------------------Insert Figure 4 about here -------------------------------Hypothesis 4 proposed individualism/collectivism to be a moderator of the relationship between LMX and ACOC in the two countries. As shown in Figure 3 and in Table 4, while the moderation of individualism/collectivism was positive and significant in the U.S., the moderation was negative and significant in Korea. As the graph at the top of Figure 5 shows, opposite to Hypothesis 3, the relationship between LMX and ACOC is stronger for high collectivist employees (Slope=.67, Mean+1 SD) than for low collectivist ones (Slope=.09, Mean-1 S.D.) in the U.S. However, as the graph in the middle of Figure 5 shows, the relationship is stronger for low collectivist employees (Slope=.78) than for high collectivist ones (Slope=-.08) in Korea. Thus, Hypothesis 4 was accepted for Korea, but rejected for the U.S. Hypothesis 5 proposed individualism/collectivism to be a moderator of the relationship between consultation and ACOC in the two countries. As shown in Figure 3 and Table 4, while the moderation of collectivism was not significant in the U.S., the moderation

23

was positive and significant in Korea (β = 0.28). As the graph at the bottom of Figure 5 shows, although the interaction was significant in Korea (but not hypothesized) the relationship between consultation and ACOC was stronger for employees with high collectivism (Slope=.62, Mean+1 SD) than for those with low collectivism (Slope=.01, Mean-1 S.D.) in Korea. Thus, Hypothesis 5 was rejected for both countries. ------------------------------Insert Figure 5 about here -------------------------------DISCUSSION We developed configural models of LMX and organizational change for the U.S. and Korea as suggested by Tsui et al. (2007), and adopted an etic-emic approach. The results of our study confirm prior theorizing that understanding national culture and its influence on leadership may be incomplete when we depend solely on mean differences across cultures. Examining and interpreting relationships among constructs may also lead to partial conclusions without considering the configural relationships among constructs within cultures, highlighting variform functional universal (Bass, 1997). This study also shows that the relationships between ACOC and its antecedents are substantially different when considering employees’ cultural orientations (e.g., power distance and collectivism), thus extending prior literature on commitment to organizational change (Tierney, 1999; Furst & Cable, 2006; Herscovitch & Meyer, 2002). In terms of the ability to generalize the LMX and organizational change model (LMXconsultationACOC), the LMX model appears to hold in both cultures despite assumed cultural differences. However, when comparing coefficients in the two countries, we found that the relationships between the research constructs are significantly different. Specifically, the relationship between LMX and ACOC in the U.S. sample was significantly

24

higher than in Korea. Also, the relationship between consultation and ACOC in the U.S. was significantly greater than Korea. Considering that U.S. respondents reported lower power distance than Korean respondents, the differences in power distance across the two countries may help explain the differences in these relationships. These results are consistent with prior studies on LMX, which found that employees with high power distance orientations are less sensitive to LMX, have the desire to participate in decision making and have higher organizational commitment (Chen et al., 2009; Kirkman et al., 2006, Pellegrini & Scandura, 2006). In addition, these results extend current understanding of commitment to organizational change by showing the role of employees’ national culture. That is, while LMX and consultation generally play positive roles in enhancing ACOC, their roles appear more important in the U.S. than in Korea. Of perhaps even more interest is that cultural values at the individual level (power distance and individualism/collectivism) may matter even more when similar patterns in the relationships among the research constructs are found. Our results indicate that cultural values moderated the relationships in significantly different ways at the individual level in the two countries. Therefore, this study supports the idea that individuals within a society may have cultural values that deviate from societal norms (Tsui et al., 2007). Employees’ attitudes during organizational change can be contingent on cultural values. Power Distance Power distance negatively moderates the relationship of LMX with ACOC in the U.S. but not in Korea. Yet, power distance negatively moderated the relationship of consultation with ACOC in Korea but not in the U.S. Taken together, these results indicate that power distance has different relationships with LMX across these two cultures: Power distance is detrimental to LMX and ACOC in the U.S., however it is detrimental to the relationship between consultation and ACOC in Korea.

25

Our results suggest that country-specific contexts related to power distance may influence results. As the correlations show (see Table 2), power distance was significantly and negatively related to LMX in Korea and not related to LMX in the U.S. Employees with high power distance may not be confident enough to develop high quality LMX in Korea. Moreover, these employees are less likely to show ACOC. In other words, in the Korean context, power distance may play an independent role that negatively influences both LMX and ACOC. We did not find this in the U.S. Collectivism The results regarding individualism/collectivism are equally interesting. Contrary to research on collectivism and our hypotheses, collectivism positively moderated the relationships between LMX and ACOC in the U.S. and between consultation and ACOC in Korea (see the graphs at the top and bottom of Figure 5). One of the reasons why employees with high collectivism have stronger effects may be that collectivists consider commitment as a sense of duty (Triandis, 1988). Thus, in both countries, employees that are higher on collectivism have stronger relationships of ACOC with LMX and consultation. Although LMX and consultation may provide favorable conditions for employee commitment, their influences tend to be marginal for employees with lower collectivism. In addition, as shown in the middle section of Figure 5, in Korea, the interaction between collectivism and LMX negatively influences ACOC, as hypothesized. Specifically, employees with low collectivism have higher levels of ACOC, compared to those with high collectivism. Post-Hoc Analyses: Three-Way Interactions To reinforce the above arguments related to the cultural values, we conducted posthoc analyses to examine the potential three-way interactions on ACOC. While the three-way interactions (LMX x Power Distance x Collectivism and Consultation x Power Distance x Collectivism) significantly affected ACOC in Korea, none of them were significant in the

26

U.S. Figure 6 shows the results of the three-way interactions for Korea. The influence of consultation on ACOC was strongest for employees with low power distance and high collectivism. Employees with low power distance and low collectivism had the next strongest. For employees with high power distance, regardless of their collectivism orientation, changes in the influence of consultation on ACOC were relatively weaker. Similar patterns were found in the three-way interactions among LMX, collectivism, and power distance. Compared to employees with high power distance, employees with low power distance had the strongest influence on ACOC at higher levels of consultation. In addition, when employees had high power distance, only employees with high collectivism rather than low collectivism had a stronger positive relationship with ACOC. Taken together, in Korea, it is possible that power distance may eliminate the positive influences of consultation and LMX on ACOC; however, employees’ collectivism may influence ACOC as a sense of duty, similar to normative commitment to organizational change (Herscovitch & Meyer, 2002)2. ------------------------------Insert Figure 6 about here -------------------------------In sum, our study suggests that examining mean differences and/or identifying the difference of relationships among research constructs across countries with the assumption that leadership is generalizable across cultures (the etic perspective) may result in incomplete conclusions. In other words, without consideration of cultural values at the individual level, LMX may be assumed to be similarly important for all individuals that have different cultural orientations in the context of organizational change. The results of this research also theoretically expand our understanding of leadership and culture for change management, while highlighting the need for configural approaches and more etic-emic studies. LIMITATIONS AND DIRECTIONS FOR FUTURE RESEARCH

27

Despite the contributions of this study, there are limitations regarding the ability to generalize the results. This study was conducted with self-report responses and with a crosssectional design. Although our analyses did not reveal a threat from common method bias, limitations related to research methods should be considered in interpreting these findings. However, Chan (2009) discusses the “urban legend” of self-report data and addressed four common concerns: Construct validity, interpretation of correlations, social desirability, and other types of performance. With respect to construct validity, following Chan’s recommendations, we employed established measures. With regard to interpretation of correlations, Chan suggests, “…inflation of the observed correlation is a possibility and not a necessity. Moreover, the impact of common method variance on the magnitude and interpretation of the observed correlation is not always clear because other counteracting effects exist” (2009: 318). In terms of social desirability, Chan states, “…social desirability responding and its effects are not as ubiquitous as it is widely believed” (2009: 324). Lastly, our study did not include measures of performance, which would have been desirable. We focused on ACOC whose construct is best measured from the employee perspective (Herscovitch & Meyer, 2002). Therefore, this should ease concerns regarding the self-report nature of our data. Additionally, our analyses detected significant mediation and moderation effects, which are not possible for respondents to anticipate or fake good (Schriesheim & DeNisi, 1981). This lends some confidence that our findings were less affected by self-report bias. The configural approach highlights understanding the influences of individual characteristics within a group as well as country characteristics (Tsui et al., 2007). Moreover, the data from the U.S. and Korea had different demographic characteristics in terms of rank, age, marital status, and education. Although we included these variables as covariates in the analysis, their influence on the results may not be completely eliminated. Especially since the

28

interactions of country membership with age and education were found to be significantly associated with power distance and collectivism and since respondents’ rank was positively related to all constructs, readers should consider these influences while interpreting the results of this study. In addition, the responses from the U.S. were collected from a panel and responses from Korea were gathered from organizations (however both samples were collected online). Thus, the compatibility between the two samples and the differences in data collection methods between the two countries may lead to issues in interpreting the results. Further, since the two countries may have different economic situations and their samples may be involved in different industry sectors, the variables related to these economics and industry sectors may influence the results. The sample sizes of the two countries are relatively small and may not reflect the overall populations. Although the samples in both countries are representative in terms of age, education, and gender, they are white-collar workers (based upon Korean and U.S. Census data). Thus, demographic characteristic should be considered while interpreting the results. Overall, future research should collect data using similar data collection methods and additional representative samples and also include industry sector as a control variable to eliminate the potential influence of industry. Despite these limitations, our findings have implications for future theory and research. The results of this study suggest that researchers focus more on the nuances of national culture (i.e., adopting a etic-emic configural approach), rather than comparing the U.S. with other cultures by reporting mean differences. Although prior studies have provided evidence of national cultural differences in leadership at the country level, this study shows that individuals may act differently than the cultural prototypes of their countries. When we examined the moderating role of cultural values in the relationships among the constructs using a finer-grained configural analysis, the cultural value of collectivism played different roles across countries. Future research needs to identify why/how collectivism plays different

29

roles across countries by questioning the fundamental meaning of collectivism to employees within countries. In general, we found that power distance was higher in Korea than in the U.S. However, there is more to the story. Since managers with high power orientations are less likely to seek information from their subordinates (Dickson et al., 2003), supervisors’ consultation with a few subordinates based on low power distance orientations suggests their role as “trusted assistants” as the LMX model predicts. However, employees with high power distance may not participate in the decisions during organizational change. Even if they do participate in decision-making, they may consider their participation to be nominal and expect the executives to make the decisions. Indeed, the significant negative correlation between power distance and consultation in the Korean sample further supports the above argument. Thus, our results suggest that power distance appears to eliminate the positive effect of consultation in Korea. However, in the U.S., a positive correlation between power distance and consultation means that employees with higher respect for authority were more likely to feel involved (and appreciate this involvement). The correlation with collectivism, while unexpected, does indicate that those who had more respect for authority were also more group-oriented. Hofstede conceptualized these as distinct cultural values and the U.S. is actually moderately high on power distance (Hofstede, 2001). So it is possible for respondents from a diverse sample to have a positive and significant relationship between power distance and collectivism. The traditional assumption that Koreans are collectivists and U.S. respondents are individualists was not borne out by our findings. In addition, though most prior studies including this study adopt Hofstede’s definitions of power distance and individualism/collectivism, these definitions do not focus on situational differences, individual variability, or the scalability of cultural values. They were also initially

30

conceptualized at the country level. Thus, future research should conceptualize cultural values at the individual level with consideration of specific contexts surrounding individual employees. In addition, cultural values should be investigated at the individual level, rather than only depending on mean differences across cultures for understanding the influences of culture on leadership behavior. Future research needs to investigate possible differentiation of leadership behavior of individualist employees in collectivistic countries from those of individualist employees in individualistic countries. For instance, cultural tightness-looseness defined as the extent to which social norms influence individuals in a society or the extent to which a society is likely to allow individuals to deviate from its norms (Gelfand, Nishii & Raver, 2006) may explain the differences in leadership behaviors. Since Hofstede (2001) and the GLOBE study (House et al., 2004) both argued that the two countries differ on the masculinity-femininity dimension and prior studies have shown that there are variations in leadership perceptions across cultures (Scott & Brown, 2006), future research can further investigate this dimension when it compares leadership and outcomes across countries. Additionally, our study focused on one leadership theory, LMX. We encourage future studies to follow a similar approach (etic-emic treatment) when examining other theories of leadership. IMPLICATIONS FOR PRACTICE This study provides important implications for practitioners managing organizational change in different cultural settings in multinational contexts. When firms implement organizational change, employees may resist the change and lack ACOC, which can hinder the firms from achieving change goals. Cultural differences may be one of the reasons that many multi-national companies are not able to implement organizational changes that have been successful in the headquarters. In particular, local offices of multinational firms may follow the change process directed by headquarters located in different countries and

31

intentionally or unintentionally ignore issues related to cultural differences. Even if firms acknowledge the differences, they may overlook differences in the cultural values of individual employees. Hence, leaders in local offices are critical because they are in charge of the process of organizational change. Our results suggest that culture values inform the change implementation process at the country level as well as at the individual level. Multinational firms need to develop leaders, in both local and main offices, that possess the necessary managerial skills to deal with cultural issues. Based on traditional studies of national culture, managers might rely on stereotypes of employees across local countries (e.g., Koreans are collectivists and have high power distance and U.S. employees are the opposite). However, this study supports the contention that such stereotypes may have limited usefulness for managers since individuals may behave in different ways than the culture of the societies they are embedded in. Even when managers consider culture at the country level, they may understand that LMX and consultation are equally important for all employees. They may also believe that LMX and consultation are less important for Korean employees and more important for US employees. As the results of this study suggest, managers may need to focus on individuals’ cultural values, and strategically manage different types of employees. For Korean employees, leaders need to encourage subordinates with higher power distance to participate in discussions during organizational change in order to increase their commitment to the change. For U.S. employees, leaders need to build high quality relationships with subordinates with low power distance to enhance their commitment to organizational change. With regard to collectivism, managers in the U.S. need to concentrate their efforts on building LMX specifically with collectivistic employees since such high quality relationships can enhance employees’ commitment to organizational change. Managers in Korea are required to use different tactics based on employees’ collectivism. Specifically, for

32

employees with high collectivism, since their commitment to organizational change can be enhanced based on consultation, leaders need to encourage these employees to engage at the team level. For employees with low collectivism, Korean managers may need to focus more on LMX relationships. The above differentiation tactics based on subordinates’ cultural values is in line with the LMX approach to leadership in which unique working relationships are developed with each employee rather than adopting an average leadership style. Each employee is treated as unique in the LMX approach and his/her national culture is part of this uniqueness. Hence, the overall results of this study help leaders manage multicultural teams, when they are in geographically dispersed areas. CONCLUSION The Hofstede and GLOBE approaches to the study of national culture provided a rich base of research that generated a great deal of interest in the impact of cultural differences on leadership. However, our comparative analysis of etic versus emic approaches to the study of cultural differences demonstrates that conclusions based only upon etic research may be incomplete. Cultures do have consequences, and our results suggest that assuming cultural generalizability of U.S.-derived models across cultures may limit our explanations of crosscultural leadership. It would be productive to develop an in-depth understanding of the configural properties within a culture and examine emic influences on leadership and organizational outcomes of interest. We discovered that theories and practices for leadership and organizational change might not generalize across cultures. A more complete understanding will include both within-culture as well as between-culture influences on follower commitment to change.

33

References Armenakis, A.A., Harris, S.G. & Mossholder, K.W. (1993). Creating readiness for organizational change. Human Relations, 46, 681–703. Avolio, B.J., Walumbwa, F.O. & Weber, T.J. (2009). Leadership: Current theories, research, and future directions. Annual Review of Psychology, 60, 421-449. Bass, B. M. (1997). Does the Transactional-Transformational Leadership Paradigm Transcend Organizational and National Boundaries? American Psychologist, 52, 130-139. Bies, R. J. & Shapiro, D. L. (1988). Voice and justification: Their influence on procedural fairness judgments. Academy of Management Journal, 31, 676-685. Blau, P. M. (1964). Exchange and power in social life. Transaction Publishers. Bommer, W. H., Rich, G. A. & Rubin, R. S. (2005). Changing attitudes about change: Longitudinal effects of transformational leader behavior on employee cynicism about organizational change. Journal of Organizational Behavior, 26(7), 733-753. Botero, I. C. & Van Dyne, L. (2009). Employee voice behavior interactive effects of LMX and power distance in the United States and Colombia. Management Communication Quarterly, 23, 84-104. Brettel, M., Engelen, A., Heinemann, F. & Vadhanasindhu, P. (2008). Antecedents of market orientation: A Cross-cultural comparison, Journal of International Marketing, 16(2), 84119. Brislin, R. W. (1980). Translation and content analysis of oral and written material. In H. C. Triandis & J. W. Berry (Eds.), Handbook of cross-cultural psychology (Vol. 2, pp. 389444), Boston: Allyn and Bacon. Chan, D. (2009). So why ask me? Are self-report data really that bad? In C.E. Lance & R.J. Vandenberg (Eds.), Statistical and methodological myths and urban legends: Doctrine, verity, and fable in organizational and social sciences, (pp. 309-332), New York: Taylor & Francis Group. Chen, Z.X. & Aryee, S. (2007). Delegation and employee work outcomes: An examination of the cultural context of mediating processes in China. Academy of Management Journal, 50, 226-238.

34

Chen, Y., Friedman, R., Yu, E., Fang, W. & Lu, X. (2009). Supervisor–subordinate guanxi: Developing a three dimensional model and scale. Management and Organization Review, 53, 375-399. Clugston, M., Howell, J. P. & Dorfman, P. W. (2000). Does cultural socialization predict multiple bases and foci of commitment? Journal of Management, 26(1), 5-30. De Cuyper, N., De Witte, H., Vander Elst, T. & Handaja, Y. (2010). Objective threat of unemployment and situational uncertainty during a restructuring: Associations with perceived job insecurity and strain. Journal of Business and Psychology, 25, 75-85. Den Hartog, D. N., House, R. J., Hanges, P. J., Ruiz-Quintanilla, S. A. & Dorfman, P. W. (1999). Culture specific and cross-cultural generalizable implicit leadership theories: Are attributes of charismatic/transformational leadership universally endorsed? The Leadership Quarterly, 10, 219-256. Detert, J. R. & Burris, E. R. (2007). Leadership behavior and employee voice: is the door really open? Academy of Management Journal, 50, 869-884. Dickson, M. W., Den Hartog, D. N. & Mitchelson, J. K. (2003). Research on leadership in a cross-cultural context: Making progress, and raising new questions. The Leadership Quarterly, 14(6), 729-768. Dorfman, P. W. & Howell, J. P. (1988). Dimensions of national culture and effective leadership patterns: Hofstede revisited. Advances in International Comparative Management, 3, 127-150. Dulebohn, J., Bommer, W. H., Liden, R. C., Brouer, R. L. & Ferris, G. R. (2011). A metaanalysis of antecedents and consequences of leader-member exchange: Integrating the past with an eye toward the future. Journal of Management, 38(6), 1715-1759. Edwards, J. R. & Lambert, L. S. (2007). Methods for integrating moderation and mediation: A general analytical framework using moderated path analysis. Psychological Methods, 12, 1-22. Ensari, N. & Murphy, S. E. (2003). Cross-cultural variations in leadership perceptions and attribution of charisma to the leader. Organizational Behavior and Human Decision Processes, 92, 52-66.

35

Erdogan, B. & Liden, R. (2006). Collectivism as a moderator of responses to organizational justice: Implications for leader-member exchange and ingratiation. Journal of Organizational Behavior. 27, 1-17. Erez, M. & Earley, P.C. (1987). Comparative analysis of goal-setting strategies across cultures. Journal of Applied Psychology, 72, 658-665. Eylon, D. & Au, K. Y. (1999). Exploring empowerment cross-cultural differences along the power distance dimension. International Journal of Intercultural Relations, 23(3), 373385. Fertig, G. (1996). Investigating the process of culture change from an anthropological perspective. The Social Studies, 87(4), 165-170. Fischer, R. & Mansell, A. (2009). Commitment across cultures: A meta-analytical approach. Journal of International Business Studies, 40(8), 1339-1358. Furst, S. A. & Cable, D. M. (2008). Employee resistance to organizational change: Managerial influence tactics and leader-member exchange. Journal of Applied Psychology, 93, 453. Fu, P.P., Kennedy, J., Tata, J., Yukl, G., Bond, M.H., Peng, T.K., Srinivas, E.S., Howell, J.P., Prieto, L., Koopman, P., Boonstra, J.J., Pasa, S., Lacassagne, M.F., Higashide, H. & Cheosakul, A. (2004). The impact of societal cultural values and individual social beliefs on the perceived effectiveness of managerial influence strategies: A meso approach’, Journal of International Business Studies, 35(4), 284–305. Gelfand, M. J., Nishii, L. H. & Raver, J. L. (2006). On the nature and importance of cultural tightness-looseness. Journal of Applied Psychology, 91, 1225-1244. Golden, T. D. & Veiga, J. F. (2008). The impact of superior–subordinate relationships on the commitment, job satisfaction, and performance of virtual workers. The Leadership Quarterly, 19, 77-88. Graen, G. B. & Scandura, T. A. (1987). Toward a psychology of dyadic organizing. Research in Organizational Behavior, 9, 175-208. Graen, G. B. & Uhl-Bien, M. (1995). Relationship-based approach to leadership: Development of leader-member exchange (LMX) theory of leadership over 25 years: Applying a multi-level multi-domain perspective. The Leadership Quarterly, 6(2), 219247. 36

Herold, D. M., Fedor, D. B. & Caldwell, S. D. (2007). Beyond change management: A multilevel investigation of contextual and personal influences on employees' commitment to change. Journal of Applied Psychology, 92(4), 942-951. Herscovitch, L. & Meyer, J. P. (2002). Commitment to organizational change: Extension of a three-component model. Journal of Applied Psychology, 87(3), 474-487. Hofmann, D. A., Morgeson, F. P. & Gerras, S. J. (2003). Climate as a moderator of the relationship between leader-member exchange and content specific citizenship: Safety climate as an exemplar. Journal of Applied Psychology, 88(1), 170. Hofstede, G. (1980). Culture’s consequences: International differences in work-related values. Newbury Park, CA: Sage. Hofstede, G. (2001). Culture’s consequences: Comparing values, behaviors, institutions, and organizations across nations (2nd ed.). Thousand Oaks, CA: Sage. Hogg, M. A. & Terry, D. J. (2000). Social identity and self-categorization processes in organizational contexts. Academy of Management Review, 25, 121-140. House, R. J., Hanges, P. J., Javidan, M., Dorfman, P. & Gupta, V. (Eds.). (2004). Culture, leadership, and organizations: The GLOBE study of 62 societies. Newbury Park, CA: Sage. Hui, M.K., Au, K. & Fock, H. (2004). Empowerment effects across cultures, Journal of International Business Studies, 35, 46–60. Jung, J. M., Polyorat, K. & Kellaris, J. J. (2009). A cultural paradox in authority-based advertising. International Marketing Review, 26(6), 601-632. Kwak, W. J. (2012). Charismatic leadership influence on empowered and less empowered followers’ voice: a mediated moderation model. Journal of Leadership, Accountability and Ethics, 9(1), 56-70. Kim, J. (2012). Sexual coercion across cultures: an examination of prevalence, perceptions, and consequences of sexual coercion in Korea and the United States, Master Dissertation, San Diego State University. Kim, T. Y. & Leung, K. (2007). Forming and reacting to overall fairness: A cross-cultural comparison. Organizational Behavior and Human Decision Processes, 104(1), 83-95.

37

Kim, Y. & Kim, S. Y. (2010). The influence of cultural values on perceptions of corporate social responsibility: Application of Hofstede’s dimensions to Korean public relations practitioners. Journal of Business Ethics, 91(4), 485-500. Kirkman, B. L., Chen, G., Farh, J., Chen, Z. X. & Lowe, K. B. (2009). Individual power distance orientation and follower reactions to transformational leaders: A cross-level, cross-cultural examination. Academy of Management Journal, 52, 744-764. Kirkman, B. L., Lowe, K. B. & Gibson, C. B. (2006). A quarter century of culture's consequences: A review of empirical research incorporating Hofstede's cultural values framework. Journal of International Business Studies, 37, 285-320. Kirkman, B. L. & Shapiro, D. L. (2001). The impact of cultural values on job satisfaction and organizational commitment in self-managing work teams: The mediating role of employee resistance. Academy of Management Journal, 44(3), 557-569. Lam, S.S.K., Chen, X.P. & Schaubroeck, J. (2002). Participative decision making and employee performance: The moderating effects of allocentrism/idiocentrism and efficacy. Academy of Management Journal, 45, 905-914. Liden, R. C. (2012). Leadership research in Asia: A brief assessment and suggestions for the future. Asian Pacific Journal of Management, 29, 205-212. Liden, R. C., Sparrowe, R. T. & Wayne, S. J. (1997). Leader-member exchange theory: The past and potential for the future. In R.F. Gerald (Ed.), Research in personnel and human resources management (pp. 47-119). US: Elsevier Science. Loi, R., Lai, J. Y. & Lam, L. W. (2012). Working under a committed boss: A test of the relationship between supervisors' and subordinates' affective commitment. The Leadership Quarterly, 23, 466-475. Lonner, W. J. (1980). The search for psychological universals. In H. C. Triandis & W. W. Lambert, (Eds.), Handbook of Cross-Cultural Psychology (Vol. 1, pp. 143-204). Boston: Allyn and Bacon. Marcus, J. & Le, H. (2013). Interactive effects of levels of individualism-collectivism on cooperation: A Meta-analysis. Journal of Organizational Behavior, 34, 813-834. Meyer, J. P., Srinivas, E. S., Lal, J. B. & Topolnytsky, L. (2007). Employee commitment and support for an organizational change: Test of the three‐component model in two cultures. Journal of Occupational and Organizational Psychology, 80(2), 185-211. 38

Morgan, D. & Zeffane, R. (2003). Employee involvement, organizational change and trust in management. International Journal of Human Resource Management, 14, 55-75. McSweeney, B. (2002). Hofstede’s model of national cultural differences and their consequences: A triumph of faith-a failure of analysis. Human Relations, 55, 89-118. Morris, M. W., Leung, K., Ames, D. & Lickel, B. (1999). Views from inside and outside: Integrating emic and etic insights about culture and justice judgment. Academy of Management Review, 24(4), 781-796. Muthen, L. K. & Muthen, B. O. (2002). Mplus VERSION 2.12: Addendum to the Mplus Users’s Guide. Los Angeles: Muthen & Muthen. Ng, T.W.H. & Feldman, D.C. forthcoming. Idiosyncratic deals and voice behavior. Journal of Management. Parnell, J. A. & Hatem, T. (1999). Cultural antecedents of behavioural differences between American and Egyptian managers. Journal of Management Studies, 36, 399-418. Pellegrini, E. K. (2006). A cross-cultural investigation of the relations among organizational justice, paternalism, delegation and leader-member exchange (LMX) relationships. Ph.D. Dissertation. University of Miami. Pellegrini, E. K. & Scandura, T. A. (2006). Leader–member exchange (LMX), paternalism, and delegation in the Turkish business culture: An empirical investigation. Journal of International Business Studies, 37(2), 264-279. Pellegrini, E. K., Scandura, T. A. & Jayaraman, V. (2010). Cross-cultural generalizability of paternalistic leadership: An expansion of leader-member exchange theory. Group & Organization Management, 35(4), 391-420. Piccolo, R. F. & Colquitt, J. A. (2006). Transformational leadership and job behaviors: The mediating role of core job characteristics. Academy of Management Journal, 49(2), 327340. Piderot, S.K. (2000). Rethinking resistance and recognizing ambivalence: A multidimensional view of attitudes toward an organizational change. Academy of Management Review, 10, 783–794.

39

Pillai, R., Scandura, T. A. & Williams, E. A. (1999). Leadership and organizational justice: Similarities and differences across cultures. Journal of International Business Studies, 30, 763-779. Preacher, K. J., Rucker, D. D. & Hayes, A. F. (2007). Addressing moderated mediation hypotheses: Theory, methods, and prescriptions. Multivariate Behavioral Research, 42, 185–227. Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y. & Podsakoff, N. P. (2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88, 879-903. Ramesh, A. & Gelfand, M. J. (2010). Will they stay or will they go? The role of job embeddedness in predicting turnover in individualistic and collectivistic cultures. Journal of Applied Psychology, 95(5), 807-823. Richards, D. A. & Hackett, R. D. (2012). Attachment and emotion regulation: Compensatory interactions and leader–member exchange. The Leadership Quarterly, 23(4), 686-701. Robert, C., Probst, T. M., Martocchio, J. J., Drasgow, F. & Lawler, J. J. (2000). Empowerment and continuous improvement in the United States, Mexico, Poland, and India: Predicting fit on the basis of the dimensions of power distance and individualism. Journal of Applied Psychology, 85, 643-658. Rockstuhl, T., Dulebohn, J. H., Ang, S. & Shore, L. M. (2012). Leader-member exchange (LMX) and culture: A meta-analysis of correlates of LMX across 23 countries. Journal of Applied Psychology, 97, 1097-1130. Sagie, A. & Aycan, Z. (2003). A cross-cultural analysis of participative decision-making in organizations. Human Relations, 56, 453-473. Schaffer, B.S. & Riordan, C.M. (2003). A review of cross-cultural methodologies for organizational research: A best-practices approach. Organizational Research Methods, 6(2), 169-215. Schaubroeck, J. & Lam, S. S. (2002). How similarity to peers and supervisor influences organizational advancement in different cultures. Academy of Management Journal, 45(6), 1120-1136. Schaubroeck, J. M. & Shao, P. (2012). The role of attribution in how followers respond to the emotional expression of male and female leaders. The Leadership Quarterly, 23(1), 27-42. 40

Schriesheim, C.A. & DeNisi, A.S. (1981). Item presentation as an influence on questionnaire validity: A field experiment. Educational and Psychological Measurement, 40, 175-182. Schyns, B., Paul, T., Mohr, G. & Blank, H. (2005). Comparing antecedents and consequences of leader-member exchange in a German working contract to findings in the US. European Journal of Work and Organizational Psychology, 14, 1-22. Scott, K. A. & Brown, D. J. (2006). Female first, leader second? Gender bias in the encoding of leadership behavior. Organizational Behavior and Human Decision Processes, 101, 230-242. Stanton, J. M. & Weiss, E. M. (2002). Online panels for social science research: An introduction to the StudyResponse project. Syracuse, NY: Syracuse University, School of Information Studies, Tech. Rep, (13001). Tangirala, S. & Ramanujam, R. (2012). Ask and you shall hear (but not always): Examining the relationship between manager consultation and employee voice. Personnel Psychology, 65(2), 251-282. Taras, V., Kirkman, B. L. & Steel, P. (2010). Examining the impact of culture’s consequences: A three-decade, multilevel, meta-analytic review of Hofstede’s cultural value dimensions. Journal of Applied Psychology, 95, 405-439. Tierney, P. (1999). Work relations as a precursor to a psychological climate for change: The role of work group supervisors and peers. Journal of Organizational Change Management, 12(2), 120-134. Triandis, H. C. (1988). Collectivism v. individualism: Are conceptualization of a basic concept in cross-cultural social psychology. In G. K. Verma & C. Bagley (Eds.), Cross-cultural studies of personality, attitudes and cognition (pp. 60-95). London: Macmillan. Triandis, H. C. & Gelfand, M. J. (1998). Converging measurement of horizontal and vertical individualism and collectivism. Journal of Personality and Social Psychology, 74, 118128. Triandis, H. C. 1995. Individualism & collectivism. Westview Press. Tsui, A. S., Nifadkar, S. S. & Ou, A. Y. (2007). Cross-national, cross-cultural organizational behavior research: Advances, gaps, and recommendations. Journal of Management, 33, 426-478. 41

Vandenberg, R.J. & Lance, C.E. (2000). A review and synthesis of the measurement invariance literature: Suggestions, practices and recommendations for organizational research. Organizational Research Methods, 3(1), 4-70. Van Dam, K., Oreg, S. & Schyns, B. (2008). Daily work contexts and resistance to organisational change: The role of leader-member exchange, development climate, and change process characteristics. Applied Psychology: An International Review, 57(2), 313334. Whelan-Berry, K.S., Gordon, J.R. & Hinings, C.R. (2003). Strengthening organizational change processes: Recommendations and implications from a multi-level analysis. Journal of Applied Behavioral Science, 39, 186–207. Yukl, G. & Fu, P. P. (1999). Determinants of delegation and consultation by managers. Journal of Organizational Behavior, 20(2), 219-232. Zhang, X., Fu, P., Xi, Y., Li, L., Xu, L., Cao, C., Li, G., Ma, L. & Ge, J. (2012). Understanding indigenous leadership research: Explication and Chinese examples. The Leadership Quarterly, 23, 1063-1079.

42

Table 1. Respondents’ demographic information Korea US Gender Male 55% 56% Female 45% 44% Age Under 21 0% 1% 22-34 42% 35% 35-44 32% 29% 45-54 22% 18% 55-64 4% 15% Over 65 0% 2% Marital Single 36% 22% Status Married 63% 66% Domestic Partner 0% 2% Divorced/Separated/Widowed 1% 9% Education High School 10% 13% Junior College 26% 13% Bachelors 53% 47% Masters 8% 20% Ph.D. 2% 7% Rank Entry Level 38% 35% Assistant Managerial Level 17% 28% Managerial Level 25% 28% Director Level 14% 7% Others 6% 2%

Constructs

1

Table 2. Correlations 2 3 4 5

0.02 0.03 -0.14 0.08

7

8

1.

Gender

2.

Age

-0.40

3.

Education

-0.06

0.02

4.

Marital Status

-0.37

0.60

0.03

5.

Rank

-0.24

0.61

0.12 0.35

6.

LMX

-0.06

0.16

0.01 0.08 0.25

7.

ACOC

-0.05

0.12

-0.03 0.03 0.24

0.46

8.

Consultation

-0.10

0.25

0.05 0.11 0.34

0.53

9.

Power Distance

-0.05

0.07

-0.07 0.04 0.07 -0.30 -0.28 -0.18

-0.04

0.17

0.03 0.11 0.25

10. Collectivism

-0.05

6

9

10

0.08

-0.04 -0.09

-0.16

-0.28 0.37 0.01 -0.03 0.04

-0.02 -0.07

0.06

0.04 0.29

0.11

0.03

0.19

-0.05

0.09

0.03

0.03

0.09

0.04

-0.08

-0.01

0.22

0.22

0.27

0.25

0.26

0.47

0.40

0.12

0.27

0.48

0.06

0.15

0.22

0.29

0.28

0.50

0.19

Above Diagonal for US, and below diagonal for Korea Absolute value of correlations higher than 0.14 is significant at p