تنظیم صحنه برای رهبری موثر: سابقه رفتار رهبری تحول گرا

Although transformational leadership behavior (TLB) has been linked to a number of positive organizational outcomes, research regarding the antecedents of such behavior is limited. Guided by Ajzen and Fishbein's theory of reasoned action [Psychological Bulletin 84 (1977) 888], we investigated two potentially relevant antecedents to performing TLB: cynicism about organizational change (CAOC) and the leader's social context—specifically peer leadership behavior. We hypothesized that CAOC would negatively predict TLB, while peer leadership behavior would positively predict TLB. Further, we expected that peer leadership behavior would have a positive moderating effect on leader CAOC. Data were gathered from 227 managers from multiple organizations and their 2247 subordinates. Findings supported the proposed hypotheses. Cynicism and peer leadership behavior explained nearly one quarter (24%) of the variance in TLB. Further, it appears that both CAOC and TLB may be malleable in organizational contexts. Implications for leadership research and practice are discussed.

After more than 20 years of accumulated research evidence, there is little doubt that transformational leadership behavior (TLB) is related to a wide variety of positive individual and organizational outcomes. Indeed, TLB has been empirically linked to increased employee satisfaction (e.g., Podsakoff, MacKenzie, Moorman, & Fetter, 1990), organizational commitment (e.g., Bycio, Hackett, & Allen, 1995), satisfaction with supervision (e.g., Podsakoff et al., 1990), extra effort (e.g., Seltzer & Bass, 1990), turnover intention (e.g., Bycio et al., 1995), organizational citizenship (e.g., Podsakoff, MacKenzie, Paine, & Bachrach, 2000), and overall employee performance (e.g., Yammarino, Spangler, & Bass, 1993). Moreover, the effects of TLB appear to be potent across management levels (e.g., Howell & Avolio, 1993), work environments (e.g., Bass, 1985), and national cultures (e.g., Bass, 1997). While researchers have learned a great deal about the effects of leader behavior, relatively little is know about its genesis. Put simply, it remains unclear why certain people engage in TLB while others do not. TLB, like most forms of leader behavior, is traditionally viewed as an independent variable exerting influence “downstream” (Bass, 1995). However, if we are to understand how to influence, improve, or modify the frequency and/or display of TLB downstream, it is necessary to shift our focus “upstream” towards the study of TLB as a dependent variable. With increased attention being paid to successful change management in the last decade (e.g., Cascio, 1993 and Katzenbach et al., 1995), organizations are accepting TLB as an important component of leading such change (e.g., Atwater & Bass, 1994, Burke & Litwin, 1992 and Worley et al., 1996). Unfortunately, organizations hoping to increase TLB have little, if any, empirical evidence to guide such change efforts. Leadership training has shown some promise in inculcating a knowledge and understanding of TLB (e.g., Barling, Weber, & Kelloway, 1996), but recent evidence suggests that the application of trained skills in the workplace is often heavily dependent on the organizational context (Baldwin & Magjuka, 1997). To date, the characteristics of the organizational context that facilitate, and are conducive to the practice of, TLB remain unspecified. That is not to say, however, that the transformational leadership literature is totally devoid of antecedent-oriented thinking. To the contrary, significant theoretical work has explored many important factors of transformational leadership and related styles such as charismatic leadership. In their forward to a special transformational/charismatic leadership issue of The Leadership Quarterly, Hunt and Conger (1999) noted that many of the contributors dealt with the notion of contextual and facilitating variables in relationship to TLB. Recent work by Shamir and Howell (1999) argued that some macro contextual variables (e.g., organizational life cycle, structure and culture) and meso contexts (e.g., goals, tasks, and technology) have a significant impact on the emergence and/or facilitation of charismatic leadership. Similarly, Bass (1985) proposed that collectivistic societies and levels of leadership might be important antecedents to transformational behavior. Conger and Kanungo (1987) purported that charismatic leadership should be related to certain behavioral components such as likableness, expertise, power base, and environmental sensitivity. Finally, as with much of the leadership research, predictions regarding leader dispositions have been offered. For instance, Bass (1998, p. 122) commented, “When it comes to predicting transformational leadership…there is no shortage of personality expectations. However, the empirical support has been spotty.” Bass continues by describing a long list of personality factors (e.g., locus of control, and conscientiousness), attitudes (e.g., optimism), and cognitions (e.g., moral reasoning) possibility associated with TLB. Consistent with Bass' assertions, Atwater and Yammarino (1993) found that a set of personal attributes (intelligence, warmth, conformity, sensing/intuition, thinking/feeling, emotional coping, behavioral coping, and athletic experience) accounted for 28% of the variance in TLB. Howell & Avolio (1993) found that leaders' internal locus of control was significantly related to intellectual stimulation and individualized consideration. Others have shown that variables, such as moral reasoning and physical fitness (Atwater, Dionne, Camobreco, Avolio, & Lau, 1998), emotional intelligence (Barling, Slater, & Kelloway, 2000), and impression management (Sosik, Avolio, & Jung, 2002), are all significantly related to TLB or charismatic leadership. Still other studies point to the potential fruitfulness of biodata as an antecedent to TLB (e.g., Avolio, 1994, Bass et al., 1987 and House et al., 1991). Although the review above highlights some important antecedent variables, two important limitations exist. First, the existing empirical literature tends to focus upon personal characteristics that are relatively difficult to change. While this basic research may be useful from a selection perspective, it presents substantial difficulties for those charged with training and developing people in existing management positions. Second, few studies have simultaneously examined the interaction between a malleable individual-level variable and the social context in which the person works. The purpose of the present study, then, is to explore antecedents of TLB across a variety of organizational contexts. However, with little empirical precedent, it was our belief that a strong theoretical basis for the inclusion of variables was essential. Rather than throw a host of typical measures “into the hopper,” our goal was to utilize a mainstream theoretical framework to guide the selection of variables that could be theoretically linked to the demonstration of TLB and then proceed from there. In organizations, one theory that has demonstrated utility in helping predict behavior from individual attitudes and context is the theory of reasoned action (Fishbein & Ajzen, 1975). From a reasoned action perspective, the best predictors of volitional behaviors are behavioral intentions. For example, the propensity to leave an organization (i.e., turnover intent) tends to be a stronger predictor of turnover than other variables such as satisfaction with pay or job involvement (Griffeth, Hom, & Gaertner, 2000). Further, the theory of reasoned action states that intentions are a function of two determinants: an individual's attitude toward the behavior and his/her perception of social pressure to perform (or not perform) the behavior. Fishbein and Ajzen (1975) refer to the latter determinant as “subjective norms.” As Ajzen (1988, p. 117) noted, “people intend to perform a behavior when they evaluate it positively and when they believe that important others think they should perform it.” Ajzen (1991) later added a third determinant of behavioral intention called “perceived behavioral control,” referring to the perceived ease or difficulty in performing a behavior based on prior experience, personal abilities, and external obstacles. Finally, when attitudes are matched to the appropriate level of specificity (i.e., general attitudes predicting general behavior and specific attitudes predicting specific behaviors; Ajzen & Fishbein, 1977) and are highly personally relevant (Leippe & Elkin, 1987), behavioral prediction improves (Petty, 1995). The theory of reasoned action provides a basis for selecting appropriate antecedents of TLB. In our case, leader attitudes likely to be predictive of TLB are those (1) that are matched to the construct, (2) highly relevant and (3) indicative of one's intention to perform the behavior. Because organizational change is a central element of TLB, one attitudinal construct that likely captures a leader's behavioral intention is cynicism about organizational change (CAOC) (Reichers, Wanous, & Austin, 1997). It is important to note that CAOC is conceptualized as a “state” or attitudinal measure as opposed to the more trait-oriented conceptualization of cynicism consistent with the MMPI and other personality assessments. Additionally, following the theory of reasoned action, to enhance the prediction of TLB, we should consider a leader's subjective norms and perceived behavioral control. In the organizational setting, a leader's peer group is likely to provide insight into both the leader's subjective norms and perceived behavioral control. To that end, it should be noted that we do not provide a direct test of Ajzen and Fishbein's theory; rather, we use it as a guide for the development of hypotheses. In the following section, we expand this rationale and outline the hypotheses for testing.

5.1. Synthesis of findings Considerable prior research has shown that TLB is related to a variety of positive organizational outcomes. Accordingly, the primary objective of this study was to better understand why certain managers engage in TLB while others do not. Across 227 practicing managers in a number of companies and contexts, we were able to explain about one quarter of the variance in TLB as a function of two variables: (1) those managers' level of cynicism toward organizational change and (2) their peers' use of leadership behaviors. While there are obviously other factors that contribute to the practice of TLB, the present study is an important step in specifying the social context that facilitates or inhibits such behavior. Perhaps the most compelling finding is that management members who have higher CAOC are less likely to engage in TLB. Thus, managers were in fact more likely to perform TLB when they believed that positive change was possible. Conversely, leaders with cynical attitudes regarding the potential for organizational change were significantly less likely to exhibit TLB. In short, the attitude of managers regarding their organizational context differentiates those who engage in TLB from those that do not. A second important finding was that the level of TLB exhibited by managers was positively related to the TLB of the managerial peers. The influence of peers has a rich history in behavioral research, dating at least to the Hawthorne studies (Roethlisberger & Dickson, 1939), and is also entirely consistent with the theory of reasoned action. As Ajzen (1988, p. 117) noted, “people are more likely to perform a behavior when they believe that important others think they should perform it.” Yet, however theoretically and intuitively appealing, peer influence has been curiously neglected in the empirical literature. Indeed, the present study is one of the first reports of which we are aware that has documented leadership outcomes associated with this type of peer behavior. A third finding that warrants highlighting is the significant interaction of cynicism and peer influence on TLB. Specifically, the presence of peers performing TLB in the environment of a cynical leader reduces the consequences of that leader's CAOC. Put another way, greater exhibition of TLB throughout an organization appeared to have the benefit of “canceling out” some of cynicism's negative effects on leaders. 5.2. Implications Few things matter more in the life of a firm than how people feel about how they are managed and led. Indeed, there is growing evidence that organizations that foster good leadership outperform their rivals (Pfeffer, 1994). Results of the present study point to several implications for how organizations view and manage their leadership development. First, CAOC (particularly among managers—who have much greater visibility than nonmanagers do) is likely to be a “poison” to the success of any TLB development initiative. Addressing the cause and cure for such cynicism is therefore critical for any organization that is hoping to build some TLB momentum. From a practical perspective, one encouraging feature of CAOC is its believed malleability; that is, unlike more stable attributes (e.g., neuroticism), some recent evidence suggests that cynicism can be influenced and reduced. A number of strategies for addressing cynicism have been proposed, and most deal with the ways in which change is communicated and the level of participation of managers in the change process Kotter, 2002 and Reichers et al., 1997. In our view, the present findings do not suggest the removal or censure of highly cynical managers (those who express doubt that organizational changes will be successful). Indeed, we contend that such an approach would often be shortsighted and ultimately counterproductive. In fact, recent research (McClough, Rogelberg, Fisher, & Bachiochi, 1998) suggests that organizational members who are cynical are often highly engaged and care deeply about organizational issues. Thus, a more fruitful approach would be to address cynicism directly in an attempt to redirect the negative energy exerted towards organizationally productive activities. Much like how former drug users often make the most compelling advocates of drug prevention, converted cynics can be among the most credible and persuasive champions of organizational change efforts. Second, the influence of peers discovered here suggests that leadership is as much a collective activity as it is an individual or heroic one. Although there is much recent interest in “top-grading” (i.e., the identification of top managerial talent) and high-potential programs (Smart, 1999), our findings point to the continued importance of the broad development of managerial talent. The old proverb that “it takes a village to raise a child” may be an apt analogue to the importance of an overall leadership culture, vis-à-vis a set of isolated and prominent individual leaders, for inducing TLB. Creating a context that develops, nurtures, and rewards the use of TLB across the full cadre of management members in an organization may well be a more effective course of action than more selective management development approaches. Taken together, the present findings add empirical support and specification of key factors to the already intuitive importance of organizational context in inducing desirable leadership behavior. Although training/educating managers in the “how,” “when” and “why” of TLB is well advised, it is unlikely to be sufficient in the absence of a context supporting TLBs. Recent work on the transfer of training has found that contextual factors can easily overwhelm the effects of the best planned and delivered learning experience. In contrast, a favorable context can enhance even suboptimal training Baldwin & Magjuka, 1997 and Rouiller & Goldstein, 1993. Of course, recognition of the importance of context to leadership is hardly new; Fiedler (1967) and others have made that case for many years. The vast majority of existing leadership research focuses upon the context moderating the effectiveness of certain behaviors as is the case in House's (1971) path goal theory, Kerr's (1977) substitutes for leadership approach, or Hersey and Blanchard's (1969) situational leadership theory. The findings of this study, however, address a more primary issue than the role of context in the situational models of the 1970s; that is, we now provide some evidence for why leader behavior occurs in the first place. 5.3. Future research directions Although this study has promising results and some positive methodological strengths (e.g., multiple sources of data, large multiorganization sample, multiple raters of behavior, etc.), two limitations warrant specific mention. First, the research design was cross-sectional, making any interpretations of causal direction tentative. For example, it is possible that leaders who engage in transformational behavior are perhaps more likely to identify with organizational change efforts and thus likely to be less cynical about organizational change. Similarly, the TLB of the focal leader could be influencing the TLB of his or her peers (rather than the opposite). Second, leadership is an area that garners great attention in both the academic and popular literature, and we suspect that there is a host of variables that others might postulate to predict TLB. We included two in our investigation, and the inclusion of other variables (e.g., abilities, perception of followers, boss's behavior, motivations) will certainly expand some of the current findings. When using samples of busy, overcommitted managers, the challenge is to efficiently collect those data (and just those data) which will explain the behaviors of most conceptual and practical import. The current study explained considerable variance in TLB, but there are certainly a number of other potential predictors that warrant investigation. For the future, we contend that shifting the focus “upstream” (i.e., toward antecedents) is not only warranted but has the potential to make strong contributions towards explaining why some leaders engage in TLB and others do not. One promising line of inquiry is the continued investigation of social context and attitudes to predict leader behavior. For example, the behavior of the leader's boss and behavioral feedback received from the leader's followers may contribute to strong normative situations. Similarly, with respect to attitudes, other change-oriented attitudes, such as organizational commitment, turnover intention, and resistance to change, may broaden the understanding of the antecedents to TLB. In addition, inasmuch as TLB is often driven from the leader's personal values, a focus on leader's values and congruence or shared values (Jones & George, 1998) with peers may be an important antecedent condition. It should be noted, however, that the present study and the above areas for research focus on proximal contextual variables that exist at the individual or group level. An ultimately more challenging task for researchers will be to understand antecedent conditions of TLB at the more distal, organizational level (Pawar & Eastman, 1997). The notion of “TLB-friendly” cultures is appealing but also wrought with complexity of research design and measurement. It will be creative empirical explorations into those domains that may well provide the most useful insights as to how best to set the stage for managers to engage in TLB.