روابط بین رفتار تشویق و تنبیه رهبر و نگرش ها، ادراکات و رفتارهای پیرو: بررسی فراتحلیلی از پژوهش های موجود و جدید

Despite decades of research on the relationships between leader reward and punishment behaviors and employee attitudes, perceptions, and performance, no comprehensive examination of these relationships has been reported in the literature. This paper reports the results of two studies that address this issue. In the first study, data from 20 new samples were gathered on the relationships between leader reward and punishment behaviors and some criterion variables that have not been examined extensively in previous research. In the second study, a meta-analytic review was conducted incorporating both the new and existing research in order to provide estimates of the bivariate relationships between these leader behaviors and a variety of employee criterion variables across 78 studies containing 118 independent samples. Results of regression analyses designed to control for the effects of the other leader behaviors showed that: (a) the relationships between leader reward and punishment behaviors and employee attitudes, perceptions, and behaviors were more functional when the rewards or punishments were administered contingently than when they were administered non-contingently, and (b) these leader reward and punishment behaviors were strongly related to two variables (employees’ perceptions of justice and role ambiguity) that were expected to be key mediators of the relationships between these leader behaviors and the employee criterion variables. In addition, meta-analytic evidence from longitudinal studies suggested that the same leader behavior can be a cause of some employee criterion variables, and a consequence of others. Implications of these findings for future research in the area are discussed.

Ever since leader reward and punishment behaviors were introduced into the field in the 1970s (cf. Scott, 1977, Sims, 1977 and Sims and Szilagyi, 1975) they have been seen as central to the role of leaders, because they are important determinants of employee attitudes, perceptions and behavior. For example, in what he termed a reinforcement analysis of leadership, Sims (1977) proposed that leadership itself may be viewed as the management of reinforcement contingencies in work settings, and that the administration of reinforcing events contingent upon desirable or appropriate forms of employee behavior is critical to the development and maintenance of employee performance. From this perspective, positive reinforcers made contingent upon appropriate task behaviors should increase subordinate performance, while negative reinforcers (or aversive stimuli) should increase escape and/or avoidance responses on the part of employees. If these escape responses include functional task behaviors, then negative feedback may also lead to increases in employee performance. Similar predictions regarding the effects of leader contingent and non-contingent reward behaviors were also incorporated into the path-goal model of leadership (cf. Evans, 1970, Fulk and Wendler, 1982, House, 1971 and House and Mitchell, 1974). Path-goal theory suggests when a leader establishes a close linkage between subordinate performance and rewards it will increase performance, because such linkages increase the subordinate’s perception of instrumentality (cf. Evans, 1970 and House, 1971), and increase job satisfaction, because high performers should receive rewards of greater magnitude and frequency. Leader reward and punishment behavior is also the heart of what is called transactional leadership (cf. Avolio, 1999, Bass, 1985, Burns, 1978 and Howell and Avolio, 1993). According to Burns, 1978, Bass, 1985 and Bass, 1998, and others (cf. Atwater et al., 1997 and Avolio, 1999), transactional leaders are those who establish a “give and take” relationship with their subordinates in which the leader provides rewards to employees in exchange for their performance on the job. In their theories, transactional leaders can motivate followers either by clarifying expectations and identifying the rewards that they will receive for meeting these expectations (e.g., using contingent rewards), or by taking corrective actions when followers do not perform effectively. More specifically, Bass (1985, p. 122) argues that: Directly or indirectly, leaders can provide rewards for progress toward … goals or for reaching them. Or, they can impose penalties for failure ranging from negative feedback to dismissal. Such positive and aversive contingent reinforcement are seen as the two ways managers in organizational settings engage in transactional leadership to motivate employees. Contingent positive reinforcement…reinforces…employee performance. Contingent aversive reinforcement is a manger’s reaction to an employee’s failure to achieve the agreed-upon performance. The manager’s reaction signals the need to…modify or change the employee’s behavior. It signals the need for a reclarification of what needs to be done and how (Bass, 1985, p. 122). However, despite the obvious importance of leader reward and punishment behavior to these theories of leadership, no comprehensive meta-analytic review of the relationships between these key managerial behaviors and subordinate criterion variables exists. Of course, this is not to say that this important area has been totally neglected. Indeed, two meta-analyses (Lowe et al., 1996 and Judge and Piccolo, 2004) have been conducted, and both show the potential importance of contingent reward behavior in determining leadership effectiveness. For example, Lowe et al. (1996) reported that transactional leadership (in the form of contingent reward behavior) had a mean corrected correlation of .41 with leadership effectiveness, and Judge and Piccolo (2004) found that contingent reward behavior had stronger relationships than transformational leadership with three of the six criterion variables they examined (follower job satisfaction, motivation, and leader job performance), leading them to conclude that future research should study these behaviors in more detail. Unfortunately, both the Lowe et al. (1996) and the Judge and Piccolo (2004) meta-analyses are somewhat limited in scope with respect to the forms of leader reward and punishment behavior and the range of criterion variables they examined. For example, because these meta-analytic reviews focused on the transformational/transactional leadership model, they were limited to studies that included both transactional and transformational leadership behavior. Thus, research examining leader reward and punishment behavior using alternative leadership frameworks, such as the path-goal or reinforcement approaches, were outside the scope of these reviews. This is an important issue, because there have actually been more studies conducted on leader reward and punishment behavior using other conceptual frameworks, than there have been using the transformational/transactional approach. In addition, because of their interest in the transformational/transactional leadership model, the Lowe et al. (1996) and Judge and Piccolo (2004) meta-analyses focused on a relatively limited set of criterion variables, and did not consider the relationships between leader reward and punishment behaviors and important employee criterion variables like role perceptions (e.g., role ambiguity), justice perceptions (e.g., distributive justice, procedural justice), employee attitudes (e.g., organizational commitment, trust in one’s supervisor), and employee behaviors (e.g., task performance, extra-role behaviors). Finally, these meta-analyses did not include all four forms of leader contingent and non-contingent reward and punishment behavior. This is noteworthy because the predicted relationships between contingent forms of leader reward and punishment behaviors and employee attitudes, perceptions, and behaviors are substantially different from the predicted relationships between non-contingent forms of leader reward and punishment behavior and these same criterion variables; and because it is important to understand the relative magnitudes of their unique effects on these criterion variables. Therefore, the purpose of this paper is to report the results of two studies that provide a more comprehensive empirical summary of what is known about (a) the relationships between leader contingent and noncontingent reward and punishment behaviors and employee attitudes, role perceptions and performance, and (b) their relative (unique) effects on these criterion variables. More specifically, the goal of Study 1 was to gather data on the relationships between leader reward and punishment behaviors and a wide variety of criterion variables, some of which have not been examined extensively in previous research in the field. In Study 2, we provide a meta-analytic review incorporating all of the published data on the relationships between leader reward and punishment behavior and a wide variety of employee attitudes, perceptions and behaviors, plus the previously unpublished data from Study 1. In contrast to the limited scope of previous meta-analyses, an attempt was made in Study 2 to include every published article that has examined leader reward and punishment behavior, regardless of the theoretical orientation of the authors. Before presenting the results of our two studies, we will first briefly review several hypotheses regarding the differential relationships expected between leader contingent and non-contingent reward and punishment behaviors and employee criterion variables included in this research.

In spite of these limitations, the findings of our Avolio et al., 1999, Avolio and Howell et al., 1999, Avolio et al., 1991, Bare, 1978, Bass and Avolio, 1989, Bass and Yammarino, 1991, Bateman and Strasser, 1984, Brass and Oldham, 1976, Bycio et al., 1995, Deluga, 1991, Dionne et al., 2002, Dirks and Ferrin, 2002, Dorfman et al., 1997, Dubinsky et al., 1994, Dubinsky et al., 1995, Eagley et al., 2003, Ehrlich et al., 1990, George, 1995a, George, 1995b, Geyer and Streyer, 1998, Goodwin et al., 2001, Hackman et al., 1992, Hater, 1988, Hinkin and Schrieshiem, 1994, Howell and Hall-Merenda, 1999, Hunt et al., 1978, Jackofsky and Slocum, 1988, Judge and Bono, 2000, Kane and Tremble, 2000, Keller and Szilagyi, 1976, Kirby et al., 1992, Koh et al., 1995, Kohli, 1985, Oldham, 1976, Pearce and Sims, 2002, Podsakoff et al., 1986, Podsakoff and MacKenzie, 1995, Podsakoff et al., 1993, Podsakoff et al., 1993, Podsakoff and Todor, 1985, Roush and Atwater, 1992, Salam et al., 1997, Scully and Sims, 1994, Silins, 1992, Singer, 1985, Singer and Singer, 1989, Sosik et al., 2002, Spangler and Braiotta, 1990, Strasser et al., 1981, Szilagyi, 1980a, Szilagyi, 1980b, Turner et al., 2002, Vandenberghe et al., 2002, Waldman et al., 1987, Wayne et al., 2002, Williams et al., 1992, Wofford et al., 1998, Yammarino and Dubinsky, 1994, Yammarino et al., 1997, Yammarino et al., 1998, Yukl and Kanuk, 1979 and Zohar, 2002 suggest that leader reward and punishment behaviors have significant unique effects on a variety of important employee attitudes, perceptions and behaviors, and that the manner in which leaders administer rewards and punishments is a critical determinant of the effectiveness of these leader behaviors. In addition, perceptions of justice and role ambiguity were identified as potentially important mediators of the effects of these leader behaviors on employee criterion variables. Generally speaking, the results confirm the importance of leader reward and punishment behaviors and the central role they have been given in theories of leadership, and suggest several new avenues for future research.