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 002001085
007 cr mnu|||uuuuu
008 090428s2008 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002563
O'Brien, Kimberly E.
A stressor-strain model of organizational citizenship behavior and counterproductive work behavior
h [electronic resource] /
by Kimberly E. O'Brien.
[Tampa, Fla] :
b University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains 69 pages.
Dissertation (Ph.D.)--University of South Florida, 2008.
Includes bibliographical references.
Text (Electronic dissertation) in PDF format.
ABSTRACT: Prior research has attempted to develop a model of organizational citizenship behaviors (OCB) and counterproductive work behaviors (CWB), but limited testing remains a problem. The purpose of the current study is to examine OCB and CWB from a job stressor-strain approach. The sample consisted of 235 employees throughout the United States and their supervisors. Results of the study suggested OCB and CWB are affected by stressors (including interpersonal conflict, low interactional justice, job demands, and organizational constraints). Additionally, trait emotion and attributional styles affect the amount of stressors perceived. The implications as well as limitations of the study are discussed.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
Advisor: Tammy D. Allen, Ph.D.
t USF Electronic Theses and Dissertations.
A Stressor Strain Model of Organizational Citizenship Behavior and Counterproductive Work Behavior by A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department o f Psychology College of Arts and Sciences University of South Florida Major Professor: Tammy D. Allen, Ph.D. Walter Borman, Ph.D. Chu Hsian g (Daisy) Chang, Ph.D. Paul E. Spector, Ph.D Joseph Vandello, Ph.D. Date of Approval: June 27, 2008 Ke ywords: Contextual performance, conflict, stress, attributions, emotion Copyright
Acknowledgements This research was supported in part by a National Institute for Occupational Safety and Health (NIOSH) training g rant (No. T42 CCT412874) from the Sunshine Education and Research Center at the University of South Florida. The contents are solely the responsibility of the author(s) and do not represent the official views of NIOSH. I would also like to acknowledge my advisor, Dr. Tammy Allen. Her guidance also like to thank Drs. Vandello and Stark, who I also feel contributed to my professional development, and my committee member s who greatly helped me to improve this project.
i Table of Contents List of Tables ii List of Figures iii Abstract iv Chapter One : Introduction 1 OCB and CWB Background 2 Antecedents of OCB and CWB 4 Job Stressors in OCB and CWB 5 The Role of Personality in the Stressor Strain Relationship 1 0 The Role of Job Stressors in OCB and CWB 1 3 Chapter Two: Methods 18 Participants and Procedures 18 Measures 19 Chapter Three: Results 23 Descriptive Statistics 23 Hypothesis Testing 23 Chapter Four: Discussion 27 Limitations 29 Future Research 30 Conclusion 32 References 34 Appendices 51 Appendix A: Questio nnaire Given to Participants 52 Appendix B: Questio nnaire Given to Supervisors 67 About the Author End Page
ii List of Tables Table 1 Intercorrelations among Time 1 variables (lower diagonal) and Time 2 variables (upper diagonal) 4 5 Table 2 Correlations between Time 1 and Time 2 data. 46 Table 3 Variable means, standard deviations, and correlations with supervisor report data 47 Table 4 Results of bootstrapped Sobel tests 48
iii List of Figures Figure 1. Job Stressor Mediated Model of OCB and CWB 49 Figure 2. Job Stressor Mediated Mode l of Target Based OCB and CWB 50
iv A Job Stress Model of Organizational Citizenship Behavior and Counterproductive Work Behavior ABSTRACT Prior research has attempted to develop a model of organizational citizenship beha viors (OCB) and counterproductive work behaviors (CWB), but limited testing remains a problem T he purpose of the current study is to examine OCB and CWB from a job stressor strain approach. The sample consisted of 235 employees throughout the United Stat es and their supervisors Results of the study suggest ed OCB and CWB are affected by stressors (including interpersonal conflict, low interactional justice, job demands, and organizational constraints). Additionally, trait emotion and attributional styles affect the amount of stressors perceived The implications as well as limitat ions of the study are discussed.
1 Chapter One Introduction Although many workplace activities are highly regulated, some employee behaviors allow for more discretion. These mor e discretionary behaviors include organizational citizenship behavior (OCB) and counterproductive work behavior (CWB). OCBs are actions that contribute to the organizational, socia l, and psychological context of the workplace, such as volunteering to accl imate new employees or enhancing the reputation of the organization (Borman & Motowidlo, 1993). For the most part, OCB is thought to benefit the organization (Bolino, Turnley, & Niehoff, 2004). Conversely, CWB refers to intentional acts that are harmful to the organization such as taking unnecessary breaks, stealing, or aggression (Fo x & Spector, 2005). Because these constructs are both considered voluntary work behaviors, researchers have begun to develop models that describe or explain OCB and CWB ( e. g., Kelloway, Loughling, Barling & Nault, 2002; Lee & Allen, 2002; Miles, Borman, Spector, & Fox, 20 02; Spector & Fox, 2002). These models explain OCB and CWB as a function of organizational environment, organizational attitudes, emotion, and personality traits, but none have received unequivocal support. Previous studies have stressed the importance of further developing these models in an attempt to better understand these present study is to extend this research by developing a model of the role of job stressors in OCB and CWB, based
2 on previous empirical and theoretical investigation. Prior investigation of job stress ors in OCB and CWB is limited, but suggests that future research in th e area would be beneficial (e.g. Miles et al., 2002) Specifically, there has been only preliminary testing of the role of job stress in OCB and CWB, and research in this area could be much improved through more rigorous design (e.g. longitudinal testing). The lack of research in the area may be due to research that has suggested t hat employee performance and well being are conflicting organizational goals (Fox & Spector, 2002). However, more recent theory has implicated employee well being in organization al outcomes such as task performance (e.g., Judge, Thoresen, Bono, & Patton, 2001) and counterproductive work behaviors (e.g., Fox, Spector, & Miles, 2001). The current study will propose a model o f OCB and CWB in which job stressors mediate the relations hip between personality variables and these voluntary behaviors. These meditational relationships will be tested using path analysis and bootstrapped Sobel tests. OCB and CWB Background OCB and CWB are discretionary actions by employees that affect organi zations in a variety of ways. OCBs are employee activities that support the social, psychological, or environmental context of an organization, but are not part of the formal job requirements (Borman & Motowidlo, 1993). OCBs do, however, contribute to the to cooperate (Podsakoff & MacKenzie, 1997). Many researchers suggest that OCB has two factors based on the target of the behavior ( e.g., LePine, Erez, & VanDyne, 2002;
3 organizationally targeted behaviors, such as enhancing the reputation of the organization, are referred to as OCB Organizational (OCB O), whereas interpersonally targeted OC B, such as helping to acclimate a new employee, are referred to as OCB Interpersonal (OCB I). Conversely, CWB consists of acts that harm or are intended to harm organizations or people in organizations (e.g., aggression, hostility, sabotage, theft, and w ithdrawal). CWB is potentially a serious organizational problem, given that 75% of employees report having stolen from their employers at least once and CWB can cost $6 to $200 billion annually (cf. Aquino, Lewis, & Bradfield, 1999). Similar to OCB CWB c an be differentiated according to the target of the behavior. The target of CWB can be either the organization (CWB O) or other employees (CWB I; Berry, Ones, & Sackett, 2007). In a factor analysis assessing the overlap between OCB, CWB, and task related behaviors, a three factor solution (OCB, CWB, and task behaviors) fit better than a two factor model that combined any of the constructs or a four factor model that included a common method factor (Kelloway et al., 2002). In another study, Rotundo and Sac kett (2002) concluded that contextual performance and counterproductive performance represent distinct dimensions of job performance. This supports the view of OCB and CWB as distinct, correlated constructs. Although OCB and CWB appear to be opposite ends of a voluntary behavior spectrum, meta analytic research has found only a moderate negative correlation ( r = .27; Dalal, 2005). Thus, it appears that OCB and CWB are not opposing ends of a continuum of voluntary behaviors. Furthermore, OCB and CWB appea r to have differential
4 relationships with other variables, including personality traits and organizational attitudes press). Consequently, the current study focuses on developing a model of OCB and CWB, based on the premise that there are both similarities and differences between the constructs. Antecedents to OCB and CWB Prior research has investigated the antecedents of OCB and CWB, including organizational attitudes and individual differences. Several studies hav e identified organizational attitudes that are consistently related to both OCB and CWB. In one such study a dominance analysis was used to investigate previously established correlates of indicated that job satisfaction, organizational support, and organizational justice received support as antecedents to both OCB and CWB. Results from a meta analysis support this finding showing that high job satisfaction, organizational commitment, and organizational justice are among the organizational attitudes that have the most support as theoretical antecedents of increased OCB and decreased CWB (Dalal, 2005). Fewer individual difference variables have received consistent testing as antecedents t o OCB and CWB. Consequently, there is less consensus regarding what individual difference variables relate to OCB and CWB. For example, one study found that lower positive affect, as well as higher negative affect and trait anger were related to more CWB but only positive affect was related to more OCB ( Miles et al., 2002 ). Another study found that positive affect was related to OCB I and OCB O, but neither positive nor negative affect was related to CWB (Lee & Allen, 2002). A qualitative
5 review of the literature identified conscientiousness, positive affect, and negative affect as the individual difference variables consistently linked to OCB and to CWB ; however, magnitudes of these relationships ranged from .10 to .41 (Dalal, 2005) In another review of the literature, conscientiousness, trait anger, and locus of control were found to be the most supported correlates of OCB and CWB The overall lack of consensus regarding which individual differences are related to OCB and CWB may be partially due to a relatively limited selection of personality variables that has been studied in terms of OCB (Borman, Penner, Allen, & Motowidlo, 2001). Furthermore, the study of individual difference variables related to CWB has focused on specific CWB behaviors (e.g., theft, sabotage, organizational retaliatory behavior, turnover, alcohol abuse), making it difficult to generalize study results to overall CWB. In summary, previous research has been able to identify some shared antecedents of OCB and CWB. Although there seems to be little dispute that organizational attitudes are correlated with OCB and CWB, there is less consensus regarding which individual differences are correlated with OCB and CWB, and to what degree these individual di fferences are related to OCB and CWB. Furthermore, very little research has looked at the role of job stressors in OCB and CWB. In order to address this gap in the literature, the current study aims to examine the relationship between a broader range of correlates of OCB and CWB, including individual differences and job stress. Job Stressors in OCB and CWB The effectiveness of an organization depends on the well being of its employees, as unhealthy, stressed, or injured workers are likely to be less effic ient and productive
6 (Sauter, Lim, & Murphy, 1996) One particularly salient health factor is job stress. Researchers have documented many negative consequences (strains) that result from job stressors such as workplace aggression, job dissatisfaction, an d negative emotion (Hershcovis et al., 2007; Miles et al., 2002). However, research investigating OCB and CWB as strains has been limited. Whereas CWB has been studied as a strain, decreased OCB has received little attention as an outcome of job stressor s. Consequently, on e potential way to improve our understanding of how employee well being relates to organizational effectiveness is to investigate the relationship between job stressors and voluntary work behaviors There are several reasons job stresso rs may lead to decreased OCB and increased CWB. For example, rational processing may be deferred under situations of stress, according to cognitive reasoning theory and self regulation theory ( e.g., Martinko, Gundlach, & Douglas, 2002). Specifically, a person may use the majority of his or her cognitive resources in order to cope with a job stressor, making it impossible to attend to additional demands, such as rational processing (e.g., avoiding aggression). Alternatively, interpersonal stressors, such as interpersonal conflict or low interactional justice (the degree to which a person is treated with politeness, dignity, and respect), may deter employees from engaging in OCB while encouraging CWB through social exchange theory. Social exchange theory p osits that people use of a subjective cost benefit ratio in their relationships, so that when a person perceives the costs of a relationship as outweighing the perceived benefits, the person will choose to leave the relationship. This may be evidenced as decreased willingness to help the other person
7 (less OCB), increased withdrawal (a form of CWB). Furthermore, interpersonal stressors have been shown to lead to aggression or retaliation in response to perceived attacks (e.g., Spector & Fox, 2005) Anoth er job stressor that may result in decreased OCB and increased CWB is organizational constraints. Restrictive organizational constraints may be perceived as a violation of the psychological contract, which can potentially lead to the desire f or retaliatio n (Aquino, Tripp, & Bies, 2001), abuse towards coworkers (Hoobler & Brass, 2006 ) or other strains Work overload is another job stressor that may lead to decreased opportunities to engage in OCB, and increase the amount of job withdrawal Work overload may make it necessary to withhold effort in order to cope with job demands. In summary, the job stressors of interpersonal conflict, interactional justice, organizational constraints, and job demands may relate to OCB and CWB. Previous studies have impli cated the ro le of job stressors in the voluntary behaviors of OCB and CWB For example, it has been hypothesized that job stressors and other environmental characteristics are appraised by employees and can lead to an emotional response which in turn lea ds to OCB and CWB depending on several other factors ( e.g., personality; Spector & F ox, 2002). In a partial test of this model, one study found that certain job stressors (interpersonal conflict, interactional justice, organizat ional constraints, work ov erload) are related to increased CWB, and surprisingly, increased OCB ( M iles, Borman, Spector, & Fox, 2002 ) Although the authors suggest this counterintuitive relationship exists because job stressors allow the opportunity to persevere, this finding may i nstead be a function of the particular items included in the OCB measure used. Specifically, some of the OCBs included in this study may simply
8 be more likely to occur under condit ions of stress. For example, employees may not have the good of the organizational environment. Although this study and others (e.g, Bolino and Turnley, 2005) have found a positive relationship between OCB and job stress, there is also contradictory evidence Specifically, there is support that job stressors, such as interpersonal conflict, interactional justice, work overload, and organizational constraints, are related to decreased OCB and increased CWB. For example, one study found that interpersonal conflict can lead to decreased OCB I and OCB O (Zellars, Tepper, & Duffy, 2002). Prior theory has also supported the role of job stressors in OCB. In situations of inte rpersonal stressors, for example, employees may not perceive social support from coworkers. According to social exchange theory, these employees may be less likely to provide OCB to their peers (Adams, 1965). Likewise, when the organization does not prev ent work overload or organizational constraints, this may be perceived as a violation of the psychological contract and lead to less OCB (e.g., Coyle Shapiro & Conway, 2005 ) Previous research investigating the relationship between job stressors and CWB h as shown that, for example, interpersonal conflict has been studied in various forms (e.g., incivility, bullying, perceived victimization) and has been shown to be positively related to CWB. Specifically, increased CWB is correlated with bullying experien ced (e.g., Ayoko, Callan, & Hartel, 2003), perceived victimization (e.g., Aquino, Tripp, &
9 Bies, 2001; Jockin, Arvey & McGue, 2001), and low interactional justice (e.g., Aquino, Galperin, & Bennett, 2004; Skarlicki & Folger, 1997). Other types of job stres sors have been linked to CWB. For example, previous research has linked work overload to job withdrawal, a form of CWB (Spector & Jex, 1998). Likewise, organizational constraints have been related to increased CWB (e.g., Penney & Spector, 2005; Storms & Spector, 1987). In general, job stressors such as interpersonal conflict, poor interactional justice, organizational constraints, and job demands are well supported antecedents to CWB. In summary, job stressors have been implicated as an antecedent to OCB and CWB, but this relationship has not been tested extensively. Prior theory and empirical testing has suggested interpersonal conflict, interactional justice, organizational constraints, and job demands as potential antecedents of OCB and CWB. Consequen tly, the current study aims to extend the literature by further investigating the relationship. Hypothesis 1a: Employees who report more interpersonal stressors (higher interpersonal conflict and lower interactional justice) and organizational stress ors ( h igher organizational constraints and job demands) will report less OCB and more CWB. Prior research suggests that the antecedents of OCB and CWB may be related to the target of the behaviors. In other words, OCB I and OCB O, as well as CWB I and CWB O, ma y have different antecedents. A meta analysis suggests that job stressors will be related to different types of OCB and CWB, based on target (Hershcovis et al., 2007). Specifically, interpersonal stress may lead to decreased OCB I and increased CWB I, co nsistent with social exchange theory. Furthermore, organizational stress, including
10 work load and organizational constraints, have been shown to be related to decreased OCB O and increased CWB O (Hershcovis et al., 2007). This is consistent with researc h on the psychological contract because prior research shows a relationship between violation of the psychological contract and voluntary behaviors (e.g. Coyle Shapiro & Conway, 2005; Hoobler & Brass, 2006) Consequently, the following hypotheses are pro posed. Hypothesis 1b: Employees who report more interpersonal stressors ( higher interpersonal conflict and lower interactional justice) will report less OCB I and more CWB I. Hypothesis 1c: Employees who report more organizational stress ors ( higher organiz ational constraints and job demands) will report less OCB O and more CWB O. The Role of Personality in the Stressor Strain Relationship Individual differences have been shown to be related to reports of job stressors ( e.g., Chen & Spector, 1991). Trait emo tion, for example, has been implicated as the mechanism responsible for the relationship between job stressors and OCB/CWB in prior theory ( e.g., Spector & Fox, 2002 ) and empirical investigations (e.g., Lee & Allen, 2002; Miles et al., 2002) Consequently one avenue for exploring the relationship between individual differences and job stressors is trait emotion. Trait emotion represents a be reached in order to reac t to a particular stimulus with that emotion (Lord, Klimoski, & Kanfer, 2002). In other words, people high in a trait emotion will be more likely to feel
11 that emotion on average and also more likely to perceive a stimulus as causing that emotion. Trait n egative emotion arousal, such that he/she will be more likely to perceive stress For example, negative affectivity has received extensive support as a correlate of job stressors (e.g., Penney & Spector, affectivity and job stressors have been well established, but specific trait emotions have received less empirical scrutiny as a correlate of job stress. Although no specific trait emot ion has received a great deal of testing as a correlate of job stress, overall, trait hostility and trait anger have been implicated as potential correlates of job stress. Trait anger represents the average amount or baseline level of anger that a person experiences. For example, prior research has shown that trait et al., 1998; Fox, Spector, & Miles, 2001). Trait hostility, another negative trait emotion, is the a verage amount of negative beliefs about others, including suspiciousness and cynicism, and is a facet of aggressiveness. Aggressiveness and hostility have also been related to perceiving interpersonal conflict, a job stressor (e.g., Hutri & Lindeman, 2002 ; Kiewitz & Weaver, 2001). Other individual differences, such as attributional style, may also be relevant to job stress. Attribution theory states that people are constantly aware of their environment and forming attributions regarding many aspects of ev ents that occur in their lives (e.g., Weiner, 1980). In the workplace, such attributions have been linked to many organizational outcomes. For example, attributions of unfairness has been linked to job
12 satisfaction, organizational commitment, organization al citizenship behavior, job withdrawal, and task performance (e.g., Cohen Charash & Spector, 2001; Colquitt, Conlon, Wesson, Porter, & Ng, 2001 ). Because attributions have been shown to be related to organizational outcomes, trait attributional style may be relevant to organizational outcomes. Furthermore, previous research has shown evidence that trait attributions may be relevant to perceived job stress. For example, external locus of control has been shown to be related to reports of job stressors ( e.g., Spector et al ., 2002). External locus of control is an individual difference that describes the degree to which people attribute consequences in their lives, both good and bad, to themselves (internal locus of control) versus other people or fate (e xternal locus of control). Because the Job Demands Control model (Karasek, 1979) than events within his or her control general attributions about control will prob ably be related to the reporting of stressors. Although the synergistic effect of job demands and personal control proposed in the Job Demands Control model has received inconsistent empirical confirmation, a main effect of external locus of control on in creased reporting of job stressors has been empirically supported (e.g., Fox, Spector, & Miles, 2001; Likewise, equity preference may influence how fair an employee perceives his/her environment. Equity prefere nce describes how much a person desires outcomes (e.g. pay) in a relationship (Huseman, Hatfield, & Miles, 1987). People who are more entitled prefer a higher amount of a reward in return for their efforts than do benevolents,
13 who generally prefer to give more than they receive. Consequently, people who are more entitled will be more likely to perceive an exchange as unfair, and because unfairness can be a job stressor ( e.g., Fox, Spector, & Miles, 2001 ; Francis, 2003; Zohar, 1995 ), entitled employees may be more likely to perceive and report job stress. The current study extends previous research by investigating the role of specific trait emotion and attributional style in job stress. Although these individual differences have been implicated as correla tes of job stressors, they have not received extensive testing. Because the majority of these studies focus on interpersonal conflict, interactional justice, job demands, and organizational constraints as stressors (e.g. Fox, Spector, & Miles, 2001; Spect the current study. Specifically it is expected that trait hostility, trait anger, locus of control, and entitled equity preference will be related to reports of job stress. Hypothesis 2: Employees with greater trait anger, trait hostility, external locus of control and entitled equity preference will r eport more job stressors, including lower interactional justice and higher interpersonal conflict, organizational constraints, and job demands The R ole of Job Stressors in OCB and CWB Building a model of OCB and CWB is important to furthering our understanding of these workplace behaviors. Previous models of OCB and CWB have focused on job affect and job cognitions (Lee & Allen, 2002) or emotion (Spe ctor & Fox, 2002). Neither study has received overwhelming empirical support. For example, one study used data from 149 registered nurses and their coworkers to study the relationship between job
14 attitudes (intrinsic satisfaction, procedural justice, pay cognitions, and work schedule load) and OCB I, OCB O, and CWB (Lee & Allen, 2002). None of the job attitudes were related to OCB I, intrinsic satisfaction and procedural justice related to OCB O, and pay cognitions related to CWB. Several of the discret e emotions predicted OCB I, OCB O, and CWB. However, only two of the six relationships between trait affectivity and voluntary behaviors were significant (positive affect was correlated with OCB I and OCB O). In general, this model was not supported, but does indicate that future research should consider a broad range of individual differences and specific emotion when investigating a model of OCB and CWB. Other previous theory has focused on the mediating role of emotion in the relationship between job s tressors and voluntary behaviors (Spector & Fox, 2002). However, such research has not received extensive empirical support. A test of portions of this model was generally supportive, except that OCB unexpectedly correlated positively with job stressors. Job stressors (interpersonal conflict, interactional justice, organizational constraints, and job demands) were related to OCB and CWB, and trait affect (anger, positive affectivity, and negative affectivity) accounted for unique variance above and beyon d the job stressor variables. This model provides some evidence that job stressors and trait emotion may be related to OCB and CWB. However, the study tested a relatively narrow set of individual differences and suggested that future research include mor e varied individual differences. Consequently, the current model aims to expand the individual differences studied in OCB and CWB research. Prior research and theory has suggested that certain individual differences are
15 related to OCB and CWB (e.g., Aquin o, Tripp, & Bies, 2001; Hershcovis et al., 2007; Neuman & Baron, 1998; Spector & Fox, 2002; Storms & Spector, 1987). T rait hostility and trait anger are emotions that have been implicated as correlates of OCB and CWB len, in press; Spector & Fox, 2002). Furthermore, OCB and CWB may be related to attributional style, including locus of control (e.g., Bennett, 1998; Borman et al., 2001; Fox & Spector, 1999) and equity preference (e.g., Kickul & Lester, 2001; Kwak, 2006; Liu, 2006; Mason & Mudrack, 1997; Shore, Sty, & Strauss, 2006). Consequently, certain individual differences will likely be related to OCB and CWB. Hypothesis 3: Greater trait hostility, trait anger, external locus of control, and entitled equity prefere nce will be related to less OCB and more CWB. Although previous models of OCB and CWB have received some empirical support, an extensive investigation is necessary to provide further evidence for these models. For example, prior theory of the role of job stressors in OCB and CWB has viewed emotion as an outcome of stress, thereby leading to OCB and CWB (Spector & Fox, 2002). Other research has suggested that individual differences, including trait emotion, may predispose a person to report job stressors ( Fortunato & Harsh, 2006, Spector & Fox, 2002). Negative emotion and attributional style have been shown to affect the way people perceive their environments, and may consequently lead to perceived job stressors in various ways (e.g., directly or by affecti their environments; Spector, Zapf, & Chen, 2000). This relationship has not received adequate empirical scrutiny despite theoretical
16 and empirical support. Furthermore, these analyses used cross sectional data. Cross sectional data has been shown to generate biased estimates of longitudinal mediation parameters, even under ideal circumstances (Maxwell & Cole, 2007). Due to the lack of extensive testing of a job stress model of OCB and CWB, future testing of the mediating roles of th ese variables would benefit from a longitudinal study design. This type of scrutiny will allow researchers to rule out other alternatives and establish a temporal precedence (although temporal precedence does not, by itself, imply causality). Furthermore separation of the predictor and criterion helps establish stability of the effect by removing the daily effects of mood. Based on previous empirical support and prior theory, it is likely that individual differences will lead to reported job stressors, which will in turn influence employee engagement in OCB and CWB (Figure 1). Hypothesis 4a: Job stressors (lower interactional justice and higher interpersonal conflict, organizational constraints, and job demands), will mediate the relationships between i ndividual differences (trait hostility, trait anger, external locus of control, and entitled equity preference) and OCB/CWB. Previous research on OCB and CWB has shown support for target based llen, in press). For example, meta analytic research has shown that OCB I and OCB O, as well as CWB I and CWB O, have differential relationships with certain antecedents (Dalal, 2005). Furthermore, a meta analysis of job stressors and CWB has shown that certain types of job stressors may be related more strongly to certain types of CWB (Hershcovis et al., 2007). Specifically, interpersonal conflict was more strongly related to CWB I than to
17 CWB O, and organizational stressors were more strongly related t o CWB O than to CWB I. This previous empirical research shows support for a target based model of job stressors as related to OCB and CWB. Furthermore, there is theoretical evidence that a target based model of OCB and CWB would provide better fit than an overall model of OCB and CWB. Specifically, social exchange theory would suggest that employees will engage in OCB or CWB towards coworkers (i.e. interpersonally directed) when they have been affected by other coworkers. Consequently, interpersonal st ressors may be related to decreased OCB I and increased CWB I. Conversely, breach of the psychological contract, including excessive job demands or organizational constraints, may relate to retaliation against the organization or decreased motivation to he lp the organization. An employee who has had a psychological contract breach may engage in less OCB O or more CWB O. Consequently, the target based model of OCB and CWB may provide greater insight into these relationships (Figure 2). Hypothesis 4b: Inter personal stressors (lower interactional justice and higher interpersonal conflict) will mediate the relationships between individual differences (trait hostility, trait anger, external locus of control, and entitled equity preference) and OCB I / CWB I. Hy pothesis 4c: Organizational stressors (higher organizational constraints and job demands) will mediate the relationships between individual differences (trait hostility, trait anger, external locus of control, and entitled equity preference ) and OCB O / CW B O.
18 Chapter Two Method Participants and Procedure Participants were recruited through the Syracuse University Study Response Project. This organization connects researchers with participants who have signed up with them in order to complete online su rveys in exchange for payment or raffle entry. This recruitment process was chosen based on its use in previous studies (e.g., Dennis & Winston, 2003; Piccolo & Colquitt, in press; Van Ryzin, 2004; Vodanovich, Wallace, & Kass, 2005) and prior validation o f online samples (e.g., Frame & Beaty, 2000; Stanton, 1998; Yost & Homer, 1998). Approximately 2 5,000 potential participants were emailed to determine eligibility ( worked 30 or more hours per week, have been mentored at some point in their career, and had a supervisor to whom they could email a survey). The 700 people who responded and met all criteria receive d an email invitation to complete a questionnaire twice (about 20 minutes each with a two week break) for ten dollars (Appendix A) Two weeks a fter the Time 2 data collection, participants were asked to email a short (2 5 minute) measure to their supervisors (Appendix B) In this Time 3 data collection, levels of OCB and CWB. Upon completion of the questionnaires, the Study Response
19 group provided $10 to each participant At Time 1, 571 people filled out the survey. After removing participants who worked less than 20 hours a week, participants who reported participa nt numbers that could not be matched, and duplicate data, there were 424 responses. These 424 people were emailed two weeks later to participate for Time 2 data collection. The similar criteria (e.g., ID matching, minimum hours worked per week) were used t o prepare this data. At Time 2, 277 responses were usable. Time 3 responses resulted in a final sample size of 212 self supervisor pairs. The participants were 5 7.2 % female, ethnically heterogeneous ( 150 White/Caucasian, six Bl ack/African American, 35 A sian, nine Hispanic, and the remaining were other ethnicities), and on average 37. 12 years old ( sd = 9.3 6 ). The average tenure within the organization was 67.45 months ( sd =77.16). Participants were employed in a variety of occupations (e.g., retail, child care, paralegal, administrative). Their supervisors were 46.2 % female, ethnically heterogeneous (69.2% White/Caucasian, 5.6% Black/African American, 19.0% Asian, 4.1% Hispanic, and the remaining were other ethnicities), and on average 42.8 3 years old ( SD = 10. 77 ). On average, the supervisors reported knowing the participant for 59.4 8 months ( SD = 74.62 ). Measures Demographics. Participants and supervisors reported demographic information, including their age, gender, race, as well as job information suc h as number of hours worked per week, type of job, organizational tenure, and job tenure. Trait hostility. Hostility was measured using the 8 item hostility subscale of the Buss Perry Aggression Questionnaire (1992). Participants responded to items such
20 ther peopl e always seem to get the point Likert scale. Higher responses indicate more hostility. Correlation alpha at time 1 was .90, at time 2 was .93. Trait anger. The 10 item t rait anger subscale of the revised State Trait Anger Expression Inventory (Spielberger, 1988) was used to measure this construct. I am quick a four point scale (1= not at all, 4 = very much so). Higher scores indicate higher levels of trait anger. At time 1, the coefficient alpha was .90, and .93 at time 2. Locus of control. To assess locus of control in the work domain, the Work Locus of Control scale (Spector, 1988) was used On this 16 item Likert scale, respondents point scale. Several items are reverse scored, in the direction such that higher scores indicate an internal locus of control. The coefficient alphas at time 1 and time 2 were .84 and .82, respectively. Equity preference. The 16 item Equity Preference Questionnaire (Sauley & Bedeian, 2000) was chosen to measure this construct. Participants reported how well point Likert scale. Higher scores indicate higher levels of entitled equity preference after reverse scoring several ite ms. The coefficient alpha was .86 at time 1 and .82 at time 2. Interpersonal conflict. The Interpersonal Conflict at Work Scale (ICAWS) was used
21 are rated on a 5 er scores represent more interpersonal conflict. At time 1, the coefficient alpha was .80, and .84 at time 2. Interactional justice The four item interactional justice factor from the point Likert scale. Higher scores represent grea ter interactional justice. The coefficient alphas were .94 and.92 at time 1 and time 2, respectively. Job demands. The Quantitative Workload Inventory (QWI) is a measure of the amount of work in a job, and was chosen to represent this construct. The scale includes H participants rate d using a 5 er scores represent higher job demands. The coefficient alphas were .88 at time 1 and .92 at time 2. Organizational constraints. The Organizational Constraints Scale (OCS) used in Spector and Jex (1998) was chosen to assess this construct. Eleven items, representing the 11 areas of con were presented to participants. Participants indicate d ) makes it difficult or impossible for them to do their jobs. Respondent s use a 5 point frequency constraints. At time 1 and time 2, the coefficient alphas were .91 and.93, respectively.
22 Organization al citizenship behavior. OCB was assessed using Williams and Anderson item (1991) survey. OCB I and OCB O are each measured with seven items on which the participant and supervisor report to how often the participant engages in certain activities, s uch as help ing others who have been absent Responses were provided on 7 point frequency scale that range s from Higher scores reflect greater OCB. For OCB I, the coefficient alphas were .91, .92, and .96 at time 1, time 2, and ti me 3, respectively. For OCB O, the coefficient alphas at time 1, time 2, and time 3 were .87, .90, and .94. The overall OCB coefficient alpha at time 1 was .91, at time 2 was .93, and at time 3 was .97. Counterproductive work behavior. Robinson and Benne item CWB measure was chosen to represent the construct of CWB. Participants and supervisors respond ed on a 1 7 scale (never every day) how often the participant engage s in behaviors ven items that represent CWB I and 12 that represent CWB O Although there is some concern that items from this scale may overlap with OCB items, prior research investigating the factor structure of OCB and CWB support the use of these scales without mod ifications I were .91, .92, and .96. The coefficient alphas for CWB O were .94, .95, and .98 at time 1, time 2, and time 3, respectively. For overall CWB, the coefficient alphas were .96 at time 1, .96 at time 2, and .98 at time 3.
23 Chapter Three Results Descriptive Statistics Means, standard deviations, and intercorrelations among study variables are provided in Tables 1 3. Correlations among Time 1 variables appear in the lower half of Table 1, whereas Time 2 correlations appear in the upper half of Table 1. Correlations between Time 1 and Time 2 variables appear in Table 2. Table 3 shows the correlations between Time 3 supervisor report data and the self reported dat a at both Time 1 and Time 2, as well as the means and standard deviations for all study variables. Hypothesis Testing Hypothesis 1a proposed that employees who report more interpersonal stressors (higher interpersonal conflict and lower interactional justi ce) and organizational stressors (higher organizational constraints and job demands) would report less OCB and more CWB. This was partially supported using Time 2 self reported stressors and Time 3 supervisor reported behaviors (Table 3), consistent with the proposed model Supervisors rated participants who reported lower interactional justice and higher interpersonal conflict and organizational constraints as engaging in less OCB ( r = .41, .29, .24, p <.001 respectively) and more CWB ( r = .33, .58, .42, p <.001 respectively). Job demands were not correlated with either behavior.
24 Hypotheses 1b and 1c focused on the interpersonal and organizational targets of OCB and CWB. Specifically, Hypothesis 1b proposed that employees who report more interpersonal stressors (higher interpersonal conflict and lower interactional justice) will report less OCB I and more CWB I. Likewise, Hypothesis 1c proposed that employees who report more organizational stressors (higher organizational constraints and job demands) w ould report less OCB O and more CWB O. These hypotheses were partially supported using correlations from Time 2 self reported stressors and Time 3 supervisor reported behaviors (Table 3) Supervisors rated participants who reported lower interactional ju stice and higher interpersonal conflict as engaging in less OCB I ( r = .36, .23, p <.001, respectively) and more CWB I ( r = .32, .56, p <.001 respectively). Likewise, supervisors rated participants who reported higher organizational constraints as engaging in less OCB O ( r = .29, p <.001 ) and more CWB O ( r = .42, p <.001 ). Job demands were again not co rrelated with the workplace behaviors. Hypothesis 2, which stated that employees with greater trait anger, trait hostility, external locus of control and entitle d equity preference will report more job stressors, including lower interactional justice and higher interpersonal conflict, organizational constraints, and job demands was partially supported using Time 1 self reported personality and Time 2 self reporte d job stressors. Data showed that correlations were consistent with the hypotheses, with the exception of relationships including job demands (Table 2 ). Trait anger correlated with interactional justice (r= .22, p< .001), interpersonal conflict (r= .40, p< .001), and organizational constraints (r= .37, p< .001). Trait hostility correlated with interactional justice (r= .42, p< .001), interpersonal
25 conflict (r= .50, p< .001), and organizational constraints (r= .37, p< .001). Internal locus of control correlat ed with interactional justice (r= .40, p< .001), interpersonal conflict (r= .31, p< .001), and organizational constraints (r= .33, p< .001). Entitled equity preference correlated with interactional justice (r= .42, p< .001), interpersonal conflict (r= .36, p< .001), and organizational constraints (r= .23, p< .001). Hypothesis 3 was supported using Time 1 self reported personality and Time 3 supervisor reported behaviors (Table 3). Specifically, gr eater trait hostility, trait anger, external locus of control, and entitled equity preference was r elated to less OCB and more CWB ( absolute value of correlations ranged from .24 .48, p <.001). To test the overall pattern of relationships as proposed in Hypothesis 4a, a path analysis using the Time 1 self report d ata was performed. Time 1 data was chosen based on the larger sample size ( n =358). This resulted in an identified model. However, the fit statistics were not satisfactory ( RMSEA =.20, NFI = .70, NNFI = .51, CFI = .71), and parameter estimates were low. T he individual relationships proposed in Hypotheses 4a c were then tested using bootstrapped Sobel analyses. Because most samples violate the assumption of normality, bootstrapping methods are generally preferred (Preacher & Hayes, 2004). Bootstrapping is a process that generates randomly sampled observations with replacement from the data set, and computes the statistic of interest in each resample. This process is repeated many times in order to approximate the sampling distribution of the statistic. Thi s statistic can then be used in hypothesis testing that requires fulfillment of distributional assumptions.
26 The bootstrapping procedure is performed using the raw data in a process based on the Sobel test. To perform this procedure, a command set is exec uted in SPSS syntax, activating a macro (Preacher & Hayes, 2004). Larger resamples require more time, but provide more accurate estimates. Because of the large number of hypothesis tests, 1,000 resamples was chosen to balance estimation accuracy and compu tational workload. Consequently, an alternative test of the mediational hypotheses 4a c is based on the output from the bootstrapping macro (Table 4). Results from the bootstrapped analyses are mixed. Overall, the indirect effect of personality on super visor reported CWB, mediated by stressors (interpersonal conflict, low interactional justice, and organizational constraints) was significant except for hostility and interactional justice. In this case, the direct effect was not significantly reduced whe n mediated by interactional justice. The patterns for CWB I and CWB O were identical to that of overall CWB, consistent with the findings from the exploratory factor analysis. The pattern for OCB was inconsistent. In all cases, trait anger was mediated by stressors (interpersonal conflict, low interactional justice, and organizational constraints). Equity preference and hostility were not mediated by interpersonal conflict, and hostility and locus of control were not mediated by organizational constrai nts. The pattern is more predictable when interpreting the interpersonal and organizational dimensions of OCB separately. Specifically, OCB I was mediated only by interactional justice for all personality traits. Conversely, the relationship between per sonality and OCB O was mediated by all stressors (interpersonal conflict, low interactional justice, and organizational constraints) except for the hostility/interpersonal conflict relationship.
27 Chapter Four Discussion The purpose of the current stud y was to propose and test relationships within a stressor strain model of OCB and CWB. The study provides evidence that OCB and CWB be influenced by organizational stressors, and that personality, particularly trait specific emotion and attributional styl e, relate to job stressors and strains. Previous empirical research on the relationship between job stressors and OCB has been inconsistent, and this current study provides further evidence of a negative relationship between job stressors and OCB. Furth ermore, the study of specific, trait emotion and attributional style goes beyond the relatively narrow set of personality variables used to study job stressors and OCB/CWB in the past, contributing to our understanding of how personal characteristics influ ence the organizational environment. Based on the zero order correlations, the relationships between personality and organizational behaviors are consistent with previous empirical and theoretical research. Specifically, previous reviews have encouraged the study of a broader range of personality characteristics, including discrete emotion (e.g. Lee & Allen, 2002), in the study of OCB and CWB. The current research supports the personality variables of trait anger, trait hostility, locus of control, and equity preference as correlates of OCB and CWB. Likewise, the job stress literature has focused on trait negative affect. Future
28 research in the area of specific personality characteristics may be beneficial in our understanding of how perception and vul nerability affect the experience and reporting of organizational stressors. This study used a longitudinal design, as suggested by previous reviews in the area (e.g., Dalal, 2005). Recent research has shown that as little as 10% of mediational studies use full longitudinal designs (Maxwell & Cole, 2007). Collecting data at multiple time points has several advantages. First, separation of the predictor and criterion can result in effects that have greater stability and generalizeability. Also, collecting the independent and dependent variables at multiple time points can help rule out plausible alternatives such as autoregressive models. Future structural equation modeling analyses can be conducted with the data to examine autoregressive effects. Another strength of the study design was the multi source data. Some research has suggested that collecting data exclusively from one source may, under certain circumstances, lead to bias. Due to this potential problem, a vast amount of OCB and CWB research has used supervisor or peer reports. However, it is ultimately unknown whether self peer or superisor reported data is the most accurate. Although the study did not address this question, this study supports previous research (e.g., Dalal, 2005; LePine model may vary based on source of data (self report vs. supervisor support). This provides further evidence that supervisor reports may not necessarily be more appropriate than sel f reports. Specifically, correlations within self reported variables and between self and supervisor reported variables varied greatly. Supervisor reports showed less
29 distinction between interpersonal and organizational dimensions of the organizational behaviors, although self reported data factored as expected, consistent with previous research (Dalal, 2004). This may indicate that supervisors rely on halo effect when rating these behaviors. Furthermore, it is unclear if the self reported relationship s are inflated, being generally stronger than self and supervisor reported data, or if the multiple source correlations are attenuated. Overall, this study also provides implications for the factor structure of OCB and CWB. Specifically, the pattern o f correlations indicates that there is a negative relationship between OCB and CWB. Furthermore, exploratory factor analysis showed that OCB and CWB are separate scales, contributing to the evidence that this is little scale or construct overlap. Limita tions A major limitation was the analysis method used. Structural equation modeling is a more appropriate test of model fit than path analysis or testing of individual relationships. However, given the relatively low correlations among observed variables (e.g., organizational constraints and job demands, or trait anger and trait hostility), these measures cannot adequately represent a latent variable (e.g., organizational stressors or trait emotion). Future research designed at identifying and measuring indicators of personality and job stressors would provide broader analysis options and potentially contribute to supporting a stressor strain model of OCB and CWB. A notable limitation was the surprising lack of correlations with the job demands measure. Although previous research has used this scale successfully, the current study
30 showed almost no correlations between the job demands measure and the other study variables in any time points. This may indicate a problem with the data collection. However, because the null results are limited to the job demands measure, we expect that any such problems do not affect the remaining correlations, or at worst, attenuate the relationships. Despite the positive correlations found in this study between OCB and job stressors, previous empirical research has been inconsistent in showing if the relationship between OCB and CWB is positive or negative. Theory has put forth that excessive OCB may cause job stress by increasing job demands, whereas other theory posits th at job stressors may violate the psychological contract and discourage OCB. In other words, employees who perceive high job stressors may also perceive violation of the psychological contract, leading to decreased OCB, whereas employees with low job stres sors may engage in OCB excessively, leading to increased job stress. A cyclical relationship could explain the contradictory correlational evidence and possibly be tested with a longitudinal design over many time points. This study used only three time p oints and is unable to address this research question. Future Research The OCB and CWB areas are limited by the lack of experimental research. A limited number of studies have used quasi experimental design (Greenberg, 1990) or vignettes (Scott & Colquitt 2007), but there is a dearth of experimental studies that include OCB or CWB as dependent variables. Experimental studies within an organization would be subject to ethical questions, pragmatic difficulties, and other such
31 obstacles. Conversely, experi mental lab studies are thwarted by the lack of lab measures of OCB and CWB. Future research that validates such measures, perhaps based on the prosocial and antisocial social psychology literature, would be pivotal in our understanding of causality in OCB and CWB. For example, an experimental lab study would be better able to distinguish if stressors precede emotions (as suggested in Spector & Fox, 2002) and if trait emotion precedes stressors (as suggested in the current study). It is, of course, possib le that trait emotion precedes stressors, which in turn precede state emotion. This relationship, too, could be tested in experimental research. Experimental research in the area of self esteem maintenance is one area that may prove fruitful. Research in social psychology shows that ego threat (for example, a negative public evaluation) can result in several reactions, including overcompensation, withdrawal, or aggression. This may be exhibited in the workplace as OCB and CWB in reaction to a negative pe rformance evaluation. Although experimental manipulation of feedback would be unethical in an organization, a lab experiment would allow researchers to manipulate feedback value (positive, negative, or neutral) and measure if OCB or CWB occurs following t he feedback. However, this experiment cannot be conducted without adequate lab measures of OCB and CWB. Another potential way to study these causal relationships is with a real time diary study. Employed participants could fill out a survey of personalit y traits prior to inception of diary keeping. Then, for a period a week, participants could use a programmed cell phone, PDA, or computer to log their emotions and activities over the past hour. The activities could be coded as OCB, CWB, or other. This would allow
32 researchers to better understand how trait emotion and mood affect OCB and CWB. Likewise, cortisol testing would help us understand the role of stress in OCB and CWB. Cortisol is a chemical produced by body when under stress and is relatively easy to measure. Employed participants could complete a survey of their personality constructs and general stress level prior to the cortisol testing. Then, for three days, participants could prepare their cortisol tests with a simple, painless cheek swa b. The participant mails the sample in a prepackaged mailer to a laboratory, which then provides researchers with the results. Each day, the participants could also report their OCB and CWB, so that researchers can investigate if there is a link between cortisol level and their rates of OCB and CWB. In addition to research suggestions, this study provides implications for practice. In this model, attributions were related to important workplace outcomes. Because attributions can be c hanged, organizations should consider using attribution training to help encourage positive workplace behavior and discourage detrimental workplace behavior. Previous interventions have been used in the clinical area and warrant investigation in the workp lace. Conclusion Overall, this study proposed and partially supported a stressor strain model of OCB and CWB. Despite some limitations, the data include multiple reporting sources and longitudinal design, consistent with previous research suggestions. In general, there was some support for this model, mostly from the bootstrapped Sobel analyses. Furthermore, the results provide further support for the model hypothesized by Spector and Fox (2002).
33 Specifically, part of the Spector and Fox (2002) model sta tes that control perceptions and personality influence appraisal of the work environment as stressful. This is consistent with the current study, which measured control attributions (locus of control) and personality (equity preference, trait anger, and t rait hostility) and showed that these relate to reporting of job stressors. Due to the self reported nature of the stressors, it is fair to opposed to objective workload and conflict. In the Spector and Fox (2002) model, appraisal then leads to emotion, which then effects OCB and CWB. Although the overall model could not be tested, the data provides preliminary support for the hypothesized model.
34 References Adams, J. S (1965). Inequity in social exchange. In L. Berkowitz (Ed.), Advances in e xperimental social p sychology (Vol. 2, pp. 267 299). New York: Academic Press. Aquino, K., Galperin, B., & Bennett, R. (2004). Social status and aggressiveness as moderators of th e relationship between interactional justice and workplace deviance. Journal of Applied Social Psychology 34 1001 1029. Aquino, K., Lewis, M. U., & Bradfield, M. (1999). Justice constructs, negative affectivity, and employee deviance: A proposed model and empirical test. Journal of Organizational Behavior, 20, 1073 1091. Aquino, K., Tripp, T., & Bies, R. (2001). How employees respond to personal offense: The effects of blame attribution, victim status, and offender status on revenge and reconciliation i n the workplace. Journal of Applied Psychology 86 52 59. Ayoko, O., Callan, V., & Hrtel, C. (2003). Workplace conflict, bullying and counterproductive behaviors. International Journal of Organizational Analysis 11 283 301. Baron, R., & Kenny, D. (19 86). The moderator mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology 51 1173 1182.
35 Bennett, R. J. (1998) Perceived powerlessness as a cause of employee deviance. In R.W. Kelly & J. M. Collins (Eds.), Dysfunctional behavior in organizations: Violent and deviant behavior (pp. 231 238). Stamford, CT: JAI Press. Berry, C., Ones, D., & Sackett, P. (2007). Interpersonal devianc e, organizational deviance, and their common correlates: A review and meta analysis. Journal of Applied Psychology 92 410 424. Bolino, M. C., Turnley, W. H., & Niehoff, B. P. (2004). The other side of the story: Reexamining prevailing assumptions about organizational citizenship behavior. Human Resource Management Review, 14, 229 246. Bongard, S., & al'Absi, M. (2005). Domain specific anger expression and blood pressure in an occupational setting. Journal of Psychosomatic Research 58 43 49. Borman, W C., & Motowidlo, S. J. (1993). Expanding the criterion domain to include elements of extrarole performance. In N. Schmitt & W. C. Borman (Eds.), Personnel selection in organizations (pp. 71 98). San Francisco: Jossey Bass. Borman, W. C., Penner, L. A., Allen, T. D., & Motowidlo, S. J. (2001). Personality predictors of citizenship performance. International Journal of Selection and Assessment, 9, 52 69. Brondolo, E., et al. (1998). Anger related traits and response to interpersonal conflict among New York City traffic agents. Journal of Applied Social Psychology 28 2089 2118. Buss, A., & Perry, M. (1992). The Aggression Questionnaire. Journal of Personality and
36 Social Psychology 63 452 459. Chen, P., & Spector, P. (1991). Negative affectivity as the underlying cause of correlations between stressors and strains. Journal of Applied Psychology 76 398 407. Cohen Charash, Y., & Spector, P. E. (2001). The role of justice in organizations: A meta analysis. Organizational Behavior and Human Decision Processes, 86 278 321. Colquitt, J. (2001 ). On the dimensionality of organizational justice: A construct validation of a measure. Journal of Applied Psychology, 86, 386 400 Colquitt, J. A., Conlon, D. E., Wesson, M. J., Porter, C. O., & Ng, Y. K. (2001). Justice at the millenium: A meta analytic review of 25 years of organizational justice research. Journal of Applied Psychology, 86, 425 445. Coyle Shapiro, J. A., & Conway, N. (2005). Exchange relationships: Examining psychological contracts and perceiv ed organizational support. Journal of Applied Psychology, 90, 774 781. Dalal, R. S. (2005). A meta analysis of the relationship between organizational citizenship behavior and counterproductive work behavior. Journal of Applied Psychology, 90, 1241 1255. leadership instrument. Leadership and Organization Development Journal, 24, 455 459. Fan, X., B. Thompson, & L. Wang. (1999). Effects of sample size, estimation method,
37 and mo del specification on structural equation modeling fit indexes. Structural Equation Modeling 6, 56 83. Fortunato, V., & Harsh, J. (2006). Stress and sleep quality: The moderating role of negative affectivity. Personality and Individual Differences 41 82 5 836. Fox, S., & Spector, P. E. (1999). A model of work frustration aggression. Journal of Organizational Behavior, 24 915 931. Fox, S., & Spector, P.E. (2005). Counterproductive work behavior: Investigations of actors and targets. Washington D.C.: APA Press. Fox, S., Spector, P. E., & Miles, D. (2001). Counterproductive work behavior (CWB) in response to job stressors and organizational justice: Some mediator and moderator tests for autonomy and emotions. Journal of Vocational Behavior, 59, 291 309 Frame, J. H., & Beaty, J. C. (2000, April). An empirical investigation of high technology survey methods: Paper and pencil, email, and web based Which is better? Paper presented at the annual meeting of the Society for Industrial and Organizational Psyc hology, New Orleans, LA. Francis, L. (2003). Organizational justice, sensitivity to injustice and the experience of stress. Dissertation Abstracts International: Section B: The Sciences and Engineering, 64(1 B), 451. Greenberg, J. (1990). Employee theft a s a reaction to underpayment inequity: The hidden cost of pay cuts. Journal of Applied Psychology, 75, 561 568. Hershcovis, M., et al. (2007). Predicting workplace aggression: A meta analysis. Journal of Applied Psychology 92 228 238.
38 Hoobler, J., & Br ass, D. (2006 ). Abusive s upervision and f amily u ndermining as d isplaced a ggression. Journal of Applied Psychology, 91, 1125 1133. Huseman, R., Hatfield, J., & Miles, E. (1987 ). A new perspective on equity theory: The equity sensitivity construct. Academy o f Management Review 12 222 234. Hutri, M., & Lindeman, M. (2002). The role of stress and negative emotions in an occupational crisis. Journal of Career Development 29 19 36. Jockin, V., Arvey, R., & McGue, M. (2001). Perceived victimization moderates self reports of workplace aggression and conflict. Journal of Applied Psychology 86 1262 1269. Judge, T., Thoresen, C., Bono, J., & Patton, G. (2001). The job satisfaction job performance relationship: A qualitative and quantitative review. Psychologic al Bulletin 127 376 407. Karasek R. ( 1979 ). Job demands, job decision latitude and mental strain: Implications for job redesign. Administrative Science Quarterly, 24, 285 306 Kelloway, E. K., Loughlin, C., Barling, J., & Nault, A. (2002). Self reporte d counterproductive behaviors and organizational citizenship behaviors: Separate but related constructs. International Journal of Selection and Assessment, 10, 143 151. Kickul, J., & Lester, S. (2001). Broken promises: Equity sensitivity as a moderator be tween psychological contract breach and employee attitudes and behavior. Journal of Business and Psychology 16 191 217. Kiewitz, C., & Weaver, J. (2001). Trait aggressiveness, media violence, and perceptions
39 of interpersonal conflict. Personality and In dividual Differences 31 821 835. Kwak, A. (2006). The relationships of organizational injustice with employee burnout and counterproductive work behaviors: Equity sensitivity as a moderator. Dissertation Abstracts International: Section B: The Sciences and Engineering, 67(2 B), 1190. Lee, K., & Allen, N. J. (2002). Organizational citizenship behavior and workplace deviance: The role of affect and cognitions. Journal of Applied Psychology, 87, 131 142. LePine, J. A., Erez, A., & Johnson, D. E. (2002). The nature and dimensionality of organizational citizenship behavior: A critical review and meta analysis. Journal of Applied Psychology, 87, 52 65. Liu, Y. (2006). Investigating turnover intention among emergency communication specialists. Dissertation Ab stracts International: Section B: The Sciences and Engineering, 66(11 B), 6322. Lord, R., Klimoski, R., & Kanfer, R. (Eds) (2002). Emotions in the Workplace: Understanding the Structure and Role of Emotions in Organizational Behavior. San Francisco, CA: Jo ssey Bass. Martinko, M. J., Gundlach, M. J., & Douglas, S. C. (2002). Toward an integrative theory of counterproductive workplace behavior: A causal reasoning perspective. International Journal of Selection and Assessment, 10, 36 50 Mason, E., & Mudra ck, P. (1997). Are individuals who agree that corporate social responsibility is a 'fundamentally subversive doctrine' inherently unethical?
40 Applied Psychology: An International Review 46 135 152. Maxwell, S. E., & Cole, D. A. (2007). Bias in cross sec tional analyses of longitudinal mediation. Psychological Methods, 12, 23 44. Miles, D. E., Borman, W. E., Spector, P. E., & Fox, S. (2002). Building an integrative model of extra role work behaviors: A comparison of counterproductive work behavior with o rganizational citizenship behavior. International Journal of Selection and Assessment, 10, 51 57. O'Brien, K. E., & Allen, T. D. (In press) Relative importance of correlates of organizational citizenship behavior and counterproductive work behavior. Human Performance. Measuring n orms for w orkplace d eviance and c itizenship b ehavior Paper presented at the 23 rd annual Society of Industrial/Organizational Psychologists conference. Organ, D. W. (1997). Organizational citizenship behavior: It's construct clean up time. Human Performance, 10, 85 97. Penney, L., & Spector, P. (2005). Job stress, incivility, and counterproductive work behavior (CWB): The moderating role of negative affectivity. Journal of Organizational Be havior, 26, 777 796. Perrew, P. (1986). Locus of control and activity level as moderators in the quantitative job demands satisfaction/psychological anxiety relationship: An experimental analysis. Journal of Applied Social Psychology 16 620 632. Peters L. H., & O'Connor, E. J. (1980). Situational constraints and work outcomes: the
41 influences of a frequently overlooked construct. Academy of Management Review, 5, 391 397. Piccolo, R. F., & Colquitt, J. A. (In press). Transformational leadership and job b ehaviors: The mediating role of core job characteristics. Academy of Management Journal Podsakoff, P., & MacKenzie, S. (1997). Impact of organizational citizenship behavior on organizational performance: A review and suggestions for future research. Human Performance, 10, 133 151. Preacher, K., & Hayes, A. (2004 ). SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, Instruments & Computers, 36 717 731. Robinson, S. L., & Bennett, R. J. (1995). A typology of deviant workplace behaviors: A multidimensional scaling study. Academy of Management Journal, 38, 555 572. Rotundo, M., & Sackett, P. (2002). The relative importance of task, citizenship, and counterproductive performance to global ratings of job performance: A policy capturing approach. Journal of Applied Psychology, 87, 66 80. Sauley, K., & Bedeian, A. (2000). Equity sensitivity: Construction of a measure and examination of its psychometric properties. Journal of Management, 26, 885 910. Sa uter, S.L., Lim, S.Y., & Murphy, L.R. (1996). Organizational health: A new paradigm for occupational stress research at NIOSH. Japanese Journal of Occupational Mental Health, 4, 248 254. Scott, B., & Colquitt, J. (2007). Are organizational justice effects bounded by individual
42 differences? An examination of equity sensitivity, exchange ideology, and the Big Five. Group and Organization Management, 32, 290 325. Shore, T., Sty, T., & Strauss, J. (2006 ). Leader responsiveness, equity sensitivity, and employee attitudes and behavior. Journal of Business and Psychology, 21, 227 241. Skarlicki, D. P., & Folger, R. (1997). Retaliation in the workplace: the roles of distributive, procedural, and interactional justice. Journal of Applied Psychology, 82, 434 443. S pector, P. E. (1988). Development of the Work Locus of Control Scale. Journal of Occupational Psychology, 61, 335 340. Spector, P., et al. (2002). Locus of control and well being at work: How generalizable are Western findings? Academy of Management Journ al, 45, 453 466. Spector, P., & Fox, S. (2002). An emotion centered model of voluntary work behavior: Some parallels between counterproductive work behavior and organizational citizenship behavior. Human Resource Management Review, 12, 269 292. Spector, P ., & Fox, S. (2005). The Stressor Emotion Model of Counterproductive Work Behavior. Washington, DC, US: American Psychological Association. Spector, P., & Jex, S. (1998). Development of four self report measures of job stressors and strain: Interpersonal Conflict at Work Scale, Organizational Constraints Scale, Quantitative Workload Inventory, and Physical Symptoms Inventory. Journal of Occupational Health Psychology, 3, 356 367. Spector, P., & O'Connell, B. (1994). The contribution of personality traits, negative
43 affectivity, locus of control and Type A to the subsequent reports of job stressors and job strains. Journal of Occupational and Organizational Psychology, 67, 1 12. Spector, P., Zapf, D., Chen, P., & Frese, M. (2000). Why negative affectivity s hould not be controlled in job stress research: Don't throw out the baby with the bath water. Journal of Organizational Behavior, 21, 79 95. Spielberger, C. D. (1988). The revised and expanded STAXI 2. Lutz FL: Psychological Assessment Resources. Stan ton, J. (1998). An empirical assessment of data collection using the Internet. Personnel Psychology 51 709 725. Storms, P. L., & Spector, P. E. (1987). Relationships of organizational frustration with reported behavioral reactions: The moderating effec t of locus of control. Journal of Occupational Psychology, 60, 227 234. Van Ryzin, G. (2004). The measurement of overall citizen satisfaction. Public Performance and Management Review, 27, 9 28. Vodanovich, S. J., Wallace, J. C., & Kass, S. J. (2005). A co nfirmatory approach to the factor structure of the boredom proneness scale: Evidence for a two factor short form. Journal of Personality Assessment, 85, 295 303. Weiner, B. (1980 ). A cognitive (attribution) emotion action model of motivated behavior: An an alysis of judgments of help giving. Journal of Personality and Social Psychology 39 186 200. Williams, L. J., & Anderson, S. E. (1991). Job satisfaction and organizational
44 commitment as predictors of organizational citizenship and in role behaviors. Jo urnal of Management, 17, 601 617. Yost, P.R. & Homer, L.E. (1998, April). Electronic versus p aper s urveys: Does the m edium a ffect the r esponse? Paper presented at the annual meeting of the Society for Industrial and Organizational Psychology. Dallas, TX. Zellars, K. Tepper, B., & Duffy, M. (2002 ). Abusive supervision and subordinates' organizational citizenship behavior. Journal of Applied Psychology 87 1068 1076. Zohar, D. (1995 ). The justice perspective of job stress. Journal of Organizational Beha vior 16 487 495.
45 Table 1 I ntercorrelations among Time 1 variables (lower triangle) and Time 2 variables (upper triangle). LOC EPQ Anger Hostile IC IJ JD OC OCB I OCB O CWB I CWB O OCB CWB LOC --.55 .35 .45 .32 .43 .03 .35 .36 .36 .32 .32 39 .33 EPQ .61 --.43 .40 .41 .41 .02 .31 .39 .54 .38 .42 .50 .41 Anger .36 .22 --.65 .50 .34 .07 .43 .28 .34 .52 .59 .34 .58 Hostile .40 .31 .62 --.48 .50 .01 .42 .28 .35 .44 .49 .34 .48 IC .25 .30 .20 .32 --.49 .33 .67 .09 .18 .61 .57 .15 .60 IJ .42 .35 .28 .44 .44 --.03 .36 .48 .53 .42 .38 .55 .41 JD .05 .13 .05 .05 .24 .07 --.52 .15 .14 .12 .06 .16 .08 OC .33 .19 .35 .36 .51 .43 .47 --.06 .17 .45 .46 .13 .46 OCB I .32 .35 .17 .27 .19 .36 .13 .10 --.69 .10 .15 .93 .12 OCB O .37 .45 .26 .25 .23 .48 .14 .15 .73 --.26 .34 .91 .31 CWB I .36 .39 .55 .46 .30 .30 .04 .29 .11 .29 --.86 .19 .95 CWB O .35 .40 .54 .51 .30 .29 .01 .30 .12 .31 .88 --.26 .98 OCB .37 .43 23 .28 .22 .45 .14 .14 .93 .92 .21 .22 --.23 CWB .36 .41 .56 .51 .31 .30 .02 .31 .12 .31 .96 .98 .23 --Notes. LOC= locus of control, EP= equity preference, Anger= trait anger, Hostile= trait hostility, IC= interpersonal conflict, IJ= i nter actional justice, JD= job demands, OC= organizational constraints. N =205 212. r > .12 p <. 05. r > .15, p < .01. r > .21, p <.001
46 Table 2 Correlations between Time 1 and Time 2 data. IC T2 JD T2 OC T2 IJ T2 LOC T2 EP T2 Hostile T2 Anger T2 OCB I T2 OC B O T2 CWB I T2 CWB O T2 OCB T2 CWB T2 IC .58 .14 .35 .42 .27 .34 .32 .23 .14 .18 .42 .38 .17 .41 JD .10 .59 .24 .09 .02 .00 .01 .04 .10 .12 .03 .02 .12 .00 OC .43 .27 .55 .47 .33 .19 .30 .27 .14 .17 .35 .28 .17 .31 IJ .50 .14 .46 60 .39 .34 .44 .33 .27 .34 .33 .32 .33 .32 LOC .31 .03 .33 .40 .77 .53 .42 .36 .36 .38 .29 .30 .40 .31 EP .36 .13 .23 .42 .52 .72 .28 .26 .35 .43 .33 .32 .42 .34 Hostile .50 .02 .37 .42 .45 .40 .76 .54 .27 .33 .37 .42 .32 .42 Anger .40 .04 .37 .22 .30 .34 .50 .73 .26 .31 .36 .41 .31 .40 OCB I .14 .09 .06 .37 .31 .37 .28 .23 .66 .46 .05 .07 .62 .05 OCB O .25 .09 .18 .49 .34 .50 .28 .31 .53 .61 .17 .21 .62 .19 CWB I .53 .10 .39 .34 .32 .47 .42 .52 .21 .39 .62 .55 .32 .60 CWB O .54 .07 .41 .35 .33 .48 .43 .51 .24 .41 .56 .61 .35 .61 OCB .21 .10 .13 .46 .35 .46 .30 .29 .64 .58 .11 .15 .67 .13 CWB .56 .08 .42 .37 .34 .50 .44 .54 .24 .42 .61 .61 .35 .63 Notes. T2= Time 2. LOC= locus o f control, EP= equity preference, Anger= trait anger, Hostile= trait hostility, IC= interpersonal conflict, IJ= interactional justice, JD= job demands, OC= organizational constraints. N =205 212. r > .12 p <. 05. r > .15, p < .01. r > .21, p <.001.
47 Table 3 Variable means, standard deviations, and correlations with supervisor report data. Time Mean SD OCB I T3 OCB O T3 CWB I T3 CWB O T3 OCB T3 CWB T3 IC 1 75.03 12.78 .26 .33 .39 .41 .31 .41 JD 1 45.34 16.01 .02 .01 .04 .02 .01 .03 OC 1 24.98 10.41 .16 .21 .37 .37 .19 .38 IJ 1 23.94 11.86 .37 .43 .30 .29 .41 .30 LOC 1 6.29 3.13 .37 .39 .30 .32 .40 .32 EP 1 22.17 5.67 .38 .40 .36 .39 .40 .38 Hostile 1 15.82 5.07 .34 .35 .45 .48 .36 .48 Anger 1 21.84 9.71 .23 .23 .41 .41 24 .42 OCB I 1 37.02 8.48 .63 .47 .13 .13 .57 .13 OCB O 1 39.66 7.94 .46 .47 .17 .17 .49 .18 CWB I 1 13.68 9.37 .26 .32 .58 .52 .30 .55 CWB O 1 22.68 14.95 .27 .34 .55 .58 .31 .58 OCB 1 76.67 15.26 .59 .51 .16 .16 .57 .16 CWB 1 36.45 2 3.83 .28 .35 .58 .58 .33 .59 IC 2 75.09 13.33 .23 .32 .56 .58 .29 .58 JD 2 45.96 15.92 .01 .04 .08 .07 .02 .08 OC 2 24.64 10.56 .18 .29 .40 .42 .24 .42 IJ 2 23.64 11.71 .36 .42 .32 .33 .41 .33 LOC 2 6.00 3.06 .37 .39 .32 .34 .40 .34 EP 2 22.52 5.48 .45 .50 .42 .46 .49 .46 Hostile 2 15.17 5.3 .37 .38 .39 .41 .39 .41 Anger 2 21.88 9.86 .26 .27 .42 .43 .27 .44 OCB I 2 36.3 8.84 .56 .49 .16 .17 .55 .17 OCB O 2 39.27 8.2 .46 .50 .26 .28 .50 .28 CWB I 2 13.33 9.23 .22 .29 .72 .66 .27 .70 CWB O 2 22.79 14.57 .20 .28 .65 .67 .25 .68 OCB 2 75.58 15.67 .56 .53 .23 .25 .57 .24 CWB 2 35.86 22.85 .21 .30 .70 .69 .27 .71 OCB I 3 38.25 9.23 --OCB O 3 39.72 8.94 .85 --CWB I 3 12.53 9.38 .39 .44 --CWB O 3 20.91 15.55 .41 .47 .92 --OCB 3 77.97 17.48 .96 .96 .43 .46 --CWB 3 33.43 24.46 .41 .47 .97 .99 .46 --Notes. T3= Time 3. LOC= locus of control, EP= equity preference, Anger= trait anger, Hostile= trait hostility, IC= interpers onal conflict, IJ= interactional justice, JD= job demands, OC= organizational constraints. N= 205 212. r > .12 p <. 05. r > .15, p < .01. r > .21, p <.001.
48 Table 4 Results from bootstrapped Sobel tests CWB T3 CWB I T3 CWB O T3 IV M Mean LB 95% CI UB 9 5% CI Mean LB 95% CI UB 95% CI Mean LB 95% CI UB 95% CI Anger IC .46 .19 .79 .17 .07 .29 .29 .12 .50 Anger IJ .14 .02 .30 .05 .01 .11 .09 .01 .20 Anger OC .27 .09 .48 .10 .04 .80 .18 .06 .32 EP IC .29 .14 .47 .10 .05 .17 .18 .08 .30 EP IJ .13 .01 .27 .05 .01 .11 .08 .01 .18 EP OC .13 .04 .24 .05 .01 .09 .08 .03 .15 Hostile IC .47 .23 .76 .18 .09 .29 .29 .13 .46 Hostile IJ .14 .02 .33 .06 .00 .14 .09 .01 .22 Hostile OC .22 .08 .41 .08 .03 .15 .14 .05 .26 LOC IC .31 .49 .16 .12 .18 .06 .19 .31 .11 LOC IJ .19 .35 .03 .07 .13 .02 .12 .22 .02 LOC OC .23 .40 .09 .08 .14 .03 .14 .25 .06 OCB T3 OCB I T3 OCB O T3 IV M Mean LB 95% CI UB 95% CI Mean LB 95% CI UB 95% CI Mean LB 95% CI UB 95% CI Anger IC .15 .28 .04 .06 .12 .01 .09 .17 .03 Anger IJ .14 .25 .03 .07 .13 .02 .07 .14 .02 Anger OC .11 .23 .01 .04 .09 .02 .08 .14 .02 EP IC .06 .14 .00 .02 .06 .01 .04 .09 .01 EP IJ .13 .23 .06 .06 .11 .02 .07 .12 .04 EP OC .04 .09 .0 1 .01 .04 .01 .03 .06 .01 Hostile IC .11 .26 .04 .03 .10 .05 .08 .15 .00 Hostile IJ .20 .32 .09 .09 .15 .04 .11 .18 .05 Hostile OC .07 .17 .02 .02 .06 .03 .06 .11 .01 LOC IC .08 .00 .16 .03 .01 .07 .05 .01 .01 LOC IJ .17 .0 8 .26 .08 .03 .13 .09 .04 .14 LOC OC .06 .02 .14 .01 .02 .06 .04 .01 .06 Notes. IV= independent variables measure at Time 1, M= mediating variables measured at time 2, T3= time 3, Anger= trait anger, EP= equity preference, Hostile= trait hostility, LO C= locus of control, IC= interpersonal conflict, IJ= interactional justice, OC= organizational constraints, LB 95% CI= lower bound 95% confidence interval, UP 95% CI= upper bound 95% confidence interval. Gray cells indicate that the confidence interval i ncludes zero.
49 Figure 1. Job Stressor Mediated Model of OCB and CWB. Personality Variables Trait Anger Trait Hostility Locus of Control Equity Sensitivity Job Stressors Interpersonal Conflict Organizational Constraints Job Demands Voluntary Behaviors OCB CWB
50 Figure 2. Job Stressor Mediated Model of Target Based OCB and CWB. Personality Variables Trait Anger Trait Hostility Locus of Control Equity Sensitivity Interpersonal Stressors Interpersonal Conflict Voluntary Behaviors OCB I CWB I Organizational Stressors Organizational Constraints Job Demands Voluntary Behaviors OCB O CWB O
52 Appendix A Self Report Survey
53 Appendix A continued
54 Appendix A continued
55 Appendix A continued
56 Appendix A continued
57 Appendix A continued
58 Appendix A continued
59 Appendix A continued
60 Appendix A continued
61 Appendix A continued
62 Appendix A continued
63 Appendix A conti nued
64 Appendix A continued
65 Appendix A continued
66 Appendix A continued
67 Appendix B Supervisor Report Survey
68 Appendix B continued
69 Appendix B continued
About the Author ee in Honors Psychology and Sociology from the University at Albany, State University of New York in 2002. She was awarded a Presidential Fellowship to attend the Ph.D program at the University of rved as Vice President of the Graduate and Professional Student Organization as well as the Psychology Graduate Student Organization. She has also coauthored four publications in peer reviewed journals, a book chapter, and approximately 20 conference pres entations.