ORGANIZATIONAL CHANGE RECIPIENTS AND CHOOSING AN OPINION LEADER: A MIXED METHODS INVESTIGATION Except where reference is made to the work of others, the work described in this dissertation is my own or was done in collaboration with my advisory committee. This dissertation does not include proprietary or classified information. _____________________________________ Dean C. Vitale Certificate of Approval: ________________________ ________________________ Hubert S. Feild Achilles A. Armenakis, Chair Torchmark Professor Pursell Professor Management Management ________________________ ________________________ Stanley G. Harris George T. Flowers Torchmark Professor Interim Dean Management Graduate School ii ORGANIZATIONAL CHANGE RECIPIENTS AND CHOOSING AN OPINION LEADER: A MIXED METHODS INVESTIGATION Dean C. Vitale A Dissertation Submitted to the Graduate Faculty of Auburn University in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy Auburn, Alabama August 9, 2008 ii iii ORGANIZATIONAL CHANGE RECIPIENTS AND CHOOSING AN OPINION LEADER: A MIXED METHODS INVESTIGATION Dean Christopher Vitale Permission is granted to Auburn University to make copies of this dissertation at its discretion upon request of individuals or institutions and at their expense. The author reserves all publication rights. ____________________ Signature of Author ____________________ Date of Graduation iv DISSERTATION ABSTRACT ORGANIZATIONAL CHANGE RECIPIENTS AND CHOOSING AN OPINION LEADER: A MIXED METHODS INVESTIGATION Dean C. Vitale Doctorate of Philosophy, August 9, 2008 (M.A., Webster University, 1998) (B.S.B.A., University of Florida, 1988) 112 Typed Pages Directed by Achilles A. Armenakis To better understand the phenomenon of opinion leadership within the context of organizational change, a study of employees was conducted across two banks in the midst of a corporate merger. Qualitative and quantitative data on organizational culture and employee self-report and other-reported attitudes and personality are analyzed from interviews and survey questionnaires. Results suggest significant differences in mid- merger change attitudes between legacy banks. Employees of the dominant bank were more likely to describe their referent organizational opinion leaders (OOL) with work- related terms, whereas employees of the dominated bank were more likely to describe their respective opinion leaders with interpersonal terms expressing positive affect. Neither the strength of influence wielded by organizational opinion leaders (OOL), nor v the leadership style of OOL differed between banks. However, a multi-level analysis suggests OOL job performance predicts OOL strength of influence. vi ACKNOWLEDGEMENTS I am honored and blessed to have reached this milestone. I look forward to the passionate pursuit of learning, teaching, research, and service for the rest of my life. More than any other gift, I am grateful the Lord has blessed me with a passion for the learning process. I am also truly thankful for my wonderful family. Lisa, my wife, and my daughters, Christina and Samantha have fulfilled my life more than I could ask for. Without their love and support, the completion of this dissertation would be nothing. Special thanks to Lisa for her enduring counsel and endless patience with me. I must also acknowledge and thank my professors, Achilles Armenakis, Junior Feild, Stan Harris, Bill Giles and a number of other Auburn Faculty for the guidance and respect they have given me over the last three years. I would also like to thank my mother and stepfather. Their love and support has always been bedrock for me, even in turbulent times. Finally, with his life and his passing, my father continues to provide me the ceaseless motivation to strive for things I once thought beyond my reach. vii Style manual used: Publication Manual of the American Psychological Association (5 th ed.). (2001). Washington, DC: American Psychological Association. Computer software used: Microsoft Word 2003, SPSS 13.0, HLM 5.0. viii TABLE OF CONTENTS LIST OF TABLES AND FIGURES??????????????...........??...... x CHAPTER 1 LITERATURE REVIEW AND HYPOTHESES??????................ 1 Purpose???...??????????????????............??..?.............. 5 Background of Opinion Leadership??????...????..?.................................. 6 Opinion Seeking????????????????..????.............................10 Organizational Change Recipients??????...?...????..?........................... 11 Merger as Organizational Change????????..??..???........................... 13 Conflicting Cultures????????..????????..??............................. 14 Dominant and Dominated?????..??????????..??........................ 15 Affiliation?????..??????????????????........................... 17 Network Perspective????..?????????????..??........................ 18 Choosing a Leader???????..???????????..?........................... 19 Leadership Style?????...??????????????..?......................... 22 Uncertainty and Influence??????????????????........................ 23 CHAPTER 2 METHODS??????????..???????..???............ 26 Overview of Research Design?????????????..??........................... 26 Organizational Context???????????????????........................ 27 Phase 1, Data Collection???????????????..??............................ 29 Phase 1, Measures??????...???..????????..??......................... 30 Phase 1, Analysis??...????????.????????.??........................ 32 Phase 2, Data Collection?.???????.??????.?..??......................... 34 Phase 2, Measures????.??..?????????????............................. 36 Phase 2, Analysis???????????????????....?......................... 43 CHAPTER 3 RESULTS?????????????????????.............. 50 Phase 1, Hypothesis???...????..??????????.?........................... 50 Phase 2, Hypotheses....??.???..?????????????......................... 55 Phase 2, Research Questions?.????..?????????..?...........................65 CHAPTER 4 DISCUSSION??..????????????????..?............ 70 Mergers and Employee Attitudes..................................................................................... 71 Describing OOLs............................................................................................................. 71 OOL Influence ?????????..????????????.......................... 74 Limitations....................................................................................................................... 75 Implications and Areas for Future Study???????????............................. 77 ix REFERENCES??????????????..?????????.................. 82 x LIST OF TABLES AND FIGURES Tables 1. Summary of hypotheses and research questions????????......................... 25 2. Demographic profile of interview participants??????.?.............................. 30 3. Interview questions????.???????????.??............................... 32 4. Summary of measures??????????????.????........................ 38 5. Survey item choices for top three organizational opinion leader attributes................43 6. Organizational opinion leader traits typology.............................................................45 7. Example comments by change recipients used to describe organizational opinion leaders................................................................................... 46 8. Responses comparing cultures between pre-merger banks........................................ 52 9. Responses comparing cultures between pre-merger banks and the post-merger bank.......................................................................................... 53 10. Reliabilities, descriptive statistics, and partial correlations of job attitudes and merger perceptions..................................................................... 56 11. Overall MANOVA for pre-merger bank of employment............?......................... 57 12. Univariate analyses for pre-merger bank of employment ??................................ 58 13. Classification results for pre-merger bank of employment ?..?............................ 59 14. OOL trait category frequencies across banks??????...?............................ 60 xi 15. Styles of OOLs chosen by change recipients at the dominant and dominated banks ????????.?........................................... 64 16. HLM descriptive statistics and correlations for mode l............................................ 66 17. HLM descriptive statistics and correlations for model 2?...................................... 67 18. Summary of HLM variables predicting OOL strength of influence.................................................................................................. 68 19. Summary of hypotheses and research questions support.????......................... 69 20. A manager?s guide to recognizing OOLs................................................................. 80 Figures 1. Summary Phase 1 findings identifying the dominant and dominated bank................................................................................................... 54 2. Most important OOL attributes..............????.?????................................ 61 3. Most important OOL attributes by pre-merger bank?????.............................. 62 1 CHAPTER 1 LITERATURE REVIEW AND HYPOTHESES Management?s use of respected peers to increase the probability of successful organizational change has long been recognized in the field of organizational science (Ryan & Gross, 1943). Changes ranging in magnitude from fundamental (e.g., organizational re-structuring) to incremental (e.g., continuous improvement programs), are often spread horizontally via peer networks (Rogers, 2003). Persuasive coworkers can promote changes within an organization by influencing group members. As respected peers, opinion leaders in organizations often act as horizontal change agents spreading positive change attitudes and information throughout the organization (Armenakis & Fredenberger, 1997; Mohrman, Tenkasi, & Mohrman, 2003). Equally as compelling, as Kahn, Cross, and Parker (2003) have shown, informal networks led by opinion leaders can be instrumental in resisting change. In fact, the success of change in an organization may very well depend on identifying opinion leaders and gaining their support to influence group members towards change adoption (Maienhofer & Finholt, 2002; Rogers, 2003). The advantage of comprehending the ?unseeable,? that is, learning more about the informal networks of persuasion and influence that exist in every organization, is useful to both practitioners and scholars alike. Understanding that influence and leadership are not reserved for the highest levels of the organization, but extend throughout the organization, even to the lowest levels, is 2 crucial to recognizing the importance of informal influence networks within an organization (Dicke, 2004). Smith (2005, p. 565) suggested that an organization?s strategic success is ? ?critically dependent on its having the capability to visualize relationship networks among its employees, and the means to identify and leverage as appropriate, patterns of positive or negative influence.? According to most managers and executives, their ability to successfully plan and execute strategy is reliant on their knowledge of their organization?s workforce. The consulting organization McKinsey, warned that ?companies shouldn?t focus so much on formal structures that they ignore the informal ones? (Cross, Parise, & Weiss, 2007, p. 31). Familiarity with existing interpersonal influence networks can assist change managers in facilitating change within an organization. Larkin and Larkin (1996) reinforced this notion, insisting that confidence in the formal communication structure of the organization is misplaced; it is the credible informal sources that most influence employees when old practices are challenged. In their study of accelerating innovations in organizations, Valente and Davis (1999) suggested that ?A network should be ? used rather than ignored when creating programs? (p. 56). Accordingly, identifying these ?possessors of influence? and understanding the flow of persuasive communication within an organization is important to global and local change agents (Lam & Schaubroeck, 2000; Vishwanath, 2006). Put simply, a change manager?s awareness and effective use of the informal communication network can make or break a change effort. Despite the importance placed on the need to better understand informal communication and the flow of peer influence within organizations in the context of organizational change, relatively little research exists on the subject (Oreg, 2006). The 3 vast majority of research and practitioner literatures in the field of organizational change focus on understanding change recipient reactions as an effect of organizational-level variables (e.g., Armenakis & Harris, 2002; Ashford, 1988; Coch & French, 1948; Goltz & Hietapelto, 2002; Lines, 2004) and to a lesser extent individual differences (e.g., Cunningham et al., 2002; Judge, Thoreson, Pucik, & Welbourne, 1999; Oreg, 2003). Consequently, whereas employees? resistance to change has been predicted by a number of context-specific antecedents, individual traits, or a combination thereof (e.g., Wanberg & Banas, 2000), the impact of social influence on change recipients has received relatively short shrift (for exceptions, see Lam & Schaubroeck, 2000; Oreg, 2006). Garnering support among organizational members is generally viewed as critical to the successful implementation of significant organizational change initiatives (Armenakis, Harris, & Mossholder, 1993; Burke, 2002; Coch & French, 1948; Kahn et al., 2003). Recognizing this imperative, researchers have focused much of their attention on better understanding factors that form employees? reactions, hoping to improve the organization?s ability to influence the extent of support or acceptance for a specific change program (Pederit, 2000). The literature documenting how best to launch and manage change (both research- and practitioner-oriented) is substantial. Yet, despite the well-documented scholarly advancement in organizational theory, most change initiatives in the real-world fall short of expectations (Burke 2002; Gilmore, Shea, & Useem, 1997; Probst & Raisch, 2005). Thus, the high failure rate of planned change programs begs the question, ?What don?t we know about change recipient reactions?? The identification of opinion leader descriptions in organizations experiencing differing conditions of change will help change managers avoid what Rogers (2003) 4 described as ?wearing out the credibility of opinion leaders? (p. 388). This loss of credibility results from being too frequently identified with management and only parroting the party line. Hence managers, as change agents, are advised to use opinion leaders judiciously, so as to preserve their influence. Research has suggested that developing the capacity to visualize the relationship network among employees and leverage appropriate patterns of influence is critical to managerial effectiveness (Smith, 2005). Understanding how change recipients? change uncertainty and attitudes towards the organization are related to their opinion seeking behaviors should help managers identify the appropriate individuals to diffuse change throughout the organization. The existence of leaders is predicated on the reality of followers. Thus, each individual determines whom he or she recognizes as a leader. This dissertation examines how change recipients? attitudes and work context may be related to the characteristics of those whom among their co-workers they choose to listen to, and the extent to which they are then influenced by this referent leader. The present research introduces the concept of the organizational opinion leader (OOL) to refer to the organizational change recipient?s referent leader. By virtue of identifying the key follower-nominated attributes of influential co-workers within the workplace, the present research will assist managers in recognizing the individuals more likely to be opinion leaders within the organization. Further development of an organizational opinion leadership concept will extend and expand the current marketing-borne construct of consumer opinion leadership to the organization and increase its usefulness for both scholars and practitioners. The present study may assist change managers to better understand the relative importance of informal networks to formal chains of authority with respect to planned change. This 5 dissertation responds, in part, to Lam and Schaubroeck?s (2000) observation that ?Future studies could use more naturalistic methodologies (perhaps qualitative tools) to understand ?what it is about opinion leaders that make them influential to their peers? (p. 994). Purpose The present mixed methods study addressed organizational opinion leadership in the context of organizational change. A triangulation mixed methods design was used, a type of design in which different, but complementary data are collected on the same topic (Creswell, 2003; Paul, 1996). In this study, qualitative interviews, written responses to open-ended questionnaire items, and archival data were used to explore organizational opinion leader (OOL) attributes in the context of organizational change for the employees of two banks engaged in a corporate merger. Following qualitative data collection, a written questionnaire containing closed- and open-ended items was used to test the effects of change recipient post-merger attitudes and follower-reported attributes of opinion leaders. Both qualitative and quantitative data were collected so as to bring together the strengths of both forms of research to corroborate results. Overall, the intent of the present study was to gain a better understanding of the phenomenon of opinion leadership in the context of a focal organizational change involving the merger of two banks. The specific goals were: (a) develop a profile of OOL traits for each merger- partner bank and compare the profiles, and (b) determine which traits are associated with the OOL?s strength of influence and compare the significant traits across merger partner banks. 6 The present study addresses the investigation of OOL using a three-fold approach. First, borrowing from existing consumer behavior and organizational change literatures, the concepts of opinion leadership and change are reviewed. Traditional consumer opinion leadership theory is described and adapted to a concept of organizational opinion leadership appropriate to the interpersonal relationships existing within an organization. Secondly, the merger of the two participant banks is hypothesized as an extreme form of organizational change and its effects on individuals is examined. Change recipient attitudes are discussed as they may relate to OOL attributes in the context of large-scale change and specific hypotheses and research questions are developed. Finally, inductively- and theoretically-developed OOL attributes are explored and empirically tested across change recipients in both banks. Background of Opinion Leadership Opinion leadership has been conceptualized and measured in a variety of ways since Lazarsfeld, Berelson, and Gaudet (1948) first introduced the concept to explain voter preferences in a presidential campaign. Their work served as the basis for the opinion leadership construct and was further developed in the consumer behavior literatures by Katz and Lazarsfeld (1955), Cartwright and Zander (1960), Rogers (1961), and Rogers and Cartano (1962). A general definition of opinion leadership was offered by Rogers and Cartano: ?opinion leaders are individuals who exert an unequal amount of influence on the decisions of others? (p. 435). King and Summers (1970) developed the first version of the opinion leadership construct; measured with a scale later revised by Childers (1986), which remains the most common measure of opinion leadership (Flynn et al., 1996; Lam & Schaubroeck, 2000). Rogers and Cartano (1962) clearly defined 7 influence as the salient effect of opinion leadership. However, the examination of the evolved opinion leader construct as reflected in both the King and Summers (1970) and Childers (1986) measures, reveals items more likely to measure an inclination to communicate, rather than a capacity to influence other consumers (Flynn et al., 1996). While opinion leader influence generally requires communication between the influencer and the recipient, communication is most appropriately viewed as a process, whereas influence is an outcome (Rogers, 2003). As such, it appears that influence is fundamental to the theory of opinion leadership (Yale & Gilly, 1995). Care should be taken as change scholars increasingly appropriate the concept of opinion leadership (e.g., Lam & Schaubroeck, 2000). Organizational change research has examined opinion leadership as it relates to change agency. For example, using opinion leaders as change agents has been shown to, increase employee performance in the service sector (Lam & Shaubreck, 2000), boost employee participation in knowledge sharing programs (Smith, 2005), and influence justice perceptions (Umphress, Labianca, Brass, Kass, & Scholten, 2003). Traditional measures of consumer opinion leadership focus on purchase behaviors (early adopter/non adopter), product familiarity, demographics, and status (Rogers, 2003). Owing to their lack of interdependent and interpersonal dynamics, consumer behaviors may not be wholly relevant to opinion leadership, or influence, within an organizational context. Consumer opinion leadership, as an artifact of the macro-oriented consumer behavior research, ignores the interpersonal aspects of communication and influence endemic to an organization. Therefore, as the concept of opinion leadership is adopted by organizational change scholars, it is necessary that the construct be further developed to reflect organizational reality. Indeed, 8 further advancement of opinion leadership as conceptualized in an organizational context may be constrained by limitations in the current consumer theory conceptualization (Flynn, Goldsmith, & Eastman, 1996). While the traditional theory of opinion leadership is well-established in the consumer behavior, marketing, and innovation diffusion literatures (Keller & Berry, 2003; Rogers, 2003; Weimann, 1991, 1994), it lacks key conceptual dimensions relevant to the interpersonal communications present in the every-day workplace (Yale & Gilly, 1995). By virtue of its scholarly roots in marketing and consumer theory and research, the existing opinion leadership and peer influence research has focused on new-product communication and innovation diffusion in larger social systems (see Rogers, 2003 for a complete review). Diffusion of innovation and consumer theory literatures describe the conveyance of influence through a largely impersonal mechanism in which structural parameters such as centralization and decentralization play key roles, but leadership qualities like trustworthiness and sympathy, for example, and social concepts such as interdependence and reciprocation are noticeably lacking (e.g., Goldsmith & DeWitt, 2003). It is also unclear if the traditional opinion leadership literatures distinguish between the cause and effect of trait variables. For instance, early adoption of innovation is often described as a trait of opinion leaders, despite the lack of demonstrated influence by the supposed influencer (Flynn et al., 1996). That is, it is assumed in much of the consumer behavior literature that early adoption equates to opinion leadership (e.g., Venkatraman, 1990). Opinion leaders are typically perceived as persons of a social status that is similar to, or slightly higher than, that of the person they influence (Rogers, 2003). Lam and 9 Schaubroeck (2001), in their study of frontline opinion leaders as change agents suggested that employees may be more readily influenced by the opinions of those with whom they work. Much as consumers turn to influential others for advice during their decision-making process (Chan & Misra, 1990), employees are likely to turn to influential peers in the face of uncertainty created by organizational change (Ryan & Gross, 1943; Umphress et al., 2003). According to social comparison theory (Festinger, 1954), people compare themselves to other similar people, and these comparisons have a greater effect when a social-based comparison is not available (Kilduff, Tsai, & Hanke, 2006). Research investigating the influence processes between friend and acquaintances has suggested that workgroup peers are readily available as referent others (e.g., Krackhardt & Porter, 1986; Newcomb, Koenig, Flacks, & Warwick, 1967). The identity of opinion leaders (chosen by peers) has been shown to vary according to situational influences (Umphress et al., 2003), such as personal attributes related to trust and sociability (Katz & Lazarfeld, 1955); innovativeness (Rogers, 2003); and knowledge in the specific areas of interest (Montgomery & Silk, 1971; Reynolds & Darden, 1971). Opinion leaders in these varying contexts are described as influential peers and identified as the individuals at the center of their organization?s communication network (Lam & Schaubroeck, 2000; Rogers, 2003). Although the organizational change research on opinion leadership is less developed than its consumer-driven counterpart, it has recently received increased attention in the form of research investigating peer influence, advice networks, and change agency among fellow employees (e.g., Bono & Anderson, 2005; Lam & Schaubroeck, 2000; Oreg, 2006). 10 Opinion Seeking It is no surprise that the vast majority of research on opinion leadership has focused on the leaders, rather than opinion seekers. The reality that the subsistence of leaders is predicated on the existence of followers notwithstanding, the historic quest to decode leadership has trumped the study of followers. Similarly, organizational change studies have focused primarily on the change agent perspective. Whereas change recipients? interpretations of change in organizations are given little attention (Bartunek, Rousseau, Rudolph, & DePalma, 2006), it is change recipient reactions, generally portrayed as support, compliance, or resistance, which has received most of the focus (e.g., Armenakis, Harris, Cole, Fillmer, & Self, 2007; Oreg, 2003). An exception to the change agent focused research is the study of change recipient emotions during organizational change and sensemaking in the organization (e.g., Bartunek et al., 2006; Mossholder, Settoon, Armenakis, & Harris, 2000). It is the change recipient?s search for information, support, advice, or opinion, to facilitate their sensemaking amidst change that obliges interpersonal communication (Roberts, Dutton, Spreitzer, Heaphy, & Quinn, 2005). Opinion seeking is described in the present study as a mechanism activated by change recipients so as to contribute to their interpretation of, and subsequent reaction to change. Opinion seeking behavior is not limited to the explicit verbal solicitation of opinions. Opinion seekers may be inspired by their referent others to more subtle behaviors. Opinion seeker behaviors include seeking advice, assurances, and confirmation of feelings, as well as imitation of the opinion leader (Engel, Blackwell, & Miniard, 1990; Flynn et al., 1996). Whether the opinion seeker?s methods are more 11 active or merely passive, the opinion seeker is influenced to varying degrees. The change recipient?s act of opinion seeking can be viewed as a behavioral outcome in response to organizational change. This perspective is consistent with Oreg?s (2006) typology of change recipient reactions as affective, cognitive, and behavioral. As Lewis, Agarwal, and Sambamuthy (2003) observed, ?Via internalization the individual may incorporate the opinion of a referent peer as part of their own belief structure. Via identification the individual will seek to believe and act in a manner similar to those possessing referent powers? (p. 662). It follows that influential co-workers, identified as OOLs in the present study, are likely to influence change recipient attitudes about organizational change outcomes. Whereas it is tempting to classify individuals as either opinion leaders or opinion seekers, it is flawed sorting, in that opinion leaders within an organization are also opinion seekers. Individuals within an organization routinely exhibit opinion leading and influence behaviors, as well as opinion seeking behaviors (Flynn et al., 1996). Given that opinion seekers in the present study were investigated in the context of organizational change, opinion seeking is viewed as a behavior of change recipients. Thus, the terms ?opinion seeker? and ?change recipient? are treated synonymously for the purposes of the proposed study. Organizational Change Recipients Change recipients? attitudinal and behavioral reactions to organizational change have received an abundance of scholarly attention. It is generally accepted that when faced with change, employees experience varying degrees of stress and engage a number of cognitive and coping processes so as to reduce the stress associated with the change 12 (Armenakis, Bernerth, Pitts, & Walker, 2007; Burke, 2002; Judge et al., 1999; Oreg, 2006; Rafferty & Griffin, 2006). An individual?s attitude toward change is recognized as having both a cognitive (knowing) and affective (feeling) component (Ashkanasy, Hartel, & Daus, 2002; Mossholder et al., 2000). Employee reactions to organizational change efforts are often affect-laden with fears of uncertainty and failure (Armenakis & Bedeian, 1999; Callan, 1993; Coch & French, 1948). Research suggests the anxiety created by ?fear of the unknown? is a primary source for initial change resistance (Armenakis & Fredenberger, 1997; Miller & Monge, 1985; Schweiger & DeNisi, 1991). Accordingly, the change recipient?s perception of change magnitude is linked with their perceived uncertainty regarding the focal change (Armenakis & Harris, 2002; Harris & Mossholder, 1996). It has become axiomatic to say that providing information to the change recipient is the number one prescription to reduce change uncertainty, which is generally regarded as the key pre-cursor to change resistance (Armenakis & Fredenberger, 1997; Judge et al., 1999). Implicit in this observation is the change recipient?s desire for information to fill perceived knowledge voids (Armenakis & Harris, 2002). In times of change, either due to the inability of organization leadership to provide enough information, or because of a perceived lack of management credibility, peers often seek information horizontally from peers rather than vertically in the organization (Bordia, Jones, Gallois, Callan, & Difonzo, 2006). Much of this communication can be detrimental to the organization. Bordia et al. (2006), in a study of rumor proliferation and organizational change stress, found a significantly greater number of negative rumors foretelling undesirable outcomes (e.g., layoffs as a result of an organizational merger), rather than positive rumors 13 describing a wished-for outcome (e.g., there is going to be a larger than usual Christmas bonus) circulating in organizations undergoing change. Merger as Organizational Change Since the advent of merger-mania in the 1980s, organizational combinations have remained a very popular business strategy for attaining growth and diversification (Fowler & Schmidt, 1988; Houghton, Anand, & Neck, 2003). Mergers can be described as acute forms of organizational change, and as such, are often perceived by employees as threatening, due to feelings of vulnerability and fears of losing security (Saunders & Thornhill, 2003). Employees in one, or both firms, often feel ?stressed, disoriented, frustrated, confused, and even frightened? due to the fear of large-scale changes in organizational structures, systems, processes, and cultures (Buono & Bowditch, 1989, p. 3). On a personal level, these feelings may lead to a sense of loss and myriad mental health problems, to include depression or even suicide (Cartwright & Cooper, 1993; Folkman, Lazarus, Gruen, & DeLongis, 1986). On an organizational level, mergers have come to be associated with a variety of detrimental workplace attitudes and behaviors such as lowered morale, job dissatisfaction, disloyalty, distrust of the organization, acts of sabotage and petty theft, increased absenteeism and turnover, and work team dysfunction (Buono & Bowditch, 1989; Marks, 2006; Sinetar, 1981). Research on mergers suggests change recipients? adverse attitudes and behaviors are caused by stress that is triggered by the expectancy of change, and fear of the future, rather than the merger itself (Cartwright & Cooper, 1993). Thus, it is more the fear of the unknown created by the prospect of change, rather than the change itself that often affects 14 employees. As Buono and Bowditch (p. 11, 1989) pointed out, merger-related stress results from legitimate questions on employees? minds during the merger period: Will my employment be terminated? Will I get a promotion? A demotion? Will I have to relocate? What will happen to my benefits? What kind of organization will we be? What is our mission? What are the values of the new organization? What types of transition will the firm undergo? How are we all going to deal with this change? What does all this mean to me? Consequently, employees affected by a merger often experience what Marks and Mirvis (1986) described as ?merger syndrome,? a highly defensive, pessimistic response to the uncertainties of the future. Interestingly, due to these uncertainty-induced fears, mergers are deemed a ?stressful life event? for employees, even when there is thought to be a high degree of cultural compatibility between the partnering organizations (Cartwright & Cooper, 1993, p. 327). Conflicting Cultures Organizational culture is generally referred to as a system of shared meanings held by members that distinguishes the organization from other organizations (Schein, 2004). According to consultants A.T. Kearney, more than half of mergers fail to reach the value goals set by top managers (Gumbel, 2003). The primary cause of failure is conflicting organizational cultures (Robbins & Judge, 2007). For example, the failed merger of AOL and Time Warner is commonly attributed to culture clash. As one observer noted, ?In some ways the merger of AOL and Time Warner was like the marriage of a teenager to a middle-aged banker. There were open collars at AOL. Time Warner was more buttoned-down? (Gail, 2003, p. 2). Each organizational culture is 15 unique, not unlike personality is specific to an individual (Schein, 2004). Cultures can also be differentiated in terms of strength. In a strong culture, the organization?s core values are both intensely held and widely shared (Wiener, 1988). The differences between cultures in the context of the merger are relevant to the present study as they result in employee attitudes and behaviors toward the new post-merger organization. Dominant and Dominated Partners In almost all cases, mergers affect one party more than the other; rarely is there a compromise between cultures (Buono & Bowditch, 1989). Moreover, Cartwright and Cooper (1992) argued that merger is primarily a legal term, and that from a psychological perspective, this type of corporate combination is more accurately referred to as a takeover. They posit that the post-merger organization will embody the culture of the dominant pre-merger partner organization. While the merger partners may go to great lengths to convey the notion of equality, one partner usually dominates the other because it is bigger, wealthier, more viable, or otherwise more powerful and influential than its partner (cf. Rentsch & Schneider, 1991). Owing to its role as the acquiring organization, the dominant organization is more influential in shaping the merged organization than the dominated organization. Thus, the merged organization is likely to resemble the dominant partner, more so than the dominated partner. Whereas the employees of the dominant organization are more likely to have a greater sense of continuity because they expect the post-merger organization to be their organization, employees in the dominated merger partner are expected to perceive the greatest degree of change (Van Knippenberg & Van Leeuwen, 2001). In addition to change recipient attitudes, a variety of cultural indicators, ranging from leadership 16 turnover associated with the merger to new policies and procedures put in place for the post-merger entity, may be used to identify the dominant and dominated merger partners (Buono, Bowditch, & Lewis, 1985). Based on the preceding discussion, it would seem reasonable to suggest the merged organizational culture investigated in the present study will resemble one of the merger partner cultures more than the other. Hypothesis 1: Employee perceptions of the magnitude of merger-induced changes will vary according to their pre-merger bank of employment, such that the pre-merger partner banks can be described as dominant and dominated. Schein (2004) recognized organizational climate as an authentic measure for culture analysis, suggesting ?The climate of the group is an artifact of the deeper cultural levels? (p. 26). In turn, as organizational climate is generally reflective of members? shared attitudes (cf.; Hicks-Clarke & Iles, 2000; Lewin K, Lippitt & White, 1939; Mayer, Nishii, Schneider, & Goldstein, 2007), the analysis of change recipient attitudes is deemed particularly suitable to the exploration of organizational culture differences. Given the well-documented evidence linking affective responses (of the type created during large-scale organizational change) to job attitudes (Mossholder et al., 2000; Staw, Sutton, & Pelled, 1994; Weiss & Cropanzano, 1996), it is estimated that change recipients? uncertainty regarding the merger, as well as job attitudes, will vary according to whether they were employees of the dominant, or dominated, merger partner. Furthermore, it is expected that employees of the dominated organization will possess more adverse job and merger attitudes. Hypothesis 2: Employees of the dominated bank will report lower job satisfaction, lower perceived organizational support, greater merger uncertainty, less 17 merger valence, less merger efficacy, and higher intention to turnover than employees of the dominant bank. Affiliation Communication exists because, as Aristotle (350BCE/1995, p. 12) famously wrote, ?man is a ? social animal.? Furthermore, as socially discriminate creatures, we have some degree of choice regarding with whom we have interdependent social relations. Face-to-face communication has consistently been judged to be more persuasive than other forms of communication (Mascitelli, 2000). Our judgments regarding another?s suitability for affiliation is central to what it means to be social. Affiliation choices are necessarily discriminate as individuals will choose some persons over others (Kurzban & Neuberg, 2005). In a seminal study of the likableness ratings of 555 personality adjectives, Anderson (1968) found that the 10 most likeable characteristics were sincerity, honest, understanding, loyal, truthful, trustworthy, intelligent, dependable, open-minded, and thoughtful. Cottrell, Neuberg, and Li (2007) further explored Anderson?s work, placing it in a socio-functional context. According to Cottrell et al. (2007) the desired personal attributes of individuals with whom individuals choose to affiliate vary according to relationship type and context. For example, their findings indicated that the follower-perceived importance of the affiliate trait trustworthiness was stable across all participant relational groups (e.g., work teams, athletic groups, family), whereas the perceived importance of attractiveness as an affiliate trait, varied according to relational group. Implicit in this discussion of valued traits is the recognition that a hierarchy of attributed characteristics for affiliates also applies to opinion leaders within an 18 organization. Both the follower-rated attributes of OOLs and the level of influence wielded by the OOLs are of interest in the present study. As previous research has demonstrated, the perceived value of these characteristics varies according to the specific context (Cottrell et al., 2007). The merger of two banks and differing change recipient reactions to the merger, form the contextual landscape for this study. A Network Perspective An organization?s communication (or social) network consists of interconnected individuals linked by informal patterned flows of information that are described as social ties (Tenkasi & Chesmore, 2003). Social network ties can be viewed as conduits for the flow of information between individuals within an organization (Monge & Contractor, 2001). An analysis of social ties in aggregate reveals a snap-shot of informal linkages in the organization. The ties differ in the amount and type of information and social support they provide to the actor. According to Umphress et al. (2003) people will seek out different social ties based on the type of information or support they desire. Thus, networks are created at the individual level through the activation of social ties. Differentiating between the instrumental and the expressive network enables a comparison of the cognitive and affective influence patterns in the organization (e.g., Fombrun, 1982; Umphress et al., 2003). On the one hand, instrumental ties are described as work-related ties, and can be explained as links used by employees to obtain information, advice, and resources necessary to complete a task (Balkundi & Harrison, 2006; Ibarra, 1993). On the other hand, expressive ties are described as ?expressions of interpersonal affect,? and are characterized by positive affect (Umphress et al., p. 742); whereas instrumental ties are information and cognition based. Expressive ties can be a 19 source of social support, may provide a sense of identity and personal belonging, and are used to convey normative expectations (Coleman, 1988). In their study of the effects of perceived ambiguity on organizational justice ties, Umphress et al. (2003) demonstrated employees? increased reliance on expressive ties rather than instrumental ties when experiencing ambiguity. Similarly, organizational change researchers have prescribed communicating information to change recipients as the key to reducing perceived uncertainty thereby reducing resistance to change (Armenakis & Fredenberger, 1997; Judge, et al., 1999). Choosing a Leader A key assumption of the present research is that individual preferences for particular types of leaders will vary. An abundance of theory and literatures regarding social influence (e.g. Byrne, 1969; Ibarra, 1993) and motivation (e.g., Lawler, 1973; McClelland, 1985; Miner, 1978) records the considerable influence of similarity attraction and need-satisfaction on attitudes. Kouzes and Posner (1995) suggested that individuals are most likely to be motivated by leaders who provide a means toward fulfillment. Meeting one?s intangible needs such as learning, self-worth, pride, and competence serves to intrinsically motivate the individual to a point of enjoyment (Fox, 1994). It is this state of, or the prospect of, enjoyment that is challenged in the face of change (Kouzes & Posner, 1995). Indeed, as discussed, the prospect of large-scale change, such as that which might occur as a result of the merger, threatens far more than enjoyment; fears of termination and demotion often run rampant (Buono et al., 1985). Whereas the need-satisfaction concept is well-represented in the scholarly and practitioner literatures regarding the effectiveness of formal leaders in the face of 20 organizational change, less is understood as it applies to informal leadership (Bono & Anderson, 2005). The similarity-attraction paradigm suggests that similar individuals will be interpersonally attracted to one another (Baugh & Graen, 1997; Chattopadhyay, Glick, Miller, & Huber, 1999; Festinger, 1954). Support for this theory is found in two parallel streams of literatures. Relational demography research suggests individuals? perceived similarities and dissimilarities with the demographic characteristics of fellow workgroup members affect their behavior (Chatman, Polzer, Barsade, & Neale, 1998; Tsui, Xin, & Egan, 1995). The literature also suggests that demographic similarity (Tsui & O?Reilly, 1989) and attitudinal similarity (Byrne, 1969) are important predictors of behavioral outcomes such as performance, satisfaction, leadership, and communication (Blau, 1985; Parsons & Liden, 1984). Individuals are also prone to judge those with whom they share similar demographic characteristics as possessing the same values, principles, and beliefs as themselves (McGrath, 1998; Milliken & Martins, 1996). More recently, communications and social network literatures have suggested similar findings; generally describing the tendency for communications to flow between similar individuals as homophily (Feder & Savastano, 2005). Characteristic similarities such as age, gender, wealth, occupation, and other socio-economic attributes have been shown to affect the influence opinion leaders have on their peers (Chatman et al., 1998; Feder & Savastano, 2005; Valente & Davis, 1999). According to Weimann (1994), ?People turn to seek advice from their peers, from individuals of the same background, interests, and values? and furthermore ?The flow of information and influence is likely to be rather horizontal? (p. 58). Based on the preceding discussion, it is expected that 21 employees in both banks, as change recipients, will value OOL attributes that (a) are affectively positive and interpersonal, and (b) reflect similarity with self. Accordingly, they will describe their OOL more often with these interpersonal attributes than with work-related attributes. Hypothesis 3: Employees will tend to describe organizational opinion leaders in terms of positive interpersonal affect and similarity-with-self, rather than work-related attributes. Additionally, as employees at the dominated bank are expected to have experienced a larger degree of organizational change it is thought they will be more likely to value interpersonally affective OOL attributes than their counterparts at the dominant bank. Conversely, employees at the dominant bank will be more likely to describe their OOLs with work-related attributes, than the employees at the dominated bank. Hypothesis 4a: Employees of the dominated bank will tend to describe organizational opinion leaders in terms of positive interpersonal affect and similarity- with-self, more so than employees of the dominant bank. Hypothesis 4b: Employees of the dominant bank will tend to describe organizational opinion leaders in work-related terms, more so than employees of the dominated bank. Given the opinion leadership theory discussed, it is expected that OOLs will be of elevated status among their peers (Rogers, 2003). As the present study is interested in further describing the attributes of OOLs, the following hypothesis is offered. Hypothesis 5: Employees at both banks will describe their organizational opinion leaders as holding a higher position at work than themselves 22 Leadership Style Leadership style refers to the manner in which leaders relate to followers. Many different labels have been used to categorize leadership styles. However, most classification schemes can be broken into three categories: task-oriented, relationship- oriented, and charismatic (Ehrhart & Klein, 2001). The relationship-oriented leadership style is associated with actions directed toward maintaining good interpersonal relationships between the leader and his or her followers. Relationship-oriented leaders treat subordinates with kindness and respect, emphasize communication with and listening to subordinates, show trust and confidence in subordinates, and provide recognition and show appreciation for subordinates' contributions. The leader/follower relationship is characterized by mutual trust, a high regard for subordinate's feelings and respect for their ideas. Whereas relationship-oriented leaders place importance on the social support of their followers, the task-oriented leader?s actions are concerned primarily with task accomplishment. Task-oriented leaders are likely to guide subordinates in setting performance goals, plan and schedule work, provide necessary supplies, equipment, and technical assistance, and coordinate subordinate activities. Lastly, charismatic leaders tend to focus on motivation. They communicate high performance expectations to followers, exhibit confidence in followers' abilities to reach goals, and articulate value-based overarching vision and collective identity (Ehrhart & Klein, 2001). Based on the understanding of the relationships between change recipients? uncertainty, job attitudes, and followers? preferences for varying leadership styles, it is expected that during an organizational merger the change recipients of the dominated 23 bank are more likely to seek emotional support than the employees of the dominant bank. Thus, the employees at the dominated bank will tend to describe their nominated OOLs with terms that are relationship-oriented, rather than task- or charismatic-oriented, more so than the employees at dominant bank. Hypothesis 6: Employees of the dominated bank will be more likely to describe their organizational opinion leaders as relationship-oriented, whereas employees of the dominant bank will be more likely to describe their organizational opinion leaders as task-oriented. Uncertainty and Influence Bordia et al. (2006) characterized uncertainty as a sense of doubt or confusion about ambiguous current events and their import for the future. Uncertainty is further described as an adverse psychological state that motivates uncertainty reduction strategies such as information seeking behaviors (Ashford & Black, 1996; Rosnow, 1991). As previously discussed, these fears for the future coupled with a desire for information and a perceived lack of available trustworthy information, motivate opinion seeking behaviors. The traditional social influence paradigm (Deutsch & Gerrard, 1955) suggests that when employees experience uncertainty as to how to react to a given stimulus (e.g., a new program or policy change), they often make use of social comparisons in an attempt to either gather additional (or more credible) information upon which to base their reaction, or to better assess which mode of action will help them construct a more positive self-image (Bamberger & Biron, 2007). In a study of the effects of group norms on excessive absenteeism, Bamberger and Biron concluded that referent peers influenced 24 absence behavior above and beyond larger organizational-level norms. In their research on self-social comparison, Hogg and Abrams (1993) suggested that change recipient uncertainty promotes in-group identification; thus there is a tendency to rely more strongly on the members of their peer referent group. Thus, it is predicted: Hypothesis 7: Organizational opinion leaders in the dominated bank will be described as more influential than organizational opinion leaders in the dominant bank. Also relevant to the present study is the relationship of specific trait descriptions of in situ OOLs, as they may relate to the strength of OOL influence. The effects of legacy bank employment as a moderator of this relationship also bear investigation. Given the extant social-psychological literatures presented, it is difficult to ascertain the salient traits embodied by an OOL as they affect influence. Lacking sufficient theory in an applied organizational setting to support a conceptual hypothesis regarding these relationships, it seems worthwhile to answer the following questions: Research Question 1: Which attributes of the organizational opinion leaders are positively associated with their strength of influence over followers? Research Question 2: Do the organizational opinion leader attributes associated with strength of influence vary according to pre-merger bank of employment? Table 1 provides a summary of the research hypotheses and the research questions for the present study. The first and second hypotheses assess the impact of the corporate merger on change recipients? job attitudes and merger perceptions in both banks. Hypotheses 3-6 investigated change recipient descriptions of OOL attributes and assesses any differences due to pre-merger bank of employment. Hypothesis 7 and the research questions explored predictors of OOL strength of influence. 25 Table 1 Summary of Hypotheses and Research Questions Study Phase Hypothesis 1 Employee perceptions of the magnitude of merger-induced changes will vary according to their pre-merger bank of employment, such that the pre-merger partner banks can be described as dominant and dominated. 1 Hypothesis 2 Employees of the dominated bank will report lower job satisfaction, lower perceived organizational support, greater merger uncertainty, less merger valence, less merger efficacy, and higher intention to turnover than employees of the dominant bank. 2 Hypothesis 3 Employees will tend to describe organizational opinion leaders in terms of positive interpersonal affect and similarity-with-self, rather than work-related attributes. 2 Hypothesis 4a Employees of the dominated bank will tend to describe organizational opinion leaders in terms of positive interpersonal affect and similarity-with-self, more so than employees of the dominant bank. 2 Hypothesis 4b Employees of the dominant bank will tend to describe organizational opinion leaders in work-related terms, more so than employees of the dominated bank. 2 Hypothesis 5 Employees at both banks will describe their organizational opinion leaders as holding a higher position at work than themselves. 2 Hypothesis 6 Employees of the dominated bank will be more likely to describe their organizational opinion leaders as relationship- oriented, whereas employees of the dominant bank will be more likely to describe their organizational opinion leaders as task-oriented. 2 Hypothesis 7 Organizational opinion leaders in the dominated bank will be described as more influential than organizational opinion leaders in the dominant bank. 2 Research Question 1 Which attributes of the organizational opinion leaders are positively associated with their strength of influence over followers? 2 Research Question 2 Do the organizational opinion leader attributes associated with strength of influence vary according to pre-merger bank of employment? 2 26 CHAPTER 2 METHODS Overview of Research Design This study employed a sequential mixed method research design (Creswell & Plano Clark, 2007) to collect and analyze data relating to change recipients? choices of opinion leaders in the midst of an organizational merger. The use of mixed methods has been recommended by many researchers (Creswell & Plano Clark, 2007, Onwuegbuzie & Johnson, 2006; Tashakkori & Teddlie, 2003) who noted the combination of qualitative and quantitative data provides a better understanding of research problems than either approach alone. The integration of methodologies is especially advocated when studying the sort of complex phenomena (change recipient attitudes and perceptions of influence) presented in this research (Creswell, 2003). The sequential design utilized in the present study consisted of two distinct data collection phases: (a) semi-structured interviews and archival data gathering (qualitative), followed by (b) the administration of a written mixed survey instrument, which contained both close- and open-ended items. The qualitative data collected in Phase 1 provided employee perceptions of organizational culture via semi-structured interviews, as well as the examination of various cultural artifacts (Schein, 2004) such as dress codes, facilities, decorum, and customer service policies. The same types of data were also collected to describe the anticipated post-merged organizational culture. The interview and archival 27 data were then analyzed and used for two distinct comparisons. First, the cultures of the pre-merger banks were compared and contrasted. Then, each pre-merger bank culture was compared in turn, to the post-merger bank culture. The comparison between respective pre-merger cultures and the post-merger bank culture informed the finding regarding the relative dominance of each bank in the merger. A data collection and a series of qualitative and quantitative analyses were then performed in Phase 2 to test the remaining hypotheses and answer the research questions. The collection in Phase 2 was administered via a written survey instrument with close-ended items measuring various self-reported job attitudes and other-reported perceptions, as well as open-ended items soliciting descriptions of a focal other (i.e., their organizational opinion leader). The analyses of data from Phase 2 served to corroborate findings from Phase 1 and are accountable for the remaining conclusions of this study. The rest of the Method chapter of this dissertation will unfold as follows: (a) the organizational setting is described to provide context for this study, (b) Phase 1 (qualitative) data collection, measures, analysis are presented, followed by (c) Phase 2 (quantitative and qualitative) data collection, measures, and analyses. Organizational Context Data for this study were collected in a large commercial bank where employees were experiencing large-scale change in the form of a merger. Prior to the merger, Bank A had over $85 billion in assets and operated in more than 1100 offices with approximately 18,000 employees; whereas Bank B had assets of approximately $50 billion, 650 branches, and 12,000 employees. The sample for this study was confined to a regional segment of the bank employing 1,200 associates across 120 branches (Banks A 28 and B combined). The corporate merger integration plan allowed for 16 months from the announcement to completion. Phase 1 data were collected six months after the merger announcement, followed by Phase 2 data collection six months later. Thus, both data collections took place during the merger period. While most merger-related changes had been initiated, many training, staffing, policy and procedural changes were still being implemented. At the time of my study, various senior managers from both banks had been re-assigned to the other bank (i.e., managers at Bank A branches transferred to Bank B branches and vice versa) to facilitate cultural integration, and training on the backbone operating system software was ongoing. Branch signage changes took place subsequent to data collection. Of note, management had indicated there would be no downsizing associated with the merger in this region, and at the time of this study, no merger-related involuntary job-loss was observed. I was asked to participate as part of an organizational assessment team requested by the senior management of the post-merger bank. Bank leadership desired feedback from employees regarding merger progress. Consequently, I and four colleagues were engaged as no-fee consultants to perform an organizational assessment. Our team provided feedback to bank leadership via a progress brief after Phase 1 and a final brief after completion of Phase 2. Employee comments were presented in aggregate. Any information that might result in identification of participants (e.g., direct quotes referencing work locations, etc.) was removed prior to the presentation of study results to the bank leadership. In summary, the present study developed in two phases. The purpose of Phase 1 was (a) to determine whether one of the pre-merger organizations was a dominant partner 29 in the merger process, and (b) if so, to identify the respective dominant and dominated banks. Phase 2 involved assessing the differences in job attitudes and merger perceptions at the dominated and dominant banks, followed by the exploration and testing of how OOL attributes may vary according to the bank (dominant or dominated) and OOL strength of influence. Phase 1 Data Collection Interviews were conducted to gain a general understanding of participant perceptions of their pre- and post-merger organizations as well as their attitudes and salient issues regarding the merger. Potential interviewees were selected so as to provide a breadth of functional experience within their respective organizations. The purposive sample of branches included representation from all job levels in the organization: manager, assistant manager, platform, and teller positions. Potential participants were given the purpose of the study, reminded that their participation was voluntary, and assured of the confidential nature of their responses. Each prospective interviewee signaled their desire to participate by agreeing to proceed with the interview having read the information letter provided. No potential interviewees declined to participate. Interviews lasted between 45-50 minutes, and notes were made during the interview. The interviews took place in the participant's private office or a private conference room using a standard protocol for each interview. The author and two doctoral students with experience in interview data collection each conducted 16 face-to-face interviews, for a total of 48 interviews. Interviewee responses were recorded via hand-written notes. The interview notes were then transcribed and entered into a database as the data record for the interview. We consulted with one another before and after each interview session, so 30 as to ensure consistency of queries and response notes format between interviewers. Table 2 provides a position profile of interview participants. All interviews were conducted over a one month period across 12 locations (six Bank B branches and six Bank A branches) in the corporate region of interest. Table 2 Demographic Profile of Interview Participants Number of Interviews Job Position Bank B Bank A Branch Manager 5 7 Inside Sales Manager 9 11 Teller 10 6 Total Interviews 24 24 N = 48 Secondary data were sought in an effort to compare external data with the qualitative content analysis results, as the triangulation of data from primary (investigator-originated) and secondary (external) sources can serve to increase validity (Onwuegbuzie & Johnson, 2006). Media, to include open-press articles (e.g., local newspapers, business finance periodicals and websites) and intra-bank briefings were searched for mentions of the focal merger. Articles were sorted and reviewed for content by the primary author. Phase 1 Measures The semi-structured interview script contained a total of 12 questions, eight of which were relevant to the present research (see Table 3). First, participants were asked to describe their job and discuss their day-to-day tasks. The second set of questions was retrospective in an effort to capture recollected perceptions of their organization prior to the merger. The third set of questions was identical to the second, but referenced their 31 present circumstance in the new organization (in transition to post-merger bank). Finally, to draw out contrasts between the past and current workplace environment, two merger- specific questions were asked: ?Regarding the merger, what do you feel good about?? and ?Do you have concerns about the merger?? The interviews concluded by asking if the participants had any additional comments. The general nature of the questions was designed to elicit salient perceptions and attitudes of the interviewees. None of the interview questions explicitly contained the term ?culture.? Thus, the interviewees were not steered during the interviews to comment on culture per se. Consequently, comments referring to culture changes taking place during the interview may be viewed as relatively important to participants, and relatively free from priming effects (cf. Vitale, Armenakis, & Feild, 2008). The inherent weakness of retrospective methods is well documented (e.g., see Smith, 1975). While primary criticism of such data is that they become less accurate with the passage of time, evidence indicates that if the information is collected within two years of the studied event, there does not appear to be an accuracy or bias problem (cf. Gutek, 1978, Herrmann, 1994). Interviews for the present study were completed within six months of the focal merger announcement, which was about a third of the way into the 18 month implementation plan. Moreover, research suggests that the traumatic nature of mergers and acquisitions has the effect of sharpening rather than dulling the memory of those impacted (Cartwright & Cooper, 1993; Sinetar, 1981). Therefore, while a pre- post-comparative research design is ideal, retrospection is understood to be less problematic in merger research, particularly as it concerns culture comparisons as 32 existing values and practices become more salient to the individual when they are threatened or faced with change (Louis, 1990). Table 3 Interview Questions Perceptions of the Pre-merger Organization 1. Prior to the merger, how did you feel about being a [Bank B or Bank A, as appropriate] employee? 2. What felt good about being employed by them, what were you happy with or proud of? 3. What bothered, concerned, or embarrassed you about being a [Bank B or Bank A] employee? Perceptions of the Post-merger Organization 4. Since the merger, how do you feel about being an employee of the new bank? 5. What makes you feel good about being employed by them (what were you happy with or proud of)? 6. What bothers, concerns, or embarrasses you about it? What could be done to make you feel happier or prouder? Perceptions of the Merger 7. When you think about the merger, what do you feel good about? 8. What about the merger concerns you? What issues need to be addressed? Phase 1 Analysis The qualitative data gathered from face-to-face interviews were used in two ways. First, the data were analyzed to compare and contrast the organizational cultures of Bank A and Bank B. By identifying key cultural attributes of both of the pre-merger organizations (cf. Schein, 2004) and comparing them with secondary cultural data gathered from internal organization briefings and open press articles, a profile of key differences and similarities in pre-merger cultures was developed. Finally, the culture of each bank was compared, in turn, with the organizational culture of the post-merger 33 organization. As discussed earlier in this study, the pre-merger partner that most resembles the merged organization is seen as the dominant partner (Van Knippenberg & Van Leeuwen, 2001). This theoretically-supported criterion enabled the subsequent identification of the dominant and dominated organization. The employees? perspectives were content analyzed and theme-coded using semi- inductive analysis (Golden-Biddle & Locke, 2006) whereby theoretically supported themes are identified. I identified four major categories, developed category theme definitions, and examined the categories for consistency in meaning and context. The categories developed in order to draw comparisons between banks and their pre- and post-merger conditions were as follows: (a) Differences between pre-merger partner cultures; (b) Similarities between pre-merger partner cultures; (c) Differences between pre-merger and post-merger cultures, and (d) Similarities between pre-merger and post- merger cultures. Drawing from Schein?s (2004) conceptualization of culture as discussed previously, comments relevant for analysis were identified as those that compared or contrasted day-to-day activities, expectations of performance, interpersonal relationships, expectations of social conduct at work, values, perceived image of the banks, processes, goals, strategies, or philosophies. Comments were coded for sorting if (a) the respondents compared or contrasted either their own recollected pre-merger conditions with their current or expected future, or (b) their own pre-merger conditions with their perceptions of the other bank?s pre-merger conditions. The responses to the interviews were then coded and content analyzed based on the comments contained in the responses. The content analysis in this study identified a comment as the minimum size of a recordable unit. Krippendorf (2004) described 34 recordable units as ?units that are distinguished for separate description?, and coding? (p. 100). Each comment was sorted by each interviewer using spreadsheet software into one of the four categories using guidelines for analyzing qualitative data (Miles & Huberman, 1994). The categorization of comments by theme was conducted separately and blindly by the primary researcher and an outside researcher familiar with content analysis. Coders worked independently of one another, and each coded the 74 total comments provided by the 48 interviewees. Coders agreed on the categorization of 69 of 74 comments resulting in a raw agreement rate of 93%. To account for agreement above chance, the intercoder-reliability was calculated using Cohen?s Kappa and found to be acceptable (0.84; Cohen & Cohen, 1983). Each response was then coded with a ?1? for each category containing a comment and a zero, if otherwise. For example, if the response contained a comment coded into the ?Differences between pre-merger partner cultures? and also a comment ?Similarities between pre-merger and post-merger cultures,? then the response was counted in both of those categories. This hypothesis called for a test to determine the significance of a difference of proportion between two independent samples. Because both measures are dichotomous and the N was small, the Fisher? exact probability test was selected (Siegel, 1956). Phase 2 Data Collection Having collected data via interviews to determine the relative dominance among pre-merger bank partners, quantitative and qualitative data from a written questionnaire were collected and used to test the remaining six hypotheses and answer the two research questions in the present study. While recognizing the banks can now be considered 35 merged in some sense, distinguishing employees by their legacy bank of employment remains central to the hypotheses and research questions in the present study. Therefore, throughout Phase 2, the legacy bank of employment, also previously referred to as ?pre- merger partners? or ?Banks A or B,? in the context of the merger, will be identified as the ?dominant? or ?dominated? bank. A written questionnaire, containing both close- and open-ended items, was administered to tellers at both banks. To reduce confounds due to formal leader hierarchies, only tellers were asked to participate. Branch managers were asked to deliver the surveys (provided in sealed envelopes only identified on front by job type, i.e., teller) to each teller and offer them time at work to complete the survey. All responses were anonymous. Of the 346 employees across all 98 branches invited to complete the survey used in Phase 2, 135 useable surveys were returned for a response rate of 39%. Sixty-eight of the 98 branches (69%) had at least one survey returned. Of the 135 surveys returned, 51% came from individuals who had previously worked for the dominant bank while 49% previously worked for the dominated bank. Average tenure was 4.8 years at the dominant bank and 4.7 years at the dominated bank. The questionnaire was administered with six objectives in mind: (a) corroborate the identification of the dominant and dominated merger partners by statistically testing for the differences between change recipient job attitudes at each bank (Hypothesis 2); (b) find the character attributes that change recipients in each organization use to describe their OOL (Hypotheses 3, 4a, 4b, and 5); (c) determine through the content analysis of responses to open-ended items whether employees of each organization were more likely to select OOLs they described as relationship-oriented, task-oriented, or charismatic 36 (Hypothesis 6); (d) determine whether OOLs at the dominated bank are perceived as more influential (Hypothesis 7); (e) test the OOL attributes that are significantly correlated with the OOL?s strength of influence (Research Question 1); and (f) determine whether those attributes associated with the OOL strength of influence vary between the dominant and dominated banks (Research Question 2). Phase 2 Measures A variety of qualitative and quantitative measures were used to test the hypotheses and answer the research questions. (see Table 4 for a summary list of quantitative measures and sources). Consistent with Krippendorf (2004), the tabulation of comments into similar themes was used to comprehend the qualitative data. Tabulations were used to produce tables of absolute frequencies and relative frequencies, which were then tested using a variety of both parametric and nonparametric tests. Subsequent to the self-report items in the written questionnaire, respondents to the written survey were reminded that their responses would remain anonymous and confidential. Confidentiality in network research is consistent with ethical concerns regarding social network analysis research in organizations and helps to mitigate participant concerns of confidentiality (Borgatti & Moina, 2003). Participants were then informed that the remaining questions focused on the person whom they named (or thought of) as their most influential person at work. Specifically the item read, At work, what person?s opinion do you most value as you form your own attitude towards changes such as the merger, new policies, procedural changes, or new programs? Please provide the first name and last initial of that individual in the text box below 37 and keep this individual in mind as you answer the following questions. Please be reminded that all of your answers will remain CONFIDENTIAL. If you prefer not to write the name of the individual below, please keep him/her in mind as you answer the following questions. The latter portion of the aforementioned question was asked so as to increase the likelihood that respondents referred to a focal peer, not an ideal individual (cf. Cottrell et al., 2007), even if they chose not to explicitly name the individual. The use of follower-ratings of leader performance is consistent with the study of emergent leadership (Bales, 1950; Judge, et al., 2002). The psychological literatures have traditionally defined leadership in terms of ?emergence ?exercising influence in a group of strangers or attaining high status in a social system? (Kaiser, Hogan, & Craig, 2008, p. 97). Measures were used to assess both the change recipient?s self-reported attitudes, his or her perceptions (follower rating) of their OOL, as well as similarity between change recipient and OOL. In some instances, to make inferences regarding similarity, a measure of difference was also calculated. Participants were asked in the latter portion of the questionnaire to choose and report on a referent other (i.e., their OOL) in terms of job performance, job satisfaction, conscientiousness, agreeableness, and extraversion. A difference score was assessed by calculating the difference between the change recipients? self-reported attitude and their perception of the attitude of their respective OOLs. Difference was computed as described by Kenny, Kashy, and Cook (2006, p. 322), using the sum of the squared differences between measures (d 2 ). Identical measures were used where a difference measure was calculated; item verbiage differed only in terms of the referent. For example, the item in the self-report job performance scale, ?I ask for challenging work assignments,? reads, ?He/she asks for challenging work assignments? when referring to the OOL. Quantitative measures are presented, followed by explanations of qualitative measures. Table 4 Summary of Measures Measure Source No. items Self-report Demographics Previous bank, Position, Tenure 3 Job Satisfaction Camman, Fichman, Jenkins, & Klesh (1983) 3 Perceived Organizational Support Lynch, Eisenberger, & Armeli (1999) 8 Job Involvement Kanungo (1982) 9 Intention to Turnover Bentein, Vandenberg, Vandenberghe, & Stingelhammer (2005) 4 Change Recipients Belief Scale Armenakis, Bernerth, Pitts, & Walker (2007) Merger Valence 5 Merger Efficacy 5 Merger Uncertainty Ashford (1988) 4 Other-report OOL Attributes Cottrell, Neuberg, & Li (2007); Goldberg (1999); Interviews (present study) 18 OOL Performance Van Scotter, Motowidlo, & Cross, T.C. (2000) 8 OOL Job Involvement Kanungo (1982) OOL Level cf. Marin & Hampton (2007) 1 Dissimilarity (calculated between respondent and OOL) Big Five Personality (other- rated) Goldberg, L. R. (1999) Conscientiousness Similarity 4 Agreeableness Similarity 4 Extraversion Similarity 4 Job Performance Similarity Van Scotter, Motowidlo, & Cross, T.C. (2000) 8 Job Satisfaction Similarity Camman, Fichman, Jenkins, & Klesh (1983) 3 38 39 Job satisfaction and job satisfaction similarity. The Cammann, Fichman, Jenkins, and Klesh (1983) 3-item measure of global job satisfaction was administered both as a self-report and as an other-directed measure of the respondent?s OOL. Sample items for the self-report were ?In general, I like working here? and ?All in all, I am satisfied with my job here.? Item examples written in other-directed form include ?In general, he/she seems to like working here? and All in all, he/she seems to be satisfied with his/her job.? Items were rated on 6-cell response format ranging from 1 (strongly disagree) to 6 (strongly agree). Previous research (reviewed in Fields, 2002, p. 5) reported acceptable internal consistencies of this measure; coefficient alpha ranged from .67 to .95. Coefficient alphas for the self-report and other-report scales in the present study were .85 and .89, respectively. Job satisfaction similarity was also computed as explained above following the procedure described by Kenny et al. (2006). Perceived organizational support. Employees reported the level of support they received from the organization using the eight-item, short version of the Survey of Perceived Organizational Support (SPOS) originally developed by Eisenberger, Huntington, Hutchison, and Sowa (1986). A confirmatory factor analysis reported by Lynch, Eisenberger, and Armeli (1999) of the eight-item measure used in the present study indicated a unitary factor structure and high internal consistency (Cronbach?s ? = .90). Sample items include ?My organization shows very little concern for me [reverse scored]? and ?My organization values my contributions to its wellbeing.? Respondents indicated the extent of agreement with each statement on a 5-cell response format (1=strongly disagree; 5= strongly agree). Coefficient alpha for this scale in the present study was .93. 40 Merger uncertainty. Self-reported uncertainty resulting from the merger was evaluated using four items adopted from Ashford?s (1988) study of organizational change mechanisms. The items were chosen because the scale was designed to measure the degree to which employees could predict the impact of the change on their own situations. A 5-cell response format was used, ranging from (1) strongly disagree to (5) strongly agree. Sample items for this measure included ?The ongoing changes here at the bank make it difficult to predict how well I can do here? and ?With all the changes here at the bank, it is difficult to know what to do to get ahead.? Coefficient alpha for this scale was .85. Merger valence and merger efficacy. Change recipients? attitudes toward the merger, specifically regarding their belief that the merger would benefit or be detrimental to them (valence) and their confidence that they could successfully perform the new tasks required by the merger (efficacy) were measured using items adopted from the Organizational Change Recipients? Beliefs Scale (OCRBS, Armenakis et al., 2007). A 5- cell response format was used for both scales, ranging from (1) strongly disagree to (5) strongly agree. Merger valence was measured using five items including ?The merger will benefit me? and ?I feel the merger will offer me a lot of opportunities for advancement.? Coefficient alpha for the merger valence scale was .85. The five items measuring Merger efficacy included ?I can implement this merger in my job? and ?I believe we can successfully implement this merger.? Coefficient alpha for the merger efficacy scale was .93. Intention-to-turnover. The intention to voluntarily leave the organization was measured with four items, including, ?I often think about quitting this organization,? and 41 ?I have thought of resigning in the last 9 months.? These items were adapted from research by Bentein, Vandenberg, Vandenberghe, and Stingelhammer (2005), where they reported coefficient alphas of .83, .87, and .88 across respective Time 1, Time 2, and Time 3 data collections. Items were rated on 5-cell response format ranging from 1 (Strongly disagree) to 6 (Strongly agree). Coefficient alpha for this scale was .92. Organizational opinion leadership style. This open-ended item was an introspective question, designed to evoke thoughtful consideration of leader attributes and constituted the variable of interest for the Hypothesis 6. The item read, Briefly describe in your own words why you chose the person you identified above. What is it about them that makes you willing to listen to and be influenced by them? (write your answer here and continue on the last page if needed). Responses were sorted according to a framework adopted from the current leadership literature (cf. Ehrhart & Klein, 2001). Based on the content analysis of each description, the referent OOL was placed into one of three categories: (a) task-oriented, (b) relationship-oriented, and (c) charismatic. Each response was considered and coded as a whole by independent raters. The category definitions used, were as described previously. Task-oriented leaders? descriptions were concerned primarily with scheduling work, setting goals, and providing supplies, while relationship-oriented leader were described as providing social support to their followers and interpersonal communications. Finally, charismatic leaders tend to focus on motivation. OOL job involvement and job involvement similarity. Kanungo?s (1982) 10-item measure of job involvement was administered as an other-report directed toward the nominated OOL. Example items included ?Most of his/her personal goals are job- 42 oriented,? ?He/she lives, eats, and breathes his/her job,? and ?Most of his/her interests are centered around his/her job. Items were rated on 6-cell response format from 1 (strongly disagree) to 6 (strongly agree). Coefficient alpha for the scale was .85. Job involvement similarity was also calculated as described by Kenny, et al. (2006). OOL job performance and job performance similarity. Van Scotter, Motowidlo, and Cross?s (2000) 8?item job performance measure was administered as an other-report. Items were rated on a 6-cell response format from 1 (Very unlikely) to 6 (Very likely). Example items for job dedication included ?He/she pays close attention to important details? and ?He/she offers to help others at work.? The coefficient alpha for the scale was .89. Job performance similarity was also calculated according to Kenny et al.?s (2006) procedure. OOL top three attributes. Change recipient descriptions of nominated OOL attributes were assessed using 18 single-item measures developed in accordance with a survey of leadership literatures (cf. Anderson, 1968; Cottrell et al., 2007; Goldberg, 1999; Judge, Bono, Ilies, & Gerhardt, 2002; Rokeach, 1973; Washington, Sutton, & Feild, 2006) and the content analysis results of qualitative data collected in Phase 1. The 18 items included adjectives (for complete list, see Table 5) that are related to factors of personality, e.g., extraversion and conscientiousness, as well as traits such as ?feelings are similar,? ?trustworthy,? and ?supportive? cited by interview participants when asked to describe what it is about the OOLs they described that makes them influential. Participants were asked to ?Please pick three traits from the table that are the most important reasons you value the opinion of the person you have named.? 43 Table 5 Survey Item Choices for Top Three Organizational Opinion Leader Attributes caring calm job expert in-the-know out-going supportive he/she makes things happen we have similar opinions trustworthy Intelligent happy honest we have similar backgrounds we have similar personalities we have similar feelings we have similar job experience same gender we are similar in age Opinion leader influence. Influence was measured using four items adapted from Flynn et al. (1996) and two items developed specifically for the proposed research. Sample items from Flynn et al. were ?Compared with others at work, how likely are you to ask your most-influential peer about new changes (e.g., new policies, procedures, programs)?? and ?Overall, regarding organizational changes how likely are you to use the most-influential peer?s advice?? Items were rated on 5-cell response format ranging from (1) very unlikely to (5) very likely. In research refining the King and Summers? (1970) opinion leadership scale, Flynn et al. (1996) reported coefficient alphas of no less than .84 in four studies. Coefficient alpha for this scale was .85. OOL relative position level. The position of the OOL as compared to the change recipient was measured using the single item, ?Compared to you, what level of position does this person occupy in Merged Bank?? Three choices were provided: (1) Lower than my position, (2) Same, and (3) Higher than my position. Phase 2 Data Analyses Hypothesis 2. To corroborate the findings of Phase 1 and test the hypothesis that change recipients employed by the dominated merger partner are more affected by change and consequently tend to have more adverse job attitudes and merger perceptions, 44 a one-way multivariate analysis of variance (MANOVA) to determine the effects of pre- merger bank of employment on the six dependent variables of this study was conducted. Subsequent univariate and discriminant analyses were also performed. Hypotheses 3. Participants were asked to ?Pick three traits from the table above [18 item attribute list] that are the most important reasons you value the opinion of the person you have named.? The frequency of each attribute (total number of times each attribute was listed in any of the three blanks) was calculated. To test whether employees will tend to describe organizational opinion leaders in terms of expressive or instrumental ties as proposed by Umphress et al (2003), category definitions were developed consistent with the typology of social ties, wherein expressive ties are characterized by interpersonal affective communications and similarity ?with-self; and instrumental ties are described as work-related and knowledge-oriented. The attributes were sorted independently by the primary author, and a fellow graduate student familiar with the content analysis of quantitative data, into each category (i.e., expressive, interpersonal, similarity (EIS) and instrumental, work-related, knowledge (IWK) traits). Initially we agreed on the classification 17 of 18 terms and after discussion, agreement was reached on the final item (see Table 6, for the final attribute categorization). Similarity traits were aggregated into a single category. This consolidation is theoretically consistent with the similarity-attraction and homophily literatures that cite a wide variety of similarity types (e.g., attitudinal, demographic, physical) as affecting social attraction. 45 Table 6 Organizational Opinion Leader Traits Typology Expressive, Interpersonal, Similarity (EIS) Instrumental, Work-related, Knowledge (IWK) trustworthy job expert honest in-the-know supportive calm caring makes things happen out-going Intelligent happy Similarity-with-self The frequencies of responses containing OOL traits in each category were tallied, summed, and compared across both banks. A Differences of proportions test (Bruning & Kintz, 1968) was then conducted between categories to determine if change recipients were more likely to describe OOLs in expressive, interpersonal, or similarity terms rather than traits described as instrumental or work related. Hypotheses 4a & 4b. The next step was to investigate the specific traits that change-recipients described as the most important regarding their referent OOL and determine whether the traits varied according to previous bank of employment. To examine differences in these specific OOL traits between banks, the frequency of each attribute was summed and a Differences of proportions test (Bruning & Kintz, 1968) was performed between banks on each attribute. Hypothesis 5. An analysis to determine the frequency with which individuals described their selected OOL as holding a position of higher-level than themselves at work was conducted. The tabulations in each category (i.e. lower level than me, same level as me, higher level than me) were then subject to a Difference of proportions test (Bruning & Kintz, 1968) to test for significant differences in the OOL ?levels? reported. 46 Hypothesis 6. OOL leadership style was assessed via content analysis of the open-ended item that asked participants to describe ?what is it about your OOL that makes you willing to listen and be influenced by them?? Three categories of leader-style (task-oriented, relationship-oriented, and charismatic) were defined according to the typology described by Ehrhart and Klein (2001). These categories were used to sort the descriptions of the referent OOLs. Categorizing each OOL was accomplished using a three-step coding process: each comment was coded and sorted in the appropriate category, and then each response was coded according to whether the comments described a task-oriented, relationship-oriented, or charismatic leader. Table 7 provides some examples of the responses analyzed and the subsequent categorization of the comments. Finally, the classification of each opinion leader (N = 87) was determined by comparing the number of comments in each category. Table 7 Example Comments by Change Recipients Used to Describe OOLs Leader Style Task-oriented ? Very knowledgeable about customer service and how to handle situations with good attitude. ? The person I chose is a hard worker. She is the first at work and the last to leave work. She practices what she preaches. ? Professional, positive attitude. Relationship- oriented ? This person understands that we share the same feelings about the company. ? She listens, is kind and understanding. ? I listen to her because she has time to listen to my concerns. Charismatic ? She is intelligent, supportive, wise, gives good advice, has high goals for employees and the bank. Responses were unitized into comments by the primary author. The categorization of comments was subsequently conducted separately and blindly by the 47 primary researcher and an outside researcher familiar with content analysis. Coders worked independently of one another, and each coded the 154 comments contained in 87 responses. Coders agreed on the categorization of 130 of 154 comments, resulting in an agreement rate of 84%. To account for agreement above chance, the intercoder- reliability was calculated using Cohen?s Kappa (Cohen & Cohen, 1983) and found to be acceptable (0.83). Each response was then coded with a leader-style according to the coded comments. Coders agreed on 79 of 87 OOL leader-style categorizations (90% agreement). Cohen?s Kappa was acceptable (.82). A Fisher?s exact probability test (Fisher, 1970) was used to determine if there was a systematic association between previous bank (i.e., dominant or dominated) and OOL leader-style. Hypothesis 7 and research questions. Strength of OOL influence was investigated to determine if OOL strength varies, on average, between banks (Hypothesis 7) and to examine its possible predictors (Research Questions 1 & 2). OOL strength of influence was investigated using Hierarchical Linear Modeling (HLM). An advantage of HLM is that it allows the examination of relationships at different levels while maintaining the appropriate level of analysis (Hofmann & Gavin, 1998). HLM is particularly appropriate to test cross-level relations when individuals are nested within groups (Raudenbush, Byrk, Cheong, & Congdon, 2004), as was the case in this study. (The change recipients were nested in branches, which were in turn, nested by previous bank of employment.) HLM recognizes that individuals within one group may be more similar to one another than to individuals in another group (Raudenbush et al., 2004). This provides the rationale for inclusion of the following level 2 variables: size, job satisfaction, perceived organizational support, the merger perception variables (uncertainty, valence, and 48 efficacy), and turnover intention. As Gavin and Hofmann (2002, p.16) described, ?HLM is used to assess: (1) the influence of predictors at both the individual and group level on an individual-level outcome, and (2) the moderating effects of group-level variables on relationships between individual-level variables.? In the present study the main effects of various OOL attributes (level 1), branch-level employee attitudes (level 2), and previous bank employment (level 3) on OOL influence were tested in order to answer Research Question 1. The moderating effects of branch and previous bank on the relationship between OOL attributes and OOL influence were tested to answer Research Question 2. HLM is a multi-step process that first examines relationships among variables at the individual level. Then, the level 1 intercept and slope parameters for each variable are regressed onto group level variables in a level 2 equation. These values are then regressed onto level 3. To investigate research question 2, the general model for the 3 level data was specified as follows: Level 1: Y ijk (OOL influence) = ? 0jk (OOL Performance) + ? 1jk (OOL Job Involvement) + ? 2jk (OOL Job Satisfaction) + ? 3jk (OOL Level )+ ? 4jk (DiffAgreeableness) + ? 5jk (Diff Extroversion)+ ? 6jk (DiffConscien) + ? 7jk +(DiffPerformance) + e Level 2: ? 0jk (Branch) = ? 00k + ? 01k (SIZE) + ? 02k (JobSat) + ? 03k (POS) + ? 04k (MergerUncert) + ? 05k (MergerValence) + ? 06k (MergerEfficacy) + ? 07k (TurnoverIntent)+ r 0jk Level 3: ? 00k (Previous bank) = ? 000 + u 000 49 The predictor variables in the level 1 equation were centered to reduce the multicollinearity between lower-order and higher-order terms (cf. Hofmann & Gavin, 1998; Yee Ng & Van Dyne, 2005). Consistent with Gavin and Hofmann (2002), the level-2 variables (branch-level) were individual scores aggregated at the branch level. Final-step beta values are reported, and because the research questions are not directional, two-tailed tests were used to interpret the results. 50 CHAPTER 3 RESULTS Phase 1 Hypothesis Hypothesis 1 predicted that employee attitudes toward the merger would enable the identification of the dominant and dominated merger partners (Bank A and Bank B). To test this hypothesis, the results of the content analysis of the employee face-to-face interviews were compared between employees grouped by pre-merger employment. Interview responses indicated that Bank A and Bank B each possessed strong cultures (Schein, 2004) that differed in some respects. For example, when describing their own bank in retrospect, ?what felt good about being employed by your former bank, what are you most proud of?? employees from Bank A tended to describe their relationships with customers and co-workers with comments like, ?I enjoyed coming to work everyday, meeting my customers; good co-workers makes your job more exciting,? whereas Bank B employees tended to explicitly refer to the Bank B culture with comments like, ?Bank B is a team environment; my home away from home? and ?I like the culture, it is sales-driven.? The culture strength was indicated by the overwhelmingly positive feelings conveyed by employees toward their legacy banks. Strong pre-merger cultures were also evident in the responses of individuals when they compared their legacy bank with the current post-merger bank. For example, responses to the open-ended item ?since the merger, how do you feel about being an 51 employee of the post-merger bank?? included, ?This is a case of a white collar bank merging a with blue collar bank; Bank B had a conservative dress code ? suits, whereas Bank A had khakis and t-shirts.? Many responses also included an us-versus-them perspective as reflected in the following statement, ?We are changing the name, but keeping the corporate culture. We heard that Bank A employees were scared and afraid of the new goals, because the goals are sales oriented and too aggressive. We are focused on new account generation and require more professional dress.? Table 8 illustrates the number of employees reporting their perceived differences and similarities between the two bank cultures prior to the merger. Results suggested more individuals, regardless of their pre-merger bank of employment, reported cultural differences, rather than similarities, between the two pre-merger banks. This finding was not unexpected, given that interviewees were not explicitly prompted to compare the two pre-merger banks, thus differences would be more salient and more likely to be mentioned (Schwarz & Sudman, 1996). Nonetheless, the relatively large number of interviewees reporting differences supports the notion that differences between the two pre-merger cultures existed. A Fisher?s exact probability test (Fisher, 1970) indicated no significant relationship (p > .05) between Bank and comparison type (i.e., differences and similarities). Thus, the proportions of interviewees reporting differences and similarities between pre-merger cultures were consistent across banks. However, determining the dominant and dominated bank (Hypothesis 1), required additional analyses of employees? attitudes and perceptions of the pre- and post merger conditions at their respective banks. 52 Table 8 Responses Comparing Cultures Between Pre-merger Banks Bank Differences Similarities Bank A a 16 1 Bank B b 13 1 a n = 24 b n = 24 The next step was to compare perceptions among bank employees of the congruence between their respective pre-merger bank cultures and their expectations of, or current experiences regarding, the post-merger culture as conveyed during the interviews. As Table 9 shows, 22 of the 24 Bank A employees mentioned differences they perceived between their pre-merger culture and their post-merger culture. This finding is contrasted by the observations of the employees at Bank B, where only 1 of 24 Bank B employees reported differences between their pre-merger and post-merger cultures. Also, 19 of 24 Bank B employees cited similarities between their culture and the post-merger culture, while only 1 of the Bank A employees described similarities. A Fisher?s test (Fisher, 1970) indicated a significant difference (p < .001) between comparison type (differences and similarities) across banks. Thus, there are a relatively high percentage of perceived culture differences between the pre-merger and the post- merger cultures by Bank A employees, and a relative lack of differences reported by Bank B employees. 53 Table 9 Responses Comparing Cultures Between Pre-merger Banks and the Post-merger Bank Bank Differences Similarities Bank A a 22 1 Bank B b 1 19 a n = 24 b n = 24 Interview quotes highlighting the differences between pre-merger cultures included comments such as ?We here at Bank A have more of a friendly laid back environment than Bank B,? ?This bank is totally different from how it used to be. Now it is more like telemarketing,? and ?The Bank A folks need to step it up, historically, Bank B managers have worked harder.? Typical interview comments indicative of dominant or dominated status included, ?We are Bank B with the Bank A name. All the changes are in the direction of Bank B. The aggressiveness is affecting customers? interests,? ?Because our name is not changing, the community feels like the new Bank A is going to be like the old Bank A. They do not understand that we will be a heart and core of Bank B with an overlay of Bank A,? and ?We didn?t meet in the middle, the change is biased towards Bank B?s ways of doing things.? Secondary data findings, though limited, supported the interview content analysis results. Cultural artifact data reflective of the work group climate, included dress codes and facility condition. A review of pre-merger dress codes found the policy for Bank B was suit and tie Monday thru Thursday, business casual on Fridays. Women were expected to wear business appropriate dresses, suits, skirts or slacks Monday thru Thursday, and business casual on Fridays. This contrasted with Bank A?s policy of open- collar and khakis (Vault, 2007). The primary author and fellow investigators also noted during their visits to 12 branches that Bank B?s facilities were in markedly better condition (furniture, fixtures, wall, and floor coverings). Taken together, the analysis of interviews and secondary data indicated that (a) there is evidence that employee perceptions of the pre-merger bank cultures differed noticeably according to their legacy bank of employment; and (b) individuals previously employed by Bank A reported the greatest degree of culture change as the post-merger bank was forming. Considering that the dominated organization is described as the pre- merger partner that experiences the largest degree of change (Van Knippenberg & Van Leeuwen, 2001) the results revealed Bank A and Bank B as the dominated and dominant bank, respectively (see Figure 1). Thus, Hypothesis 1 was supported. Figure 1. Summary Phase 1 findings identifying the dominant and dominated bank 54 55 Phase 2 Hypotheses Hypothesis 2 predicted that employees of the dominated bank will report lower job satisfaction, less perceived organizational support, greater merger uncertainty, lower merger valence, lower merger efficacy, and higher intention to turnover than employees of the dominant merger partner. Table 10 lists the partial correlations between the antecedent variable (pre-merger employment), and outcome variables while controlling for tenure. Tenure (in years) was entered as a control variable in the correlation as it might be responsible for observed covariation of job attitudes and merger perceptions (Lee & Wilbur, 1985). The significant partial correlations between bank and each of the attitude variables provided initial support for the hypothesis (r?s between .22 and .49, p < .05). Table 10 Reliabilities, Descriptive Statistics, and Partial Correlations of Job Attitudes and Merger Perceptions Mean S.D. 2 3 4 5 6 7 8 Control variable 1. Years Tenure 5.14 8.17 Antecedent 2. Bank (A or B) dummy variable .49 .50 Outcome variables 3. Job Satisfaction 3.73 .98 -.35** (.85) 4. Perceived Organizational Support 3.35 .83 -.25** .72** (.93) 5. Merger Uncertainty 2.77 .93 .40** -.67** -.66** (.85) 6. Merger Valence 2.80 .91 -.27** .68** .71** -.65** (.85) 7. Merger Efficacy 3.79 .79 -.53** .65** .62** -.71** .68** (.93) 8. Turnover Intention 2.53 1.18 .42** -.81** -.74** .62** -.74** -.67** (.92) Note. coefficient alpha reliabilities are shown on the diagonal, N = 111 (listwise deletion) * p < 0.05 level (2-tailed). ** p < 0.01 level (2-tailed). 56 57 However, noting the significant correlations between dependent variables a one- way multivariate analysis of variance (MANOVA) was deemed appropriate to further test the hypothesis since it allows for a simultaneous test of differences among multiple dependent variables and consequently has more power to detect differences than a univariate test (Field, 2005). Thus, a MANOVA of differences between bank groups was used to determine the effects of pre-merger bank of employment on the six dependent variables of this study. The demographic variable tenure was treated as a control variable in the analysis. The overall MANOVA results expressed in Table 11 indicated significant differences (Wilks Lambda = .28, F(8, 111) = 36.11, p < .001) between banks. Table 11 Overall MANOVA for Pre-merger Bank of Employment a Effect Value F b Sig. Intercept Pillai's Trace .997 4103.109 .000** Wilks' Lambda .003 4103.109 .000** Hotelling's Trace 298.408 4103.109 .000** Roy's Largest Root 298.408 4103.109 .000** Bank Pillai's Trace .450 11.263 .000** Wilks' Lambda .550 11.263 .000** Hotelling's Trace .819 11.263 .000** Roy's Largest Root .819 11.263 .000** * p < 0.05 level (2-tailed). ** p < 0.01 level (2-tailed). a Dependent variables are job satisfaction, perceived organizational support, merger uncertainty, merger valence, merger efficacy, and intention to turnover b Exact statistic Additionally, a univariate analysis (see Table 12) indicated mean difference between banks was significant (p < .05) for of each of the six dependent variables. 58 Table 12 Univariate Analyses for Pre-merger Bank of Employment Source Dependent Variable Type III Sum of Squares df Mean Square F Sig. Corrected Model Years Tenure 36.273 a 1 36.273 .517 .473 Job Satisfaction 11.675 b 1 11.675 14.181 .001*** Perceived Operational Support 3.477 c 1 3.477 5.339 .023* Merger Uncertainty 15.368 d 1 15.368 21.066 .001*** Merger Valence 4.224 e 1 4.224 5.135 .025* Merger Efficacy 18.301 f 1 18.301 39.009 .001*** Turnover Intention 28.810 g 1 28.810 25.271 .001*** * p < 0.05 level (2-tailed). ** p < 0.01 level (2-tailed). *** p < 0.001 level (2-tailed). a R Squared = .004 (Adjusted R Squared = -.004) b R Squared = .108 (Adjusted R Squared = .100) c R Squared = .044 (Adjusted R Squared = .035) d R Squared = .153 (Adjusted R Squared = .145) e R Squared = .042 (Adjusted R Squared = .034) f R Squared = .250 (Adjusted R Squared = .244) g R Squared = .178 (Adjusted R Squared = .171) A discriminant analysis is recognized as an essential follow-on to MANOVA (Field, 2005). A discriminant analysis was recognized as especially useful for this test because (a) the dependent variable (bank) is categorical, and (b) it takes into account the interrelationships of the independent variables (Cohen & Cohen, 1983). The discriminant analysis result was significant between banks (Wilks Lambda = .642, N = 119, df = 7, ? 2 = 50.31, p < .001). The results of the MANOVA, univariate, and discriminant analyses indicated that differences in the employee attitudes and perceptions could be explained by bank employment. Further support is shown in Table 13, where the values in the diagonal of the classification table reflect the correct classification of individuals into groups based on their scores on the dimensions of job satisfaction, perceived organizational support, merger uncertainty, merger valence, merger efficacy, and intention to turnover. As an example, the dimension scores correctly predicted 59 employment by the dominated bank 84.7% of the time, whereas employment by the dominant bank was predicted accurately 88.3% of the time. Table 13 Classification Results for Pre-merger Bank of Employment Previous Bank Predicted Group Membership Dominant Dominated Total Original Count Dominant 53 7 60 Dominated 9 50 59 % Dominant 88.3 11.7 100.0 Dominated 15.3 84.7 100.0 Note: 86.6% of original grouped cases correctly classified Therefore, Hypothesis 2, which stated that employees of the dominated bank would report lower job satisfaction, less perceived organizational support, less merger valence, less merger efficacy, greater merger uncertainty, and higher intention to turnover than employees of the dominant bank was wholly supported. The support for Hypothesis 1 in Phase 1, and Hypothesis 2 in Phase 2, that found significant perceptual and attitudinal differences between change recipients based on pre-merger bank of employment provided a basis for the remaining hypotheses tests in which change recipients? attributions of OOLs are compared across the dominant and dominated banks. Hypothesis 3 posits that overall, employees at both banks will more often describe their OOLs? most important attributes with traits characterized by positive interpersonal affect and similarity, rather than work related attributes. The number of respondents listing each trait was summed and the number of responses containing traits defined as expressive/interpersonal/similar (EIS) or instrumental/work-related/knowledge (IWK) were then sorted and tabulated. The totals for each category are shown in the ?combined column? of Table 14. A difference of proportions test (Bruning & Kintz, 1968) indicated 60 that overall, change recipients (z = 10.87, p < .001) more often described their OOLs with EIS terms. Thus, Hypothesis 3 is supported in that employees from both banks tended to describe their OOL?s most important attributes with EIS traits rather than IW traits. Table 14 OOL Trait Category Frequencies Across Banks Trait Type Dominant Bank Dominated Bank Combined n = 57 n = 58 N = 115 Expressive, Interpersonal, Similar EIS) 101 114 216 Instrumental, work-related (IWK) 58 36 94 Hypothesis 4a states employees at the dominated bank will be more likely than those at the dominant bank to use specific attributes characterized as EIS traits to describe their selected OOLs, whereas Hypothesis 4b expects that employees at the dominant bank will be more likely to use specific traits characterized as IWK to describe their selected OOLs than their counteparts at the dominant bank. Testing these hypotheses was conducted in two steps. First, a Fisher?s test (Fisher, 1970) was conducted across trait categories and bank type. Results indicated a significant difference (p = .019) in proportion of trait types between banks. That is, as Table 14 shows, employees at the dominant bank are approximately twice as likely (i.e., 102/58) to describe their OOLs? most important attributes with EIS rather than IW traits, whereas employees from the dominated bank are more than three times as likely(i.e., 114/36) to describe their OOLs with EIS traits rather than IWK traits To provide additional insight as to the differences between banks regarding each descriptive trait, the number of respondents listing each trait was summed, and percentages of responses containing each trait were calculated. Figure 2 illustrates that overall; change recipients most often described their respective OOLs as trustworthy (53%), honest (43%), supportive (39%), and caring (32%). Furthermore, change recipients also referred to their OOLs as a job expert (23%), intelligent (22%), in-the- know (12%), makes-things-happen (10%), calm (9%), happy (8%), outgoing (6%), knowledgeable (6%), and similar to me (6%). Figure 2. Most important OOL attributes (N = 115) 0.00% 10.00% 20.00% 30.00% 40.00% 50.00% 60.00% Tr u s t wor t hy Ho ne s t S uppo rt i ve C a ri ng J ob E x per t In te l l i ge nt I n -t he- kno w Ma k e s - t h i ng s - ha pp en Ca l m Ha ppy O u t g oi ng K now le dgea bl e Sim ila r t o me OOL Attributes (follower-rated) % Responses To test for differences between banks in the percentage of responses regarding each OOL attribute, the percentages for each attribute were calculated in each of the dominant (n=57) and dominated (n=58) bank groups. A two-tailed test for Significance between proportions (Bruning & Kuntz, 1968) was used to calculate a z score for each attribute. The percentages of respondents listing the attributes trustworthy (z = -2.15, p < .05), supportive (z = 2.35, p < .05), job expert (z = 4.76, p < .01), makes-things-happen (z = -2.77, p < .01), and calm (z = 3.05, p < .01) differed significantly between banks (see Figure 3). The results indicated that change recipients previously employed by the 61 dominated bank more often described their respective OOLs as trustworthy, supportive, and as individuals that make things happen. On the other hand, those previously employed by the dominant bank were more likely than their counterparts at the dominated bank to describe their OOLs as job experts and calm. Of these attributes, trustworthy and supportive were EIS traits, whereas, makes things happen, job expert, and calm were IWK traits. In summary, employees of the dominated bank, significantly more so than the dominant bank, described their OOLs with two EIS traits (i.e., trustworthy and supportive) and one IWK trait (makes-things-happen). Additionally, employees of the dominant bank more often described their OOLs with two IWK traits (i.e., job expert and calm) than did their counterparts at the dominated bank. Thus, hypothesis 4a and 4b are supported. Figure 3. Most important OOL attributes by pre-merger bank (N = 115) 0% 10% 20% 30% 40% 50% 60% 70% Tr u s tw o r t h y* H o nes t S upp or t iv e * Ca r i n g J ob E x per t * * In t e l l ig e n t I n-t h e -k n ow Ma k e s- t hi ng s - ha pp en* * Ca l m * * Ha p p y O u t go ing Kn o wl ed g ea bl e Si mi l a r t o m e OOL Attributes (follower-rated) % R es pons es Dominant Bank (n=57) Dominated Bank (n=58) * p < 0.05 level (2-tailed). ** p < 0.01 level (2-tailed). 62 63 Hypothesis 5 holds that employees will describe their OOLs as holding a higher position at work than themselves. A frequency analysis of the responses (N = 115) revealed that 90.4% of the participants (n = 104) indicated their identified OOL as holding a higher position. By comparison, of the remaining choices, ?at the same level? was chosen by 9.6% of the respondents (n = 11), and ?at a lower level than me? was not selected. A Differences of proportions test statistically verified that a significantly (z = 34.01, p < .001) higher number of OOLS were described as ?above my level at work? by their respective nominators. Thus, Hypothesis 5 is supported. That is, on average, employees were more likely to describe their OOLs as at a higher-level position than themselves. Interestingly, analysis of the open-ended item in the questionnaire revealed 10% of the respondents described their OOLs as their ?boss,? ?manager,? or ?supervisor.? This finding is addressed further in the discussion section of this study. Hypothesis 6 predicted that employees of the dominated bank would be more likely to select OOLs they describe as relationship oriented, whereas employees of the dominant bank would be more likely to select OOLs they describe as task-oriented. To test this hypothesis, the open-ended responses of participants describing OOLs were compared across the dominant and dominated employee groups. The findings illustrated in Table 15 indicate that of the 87 OOLs chosen, 40 were described as task-oriented (20 at each bank), 40 were relationship-oriented (21 at the dominant bank and 19 at the dominated bank) and 7 were charismatic (3 at the dominant bank and 4 at the dominated bank). 64 Table 15 Styles of OOLs Chosen by Change Recipients at the Dominant and Dominated Banks Qualitative OOL Style Pre-merger Bank Task-oriented Relationship-oriented Charismatic Dominant 20 21 3 Dominated 20 19 4 Total 40 40 7 Note. N=87 A Fisher?s test (Fisher, 1970) found no significant association (p = .49) between bank type and OOL leader-style. In fact, as Table 15 shows, frequencies were nearly identical between banks and leaders-styles. This finding suggests that the OOLs sought by tellers are equally likely to be task- or relationship-oriented, regardless of bank. Thus, Hypothesis 6 was not supported, in that change recipient choice of OOL style did not vary according to pre-merger employment. In fact, participants were equally likely to describe their OOL as task- or relationship-oriented, regardless of whether they were employed by the dominant or dominated Bank. Note that the low number of charismatic OOLs described is not surprising; given the extensive verbiage necessary to convey the unique connection between a follower and charismatic leader (cf. Ehrhart & Klein, 2001). For example to be considered a charismatic leader, respondents would have needed to report with comments describing the OOL?s vision, goals, or concern for performance and a sense of connectedness with the leader. Hypothesis 7 posited that change recipients at the dominated bank would report their OOLs as more influential than would the change recipients at the dominant bank. To assess the effect of bank on OOL influence, an unconditional HLM (null model) was tested to estimate the total systematic variance in the outcome variable. This model is analogous to a one-way analysis of variance where the variable of interest is the 65 dependent variable (OOL influence) and bank is used as the independent variable. HLM results for the null model were non-significant for between group variance at the bank- level (? = .001, df = 1, ? 2 = .47, p > .50). Thus, there is not enough evidence to suggest that OOL strength of influence varied according to bank type, and as a result, Hypothesis 7 was not supported. In other words, OOL influence did not vary according to bank of employment. The null group?level 2 test indicated that the branch (level 2) effect on the OOL strength of influence was also non significant (? = .001, df = 67, ? 2 = 70.80, p = .35). Thus, branch size, nor any other branch level variable, accounted for sufficient variance to be deemed significant. Phase 2 Research Questions Research Question 1 posed the following question: ?Which, if any attributes of the OOLs are positively associated with the OOLs? strength of influence over followers?? Research Question 2 sought to explore whether OOL attributes associated with strength of influence varied according to previous bank of employment. Correlations were computed for all variables within their respective level of analysis to determine zero- order relationships within each level (e.g., individual or group level). Table 16 presents descriptive statistics and correlations of individual-level variables. Table 17 presents this information at the group level of analysis. Table 16 Descriptive Statistics and Correlations for Model 1 (Individual-Level Variables, N=114) Variable M SD 1 2 3 4 5 6 7 8 Outcome variable 1. OOL Influence 5.21 .72 OOL Traits 2. OOL Performance 5.55 .55 .31** 3. OOL Job Involvement 3.33 .67 .13 .19* 4. OOL Job Satisfaction 3.78 .84 .13 .09 .49** 5. OOL Level 2.90 .30 .05 .24* .30** .04 Difference with OOL a 6. Agreeableness .43 .70 -.04 -.17 .07 .15 -.22* 7. Extraversion 1.13 1.71 .19* -.06 .04 .19* -.12 .15 8. Conscientious .49 .93 -.06 -.02 -.06 .07 .01 .33** -.01 9. Performance .56 1.10 -.05 .01 .09 .09 -.02 .12 -.01 .32** * p < 0.05 level (2-tailed). ** p < 0.01 level (2-tailed). a Calculated as d 2 between respondents? self-ratings and other (OOL) directed ratings 66 67 Table 17 Descriptive Statistics and Correlations for Model 2 (Branch-Level Variables) Variable M SD 1 2 3 4 5 6 1. Size 14.11 16.80 2. Job satisfaction 3.78 .78 .02 3. POS 3.34 .64 .06 .74** 4. Merger uncertainty 2.73 .81 -.05 -.69** -.65** 5. Merger valence 2.81 .84 .00 .66** .73** -.64** 6. Merger efficacy 3.83 .62 .07 .66** .65** -.72** .69** 7. Turnover intent 2.57 1.07 -.03 -.81** -.73** .58** -.80** -.67** * p < 0.05 level (2-tailed). ** p < 0.01 level (2-tailed). HLM analysis indicated the traits of OOL job performance (? = .64, p = .001) and job performance difference (between OOL and change recipient) (? = .29, p = .020) were significant predictors of OOL influence (see Table 18). The level 1 model was significant (? = .078, df = 114, ? 2 = 104.06, p = .003). In answer to Research Question 1: on average, the job performance of an OOL, as perceived by their followers, was positively associated with the reported OOL strength of influence. Thus, OOL job performance and job performance similarity were significant predictors of OOL influence in both banks. Also, the similarity between change recipients? self-reports of job performance and their other-reports of their respective OOL job performance were positively related to the OOLs? strength of influence. In answer to Research Question 2, to determine whether differences in OOL influence existed between banks or branches, the results for the same HLM model were analyzed. Branch and bank were modeled as level 2 and level 3 variables respectively. As Table 18 indicates, neither branch nor bank (p = .82, p > .50, respectively) were significant predictors of OOL influence. 68 Table 18 Summary of HLM Variables Predicting OOL Strength of Influence Variable Coefficient SE p Level 1 a (Individual) OOL Performance 0.64 0.16 .001** OOL Job Involvement 0.09 0.17 .580 OOL Job Satisfaction 0.06 0.15 .700 OOL Level 0.15 0.33 .650 Agreeableness Difference 0.07 0.14 .600 Extraversion Difference 0.10 0.06 .090 Conscientious Difference 0.08 0.14 .540 Performance Difference -0.29 0.12 .020* Level 2 b (Branch) Size 0.00 0.00 .820 Level 3 c (Bank) Intercept 0.00 0.00 >.500 * p < 0.05 level (2-tailed). ** p < 0.01 level (2-tailed). a N = 114, b n = 66, c n = 2 Overall, support for hypotheses in the present study, was mixed. A summary of hypotheses and support is depicted in Table 19. On the one hand, OOLs were described differently, in different organizations that were marked by differing magnitudes of change. Individuals employed by the organization that experienced the greatest degree of change tended to describe their OOLs with affective terms, such as trustworthy and supporting; more so than their counterparts at the less change-affected bank. On the other hand, counter to expectations, the OOLs? strength of influence did not vary according to pre-merger organization. OOL job performance and similarity of job performance between OOL and change recipient predicted OOL strength of influence. . 69 Table 19 Summary of Hypotheses and Research Questions Hypothesis 1 Employee perceptions of the magnitude of merger-induced changes will vary according to their pre-merger bank of employment, such that the pre-merger partner banks can be described as dominant and dominated. Supported Hypothesis 2 Employees of the dominated bank will report lower job satisfaction, lower perceived organizational support, greater merger uncertainty, less merger valence, less merger efficacy, and higher intention to turnover than employees of the dominant bank. Supported Hypothesis 3 Employees will tend to describe organizational opinion leaders in terms of positive interpersonal affect and similarity-with-self, rather than work-related attributes. Partially Supported Hypothesis 4a Employees of the dominated bank will tend to describe organizational opinion leaders in terms of positive interpersonal affect and similarity-with-self, more so than employees of the dominant bank. Supported Hypothesis 4b Employees of the dominant bank will tend to describe organizational opinion leaders in work-related terms, more so than employees of the dominated bank. Supported Hypothesis 5 Employees at both banks will describe their organizational opinion leaders as holding a higher position at work than themselves. Supported Hypothesis 6 Employees of the dominated bank will be more likely to describe their organizational opinion leaders as relationship-oriented, whereas employees of the dominant bank will be more likely to describe their organizational opinion leaders as task-oriented. Not Supported Hypothesis 7 Organizational opinion leaders in the dominated bank will be described as more influential than organizational opinion leaders in the dominant bank. Not Supported Research Question 1 Which attributes of the organizational opinion leaders are positively associated with their strength of influence over followers? OOL Job Performance & Job Performance Similarity Research Question 2 Do organizational opinion leader attributes associated with strength of influence vary according to pre-merger bank of employment? No 70 CHAPTER 4 DISCUSSION This dissertation examined how change recipients? attitudes and organizational context may be related to the characteristics of their referent leaders (i.e., opinion leaders) at work, and the extent to which they are then influenced by these people. The variables of interest in this study included (a) change recipients? attitudes during large-scale organizational change, (b) the follower-perceived descriptive traits of opinion leaders, and c) the strength of influence, as attributed by change recipients, to opinion leaders. Opinion leaders can either be effective agents of change for organizations or prove to be major obstacles. Thus, opinion leaders are a potential ally that, through their greater social influence, may assist in institutionalizing change in an organization. Despite their utility, much is to be learned about what makes opinion leaders influential to their peers (Lam & Schaubroeck, 2000). This dissertation borrowed from existing research in the consumer theory, leadership, and organizational change disciplines to craft the investigation of the organizational opinion leadership phenomenon. The mixed method design allowed for corroboration of findings regarding the culture differences affecting a merger and the associated job attitudes and merger perceptions experienced by the change recipients in both pre-merger organizations. This section discusses the findings related to the specific hypotheses and research questions addressed. Implications and limitations of this study, as well as areas for future research are also discussed. 71 Mergers and Employee Attitudes A corporate merger between two banks provided the organizational context which allowed for additional comparisons to be made between employee groups at the affected banks. The corporate merger was generally perceived by change recipients as either ?a big deal? or ?not much of a change,? depending on which pre-merger partner they worked for. The analysis of interview data and subsequent statistical tests of change attitudes and merger perceptions provided compelling evidence for the dominant/dominated theory of mergers as suggested by Buono & Bowditch (1985). The employees of the dominated bank clearly reported greater uncertainty and more adverse job attitudes and merger perceptions than did employees at the dominant bank. The recognition of similar attitudinal responses of employee groups to the focal merger provided an opportunity to compare and contrast bank employee groups, in addition to drawing conclusions regarding organizational opinion leadership overall. Describing OOLs In their study of desirable traits in others, Cottrell et al. (2007) concluded that trustworthiness was deemed the most important trait that individual?s ascribed to their ideal person. They found trustworthiness to be the most important trait across six of eight affiliate groups to include sports team, family teams, and employees. Thus, it was not surprising to find that trustworthiness was judged by change recipients to be the most important OOL attribute. Indeed, two of the next three attributes in the resultant hierarchy of OOL descriptions (i.e., supportive, caring) reflect a similar positive interpersonal affect. Honesty, which was ranked second overall, is supported by Anderson?s (1968) finding that it ranked second out of the of ten ?most likeable? 72 characteristics from a list of 555 personality adjectives. Furthermore, given the well- documented requirement for credibility in interpersonal relationships, the importance of honesty is understandable. Of note, the descriptions of OOLs were of actual focal others, and as such, contribute some practical relevance to the sociological study of ideal, or imaginary, others. Interestingly, despite being aggregated as a common trait, ?similar to me? was not listed prominently by change recipients as an OOL attribute. This finding seems at odds with the similarity-attraction concept. Perhaps the direct recognition of similarity required by this item biased the response. The precise mechanism for the diminished importance of similarity remains unclear. However, the finding supports Cottrell et al.?s (2007) assertion that based on methodological issues, the effect of similarity on desire for affiliation has been over-rated. The comparison of the most important OOL attributes across dominant and dominated banks provides support for the position that change recipients in the group that is considered aggrieved will describe their respective OOLs differently than the group reporting less change. Change recipients? descriptions of OOLs varied significantly across the bank regarding the attributes of trustworthy, supportive, job expert, makes- things-happen, and calm. The dominated bank employees described their nominated OOLs as trustworthy, supportive, and makes-things-happen more often than their counterparts, while employees of the dominant bank reported their opinion leaders as job experts and calm, more often than employees of the dominated bank. Taken together, theories of need-satisfaction motivation (Kouzes & Posner, 1995), change uncertainty, and the typology of social ties described by Umphress, et al. (2003) support the notion that change recipients experiencing uncertainty and possessing negative attitudes will 73 seek support from individuals they describe as trustworthy and supportive. Thus, it is understandable that associates experiencing less job satisfaction and less organizational support may seek emotional reassurance from their nominated OOL. Also, as frustrated change recipients may experience a sense of lack of efficacy, or loss of control, it is notable that they may seek an OOL who can make-things-happen, whereas those that are lesser affected by change may not seek instrumentality in the OOL. OOLs were overwhelmingly described as at a ?higher-level position.? However, as discussed, only 10% of the total identified their boss, manager, or supervisor as their OOL in the open-ended response block. This finding is in agreement with Rogers (2003) suggestion that opinion leaders hold a slightly elevated status relative to their followers. The OOL is most likely a more senior teller, as opposed to someone outside the teller position. This finding is supported by the large number of responses to the open-ended item citing ?more-experienced in my position? or ?? at my job? as an important descriptor of the OOL. The describing of higher-level individuals as OOLs suggests the seniority hierarchy may be working as it is designed. That is, associates in higher positions of are being sought for their advice and opinions. Though the results do not provide for precise conclusions regarding whether the referenced OOLs are directly above the employees (i.e., their supervisors), it appears that the employee opinion seeking could be described as at least diagonally vertical. That is, their opinion leader is positionally above them in some respect. While this finding may seem ordinary, it is often not the case (Larkin & Larkin, 1996). For organizations experiencing turmoil, hierarchical communication flow is often weak (Monge & Contractor, 2001). In times of change, either due to the inability of organization leaders to provide enough information 74 or because of a perceived lack of management credibility, peers often seek information from peers rather than vertically in the organization. Being outside the ?communication loop? diminishes the ability of the leader to influence followers. In fact, the content of the informal communication network (e.g., rumors) can be detrimental to the organization (Schweiger & DeNisi, 1991). As such, this finding that the informal and formal influence networks at the teller-level are generally aligned bodes well for the post-merger bank. The expectation that leadership style of OOLs at the dominated bank would be described as more relationship oriented, versus task oriented and that the reverse would be true for the dominant bank (more task oriented, fewer relationship oriented) was not borne out. Task- and relationship-oriented leaders were distributed evenly across the sample and former bank employer. This finding suggests that the OOLs employees tend to go to are equally likely to be task- or relationship-oriented, regardless of previous bank affiliation. A possible explanation for the contrary results is the interpersonal nature of the item. Participants were asked to describe their interactions in their own words. It is reasonable to expect folks to under-report social intimacy in this context (Knobloch & Carpenter-Theune, 2004). OOL Influence OOL influence was explored as a research question. A variety of dependent variables were tested via multi-level analysis. Results indicated job performance as a significant predictor of OOL strength of influence. It appears that OOLs perceived as performing well are able to wield more influence among their respective followers. This is not surprising given the credibility and expert power generally associated with high job 75 performance. Interestingly, the similarity in job performance between the follower and their respective OOL was also positively associated with OOL strength of influence. Taken together individuals who perceived their OOLs as high performers, similar to themselves, tended to rate their respective OOLs as being more influential. This finding was true across both banks. No significant differences existed between banks or bank branches. This result may shed light on the ability of managers to successfully identify opinion leaders in their organization (cf. Lam & Schaubroeck, 2000) Limitations This study was conducted in a field setting within a small organization. As such, generalizability beyond this specific context should be viewed with caution. As a cross- sectional study, the diagnostic findings are subject to the vagaries inherent in a one-shot analysis of organizational change. Indeed, capturing organizational change within a single time frame has been criticized as unrealistic (Golden-Biddle & Locke, 2006). No conclusions should be drawn regarding causality or directional relationships among variables predicting OOL influence. While change recipient attitudes were discussed as affecting the described attributes of OOLs, an obligatory reciprocal relationship exists whereby the OOL influence affects change recipient attitudes. The role that pre-merger bank cultures played in the selection and description of OOLs in the present study is unclear and remains a limitation. Teasing out culture effects from the merger change effects would require a systematic assessment of the two cultures, which was not possible at the time of this study. However, the item, ?At [work], what person?s opinion do you most value as you form your own attitude towards changes such as the merger, new policies, procedural changes, or new programs?? expressly asks 76 the participant to consider the merger, and other changes, for the nomination of the OOL. In their research on mergers and acquisitions, Buono and Bowditch (1989) emphasized that it is the threat of anticipated changes that aggrieves merger-affected individuals. While this study cannot rule out possible effects of pre-merger cultural style on OOL choice, it seems plausible that the varying degrees of organizational change reported by change recipients in the dominant and dominated banks contributed to the systematic differences in OOL descriptions. As with any study employing self-report measures, caution is in order when interpreting the results. Common-method bias may have inflated the relationships of interest. To mitigate this same-source threat to validity, qualitative data from interviews in the first phase of the study and quantitative measures in the second phase were triangulated to identify the dominant and dominated bank and provided key traits of individuals described as OOLs (Onwuegbuzie & Johnson, 2006). To further mitigate the threat of common method variance future research would benefit by incorporating measures of OOL influence other than self report. For example, investigator observation of OOL manifest influence, or network measurement of centrality (see for example, Bono & Anderson, 2005) have provide empirically validated measures of interpersonal influence. Concerns are reduced somewhat in that there were a large number of small correlations among the independent variables in Level 1 of the HLM model (see Table 16), which is inconsistent with mono-method bias (Lindell & Whitney, 2001). The present study used a dyadic approach to the identification of organizational opinion leaders. The requirement for anonymity precluded the data collection necessary for network mapping via social network analysis (SNA). The identification of opinion 77 leaders through SNA would provide a sense of magnitude of influence through the analysis of in-degree of centrality for the identified opinion leaders. Future studies should make use of SNA to identify additional attributes of OOLs. Identification of OOL characteristics will inevitably lead to the discussion of OOL development. That is, the possibility of developing attributes found in OOLs. The leadership style typology of task oriented, relationship oriented, and charismatic, (cf. Ehrhart & Klein, 2001) was chosen to provide contrast between OOL styles. However, no differences were observed in OOL style between banks. A more comprehensive evaluation of leadership styles (see for example, Kouzes and Posner?s (1995) typology of leader dimensions) may have enabled a better comparison between OOLs. Though given the open-ended nature of the qualitative data collected, deriving a comprehensive evaluation of leadership style was not possible from the analysis of the answers. Future studies, utilizing a more thorough measure of leadership style, may provide additional insight as to possible differences of OOL style across organizations. Implications and Areas for Future Study The present research moved beyond exploring information seeking behaviors to influence seeking outcomes and provides insight into the key antecedents of choosing an opinion leader. It is probably worth noting again that most organizational change interventions fail, at least in part, due to the resistance of those asked to adopt the change (Porras & Robertson, 1992). This study may help managers see the unseeable. By virtue of identifying the key follower-nominated attributes of influential peers within the workplace, the present research will assist managers in recognizing the individuals more likely to be OOLs within the organization. 78 An increased understanding of the mechanisms by which opinion leaders influence others in an organization should help managers increase their ability to leverage these potential change agents, thereby increasing their chances of successfully implementing change. The belief that ?opinion leader? moves beyond the individual and becomes an ?identity? that can then be described by archetypal characteristics lends credence to the recognition of broad attributes associated with opinion leaders (Smith, 2005). The present study contributes to the organizational change and opinion leader literatures by adding to the discussion some representative attributes of organizational opinion leaders. A profile of OOL characteristics and considerations for identification of OOLs was developed from the findings of the present study and is presented in Table 20. Understanding how attitudes and perceptions among change recipients and opinion leaders are related to informal influence networks, specifically the choice of OOLs, also contributes to the development of opinion leadership theory in an organizational change context. Interestingly, though the present research tested the similarity between change recipients and their nominated OOLs with respect to three of the Big Five personality constructs (i.e., conscientiousness, agreeableness, and extraversion) as predictors of OOL influence, none were found to be statistically significant. These findings were reinforced by the low numbers of individuals that chose OOL similarity to themselves as one of the top three OOL traits. This result seems contrary to the similarity-attraction paradigm and homophily research that has found support for similarity of demographic characteristics, personality, and values, as predictors of outcomes, to include performance, leadership and communication (Blau, 1985; Parsons & Liden, 1984; Tsui & O?Reilly, 1989). 79 However, research does suggest that the similarity-attraction relationship, specifically in regards to the environs of organizations is complex (Chatman & O?Reilly, 2004). Social psychological research also suggests that the strength of similarity?attraction relationships is conditional. For example, the kind of similarity shared (Tesser, Millar, & Moore, 1988), and self-monitoring personality (Jamieson, Lydon, & Zanna, 1987) have been found to influence similarity?attraction relationships. Thus, similarity-attraction may be diminished as a predictor of socialization when knowledge or trust is at stake in the organization. As Smith (2005, p. 567) points out, trust and confidence in opinion leaders is a result of their repeated demonstration of attaining desired outcomes for the followers enroute to the opinion leader ?achieving legitimacy?. 80 Table 20 A Manager?s Guide to Recognizing OOLs ? OOLs may support or oppose a focal organization change. Place emphasis on identifying and engaging opposing OOLs. ? OOL attributes, as perceived by the followers a , include the following: ? OOLs are thought to be trustworthy and honest, which underscores the requirement for credibility. ? OOLs are viewed as supportive and caring by those who go to them for opinions and advice. ? OOLs are perceived to possess the ability to make things happen, particularly by those change recipients who may be discontent with their present or expected circumstance. ? OOLs are described by change recipients as job experts, more so by those who experience a lesser degree of change. ? OOLs are perceived as calm, mainly by those less affected by change. ? OOLs are likely to be at a higher level in the work place than the change recipient, but probably in the same job position. ? OOLs express effective communication skills to their followers. b ? OOLs tend to be relatively high performers. ? The strength of OOLs? influence is increased by OOLs? perceived job performance, particularly in job performance as it relates to initiative and helping others perform. ? Change recipients perceiving greater job performance similarity with the OOL tend to be more influenced by him/her. a Should not assume the management perception is the same. Repeated participation of OOLs as change agents may influence change recipients? perceptions of OOL credibility (see ?wearing out of the opinion leaders? (Rogers, 2003, p. 388)). b Not explicitly measured in the present study, but consistent with the consumer theory and innovation diffusion literatures (cf. Childers, 1986; King & Summers, 1970; Rogers, 2003). Large scale organizational change often produces haves and have-nots (Rafferty & Griffin, 2006). As this study demonstrated, these groups often exhibit different job attitudes and change perceptions. The information that change recipients solicit from OOLs assists in the sensemaking process, ultimately affecting the change recipients? attitudes toward the change (Lam & Schaubroeck, 2000). Mapping affected stakeholders in an organizational change and appreciating the magnitude of its effect on groups will 81 help change managers differentiate between OOL traits deemed desirable by change recipients. The opinion leader traits described in this study are both consistent with, and expand, the current consumer-borne conceptualization of opinion leadership. It appears that the empirically-tested traits of trustworthiness, honesty, supportiveness, and caring, as well as the organizationally-relevant qualities of high job performance and status in the workplace provide a good starting point for OOL identification. Job expertise and even temperament play a larger role when organizational members are experiencing lesser change impact. As organizational change researchers continue to investigate the mechanisms by which change recipients are influenced, considerations to the interpersonal relationships that begin networks bear further study. Future research should also focus on how change recipients? personality and organizational context are related to the characteristics of those among their peers they choose to listen to, and the extent to which they are then influenced by this referent peer. Specifically, individual differences, including core self-evaluation, self-monitoring, conscientiousness, and extraversion, should be investigated with the context variables of organization trust and change uncertainty to determine their relationships with both, the characteristics of the chosen most-influential peer, and the magnitude of peer influence reported by the change recipient. Understanding how organizational context and individual differences among opinion seekers are related to informal influence networks, specifically the choice of influential peers, contribute to the development of opinion leadership theory in an organizational context. 82 REFERENCES Anderson, N. H. (1968). Likableness ratings of 555 personality-trait words. Journal of Personality and Social Psychology, 9, 272-279. Aristotle, (1995). Politics. Books I and II / Aristotle. (T. J. Saunders, Trans.). New York: Clarendon Press. (Original work published 350 B.C.) Armenakis, A. A. & Bedeian, A. G. (1999). Organizational change: A review of theory and research in the 1990s. Journal of Management, 25, 293-315. Armenakis, A. A., Bernerth, J.B., Pitts, J. P., & Walker, H. J. (2007). Organizational change recipients? belief scale: Development of an assessment instrument. The Journal of Applied Behavioral Science, 43, 481-504. Armenakis, A. A., & Fredenberger, W. B. (1997). Organizational change readiness practices of business turnaround change agents. Knowledge and Process Management, 4, 143-152. Armenakis, A. A., & Harris, S. G. (2002). Crafting a change message to create transformational readiness. Journal of Organizational Change Management, 15, 169-183. Armenakis, A. A., Harris, S. G., & Mossholder, K. W. (1993). Creating readiness for organizational change. Human Relations, 46, 681-704. Armenakis, A. A., Harris, S. G., Cole, M. S., Fillmer, J. L., & Self, D. R. (2007). A top management team?s reactions to organizational transformation: The diagnostic 83 benefits of five key change sentiments. Journal of Change Management, 7, 273- 290. Ashford, S. J. (1988). Individual strategies for coping with stress during organizational transitions. The Journal of Applied Behavioral Science, 24, 19-36. Ashford, S., & Black, J. (1996). Proactivity during organizational entry: The role of desire for control. Journal of Applied Psychology, 81, 199-214. Ashkanasy, N. M., Hartel, E. J., & Daus, C. S. (2002). Diversity and emotion: The new frontiers in organizational and behavior research. Journal of Management, 28, 307-338. Balkundi, P., & Harrison, D. A. (2006). Ties, leaders, and time in teams: Strong inference about network structure?s effects on team viability and performance. Academy of Management Journal, 49, 49-68 Bamberger, P., & Biron, M. (2007). Group norms and excessive absenteeism: The role of peer referent others. Organizational Behavior and Human Decision Processes, 103, 179-196. Bartunek, J. M., Rousseau, D. M., Rudolph, J. W., & DePalma, J. A. (2006). On the receiving end: Sensemaking, emotion, and assessments of an organizational change initiated by others. The Journal of Applied Behavioral Science, 42, 182- 206. Baugh, S., & Graen, G. (1997). Effects of team gender and racial composition on perceptions of team performance in cross-functional teams. Group & Organization Management, 22, 366?383. 84 Bentein, K., Vandenberg, R., Vandenberghe C., & Stinglhamber, F. (2005). The role of change in the relationship between commitment and turnover: A latent growth modeling approach. Journal of Applied Psychology, 90, 468?482 Blau, G. (1985). Relationship of extrinsic, intrinsic, and demographic predictors to various types of withdrawal behaviors. Journal of Applied Psychology, 70, 442- 450. Bono, J. E., & Anderson, M. H. (2005). The advice and influence networks of transformational leaders. Journal of Applied Psychology, 90, 1306-1314. Bordia, P., Jones, E., Gallois, C., Callan, V. J., & Difonzo, N. (2006). Management are aliens: Rumors and stress during organizational change. Group & Organization Management., 31, 601-621. Borgatti, S. P., & Molina, J. L. (2003). Ethical and strategic issues in organizational social network analysis. The Journal of Applied Behavioral Science, 39, 337-349. Bruning, J. L., & Kintz, B. L. (1968). Computational handbook of statistics. Glenview, IL: Scott, Foresman and Co. Buono, A. F., & Bowditch, J. L. (1989). The human side of mergers and acquisitions. San Francisco: Jossey-Bass. Buono, A. F., Bowditch, J. L., & Lewis, J. W., III. (1985). When cultures collide: The anatomy of a merger. Human Relations, 38, 477-500. Burke, W. (2002). Organization change: Theory and practice. Thousand Oaks, CA: Sage Publications. Byrne, D. (1969). Attitudes and attraction. In L. Berkowitz (Ed.), Advances in experimental social psychology, 52, 1167-1170. 85 Callan, V. J. (1993). Individual and organizational strategies for coping with organizational change. Work and Stress, 7, 63-75. Camman, C., Fichman, M., Jenkins, D., & Klesh J. (1983). Assessing the attitudes and perceptions of organizational members. In S. Seashore, E. Lawler, P. Mirvis, & C. Cammann (Eds.). Assessing organizational change: A guide to methods, measures and practices. New York: John Wiley. Cartwright, S., & Cooper, C. L. (1993). The psychological impact of merger and acquisition on the individual: A study of building society managers. Human Relations, 46, 327-348. Cartwright, D., & Zander, A. (1960). Group dynamics: Research and theory (2 nd ed.). Evanston, IL: Row, Peterson. Chan, K. K., & Misra, S. (2001). Characteristics of the opinion leader: A new dimension. Journal of Advertising, 19, 53-60. Chatman, J. A., & O'Reilly, C. A. (2004). Asymmetric reactions to work group sex diversity among men and women. Academy of Management Journal, 47, 193? 208. Chatman, J., Polzer, J., Barsade, S., & Neale, M. (1998). Being different yet feeling similar: The influence of demographic composition and organizational culture on work processes and outcomes. Administrative Science Quarterly, 43, 749?780. Chattopadhyay, P. (1999). Beyond direct and symmetrical effects: The influence of demographic dissimilarity on organizational citizenship behavior. Academy of Management Journal, 42, 273?287. 86 Childers, T. L. (1986). Assesment of the psychometric properties of an opinion leadership scale. Journal of Marketing Research, 23, 184-188. Coch, L., & French, J. R. P., Jr. (1948). Overcoming resistance to change. Human Relations, 1, 512-525. Cohen, J., & Cohen, P. (1983). Applied multiple regression/correlation analysis for the behavioral sciences (2 nd ed.). Hillsdale, NJ: Erlbaum. Cottrell, C. A., Neuberg, S. L., & Li, N. P. (2007). What do people desire in others? A sociofunctional perspective on the importance of different valued characteristics. Journal of Personality and Social Psychology, 92, 208?231. Creswell, J. W. (2003). Research design: Qualitative, quantitative, and mixed methods approaches (2 nd ed.). Thousand Oaks, CA: Sage. Creswell, J. W., & Plano Clark, V. L. (2007). Designing and conducting mixed methods research. Thousand Oaks, CA: Sage Cross, R. L., Parise, S., & Weiss, L. M. (2007). The role of networks in organizational change. The McKinsey Quarterly, 3, 28-41. Cunningham, C. E., Woodward, C. A., Shannon, H. S., MacIntosh, J., Lendrum, B., Rosenbloom, D., & Brown, J. (2002). Readiness for organizational change: A longitudinal study of workplace, psychological and behavioural correlates. Journal of Occupational and Organizational Psychology, 75, 377?392. Deutsch, M., & Gerrard, H. (1955). A study of normative and informational social influences upon individual judgments. Journal of Abnormal and Social Psychology, 51, 629?636. 87 Dietz, G., & Den Hartog, D. N. (2006). Measuring trust inside organizations. Personnel Review, 35, 557-588. Dicke, L. A. (2004). Street-level leadership in the public sector: A case study of the challenges. International Journal of Organization Theory and Behavior, 7, 231- 251. Donnellan, M. B., Oswald, F. L., Baird, B. M., & Lucas, R. E. (2006). The mini-IPIP scales: Tiny-yet-effective measures of the big five factors of personality. Psychological Assessment, 18, 192-203. Ehrhart, M. G., & Klein, K. J. (2001). Predicting followers' preferences for charismatic leadership: The influence of follower values and personality. Leadership Quarterly, 12, 153-179. Eisenberger, R., Huntington, R., Hutchison, S., & Sowa, D. (1986). Perceived Organizational Support. Journal of Applied Psychology, 71, 500-507. Engel, J. F., Blackwell, R. D., & Miniard, P. W. (1990). Consumer Behavior (6 th ed.). Chicago: Dryden Feder, G., & Savastano, S. (2005). The role of opinion leaders in the diffusion of new knowledge: The case of integrated pest management. World Development, 34, 1287-1300. Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7, 117- 140. Field, A. (2005). Discovering statistics using SPSS (2 nd ed.). London: Sage. Fields, D. L. (2002). Taking the measure of work: A guide to validated studies for organizational research and diagnosis. Thousand Oaks, CA: Sage. 88 Fisher, R. A. (1970). Statistical methods for research workers (14 th ed.). Hafner Publishing. Flynn, L. R., Goldsmith, R. E., & Eastman, J. K. (1996). Opinion leaders and opinion seekers: Two new measurement scales. Academy of Marketing Science, 24. 137- 147. Folkman, S., Lazarus, R.S., Gruen, R.J., & DeLongis, A. (1986). Appraisal, coping, health status, and psychological symptoms. Journal of Personality and Social Psychology, 50, 571-579. Fombrun, C. J. (1982). Strategies for network research in organizations. Academy of Management Review, 7, 280-291. Fowler, K., & Schmidt, D. (1988). Tender offers, acquisitions, and subsequent performance in manufacturing firms. Academy of Management Journal, 31, 962- 974. Fox, M. (1994). Reinvention of work: A new vision of livelihood for our time. San Francisco: Harper. Gail, S. F. (2003, January 30). Memo to AOL Time Warner: Why mergers fail?case studies [Electronic version]. Workforce. Retrieved January 15, 1008, from www.workforce.com/archive/feature/23/39/96/index.php - 55.0Kb Gavin, M. B., & Hofmann, D. A. (2002). Using hierarchical linear modeling to investigate the moderating influence of leadership climate. Leadership Quarterly, 13, 15-33. Gilmore, T. N., Shea, G. P., & Useem, M. (1997). Side effects of corporate cultural transformations. The Journal of Applied Behavioral Science; 33, 174-181. 89 Goldberg, L. R. (1992). The development of markers for the big five factor structure. Psychological Assessment, 4, 26-42. Goldberg, L. R. (1999). A broad-bandwidth, public-domain, personality inventory measuring the lower-level facets of several five-factor models. In I. Mervielde, I. Deary, F. De Fruyt, & F. Ostendorf (Eds.), Personality Psychology in Europe, 7, 7-28. Tilburg, The Netherlands: Tilburg University Press. Golden-Biddle, K., & Locke, K. D. (2006). Composing qualitative research. Thousand Oaks, CA: Sage. Goldsmith, R. E., & De Witt, T. S. (2003). The predictive validity of an opinion leadership scale. Journal of Marketing Theory & Practice, 11, 28-36 Goltz, S. M., & Hietapelto, A. (2002). Using the operant and strategic contingencies models of power to understand resistance to change. Journal of Organizational Behavior Management, 22, 3-22. Gumbel, P. (2003). Return of the urge to merge [Electronic version]. Time Europe Magazine, Retrieved December 12, 2007, from http://www.time.com/time/europe/magazine/article/0,13005,901030721- 464418,00.html Gutek, B. A. (1978). On the accuracy of retrospective attitudinal data. Public Opinion Quarterly, 42, 390-401. Harris S. G., & Mossholder, K. W. (1996). The affective implications of perceived congruence with cultural dimensions during organizational transformation. Journal of Management, 22, 527-547. 90 Herrmann, D. J. (1994). The validity of retrospective reports as a function of the directness of the retrieval process. In N. Schwarz & S. Sudman (Eds), Autobiographical memory and the validity of retrospective reports, (pp. 21?37). Berlin: Springer-Verlag. Hicks-Clarke, D., & Iles, P. (2000). Climate for diversity and its effects on career and organizational attitudes and perceptions. Personnel Review, 29, 324-347. Hogg, M. A., & Abrams, D. (1993). Towards a single-process uncertainty reduction model of social motivation in groups. In M. A. Hogg & D. Abrams (Eds.). Group motivation: Social psychological perspectives (pp. 173?190). London: Harvester/Wheatsheaf. Houghton, J. D., Anand, V., & Neck, C. P. (2003). Toward a framework of corporate merger processes and outcome: a behavioral perspective. International Journal of Public Administration, 26, 97-117. Ibarra, H. (1993). Personal networks of women and minorities in management: A conceptual framework. Academy of Management Review, 18, 56-87. Jamieson, D. W., Lydon, J. E., & Zanna, M. P. (1987). Attitude and activity preference similarity: Differential bases of interpersonal attraction for low and high self- monitors. Journal of Personality and Social Psychology, 53, 1052?1060. Judge, T. A., Bono, J. E., Ilies, T., & Gerhardt, M. W. (2002). Personality and leadership: A qualitative and quantitative review. Journal of Applied Psychology, 87, 765? 780. 91 Judge, T. A., Thoreson, C. J., Pucik, V., & Welbourne, T. M. (1999). Managerial coping with organizational change: A dispositional perspective. Journal of Applied Psychology, 84, 107-122. Kahn, W. A., Cross, R., & Parker, A. (2003). Layers of diagnosis for planned relational change in organizations. The Journal of Applied Behavioral Science, 39, 259-280. Kaiser, R. B., Hogan, R., & Craig, S. B. (2008). Leadership and the fate of organizations. American Psychologist, 63, 96-110. Kanungo, R. N. (1982). Measurement of job and work involvement. Journal of Applied Psychology, 67, 341?349. Katz, E., & Lazarsfeld, P. F. (1955). Personal influence. New York: The Free Press. Keller, E. B., & Berry, J. L. (2003). The influentials: One American in ten tells the other nine how to vote, where to eat, and what to buy. New York: Free Press. Kenny, D. A., Kashy, D. A., & Cook, W. L. (2006). Dyadic data analysis. New York: The Guilford Press. Kilduff, M., Tsai, W., & Hanke, R. (2006). A paradigm too far? A dynamic stability reconsideration of the social network research paradigm. Academy of Management Review, 31, 1031-1048 King, C. W., & Summers, J. O. (1970). Overlap of opinion leadership across consumer product categories. Journal of Marketing Research, 7, 43-50. Kline, R. B. (2005) Principles and practice of structural equation modeling (2 nd ed.). New York, NY: The Guilford Press. 92 Knobloch, L. K., & Carpenter-Theune, K. E. (2004). Topic avoidance in developing romantic relationships: Associations with intimacy and relational uncertainty. Communication Research, 31, 173-186. Kouzes, J. M., & Posner, B. Z. (1995). The leadership challenge: How to keep getting extraordinary things done in organizations. San Francisco: Jossey-Bass. Krackhardt, D., & Porter, L. T. (1986). The snowball effect: Turnover embedded in communication networks. Journal of Applied Psychology, 71, 1?6. Krippendorf, K. (2004). Content analysis: An introduction to its methodology. Thousand Oaks, CA: Sage Publications. Kurzban, R., & Neuberg, S. L. (2005). Managing ingroup and outgroup relationships. In D. M. Buss (Ed.), Handbook of evolutionary psychology, 653-675. New York: Wiley. Lam, S. K., & Schaubroeck, J. (2000). A field experiment testing frontline opinion leaders as change agents. Journal of Applied Psychology, 85, 987-995. Larkin, T. J., & Larkin, S. (1996, May-June). Reaching and changing frontline employees. Harvard Business Review, 95-104. Lawler, E. E. (1973). Motivation in work organizations. Belmont, CA: Brooks/Cole. Lazarsfeld, P. F., Berelson, B., & Gaudet, H. (1948). The people?s choice. New York: Columbia University Press. Lee, R., & Wilbur, E. R. (1985). Age, education, job tenure, salary, job characteristics, and job satisfaction: A multivariate analysis. Human Relations, 38, 781-792 93 Lewin K., Lippitt, R., & White, R. K. (1939). Patterns of aggressive behavior in experimentally created social climates. Journal of Social Psychology, 22, 272- 280. Lewis, W., Agarwal, R., & Sambamurthy, V. (2003). Sources of influence on beliefs about information technology use: An empirical study of knowledge workers, MIS Quarterly, 27, 657-678. Lindell, M. K., & Whitney, D. J. (2001). Accounting for common method variance in cross-sectional research designs. Journal of Applied Psychology, 86, 114-121. Lines, R. (2004). Influence of participation in strategic change: Resistance, organizational commitment and change goal achievement. Journal of Change Management, 4, 193-215. Louis, M. R. (1990). Acculturation in the workplace: Newcomers as lay ethnographers. In B. Schneider (Ed.), Organizational climate and culture (pp. 85-129). San Francisco: Jossey-Bass. Lynch, P., Eisenberger, R., & Armeli, S. (1999). Perceived organizational support: Inferior versus superior performance by wary employees. Journal of Applied Psychology, 84, 467-483. Maienhofer, D., & Finholt, T. (2002). Optimal targets for change agents: a computer simulation of innovation diffusion. Computational and Mathematical Organization Theory, 4, 259-281. Marin, A., & Hampton, K. N. (2007). Simplifying the personal network name generator. Field Methods, 19, 163-193. 94 Marks, M. L. (2006). Workplace recovery after mergers, acquisitions, and downsizings: Facilitating individual adaptation to major organizational transitions. Organizational Dynamics, 35, 384-399. Marks, M. L. & Mirvis, P. (1986). The merger syndrome. Psychology Today, 19, 36-42. Mascitelli, R. (2000). From experience: Harnessing tacit knowledge to achieve breakthrough innovation. Journal of Product Innovation Management, 17, 179? 193. Mayer, D., Nishii, L., Schneider, B., & Goldstein, H. (2007). The precursors and products of justice climates: group leader antecedents and employee attitudinal consequences. Personnel Psychology, 60, 929-964. McGrath, J.E. (1998). A view of group composition through a group-theoretic lens. In D.E. Gruenfeld (Ed.), Research on managing teams and groups: Composition (pp. 255-272). Greenwich, CT: JAI Press. McClelland, D. C. (1985). Human motivation. Glenview, IL: Scott Foresman. Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: An expanded sourcebook (2 nd ed.). Thousand Oaks, CA: Sage Publications. Miller, K. I., & Monge, P. R. (1985). Social information and employee anxiety about organizational change. Human Communication Research, 11, 365-386. Milliken, F.J., & Martins, L.L. (1996). Searching for common threads: Understanding the multiple effects of diversity on organizational groups. Academy of Management Review, 21, 402-443. Miner, J. B. (1978). Twenty years of research on role motivation theory of managerial effectiveness. Personnel Psychology, 31, 739-760. 95 Monge, P., & Contractor, N. (2001). Emergence of communication networks. In F. Jablin, & L. Putnam (Eds.), Handbook of organizational communication (2 nd ed., pp. 440-502). Thousand Oaks, CA: Sage. Montgomery, D. B., & Silk, A. (1971). Clusters of consumer interests and opinion leaders? spheres of influence. Journal of Marketing Research, 8, 317-321. Mohrman, S. A., Tenkasi, R. V., & Mohrman, A. M., Jr. (2003). The role of networks in fundamental organizational change: A grounded analysis. The Journal of Applied Behavioral Science, 39, 301-324. Mossholder, K. W., Settoon, R. P., Armenakis, A. A., & Harris, S. G. (2000). Emotion during organization transformations: An interactive model of survivor reactions. Group & Organization Management, 25, 220-243. Nagy, M. S. (2002). Using a single-item approach to measure facet job satisfaction. Journal of Occupational and Organizational Psychology, 75, 77-86. Nebus, J. (2006). Building collegial information networks: A theory of advice network generation. Academy of Management Review, 31, 615-637. Newcomb, T. M., Koenig, K. E., Flacks, R., & Warwick, D. P. (1967). Persistence and change: Bennington College and its students after twenty-five years. New York: Wiley. Nisbet, E. C. (2005). The engagement model of opinion leadership: Testing validity within a European context. International Journal of Public Opinion Research, 18, 3-30. Onwuegbuzie, A. J., & Johnson, R. B. (2006). The validity issue in mixed research. Research in the Schools, 13, 48-63. 96 Oreg, S. (2003). Resistance to change: Developing an individual differences measure. Journal of Applied Psychology, 88, 587?604. Oreg, S. (2006). Personality, context, and resistance to change. European Journal of Work and Organizational Psychology, 15, 73-101. Parsons, C. K., & Liden, R. C. (1984). Interviewer perceptions of applicant qualifications: A multi-variate field study of demographic characteristics and non- verbal cues. Journal of Applied Psychology, 69, 557-568. Paul, J. (1996). Between-method triangulation in organizational diagnosis. International Journal of Organizational Analysis 4, 135-156. Piderit, S. K. (2000). Rethinking resistance and recognizing ambivalence: A multidimensional view of attitude toward an organizational change. Academy of Management Review, 25, 783-794. Porras, J. L, & Robertson, P. (1992). Organization development: Theory, practice, and research. In M. D. Dunnette & L. M. Hough (Eds.), Handbook of industrial and organizational psychology, (2 nd ed., pp. 719-822). Palo Alto, CA: Consulting Psychological Press. Probst, G., & Raisch, S. (2005). Organizational crisis: The logic of failure. Academy of Management Review, 19, 90-105. Rafferty, A. E., & Griffin, M. A. (2006). Perceptions of organizational change: A stress and coping perspective. Journal of Applied Psychology, 91, 1165-1162. Raudenbush, S. W., Byrk, A. S., Cheong, Y. F., & Congdon, R. T., Jr. (2004). HLM 6: Linear and nonlinear modeling. Lincolnwood, IL: Scientific Software International, Inc. 97 Rentsch, J. R., & Schneider, B. (1991). Expectations for combination in postcombination life: A study of responses to merger and acquisition scenarios. Journal of Applied Social Psychology, 21, 233-252. Reynolds, F. D., & Darden, W. R. (1971). Mutually adaptive effects of interpersonal communication. Journal of Marketing Research, 8, 449-454. Robbins, S. P. & Judge, T. A. (2007). Organizational behavior (12th ed.). Upper Saddle River, NJ: Prentice Hall. Roberts, L. M., Dutton, J. E., Spreitzer, G. M., Heaphy, E. D., & Quinn, R. E. (2005). Composing the reflected best-self portrait: Building pathways for becoming extraordinary in work organizations. Academy of Management Review, 30, 712- 736. Rogers, E. M. (2003). Diffusion of innovations (5 th ed.). NY: Free Press. Rogers, E. M., & Cartano, D. G. (1962). Methods of measuring opinion leadership. Public Opinion Quarterly, 26, 435-441. Rokeach, M. (1973). The nature of human values. New York: Free Press. Rosnow, R. L. (1991). Inside rumor: A personal journey. American Psychologist, 46, 448-496. Ryan, B., & Gross, N. C., (1943). The diffusion of hybrid seed corn in two rural communities. Rural Sociology, 8, 15-24. Saunders, M. N. K., Thornhill, A. (2003). Organisational justice, trust and the management of change: An exploration. Personnel Review, 33, 174-86. Schein, E.H. (2004). Organizational culture and leadership. San Francisco: Jossey-Bass. Schwarz, N., & Sudman, S. (1996). Answering questions. San Francisco: Jossey-Bass. 98 Schweiger, D. M., & DeNisi, A. S. (1991). Communication with employees following a merger: A longitudinal field experiment. Academy of Management Journal, 34, 110-135. Siegel, S. (1956). Nonparametric statistics: For the behavioral sciences. New York: McGraw-Hill Book Co. Sinetar, M., (1981). Mergers, morale and productivity. Personnel Journal, 60, 863-868. Smith, H. W. (1975). Strategies of social research, the methodological imagination. New Jersey: Prentice Hall. Smith, P. A. C. (2005). Knowledge sharing and strategic capital: The importance and identification of opinion leaders. The Learning Organization, 12, 563-574. Staw, B. M., Sutton, R. U., & Pelled, L. H. (1994). Employee positive emotion and favorable outcomes at the workplace. Organizational Science, 5, 51-71. Stevenson, W. B. (2003). Introduction: Planned organizational change and organizational networks. The Journal of Applied Behavioral Science, 39, 238-242. Tashakkori, A. & Teddlie, C. (2003). Handbook of mixed methods in social and behavioral research. Thousand Oaks, CA: Sage. Tenkasi, R. V., & Chesmore, M. C. (2003). Social networks and planned organizational change. The Journal of Applied Behavioral Science, 39, 281-300. Tesser, A., Millar, M., & Moore, J. (1988). Some affective consequences of social comparison and reflection processes: The pain and pleasure of being close. Journal of Personality and Social Psychology, 54, 49?61. Tsui, A. S., & O?Reilly, C. A., III. (1989). Beyond simple demographic effects: The 99 importance of relational demography in superior?subordinate dyads. Academy of Management Journal, 32, 402?423. Tsui, A. S., Xin, K. R., & Egan, T. D. (1995). Relational demography: The missing link in vertical dyad link-age. In S. E. Jackson & M. N. Ruderman (Eds.). Diversity in work teams: Research paradigm for a changing workplace (pp. 97-129). Washington DC: American Psychological Association. Umphress, E. E., Labianca G., Brass D. J., Kass, E., & Scholten, L. (2003). The role of instrumental and expressive social ties in employees? perceptions of organizational justice. Organization Science, 14, 738-753. Valente, T., & Davis, R. (1999). Accelerating the diffusion of innovations using opinion leaders. Annals of the American Academy of Political and Social Science, 1, 55? 67. Van Knippenberg, D., & Van Leeuwen, E. (2001). Organizational identity after a merger: Sense of continuity as the key to postmerger identification. In M. A. Hogg & D. J. Terry (Eds.), Social Identity Processes in Organizational Contexts (pp. 249- 264). Philadelphia: Psychology Press. Van Scotter, J. R., Motowidlo, S. J., & Cross, T. C. (2000). Effects of task performance and contextual performance on systemic rewards. Journal of Applied Psychology, 85, 526 - 535. Vault, Inc. (2007, October 10). Job survey: Vice president and private banking relationship manager. Retrieved January 8, 2008, from www.vault.com/survey/occupational/AmSouth-Bank-9962.html 100 Venkatraman, M. P. (1990). Opinion leadership, enduring involvement and characteristics of opinion leaders: A moderating or mediating relationship? Advances in Consumer Research, 17, 60-67. Vishwanath, A. (2006). The effect of the number of opinion seekers and leaders on technology attitudes and choices. Human Communication Research, 32, 322-355. Vitale, D. C., Armenakis, A. A., & Feild, H. S. (2008). Integrating qualitative and quantitative methods for organizational diagnosis: Possible priming effects? Journal of Mixed Methods Research, 2, 87-105. Wainright, J. L., & Patterson, C. J. (2008). Peer relations among adolescents with female same-sex parents. Developmental Psychology, 44, 117-126. Wanberg, C. R, & Banas, J. T. (2000). Predictors and outcomes of openness to changes in a reorganizing workplace. Journal of Applied Psychology, 85, 132-145 Washington, R. R., Sutton C. D., & Feild, H. S. (2006). Individual differences in servant leadership: The roles of values and personality. Leadership & Organization Development Journal, 27, 700-716. Weimann, G. (1991). The influentials: Back to the concept of opinion leaders? Public Opinion Quarterly, 55, 267-279. Weimann, G. (1994). The influentials. State University of New York Press, Albany, NY. Wiener, Y. (1988). Forms of value systems: A focus on organizational effective. The Academy of Management Review, 13, 534-545. Witt, L. A., & Ferris, G. R. (2003). Social skill as moderator of the conscientiousness? performance relationship: Convergent results across four studies. Journal of Applied Psychology. 88, 809?820. 101 Yale, L. J., & Gilly, M. C. (1995). Dyadic perceptions in personal information search. Journal of Business Research, 32, 225-237. Yee Ng, K., & Van Dyne, L. (2005). Antecedents and performance consequences of helping behavior in work groups. Group & Organization Management, 30, 514- 540.