|USFDC Home | USF Electronic Theses and Dissertations||| RSS|
This item is only available as the following downloads:
Leadership Style and the Link with Counterproductive Work Behavior (CWB): An Investigation Using the Job-Stress/CWB Model by Kari Bruursema A thesis submitted in partial fulfillment of the requirements for the degree of Master of Arts Department of Psychology College of Arts and Sciences University of South Florida Major Professor: Paul Spector, Ph.D. Jonathan Rottenberg, Ph.D. Joseph Vandello, Ph.D. Date of Approval December 13, 2004 Keywords: deviance, theft, sabotage, aggressi on, transformational, tr ansactional, abuse, justice, conflict, constraints Copyright 2005, Kari Bruursema
Dedication For beneficent people everywhere who agree to fill out surveys for research purposes. Bless them.
Acknowledgements Many thanks to my major professor, Dr. Paul E. Spector, who has examined, edited, and encouraged my writing from the very first, ve ry ghastly literature review and who had the patience and fortitude to revise this work countless times. My gratitude also extends to my committee members, Dr. Jonathan Rottenberg and Dr. Joseph Vandello, who provided fresh perspective, provocative ques tions, and a few important amendments to the study. I also appreciate the inspirational motivation provided by my colleagues, Angeline Ping Shin Goh, M.A.; Stacey Kessler, M.A.; and Lora Levett, M.A. Angel, my mentor, was the first to show me how to be a graduate student, how to conduct research, and how to write a scientific paper. And Stacey and Lora answered numerous questions along the way and led by assiduous, diligent, hard-working example, showing me that with a little more concentration, a completed masters thesis was possible. And finally, a sincere thank you to Dr. Tammy Allen whos e course in leadership was where the proposal for this study was born, and who was the fi rst to say, I think you should do the study.
i Table of Contents List of Tables.............................................................................................iii List of Figures............................................................................................iv Abstract........................................................................................................v Chapter One.................................................................................................1 Introduction............................................................................................1 The relationship between job stressors and CWB...........................3 Leadership and the work environment.............................................5 Leadership and its effects on CWB.................................................7 Transactional and transf ormational leadership and CWB...............9 Leadership style and the link with justice......................................15 Transactional leader ship and organizational CWB........................16 Study Objectives............................................................................17 Chapter Two...............................................................................................20 Method.................................................................................................20 Participants....................................................................................20 Procedure......................................................................................20 Measures.......................................................................................21 Leadership Style....................................................................21 Conflict.................................................................................22 Constraints............................................................................23 Justice....................................................................................23 Affect....................................................................................23 Autonomy.............................................................................24 Counterproductive Work Behavior.......................................24 Chapter Three.............................................................................................25 Results..................................................................................................25 Chapter Four..............................................................................................40 Discussion............................................................................................40 Hypothesis 1: Relationships Among the Stressors, Leadership Style, Negative Emotions, and CWB..........................41 Hypothesis 2: The Mediating Role of Negative Emotions............45 Hypothesis 3: Differential Relati onships Between Leadership Style and CWBO and Leadership Style and CWBP......................46 Hypothesis 4: Moderation by Jus tice of the Leadership-CWB Relationship...................................................................................47 Convergence between Self and Coworker Reports.......................49
ii Limitations.....................................................................................49 Conclusions....................................................................................51 References..................................................................................................52 Appendix: Study Questionnaire.................................................................59
iii List of Tables Table 1. One way ANOVAs for examining differences in 3 samples......25 Table 2. Descriptive statisti cs for all study variables...............................26 Table 3. Correlations among indepe ndent and dependent variables.........29 Table 4. Correlations among dependent variables....................................31 Table 5. Correlations among independent variables.................................32 Table 6. Agreement between sources.......................................................33 Table 7. Analysis of mediati ng role of negative emotion.........................34 Table 8. Results for Hotelling-Williams t-tests for Dependent Correlations.................................................................................37 Table 9. Results for moderated regression analysis with procedural justice as moderator..........................................38
iv List of Figures Figure 1. Spector and Foxs CWB Model..................................................2 Figure 2. The Proposed Model....................................................................4 Figure 3. Leadership/Justice Interaction...................................................38
v Leadership Style and the Link with Counterproductive Work Behavior (CWB): An Investigation Using the Job-Stress/CWB Model Kari Bruursema ABSTRACT Relations among job stressors, leadership style, emotional re actions to work, counterproductive work behavior (CWB), and autonomy were i nvestigated. Participants representing a wide variety of jobs were surveyed. Results indicate that transactional leadership style is related to negative emotions and occurr ence of CWB. Relationships between variables were mediated by emotions.
1 Chapter One Introduction Violence, theft, sabotage and other fo rms of counterproductive work behavior (CWB) are enormously costly to organizations from financial, image, and human capital perspectives. The U.S. Chamber of Commerce reports that $50 billion are lost annually by U.S. organizations due to employee thef t and fraud (U.S. Chamber of Commerce, 2002). The Chamber of Commerce (2002) also st ates that 20% of bus inesses fail due to internal theft and fraud. Na tionally, as many as six people are murdered every month at the hands of a co-worker or former co-worker (U.S. Department of Labor, 1996). Moreover, the U.S. Occupational Safety and Health Administration (OSHA), a division of the Department of Labor (DOL), reports that workplace violence costs U.S. companies 500,000 employees per year in voluntary and involuntary turnover. Due to its considerable harm, CWB, or intentional acts by employees to inflict harm on the organization or its members (Spector & Fox, 2002), is an important topic for organizations to understand and deal with. The purpose of this study was to addre ss the occurrence of CWB as a function of leader style. The explicit focus was on the e ffects of leadership on emotional reactions of subordinates and on their reports of committing ac ts of CWB. The influence of type of leadership on subordinates CWBs was inve stigated using an em otion/stress/CWB model that has been widely tested in the literature. Thus, this st udy served as a replication and extension of the model. A summary of th e hypothesized relationships among variables in this study is presented in figure 2.
2 The model, put forth by Spector and F ox (1999), casts CWB as a response to various stressors at work. This integrated CWB/job stress model has been well supported by recent work (e.g. Fox, Spector, & Miles, 2001; Spector & Fox, 2002; Goh, Bruursema, Spector, & Fox, 2003). In this model, threats to well-being, or stressors, induce negative emotional states like anger or anxiety a nd these emotions, which are the affective outcomes of stressors, lead to strains. Strains are outcomes of the job stress process that can be physical (e.g. headache), psychological (e.g. job dissatisfacti on), or behavioral (e.g. work withdrawal). CWB is a manifesta tion of a behavioral strain (Fox, Spector, & Miles, 2001). In short, negative perceptions of the work environment (i.e. stressors) relate to negative emotion, which is positively correlated with CWB. Taken as a whole, this research has demonstrated that an organizational focus on creating a positive environment as well as monitoring and mana gement of employee emotion may be an effective way to address the occurrence of CWB. Figure 1. Spector and Foxs CWB Model Perceived Stressors Negative Emotion Control Perceptions Counterproductive work behavior
3 The relationship between job stressors and CWB Job stressors are events that are interp reted as threats to ones well being and induce negative emotional reactions (Spect or, 1998). Organizational constraints are situations at work that inhibit task performance (Peters & OConnor, 1980). Organizational constraints have been con ceptualized as job stressors in the job stress/CWB model (Spector & Jex, 1998). The relationship between organizational constraints and CWB has been demonstrated. Specifically, constraints have been linked to acts of aggression, hostilit y, sabotage, theft, and withdr awal (Chen & Spector, 1992; Storms & Spector, 1987). Fox et al. (2001) also reported a correlation of .47 between organizational constraints and negative emo tion, thereby showing further support for its place as a stressor in the job stress/CWB model. Interpersonal conflict, or getting into arguments with co-workers, also has a demonstrated relationship with various kinds of CWB (e.g. Chen & Spector, 1992). Not only is conflict one of the most widely cited job stressors (Keenan & Newton, 1984), it also shows a strong relationship with ne gative emotion (r=.49; Fox et al., 2001). Justice, another type of job stressor, sp eaks to the perceived fairness of processes (in the case of procedural just ice) and outcomes (in the case of distributive justice) at work. Research on distributive and procedural justice has established that they contribute greatly to employee decisions to engage in CWB. Correlations of -.29 with both procedural and distributive justice and CWB (G oh et al., 2003) show that justice is an important stressor in the job-stress/CWB process.
4 According to the model (Spector & F ox, 1999), stressors have their effects on CWB through perceptions of control and aut onomy, and through emoti ons. The role of negative emotions in the model was discusse d previously and has been supported by tests of mediation (e.g. Fox et al., 2001 & Storms & Spector, 1987); however, the role of control in the process is more ambiguous. Control, or the extent to which individuals perceive that they have the ability to cope with and manage thr eats (Fox et al., 2001), has a demonstrated relationship with levels of employee stressors and physical strains ranging in seriousness from soma tic symptoms such as headaches to cardiovascular disease (Spector, 2002). However, attempts to place it as a moderator in the jobstress/CWB model have met with mixed resu lts (e.g. Fox et al., 2001). The relationship between perceptions of control and CWB is still being examined because as Allen and Greenberger (1980) pointed out, nonconstructiv e behavioral responses (such as CWB) are more likely when a person perceives low c ontrol of the situation. Therefore, the jobstress/CWB model posits that an individual interprets the en vironment, has an emotional response, and a belief about how much contro l he or she has over that environment, and then chooses to engage, or not engage, in CWB. Figure 2. The proposed model Leadership style Negative emotions CWB Procedural justice Control perceptions
5 Leadership and the work environment The role of leadership in creating the work environment is well established. For example, Dionne, Yammarino, Atwater, and Ja mes (2002) concluded that leadership is important to subordinate performance, satisf action and other outcomes regardless of the other individual, task, and organizational variables once thought to substitute for leadership. Specifically, the researchers f ound that leader member exchange (LMX), defined as the dyadic relationship between a leader and a subordinate (Graen & Cashman, 1975), and likeability of the leader correlate d significantly (r = .33 for LMX and r = .29 for likeability of the leader) w ith the performance indicator, namely, group effectiveness. On the other hand, none of the variables thought to substitute for leadership, defined as negating a leaders ability to positively or negatively influe nce subordinate attitudes and effectiveness (Dionne et al., 2002), correla ted in any significa nt way with group performance. These variables included fo rmalization of the or ganization, organization inflexibility, subordinate c ontrol, spatial difference betw een subordinates and leaders, subordinate indifference toward rewards, a nd subordinate professional orientation. In order to eliminate common-method bias, Dionne and colleagues used different subordinates to provide ratings of leader behaviors, subs titutes for leadership, and performance criterion. The lack of significant findings when data we re collected in this way led the researchers to conclude that prior significant effects in substitutes literature may be merely a statistical artifact, result ing from common-source bias (Dionne et al., 2002).
6 Other studies have also e xplored high LMX relationshi ps and their influence on subordinate and organizational outcomes. These studies seem to underscore the value of good leadership in effecting positive outcomes. In a meta-analysis by Gerstner and Day (1997), LMX was correlated .41 with subordina te/member performance ratings, .62 with satisfaction with supervision, .46 with overall job satisfaction, and .35 with organizational commitment. In a slightly different vein, Tierney, Bauer, and Potter (2002) found that leader member exchange re lated positively to subordinate willingness to perform extra-role behaviors. Extra-role behaviors are de fined as helpful, beneficial behaviors that go beyond an employees fo rmal work requirements. Outside of the LMX domain, other studies of leadership have shown that what a leader does and the feelings he or she cr eates in followers has important effects on follower behavior. For instance, George (1995) found that leader positive mood predicted group performance even after co ntrolling for group positive affective tone. Similarly, Williams, Podsakoff, and Huber (1 992) found that subordinate ratings of leader behaviors correlated with subordinate satisfaction with supe rvision, performance, and organizational commitment. Further evidence demonstrating the importance of the leader to follower outcomes comes from work on bad or abusive leader ship. For instance, Xin and Pelled (2003) found that task and particularly emotional conflict between superv isors and subordinates had negative associations with subordinate evaluations of leader behaviors. Another study examining the effects of conflict found that conflict with supervisors was negatively related to organizationally rele vant variables such as job satisfaction,
7 organizational commitment, and turnover inte ntions (Frone, 2000) Tepper (2000) not only examined subjective perceptions of leader behaviors but also objective indicators of subordinate satisfaction in his study of the consequences of abusive supervision. Abusive supervision referred to subordi nate perceptions of the extent to which supervisors engaged in the sustained display of hostil e verbal and nonverbal behaviors, excluding physical contact. The researcher found that reports of abusive supervision correlated negatively with job sa tisfaction (r = -.35), normative comm itment (r = -.27), and affective commitment (r = -.24), while correlating positively with self-re ports of emotional exhaustion (r = .36), work to family conflict (r = .22), and anxiety (r = .21). Taken as a whole, this research reveal s that poor leader-subordinate relationships have harmful effects on subordinates and the overall work environment just as positive leadersubordinate relations have beneficial effects. Leadership and its effects on CWB Despite the repeated finding that l eadership exerts important effects on subordinates, only a few studies have looked at characteristics of th e leader or leadersubordinate relationship as pr edictors of CWB. One study looking at this relationship examined the effects of high or low leader member exchange (LMX) on citizenship and retaliation behaviors (Townsend, Philips, & El kins, 2000). The theoretical basis of LMX is that dyadic supervisor-subordinate relati onships and work roles are negotiated over time through many interactions in which both supervisor and subordinate determine the type and quality of the rela tionship (Bauer & Green, 1996). High-quality leader-member exchange relationships have been associ ated with many positive outcomes including
8 citizenship behaviors, subordi nate satisfaction, and subordi nate promotions (Bauer & Green, 1996). In an attempt to examine th e flip side, Townsend and colleagues (2000) looked at outcomes of poor LMX relationships. They found that supervisors reported a higher incidence of CWB (whi ch they termed retaliatory behaviors) against the organization among subordinates in poor exchange relationships. High LMX relationships, on the other hand, were negatively correlated with supervisor reports of subordinate retaliation. This research suggests that leaders do have some impact on subordinate readiness to commit retaliator y acts that fit the definition of CWB. Teppers (2000) aforementioned work on outcomes of abusive supervision spoke to this relationship as well. He found that self-reports of abusive supervision correlated with many psychological strains such as anxiety, depression (r =.18), and emotional exhaustion. However, this research did not examine the effects of abusive supervision on behavioral strains such as CWB. Marrs (2000) found that verbal aggre ssion from supervisory sources, both witnessed and experienced, is negatively related to the affective outcomes of job satisfaction, affective organiza tional commitment, organizatio nal citizenship behaviors, trust in management, and positively related to stress. Moreover, verbal aggression from supervisors is associated with higher leve ls of deviant acts (CWB) on the part of organizational members and is associated with higher levels of intentions to leave the organization (Marrs, 2000). In order to unders tand deviance as Marrs conceptualized it, it is necessary to refer to Robinson and Bennetts (1995) work. Robinson and Bennett break deviance into four distinct categorie s: production deviance (e.g. purposely working
9 slowly), property deviance (e.g. wrecking suppl ies), political devian ce (e.g. manipulating gossip to affect promotions), and personal a ggression (e.g. beating someone up at work). Since deviance is operationally similar to CWB, relationships am ong the variables in Marrss study should be similar when CWB is used as the dependent variable instead of deviance. Further support for this contention comes from a study by Penney (2003) who found a correlation of .468 between self-reports of experienced incivility and self-reports of CWB. Incivility is defined as low intensity antisocial behavior that occurs at work (Pearson, Andersson, & Wegner, 2001). Although this study did not make a distinction between supervisor and other/co-worker sour ces of incivility, it shows that even lowgrade negativity has detrimental effects on employee willingness to commit CWB. In a somewhat different ve in, Giesburg (2001) examined employee perceptions of the causes and prevention of workplace viol ence and sabotage. In his study, 80% of employees stated that better communi cation by management could prevent the proliferation of workplace violence. This fi nding indicates that employees look to their leaders to improve the flow of communication and that they hold their leaders responsible when things go awry. Therefore, leadership creates the work environment both in terms of objective productivity, as described previ ously, and in terms of subjective employee perceptions, as this study indicates. Transactional and Transformational Leadership and the link with CWB Transformational leadership is the instilling of pride, self-respect and faith in the leader and is centered on the articulation of a vision for the organization (Masi & Cooke,
10 2000). Conversely, transactional leadership is characterized by the exchange of one thing of value for another between leader and subor dinates and careful correction of mistakes by the leader (Masi & Cooke). Bass (1985) operationalized the two types of leadership into multiple dimensions. Transformational leadership was operationalized as charisma, inspirational motivation, inte llectual stimulation, and indi vidualized consideration. Transactional leadership was operationalized into three dimensions: management by exception, contingent reward, and laissez-fair e (i.e. passive management by exception). Dimensions such as these have been empiri cally supported but the ex act factor structure has varied across samples (e.g. Avolio, 1999; Hater & Bass, 1988). There is considerable evidence that tran sformational leadership is effective in promoting positive follower and organizational results. Survey studies using the MLQ (Multifactor Leadership Questionnaire) and similar questionnaires find that transformational leadership re lates positively with subordinate satisfaction, motivation, and performance (Bass, 1996; Wofford, Whittington, & Goodwin, 2001). Moreover, Sparks and Shenk (2001) found that transfor mational leadership did indeed transform followers by encouraging them to see the hi gher purpose in their wo rk. They also found positive relationships between belief in th is higher purpose and job satisfaction, group cohesion, and subordinate effort (Sparks et al., 2001). Through structural equation modeling, McColl and Anderson (2002) found that transforma tional leadership has a significant direct influence on frustration a nd optimism, with the negative influence on frustration exerting a stronger effect on perfor mance than the positive effect on optimism. The emotion, frustration, and the belief, op timism, exert direct effects on performance
11 and fully mediate the relationship between tr ansformational leadersh ip and performance (McColl et al., 2002). This finding, with its emphasis on the importance of emotion, also lends support to the idea that leadership style could be a stressor in the Spector/Fox (1999) model. Elsewhere, researchers examined the effect s of type of leadership on subordinate motivation, commitment to quality, organi zational productivity, and self-image. A significant positive relationship was found be tween transformational leadership and subordinate motivation, while negative rela tionships were found be tween transactional leadership and both commitment to quality and organizational productivity (Masi & Cooke, 2000). These results imply that transa ctional leadersh ip may be related to CWB in that it relates negatively with both quality and quantity of work. Perhaps the lowered commitment to quality is expressed in sabotage or wasting supplies while lowered organizational productivity is due to CWBs such as taking longer breaks, purposefully slow work, showing up to work late, or theft. The particul ar question of how transactional leadership relates with CWB has not been addressed in the research, however. Another study addressed the impact of transformational and transactional leadership on sales performance and citi zenship behaviors among sales agents. The researchers found that transformational leader behaviors had stronge r relationships with both sales performance and citizenship behavi or than transactional leader behaviors (MacKenzie, Podsakoff, & Rich, 2001). Though this study found that transactional
12 leadership was associated to a lesser degr ee with positive outcomes, it stopped short of addressing the possible negative conseque nces of transactional leadership. In other literature, however, the detrimen tal effects of transactional leadership have been examined. In a study examining the effects of transactional and transformational leadership on consolidated business unit performance, Howell and Avolio (1993) found that contingent rewa rd and active management by exception, two facets of transactional lead ership, correlated negatively with consolidated unit performance (r = -.25 and -.41 respectively). Th e other facet of transactional leadership measured, namely, contingent reward be havior, correlated positively with unit performance (r = .37). This finding went contra ry to the researchers expectations that transformational leadership would be unifo rmly positively related to unit performance while transactional leadership would be uniformly negatively related. The authors suggested that it may have been due to problem s with the contingent reward scale. Later researchers found that the scale loaded on tw o separate factors, implicit and explicit rewards, and that the implicit factor loaded on other transformationa l leadership scales while the explicit factor re lated to transactional lead ership (Goodwin, Wofford, & Whittington, 2001). As predicted by Howell et al. (1993), transformational leadership correlated positively with consolidated unit performance. The specific components of transformational leadership examined we re charisma (r = .34 with performance), intellectual stimulation (r = .26) and individualized consider ation (r = .36; Howell et. al, 1993). This research shows that outcomes of tr ansactional leadership can be negative, at least in the performance domain.
13 In a meta-analysis by Podsakoff, MacKen zie, and Bommer (1996), substitutes for leadership (e.g. professional or ientation, indifference to rewa rds) accounted for more of the variance in criterion variab les (e.g. job satisfaction, orga nizational commitment) than did leader behaviors. This finding should not be surprising, however, as all of the leader behaviors examined in the meta-analyzed studie s were strictly transactional. The seven examined leader behaviors were leader cl arification, specifica tion of procedures, supportive leader, contingent reward, conti ngent punishment, noncontingent reward, and noncontingent punishment. None of the studies included in the meta-analysis investigated such leader be haviors as individualized consideration, inspirational motivation, idealized influence, or intellect ual stimulation. This meta-analysis thereby demonstrated that it is possible to substitute for transactional leader behaviors, and it established that the substitutes have stronge r influences on such outcome variables as organizational citizenship behaviors (OCB), in-role performance, and organizational commitment than do transactional leader behavi ors. Yet the analysis did not show that there is any substitute for transformational leadership. In order to conceptualize transactional leader ship as a stressor, it should fit the definition as being a situation that elicits ne gative emotional reactions (Spector, 1998). Transactional leadership, part icularly active management by exception, could be stressful to subordinates because it involves vigilant attention to subordinate mistakes. Active management by exception is operationally defi ned as looking for mistakes or enforcing rules to avoid mistakes (Yukl, 1999). To subor dinates, this type of monitoring could be interpreted as controlling and intrusive. The leaders careful correc tion of mistakes could
14 also cause the subordinate to lose faith in his or her own abilitie s, creating low selfefficacy and a sense of learned helplessness. Passive management by exception, or waiting until problems are serious before the leader responds to th e subordinate (Yukl, 1999), could also be very stress ful to subordinates. The empl oyee could feel that he or she is being persecuted or that the leader fails to notice his or her pos itive contributions. Similarly, the lack of transformational le adership could be considered just as critical of a stressor. A leader who lacks charisma, defined as the instilling of pride, faith and respect, a gift for seeing what is important, and the ab ility to transmit a sense of mission (Lowe et al., 1996), may leave followe rs without a sense of the bigger picture and without pride and faith in the organizati on and its goals. This could lead to such negative emotions as boredom or discouragemen t and also to CWB. A leader who lacks individualized consideration, defined as delegation of lear ning projects, coaching, and teaching (Lowe et al., 1996), may cause followers to feel that the workplace is impersonal or that the leader does not notice them as an indi vidual. This could lead to anger or sadness and in turn to CWB. Finall y, a leader who does not provide intellectual stimulation, defined as the emphasis of pr oblem solving skills and logical reasoning (Lowe et al., 1996), to followers may cause them to feel a host of negative emotions including boredom or anxiety. In order to establish it as a stressor, the link between leadership style and negative emotions will be examined. Though research has established that tran sactional leadership can have negative consequences, it is not clear that we can desc ribe it as a stressor unless it is known to relate to negative emotions and CWB in a si milar fashion as othe r stressors such as
15 constraints, justice, and conflic t. Therefore, linking leadersh ip style with other stressors and outcomes of the job stress/CWB model is an important goal of the present study. Leadership style and the link with justice Organizational justice is concerned with fair treatment of people in organizations (Muchinsky, 2000). Two types of justice are distributive, or peoples perceptions of fairness of outcomes received by self and ot hers, and procedural, or perceptions of fairness in the process that determines out comes (Fox et al., 2001). Although procedural justice has demonstrated stronger relati onships to emotions and CWB (e.g., CohenCharash & Spector, 2001; Fox, Spector, Miles, 2001), both types of justice have shown significant correlations within Spector a nd Foxs (1999) job-stress/CWB model. Distributive justice co rrelated -.38 with negative emo tion while procedural justice correlated -.44. In addition, distributive just ice correlated -.17 w ith organizational CWB but non-significantly (r = -.09) with personal CWB. Procedural justice correlated -.26 and -.15 with CWBO and CWBP respectiv ely (Fox, Spector, & Miles, 2001). In the leadership literature, there is a demonstrated relationship between transformational leadership and OCB thr ough justice (Rajnandini Schriesheim, & Williams, 1999). The structural model developed by the researchers showed a relationship fully mediated by justice and trus t. However, research to date has not examined the relationship of CWB to transformational leadership and justice. In terms of the job-stress/CWB model, one would expect an interaction between the environmental stressors of transactional l eadership style and procedural justice in the relationship with the strain, namely, CWB. This relationship is e xpected because of a
16 prior study in this area. This study examin ed the effects of procedural justice and charismatic leadership on cooperation and OCB. The results indicated that charismatic leadership and procedural justice both exer ted positive effects on cooperation, but the two variables interacted so that their effects were stronger together than alone (De Cremer & van Knippenberg, 2002). Another interaction observed in this study was that when procedural justice was low, charismatic leader ship was associated with higher levels of OCB, but when procedural justice was hi gh, the extent to which the leader was charismatic did not matter. A similar interaction is expected in the relationship with CWB; When procedural justice is high, leader style matters less than when justice is low. Transactional Leadership and Organizational CWB In Spector and Foxs (1999) model, ther e is a distinction between two types of CWB. This distinction was first deve loped by Robinson and Bennett (1995) in a multidimensional scaling study where, by usi ng both rational and empirical methods, they derived a typology of deviant workplace behaviors. Results show that deviant behaviors differ on two dimensions: minor ve rsus serious and in terpersonal versus organizational. Organizational CWB (CWBO) refers to all behavi ors directed at the organization as a whole (e.g. stealing money from the cash register); personal CWB (CWBP) covers behaviors dir ected at individuals within the organization (e.g. stealing money from a co-workers purse). Fox et al.s (2001) study showed a significantly stronger negative relationship between pro cedural justice and CW BO (r = -.26) than procedural justice and CWBP (r = -.15). Sin ce leadership style is directly related to
17 organizational processes, much like justice, relationships between the variables may be similar. In a similar vein, Bruk (2003) found that s ubordinate, self-repor ted conflict with supervisors correlated positively with CWBO (r = .21), but it did not correlate with CWBP. These findings suggest that employees target thei r counterproductive behaviors to the source of the problem; they do not randomly respond to environmental stressors. For this reason, the relative stre ngth of relationship between transactional leadership and CWBO versus CWBP was examined in this research. Study Objectives This study sought to cast leadership style as a job stressor in Spector and Foxs (1999) model. Previous research demonstr ated the superiority of transformational leadership over transactional leadership for many work outcomes. However, transactional leadership had not been examined as part of the job stress process or as a predictor of CWB. Therefore, this study investigated whet her leadership style related with CWB in similar ways as established job stressors such as orga nizational constraints, justice, and conflict do. In s hort, this study served as a replication and extension of Spector and Foxs (1999) model. Furthermore, this study examined the effects of high and low justice in moderating the relationship between leadersh ip style and CWB. Since justice had previously only been examined as a modera tor in the leadership and OCB relationship, this inquiry will be the first of its kind. Fi nally, the study explored the differential effects of transactional leadership on the two t ypes of CWB, organizational and personal.
18 Research suggests that transactional lead ership may be a stressor in Spector and Foxs (1999) job-stress/CWB m odel. Several studies suppor t the relationship between transactional leadership and both low produc tivity and poor job at titudes (e.g. Masi & Cooke, 2000; MacKenzie, Podsakoff, & Ri ch, 2001; Howell & A volio, 1993). If transactional leadership does conform to th e model as expected, it should show similar relationships to negative emotions and CWB as other stressors previously examined in the research such as conf lict and constraints (e.g. Fox, Spector, & Miles, 2001). Therefore, the following hypothesis is proposed: Hypothesis 1: High levels of conflict, or ganizational constraints, and transactional leadership and low levels of justice and transformational leadership will be associated with high levels of negative emotions and CWB. Negative emotions mediate the relationship between stressors and CWB in Spector and Foxs model (Fox et al., 2001). Therefore, the following hypothesis is proposed: Hypothesis 2: Negative emotions will mediate the relationship between stressors/leadership style and CWB. Studies have shown that subordinates respond to stressors from the organization with organization focused CWB (i.e. CWBO) and stressors from co-workers with coworker focused CWB (i.e. CWBP) (e.g. Bruk, 2003; Penney, 2003). Since leadership style deals specifically with organizationa l processes, the fo llowing hypothesis is proposed:
19 Hypothesis 3: Leadership style will be more strongly related to organizational CWB (CWBO) than personal CWB (CWBP). It is believed based on some prior res earch, that leader style matters less to subordinates when organizational justice is high than when orga nizational justice is low. Based on a study showing that justice and char ismatic leadership interacted in their effects on OCB (i.e. DeCremer & van Knippe nberg, 2002), and because it is important to check for expected moderation effects before looking at main effects, the following hypothesis is proposed: Hypothesis 4: The effects of leadership style on CWB will be moderated by procedural justice such that when procedural justice is high, leadership style matters less to subordinate CWB than when procedural justice is low.
20 Chapter Two Method Participants Participants were 172 employees from the Tampa Bay area, recruited from three sources. Participants were asked to choose one co-worker to independently rate their common supervisor on leadership style. Of the 172 respondents, 116 returned the coworker survey as well, resulting in 116 matc hed pairs. Sixty-two of the 172 participants were male (36%); 40 participants were in managerial positions ( 23%); 132 participants were in white collar jobs (76%); and the remaining participants reported having blue collar jobs. Participants had to work at le ast 20 hours to be incl uded in the study, and on average they worked 37.7 hours per week. To ensure anonymity, no names or specific places of employment were collected. Procedure Surveys were administered to employe d graduate students in programs including education, physics, economics, business, publ ic health, and womens studies. The researcher obtained permission from instruct ors to visit graduate classes and request participation from students. Participation wa s voluntary and did not a ffect course grade. Participants were asked to recruit a co-wor ker to complete the co-worker survey and complete their own survey. They then had two options for returning the surveys to the researcher. The first option was for the part icipant to place the two surveys in the same envelope and send them via in tra-campus mail to the resear chers mail stop. The second option was to return it directly to the resear cher who would visit the class the following
21 week. Fifty-five percent of the sample was generated in this manner. The remainder of the sample is made up of two distinct groups : undergraduates who pa rticipated for extra course credit and to whom surveys were admi nistered in the same manner as the graduate students, and employees from an outside orga nization not associated with the university. The organizational participants returned their completed surveys in a manila envelope to an in-basket upon completion. The researcher picked up completed surveys from the inbasket after the sp ecified deadline. Measures Participants surveys included measures of supervisors leadership style, participants constraints, just ice, conflict, partic ipants counterproducti ve work behavior, participant autonomy, and participants posi tive and negative affect Coworker surveys contained only the measure of s upervisors leadership style. Leadership Style. Supervisors leadership st yle was measured using the MLQ Form 5x Short instrument (Bass & Avolio, 1995). This measure was chosen because it distinguishes between active and passive ma nagement by exception and because of its frequent use in the literature. Participan ts and coworkers responded based on a 5-point Likert scale (0=not at all, 4= always) to how often their superv isor displays specific leader behaviors. Five scales m easure transformational leadership and four scales measure transactional leadership (Turner et. al, 2002) The scales measuring transformational leadership are: (a) Attributed Idealized In fluence (sample item: Goes beyond his or her own self-interest for the good of the group). Average coeffi cient alpha (for participant and co-worker reports) for this facet in the present study was .81, (b) Behavioral
22 Idealized Influence (e.g. Specifies the importa nce of having a strong sense of purpose). Average coefficient alpha for this facet in the present study was .75. (c) Inspirational Motivation (e.g. Articulates a compelling vision of the future). Average coefficient alpha for this facet in the present study wa s .87. (d) Intellectual Stimulation (e.g. Seeks differing perspectives when solving problems). Average coefficient alpha for this facet in the present study was .80. (e) Individualized Consideration (e.g. Treats each of us as individuals with different need s, abilities, and as pirations). Average coefficient alpha for this facet in the present study was .81. Th e scales measuring transactional leadership are: (a) Contingent Reward (e.g. Makes cl ear what I can expect to receive if my performance meets designated standards). Av erage coefficient alpha for this facet in the present study was .78. (b) Management-by-Ex ception Active (e.g. Keeps track of my mistakes). Average coefficient alpha for this facet in the present study was .71. (c) Management-by-Exception Passive (e.g. Things have to go wrong for him/her to take action). Average coefficient alpha for th is facet in the present study was .75. (d) Laissez-faire (e.g. Avoids getting involved when important issues arise). Average coefficient alpha for this facet in the present study was .75. Conflict. Work conflict was measured using Frones (2000) modified version of the Interpersonal Conf lict at Work Scale (ICAWS; Spect or & Jex, 1998). Each set of questions measures the extent to which th e employee experiences arguments, yelling, and rudeness while interacting with the supervisor or co-workers, respectively. The scale consists of 4 items rated on a 5-point scale ranging from 1=Never to 5=Every day. High
23 scores represent high levels of conflict. Frone (2000) reported a Cronbach alpha of .86 for conflict with supervisors and .85 for conflict with co-workers. Constraints. The Organizational Constraints Scale (OCS), developed by Spector and Jex (1998), was used to measure job constr aints. This 11-item scale is based on the constraints identified by Peters and OConnor (1980). Respondents indicate on a fivepoint scale ranging from never to every day how frequently their work performance was hindered by constraints such as inadequate help from supervis ors, incorrect instructions, or lack of equipment. High scores indicate high levels of constraints. Spector and Jex (1998) reported a mean Cronbach al pha of .85 for this scale. Justice. Procedural justice was measured us ing Moormans (1991) 12-item scale. Response choices range from 1=strongly disagr ee to 5=strongly agree, with high scores representing high levels of procedural justice. Mean alpha for this scale is .94 (Moorman, 1991). Affect. The Job-Related Affective Well -Being Scale (JAWS), developed by VanKatwyk, Fox, Spector, and Kelloway (2000), measures a wide range of emotions drawn out in response to the job. Respondent s indicated how often they experience each of 20 emotional states. Response choices are in the standard 1 to 5 Likert format where a 1 indicates almost never and a 5 indicates ex tremely often or always. Therefore, high scores represent high levels of each emoti on. Ten positive emotion items are summed to yield a positive affect score and ten negativ e emotion items are added to obtain a negative affect score. Only the negative emotions sc ore were used in the hypotheses in the current study, however both were collected in order to keep the scale bala nced and to look at
24 relationships with the leadership variable s and positive emotions. VanKatwyk et al. (2000) reported a .95 coefficient alpha for this scale. For this study, coefficient alpha for the negative emotion scale was .88 while alpha for the positive emotion scale was .91. Autonomy. Work autonomy was measured using the Factual Autonomy Scale (FAS; Fox, Spector, & VanKatwyk, 1997), whic h provides items that are factual in nature and resistant to affec tive bias. The reduced seven-it em scale (as used in prior work such as Goh, et al., 2003) has a reporte d alpha of .87 (Fox et al., 2001). A sample item is: Do you have to ask permission to take a rest break? Answer choices range from 1=never to 5=always. Therefore, hi gher scores indicate less autonomy. Counterproductive Work Behavior. CWB was assessed using a behavioral checklist based on a master list compiled fr om a number of existing measures and previously used by Goh and colleagues (2003). The checklist includes as many distinct behaviors as possible without duplicating items. The 45-item list requires respondents to indicate the frequency with wh ich they engage in specific be haviors; there are 5 response choices ranging from 1=never to 5=every da y. Therefore, high scores indicate high incidence of CWB. Subscale scores were comp uted consisting of items that targeted the organization (e.g. showing up late for work) a nd behaviors targeting individuals within the organization (e.g. insulting someones work).
25 Chapter Three Results To determine if I was justified in co mbining the graduate student, undergraduate student, and non-student samples, one-way ANOVAs were run for all study variables (see Table 1). Significant differences were found for autonomy (F(2, 169) = 5.60, p = .0044), negative emotion (F(2, 169) = 4.67, p = .0106), conflict (F(2, 169) = 4.23, p = .0162), CWB (F(2, 169) = 3.79, p = .0247), transac tional leadership (F(2, 169) = 5.32, p = .0057), and hours worked per week (F(2, 169) = 10.90, p < .0001) Recall that for the autonomy variable, higher scores indicate less autonomy; therefore, the graduate sample reported significantly more autonomy (M = 15.49 SD = 6.95) than did the undergraduate sample (M = 19.58, SD = 7.43). Constraints, pr ocedural justice, di stributive justice, positive emotion, and transfor mational leadership showed no significant differences among the three groups. Since these differences were relatively minor, the samples were combined for further analysis. Table 1. One way ANOVAs for examining differences in 3 samples F(2, 169)R2 Undergraduates M (SD) Graduates M (SD) Non-students M (SD) Autonomy 5.60** .06 19.58 (7.43)a 15.49 (6.95)b 16.46 (5.96) Negative Emotion 4.67* .05 26.63 (8.39)a 24.85 (8.16) 20.71 (7.87)b Conflict 4.23* .05 12.19 (4.09)a 10.51 (2.88)b 10.61 (3.48) Transactional Leadership 5.32** .06 29.38 (8.32)a 25.96 (7.94)b 23.43 (7.94)b Transformational Leadership 1.40 .02 44.94 (19.26) 46.58 (17.81) 39.96 (18.95) Hours worked 10.90***.11 32.95 (7.74)a 39.70 (9.11)b 39.00 (6.28)b
26 Table 1. (Continued) Procedural Justice .45 .01 56.56 (17.79) 56.56 (16.00) 59.75 (14.92) Distributive Justice 3.40* .04 19.92 (5.71) 17.78 (6.64) 20.89 (6.82) Positive Emotion .16 .00 28.92 (9.06) 29.85 (9.75) 29.79 (9.14) Constraints 1.58 .02 24.92 (7.89) 27.01 (8.26) 24.61 (8.03) CWB 3.79* .04 64.67 (20.64)a 58.36(10.12)b 57.18 (12.59) Different letters across a colu mn indicate that the two mean s are significantly different from each other. Descriptive statistics were calculated fo r all study variables. Means, standard deviations, and ranges (observed and possibl e) can be found in Table 2. The observed values of most variables spanned the range of possible values. However, CWB, conflict, and both participant reported and co-worker reported tr ansactional leadership had observed ranges that were much smaller than possible. This restri ction in range could attenuate correlations with these variables. Table 2. Descriptive statis tics for all study variables IV Mean SD Observed range Possible range Coefficient alpha Constraints 26.02 8.15 11-48 11-55 .86 Procedural Justice 57.11 16.14 14-84 12-84 .95 Distributive Justice 18.89 6.53 6-30 6-30 .93
27 Table 2. (Continued) Autonomy 16.81 7.14 7-35 7-35 .87 Negative Emotion 24.63 8.43 10-48 10-50 .88 Positive Emotion 29.54 9.45 10-50 10-50 .91 CWB 59.73 13.95 44-167 44-220 .92 Conflict 11 3.42 8-24 8-40 .81 Contingent Reward 9.18 4.34 0-16 0-16 .81 Passive mngmt. by exception 5.68 4.03 0-16 0-16 .73 Active mngmt. by exception 6.98 3.76 0-16 0-16 .71 Laissez-faire 4.59 3.59 0-15 0-16 .75 Intellectual Stimulation 8.36 4.03 0-16 0-16 .82 Behavioral Idealized Influence 8.57 3.88 0-16 0-16 .73 Attributed Idealized Influence 9.14 4.32 0-16 0-16 .81 Inspirational Motivation 10.03 4.23 0-16 0-16 .86 Individualized Consideration 8.85 4.32 0-16 0-16 .82
28 Table 2. (Continued) Overall Transactional 26.39 8.28 9-48 0-64 .65 Overall Transformational 44.36 18.39 0-78 0-80 .94 Coworker Contingent reward 9.88 3.77 0-16 0-16 .75 Coworker Passive mngmt. by exception 5.23 3.89 0-15 0-16 .76 Coworker Active mngmt. by exception 7.82 3.44 0-16 0-16 .71 Coworker Laissezfaire 4.30 3.52 0-13 0-16 .75 Coworker Intellectual Stimulation 8.79 3.76 0-16 0-16 .77 Coworker Behavioral Idealized Influence 9.27 3.87 0-16 0-16 .76 Coworker Attributed Idealized Influence 9.42 4.17 0-16 0-16 .81 Coworker Inspirational Motivation 10.20 4.15 0-16 0-16 .87 Coworker Individualized Consideration 9.37 4.06 0-16 0-16 .79 Coworker Overall Transactional 27.20 7.20 12-47 0-64 .56 Coworker Overall Transformational 46.88 17.72 0-78 0-80 .94
29 Zero-order Pearson correlations were computed for both organizational and personal CWB with leadership style and other job stressors (i.e. c onstraints, conflict, justice, autonomy). These co rrelations are reported in ta ble 3. Correlations among the dependent variables are given in table 4; correlations am ong the independent variables are given in table 5. Thus, tables 3, 4, and 5 collectively give correlations among all study variables. It was also necessary to look at the de gree of relationship be tween self and coworker reports of leadership style. Particip ant and co-worker reports of leadership style were significantly correlated. All correlations were signific ant at the .001 level except for active management by exception which was significant at .01 level. Correlations for the facets ranged from r = .27 (for active management by exception) to r = .57 (for attributed idealized influence). The overall se lf and co-worker reports of leadership style also correlated significantly (r = .47) fo r both transformational and transactional leadership. Correlations between participant and co-worker reports fo r all facets of the leadership scale can be found in table 6. Table 3. Correlations among indepe ndent and dependent variables. IV CWB CWBO CWBP Negative emotions Positive Emotions Constraints .30*** .31*** .21** .52*** -.33*** Procedural Justice -.29*** -.26*** -.25*** -.54*** .29*** Distributive Justice -.17* -.15 -.14 -.46*** .26***
30 Table 3. (Continued) Autonomy .16* .18* .11 .16* -.25*** Conflict .47*** .40*** .47*** .47*** -.21** Conflict with supervisors .47*** .46*** .39*** .44*** -.23** Conflict with coworkers .37*** .24** .41*** .34*** -.08 Transactional Leadership .24** .19* .20* .30*** -.08 Transformational Leadership -.16* -.21** -.05 -.24** .35*** Passive mngmt. by exception .25** .22** .22** .41*** -.23** Laissez-faire .31*** .30*** .23** .40*** -.22** Contingent Reward -.11 -.18* -.02 -.26*** .30*** Active mngmt. by exception .07 .08 .01 .18* -.09 Individualized consideration -.20* -.28*** -.06 -.30*** .38*** Attributed Idealized Influence -.24** -.28*** -.14 -.27*** .30*** Intellectual Stimulation -.13 -.20** -.03 -.19* .27*** Behavioral Idealized Influence -.13 -.18* -.04 -.11 .23**
31 Table 3. (Continued) Inspirational Motivation -.14 -.16* -.08 -.21** .25*** Co-worker Transactional .19* .15 .17 .19* -.06 IV CWB CWBO CWBP Negative emotion Positive emotion Co-worker Transformational -.13 -.09 -.13 -.10 .16 Co-worker Contingent Reward -.15 -.13 -.13 -.21* .19* Co-worker Passive mngmt. By exception .23* .22* .16 .28** -.15 Co-worker Active mngmt. By exception .06 -.02 .14 .04 -.03 Co-worker Laissez-faire .22* .21* .16 .26** -.13 Co-worker Intellectual Stimulation -.08 -.05 -.09 -.07 .14 Co-worker Behavioral Idealized Influence -.13 -.07 -.17 .02 .07 Co-worker Attributed Idealized Influence -.11 -.09 -.10 -.13 .18 Co-worker Inspirational Motivation -.11 -.06 -.14 -.04 .13 Co-worker Individualized Consideration -.15 -.15 -.09 -.20* .17 p<.05, ** p<.01, *** p<.001
32 Table 4. Correlations am ong dependent variables DV 1 2 3 4 1. Negative emotion 2. Positive emotion -.38*** 3. CWB .52*** -.14 4. CWBO .51*** -.18* .90*** 5. CWBP .42*** -.09 .89*** .61*** p<.05, ** p<.01, *** p<.001 Table 5. Correlations among independent variables 1 2 3 4 5 6 7 8 9 10 1. Constraints 2. Procedural Justice -.47*** 3. Distributive Justice -.42*** .50*** 4. Autonomy .20** -.14 .00 5. Conflict .45*** -.40*** .38*** .29*** 6. Conflict w/ supervisor ..37*** -.39*** .36*** .27*** .88*** 7. Conflict w/ coworker .39*** -.28*** .27*** .20** .79*** .40*** 8. Transactional Leadership .25*** -.17* -.12 .30*** .40*** .36*** .31*** 9. Transformational Leadership -.30*** .49*** .41*** -.08 .34*** .38*** -.17* .15* 10. Co-worker transactional .17 -.18* -.07 .07 .19* .15 .16 .46*** .16 11. Co-worker transformational -.25** .32*** .27** -.04 -.30** -.27** -.20* .00 .49*** -.03 p<.05, ** p<.01, *** p<.001
33 Table 6. Agreement between sources. Variable Correlation between self and co-worker reports Overall Transformational Leadership .49*** Intellectual Stimulation .39*** Individualized Consideration .34*** Behavioral Idealized Influence .45*** Attributed Idealized Influence .57*** Inspirational Motivation .51*** Overall Transactional Leadership .46*** Contingent Reward .47*** Passive Management by Exception .51*** Active Management by Exception .27** Laissez-faire .45*** p<.05, ** p<.01, *** p<.001 Hypothesis 1 predicted that high levels of stressors would relate to high levels of negative emotions and high levels of CWB. Re plicating prior work, c onstraints, conflict, and procedural justice showed significant corre lations with both negative emotions and overall CWB. Autonomy and distributive ju stice showed smaller, yet significant, correlations with CWB and emotions. Exam ining the part of the hypothesis unique to this study, participant data show that participant-reported tr ansformational leadership is related negatively and significantly with CW B (r = -.16) while participant-reported transactional leadership is si gnificantly positively related with CWB (r = .24). Similarly, co-worker-reported transactional leader beha viors were significan tly positively related with CWB (r = .19), but co-worker reports of transformational leader behaviors were not significantly related to CWB (r = -.13), alt hough the correlation was in the expected direction. It is important to note here that co-wor ker reports were based on a sample size of 116 while participant reports are based on th e full sample of 172. Finally, all stressors
34 were significantly related to negative emotions and in the expected direction. Therefore, hypothesis 1 receives almost full support. Table 7. Analysis of mediati ng role of negative emotion Step 1 beta weight Step 2 beta weight Step 1 R2 Step 2 R2 R2 delta F Transactional .36** .08 .04** .27*** .23 31.73 Negative emotion .88*** Laissez-faire 1.20*** .45 .09*** .28*** .19 33.15 Negative emotion .82*** Passive management by exception .93*** .20 .07*** .27*** .20 31.85 Negative emotion .86*** Transformational -.15* -.05 .04* .28*** .24 32.08 Negative emotion .87*** Attributed Idealized Influence -.80** -.35 .06** .28*** .22 33.10 Negative emotion .85*** Individualized Consideration -.66** -.16 .04** .27*** .23 31.77
35 Table 7. (Continued) Negative emotion .88*** Constraints .54*** .08 .09*** .27*** .18 31.67 Negative emotion .86*** Conflict 2.13*** 1.42*** .26*** .36*** .10 47.37 Negative emotion .63*** Procedural Justice -.26*** .00 .09*** .27*** .18 31.46 Negative emotion .89*** Distributive Justice -.38* .19 .03* .28*** .25 32.34 Negative emotion .97*** Autonomy .33* .17 .03* .28*** .25 32.60 Negative emotion .88*** p<.05, ** p<.01, *** p<.001 To test hypothesis 2, the mediation hypothesis, the procedure recommended by Baron and Kenny (1986) was used. This procedur e entails investigating three regression models, regression of CWB on the stressor, th e proposed mediator (negative emotion) on the stressor, and the CWB on the stressor and negative emotion together. If the beta of
36 the stressor variable is signifi cant in the first model, but nonsignificant or substantially reduced in the combined model, that is a pa ttern consistent with mediation. Results are presented in Table 7. All stressor variable s except for conflict displayed full mediation by negative emotion. Hypothesis 2 was we ll-supported in al l but one case. Hypothesis 3 stated that leadership style would more strongly relate to organizational forms of CWB than to persona l forms of CWB. To test hypothesis 3, relationships among transactional leadership and CWBO and CWBP were examined, as were relationships between transformational leadership and CWBO and CWBP. In table 3, it is noteworthy that transactional leadersh ip relates positively a nd significantly with both CWBO (r = .19) and CWBP (r = .20) while transformational leadership is negatively correlated with only CWBO (r = -.21). In addition, all facets of transformational leadership except for inspirational motivation were more strongly inversely related with CWBO than with CWBP. Similar to the transformational leadership facets, the contingent reward subscale of the transact ional leadership scale showed a significantly stronger negative correlation w ith CWBO (r = -.18) than CWBP (r = -.02). Comparisons between the correlations were calculated by using Hotelli ngs t-test for dependent correlations. Results for these analyses can be found in table 8. The relationships of CWBO and CWBP with co-worke r reported leadership style we re not significant, but two facets of co-worker reported transactional leadership, passive management by exception and laissez-faire leadership, were significan tly positively correlat ed with CWBO (r = .22 and r = .21 respectively) but not CWBP (although the two correlations were not
37 significantly different from each other). Considering these findings, hypothesis 3 is partially supported. Table 8. Results for Hotelling-Williams t-tests for Dependent Correlations r CWBO r CWBP t value Transformational Leadership -.21** -.05 -2.42** Individualized Consideration -.28*** -.06 -3.41*** Attributed Idealized Influence -.28*** -.14 -2.15* Intellectual Stimulation -.20** -.03 -2.57** Behavioral Idealized Influence -.18* -.04 -2.10* Inspirational Motivation -.16* -.08 -1.19 Contingent Reward -.18* -.02 -2.41** Laissez-faire .30*** .23** 1.08 p<.05, ** p<.01, *** p<.001 Hypothesis 4 stated that procedural justice would moderate the relationship between leadership style and CWB. Hypothe sis 4 was tested using moderated regression analysis as done by Fox et al. (2001). As expected, transformational leadership was moderated by procedural just ice in its relationship with CWB. However, results for transactional leadership, procedural jus tice, and CWB did not show a significant moderator term. Results of th e moderator analysis can be found in table 9. A graph of the interaction is shown in figure 3.
38 Table 9. Results for moderated regression analys is with procedural justice as moderator. CWB Step Independent variable Unstandardized bs Total R2 Change in R2 1 Transformational Leadership .34 .04* .04* 2 Procedural justice .06 .09*** .05** 3 Transformational x Procedural -.01 .12*** .03* 1 Transactional Leadership .72 .04** .04** 2 Procedural justice -.02 .11*** .07** 3 Transactional x Procedural -.01 .12 .01 p<.05, ** p<.01, *** p<.001 Figure 3: Leadership / Justice interaction0 10 20 30 40 50 60 70 80 90 LoHi TransformationalCWB Lo Procedural Hi Procedural The significant (p = .027) interaction be tween transformational leader style and procedural justice (as pictured in figure 3), is exactly the oppos ite of the predicted interaction. Given the data, tr ansformational leader style matters less when procedural justice is low than when procedural justice is high. Moreover, this is a strong, crossover
39 interaction, so more can be stated rega rding how the effects of transformational leadership on CWB are bounded by procedural ju stice. At low levels of procedural justice, CWB rises slightly as transformational leadership increases. Conversely, at high levels of procedural justice, CWB decreas es sharply as transformational leadership increases. Essentially, the form of the re lationship between transformational CWB and leadership is very different at high and low le vels of procedural ju stice. Thus, hypothesis 4 receives no support in that there is a si gnificant interaction, but the form of the relationship is different than what was hypothesized.
40 Chapter Four Discussion The purpose of this study was to investigat e the impact of different types of leader behaviors at work on subordina te readiness to commit CWBs using an emotion centered model. Research that has examined the e ffects of transformational and transactional leadership found positive effects on motiva tion for transformational leadership and negative effects on commitment to quality and organizational productivity for transactional leadership (Masi & Cooke, 2000). Findings from many studies have demonstrated the ill effects of transactional leadership and positive outcomes associated with transformational leader ship (e.g. McColl & Anderson, 2002; Sparks et al., 2001; Mackenzie et al., 2001). Other research has indicated that employees direct their CWBs toward the source of the problem (Bruk, 2003). Because of the dearth of studies directly examining leadership styles effects on diffe rent counterproductive workplace behaviors, this study assessed leader style using tw o data sources and also assessed both organizational and personal CWBs. Given th e importance of emotion demonstrated in much prior work (e.g. Fox & Spector, 2001; Goh, Bruursema, Fox, & Spector, 2003), mediation tests were run. A nd given the role of fairness in how a leader is perceived (DeCremer & van Knippenberg, 2002), moderato r tests for justice were also conducted. Generally, results of the current study provide support for the replication and extension of the job stress/emotion/CWB m odel (Spector et al., 1998). Specifically, it was found that transactional and transformationa l leadership relate di rectly and inversely, respectively, with CWB (Hypot hesis 1); that negative em otions fully mediate the
41 relationship between leadership style a nd CWB (Hypothesis 2); that CWBO relates inversely with transformational leadership while CWBP does not (H ypothesis 3); and that transformational leadership is moderated by procedural justice in its effect on CWB (Hypothesis 4). Hypothesis 1: Relationships among the stressors, leadership style, negative emotions, and CWB. As predicted by hypothesis 1, participant repo rts of transactiona l leadership style were significantly and directly associated with negative emotions and CWB. As expected, and in accord with prior work, cons traints and conflict showed this pattern of relationships as well. Meanwhile, transf ormational leadership and distributive and procedural justice showed inverse relationshi ps with CWB and negative emotions, also as hypothesized. The co-worker reports of tran sactional leadership style were also significantly and directly related to particip ant reports of negative emotions and CWB, albeit with a smaller correlation (r = 19 for both). However, co-worker reported transformational leadership did not rise to th e level of significance in its relationships with negative emotions or CWB, although corre lations were in the expected direction (r = -.10 and -.13 respectively). Although the relationships with overall lead ership style are interesting, we are able to obtain a more detailed view of the findings when we examine the relationships with the facets of leadership and CWB. Fi rst, for both participant and co-worker sources, the transactional facets of passive manage ment by exception and laissez-faire related significantly and positively w ith participant reported nega tive emotions and CWB.
42 Passive management by exception deals mainly with the leader failing to take action in the early stages of a problem. Some exampl e items are, Waits for things to go wrong before taking action, and, Fails to interf ere until problems become serious. Laissezfaire leadership deals mainly with being unavailable when direction or assistance is needed. Some example items are, Is abse nt when needed, and, Delays responding to urgent questions. Perhaps the higher report ed incidence of CWB a ssociated with these two facets is due simply to reduced supervisor monitoring and subordinates perceptions about the reduced likelihood of being caught committing CWBs. However, passive management by exception and laissez-faire l eadership are strongly and positively related to negative emotions (r = .41 and .40 respectiv ely) and moderately and negatively related to positive emotions (r = -.23 and -.22 resp ectively); this would tend to suggest that subordinates view their independence negative ly and feel badly about this type of supervisor treatment. Active management by exception, on the other hand, shows no correlation with CWB. This facet deals mainly with supervisor scrutiny of subordinate mistakes. Some example items are, Focuses attention on irregularities, mistakes, exceptions, and deviations from standards, and, K eeps track of all mistakes. The lack of a significant correlation with CWB could be because subordi nates feel like they would have a very high likelihood of being caught but of note is the signifi cant but relatively small correlation with negative emotions (r = .18) and the lack of a significant negative correlation with positive emo tions, although in the expected direction (r = -.09). The correlation with active management by excepti on and negative emotion is significantly
43 smaller than both the laiss ez faire (t = 2.39, p<.01) and passive management by exception (t = 2.52, p<.01) correlations w ith negative emotions. Theref ore, subordinates feel worse about an absent or uninvolved supervisor th an they do about a supe rvisor who points out failures. This would seem to indicate that subordinates find some attention, even negative attention, be tter than no attention. When we li nk all this back to CWB, it makes sense that participants who report passive l eader behaviors also report more CWB than participants who report the negative leader monitoring behaviors; those who have passive leaders have more negative feelings about th eir jobs and therefore more reason to engage in CWBs. In examining the transformational leader facets, we find that only individualized consideration (r = -.20 ) and attributed ideal ized influence (r = -.24) show inverse relationships with overall CWB. Individualized consid eration is generally about individually-focused, mentori ng-type behaviors. Some sa mple items are, Spends time teaching and coaching, and, Considers me as having different needs, abilities, and aspirations from others. This variable has the highest correlation (r = .38) with positive emotions of all variab les included in the study. This fi nding, in concert with the findings for laissez-faire and passive management by exception, again seems to highlight the importance of subordinates feeling attended to rather than ignored. Throughout this study, subordinates who report more attention of any kind (positive or negative) from supervisors also report less ne gative emotion and less CWB. A fruitful area for further research would be to look more directly at leader neglecting or l eader ignoring behaviors with subordinate stress, emotions, jo b performance, and counterproductivity.
44 With respect to attributed idealized in fluence, all relationships were again significant in the expected directions (r = .27 for negative emotion, r = -.30 for positive emotion). Attributed idealized influence deals with the overa ll feeling ones supervisor projects. Some example items are, Displ ays a sense of power and confidence, and, Acts in ways that builds my respect, and Ins tills pride in me for being associated with him or her. This is different from behavi oral idealized influence, which did not show significant relationships with CW B or negative emotions, in that it deals not with what a supervisor says, but with the feeling one gets from how the supervisor behaves. This is similar to Lowe et als (1996) definition of charisma: the instilling of pride, faith, and respect, a gift for seeing what is important, and the ability to tran smit a sense of mission. It seems that lacking this subtle touch, a so rt of leading by example rather than by lesson, is perceived negatively by subordinates, and they tend to engage in more CWBs. This finding, that subordinates are less likely to commit undermining or retaliatory acts at work when they respect and admire their s upervisor, is attractive since it suggests that being a respected leader has implications fo r important organizational outcomes. More research should examine what specifically about a leader deems him or her respectworthy and what other positive outcomes this could be associated with on the organizational and the individual subordinate level. Little support was garnered for the relati onship between intellect ual stimulation or behavioral idealized influence and CWB. Ho wever, both facets were strongly related to positive emotions and significantly negatively related with CWBO (r = -.20 and r = -.18, respectively). This makes intuitive sense si nce feeling stimulated by your job may make
45 you happy and less willing to harm your organi zation, but it will proba bly matter little in terms of how you treat co-workers. Hypothesis 2: The mediating role of negative emotions It was reasoned that negative emotions were the process by which stressors and leadership style exert their effects on count erproductive behaviors. Therefore, full mediation by negative emotions of all indepe ndent-dependent variab le relationships was expected. The findings of this study provide almost full support for this hypothesis. As evidenced in table 5, all inde pendent variables except for c onflict were fully mediated by negative emotions in their e ffects on CWB. Concurrent with prior work (Bruk, 2003), conflict showed partial medi ation by negative emotion. A key takeaway message here is that all four facets of transactional and transformational leadership that displayed significant relationships with CWB were included in the mediation analysis (laisse z-faire, passive management by exception, attributed idealized influence, and individualized considerati on), and all four showed full mediation by negative emotion. This yields fu rther support for the idea that these facets of transactional leader ship are upsetting to subordinates, and that they are engaging in CWB because of these bad feelings, not simp ly because they can get away with them when no one is watching. Taken together wi th the transformational facet findings, these results would indicate that subordinates do not so much need mon itoring and control to stop engaging in CWBs, but they need emoti onal management from a leader they respect who gives them specific, individualized at tention and who is present when needed.
46 Hypothesis 3: Differential relationships between leadership style and CWBO and leadership style and CWBP Hotellings (1940) t test for dependent corr elations was used to assess differences between correlations with CWBO and CWBP and the leadership facets. No support was found for the predicted stronger relationship be tween overall transac tional leadership and CWBO (r = .19 versus r = .20 for CWBP). However, as evidenced in table 8, transformational leadership did show a mo re significantly negative relationship with CWBO than CWBP. Specifically, all transformational facets except for inspirational motivation displayed significantly stronger corr elations with CWBO than with CWBP. Interestingly, however, neither procedural nor distributive justice showed this same pattern, contradicting prior work with these variables (Fox et al., 2001). Also, one facet of transactional lead ership, namely, contingent rewar d, showed a significantly stronger relationship with CWBO than with CWBP. R easons for these results are congruent with reasons put forth for the hypotheses, employees are less likely to harm the organization when they feel positive emotions, lack ne gative emotions, and feel good about and admire their supervisor. Also, as Bruk (2003) and Penney (2003) found, employees target their CWB responses at the perc eived sources of their bad feelings. A puzzling finding is the significant correlation between laissez-faire leadership and CWBP and passive management by exception and CWBP. None of the other facets of transactional or transformational leadersh ip bore any relations hip with CWBP. One explanation for this could be that without a strong leader presence, subordinates experience fighting with co-workers and engage in power struggles. A closer look at the
47 individual CWB items that correlated signi ficantly with laissez-faire leadership and passive management by exception revealed that 5 CWB items were related to both facets of transactional leadership. Those 5 are: insulted someone about their job performance, made fun of someones personal life, refused to help someone at work, played a mean prank to embarrass someone at work, and destroyed property belonging to someone at work. It could be that there is just a negative, hostile envi ronment, perhaps similar to low morale, when subordinates feel abandoned by their supervisor. Ne gative emotions, as discussed previously are quite high when th ese two facets are high so it could be that these feelings spill over into other domains. For instance, for the conflict variable, where correlations with negative emotion are e qually high (r = .34 to r = .47), conflict with supervisors relates strongly to both CWBO a nd CWBP as does conflic t with co-workers. It could be that after a cert ain point, negative emotions are expressed to others in CWB regardless of their source. This same re lationship is present for constraints and procedural justice, two othe r variables with hi gh correlations with negative emotions. Hypothesis 4: Moderation by justice of the leadership-CWB relationship It was expected that justice would determine when the leadership-CWB relationship was strong. Specifi cally, it was predicted that le ader style would matter less when procedural justice was high than when procedural justice was low. Support was found for justice as a moderator in the rela tionship between transf ormational leadership and overall CWB, but no support was found for a moderating role of justice in the relationship between transactional leadership and CWB.
48 The finding that the interaction did not occur in the expected fashion is not completely surprising as the rationale for th e hypothesis was derived from very little prior work. However, this finding is quite interesting for two reasons. First, it is very difficult to find a significant moderator effect with such a relatively small sample (N = 172). These tests have notoriously low power as multicollinearity is generally a problem. Although it depends on the reliability of the m easurement instrument for the variables, a sample size of over 220 is generally recommen ded to detect modera tor effects (Aguinis, Boik, & Pierce, 2001). Given this, and given the strength of the interaction, we can expect that this finding is relatively robust. The graphed interaction is also interes ting because it shows that when a persons workplace is perceived as procedurally unfai r, an exceptionally transformational leader may slightly increase the o ccurrence of CWB. Though the re ally remarkable finding is that when procedural justice is high, and tr ansformational leadership is low, CWB is highest than in any other c ondition. And conversely, when pr ocedural justice is high and transformational leadership is high, the occu rrence of CWB is lowest than in any other condition. Therefore, the behavi ors of the leader and feelings projected by the leader are most important when organizational proce dures are exceptionally fair and just. More work is needed to determine the th eoretical significance of the procedural justice moderator on the leadership style/CW B or leadership styl e and other dependent organizational variables.
49 Convergence between Self and Coworker Reports Correlations between the two leadership reports showed mo derate to high agreement between sources (see table 6). Active management by exception revealed the lowest between-source correlation (r = .27); this could be because s upervisors dont keep track of all subordinates mistakes equally. Perhaps a low-performing subordinate receives more of this type of negative atte ntion than a better pe rforming subordinate. Therefore, this may reflect actual differences in leader behavior. Oddly, the highest intersource agreement coefficient was for attributed idealized influence (r = .57). This factor, measuring a sort of esoteric respectability f actor, was easiest for two co-workers to agree on. Passive management by exception and inspirational motivation were the second highest correlations at (r = .51). It seems that co-workers came to reasonable agreement regarding how indecisive and unhelpful the l eader is and how articulately the leader spells out a vision for the organization. Limitations Limitations of this study include the single source for CWB data. Although respondents would be the experts on which CWBs they engage in or do not engage in, there has been concern that they may be mo tivated to lie or misremember for social desirability purposes. Previous research tends to find good inter-source agreement for co-worker or supervisor reports of CWB (e .g. Goh et al., 2003; Bruk, 2003), and there is the question of when the sources disagree, who is more accurate? While respondents may be motivated to deflate their estimates of their own CWB, co-workers may not
50 notice all of the respondents CW Bs and/or may be motivated to inflate or deflate their estimates of the respondents CWB dependi ng on their opinion of the respondent. Further limitations are the cross-sectiona l design of the study. This could be problematic since respondents in dicated what they thought of their supervisors, how they felt about work, and how much CWB they enga ged in all at the same point in time. Reactivity could have occurred if respondents guessed that bad feelings should go with conflict, constraints, and unfai rness at work. Or, respondents could have been primed to feel badly (or good as the case may be) after thinking of all the th ings wrong with their jobs. A longitudinal design for CWB could o ffer more support for the findings in this and other CWB studies. Part of the concern about single source, single time data was addressed by having a co-worker fill out the lead ership questionnaire. Findings with this additional measure were somewhat supportive of findings with only the participant report data. A final limitation of this study is that th e co-worker questionnaires were based on a sample size of 116 rather than the full part icipant sample of 172. This is a very large decrement in sample size from which to c onduct analyses and many of the correlations could have been attenuated because of this. It was not possible to tell if the small sample size was responsible for the lack of findings with co-worker reported transformational leadership or if it was just differential perspectives between sources, but source agreement correlations would tend to implemen t the reduced sample size in the problem.
51 Conclusions The major findings of this study suggest that good leaders ar e those who lead by example, pay attention to individual people, and who respond to problems quickly, decisively, and ably. This is an over-si mplification, but worth remembering, and also worth further study using different dependent variables (e.g. orga nizational productivity, customer satisfaction, subordina te reactions) and a different theoretical perspective than the work stress angle presented here. In short, this study provided good support fo r the inclusion of transactional and transformational leadership styl e as a stressor in the job-st ress/emotion/CWB model. The strongest support was found for including la issez faire, passive management by exception, attributed idealized in fluence, and individualized c onsideration as variables in the model. It also provided a point from which to start in examining which leader behaviors are important to how an employee f eels about his or her j ob and to his or her willingness to participate in CWB. More re search should examine other kinds of leader behaviors with respect to these outcomes.
52 References Aguinis, H.; Boik, R.J.; & Pierce, C.A. (2001) A generalized solution for approximating the power to detect effects of categorical moderator variables using multiple regression. Organizational Research Methods 4, 291-323. Allen, V.L.; Greenberger, D.B. (1980). Destruct ion and perceived control. In A. Baum & J.E. Singer (Eds.), Applications of personal control Hillsdale, NJ: Erlbaum. Avolio, B.J. (1999). Full leadership development. Thousand Oaks, CA: Sage. Baron, R.M. & Kenny, D.A. (1986). The mode rator-mediator variab le distinction in social psychological research: Conceptual, strategic, and statis tical considerations. Journal of Personality and Social Psychology 51, 1173-1182. Bass, B.M. (1985). Leadership and performance beyond expectations. New York: Free Press. Bass, B.M. (1996). A new paradigm of leadership: A survey of theory and research. New York: Free Press. Bauer, T.N. & Green, S.G. (1996). Deve lopment of Leader-Member Exchange: A Longitudinal Test. Academy of Management Journal 39, 1538-1567. Bruk, V. (2003). Title? Unpublished masters thesis. Bycio, P., Hackett, R.D., & Allen, J.S. ( 1995). Further assessments of Basss (1985) conceptualization of transactional and transformationa l leadership. Journal of Applied Psychology 80, 468-478. Chen, P.Y. & Spector, P.E. (1992). Relations hips of work stressors with aggression, withdrawal, theft, and substan ce abuse: An exploratory study. Journal of
53 Occupational and Organizational Psychology 65, 177-184. Cohen-Charash, Yochi, & Spector, P.E. (2001). The role of justi ce in organizations: A meta-analysis. Organizational Behavior & Human Decision Processes 86, 278321. DeCremer, D. & van Knippenberg, D. (2002) How do leaders promote cooperation? The effects of charisma and procedural fairness. Journal of Applied Psychology 87, 858-866. Dionne, S.D.; Yammarino, F.J.; Atwater, L.E ., & James, L.R. (2002). Neutralizing substitutes for leadership theory: Lead ership effects and common-source bias. Journal of Applied Psychology 87(3), 454-464. Fox, S., Spector, P.E., & Miles, D. (2001). Counterproductive work behavior in response to job stressors and organizationa l justice: Some mediator and moderator tests for autonomy and emotions. Journal of Vocational Behavior 59, 1-19. Fox, S., Spector, P.E., & VanKatwyk, P. (1997). Objectivity in the a ssessment of control at work. Paper presented at the So ciety of Industrial and Organizational Psychology, St. Louis. Frone, M.R. (2000). Interpersonal conflict at work and psychological outcomes: Testing a model among young workers. Journal of Occupational Health Psychology 5, 246-255. George, J. M. (1995). Leader positive mood and group performance: The case of customer service. Journal of Applied Social Psychology 25, 778-794. Gerstner, C.R. & Day, D.V. (1997). Meta-ana lytic review of lead er-member exchange
54 theory: Correlates and construct issues. Journal of Applied Psychology 82, 827 844. Giesburg, J. (2001). The role of communica tion in preventing workplace sabotage. Journal of Applied Social Psychology 31, 2439-2461. Goh, A.P.S., Bruursema, K., Fox, S., & Spector P.E. (2003). Comparisons of self and co-worker reports of count erproductive work behavior. Paper presented at the meeting of the Society of Industrial a nd Organizational Psychology, Orlando, FL. Goodwin, V.L., Wofford, J.C., & Whittington, J. L. (2001). A theoretical and empirical extension of the transformati onal leadership construct. Journal of Organizational Behavior 22, 759-774. Graen, G. B. & Cashman, J. (1975). A role -making model of leadership in formal organizations: A development approach. In J.G. Hunt & L.L. Larson (Eds.), Leadership Frontiers (p. 143-165). Kent, OH: Kent State University. Hater J.J. & Bass, B.M. (1988). Superiors evaluations and subordina tes perceptions of transformati onal and transactional leadership. Journal of Applied Psychology 73, 695-702. Hotelling, H. (1940). The selection of vari ates for use in prediction, with some comments on the general problem of nuisance parameters. Annals of Mathematical Statistics 11, 271-283. Howell, J.M. & Avolio, B.J. (1993). Tr ansformational leader ship, transactional leadership, locus of control, and suppor t for innovation: Key predictors of
55 consolidated businessunit performance. Journal of Applied Psychology 78, 891902. Lowe, K.B., Kroeck, K.G., & Sivasubramaniam, N. (1996). Effectiveness correlates of transformational and transactional leadersh ip: A meta-analytic review of the MLQ literature. Leadership Quarterly 7, 385-425. MacKenzie, S.B., Podsakoff, P.M., & Ri ch, G.A. (2001). Transformational and Transactional Leadership and Salesperson Performance. Journal of the Academy of Marketing Science 29, 115-134. Marrs, Mary E.M. (2000). Antecedents a nd outcomes of verbal aggression in the workplace. Dissertation Abstracts International Section A: Humanities and Social Sciences 61, 681. Masi, R.J. & Cooke, R.A. (2000). Effects of transformational leadership on subordinate motivation, empowering norms, a nd organizational productivity. International Journal of Organizational Analysis 8, 16-47. McColl-Kennedy, J.R. & Anderson, R.D. (2002). Impact of leadership style and emotions on subordinate performance. Leadership Quarterly 13, 545-559. Moorman, R.H. (1991). Relationship between organizational justice and organizational citizenship behaviors: Do fairness per ceptions influence employee citizenship? Journal of Applied Psychology 76, 845-855. Muchinsky, P.M. (2000). Psychology applied to work. Belmont, CA: Wadsworth/Thomson Learning. Occupational Safety and Health Administra tion (1996). Workplace violence prevention:
56 Health care and social service work ers. Retrieved April 4, 2003 from http://www.OSHA.gov Pearson, C.M., Andersson, L.M., & Wegner, J.W. (2001). When workers flout convention: A study of workplace incivility. Human Relations 54, 1387-1419. Penney, L.M. (2003). Peters, L.H. & OConnor, E.J. (1980). Wor kplace violence and wo rkplace aggression: Evidence concerning specific forms, poten tial causes, and preferred targets. Journal of Management 24, 391-419. Price, J.L. & Mueller, C.W. (1986). Handbook of Organizational Measurement. Marshfield, MA: Pittman. Podsakoff, P.M., MacKenzie, S.B., & Bomm er, W.H. (1996). Meta-analysis of the relationships between Kerr and Jermiers substitutes for leadership and employee job attitudes, role perceptions, and performance. Journal of Applied Psychology 81, 380-399. Rajnandini, P., Schrisheim, C.A., & Williams E.S. (1999). Fairness perceptions and trust as mediators for transformatio nal and transactional leadership. Journal of Management 25, 897-933. Robinson, S.L. & Bennett, R.J. (1995). A t ypology of deviant wor kplace behaviors: A multidimensional scaling study. Academy of Management Journal 38, 555-572. Sparks, J.R. & Schenk, J.A. (2001). Expl aining the effects of transformational leadership: An investigation of the effects of higher-order motives in multi-level marketing organizations. Journal of Organizational Behavior 22, 849-869.
57 Spector, P.E. (1998). A control theory of the job stress process. In C.L. Cooper (Ed.), Theories of organizational stress (pp. 153-169). Oxford, UK: Oxford University Press. Spcector, P.E. (2002). Employee c ontrol and occupational stress. Current Directions in Psychological Science 11(4), 133-136. Spector, P.E. & Fox, S. (2002). An emotioncentered model of voluntary work behavior: Some parallels between counterprod uctive work behavior (CWB) and organizational citizenship behavior (OCB). Human Resources Management Review 12, 269-292. Spector, P.E. & Jex, S. M. (1998). Developm ent of four self-report measures of job stressors and strain: Inte rpersonal conflict at work scale, organizational constraints scale, quantitative workload inventory, and physical symptoms inventory. Journal of Occupati onal Health Psychology 3, 356-367. Storms, P.L. & Spector, P.E. (1987). Relatio nships of organizational frustration with reported behavioral reactions: The mode rating effects of perceived control. Journal of Occupational Psychology 60, 227-234. Tepper, B.J. (2000). Consequences of abusive supervision. Academy of Management Journal 43, 178-190. Tierney, P., Bauer, T.N., & Potter, R.E. (2002). Extra role be haviors among Mexican employees: The impact of LMX, group acceptance, and job attitudes. International Journal of Selection and Assessment 10, 292-303. Townsend, J., Phillips, J.S., & Elkins, T.J. (2000). Employee Retaliation: The neglected
58 consequence of poor leader-member exchange relations. Journal of Occupational Health Psychology 5, 457-463. Turner, N.; Barling, J.; Epitropaki, O; Butcher, V; Milner, C. (2002) Transformational leadership and moral reasoning. Journal of Applied Psychology 87, 304-311. U.S. Chamber of Commerce (2002). Employee Benefits Study. Washington, D.C. American Internationa l Group, Inc. (AIG). U.S. Department of Labor, Bur eau of Labor Statistics (1996). Census of fatal occupational injuries Washington, DC. VanKatwyk, Fox, Spector, & Kelloway (2000). Using the Job-Related Affective Well Being Scale (JAWS) to investigate aff ective responses to work stressors. Journal of Occupational Health Psychology 5, 219-230. Williams, M.L., Podsakoff, P.M., & Huber, V. (1992). Effects of group-level and individual-level variatio n in leader behaviors on subordinate attitudes and performance. Journal of Occupational and Organizational Psychology 65, 115129. Wofford, J.C., Whittington, J.L. & Goodwin, V. L. (2001). Follower motive patterns as situational moderators for transformational leadership effectiveness. Journal of Managerial Issues 13, 196-211. Yukl, G. (1999). An evaluation of concep tual weaknesses in transformational and transactional leadership theories. Leadership Quarterly 10, 285-305.
59 The following questions ask about situations and conditions in your workplace. For each statement indicate how often it occurs on your present job. How often do you find it difficult or impossible to do your job because of ... ? 1=Never 2=Once or twice 3=Once or twice per month 4=Once or twice per week 5=Every day Poor equipment or supplies. 1 2 3 4 5 1. Organizational rules and procedures. 1 2 3 4 5 2. Other employees. 1 2 3 4 5 3. Your supervisor. 1 2 3 4 5 4. Lack of equipment or supplies. 1 2 3 4 5 5. Inadequate training. 1 2 3 4 5 6. Interruptions by other people. 1 2 3 4 5 7. Lack of necessary information about what to do or how to do it. 1 2 3 4 5 8. Conflicting job demands. 1 2 3 4 5 9. Inadequate help from others. 1 2 3 4 5 10. Incorrect instructions. 1 2 3 4 5 The purpose of this section is to examine your perceptions about workplace equity In answering the following questions, think a bout the day-to-day d ecisions made about worker responsibilities, schedules, rewards, and general treatment on your present job For each statement, indicate your AGREEMENT or DISAGREEMENT: 1 = Strongly disagree 5 = Slightly agree 2 = Disagree 6 = Agree 3 = Slightly disagree 7 = Strongly agree 4 = Neither disagree or agree When decisions about other employees in general or you in particular are made in this company ... 11. requests for clarification and additional information are allowed. 1 2 3 4 5 6 7 12. you are treated with respect and dignity. 1 2 3 4 5 6 7
60 13. you are dealt with in a truthful manner. 1 2 3 4 5 6 7 14. all the sides affected by the decisions are represented. 1 2 3 4 5 6 7 15. the decisions are applied with consistency to the parties affected. 1 2 3 4 5 6 7 16. you are offered adequate justification for the decisions. 1 2 3 4 5 6 7 17. accurate information upon which the decisions are based is collected. 1 2 3 4 5 6 7 18. complete information upon which the decisions are based is collected. 1 2 3 4 5 6 7 19. opportunities are provided to appeal or challenge the decisions. 1 2 3 4 5 6 7 20. you are treated with kindness and consideration. 1 2 3 4 5 6 7 21. you are shown concern for your rights as an employee. 1 2 3 4 5 6 7 22. you are helped to understand the reasons for the decision. 1 2 3 4 5 6 7 For the next set of questions, please use the following choices: 1 = Very unfairly 2 = Unfairly 3 = Undecided 4 = Fairly 5 = Very fairly To what extent are you fairly rewarded ... 23. considering the responsibilities that you have. 1 2 3 4 5 24. taking into account the amount of education and training you have had. 1 2 3 4 5 25. in view of the amount of experience that you have. 1 2 3 4 5 26. for the amount of effort that you put forth. 1 2 3 4 5 27. for the work that you have done well. 1 2 3 4 5 28. for the stresses and strains of your job. 1 2 3 4 5
61 In your present job, how often do you have to ask permission 1 = Never 4 = Quite often 2 = Rarely 5 = Extremely often or always 3 = Sometimes 29. to take a rest break? 1 2 3 4 5 30. to take a lunch/meal break? 1 2 3 4 5 31. to leave early for the day? 1 2 3 4 5 32. to change the hours you work? 1 2 3 4 5 33. to leave your office or work station? 1 2 3 4 5 34. to come late to work? 1 2 3 4 5 35. to take time off? 1 2 3 4 5 Below are a number of statements that descri be different emotions that a job can make a person feel. Please indicate how often any part of your present job (e.g., the work, co-workers, supervisor, clients, pay) has made you feel the listed emotion, by circling the appropriate response, using the following choices. 1=Never 2=Once or twice 3=Once or twice per month 4=Once or twice per week 5=Every day 36. My job made me feel angry. 1 2 3 4 5 37. My job made me feel anxious. 1 2 3 4 5 38. My job made me feel at ease. 1 2 3 4 5 39. My job made me feel bored. 1 2 3 4 5 40. My job made me feel calm. 1 2 3 4 5 41. My job made me feel content. 1 2 3 4 5 42. My job made me feel de pressed. 1 2 3 4 5 43. My job made me feel discouraged. 1 2 3 4 5 44. My job made me feel di sgusted. 1 2 3 4 5 45. My job made me feel ecstatic. 1 2 3 4 5 46. My job made me feel energetic. 1 2 3 4 5 47. My job made me feel enthusiastic. 1 2 3 4 5
62 48. My job made me feel excited. 1 2 3 4 5 49. My job made me feel fa tigued. 1 2 3 4 5 50. My job made me feel frightened. 1 2 3 4 5 51. My job made me feel furious. 1 2 3 4 5 52. My job made me feel gloomy. 1 2 3 4 5 53. My job made me feel inspired. 1 2 3 4 5 54. My job made me feel re laxed. 1 2 3 4 5 55. My job made me feel satisfied. 1 2 3 4 5 Modified Interpersonal Conflict at Work Scales for Employees The following questions ask about your interpersonal relationships in your workplace. Please mark the number for each que stion that best indicates how often the following events occur in your present job with your superv isor or with coworkers, respectively. 1 = Never 2 = Once or Twice 3 = Once or Twice a Month 4 = Once or Twice a Week 5 = Every Day 1. How often do you get into arguments with your supervisor ? 1 2 3 4 5 2. How often does your supervisor yell at you at work? 1 2 3 4 5 3. How often is your supervisor rude to you at work? 1 2 3 4 5 4. How often does your supervisor do nasty things to you at work? 1 2 3 4 5 1. How often do you get into arguments 1 2 3 4 5
63 with your coworkers ? 2. 2. How often do your coworkers yell at you at work? 1 2 3 4 5 3. How often are your coworkers rude to you at work? 1 2 3 4 5 4. How often do your coworkers do nasty things to you at work? 1 2 3 4 5 How often have you done each of the following things on your present job? 1=Never 2=Once or twice 3=Once or twice per month 4=Once or twice per week 5=Every day 56. Purposely wasted your employers materials/supplies 1 2 3 4 5 57. Daydreamed rather than did your work 1 2 3 4 5 58. Complained about insignificant things at work 1 2 3 4 5 59. Told people outside the job what a lousy place you work for 1 2 3 4 5 60. Purposely did your work incorrectly 1 2 3 4 5 61. Came to work late without permission 1 2 3 4 5 62. Stayed home from work and said you were sick when you werent 1 2 3 4 5 63. Purposely damaged a piece of equipment or property 1 2 3 4 5 64. Purposely dirtied or littered your place of work 1 2 3 4 5 65. Stolen something belonging to your employer 1 2 3 4 5 How often have you done each of the following things on your present job?
64 1=Never 2=Once or twice 3=Once or twice per month 4=Once or twice per week 5=Every day 66. Started or continued a damaging or harmful rumor at work 1 2 3 4 5 67. Been nasty or rude to a client or customer 1 2 3 4 5 68. Purposely worked slowly when things needed to get done 1 2 3 4 5 69. Refused to take on an assignment when asked 1 2 3 4 5 70. Purposely came late to an appointment or meeting 1 2 3 4 5 71. Failed to report a problem so it would get worse 1 2 3 4 5 72. Taken a longer break than you were allowed to take 1 2 3 4 5 73. Purposely failed to follow instructions 1 2 3 4 5 74. Left work earlier than you were allowed to 1 2 3 4 5 75. Insulted someone about their job performance 1 2 3 4 5 76. Made fun of someones personal life 1 2 3 4 5 77. Took supplies or tools home without permission 1 2 3 4 5 78. Tried to look busy while doing nothing 1 2 3 4 5 79. Put in to be paid for more hours than you worked 1 2 3 4 5 80. Took money from your employer without permission 1 2 3 4 5 81. Ignored someone at work 1 2 3 4 5 82. Refused to help someone at work 1 2 3 4 5 83. Withheld needed information from someone at work 1 2 3 4 5
65 84. Purposely interfered with someone at work doing his/her job 1 2 3 4 5 85. Blamed someone at work for error you made 1 2 3 4 5 86. Started an argument with someone at work 1 2 3 4 5 87. Stole something belonging to someone at work 1 2 3 4 5 88. Verbally abused someone at work 1 2 3 4 5 89. Made an obscene gesture (the finger) to someone at work 1 2 3 4 5 90. Threatened someone at work with violence 1 2 3 4 5 91. Threatened someone at work, but not physically 1 2 3 4 5 77. Hid something so someone at work couldnt find it 1 2 3 4 5 78.Did something to make someone at work look bad 1 2 3 4 5 79.Played a mean prank to embarrass someone at work 1 2 3 4 5 80.Destroyed property belonging to someone at work 1 2 3 4 5 81.Looked at someone at works private mail/property without permission 1 2 3 4 5 82.Hit or pushed someone at work 1 2 3 4 5 83.Insulted or made fun of someone at work 1 2 3 4 5 84.Avoided returning a phone call to someone you should at work 1 2 3 4 5
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001681099
007 cr mnu|||uuuuu
008 060103s2004 flu sbm s000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001015
Leadership style and the link with counterproductive work behavior (cwb)
h [electronic resource] :
b an investigation using the job-stress/cwb model /
by Kari Bruursema.
[Tampa, Fla.] :
University of South Florida,
Thesis (M.A.)--University of South Florida, 2004.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 73 pages.
ABSTRACT: Relations among job stressors, leadership style, emotional reactions to work,counterproductive work behavior (CWB), and autonomy were investigated. Participants representing a wide variety of jobs were surveyed. Results indicate that transactional leadership style is related to negative emotions and occurrence of CWB. Relationships between variables were mediated by emotions.
Adviser: Paul Spector.
t USF Electronic Theses and Dissertations.