The Motivational Effects of Diminished Self- Esteem for ...media.terry.uga.edu/socrates/publications/2015/01/...supervision indirectly influences employees’ workplace deviance and

Journal of ManagementVol. XX No. X, Month XXXX 1 –34

DOI: 10.1177/0149206314566462© The Author(s) 2015

Reprints and permissions:sagepub.com/journalsPermissions.nav

1

The Motivational Effects of Diminished Self-Esteem for Employees Who Experience Abusive

Supervision

Ryan M. VogelPennsylvania State University–Erie

Marie S. MitchellUniversity of Georgia

This research considers two theoretical perspectives on employees’ motivation associated with diminished self-esteem from abusive supervision. The self-defense view of diminished self-esteem suggests that abusive supervision motivates destructive behavior in an attempt to reas-sert personal control and protect victims’ self-image. The self-presentational view of diminished self-esteem suggests abusive supervision motivates behavior that attempts to signal fit with and value to the workgroup and organization. On the basis of these two theoretical perspectives, we examine how employees’ diminished self-esteem from abusive supervision can motivate destruc-tive work behavior (i.e., supervisor-directed deviance, organizational deviance) and self-presen-tational behavior (i.e., putting on a façade, ingratiation). Additionally, employees’ turnover intentions, which are an indicator of employees’ psychological detachment from the organiza-tion, are considered a moderator of the effects of abusive supervision on diminished self-esteem and associated behavior such that high turnover intentions attenuate the effects. Results of two field studies and a daily diary study support the hypothesized model and show that abusive supervision indirectly influences employees’ workplace deviance and self-presentational behav-ior via diminished self-esteem. As predicted, the effects are stronger for employees with lower versus higher turnover intentions.

Keywords: abusive supervision; self-esteem; diminished self-esteem; workplace deviance; façade; ingratiation

Acknowledgments: The authors would like to thank Associate Editor Michelle Duffy and two anonymous reviewers for their helpful suggestions and supportive guidance throughout the review process.

Corresponding author: Ryan M. Vogel, Department of Management, Sam and Irene Black School of Business, Pennsylvania State University, 5101 Jordan Road, Erie, PA, 16563, USA.

E-mail: [email protected]

566462 JOMXXX10.1177/0149206314566462Journal of ManagementVogel, Mitchell / Diminished Self-Esteem From Abuseresearch-article2014

at UNIV OF GEORGIA LIBRARIES on January 16, 2015jom.sagepub.comDownloaded from

mailto:[email protected]

http://jom.sagepub.com/

2 Journal of Management / Month XXXX

There are a variety of negative consequences associated with abusive supervision, defined as employees’ perceptions of supervisors’ sustained hostile verbal and nonverbal behavior (Tepper, 2000). For example, victims of abusive supervision are less satisfied with their jobs, experience greater stress and strain, and are likely to engage in fewer productive and greater counterproductive behaviors in the workplace (e.g., Mitchell & Ambrose, 2007; Tepper; Zellars, Tepper, & Duffy, 2002). Recently, studies have found that one reason why employees react destructively to abusive supervision is because they experience a diminished sense of self (e.g., diminished self-esteem, ego depletion; Ferris, Spence, Brown, & Heller, 2012; Thau & Mitchell, 2010). Self-esteem is “an internal, subjective ‘marker’ that reflects an ongoing assessment of the individual’s inclusionary status” or sense of value within a group (Leary, 1990: 226). Because abusive supervision suggests the supervisor does not value the relationship with or accept the targeted employee (Baumeister, Smart, & Boden, 1996; Leary; Smart Richman & Leary, 2009), victims of abusive supervision experience diminished self-esteem (Burton & Hoobler, 2006). Although research within the organizational sciences has demonstrated destructive behavior is a likely reaction to diminished self-esteem from abuse (e.g., Ferris et al.; Thau & Mitchell), two streams of social psychology research have demon-strated that diminished self-esteem can promote very different types of behavioral reactions.

One stream of research, which we term the self-defense view of diminished self-esteem, suggests that when individuals experience diminished self-esteem from abuse, they become motivated to protect and defend their sense of self (Baumeister et al., 1996). In particular, they engage in destructive behaviors toward the source of diminished self-esteem because these acts allow victims to reassert their self-image and demonstrate control over the situa-tion (Baumeister et al.; Miller, 2001; Tedeschi, 2001). Consistent with these ideas, organiza-tional research (e.g., Ferris, Brown, Lian, & Keeping, 2009; Ferris et al., 2012) has shown that employees who experience a diminished sense of self from negative work experiences, such as abusive supervision, engage in deviant work behavior.

Another stream of research (Leary & Downs, 1995; Maner, DeWall, Baumeister, & Schaller, 2007), which we term the self-presentational view of diminished self-esteem, pro-poses that diminished self-esteem from abuse can motivate behavior intended to present the victim as a good fit with others in the social group, called self-presentational behavior. Diminished self-esteem from abusive and exclusionary behavior (i.e., abusive supervision) drives individuals to attempt to fortify their value to the group by engaging in acts that sug-gest to others that the target of abuse is valued and worthy of inclusion (Leary & Downs). Given these ideas, it is possible for abusive supervision and its negative effects on employ-ees’ self-esteem to motivate self-presentational behavior in addition to destructive behavior, such as deviance.

Both theoretical perspectives of self-esteem (Baumeister & Leary, 1995; Leary, 1990), however, suggest that employees’ detachment from their organization may be an important consideration in determining how damaging abusive supervision is to the sense of self. For individuals who have psychologically detached from the social group, treatment from group members does not exert a strong influence on the sense of self (Bowins, 2004). Consequently, we also consider how employees’ detachment from their organization (via turnover inten-tions; Burris, Detert, Chiaburu, 2008; Kahn, 1990) moderates and weakens the effects of abusive supervision on employees’ self-esteem and behavior.



Vogel, Mitchell / Diminished Self-Esteem From Abuse 3

The goal of our study is to present and test an expanded theoretical model on the effects of diminished self-esteem from abusive supervision (see Figure 1). Specifically, we con-sider diminished self-esteem as a mediator of the relationships between abusive supervision and victims’ behaviors. We draw from two streams of social psychology research on self-esteem (e.g., Baumeister & Leary, 1995; Leary, 1990; Leary & Downs, 1995) to explain that diminished self-esteem from abusive supervision motivates both destructive work behavior (i.e., workplace deviance) and self-presentational behavior (i.e., façade, ingratiation). Furthermore, drawing on the notion that employees with high turnover intentions have psy-chologically detached themselves from the organization (Burris et al., 2008), we examine whether abusive supervision exhibits similar diminishing effects on self-esteem and associ-ated behavioral reactions for employees who have psychologically detached from the organization.

Our work contributes to the literature in several ways. First, our research adds to the litera-ture that has examined the psychological and behavioral consequences of abusive supervi-sion (e.g., Lian, Brown, Ferris, Liang, Keeping, & Morrison, 2014; Tepper, Henle, Lambert, Giacalone, & Duffy, 2008; Thau & Mitchell, 2010). The standing literature has primarily focused on counterproductive behaviors associated with abusive supervision that may be ultimately self-defeating to the victim in the long run. Our study adds to knowledge by con-sidering how diminished self-esteem from abusive supervision may motivate victims to engage in behavior aimed at strengthening social bonds with other workgroup members. Second, our approach takes a more united and integrative view to understand the motivations of diminished self-esteem. Although prior work in social psychology has examined destruc-tive/defensive and self-presentational reactions separately, Leary (1990) called for a more integrative approach. Our work meets this call and explains how different behavioral reac-tions to diminished self-esteem from abusive supervision are possible. Third, we consider individuals’ psychological detachment from the organization as an important boundary

Figure 1Theoretical Model

AbusiveSupervision

Self-Esteem

Workplace Deviance

Self-Presentational

Behavior

TurnoverIntentions




condition influencing the extent to which abusive supervision negatively affects self-esteem and indirectly influences the strength of victims’ behavioral reactions. Similar to existing abusive supervision research on the influence of turnover intentions (Tepper, Carr, Breaux, Geider, Hu, & Hua, 2009), we suggest a high level of psychological detachment is empower-ing to victims of abusive supervision. However, our research departs from the standing litera-ture by explaining how psychological detachment allows victims to more easily deflect the impact of a supervisor’s abusive behavior because the supervisor is not an important source to these victims’ self-esteem.

Theoretical Overview

Employees’ self-esteem is particularly susceptible to treatment received from supervisors (Brockner, 1988). This is, in part, because supervisors control and distribute many resources (e.g., performance appraisals, promotions, bonuses, preferred assignments, feedback) that are important to employees. More directly, however, this is because supervisor treatment signals whether employees are perceived as valuable and effective contributors to the orga-nization (Brockner; Burton & Hoobler, 2006). Social psychology research has long found that receiving positive and favorable treatment from others in the social group suggests that the person is valued, which increases self-esteem. In contrast, being criticized, ridiculed, bul-lied, or abused by others suggests that the person is not valued and is perceived as a bother-some group member, which diminishes self-esteem (Leary, 2010; Leary, Tambor, Terdal, & Downs, 1995; Leary, Twenge, & Quinlivan, 2006; K. D. Williams, Cheung, & Choi, 2000).

Abusive supervision is persistent supervisor hostility toward employees and, as such, sig-nals a deteriorating work relationship (Tepper & Almeda, 2011) and communicates to vic-tims that they are a source of aggravation to the supervisor (Tepper, Moss, & Duffy, 2011). Consequently, abusive supervision marginalizes employees (Ashforth, 1997) and under-mines their sense of value and worth to the workgroup, which, according to theory (Leary, 2010), diminishes their self-esteem. Given the importance of supervisors’ interactions to employees’ self-worth (Brockner, 1988; Burton & Hoobler, 2006), abusive supervision should have a strong, negative effect on employees’ self-esteem.

The Influence of Diminished Self-Esteem From Abuse on Employees’ Behaviors

Scholars generally agree that all human beings wish to avoid being treated poorly because mistreatment suggests that the person is not of value to the social group (Baumeister & Tice, 1985; Leary, 1990). Consequently, victims’ sense of self-worth diminishes, which motivates attempts to strengthen or reinforce their self-esteem. Theories differ, though, on exactly how individuals seek to strengthen and fortify their self-esteem. The self-defense view suggests that diminished self-esteem from abuse motivates destructive and deviant behaviors that demonstrate that victims are standing up to the abuse (Baumeister, 1997). The self-presenta-tional view suggests that diminished self-esteem from abuse motivates efforts from victims to signal they are a valuable member of the social group (Leary). We integrate these differing theoretical perspectives into an overall theoretical model of diminished self-esteem, thereby providing “a broader analysis” (Leary, 1990: 221) on the motives and responses to dimin-ished self-esteem associated with abusive supervision. We elaborate on each view below.




The self-defense view of diminished self-esteem. The self-defense view proposes that abused employees become motivated to protect their sense of self through destructive, aggressive, or deviant acts against the source of diminished self-esteem (Baumeister, 1997; Baumeister et al., 1996). On the face of things, these hostile behaviors seem self-defeating because they are likely to provoke further abusive treatment from others (e.g., Baumeis-ter; Thau, Aquino, & Poortvliet, 2007). However, victims with diminished self-esteem who engage in destructive behavior often do not consider the potential costs of their actions (Baumeister et al.; Leary et al., 2006). Instead, their focus is on defending and boosting their self-worth by discrediting the threat to their self-esteem and standing up for themselves (Baumeister; Crocker & Park, 2004). In so doing, destructive behaviors offer a cathartic, self-gratifying response to the aversive state of diminished self-esteem from abuse (Baumeister et al.), allowing victims to assert some control over the situation (Tedeschi, 2001).

Because responding to abusive treatment with destructive behavior allows victims to pro-tect their self-worth (e.g., Baumeister, 1997; Tedeschi, 2001), it is possible that victims of abusive supervision who experience diminished self-esteem will engage in workplace devi-ance, defined as behaviors that violate organizational norms and are harmful to the organiza-tion and its members (Robinson & Bennett, 1995). For example, Chen, Ferris, Kwan, Yan, Zhou, and Hong (2013) investigated the role of workplace incivility on employees’ perfor-mance and found that incivility sets in motion a self-disengagement process that negatively affects victims’ work performance. Chen et al. concluded that incivility threatens victims’ sense of self, which motivates them to behaviorally disengage. Furthermore, Ferris and col-leagues (Ferris, Brown, & Heller, 2009; Ferris et al., 2012) found that diminished esteem within the work environment mediated the effects of supervisors’ interactions with subordi-nates on subordinates’ workplace deviance. Conversely, they found more favorable interac-tions (i.e., high leader-member exchange, high interactional justice) enhanced employees’ self-worth, which reduced workplace deviance.

Altogether, these research findings and the self-defense view suggest that victims who experience diminished self-esteem from abusive supervision will engage in workplace devi-ance targeted toward the supervisor, such as supervisor-directed deviance and organizational deviance. This is because researchers have found that victims of abusive supervision who direct deviance at their supervisor do so in outrage and as a form of retaliation (e.g., Mitchell & Ambrose, 2007; Tepper & Almeda, 2011). Furthermore, researchers have argued that orga-nizational deviance serves as a form of retaliation because these employees are actively try-ing to damage the organization that the supervisor represents (e.g., Tepper & Almeda; Tepper et al., 2008). Thus, diminished self-esteem will mediate the effects of abusive supervision on supervisor-directed deviance and organizational deviance.

The self-presentational view of diminished self-esteem. The self-presentational view (Leary, 1990; Leary & Downs, 1995) suggests that diminished self-esteem from abuse moti-vates individuals to attempt to protect and reinforce their sense of self-worth by demonstrat-ing value to and fit with the social group (e.g., Maner et al., 2007; Wolfe, Lennox, & Cutler, 1986). These self-presentational behaviors have the potential to fortify self-esteem because they are intended to signal to others that the victim belongs in the group and is able to build social bonds with others (Baumeister & Leary, 1995; Leary & Downs). Two types of self-presentational behaviors have been highlighted in the literature: victims’ acting like they embrace the social group’s values and strategic behavior aimed at making the victim seem more likeable.




To begin, diminished self-esteem motivates individuals to act like they embrace (and not begrudge) group values (Leary, 1995). By conforming to group values, even if those values are perceived to be incompatible with the victim’s own values, it is possible to build social bonds with group members and foster inclusion in the group (K. D. Williams et al., 2000). Not outwardly conforming to group values is a risky strategy for those who have diminished self-esteem, as doing so might strengthen others’ perceptions that the person does not belong, inducing further exclusionary and hostile behavior from group members (Baumeister, 1997). In addition, those with diminished self-esteem may self-censor idiosyncratic thoughts and opinions because suppressing characteristics and ideals dissimilar from those of the group may signal to others that the person fits in (Leary; Leary & Allen, 2011). Thus, the self-pre-sentational view of diminished self-esteem suggests that diminished self-esteem may moti-vate the adaptive behaviors of acting like the person embraces group values and withholding inconsistent group ideals (Leary & MacDonald, 2003).

Research has also shown that diminished self-esteem from abusive treatment prompts victims to engage in strategic self-presentational behaviors (Leary, 2001). In particular, abused individuals who experience diminished self-esteem try to protect and fortify their sense of self by presenting themselves as more likeable to others in an attempt to gain their acceptance (Leary & Allen, 2010; Leary & Downs, 1995). For example, individuals moti-vated by a desire to repair diminished self-worth engage in personal flattery or favor doing for others (Leary & Allen) because these types of behaviors are often effective in enhancing others’ interpersonal attraction to and acceptance of the actor (Gordon, 1996).

Consistent with the self-presentational view is our examination of two types of workplace behavior akin to the self-presentational behaviors described above: façade and ingratiation. Façade behavior allows victims of abuse to give the impression that they share the social group’s values. Defined as “false representations created by employees to appear as if they embrace organizational values” (Hewlin, 2003: 634), façade includes consistent efforts to modify behavior to conform to the values of the workgroup and suppress aspects of the true self (e.g., not sharing personal thoughts or feelings with others, saying things that one does not really believe, behaving consistently with organizational values, withholding personal values that conflict with those of the organization; Hewlin, 2009). Given the motivational effects of diminished self-esteem, it is possible that employees who experience diminished self-esteem from abusive supervision engage in façade as an attempt to better fit with and become accepted by members of the workgroup.

Ingratiation is a strategic form of impression management behavior that aligns with the self-presentational view (Leary & Allen, 2010); ingratiation is self-presentation behavior performed with the intention of being perceived as likeable (Jones & Pittman, 1982). Employees may engage in ingratiation as a way of enhancing diminished self-esteem because they want to be viewed as more attractive to the workgroup, particularly when they feel the need to protect themselves against attacks or negative feedback from others (Liden & Mitchell, 1988). Theorists (Liden & Mitchell; Tedeschi & Melburg, 1984) have argued that ingratiation is driven by a desire to attain positive future interactions with others in the work-group. Ingratiation is likely to be used by abused employees with diminished self-esteem to be viewed by others as likeable in an attempt to be accepted and included.

Therefore, the self-presentational view of diminished self-esteem suggests that abusive supervision will negatively influence employees’ self-esteem, which will then motivate




self-presentational behavior—behavior aimed at demonstrating the person holds the same values as others in the workgroup (façade) and is likeable (ingratiation). In line with the self-presentational view, self-esteem will mediate the effects of abusive supervision on these self-presentational behaviors.

The Attenuating Effects of Psychological Detachment

Both the self-defense and self-presentational views of diminished self-esteem suggest that the extent to which abusive treatment affects victims’ sense of self is contingent on how psy-chologically detached the victim is from the social group (Baumeister & Leary, 1995; Leary, 1995). Because assessments of individuals’ self-esteem derive from interactions with mean-ingful others, the degree to which self-esteem is affected by group members’ treatment of them depends on whether the person desires to be part of the group (Leary, 2001). Indeed, Bowins pointed out that “the pain of conflict seems irrelevant” (2004: 3) for individuals who have detached from the source of adversity. Accordingly, psychological detachment protects individuals’ self-worth from the treatment of group members. This is not to say that hostile treatment from group members (such as an abusive supervisor) may not trigger negative reactions; rather, individuals are simply more able to deflect mistreatment when they have psychologically detached from the group.

These ideas are consistent with organizational research on psychological detachment, which refers to employees’ mental disengagement from work (Sonnentag & Bayer, 2005; Tett & Meyer, 1993). Psychological detachment amounts to mentally “checking out” from stressful and offending situations at work. It is a strategy used to divert stress, allowing the individual to avoid or deflect antagonizing and aversive effects of work life. Employees who psychologically detach from work while off the job experience improved emotional states and less fatigue associated with negative work occurrences (e.g., Sonnentag & Bayer; Sonnentag, Binnewies, & Mojza, 2010). For example, the effects of workplace bullying on psychological strain have been found to be weaker for employees who have psychologically detached from work (Moreno-Jiménez, Rodríguez-Muñoz, Pastor, Sanz-Vergel, & Garrosa, 2009).

Psychological detachment while at work occurs when employees hold high intentions to quit (Burris et al., 2008; Griffeth, Hom, & Gaertner, 2000; Tett & Meyer, 1993). High turn-over intentions signify that employees no longer feel a sense of personal engagement with the organization. The detachment experience manifests in employees becoming “psychically uninvolved in tasks, cognitively unvigilant, and emotionally disconnected from others in a way that hides what they think and feel, their creativity, their beliefs and values, and their personal connections with others” (Kahn, 1990: 702). It is for these reasons that psychologi-cal detachment (viz., turnover intentions) often results in voluntary turnover behavior (Tett & Meyer). Additionally, given the disconnect employees have with their organization when they psychologically detach, these employees engage in less helpful and developmental behaviors for the organization (Burris et al.; Sagie, Birati, & Tziner, 2002) and are relatively impervious to treatment received from supervisors (Collins, Mossholder, & Taylor, 2012).

Along these lines, Tepper et al. (2009) theorized that turnover intentions are a source of empowerment for employees who experience abusive supervision. Because these employees have mentally “checked out,” hostile treatment from focal organizational members, such as supervisors, communicates little value (Kahn, 1990). Accordingly, employees with high




turnover intentions are less dependent on the supervisor, which empowers victims of abusive supervision because interpersonal treatment received from their supervisor is of little value (Tepper et al.). Underscoring this point, Collins et al. (2012) concluded that supervisory treatment is not a critical concern for those who have psychologically detached because these employees do not rely on nor do they need to leverage how current interactions with their supervisors might affect their future work circumstances. Thus, principles of psychological detachment suggest that employees who intend to leave their organization will hold little value in how their supervisor treats them. Consequently, psychological detachment should attenuate the effects of abusive supervision on self-esteem.

Given the motivational effects of diminished self-esteem on employees’ behaviors, the indirect effects of abusive supervision on victims’ behaviors should vary as a function of their psychological detachment. The extent to which victims will engage in self-defensive or self-presentational acts will depend on the extent to which abusive supervision diminishes their self-esteem. For victimized employees who intend to leave the organization, the effects of abuse will be weaker relative to those who intend to remain in the organization. Consequently, the strength of employees’ motivation to engage in various behaviors will indirectly depend on their level of detachment from the organization.

Research provides some support for our arguments. For example, in a field study of work-ing employees, Burris et al. (2008) argued that employees who have psychologically detached from the organization would display different behavioral patterns than those who are psycho-logically attached to the organization. They found that employees who intended to leave the organization were less likely to engage in proactive behaviors intended to benefit the organi-zation than their attached counterparts. Furthermore, in a series of experiments, Maner et al. (2007) found that subjects who experienced abusive treatment at the hands of others put in less effort to build and maintain social bonds with those others when future exchanges with them were less likely. Maner et al. reasoned that abused individuals’ motivation to fit in is targeted only at those with whom there is a realistic possibility of future interaction.

In sum, employees find little value in how their supervisor treats them when they have psychologically detached and intend to leave the organization. Given this, high turnover intentions will weaken the effects of abusive supervision on employees’ self-esteem and thereby diminish these employees’ motivation to protect their self-worth. Conversely, given the importance of future interactions with the supervisor when turnover intentions are lower, abusive supervision will have a stronger effect on employees’ self-esteem and indirectly bolster their motivation to protect the sense of self via self-defensive behavior (i.e., supervi-sor-directed deviance, organizational deviance) and self-presentational behavior (i.e., façade, ingratiation).

Hypothesis 1: The indirect effect of abusive supervision on (a) workplace deviance and (b) self-presentational behavior (through the mediator, self-esteem) will be moderated at the first stage by psychological detachment (i.e., turnover intentions), such that the indirect effects will be stronger when psychological detachment is lower versus higher.

Overview of Studies

To evaluate our hypotheses, we conducted three field studies of individuals working in full-time jobs across multiple organizations. Study 1 allows for a test of our model on one




form of workplace deviance (organizational deviance) and one form of self-presentational behavior (façade). Study 2 provides a constructive replication of our first study and extends the model by employing a measure of state self-esteem and including multiple forms of workplace deviance (i.e., supervisor-directed deviance, organizational deviance) and self-presentational behavior (i.e., façade, ingratiation). Study 3 provides stronger evidence for the causal order of our model, given its daily diary design. In Study 3, we assess employees’ daily reports of abusive supervision, self-esteem, workplace deviance, and self- presentational behavior to test the within-individual effects of the model. Because replicating complex interaction effects in field studies is often difficult (Aguinis, 1995), our multistudy design allows for a robust test of our hypotheses.

Study 1 Method

Sample and Procedure

Participants were recruited via the StudyResponse project (http://studyresponse.syr.edu), which is an online data collection service of working adults. Participants earned Amazon.com credit for their participation in two surveys. Participants reported on demographics, abusive supervision, and self-esteem at Time 1 and, about one month later, reported on their turnover intentions, organizational deviance, and façade at Time 2. StudyResponse sent an invitation to participate in our study from their panel; a total of 570 individuals consented to participate and completed the Time 1 survey. Our analyses included a total of 172 observa-tions with complete data for our study’s variables across the Time 1 and Time 2 surveys. Results of analysis of variance (ANOVA) tests demonstrated no significant differences on demographics or Time 1 variables (i.e., abusive supervision, self-esteem) existed between the Time 2 responders and nonresponders (Dooley & Lindner, 2003). Seventy-three percent of the sample was female, and 91% of the sample identified as White/Caucasian. On average, participants’ age was 41.6 years (SD = 11.1) and company tenure was 7.6 years (SD = 7.7). Participants came from a variety of industries (finance/insurance, health care, and education were the three most frequently cited), and 66.9% of participants were in nonsupervisory positions.

Measures

Abusive supervision. Abusive supervision was assessed using Mitchell and Ambrose’s (2007) five-item short version of Tepper’s (2000) measure. Employees’ perceptions of super-visor abuse (e.g., “My boss puts me down in front of others”) were measured on a 5-point scale from 1 (strongly disagree) to 5 (strongly agree).

Turnover intentions. Turnover intentions were assessed with Cropanzano, James, and Konovsky’s (1993) three-item measure. Respondents indicated their intentions to leave the organization (e.g., “I intend to leave this organization within the next year”) on a 5-point scale ranging from 1 (strongly disagree) to 5 (strongly agree).

Self-esteem. Self-esteem was assessed with Rosenberg’s (1965) 10-item measure. Respondents were asked to rate how they currently felt with respect to each statement (e.g.,


http://studyresponse.syr.edu



“I feel comfortable with myself”) on a 5-point scale from 1 (strongly disagree) to 5 (strongly agree).

Workplace deviance. Workplace deviance was assessed with Bennett and Robinson’s (2000) 12-item organizational deviance measure. Respondents indicated how often they engaged in each of the behaviors (e.g., “Taken an additional or longer break than is accept-able”) on a 5-point scale ranging from 1 (never) to 5 (always).

Façade. Façade was assessed with Hewlin’s (2009) five-item measure. Respondents indicated their agreement with statements regarding façade behavior on a 5-point scale from 1 (strongly disagree) to 5 (strongly agree; e.g., “I behave in a manner that reflects the organi-zation’s value system even though it is inconsistent with my personal values”).

Control variables. Previous research has found that minority status is associated with an increased likelihood of engaging in façade (Hewlin, 2009) and deviance (Aquino & Douglas, 2003). Thus, in our analyses, we controlled for employees’ gender (1 = male, 2 = female) and race (1 = White/Caucasian, 2 = Non-White/Caucasian).

Study 1 Results and Discussion

Descriptive Statistics, Correlations, and Confirmatory Factor Analyses

Table 1 shows the means, standard deviations, and correlations among the variables of Study 1. To assess the appropriateness of our measurement model, we conducted a series of confirmatory factor analyses (CFAs) using Mplus Version 7.1. Because the intention of the CFA was to determine whether the study variables were distinct for the purpose of hypothesis testing and because the number of items was very large relative to the sample size (Little, Rhemtulla, Gibson, & Schoemann, 2013), we created parcels of items for several of the con-structs with large numbers of items (e.g., self-esteem, deviance).1 We created three parcels

Table 1

Descriptive Statistics and Correlations Among Variables in Study 1

Variable 1 2 3 4 5 6 7 M SD

1. Abusive supervision (.93) 1.50 0.762. Turnover intentions .22 (.85) 2.48 1.093. Self-esteem −.40 −.20 (.89) 3.99 0.664. Façade .25 .36 −.32 (.80) 2.70 0.775. Organizational deviance .29 .15 −.39 .28 (.85) 1.47 0.456. Gendera .02 −.22 −.07 −.14 −.12 — 1.73 0.457. Raceb −.04 .12 −.07 −.04 −.05 .02 — 1.09 0.29

Note: N = 172. Correlations greater than |.19| are significant at p < .01 and those greater than |.14| are significant at p < .05, two-tailed. Reliability estimates (coefficient alphas) of scales are reported along the diagonal in parentheses.aMale = 1, female = 2.bWhite/Caucasian = 1, non-White/Caucasian = 2.




for each construct by successively assigning high and low loading items onto each of the three parcels (Little, Cunningham, Shahar, & Widaman, 2002).

The CFA results showed that our hypothesized, five-factor model provided a good fit to the data, χ2(142) = 207.69, comparative fit index (CFI) = .96, root mean square error of approximation (RMSEA) = .06, standardized root mean square residual (SRMR) = .07. This model provided a better fit to the data than the alternatives we tested, including (a) a four-factor model in which abusive supervision and self-esteem items were specified to load onto one factor, χ2(146) = 456.26, CFI = .82, RMSEA = .12, SRMR = .11, Δχ2(4) = 248.57, p < .001; (b) a four-factor model in which abusive supervision and workplace deviance items were specified to load onto one factor, χ2(146) = 422.94, CFI = .84, RMSEA = .11, SRMR = .11, Δχ2(4) = 215.25, p < .001; (c) a four-factor model in which self-esteem and workplace deviance items were specified to load onto one factor, χ2(146) = 472.64, CFI = .81, RMSEA = .12, SRMR = .11, Δχ2(4) = 264.95, p < .001; (d) a four-factor model in which self-esteem and façade items were specified to load onto one factor, χ2(146) = 421.76, CFI = .84, RMSEA = .11, SRMR = .11, Δχ2(4) = 214.07, p < .001; (e) a four-factor model in which workplace deviance and façade items were specified to load onto one factor, χ2(146) = 431.95, CFI = .83, RMSEA = .11, SRMR = .10, Δχ2(4) = 224.26, p < .001; and (f) a one-factor model, χ2(152) = 1,096.69, CFI = .45, RMSEA = .20, SRMR = .17, Δχ2(10) = 889.0, p < .001. Thus, our measures showed acceptable measurement properties for hypothesis testing.

Hypothesis Testing

We tested our hypotheses in a path analytic framework for mediated models, adapted to include moderated regression using procedures for moderated mediation (MacKinnon, Fritz, Williams, & Lockwood, 2007; Preacher, Rucker, & Hayes, 2007; see also MacKinnon, Lockwood, Hoffman, West, & Sheets, 2002; Shrout & Bolger, 2002). To test the indirect effects, we employed the MacKinnon et al. (2007) RMediation methodology (Tofighi & MacKinnon, 2011). This method computes 95% asymmetric bias-corrected confidence inter-vals (CIs) for the indirect effects using the distribution-of-the-product method. Furthermore, it takes into account the coefficient and standard error for the terms representing the effects from the independent variable to the mediator and from the mediator to the dependent vari-able. MacKinnon et al. (2007) found this method adequately accounts for the chance of Type I error and provides a more accurate estimation of the indirect effect than other methods, such as the Sobel test. The moderated indirect effects were computed at 1 SD above and below the mean of the moderator (i.e., turnover intentions) using the RMediation method. Predictor variables were mean centered (Cohen, Cohen, West, & Aiken, 2003), and the direct effects of abusive supervision were entered into equations for the dependent variables (Preacher et al.).

Table 2 shows the path estimates of the hypothesized model. Hypothesis 1 predicted the indirect effect of Abusive Supervision × Turnover Intentions on (a) workplace deviance and (b) self-presentational behavior (through the mediator, self-esteem) would be stronger when turnover intentions are lower versus higher. Supporting the first stage of our model, the results show the Abusive Supervision × Turnover Intentions interaction term was significant on self-esteem (b = 0.18, p < .001), and the relationship between abusive supervision and self-esteem was stronger when turnover intentions were lower (b = −0.58, p < .001) than higher (b = −0.19, p < .05). This is shown in Figure 2a.



12

Tab

le 2

Pat

h A

nal

ysis

Res

ult

s fo

r S

tud

y 1

Sel

f-E

stee

mF

açad

eO

rgan

izat

iona

l Dev

ianc

e

S

tep

1S

tep

2S

tep

1S

tep

2S

tep

1S

tep

2

Var

iabl

esb

SEb

SEb

SEb

SEb

SEb

SE

Gen

dera

−0.

110.

11−

0.16

0.10

−0.

25*

0.12

−0.

28*

0.12

−0.

130.

08−

0.15

*0.

07R

aceb

−0.

170.

18−

0.18

0.17

−0.

060.

18−

0.12

0.19

−0.

060.

11−

0.10

0.10

Abu

sive

sup

ervi

sion

−0.

34**

*0.

07−

0.38

***

0.07

0.26

***

0.08

0.14

0.10

0.17

**0.

060.

090.

05T

urno

ver

inte

ntio

ns−

0.04

0.05

−0.

080.

05

Abu

sive

Sup

ervi

sion

× T

urno

ver

Inte

ntio

ns0.

18**

*0.

07

Sel

f-es

teem

−0.

33**

0.10

−0.

23**

*0.

06R

2 Equ

atio

n.1

8**

.24*

**.0

9**

.15*

**.1

0*.2

0***

ΔR

2.0

6**

.06*

.10*

*

Not

e: N

= 1

72. T

able

val

ues

are

unst

anda

rdiz

ed p

ath

esti

mat

es a

nd s

tand

ard

erro

rs f

rom

the

esti

mat

ed m

odel

. Sta

tist

ical

sig

nifi

canc

e is

bas

ed o

n a

two-

tail

ed te

st.

a Mal

e =

1, f

emal

e =

2.

b Whi

te =

1, n

on-W

hite

= 2

.*p

< .0

5.**

p <

.01.

***p

< .0

01.




To determine whether the interaction carried through self-esteem to the dependent vari-ables, we decomposed the effects of abusive supervision into simple effects representing the first stage and second stage of our model at high and low levels of turnover intentions. The simple effects of self-esteem were significant on organizational deviance (b = −0.23, p < .001) and façade (b = −0.33, p < .01). After we controlled for the direct effects of abusive supervision (organizational deviance: b = 0.09, n.s.; façade: b = 0.14, n.s.), the RMediation results indicated that the indirect effect of abusive supervision on organizational deviance was stronger when turnover intentions were lower (ρ = .14, 95% CI = [.06, .25]) than higher (ρ = .04, 95% CI = [.01, .11]), supporting Hypothesis 1a. For façade, the results indicated that the indirect effect of abusive supervision was stronger when turnover intentions were lower (ρ = .19, 95% CI = [.07, .35]) than higher (ρ = .06, 95% CI = [.01, .16]), supporting Hypothesis 1b. Given the nonsignificant direct effects, the effects of abusive supervision on the depen-dent variables appear to be fully mediated by diminished self-esteem. Overall, then, our hypothesized model was supported.

Figure 2Plotted Interactions of Abusive Supervision × Turnover Intentions on Self-Esteem

3

3.2

3.4

3.6

3.8

4

4.2

4.4

4.6

4.8

5

Low High

Sel

f-E

stee

m

Abusive Supervision

Low Turnover Intentions

High Turnover Intentions

3

3.2

3.4

3.6

3.8

4

4.2

4.4

4.6

4.8

5

Low High

Sel

f-E

stee

m

Abusive Supervision



3

3.2

3.4

3.6

3.8

4

4.2

4.4

4.6

4.8

5

Low High

Sel

f-E

stee

m

Abusive Supervision



(a) Study 1 Interaction (b) Study 2 Interaction

(c) Study 3 Interaction




Supplemental Robustness Tests

Alternative model testing. To provide further support for the overall model, we compared our hypothesized model with alternatives (Edwards & Lambert, 2007). It has been argued that hostile interpersonal treatment negatively affects self-esteem regardless of expectations for future interactions with the source of hostility (see Leary et al., 2003). Thus, we compared our model to one in which turnover intentions act as a moderator at the second stage between self-esteem and the dependent variables. Alternatively, it could be argued that high turn-over intentions strengthen the direct effect from abusive supervision to behavioral responses because employees who intend to leave perceive fewer costs associated with retaliation (cf. Tepper et al., 2009). Thus, we also compared our model to a moderated direct effects model, where turnover intentions moderated the direct effects of abusive supervision on deviance and façade. The interaction terms in both of these alternative models were not significant, lending further support for our theoretical framework.

Common method variance testing. The data from Study 1 were collected via surveys and from one source, which some argue may enhance the likelihood of common method vari-ance (CMV) bias, even though other design features were integrated to minimize such a bias (e.g., using time-separated data, testing for an interaction; Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). To assess the possibility of CMV as a bias, we empirically examined the extent of the presence of CMV and its potential influence on our results by conducting latent variable analyses with a marker variable (L. J. Williams, Hartman, & Cavazotte, 2010). For the marker variable in our study, we used a measure of coworker-directed citizenship behav-ior (Lee & Allen, 2002), for example, “I willingly give my time to help coworkers who have work-related problems,” rated on a 5-point scale from 1 (never) to 5 (always). The results showed that CMV was not present and so it did not bias the parameters of our model, as evidenced by a nonsignificant chi-square difference test between the baseline and Method-C models, Δχ2(1) = 0.004, n.s.

Discussion

Study 1 provides support for our model, which unites the self-defense and self-presenta-tional views of diminished self-esteem. The results revealed the indirect effects of abusive supervision on employees’ workplace deviance and façade (via diminished self-esteem) were stronger when turnover intentions were low than high. These results suggest that the self-esteem of employees who do not intend to leave the organization is more susceptible to the diminishing effects of abusive supervision; consequently, these employees’ motivation to engage in self-defensive and self-presentational behavior is enhanced. In contrast, psycho-logical detachment (viz., high turnover intentions) seems to insulate employees’ self-esteem from the effects of abusive supervision and diminishes motives to engage in behavior to protect the self.

There are notable strengths to Study 1, such as allowing us to examine our predictions on a large and diverse sample of working adults from various industries. Furthermore, the data from Study 1 were separated in time, which can enhance causal inferences and reduce the likelihood that CMV biased the parameters of the model (Podsakoff et al., 2003). Moreover, our supplemental tests demonstrated that CMV did not bias the results. Still, like all studies, the design of Study 1 is not without limitations. For example, victim precipitation research




(see Aquino & Thau, 2009) suggests that employees lower in self-esteem may prompt abu-sive behavior from supervisors. In Study 1, both abusive supervision and self-esteem were measured at the same time, limiting causal inferences on their relationship. Furthermore, self-esteem was measured at Time 1 while turnover intentions (our moderator) were assessed at Time 2. Previous research suggests that withdrawal cognitions are a consequence of abu-sive supervision (e.g., Tepper, 2000) and, certainly, it could be argued that individuals with lower self-esteem are more likely to think about and plan on quitting their jobs (see Pierce & Gardner, 2004). Our assessment of turnover intentions at Time 2, after our assessment of self-esteem, limits causal inferences.

Study 2 attempts to address these issues, extends our model test, and provides a construc-tive replication to strengthen our conclusions. In Study 2, we measured turnover intentions at the same time as abusive supervision, both of which were measured about one month prior to assessing self-esteem. This time separation may allow us to make a stronger case for the causal direction of our model. Furthermore, rather than a measure to capture dispositional self-esteem, we used a measure that was specifically designed to capture fluctuations in self-esteem (Heatherton & Polivy, 1991), which is more consistent with theory on the motivation of diminished self-esteem (Leary & Downs, 1995). Dispositional self-esteem is theoretically distinct from state self-esteem (Leary & Downs; Leary, Haupt, Strausser, & Chokel, 1998); moreover, the measure we used to capture state self-esteem in Study 2 has been shown to be empirically distinct from measures of trait self-esteem (e.g., Heatherton & Polivy) and is more sensitive than measures of trait self-esteem to the effects of specific events (e.g., Heatherton & Polivy; Leary et al., 1995; Marigold, Holmes, & Ross, 2007). In Study 2, we also collected data on multiple forms of deviance (i.e., organizational deviance, supervisor-directed deviance) and self-presentational behavior (i.e., façade, ingratiation). Finally, in Study 2, we controlled for an individual difference that could provide an alternative explana-tion for the effects: need for affiliation (nAff). Individuals higher in affiliative needs may be more sensitive to the effects of abusive treatment (cf. Maner et al., 2007) and might be more likely to conform to group values and ingratiate their coworkers and/or less likely to engage in deviant behavior (Leary & Allen, 2010; McGhee & Teevan, 1967).

Study 2 Method


Data were collected from full-time employees recruited from a variety of industries in the United States (e.g., finance, insurance, real estate, education, and health care were the most cited). In exchange for class credit, undergraduate students in a management class at a large university in the southeastern United States were asked to recruit up to two individuals who worked full-time, reported to a supervisor, worked in an environment with coworkers (to ensure sufficient variance on ingratiation), and were willing to complete two surveys sepa-rated by approximately one month. This recruitment method is consistent with published work (e.g., Mawritz, Mayer, Hoobler, Wayne, & Marinova, 2012; Mayer, Thau, Workman, Van Dijke, & De Cremer, 2012; Mitchell & Ambrose, 2012). We followed established design protocols to ensure that the data were from the recruited working adults for the study (e.g., emphasized the importance of integrity in the responses, compared student and participant e-mail and Internet protocol addresses, asked students to provide participants’ full company contact information as a way to check that the participants met the study criteria, provided




students credit regardless of whether their recruited participant completed the surveys for the study; see Mawritz et al.; Mayer et al.; Mitchell & Ambrose). Abusive supervision, turnover intentions, nAff, and participants’ demographics were assessed at Time 1; self-esteem, orga-nizational deviance, supervisor-directed deviance, façade behavior, and ingratiation were assessed at Time 2.

A total of 393 individuals completed the Time 1 survey. Of those, 262 individuals returned the Time 2 survey (66.7% response rate). Our analyses included a total of 221 observations with complete data for our study’s variables across the Time 1 and Time 2 surveys. Results of ANOVA tests indicated no significant differences between the Time 2 responders and nonresponders on demographics or Time 1 variables (Dooley & Lindner, 2003). Fifty-one percent of the sample was female, and 85% of the sample identified as White/Caucasian. On average, participants’ age was 43.4 years (SD = 13.3), company tenure was 8.6 years (SD = 8.7), and 51.1% were in nonsupervisory positions.

Measures

In Study 2, we used the same measures of abusive supervision, turnover intentions, orga-nizational deviance, and façade as those used in Study 1.

Self-esteem. We used Heatherton and Polivy’s (1991) 14-item measure of state self-esteem, which captures short-term fluctuations in feelings of self-worth. Respondents were asked to indicate how they currently felt about themselves, for example, “I feel inferior to others at this moment” (reverse-scored), on a 5-point scale ranging from 1 (not at all) to 5 (extremely).

Supervisor-directed deviance. We assessed supervisor-directed deviance with Mitchell and Ambrose’s (2007) 10-item measure. Respondents indicated their engagement in the listed behaviors (e.g., “Acted rudely toward my supervisor”) on a 5-point scale from 1 (never) to 5 (always).

Ingratiation. We assessed ingratiation with Bolino and Turnley’s (1999) four-item mea-sure. Respondents rated the extent to which each statement accurately described their behav-ior (e.g., “Compliment my colleagues so they will see me as likeable”) on a 5-point scale ranging from 1 (very inaccurate) to 5 (very accurate).

Control variables. As in Study 1, we controlled for employees’ gender and race. We also controlled for nAff with Hill’s (1987) six-item measure. Respondents rated their agreement with each statement (e.g., “I feel like I have really accomplished something valuable when I am able to get close to someone”) on a 5-point scale from 1 (strongly disagree) to 5 (strongly agree).


Descriptive Statistics, Correlations, and CFAs

Table 3 shows the means, standard deviations, and correlations among the variables for Study 2. CFA results showed the hypothesized, seven-factor measurement model was a good




fit to the data, χ2(254) = 401.92, CFI = .96, RMSEA = .05, SRMR = .06, and was a better fit than all the alternative models we tested, including (a) a six-factor model in which organiza-tional deviance and supervisor-directed deviance items loaded onto one factor, χ2(260) = 600.41, CFI = .91, RMSEA = .08, SRMR = .07, Δχ2(6) = 198.49, p < .001; (b) a six-factor model in which façade and ingratiation items loaded onto one factor, χ2(260) = 897.10, CFI = .83, RMSEA = .11, SRMR = .11, Δχ2(6) = 495.18, p < .001; (c) a six-factor model in which abusive supervision and self-esteem items loaded onto one factor, χ2(260) = 843.45, CFI = .85, RMSEA = .10, SRMR = .11, Δχ2(4) = 441.53, p < .001; and (d) a one-factor model, χ2(275) = 2,738.49, CFI = .36, RMSEA = .20, SRMR = .20, Δχ2(21) = 2,336.57, p < .001. Thus, the variables displayed measurement properties suitable for hypothesis testing.

Hypothesis Testing

We followed identical analytical procedures to those described for Study 1. Table 4 shows the path estimates of the hypothesized model. Supporting the first stage of our model, the Abusive Supervision × Turnover Intentions interaction term was significant on self-esteem (b = 0.11, p < .05), and the relationship between abusive supervision and self-esteem was stronger when turnover intentions were lower (b = −0.36, p < .01) than higher (b = −0.14, p < .05). This is shown in Figure 2b.

We decomposed the effects of abusive supervision into simple effects at higher and lower levels of turnover intentions. The simple effects of self-esteem were significant on organiza-tional deviance (b = −0.22, p < .01), supervisor-directed deviance (b = −0.25, p < .001), façade (b = −0.39, p < .001), and ingratiation (b = −0.36, p < .01). After we controlled for the direct effects of abusive supervision (organizational deviance: b = 0.05, n.s.; supervisor-directed deviance: b = 0.17, p < .01; façade: b = 0.00, n.s.; ingratiation: b = −0.03, n.s.), the RMediation results indicated that the indirect effect on organizational deviance was stronger when turnover intentions were lower (ρ = .08, 95% CI = [.02, .18]) than higher (ρ = .03, 95%

Table 3

Descriptive Statistics and Correlations Among Variables in Study 2

Variable 1 2 3 4 5 6 7 8 9 10 M SD

1. Abusive supervision (.95) 1.34 0.672. Turnover intentions .29 (.75) 2.36 0.963. Self-esteem −.28 −.19 (.90) 3.77 0.444. Façade .06 .35 −.25 (.83) 2.35 0.735. Ingratiation −.00 .17 −.19 .18 (.87) 3.33 0.796. Organizational deviance .17 .13 −.30 .20 .14 (.87) 1.36 0.367. Supervisor-directed deviance .37 .14 −.36 .16 .07 .61 (.87) 1.30 0.408. Gendera −.05 −.04 .05 −.07 −.01 −.09 −.03 — 1.50 0.509. Raceb −.04 .07 −.04 .18 −.15 −.03 .04 .04 — 1.15 0.36

10. Need for affiliation −.16 .09 −.01 .02 .25 .04 −.08 .02 −.00 (.73) 3.50 0.56

Note: N = 221. Correlations greater than |.17| are significant at p < .01 and those greater than |.13| are significant at p < .05, two-tailed. Reliability estimates (coefficient alphas) of scales are reported along the diagonal in parentheses.aMale = 1, female = 2.bWhite = 1, non-White = 2.



18

Tab

le 4

Pat

h A

nal

ysis

Res

ult

s fo

r S

tud

y 2

Sel

f-E

stee

m

S

tep

1S

tep

2O

rgan

izat

iona

l D

evia

nce

Sup

ervi

sor-

Dir

ecte

d D

evia

nce

Faç

ade

Ingr

atia

tion

Var

iabl

esb

SEb

SEb

SEb

SEb

SEb

SE

Gen

dera

0.03

0.06

0.01

0.06

−0.

050.

050.

000.

05−

0.10

0.10

−0.

000.

10R

aceb

−0.

060.

09−

0.07

0.08

−0.

030.

070.

050.

080.

36*

0.15

−0.

35*

0.16

Nee

d fo

r af

fili

atio

n−

0.05

0.05

−0.

030.

050.

030.

05−

0.02

0.05

0.03

0.09

0.34

***

0.10

Abu

sive

sup

ervi

sion

−0.

19**

0.06

−0.

25**

0.08

0.05

0.06

0.17

**0.

060.

000.

06−

0.03

0.09

Tur

nove

r in

tent

ions

−0.

010.

04−

0.04

0.03

A

busi

ve S

uper

visi

on ×

Tur

nove

r In

tent

ions

0.11

*0.

06

Sel

f-es

teem

−0.

22**

0.07

−0.

25**

*0.

07−

0.39

***

0.10

−0.

36**

*0.

11R

2 Equ

atio

n.0

9**

.12*

*.1

1**

.21*

*.0

9**

.13*

*Δ

R2

.03*

.07*

*.0

7**

.05*

*.0

4**

Not

e: N

= 2

21. T

able

val

ues

are

unst

anda

rdiz

ed p

ath

esti

mat

es a

nd s

tand

ard

erro

rs fr

om th

e es

tim

ated

mod

el. S

tati

stic

al s

igni

fica

nce

is b

ased

on

a tw

o-ta

iled

test

. ΔR

2 val

ues

for

depe

nden

t var

iabl

es a

re c

ompa

red

to th

e eq

uati

on e

xclu

ding

sel

f-es

teem

.a M

ale

= 1

, fem

ale

= 2

.b W

hite

= 1

, non

-Whi

te =

2.

*p <

.05.

**p

< .0

1.**

*p <

.001

.




CI = [.01, .08]). For supervisor-directed deviance, the indirect effect was stronger when turn-over intentions were lower (ρ = .09, 95% CI = [.02, .20]) than higher (ρ = .03, 95% CI = [.01, .09]). Combined, these results support Hypothesis 1a. For façade, the indirect effect was stronger when turnover intentions were lower (ρ = .14, 95% CI = [.04, .27]) than higher (ρ = .05, 95% CI = [.01, .12]). For ingratiation, the indirect effect was stronger when turnover intentions were lower (ρ = .13, 95% CI = [.04, .30]) than higher (ρ = .05, 95% CI = [.01, .12]). Combined, these results support Hypothesis 1b. Given the strength of the direct effects, it appears that diminished self-esteem fully mediated the effects of abusive supervision on organizational deviance, façade, and ingratiation but only partially mediated the effects on supervisor-directed deviance. Altogether, our hypothesized model was supported.

Supplemental Robustness Tests

Alternative model testing. We again compared our hypothesized model to alternative models discussed in Study 1 to test their fit to the Study 2 data (cf. Edwards & Lambert, 2007). The interaction terms of a second-stage moderated mediation model and a moderated direct effects model were not significant, suggesting our theoretical model was a better fit to the data than these alternatives.

CMV testing. As with Study 1, we empirically assessed the degree to which CMV existed and presented a potential bias to our results (L. J. Williams et al., 2010). The marker vari-able used was a measure of coworker reconciliation attempts (Wade, 1989), for example, “I accept their humanness, flaws, and failures,” rated on a 5-point scale ranging from 1 (strongly disagree) to 5 (strongly agree). The results show that CMV was not present and so it did not bias the parameters of our model, as evidenced by a nonsignificant chi-square dif-ference test between the baseline and Method-C models, Δχ2(1) = 0.19, n.s.

Discussion

Study 2 replicated the results of Study 1 and provided converging evidence supporting theory on diminished self-esteem (Baumeister et al., 1996; Leary & Downs, 1995). Abusive supervision impaired victims’ self-esteem, motivating them to protect their self-worth through deviant and self-presentational behavior. Furthermore, abusive supervision had a weaker effect on employees’ self-esteem and behavioral reactions when employees had psy-chologically detached from the organization (i.e., held high turnover intentions).

Study 2 had notable strengths that offer more evidence in support of our overall theoreti-cal model. The effects of our model generalized to multiple forms of workplace deviance (i.e., organizational deviance, supervisor-directed deviance) and self-presentational behavior (i.e., façade, ingratiation). Furthermore, the results held while controlling for an important individual difference (i.e., nAff), which is theoretically related to the dependent variables. In Study 2, we also assessed self-esteem with a measure of state self-esteem, which is largely implicated by our theory. Importantly, even after a 1-month time lag, we found a relationship between abusive supervision and state self-esteem, which suggests the effects of abusive supervision can potentially have long-lasting effects on the sense of self.

Nevertheless, even with the strengths of the design of Study 2, there are limitations worth mentioning. For example, although we separated the collection of our predictor from




the mediator and dependent variables, inferences of causality are still limited (Stone-Romero, 2002). It is also possible that other causal mechanisms could have accounted for the effects of our model. Affective events theory (Weiss & Cropanzano, 1996) suggests that employees’ affect or job attitudes may mediate the relationship between events at work (e.g., abusive supervision) and employees’ work behavior. Indeed, Ferris et al. (2012) found that the relationship between supervisors’ treatment of employees and employees’ deviant behavior could be explained, in part, by job satisfaction and affect. Finally, turn-over researchers (Hom, Mitchell, Lee, & Griffeth, 2012) have theorized that the availabil-ity of viable job alternatives may influence the impact of turnover intentions on employee behavior.

Our Study 3 addresses these limitations and further bolsters the causal chain implied by our model. Study 3 uses a daily diary study design, assessing abusive supervision, state self-esteem, and the dependent variables each day over a period of 21 days. This design allows us to examine the within-individual effects of our model while controlling for the between-individual variance that exists. Furthermore, in Study 3, we modeled and controlled for daily job satisfaction and daily positive affect, which are alternative mediating mechanisms that could explain the effects (Ferris et al., 2012). We also controlled for employees’ perceptions of job alternatives and employees’ trait self-esteem.

Study 3 Method


Participants were recruited using advertisements placed online. To be eligible for the study, participants had to have regular interactions with supervisors and coworkers and be working in a full-time job. Furthermore, we specified that individuals should be willing to complete short daily surveys at the conclusion of their workdays over the period of 21 days. Participants were compensated $2.50 for completing each survey and received a $10 bonus if they completed at least 10 surveys. Approximately 330 individuals indicated willingness to participate in the study; 225 of these met the eligibility criteria. Because of budget con-straints, we randomly selected 85 of these individuals to participate in the study.

Participants were sent a link to an initial survey that captured their demographics, turn-over intentions, job alternatives, and trait self-esteem. Approximately 1 week later, 83 par-ticipants who had completed the initial survey were sent daily surveys during the last hour of each day they worked for 21 days. The daily surveys contained shortened measures of abu-sive supervision, state self-esteem, daily job satisfaction, daily positive affect, workplace deviance (organizational and supervisor directed), and façade. Ten participants did not com-plete surveys during the first week and were removed from the study. Because some indi-viduals worked on the weekend, it was possible for an individual to receive 21 surveys during the 3-week period, although the majority received 15. Of the 1,179 surveys sent to the remaining 73 participants, 844 were completed for an overall response rate of 71.6%. Seventy percent of the sample was female, and 65.8% of the sample identified as White/Caucasian. On average, participants’ age was 35.97 years (SD = 9.84) and company tenure was 6.4 years (SD = 4.6). Participants came from a variety of industries (finance/insurance, retail, comput-ers/information technology, and education were most cited), and 71.9% held nonsupervisory positions.




Measures

Consistent with published work employing a diary design (e.g., Dalal, Lam, Weiss, Welch, & Hulin, 2009), our procedure used shortened measures for several of our constructs to reduce survey length and rater fatigue. Unless otherwise stated, each measure was assessed on a 5-point scale ranging from 1 (strongly disagree) to 5 (strongly agree).

Daily abusive supervision. To assess abusive supervision, we used the five-item Mitchell and Ambrose (2007) measure. Respondents indicated their agreement with whether their supervisor had engaged in each of the behaviors toward them during the past day.

Turnover intentions. We assessed turnover intentions on the initial survey. Turnover intentions were measured with two items from Crossley, Bennett, Jex, and Burnfield (2007; e.g., “I will quit this organization as soon as possible”).

Daily self-esteem. We used 11 items from Heatherton and Polivy’s (1991) measure of state self-esteem. Respondents indicated their agreement with the extent to which each state-ment described how they currently felt, for example, “I am worried about what other people think of me” (reverse-scored).

Daily workplace deviance. We assessed organizational deviance with nine items from Bennett and Robinson (2000) and supervisor-directed deviance with eight items from Mitch-ell and Ambrose (2007). We asked respondents to indicate their agreement with whether they had engaged in the behaviors at work during the past day.

Daily façade. We assessed façade with the five-item measure from Hewlin (2009). Respondents rated their agreement with whether each statement accurately described their behavior over the past day.

Control variables. Daily job satisfaction was assessed with the five items from Judge, Scott, and Ilies (2006), and daily positive affect was measured with the 10-item Positive and Negative Affect Schedule (Watson, Clark, & Tellegen, 1988). Employees’ job alterna-tives and trait self-esteem were measured on the initial survey. Employees’ perceptions of job alternatives were assessed using a three-item measure by Crossley et al. (2007; e.g., “It would be fairly easy to get an alternative job offer”). Trait self-esteem was assessed with the 10-item Rosenberg (1965) measure.


Descriptive Statistics, Correlations, and CFAs

Table 5 shows the means, standard deviations, and correlations among the variables for Study 3. Following other researchers’ similar examples (e.g., Ilies, Scott, & Judge, 2006), we conducted CFA on the daily measures and found the hypothesized five-factor measurement model provided an adequate fit to the data, χ2(181) = 570.60, CFI = .93, RMSEA = .05, SRMR = .07, and was a better fit to the data than the alternative models we tested, including



22

Tab

le 5

Des

crip

tive

Sta

tist

ics

and

Cor

rela

tion

s A

mon

g V

aria

ble

s in

Stu

dy

3

Var

iabl

e1

23

45

67

89

10M

SD

Wit

hin-

indi

vidu

al (

dail

y) v

aria

bles

a

1.

Abu

sive

sup

ervi

sion

(.94

)−

.29*

*.1

1**

.15*

*.2

3**

−.1

5**

−.1

9**

1.58

0.65

2.

Sta

te s

elf-

este

em−

.48*

*(.

93)

−.1

9**

−.2

1**

−.2

2**

.05

.20*

*3.

740.

693.

F

açad

e.4

7**

−.5

1**

(.91

).1

7**

.19*

*.0

4−

.10*

*2.

410.

794.

O

rgan

izat

iona

l dev

ianc

e.4

7**

−.4

4**

.34*

*(.

89)

.60*

*.0

5−

.06

1.55

0.54

5.

Sup

ervi

sor-

dire

cted

dev

ianc

e.6

4**

−.4

8**

.43*

*.8

6**

(.91

).0

7*−

.06

1.56

0.61

6.

Job

sati

sfac

tion

.06

.11

.06

−.0

8−

.04

(.90

).3

1**

2.84

0.30

7.

Pos

itiv

e af

fect

−.0

6.2

4*.0

4−

.24*

−.1

6.4

3**

(.93

)2.

870.

73B

etw

een-

indi

vidu

al (

trai

t) v

aria

bles

b

8.

Tur

nove

r in

tent

ions

.19

.09

.13

.20

.21

−.1

3−

.20

(.80

)2.

910.

709.

Jo

b al

tern

ativ

es−

.10

.14

−.0

3−

.05

.01

.07

.07

.27*

(.83

)3.

500.

9410

. T

rait

sel

f-es

teem

−.2

8*.6

0**

−.2

6*−

.37*

*−

.26*

.22

.24*

.09

.21

(.86

)3.

820.

61

Not

e: C

orre

lati

ons

abov

e th

e di

agon

al a

re w

ithi

n-in

divi

dual

cor

rela

tion

s; c

orre

lati

ons

unde

r th

e di

agon

al a

re b

etw

een-

indi

vidu

al c

orre

lati

ons.

Cor

rela

tion

s be

twee

n th

e da

ily

vari

able

s an

d tr

ait

vari

able

s w

ere

com

pute

d by

agg

rega

ting

par

tici

pant

s’ d

aily

sco

res

and

then

cor

rela

ting

the

m w

ith

trai

t sc

ores

. R

elia

bili

ty e

stim

ates

(co

effi

cien

t al

phas

) of

sca

les

are

repo

rted

alo

ng th

e di

agon

al in

par

enth

eses

.a N

= 8

44 o

bser

vati

ons.

b N =

73

part

icip

ants

.*p

< .0

5.**

p <

.01.




(a) a four-factor model in which organizational deviance and supervisor-directed deviance items loaded onto one factor, χ2(185) = 728.43, CFI = .90, RMSEA = .06, SRMR = .07, Δχ2(4) = 157.83, p < .001; (b) a four-factor model in which abusive supervision and self-esteem items loaded onto one factor, χ2(185) = 1,338.09, CFI = .79, RMSEA = .08, SRMR = .12, Δχ2(4) = 767.49, p < .001; and (c) a one-factor model, χ2(191) = 3,979.15, CFI = .30, RMSEA = .15, SRMR = .17, Δχ2(10) = 3,408.55, p < .001. Thus, the variables were deemed suitable for hypothesis testing.

Hypothesis Testing

Testing the within-individual relationships of the hypothesized model relied on sufficient within-individual variance existing for our constructs. We calculated the amount of within-individual (σ2) and between-individual (τ00) variance for each of the daily variables. The proportion of variance residing at the within-individual level can be calculated with the equa-tion σ2 / (σ2 + τ00) (see Ferris et al., 2012). The results of these computations indicated that all of our daily variables showed sufficient variance at the within-individual level (abusive supervision = 35.2%, state self-esteem = 24.1%, façade = 31.7%, organizational deviance = 34.5%, supervisor-directed deviance = 34.5%, job satisfaction = 57.5%, positive affect = 42.0%). Thus, we proceeded with multilevel analyses.

Because our hypothesized model involved the cross-level moderating effect of turnover intentions on relationships between variables at the within-individual level, we followed other similar examples (e.g., Wallace, Butts, Johnson, Stevens, & Smith, in press) and con-ducted multilevel path analyses with robust full maximum likelihood estimation using Mplus Version 7.1. Similar to other daily diary studies (e.g., Ferris et al., 2012), we conducted all of our analyses on an unrestricted model, where error variances and correlations are not con-strained to be fixed. In our analyses, we controlled for the between-individual covariances among the daily variables to appropriately account for the between-individual variance of our constructs. The between-individual variables were grand-mean centered prior to the anal-yses. Furthermore, the direct effects of abusive supervision on the dependent variables were controlled for.

Table 6 shows the parameters of the hypothesized multilevel model. Supporting the first stage of our model, the cross-level interaction of daily abusive supervision and turnover intentions was significant (γ = 0.09, p < .05), and the relationship between daily abusive supervision and daily self-esteem was stronger when turnover intentions were lower (γ = −0.31, p < .001) than higher (γ = −0.13, p < .001). This is shown in Figure 2c.

We decomposed the within-individual effects of abusive supervision into simple effects at high and low levels of turnover intentions. The simple effects of self-esteem were significant on organizational deviance (γ = −0.17, p < .001), supervisor-directed deviance (γ = −0.18, p < .001), and façade (γ = −0.23, p < .01). After we controlled for the direct effects of abusive supervision (organizational deviance: γ = 0.07, n.s.; supervisor-directed deviance: γ = 0.15, p < .05; façade: γ = 0.06, n.s.), the results showed that the within-individual indirect effect of daily abusive supervision on organizational deviance was stronger when turnover intentions were lower (ρ = .05, 95% CI = [.02, .08]) than higher (ρ = .02, 95% CI = [.00, .05]). Similarly, for supervisor-directed deviance, the within-individual indirect effect of daily abusive super-vision was stronger when turnover intentions were lower (ρ = .06, 95% CI = [.02, .10]) than higher (ρ = .02, 95% CI = [.00, .05]). These results support Hypothesis 1a. For façade, the within-individual indirect effect of daily abusive supervision was stronger when turnover




intentions were lower (ρ = .07, 95% CI = [.02, .12]) than higher (ρ = .04, 95% CI = [.01, .08]). Thus, Hypothesis 1b was also supported.

Supplemental Analyses

As outlined above, we also assessed the role of two other potential mediators (i.e., daily job satisfaction, daily positive affect) of the relationship between daily abusive supervision and daily work behaviors. Although the direct effects of daily abusive supervision were sig-nificant on daily job satisfaction (γ = −0.25, p < .001) and daily positive affect (γ = −0.24, p < .01), turnover intentions did not moderate the effects of daily abusive supervision on these variables. Thus, the effects of daily abusive supervision on job satisfaction and positive affect are not conditional on employees’ detachment from the organization. Furthermore, the results showed that job satisfaction mediated the effects of daily abusive supervision on orga-nizational deviance (ρ = .02, 95% CI = [.00, .04]) but not on supervisor-directed deviance (ρ = .02, 95% CI = [−.00, .05]) or façade (ρ = −.00, 95% CI = [−.03, .02]). Daily positive affect did not mediate any of the effects of daily abusive supervision on the dependent variables. Combined, these results lend further support for the role of diminished self-esteem in our model.

Discussion

Study 3 replicated the results of Studies 1 and 2, providing further evidence of the theory we proposed about the motivational effects of diminished self-esteem. Over the course of the

Table 6

Within-Individual and Cross-Level Effects of the Multilevel Path Analysis for Study 3

Daily Self-Esteem Façade

Organizational Deviance

Supervisor-Directed Deviance

Variables γ SE γ SE γ SE γ SE

Intercept, γ00 3.74*** 0.08 2.41*** 0.10 1.55*** 0.06 1.57*** 0.07Job alternatives, γ01 −0.01 0.09 0.02 0.11 0.02 0.06 0.05 0.06Trait self-esteem, γ02 0.67*** 0.13 −0.34 0.18 −0.33** 0.11 −0.27* 0.12Daily abusive supervision, γ10 −0.24*** 0.04 0.06 0.05 0.07 0.04 0.15* 0.06Turnover intentions, γ03 0.14 0.09 Daily Abusive Supervision ×

Turnover Intentions, γ13

0.09* 0.04

Daily job satisfaction, γ20 0.01 0.05 −0.08* 0.04 −0.09 0.05Daily positive affect, γ30 −0.05 0.04 0.03 0.05 0.04 0.05Daily self-esteem, γ40 −0.23** 0.08 −0.17*** 0.04 −0.18*** 0.05

Note: N = 73 participants, with 844 observations. Table values are unstandardized path estimates from the unrestricted model. Statistical significance is based on a two-tailed test. Cross-level moderating effects of turnover intentions on daily job satisfaction and daily positive affect (not shown) are not significant; results available from authors.*p < .05.**p < .01.***p < .001.




diary study, abusive supervision was found to diminish victims’ self-esteem, which influ-enced victims’ self-defensive behavior (i.e., supervisor-directed deviance, organizational deviance) and self-presentational behavior (i.e., façade). Diminished self-esteem appeared to fully mediate the effects of abusive supervision on organizational deviance and façade and partially mediate the effects on supervisor-directed deviance. Abusive supervision indirectly and positively affected these behaviors (via self-esteem); however, the effects were weaker when victims had psychologically detached from the organization (i.e., they held high inten-tions to leave the organization). Notably, the results held even while controlling for variables that might offer alternative explanations for the effects of abusive supervision on employees’ behaviors.

General Discussion

Combined, the results of our three studies demonstrate the motivational effects of dimin-ished self-esteem from abusive supervision. Self-esteem involves individuals’ sense of self-worth and value. It is influenced by interactions with important others, like supervisors, because these interactions indicate whether employees are of value to the workgroup and organization (Brockner, 1988; Leary & Downs, 1995). Abusive supervision signals that vic-timized employees are not valued and, instead, are a source of aggravation to the supervisor (Tepper et al., 2011). As such, abusive supervision undermines employees’ self-esteem.

Social psychologists (Baumeister & Tice, 1985; Leary, 1990) generally agree with the notion that diminished self-esteem from abusive treatment motivates victims to engage in behavior to protect and fortify their self-worth. Scholars differ, however, in their position on how diminished self-esteem from abuse affects behavioral reactions. Our work bridges the self-defense view (which suggests diminished self-esteem from abuse motivates destructive acts) with the self-presentational view (which suggests diminished self-esteem from abuse motivates self-presentational acts). The results from our studies support this united model, showing that victims of abusive supervision not only attempted to protect their self-esteem and stand up for themselves via workplace deviance but also attempted to signal their fit with the workgroup with self-presentational behavior by acting consistent with the organi-zation’s values (i.e., façade) and endeavoring to appear to others as more likeable (i.e., ingratiation). Indeed, diminished self-esteem fully mediated the effects of abusive supervi-sion on three of the four dependent variables studied; it only partially mediated the effects on supervisor-directed deviance. Thus, it can be concluded that diminished self-esteem is a powerful explanatory mechanism in understanding the motivational potential of abusive supervision, particularly with respect to self-presentational behavior, such as façade and ingratiation.

Our results also show that the effects of abusive supervision on victims’ self-esteem and behavior depend on victims’ psychological detachment (i.e., turnover intentions) from the organization. Because employees with high turnover intentions have psychologically dis-tanced themselves from other individuals within the organization (e.g., supervisors) and from the organization itself (Burris et al., 2008; Kahn, 1990), they are better able to deflect the effects of abusive supervision on their sense of self (e.g., Bowins, 2004). Consequently, their motivation to engage in behavior to protect and fortify the sense of self-worth was reduced. In contrast, victims who have low turnover intentions are more dependent on the treatment they receive from their supervisors (Burris et al.; Kahn; Tepper et al., 2009). Given their




intention of continued employment, abusive supervision more strongly diminished these vic-tims’ self-esteem, which strengthened their motivation to protect their sense of self through deviant and self-presentational behavior.

Theoretical Implications

Our research has implications for theory. In particular, our work extends the literatures on abusive supervision and self-esteem. Prior work on abusive supervision has found that employees experience a depleted sense of self from abusive treatment, which motivates devi-ant behaviors (e.g., Ferris et al., 2012). Deviance is said to serve as a defensive act, allowing victims to stand up for themselves against the abusive supervisor. Yet these ideas derive from only one stream of literature within social psychology (Baumeister & Leary, 1995). Theory and research on self-esteem (Leary, 1990; Leary & Downs, 1995) suggest that abusive treat-ment may motivate different types of behaviors, each with the underlying goal of protecting the sense of self. We took an integrative approach, creating a unified model of self-esteem to explain how abusive supervision can motivate its victims to engage in defensive as well as self-presentational acts. We show that deviance is only one potential reaction to abuse. Our results highlight that diminished self-esteem from abuse influences defensive and self-pre-sentational behavior. Workplace deviance allows victims to stand up for their sense of worth (Baumeister, 1997; Crocker & Park, 2004). However, victims who experience diminished self-esteem may also engage in façade and ingratiation, signaling they are a good fit with the workgroup.

Our research also contributes to both the abusive supervision and social psychology litera-tures on the relevance of turnover intentions on abused employees’ self-esteem. We drew upon the self-defense (Baumeister & Leary, 1995) and self-presentational (Leary, 1990) views of self-esteem, which suggest that the extent to which individuals experience dimin-ished self-esteem from abusive treatment depends on how detached victims are from the group. We integrated these ideas with theory about the psychological detachment associated with turnover intentions (e.g., Burris et al., 2008; Kahn, 1990) to explain that psychological detachment helps to protect employees’ self-esteem from abusive supervision. Employees who intend to leave the organization do not hold as much value in their supervisor’s interac-tions with them because they have psychologically distanced themselves from the work-group and do not see themselves as dependent on that group in the future. Conversely, abusive supervision exerts a stronger influence on victim’s self-esteem when they intend to stay in the organization. Supervisors’ treatment of these individuals is a more salient and relevant source of self-esteem because these employees expect future interactions with organizational mem-bers. Our results support these ideas and show that low turnover intentions magnified the negative effects on victims’ self-esteem, strengthening self-defensive and self-presentational motivations.

Limitations and Future Research Directions

Like all studies, our research should be considered in light of some limitations. In particu-lar, our data were collected via survey design and using one focal source. Both of these design elements were important to the nature of the phenomenon being studied. That is, these individuals are most knowledgeable about how they perceive the treatment of others, their




sense of self-esteem, and their own deviant, façade, and ingratiation behaviors. Furthermore, examining the model using three samples of employees from various jobs and industries heightens the validity and generalizability of the findings (Kerlinger & Lee, 2000).

Nevertheless, we concede that collecting the data via surveys and through self-report potentially increases CMV and limits causal inferences. We attempted to proactively deal with these issues in the design of our studies, and we empirically tested for CMV. In the first two studies, separating the surveys in time by 1 month reduces the likelihood of CMV bias-ing the results (Doty & Glick, 1998; Podsakoff et al., 2003); furthermore, the results of our empirical tests indicated that CMV did not bias the results. Time separation also enhances the ability to make causal inferences. It is also important to note that causal ordering associated with our predictions is consistent with theory and experimental research: Abusive treatment decreases self-esteem, which leads to attempts to protect self-esteem via destructive and self-presentational behavior (Baumeister et al., 1996; Leary, 1990; Leary et al., 1995). Study 3 lends more confidence to the casual direction of our hypothesized model. With its daily diary design, we were able to partial out the variance in self-esteem at the between-individual level and demonstrate that on days when employees are subjected to abusive supervision, their self-esteem is likely to diminish and, subsequently, their motivation to protect and fortify their self-worth is heightened. Of course, these effects were reduced when employees had greater intentions to leave the organization.

Finally, as pointed out by an anonymous reviewer, assessing employees’ detachment from the organization with a measure of turnover intentions may ignore the possibility that some employees who intend to leave have not psychologically detached. Our use of a measure of turnover intentions as an indicator of detachment is consistent with other work in the litera-ture (e.g., Burris et al., 2008). Moreover, in Study 3, we controlled for perceived job alterna-tives to strengthen the claim that turnover intentions assess psychological detachment. Nevertheless, turnover researchers have argued that some employees who leave their organi-zations should be considered involuntary leavers (Hom et al., 2012). For example, an employee whose spouse’s job necessitates relocation of the family may still be somewhat attached to the organization despite intending to leave it. Therefore, future research might consider other ways of assessing the psychological detachment of employees.

Despite these limitations, our research also provides a nice foundation for future research. For instance, our work shows that diminished self-esteem from abuse motivates victims’ organizational deviance, supervisor-directed deviance, façade, and ingratiation. These are only four possible ways victims may protect and fortify their self-esteem. Theoretically, it is possible for victims to engage in other types of behavior (Leary & Downs, 1995). For instance, it is quite possible that diminished self-esteem prompts abused employees to engage in unethical behavior, particularly when such behavior may produce benefits for the work-group (Pillutla & Thau, 2009; Scott & Thau, 2012). Unethical acts may improve victims’ standing within the group, which could ultimately enhance self-esteem. Alternatively, it is possible that victims attempt to fortify self-worth ethically through increased constructive efforts or other behavior that benefits the workgroup (e.g., citizenship, creativity). We leave these ideas for future research to consider.

Furthermore, an underlying goal stipulated in the self-presentational view (Leary & Downs, 1995) is that victims engage in self-presentational behaviors to attain a better fit with their workgroup, which then fortifies their self-esteem. Our findings are consistent with Leary and Downs’ hypothesis that diminished self-esteem motivates behavior aimed at




reconnecting with others, which was demonstrated in façade and ingratiation. However, we did not examine whether employees were ultimately successful in increasing their belonging with others through these tactics. Creating façades may make employees appear more inter-personally desirable and trustworthy (Edwards & Cable, 2009), thereby improving the likeli-hood of forming social bonds, which would ultimately raise self-esteem. For example, Chrobot-Mason, Button, and DiClementi (2001) found that employees who falsely portray an identity that is similar to that of group members were more likely to be included by the group than were employees who avoided discussion about or who openly revealed their true selves. Even still, Hewlin (2009) found that façade is emotionally exhausting and leads to withdrawal cognitions; both of these outcomes suggest that detrimental effects to organiza-tions may also emerge as a result of employee façade. Similarly, although research suggests ingratiatory behaviors toward others are often effective at fulfilling needs for belonging and affiliation (Gordon, 1996), ingratiators run the risk of being perceived as brownnosers (Turnley & Bolino, 2001). This could result in the abused employee being alienated or ostra-cized, threatening diminished self-esteem even further. Research has shown that the effec-tiveness of ingratiation depends on individuals’ political skill and self-monitoring (Harris, Kacmar, Zivnuska, & Shaw, 2007; Turnley & Bolino). Thus, future research should investi-gate the extent to which ingratiation results in the restoration of abused employees’ self-esteem and whether some individuals are more successful than others.

Future research might also investigate factors, such as personality traits, that prompt devi-ant versus self-presentational responses associated with diminished self-esteem from abuse. For example, workplace deviance may be more likely when the victim is low on conscien-tiousness or emotional stability or high on impulsivity. Similarly, self-presentational responses could be more likely for employees higher in agreeableness or extraversion. It may also be instructive to investigate situational factors promoting these behavioral responses. Because the goal of self-presentational responses are geared to restore the sense of self through greater acceptance and inclusion in the group, we would expect to see this response more often used when coworkers are readily available for interaction, such as in jobs with high interdependence with colleagues or in the context of teamwork.

Practical Implications

Our findings add to growing evidence that abusive supervision is a significant problem in organizations and efforts should be made to reduce its incidence. As with other studies exam-ining the consequences of abusive supervision, our studies show these hostile and abusive acts towards employees have a detrimental effect on employees. Abused employees experi-ence diminished self-esteem, which consequently motivates costly behavior to organizations, such as workplace deviance (Dunlop & Lee, 2004). Furthermore, although self- presentational behavior may, on the face of things, seem beneficial to organizations, employ-ees who engage in these behaviors are masking their true thoughts, feelings, and opinions. That is, façade involves employees faking that they embrace organizational values (Hewlin, 2003), and ingratiation involves doing favors for or flattering individuals in the workgroup (Bolino & Turnley, 1999), which can include individuals who are abusive to the actor. Acting contrary to one’s values and trying to earn the favor of someone who is abusive requires energy that may detract from effort being exerted for productive purposes; moreover, these are stressful experiences (Hewlin, 2009; Meglino & Ravlin, 1998), which may further




promote costs to organizations by way of decreased commitment and retention (e.g., Schneider, Kristof, Goldstein, & Smith, 1997). Furthermore, façade may be more difficult for managers to detect and monitor than deviance. If employees are hiding frustration and pain associated with abusive supervision and disguising it as constructive behavior, organizational decision makers may not become aware of problems with the employee, supervisor, or work unit, making it more difficult to rectify the problematic supervisor.

Our results also offer insight to human resource practitioners seeking to help employees cope with abusive supervision. Self-esteem theory suggests that diminished self-esteem from abuse can be fortified by fulfilling individuals’ need for belonging (Leary, 1990; Leary & Downs, 1995). Organizations can help employees deal with abusive supervision by offering opportunities for employees to build strong social connections with their colleagues in the workplace and by presenting employees with other more adaptive and constructive ways to deal with abusive supervisors (e.g., reporting systems, open-door policies). By creating an environment that enhances and maintains adaptive and safe interactions among employees, organizations can reduce employees’ overdependence on their supervisor for the need to feel valued and help buffer the negative consequences associated with abusive supervision. Doing so could also mitigate the occurrence of abusive supervision altogether.

Conclusion

Our research adds to a growing literature focusing on the behavioral motivations associ-ated with abused employees’ diminished self-esteem (e.g., Ferris et al., 2012). The findings of three studies show that when victims of abusive supervision experience diminished self-esteem, they become motivated to engage in self-defensive behavior (organizational and supervisor-directed deviance) and self-presentational behavior (façade and ingratiation) in order to protect their sense of self-worth. Furthermore, these effects are stronger when employees intend to maintain employment in the organization (i.e., turnover intentions are low). Practitioners should be made aware that the consequences of abusive supervision are not always destructive and easy to detect; victims of abusive supervisors are also motivated to restore self-esteem via self-presentational behavior. Moreover, and contrary to their self-interest, employees who intend to remain are even more likely to display these behavioral tendencies.

Note1. An anonymous reviewer pointed out that debate exists about when the use of a parceling strategy is acceptable.

Although parceling is not acceptable when developing measures, analyzing latent means, or testing measurement invariance (Marsh, Lüdtke, Nagengast, Morin, & Von Davier, 2013), it is generally agreed that when researchers are confident in the dimensionality of a study’s measures (both by consulting evidence from the literature and through statistical tests, such as exploratory factor analysis), parcels may be used to test for discriminant validity for the purpose of hypothesis testing (e.g., Little et al., 2013; Marsh et al.). Even still, Marsh et al. suggested that additional tests, such as exploratory structural equation modeling (ESEM), might be used along with traditional CFA tests to determine the extent to which items might cross-load onto other factors. CFA imposes strict tests for factor load-ings—that they do not cross-load—while ESEM does not. To explore the possibility that some items showed sig-nificant cross-loadings, we conducted ESEM on our data. ESEM results for Study 1 supported a five-factor model, and every item loaded strongest on its intended factor. Cross-loadings were weak and over 95% were nonsignificant; those nearing significance could be predicted a priori by our hypothesized model. For Study 2, ESEM results sup-ported a seven-factor model, and item loadings were consistent with the pattern shown for Study 1. Given that the




purpose of our CFA was to establish sufficient discriminant validity of our constructs for hypothesis testing and that we were using established measures from the literature, we concluded that a parceling strategy was appropriate.

ReferencesAguinis, H. 1995. Statistical power problems with moderated multiple regression in management research. Journal

of Management, 21: 1141-1158.Aquino, K., & Douglas, S. 2003. Identity threat and antisocial behavior in organizations: The moderating effects

of individual differences, aggressive modeling, and hierarchical status. Organizational Behavior and Human Decision Processes, 90: 195-208.

Aquino, K., & Thau, S. 2009. Workplace victimization: Aggression from the target’s perspective. Annual Review of Psychology, 60: 717-741.

Ashforth, B. 1997. Petty tyranny in organizations: A preliminary examination of antecedents and consequences. Canadian Journal of Administrative Sciences, 14: 126-140.

Baumeister, R. F. 1997. Esteem threat, self-regulatory breakdown, and emotional distress as factors in self-defeating behavior. Review of General Psychology, 1: 145-174.

Baumeister, R. F., & Leary, M. R. 1995. The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117: 497-529.

Baumeister, R. F., Smart, L., & Boden, J. M. 1996. Relation of threatened egotism to violence and aggression: The dark side of high self-esteem. Psychological Review, 103: 5-33.

Baumeister, R. F., & Tice, D. M. 1985. Self-esteem and responses to success and failure: Subsequent performance and intrinsic motivation. Journal of Personality, 53: 450-467.

Bennett, R. J., & Robinson, S. L. 2000. The development of a measure of workplace deviance. Journal of Applied Psychology, 85: 349-360.

Bolino, M. C., & Turnley, W. H. 1999. Measuring impression management in organizations: A scale development based on the Ones and Pittman taxonomy. Organizational Research Methods, 2: 187-206.

Bowins, B. 2004. Psychological defense mechanisms: A new perspective. The American Journal of Psychoanalysis, 64: 1-26.

Brockner, J. 1988. Self-esteem at work. Lexington, MA: Lexington Books.Burris, E. R., Detert, J. R., & Chiaburu, D. S. 2008. Quitting before leaving: The mediating effects of psychological

attachment and detachment on voice. Journal of Applied Psychology, 93: 912-922.Burton, J. P., & Hoobler, J. M. 2006. Subordinate self-esteem and abusive supervision. Journal of Managerial

Issues, 18: 340-355.Chen, Y., Ferris, D. L., Kwan, H. K., Yan, M., Zhou, M., & Hong, Y. 2013. Self-love’s lost labor: A self-

enhancement model of workplace incivility. Academy of Management Journal, 56: 1199-1219.Chrobot-Mason, D., Button, S. B., & DiClementi, J. D. 2001. Sexual identity management strategies: An explora-

tion of antecedents and consequences. Sex Roles, 45: 321-336.Cohen, J., Cohen, P., West, S. G., & Aiken, L. S. 2003. Applied multiple regression/correlation analysis for the

behavioral sciences (3rd ed.). Hillsdale, NJ: Erlbaum.Collins, B., Mossholder, K., & Taylor, S. 2012. Does process fairness affect job performance? It only matters if they

plan to stay. Journal of Organizational Behavior, 33: 1007-1026.Crocker, J., & Park, L. E. 2004. The costly pursuit of self-esteem. Psychological Bulletin, 130: 392-414.Cropanzano, R., James, K., & Konovsky, M. 1993. Dispositional affectivity as a predictor of work attitudes and job

performance. Journal of Organizational Behavior, 14: 595-606.Crossley, C. D., Bennett, R. J., Jex, S. M., & Burnfield, J. L. 2007. Development of a global measure of job embed-

dedness and integration into a traditional model of turnover. Journal of Applied Psychology, 92: 1031-1042.Dalal, R. S., Lam, H., Weiss, H. M., Welch, E. R., & Hulin, C. L. 2009. A within-person approach to work behavior

and performance: Concurrent and lagged citizenship-counterproductivity associations, and dynamic relation-ships with affect and overall job performance. Academy of Management Journal, 52: 1051-1066.

Dooley, L. M., & Lindner, R. 2003. The handling of nonresponse error. Human Resource Development Quarterly, 14: 99-110.

Doty, D. H., & Glick, W. H. 1998. Common methods bias: Does common methods variance really bias results? Organizational Research Methods, 1: 374-406.




Dunlop, P. D., & Lee, K. 2004. Workplace deviance, organizational citizenship behavior, and business unit perfor-mance: The bad apples do spoil the whole barrel. Journal of Organizational Behavior, 25: 67-80.

Edwards, J. R., & Cable, D. M. 2009. The value of value congruence. Journal of Applied Psychology, 94: 654-677.Edwards, J., & Lambert, L. 2007. Methods for integrating moderation and mediation: A general analytical frame-

work using moderated path analysis. Psychological Methods, 12: 1-22.Ferris, D. L., Brown, D. J., & Heller, D. 2009. Organizational supports and organizational deviance: The mediating

role of organization-based self-esteem. Organizational Behavior and Human Decision Processes, 108: 279-286.

Ferris, D. L., Brown, D. J., Lian, H., & Keeping, L. M. 2009. When does self-esteem relate to deviant behavior? The role of contingencies of self-worth. Journal of Applied Psychology, 94: 1345-1353.

Ferris, D. L., Spence, J. R., Brown, D. J., & Heller, D. 2012. Interpersonal justice and workplace deviance: The role of esteem threat. Journal of Management, 38: 1788-1811.

Gordon, R. A. 1996. Impact of ingratiation on judgments and evaluations: A meta-analytic investigation. Journal of Personality and Social Psychology, 71: 54-70.

Griffeth, R. W., Hom, P. W., & Gaertner, S. 2000. A meta-analysis of antecedents and correlates of employee turnover: Update, moderator tests, and research implications for the next millennium. Journal of Management, 26: 463-488.

Harris, K. J., Kacmar, K. M., Zivnuska, S., & Shaw, J. D. 2007. The impact of political skill on impression manage-ment effectiveness. Journal of Applied Psychology, 92: 278-285.

Heatherton, T. F., & Polivy, J. 1991. Development and validation of a scale for measuring state self-esteem. Journal of Personality and Social Psychology, 60: 895-910.

Hewlin, P. F. 2003. And the award for best actor goes to . . . : Facades of conformity in organizational settings. Academy of Management Review, 28: 633-642.

Hewlin, P. F. 2009. Wearing the cloak: Antecedents and consequences of creating facades of conformity. Journal of Applied Psychology, 94: 727-741.

Hill, C. A. 1987. Affiliation motivation: People who need people . . . but in different ways. Journal of Personality and Social Psychology, 52: 1008-1018.

Hom, P. W., Mitchell, T. R., Lee, T. W., & Griffeth, R. W. 2012. Reviewing employee turnover: Focusing on proxi-mal withdrawal states and an expanded criterion. Psychological Bulletin, 138: 831-858.

Ilies, R., Scott, B. A., & Judge, T. A. 2006. The interactive effects of personal traits and experienced states on intra-individual patterns of citizenship behavior. Academy of Management Journal, 49: 561-575.

Jones, E., & Pittman, T. 1982. Toward a general theory of strategic self-presentation. In J. Suls (Ed.), Psychological perspectives on the self, vol. 1: 231-262. Hillsdale, NJ: Erlbaum.

Judge, T. A., Scott, B. A., & Ilies, R. 2006. Hostility, job attitudes, and workplace deviance: Test of a multilevel model. Journal of Management, 33: 378-394.

Kahn, R. L. 1990. Psychological conditions of personal engagement and disengagement at work. Academy of Management Journal, 33: 692-724.

Kerlinger, F. N., & Lee, H. B. 2000. Foundations of behavioral research (4th ed.). Fort Worth, TX: Harcourt College.

Leary, M. R. 1990. Responses to social exclusion: Social anxiety, jealousy, loneliness, depression, and low self-esteem. Journal of Social and Clinical Psychology, 9: 221-229.

Leary, M. R. 1995. Self-presentation. Boulder, CO: Westview Press.Leary, M. R. 2001. The self we know and the self we show: Self-esteem, self-presentation, and the maintenance of

interpersonal relationships. In G. J. O. Fletcher & M. S. Clark (Eds.), Blackwell handbook of social psychology: 457-477. Oxford, England: Blackwell.

Leary, M. R. 2010. Affiliation, acceptance, and belonging: The pursuit of interpersonal connection. In S. T. Fiske, D. T. Gilbert, & G. Lindzey (Eds.), Handbook of social psychology (5th ed.), vol. 2: 864-897. Hoboken, NJ: Wiley.

Leary, M. R., & Allen, A. B. 2010. Belonging motivation: Establishing, maintaining, and repairing relational value. In D. Dunning (Ed.), Social motivation: 37-56. New York: Psychology Press.

Leary, M. R., & Allen, A. B. 2011. Personality and persona: Personality processes in self-presentation. Journal of Personality, 79: 1191-1218.

Leary, M. R., & Downs, D. 1995. Interpersonal functions of the self-esteem motive: The self-esteem system as a sociometer. In M. Kernis (Ed.), Efficacy, agency, and self-esteem: 123-144. New York: Plenum.




Leary, M. R., Gallagher, B., Fors, E. H., Buttermore, N., Baldwin, E., Lane, K. K., & Mills, A. 2003. The invalidity of personal claims about self-esteem. Personality and Social Psychology, 81: 898-909.

Leary, M. R., Haupt, A. L., Strausser, K. S., & Chokel, J. L. 1998. Calibrating the sociometer: The relationship between interpersonal appraisals and state self-esteem. Journal of Personality and Social Psychology, 74: 1290-1299.

Leary, M. R., & MacDonald, G. 2003. Individual differences in self-esteem: A review and theoretical integration. In M. R. Leary & J. P. Tangney (Eds.), Handbook of self and identity: 410-418. New York: Guilford Press.

Leary, M. R., Tambor, E. S., Terdal, S. K., & Downs, D. L. 1995. Self-esteem as an interpersonal monitor: The sociometer hypothesis. Journal of Personality and Social Psychology, 68: 518-530.

Leary, M. R., Twenge, J. M., & Quinlivan, E. 2006. Interpersonal rejection as a determinant of anger and aggression. Personality and Social Psychology Review, 10: 111-132.

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.

Lian, H., Brown, D., Ferris, D. L., Liang, L., Keeping, L., & Morrison, R. 2014. Abusive supervision and retaliation: A self-control framework. Academy of Management Journal, 57: 116-139.

Liden, R. C., & Mitchell, T. R. 1988. Ingratiatory behaviors in organizational settings. Academy of Management Review, 13: 572-587.

Little, T. D., Cunningham, W. A., Shahar, G., & Widaman, K. F. 2002. To parcel or not to parcel: Exploring the question, weighing the merits. Structural Equation Modeling, 9: 151-173.

Little, T. D., Rhemtulla, M., Gibson, K., & Schoemann, A. M. 2013. Why the items versus parcels controversy needn’t be one. Psychological Methods, 18: 285-300.

MacKinnon, D. P., Fritz, M. S., Williams, J., & Lockwood, C. M. 2007. Distribution of the product confidence limits for the indirect effect: Program PRODCLIN. Behavior Research Methods, 39: 384-389.

MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S. G., & Sheets, V. 2002. A comparison of methods to test mediation and other intervening variable effects. Psychological Methods, 7: 83-104.

Maner, J. K., DeWall, C. N., Baumeister, R. F., & Schaller, M. 2007. Does social exclusion motivate interper-sonal reconnection? Resolving the “porcupine problem.” Journal of Personality and Social Psychology, 92: 42-55.

Marigold, D. C., Holmes, J. G., & Ross, M. 2007. More than words: Reframing compliments from romantic partners fosters security in low self-esteem individuals. Journal of Personality and Social Psychology, 92: 232-248.

Marsh, H. W., Lüdtke, O., Nagengast, B., Morin, A. J. S., & Von Davier, M. 2013. Why item parcels are (almost) never appropriate: Two wrongs do not make a right—camouflaging misspecification with item parcels in CFA models. Psychological Methods, 18: 257-284.

Mawritz, M. B., Mayer, D. M., Hoobler, J. M., Wayne, S. J., & Marinova, S. V. 2012. A trickle-down model of abusive supervision. Personnel Psychology, 65: 325-357.

Mayer, D. M., Thau, S., Workman, K. M., Van Dijke, M., & De Cremer, D. 2012. Leader mistreatment, employee hostility, and deviant behaviors: Integrating self-uncertainty and thwarted needs perspectives on deviance. Organizational Behavior and Human Decision Processes, 117: 24-40.

McGhee, P. E., & Teevan, R. C. 1967. Conformity behavior and need for affiliation. Journal of Social Psychology, 72: 117-121.

Meglino, B. M., & Ravlin, E. C. 1998. Individual values in organizations: Concepts, controversies, and research. Journal of Management, 24: 351-389.

Miller, D. T. 2001. Disrespect and the experience of injustice. Annual Review of Psychology, 52: 527-553.Mitchell, M. S., & Ambrose, M. L. 2007. Abusive supervision and workplace deviance and the moderating effects

of negative reciprocity beliefs. Journal of Applied Psychology, 92: 1159-1168.Mitchell, M. S., & Ambrose, M. L. 2012. Employees’ behavioral reactions to supervisor aggression: An examina-

tion of individual and situational factors. Journal of Applied Psychology, 97: 1148-1170.Moreno-Jiménez, B., Rodríguez-Muñoz, A., Pastor, J. C., Sanz-Vergel, A., & Garrosa, E. 2009. The moderating

effect of psychological detachment and thoughts of revenge in workplace bullying. Personality and Individual Differences, 46: 359-364.

Pierce, J., & Gardner, D. 2004. Self-esteem within the work and organizational context: A review of the organiza-tion-based self-esteem literature. Journal of Management, 30: 591-622.

Pillutla, M., & Thau, S. 2009. Actual and potential exclusion as determinants of individual’s unethical behavior in groups. In D. De Cremer (Ed.), Psychological perspectives on ethical behavior and decision-making in groups: 121-134. Charlotte, NC: IAP.




Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. 2003. Common method bias in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88: 879-903.

Preacher, K. J., Rucker, D., & Hayes, A. F. 2007. Addressing moderated mediation hypotheses: Theory, methods, and prescriptions. Multivariate Behavioral Research, 42: 185-227.

Robinson, S. L., & Bennett, R. J. 1995. A typology of deviant workplace behaviors: A multidimensional scaling study. Academy of Management Journal, 38: 555-572.

Rosenberg, M. 1965. Society and the adolescent self-image. Princeton, NJ: Princeton University Press.Sagie, A., Birati, A., & Tziner, A. 2002. Assessing the costs of behavioral and psychological withdrawal: A new

model and an empirical illustration. Applied Psychology, 51: 67-89.Schneider, B., Kristof, A. L., Goldstein, H. W., & Smith, D. B. 1997. What is this thing called fit? In N. R. Anderson

& P. Herriott (Eds.), Handbook of selection and appraisal, vol. 2: 393-412. London: Wiley.Scott, K. L., & Thau, S. 2012. Theory and research on social exclusion in work groups. In N. DeWall (Ed.),

Handbook of social exclusion: 65-73. Oxford, England: Oxford University Press.Shrout, P. E., & Bolger, N. 2002. Mediation in experimental and nonexperimental studies: New procedures and

recommendations. Psychological Methods, 7: 422-445.Smart Richman, L., & Leary, M. R. 2009. Reactions to discrimination, stigmatization, ostracism, and other forms of

interpersonal rejection: A dynamic, multimotive model. Psychological Review, 116: 365-383.Sonnentag, S., & Bayer, U. 2005. Switching off mentally: Predictors and consequences of psychological detachment

from work during off-job time. Journal of Occupational Health Psychology, 10: 393-414.Sonnentag, S., Binnewies, C., & Mojza, E. J. 2010. Staying well and engaged when demands are high: The role of

psychological detachment. Journal of Applied Psychology, 95: 964-976.Stone-Romero, E. 2002. The relative validity and usefulness of various empirical research designs. In S. G.

Rogelberg (Ed.), Handbook of research methods in industrial and organizational psychology: 77-98. London: Blackwell.

Tedeschi, J. 2001. Social power, influence, and aggression. In J. Forgas & K. Williams (Eds.), Social influence: Direct and indirect processes: 109-128. New York: Psychology Press.

Tedeschi, J., & Melburg, V. 1984. Impression management and influence in the organization. In S. Bacharach & E. Lawler (Eds.), Research in the sociology of organizations, vol. 3: 31-58. Greenwich, CT: JAI Press.

Tepper, B. J. 2000. Consequences of abusive supervision. Academy of Management Journal, 43: 178-190.Tepper, B. J., & Almeda, M. 2011. Negative exchanges with supervisors. In L. Eby & T. Allen (Eds.), Personal

relationships at work: 67-94. Washington, DC: Routledge.Tepper, B. J., Carr, J. C., Breaux, D. M., Geider, S., Hu, C., & Hua, W. 2009. Abusive supervision, intentions to

quit, and employees’ workplace deviance: A power/dependence analysis. Organizational Behavior and Human Decision Processes, 109: 156-167.

Tepper, B. J., Henle, C. A., Lambert, L. S., Giacalone, R. A., & Duffy, M. K. 2008. Abusive supervision and subor-dinates’ organizational deviance. Journal of Applied Psychology, 93: 721-732.

Tepper, B. J., Moss, S. E., & Duffy, M. K. 2011. Predictors of abusive supervision: Supervisor perceptions of deep-level dissimilarity, relationship conflict, and subordinate performance. Academy of Management Journal, 54: 279-294.

Tett, R. P., & Meyer, J. P. 1993. Job satisfaction, organizational commitment, turnover intention, and turnover: Path analyses based on meta-analytic findings. Personnel Psychology, 46: 259-293.

Thau, S., Aquino, K., & Poortvliet, P. M. 2007. Self-defeating behaviors in organizations: The relationship between thwarted belonging and interpersonal work behaviors. Journal of Applied Psychology, 92: 840-847.

Thau, S., & Mitchell, M. S. 2010. Self-gain or self-regulation impairment? Tests of competing explanations of the supervisor abuse and employee deviance relationship through perceptions of distributive justice. Journal of Applied Psychology, 95: 1009-1031.

Tofighi, D., & MacKinnon, D. P. 2011. RMediation: An R package for mediation analysis confidence intervals. Behavior Research Methods, 43: 692-700.

Turnley, W. H., & Bolino, M. C. 2001. Achieving desired images while avoiding undesired images: Exploring the role of self-monitoring in impression management. Journal of Applied Psychology, 86: 351-360.

Wade, S. H. 1989. The development of a scale to measure forgiveness. Unpublished doctoral dissertation, Fuller Theological Seminary, Pasadena, CA.

Wallace, J. C., Butts, M. M., Johnson, P. D., Stevens, F. G., & Smith, M. B. in press. A multilevel model of employee innovation: Understanding the effects of regulatory focus, thriving, and employee involvement cli-mate. Journal of Management. doi:10.1177/0149206313506462




Watson, D., Clark, L. A., & Tellegen, A. 1988. Development and validation of a brief measure of positive and nega-tive affect: The PANAS scales. Journal of Personality and Social Psychology, 54: 1063-1070.

Weiss, H. M., & Cropanzano, R. 1996. Affective events theory: A theoretical discussion of the structure, causes and consequences of affective experiences at work. Research in Organizational Behavior, 18: 1-74.

Williams, K. D., Cheung, C. K. T., & Choi, W. 2000. Cyberostracism: Effects of being ignored over the Internet. Journal of Personality and Social Psychology, 79: 748-762.

Williams, L. J., Hartman, N., & Cavazotte, F. 2010. Method variance and marker variable: A review and compre-hensive CFA marker technique. Organizational Research Methods, 13: 477-514.

Wolfe, R. N., Lennox, R. D., & Cutler, B. L. 1986. Getting along and getting ahead: Empirical support for a theory of protective and acquisitive self-presentation. Journal of Personality and Social Psychology, 50: 356-361.

Zellars, K. L., Tepper, B. J., & Duffy, M. K. 2002. Abusive supervision and subordinates’ organizational citizenship behavior. Journal of Applied Psychology, 87: 1068-1076.



Documents

The Motivational Effects of Diminished Self- Esteem for ...media.terry.uga.edu/socrates/publications/2015/01/...supervision indirectly influences employees’ workplace deviance and