When Can Forgiveness be Enhanced? It Depends on What Motivates Abusive Supervision

As Boulding (1963) explained, conflict is an awareness of the existence of discrepancies, incompatible wishes, or irreconcilable desires between or among the involved parties. In organizational settings, conflict manifests at the interpersonal, cross-departmental, and inter-organizational levels. Among these, the most studied has been supervisor-subordinate conflict (e.g., Liang et al., 2024; Schyns & Schilling, 2013; Krasikova, Green & LeBreton, 2013; Einarsen, Aasland & Skogstad, 2007; Mitchell & Ambrose, 2007). This study focuses on a specific form of supervisor-subordinate conflict - that created by the destructive leadership style of “abusive supervision” (Tepper, 2000, p. 178). While it has been presumed that abusive supervision is driven by an unwillingness of that supervisor to consider the needs of the subordinate, we examine whether some abusive behaviors may in fact be driven by concern for the subordinate (Rahim, 2011; Pruitt & Lewis, 1975) and whether claims of concern for the subordinate can diminish the level of conflict between supervisor and subordinate by enhancing forgiveness. In effect, if an employee perceives that the supervisor’s bad actions are driven by concern for the performance of the subordinate, this can serve as a kind of social account for the bad behavior (Bies & Shapiro, 1987) that can enhance a willingness to forgive (Davidson & Friedman, 1998).

We conducted three studies to test our hypotheses, two from samples in Taiwan’s military and one survey through MTurk in the U.S. Our study makes the following contribution. We introduce the idea that subordinates’ attributions of supervisory intent (whether the intent is to help the subordinate) is an important factor affecting their responses to abusive supervision, especially their willingness to forgive - which is central to conflict management (Aquino et al., 2001; Fehr et al., 2010). This work helps clarify the complex relationships among abusive supervision, the motives behind supervisors’ abusiveness, and subordinates’ cognitive and behavioral reactions to abusive supervision. In this way, it opens avenues for understanding how the conflict that comes from abusive supervision can be mitigated, and if true, what the ethical implications for companies and managers are.

Theoretical Background and Hypotheses

Agency-Communion Paradigm and Perceptions of Task Conflict Instigators

Tepper (2000, p. 178) conceptualized abusive supervision as “subordinates’ perceptions of the extent to which their supervisors engage in the sustained display of hostile verbal and nonverbal behaviors, excluding physical contact.” Abusiveness, along with other negative traits/behaviors (e.g., psychopathy or authoritarianism) of supervisors, has been labeled the “dark side of leadership” (Liu et al., 2012) and usually leads to (sometimes hidden) conflict (Michel et al., 2016; Xu et al., 2015). Abusive supervision has been shown to reduce social bonds (Xu et al., 2012), trigger retaliatory actions (Mitchell & Ambrose, 2007) and reduce trust (Vogel et al., 2015). For example, abusive supervision deteriorates the leader-member exchange (LMX) quality between supervisors and subordinates, which undermines the overall social exchange relationship. This erosion of social bonds can lead to reduced organizational citizenship behaviors (OCBs) and lower task performance, as employees withdraw their efforts and engagement in response to the poor-quality exchanges (Xu et al., 2012). Additionally, employees subjected to abusive supervision may engage in retaliatory actions, such as direct acts of defiance against the supervisor or indirect organizational deviance, driven by perceptions of injustice and reinforced by negative reciprocity beliefs (Mitchell & Ambrose, 2007). Finally, abusive supervision violates fairness principles, reducing interpersonal justice perceptions and leading employees to distrust their supervisors (Vogel et al., 2015). This lack of trust further diminishes employees’ motivation to invest in their work relationships and efforts.

Yet empirical research has uncovered psycho-physiological benefits of taking a forgiving stance toward an offenders (Cox et al., 2012; McCullough & Witvliet, 2001). According to Aquino et al. (2006), forgiveness refers to “the internal act of relinquishing anger, resentment, and the desire to seek revenge against the offender” (p. 654). In organizational contexts, forgiveness can help individuals repair damaged workplace relationships and release negative thoughts and emotions resulting from interpersonal conflict (Aquino et al., 2003).¹

Nevertheless, forgiving supervisors’ abuse can be a very difficult step for subordinates to take. According to the predictions of organizational justice theory (Greenberg, 1987), employees may have a desire to strike back and refuse to forgive when they are mistreated by supervisors (Bies & Tripp, 1996; Liu et al., 2010) in the hope that their perceived injustice can be restored (Bradfield & Aquino, 1999). This tendency may be especially likely when subordinates face supervisory abuse (Mitchell & Ambrose, 2007). Being the target of supervisor abuse is associated with anger (Mawritz et al., 2014) and distress (Tepper, 2000), both of which make forgiveness more difficult (Fehr et al., 2010). Although a negative relationship between abusive supervision and subordinate forgiveness is supported by related literature, it has received limited explicit attention in the literature (Aquino et al., 2006; Fehr et al., 2010).

One of the challenges of studying this relationship is that, while forgiveness is expected to be higher for lower levels of abuse, if there is zero abuse then there is nothing to forgive; with nothing to forgive, forgiveness might actually be very low at the lowest level of abuse. We address this challenge by focusing on general attitude of forgiveness rather than forgiveness for abuse in particular. Forgiveness is an attitude that applies to any perceived slights or missteps of a supervisor, representing a more open attitude towards the supervisor. This approach to forgiveness might best be called “generalized forgiveness.” We expect that both low abuse and zero abuse would create a context for lower generalized forgiveness than high abuse. This is an approach used, for example, in studies by Karremans et al. (2005) and Karremans and Van Lange (2008). In both studies, victims’ forgiving was seen as a general attitude towards the transgressor. Karremans et al. (2005) proposed that forgiveness is an internal shift that reshapes how people think and act socially. It strengthens relationships, encourages prosocial behavior, and supports mental health. As such, forgiveness serves as a broad psychological mechanism with important social functions. Thus, we propose:

H1. Abusive supervision is negatively related to subordinate generalized forgiveness.

Forgiveness and OCB

Organ (1988) coined the term “organizational citizenship behavior” (OCB) to describe employees’ voluntary contributions to the organization beyond their required tasks (Organ, 1997). Later, scholars proposed the concept of “organizational citizenship behavior—individuals” (OCBI)², whose target or beneficiary behaviors are directed at individuals rather than the organization (Aryee et al., 2007; Lee & Allen, 2002).

As Aryee et al. (2007) depicted, supervisor-related outcomes, such as organizational citizenship behavior, tend to be related to subordinates’ perceptions of interactional justice. Karremans and Van Lange (2008) showed that, after forgiving, victims were more likely to make personal sacrifices for their offenders than before they forgave, namely cooperating with them in social dilemma tasks. In another study, Karremans et al. (2005) found that victims who had forgiven past offenses were more willing than those who had not to volunteer for a charity organization and donate money. That is, forgiveness restores victims’ prosocial orientation.

We propose that when generalized forgiveness is enhanced, victims of abusive supervision can shift their focus away from injustice and toward recovery. During this process, victims’ sense of injustice lessens because of their decreased focus on the negative aspects of the relevant events (Barclay & Saldanha, 2016). According to Fredrickson’s broaden-and-build theory of positive emotions (Fredrickson, 2001), a forgiving person is equipped with positive emotions, which in turn promote future-oriented thinking (Fredrickson, 2001; Foo et al., 2009) and proactive behaviors (Foo et al., 2009; Grant & Ashford, 2008). We predict that employees who no longer dwell on mistreatment are likely to perform behaviors beyond their call of duty, such as helping coworkers at work. Having an attitude of generalized forgiveness should enhance subordinates’ OCB.

H2. Subordinates who approach supervisors with an attitude of generalized forgiveness will engage in higher levels of organizational citizenship behaviors.

Mediation through Forgiveness

Hypotheses 1 and 2 together suggest a mediating effect, with abusive supervision predicting forgiveness, and forgiveness in turn enhancing OCBs. We propose that victims of abusive supervisions will transfer their resentments by withdrawing helping behaviors to coworkers or supervisors such as organizational citizenship behavior, and inversely that if resentments are kept lower, there will be higher levels of organizational citizenship behavior.

H3. There is an indirect, negative effect of abusive supervision on organizational citizenship behavior through forgiveness.

Supervisor Performance Promotion Motives as Moderator

Next, we turn to a discussion of one factor that we expect to reduce subordinates’ resistance to forgive - the perception that the motivation behind a supervisor’s abusive behavior is a goal of improving subordinates’ performance. In Zhang and Bednall’s (2016) meta-analytic review, they identified four categories of antecedents of abusive supervision: (1) supervisor-related factors (e.g., stressors and negative affective state), (2) organization-related factors (e.g., aggressive norms), (3) subordinate-related factors (e.g., subordinates’ personality such as narcissism or neuroticism), and (4) demographic characteristics of supervisors and subordinates (e.g., supervisors’ age or subordinates’ age, or gender dissimilarity of subordinates and supervisors). In other words, it is possible that supervisory abuse is caused by the supervisor being stressed out, in an organization with strong norm of aggression, or he/she has a bad personality. We propose an additional subordinate-related motivation for abusive supervision, which is supervisors’ “performance promotion motives” (Liu et al., 2012).

Liu et al. (2012) argue that supervisor aggressive behaviors may not be driven by a motive to “injure” or harm the subordinate but can sometimes be driven by a motive to promote better long-term performance by the subordinate. This idea counters a core assumption of the abusive supervision literature: the belief that the offender necessarily holds malice toward the victim and does not care about the latter’s well-being (Malle & Knobe, 1997; Struthers Eaton et al., 2008). It allows for the possibility that supervisors’ abusive acts may actually be driven by subordinate-centered motives--inappropriate behavior does not necessarily imply a lack of other-centered motives and victims of abusive supervision may take that into account when judging abusive supervisors.

According to Liu et al. (2012), subordinates are eager to decipher the motives underlying supervisory abuse, and their attributions of supervisory intent are an important component of their sense-making process (Weick, 1995). When subordinates attribute (Heider, 1958; Weiner, 1995) supervisors’ abusive behaviors to a performance promotion motivations by the supervisor, it means that they see the behavior as externally driven (by the supervisor’s assessment of what that subordinate needs to do better) rather than internally drive (by, for example, the supervisor’s inherent tendency to treat all people in an uncaring and abusive way). Based on prior research (e.g., Kim et al., 2006; Weiner et al., 1987), victims’ external attributions should reduce anger at the abusive supervisor and allow for trust to be repaired (Kim et al., 2006; Kim et al., 2004).

This is consistent with the literature on social accounts, which shows that one of the primary ways to manage perceptions of injustice and reduce conflict is to shape perceptions of the intent of the harm-doer. Blame, anger, and conflict are often lessened by providing excuses that shown that the harm was not intended (cf. Riordan et al., 1983; Sitkin & Bies, 1993; Shaw et al. 2003). In the case of supervisor performance motivations, the stress experienced by the subordinate is indeed intended - that is the whole point of the performance motivation - but what is not intended is “harm”. Rather, what is intended is that abuse should increase subordinate performance, resulting in subordinate growth and success.

The idea that shaping perceptions of supervisor intent can mitigate the harm of abusive supervision can be substantiated by literature on ethical leadership (Brown et al., 2005; Treviño et al., 2000) and moral disengagement (Bandura, 1999). Brown et al. (2005, p.120) define ethical leadership as “the demonstration of normatively appropriate conduct through personal actions and interpersonal relationships.” While “abuse” in itself is unlikely to be deemed normatively appropriate, striving to enhance subordinate growth and development is likely to be deemed normatively appropriate. Tough love may be painful, but it can be portrayed to be based on caring. Such a moral framing can shape how subordinates assign meanings to such behaviors, and in turn mitigate the negative effect of supervisory abuse on forgiveness.

So far, we have focused on the issue of whether supervisors actually do care for subordinates. But there is certainly an incentive for a supervisor to falsely portray their motives as being driven by concern for the subordinate’s performance, and even false portrayals can help reduce blame against the abusive supervisor. Bandura (1999) defines “moral justification” as a cognitive reframe of unethical acts as being in the service of a greater good, which what happens with an abusive supervisor conveys the impression that their motivation is to make the subordinate more effective, which helps both the subordinate and the company.

Drawing upon the above arguments, we propose that subordinate attributions of supervisors’ performance promotion motives (regardless of actual caring) play a mitigating role in the negative relationship between abusive supervision and subordinate forgiveness. When viewing abusive supervision through a less negative lens, the abused subordinates may be more willing to forego blaming their pain on their supervisor’s disposition and let go of the conflict with him/her (cf. McCullough et al., 1998). Thus, we expect to find that abusive supervision is less negatively related to subordinate generalized forgiveness in the presence of subordinate-attributed supervisor performance promotion motives (cf. Farmanara, 2021).

H4. Abusive supervision and subordinate-attributed performance promotion motives of the supervisor interact to predict subordinate generalized forgiveness of the supervisor, such that abusive supervision has a less negative relationship with subordinate forgiveness when subordinate-attributed performance promotion motives are high as compared to when they are low.

Lastly, we integrate the above moderating prediction with the mediating hypothesis proposed earlier and propose the following moderated-mediation hypothesis.

H5. Forgiveness will mediate the interactive effect of abusive supervision and supervisors’ motives on organizational citizenship behavior. The indirect, negative effect of abusive supervision on organizational citizenship behavior through generalized forgiveness is less strong when subordinate-attributed performance promotion motives are high than when they are low.

Study 1

Method

Participants and Procedure

To test our hypotheses, we conducted a two-wave field survey with voluntary military personnel in northern Taiwan as our respondents. We obtained permission from headquarters to conduct the survey. We assured the anonymity of our survey, namely that responses would be held completely confidential and used only for research purposes. During the first wave of data collection (Time 1), we distributed the survey to 360 respondents; 343 were returned (95.3% response rate). One-and-a-half months later (Time 2), we distributed 360 survey questionnaires in total, and 295 respondents completed them (81.9% response rate). However, after we referenced respondents’ demographics (they filled out their military ID numbers in both surveys) and kept those who participated in both surveys, the final sample became 188, a much smaller sample size. The large drop in response rate was due to the fact that some respondents from the first-wave survey were not available during the time we collected the second-wave survey (e.g., they were on duty or taking the day off) or had been transferred to other units during the time interval. Among the 188 respondents, there were 170 males and 18 females, and the average age of the respondents was 23.28 years (SD = 3.49).

Measures

Since all measures were originally constructed in English, we followed a back-translation process to ensure that the Chinese versions of all items were comparable to the original version at a high degree of accuracy (Brislin, 1986). Measures of abusive supervision and subordinate-attributed performance promotion motives were included in the first-wave survey, and those of forgiveness and OCB were included in the second-wave survey. We used a five-point scale throughout the questionnaire.

Abusive supervision. Subordinates rated their supervisors’ abusive supervision using Aryee et al.’s (2007) 10-item version of the 15-item scale developed by Tepper (2000). This shortened version has been widely validated in non-Western contexts (e.g., Aryee et al., 2008; Chi & Liang, 2013; Farh & Chen, 2014). Sample items are: “My immediate supervisor tells me my thoughts or feelings are stupid,” “My immediate supervisor makes negative comments about me to others,” and “My immediate supervisor blames me to save himself/herself embarrassment.” The Cronbach’s alpha coefficient for this scale was .93.

Subordinate-attributed performance promotion motives. We used five items from an instrument developed by Liu et al. (2012) to measure subordinate-attributed performance promotion. In our instructions, we provided a definition of abusive supervision and asked respondents to reply specifically in relation to supervisory abusiveness, i.e., the extent to which they agree that the descriptions presented represent the motives of their supervisor’s abusiveness. A sample item for performance promotion motives is: “My immediate supervisor desires to elicit high performance from me.” The Cronbach’s alpha coefficient for this measure was .84.

Generalized forgiveness. We measured subordinate generalized forgiveness using four items developed by Bradfield and Aquino (1999). We modified the original items to refer to “my supervisor.” Some sample items are “I gave my supervisor back a new start, a renewed relationship” and “I wanted to be more friendly and concerned towards my supervisor.” This measure of forgiveness is not specifically about forgiveness for abuse, but should nonetheless encompass forgiveness for abuse, as well as forgiveness for supervisor’s other behaviors, e.g., miscalculated decisions or no decision that affect subordinates’ welfare, or behaviors that are unethical and undermine leadership. The Cronbach’s alpha coefficient for this scale was .87.

OCB. We adopted Lee and Allen’s (2002) eight-item OCBI scale. A sample item is, “I help others who have been absent.” The Cronbach’s alpha coefficient for this scale was .94.

Control variables. We controlled for respondents’ sex, age, and tenure with their supervisor in our analyses, because past studies have found significant effects of these variables on employees’ behavioral reactions to abusive supervision (e.g., Chi & Liang, 2013; Mitchell & Ambrose, 2007; Liu et al., 2012).

Results

Descriptive Statistics

Table 1 provides means, standard deviations, reliability coefficients, and correlations for the study variables. Consistent with our purposes, an examination of the zero-order correlations demonstrated that abusive supervision was negatively correlated with forgiveness and OCB (r = -.15, p < .05 and r = -.23, p < .01, respectively), and forgiveness was positively related to OCB (r = .42, p < .01).

Table 1 Descriptive Statistics and Correlations in Study 1

Note. n = 188; SAPPM = subordinate-attributed performance promotion motives, OCB = organizational citizenship behavior; Coefficient α reliabilities are reported in parentheses on the diagonal. ^a Gender: 0 = female, 1 = male. ^* p < .05; ^** p < .01 (two-tailed tests)

Confirmatory Factor Analyses

Prior to hypotheses testing, we used confirmatory factor analyses (CFA) procedures to test the distinctiveness of the four study variables: abusive supervision, subordinate-attributed performance promotion motives, forgiveness, and OCB. We first specified a model in which all the items loaded on their respective underlying constructs. The hypothesized four-factor model yielded a better fit into the data (χ2 = 592.21, df = 318, p < .01; root mean square error of approximation [RMSEA] = .07; standardized root-mean-square residual [SRMR] = .07; comparative fit index [CFI] = .96; Tucker-Lewis index [TLI] = .96). We further compared these fit results with those of alternative three-factor, two-factor, and one-factor models. First, the three-factor model combined forgiveness and OCB as one latent factor. Furthermore, the two-factor model included subordinate-attributed performance promotion motives, forgiveness, and OCB as one factor. Finally, the one-factor model loaded all items onto a single factor. As shown in Table 2, the four-factor model had a superior fit over each alternative model. Therefore, the CFA results supported the distinctiveness of the four study variables for subsequent analyses.

Table 2 Confirmatory Factor Analyses Results in Study 1

Note. AS = abusive supervision, SAPPM = subordinate-attributed performance promotion motives, FOR = forgiveness, OCB = organizational citizenship behavior; The χ² difference was compared with the value of the four-factor model (baseline model). ^**p < .01 (two-tailed tests).

Hypotheses Testing

We conducted hierarchical regression analyses to examine our hypotheses. As shown in Model 1 of Table 3, abusive supervision was significant in predicting forgiveness (b = -.17, p < .05). Hypothesis 1 was supported.

Table 3 Regression Results in Study 1

Note. n = 188. Employee sex: Male=1, Female=0; SAPPM = subordinate-attributed performance promotion motives, OCB = organizational citizenship behavior. † p < .1; ^* p < .05; ^** p < .01 (two-tailed tests)

Furthermore, as shown in Model 4 of Table 3, forgiveness was significant in predicting OCB (b = .34, p < .01). Hypothesis 2 was supported. Also, as shown in Table 3: (1) in Model 3, abusive supervision significantly predicted OCB (b = -.20, p < .01); (2) in Model 1, abusive supervision was significantly related to forgiveness (b = -.17, p < .05); and (3) in Model 5, when forgiveness was entered into the equation, the coefficients of abusive supervision decreased (b = -.15, p < .05), and forgiveness was still found to be positively related to OCB (b = .32, p < .01). We also conducted the Process to confirm the mediation effects, and the result showed that the indirect effect of abusive supervision on OCB through forgiveness was not statistically significant (b = -.05, BootSE = .03), with the 95% bootstrap confidence interval ranging from -.13 to .00, including zero. Although the total effect of abusive supervision on OCB was significant (b = -.20, p < .01), its direct effect after accounting for forgiveness remained statistically significant though reduced in magnitude (b = -.15, p < .05). With these findings, Hypothesis 3 was not supported.

For Hypothesis 4, we proposed that subordinate-attributed performance promotion motives would moderate the relationship between abusive supervision and forgiveness. Following Aiken and West (1991), we centered the continuous predictors before creating the interaction terms. As shown in Model 2 of Table 3, the interaction term of abusive supervision and subordinate-attributed performance promotion motives was significant in predicting subordinates’ forgiveness (b = .19, p < .05). The pattern of the interaction is shown in Figure 1. There was a stronger negative association between abusive supervision and subordinates’ forgiveness when subordinate-attributed performance promotion motives were low as compared to when they were high. Therefore, Hypothesis 4 was supported.

Figure 1 The Interactive Effect of Abusive Supervision and Subordinate-attributed Performance Promotion Motives on Forgiveness (Study 1)

For Hypothesis 5, we tested first-stage moderation models using moderated path analysis developed by Edwards and Lambert (2007). Hypothesis 5 predicted that subordinate-attributed performance promotion motives would moderate the indirect effect of abusive supervision on OCB via subordinate forgiveness. We used the information from the regression results in Models 2 and 5 of Table 3 to conduct path moderation analyses at high and low levels of subordinate-attributed performance promotion motives. The moderated path analysis results are shown in Table 4. The path estimates revealed that the indirect effect of abusive supervision on OCB via subordinate forgiveness varied across low (P_MXP_YM = -.13, p < .01) and high (P_MXP_YM = -.04, n.s.) levels of subordinate-attributed performance promotion motives; the difference between the two effects was significant ([-.04] - [-.13] =.09, p < .05). Thus, Hypothesis 5 received support.

Table 4 Moderated Path Analysis Results in Study 1

Note. n = 188. ^aP_MX: path from abusive supervision to forgiveness; P_YM: path from forgiveness to organizational citizenship behavior; P_YX: path from abusive supervision to organizational citizenship behavior. ^bLow moderator variable refers to one standard deviation below the mean of the moderator; high moderator variable refers to one standard deviation above the mean of the moderator. † p < .1; ^* p < .05; ^** p < .01 (two-tailed tests)

Discussion

The results of Study 1 showed that subordinates’ attributions of supervisors’ performance promotion motives provide subordinates with a certain degree of justification of supervisory abuse that could allow them to choose to move on and cope with uncomfortable experiences. However, Study 1 may not have precluded the possibility that supervisors’ self-centeredness lay behind such performance promotion motives. That is, it is possible that subordinates see supervisors’ emphasis on subordinate performance promotion as being for the benefit of supervisors themselves? That the enhancement of subordinates’ performance is in fact driven by supervisors’ desire to achieve their personal goals in the organization. To clarify this issue, we conducted another study to examine the perceived intentions behind supervisors’ emphasis on subordinates’ performance promotion and whether supervisor self-centeredness vs. subordinate-centeredness affects subordinates’ responses.

Study 2

In Study 1 we looked at whether supervisors were motivated by a desire to enhance subordinate performance when acting abusively. That enhanced performance could benefit the subordinate, but it also may benefit the supervisor. Subordinates’ ability to forgive may depend not just on whether their supervisors used performance enhancement to benefit them, the subordinates, or as a vehicle to accomplish the supervisors’ own personal goals. That is, is the supervisor goal of subordinate performance enhancement pro-self or prosocial?

Past studies on social motives and on pro-self vs. prosocial personality traits (also called social value orientation) (e.g., De Cremer & Van Lange, 2001; De Dreu et al., 2000) discovered differential patterns of behavior from people who assign only positive weight to outcomes for the self and negative weight to outcomes for others (termed “pro-selfs”), versus people who not only assign a positive weight to outcomes for the self, but also a positive weight to outcomes for others (termed “prosocials”). Scholars have found that such a distinction in behavioral patterns is consequential to people’s interactions with others (Olekalns & Smith, 2003; Balliet & Joireman, 2010). In our context, are supervisors’ performance promotion motives self-centered or subordinate-centered. On the one hand, although supervisors with self-centered motives are concerned about their subordinates’ performance in the long run, these supervisors’ main goal is to enhance their own benefits through heightened subordinate performance. For them, subordinates’ high performance is instrumental to attaining their own goals. If subordinates believe that abusive supervisors have self-centered motives for promoting their performance, this may inhibit their willingness to forgive.

Therefore, we propose that self- vs. subordinate-centered motives play a moderating role on the relationship between abusive supervision and subordinate forgiveness. We suggest that perceiving that the supervisor has subordinate-centered (vs. self-centered) motives will mitigate the negative impact of abusive supervision on subordinate forgiveness. We predict that abusive supervision is less negatively related to subordinate forgiveness when supervisors are seen as having subordinate-centered motives than self-centered motives. We propose:

H6. Abusive supervision and supervisors’ motives interact in predicting subordinate generalized forgiveness, such that abusive supervision has a less negative relationship with subordinate forgiveness when the supervisor’s motives are seen to be subordinate-centered (rather than self-centered).

Lastly, we integrate the above moderating prediction with the mediating hypothesis and propose the following moderated-mediation hypothesis.

H7. Forgiveness will mediate the interactive effect of abusive supervision and supervisors’ motives on organizational citizenship behavior. The indirect, negative effect of abusive supervision on organizational citizenship behavior through generalized forgiveness is less strong when supervisors’ motives are subordinate-centered than when they are self-centered.

Method

Participants and Procedure

To test our hypotheses, we conducted an experimental study with 160 undergraduates (27.5% women and 72.5% men; mean age = 20.14 years, SD = 1.37) enrolled in a northern military university in Taiwan. We assured the anonymity of our survey, namely that responses would be held completely confidential and used only for research purposes. We manipulated the independent variable with a modified version of the individual abusive supervision scenario (high vs. low) designed by Farh and Chen (2014). We further manipulated the moderator with performance promotion motives (self-centered vs. subordinate-centered) designed for this study. We randomly assigned the participants to four conditions and counterbalanced the order of manipulations for abusive supervision and performance promotion motives. Following the scenarios, participants completed manipulation checks, measures of forgiveness, and OCB. They also completed the measure of demographic information as controls.

Manipulations: Abusive supervision

We modified the version of the individual abusive supervision scenario from Farh and Chen (2014). In the original scenario, subordinates were said to have received a personalized e-mail from their direct supervisor whose content was either abusive or non-abusive in nature. We modified the content to refer to direct talks between supervisor and subordinate. In the high abuse condition, the participants were shown derogatory comments made by the supervisor along with information about the participants’ own low performance as the subordinate. In the low abuse condition, participants were only shown information about their own low performance without derogatory comments from the supervisor.

Manipulations: Performance promotion motives

After the abusive supervision manipulation in the case scenario, we continued to manipulate performance promotion motives. In the story, we wrote that the supervisor had left the office to welcome an important guest. While the supervisor was away, the respondent accidentally noticed a memo written by the supervisor to himself/herself on the table. In the self-centered condition, the memo indicated that if the subordinate’s performance goals were not accomplished, the supervisor’s promotion would be jeopardized. In the subordinate-centered condition, the memo indicated that if the subordinate’s performance goals were not accomplished, the subordinate’s promotion would be jeopardized.

Measures

As in Study 1, we used a five-point scale throughout the questionnaire. For the manipulation check of self- vs. subordinate-centered performance promotion motives condition, we asked participants to respond to a one-item question after reading the case scenario. The other scales are as follows.

Abusive supervision. As a manipulation check, after the case scenario, respondents rated how they perceived the degree of supervisors’ abusive supervision by using the same scale from Aryee et al. (2007) as in Study 1. The Cronbach’s alpha coefficient was .88.

Generalized forgiveness. Forgiveness towards supervisor was assessed with the same Bradfield and Aquino (1999) scale used in Study 1. The Cronbach’s alpha coefficient for this scale was .86.

OCB. We again adopted Lee and Allen’s (2002) OCBI scale to assess a person’s tendency to perform OCB. The Cronbach’s alpha coefficient was .96.

Control variables. We controlled participants’ gender, and age in our analyses.

The Cronbach’s alpha coefficient for this scale was .85.

Manipulation checks

First, we wanted to make sure that the manipulation was successful. Participants were randomly assigned into one of the four sets of scenarios. Analysis of variance tests indicated a main effect of the abuse supervision condition (mean high = 2.99, mean low = 2.01, F(1, 158) = 163.88, p < .01, η²= .24). Regarding the performance promotion motives condition, analysis of variance tests indicated a main effect for performance promotion motives (subordinate-centered = 4.08, self-centered = 2.84, F(1, 158) = 77.78, p < .01, η²= .79). These results provided strong evidence for the validity of the two manipulations in our scenario.

Results

Descriptive Statistics

Table 5 provides means, standard deviations, reliability coefficients, and correlations for Study 2 variables. Consistent with what we found in Study 1, the results of zero-order correlations showed that abusive supervision (dummy low = 0, high = 1) was negatively correlated with forgiveness and OCB (r = -.36, p < .01 and r = -.46, p < .01, respectively). In addition, forgiveness was positively related with OCB (r = .36, p < .01).

Table 5 Descriptive Statistics and Correlations in Study 2

Note. n = 160; Coefficient α reliabilities are reported in parentheses on the diagonal. Abusive supervision dummy low = 0, high = 1; PPM = performance promotion motives (dummy self-centered = 0, subordinate-centered = 1); OCB = organizational citizenship behavior; ^a Gender: 0 = female, 1 = male. ^* p < .05; ^** p < .01 (two-tailed tests).

Confirmatory Factor Analyses

Prior to hypothesis testing, we used confirmatory factor analyses (CFA) procedures to test the distinctiveness of the two study variables: forgiveness and OCB.

Table 6 Confirmatory Factor Analyses Results in Study 2

Note. OCB = organizational citizenship behavior; The χ² difference was compared with the value of the three-factor model (baseline model). ^**p < .01 (two-tailed tests). ^* p < .05; ^** p < .01 (two-tailed tests).

We first specified a model in which all the items loaded on their respective underlying constructs. The hypothesized two-factor model yielded a better fit into the data (χ2 = 145.54, df = 53, p < .01; root mean square error of approximation [RMSEA] = .10; standardized root-mean-square residual [SRMR] = .04; comparative fit index [CFI] = .95; Tucker-Lewis index [TLI] = .94). We further compared these fit results with one-factor models. For the one-factor model loaded all items onto a single factor. As shown in Table 6, the two-factor model had a superior fit over each alternative model. Therefore, the CFA results supported the distinctiveness of the two study variables for subsequent analyses.

Hypotheses Testing

We conducted hierarchical regression analyses to examine Hypotheses 6 and 7. For Hypothesis 6, we used the dummy variable to create the interaction term of abusive supervision (high vs. low) and performance promotion motives (self-centered vs. subordinate-centered). As shown in Model 2 of Table 7, the interaction term of abusive supervision and supervisor’s performance promotion motives was significant in predicting subordinates’ forgiveness (b = .54, p < .05). The pattern of the interaction effect is shown in Figure 2. There was a less negative association between abusive supervision and subordinates’ forgiveness when supervisors’ motives are subordinate-centered as compared to when they are self-centered. Therefore, Hypothesis 6 was supported.

Table 7 Regression Results in Study 2

Note. n = 160. Abusive supervision dummy low = 0, high = 1; PPM = performance promotion motives (dummy self-centered = 0, subordinate-centered = 1); OCB = organizational citizenship behavior; Change in R^2: Compare with the model that estimates only the control variables and abusive supervision; † p < .1; ^* p < .05; ^** p < .01 (two-tailed tests).

Figure 2 The Interactive Effect of Abusive Supervision and Supervisor’s Performance Promotion Motives on Forgiveness (Study 2)

Hypothesis 7 predicted that forgiveness mediates the interactive effect of abusive supervision and supervisors’ (subordinate vs. self-centered) motives on OCB. Before we examine the moderated-mediation effect, we first tested whether forgiveness mediated the relationship between abusive supervision and OCB. As shown in Table 7: (1) in Model 3, abusive supervision significantly predicted OCB (b = -.93, p < .01); (2) in Model 1, abusive supervision was significantly related to forgiveness (b = -.60, p < .01); and (3) in Model 5, when forgiveness was entered into the equation, the regression coefficient of abusive supervision was slightly decreased (b = -.78, p < .01), and forgiveness remained positively related to OCB (b = .24, p < .01). Hence, forgiveness mediated the relationship between abusive supervision and OCB. We then adopted the moderated path analysis approach of Edwards and Lambert (2007) to examine our moderated-mediation prediction. We used the coefficients from the regression results in Models 2 and 5 of Table 7 to conduct path moderation analyses by self-centered vs. subordinate-centered motives. The moderated path analysis results are shown in Table 8. The path estimates revealed that the indirect effect of abusive supervision on OCB via subordinate forgiveness varied across low (self-centered) (P_MXP_YM = -.21, p < .05) and high (subordinate-centered) (P_MXP_YM = -.08, p = n.s.) levels of supervisor’s motives; the difference between the two effects was significant ([-.08] - [-.21] =.13, p < .05), showing that the indirect effect was less negative when the supervisor’s motives were subordinate-centered rather than self-centered. Thus, Hypothesis 7 received support.

Table 8 Moderated Path Analysis Results in Study 2

Note. n = 160. ^aP_MX: path from abusive supervision to forgiveness; P_YM: path from forgiveness to organizational citizenship behavior; P_YX: path from abusive supervision to organizational citizenship behavior. † p < .1; ^* p < .05; ^** p < .01 (two-tailed tests).

Study 3

Method

Participants and Procedure

Since we conducted Study 1 and 2 using samples from Taiwan’s military, we collected a third sample from a Western cultural context to improve the generalizability of our findings. We recruited 200 participants through Amazon’s Mechanical Turk (MTurk). Eligible participants were at least 18 years old and worked more than 35 hours per week. To mitigate common method bias, data were collected in two waves, spaced two weeks apart (Podsakoff et al., 2003). Participants were informed that the study would involve completing multiple surveys over time. The first survey included informed consent, demographic questions, and measures of abusive supervision, performance-promotion motives and control variables. The second survey measures of organizational citizenship behavior directed toward individuals (OCBI). Participants received $1.50 for completing the first survey and an additional $2.50 for the second survey as an incentive.

After excluding those who failed the attention check in the first survey, we retained 195 valid responses. In the second wave, we received 164 completed responses. After removing subjects who failed the attention test, 159 valid responses remained yielding a follow-up rate of 82%. Among the 159 participants, 96 identified as male and 63 as female, with an average age of 32.45 years (SD = 5.35). On average, they had worked with their current supervisor for 45.7 months (SD = 28.53). In terms of racial demographics, 155 participants identified as White, 3 as African American, and 1 as Black. The sample represented a variety of industries, including High Technology (n = 57), Banking or Financial Institutions (n = 36), Media/Manufacturing (n = 27), Wholesale and Retail Trade (n = 22), Government/Education (n = 6), Shipping and Warehousing (n = 5), Service (n = 5), and Other (n = 2).

Measures

We used the same measures to assess abusive supervision, subordinate-attributed performance promotion motives, and forgiveness as Study 1. As to the scale for respondents to answer, instead of using the 5-point Likert scale throughout the questionnaire, we adopted the scales that were used in the original measures.

Abusive supervision. We used Aryee et al.’s (2007) 10-item scale with a 5-point Likert scale. The Cronbach’s alpha coefficient for this scale was .92.

Subordinate-attributed performance promotion motives. We used Liu et al.’s (2012) 5-item scale with a 7-point Likert scale. The Cronbach’s alpha coefficient for this measure was .77.

Generalized Forgiveness. We used Bradfield and Aquino’s (1999) 4-item scale with a 5-point Likert scale. The Cronbach’s alpha coefficient for this scale was .70.

OCB. We adopted Lee and Allen’s (2002) 8-item scale with a 7-point Likert scale. The Cronbach’s alpha coefficient for this scale was .60.

Control variables. As in Study 1, we also controlled for respondents’ gender, age, and years of tenure with their supervisor.

Results

Descriptive Statistics

Table 9 provides means, standard deviations, reliability coefficients, and correlations for the study variables. Consistent with our purposes, an examination of the zero-order correlations demonstrated that abusive supervision was positively correlated with subordinate-attributed performance promotion motives (r = .38, p < .01), forgiveness (r = .19, p < .01), and OCB (r = .22, p < .01). Subordinate-attributed performance promotion motives of supervisor was also positively associated with forgiveness (r = .25, p < .01), but not OCB (r = .07, n.s.). Furthermore, forgiveness was positively related to OCB (r =.30, p < .01).

Table 9 Descriptive Statistics and Correlations in Study 3

Note. n = 159; SDS = subordinate deviance toward supervisors, SAPPM = subordinate-attributed performance promotion motives, OCB = organizational citizenship behavior; Coefficient α reliabilities are reported in parentheses on the diagonal. ^a Gender: 0 = female, 1 = male. ^* p < .05; ^** p < .01 (two-tailed tests).

Confirmatory Factor Analyses

Prior to hypotheses testing, we used confirmatory factor analyses (CFA) procedures to test the distinctiveness of the four variables: abusive supervision, subordinate-attributed performance promotion motives, forgiveness, and OCB. We first specified a model in which all the items loaded on their respective underlying constructs. The hypothesized four-factor model yielded a better fit into the data (χ² = 581.44, df = 318, p < .01; root mean square error of approximation [RMSEA] = .07; standardized root-mean-square residual [SRMR] = .08; comparative fit index [CFI] = .82; Tucker-Lewis index [TLI] = .81). We further compared these fit results with those of alternative three-factor, two-factor, and one-factor models. First, the three-factor model combined forgiveness and OCB as one latent factor. Furthermore, the two-factor model included subordinate-attributed performance promotion motives, forgiveness, and OCB as one factor. Finally, the one-factor model loaded all items onto a single factor. As shown in Table 10, the four-factor model had a superior fit over each alternative model. Therefore, the CFA results supported the distinctiveness of the four study variables for subsequent analyses.

Table 10 Confirmatory Factor Analyses Results in Study 3

Note. AS = abusive supervision, SAPPM = subordinate-attributed performance promotion motives, OCB = organizational citizenship behavior; The χ² difference was compared with the value of the four-factor model (baseline model). ^**p < .01 (two-tailed tests).

Hypotheses Testing

We conducted hierarchical regression analyses to examine our hypotheses. As shown in Model 1 of Table 11, abusive supervision was positively associated with forgiveness (b = .11, p < .05), contrary to our prediction of a negative relationship. Thus, Hypothesis 1 was not supported.

Furthermore, as shown in Model 4 of Table 11, forgiveness was significant in predicting OCB (b = .19, p < .01). Hypothesis 2 was supported. Also, as shown in Table 11: (1) in Model 3, abusive supervision significantly predicted OCB (b = .09, p < .05); and (2) in Model 5, when forgiveness was entered into the equation, the coefficients of abusive supervision decreased (b = .07, p < .10), and forgiveness was still found to be positively related to OCB (b = .17, p < .01). We also conducted the Process to confirm the mediation effects, and the result showed that the indirect effect of abusive supervision on OCB through forgiveness was not statistically significant (b = .02, BootSE = .01), with the 95% bootstrap confidence interval ranging from -.00 to .05, including zero. Although the total effect of abusive supervision on OCB was significant (b = .09, p < .05), its direct effect after accounting for forgiveness dropped to marginal significance (b = .07, p < .10). With these findings, Hypothesis 3 was not supported.

We continued to test Hypothesis 4, i.e., subordinate-attributed performance promotion motives moderate the relationship between abusive supervision and forgiveness. Following Aiken and West’s (1991) recommendation, we mean-centered the continuous variables before creating interaction terms. As shown in Model 2, the interaction between abusive supervision and subordinate-attributed performance promotion motives significantly predicted forgiveness (b = .16, p < .01). As illustrated in Figure 3, when subordinates reported high levels of performance promotion motives, the positive relationship between abusive supervision and forgiveness was stronger than when attribution was low, which means that supervisory performance promotion motives made abusive supervision have a positive impact on subordinate forgiveness. Hypothesis 4 was partially supported, i.e., the moderating effect was significant with a consistent pattern, but the direction of the relationship between abusive supervision and forgiveness was not as predicted. We will discuss the possible rationale for this finding in the Discussion section.

Table 11 Regression Results in Study 3

Note. n = 159. Employee sex: Male=1, Female=0; SAPPM = subordinate-attributed performance promotion motives, OCB = organizational citizenship behavior. † p < .1; ^* p < .05; ^** p < .01 (two-tailed tests)

Figure 3 The Interactive Effect of Abusive Supervision and Subordinate-attributed Performance Promotion Motives on Forgiveness (Study 3)

Table 12 Moderated mediation Path Analysis Results in Study 3

Note. n = 159. ^aP_MX: path from abusive supervision to forgiveness; P_YM: path from forgiveness to organizational citizenship behavior; P_YX: path from abusive supervision to organizational citizenship behavior. ^b Low moderator variable refers to one standard deviation below the mean of the moderator; high moderator variable refers to one standard deviation above the mean of the moderator. † p < .1; ^* p < .05; ^** p < .01 (two-tailed tests)

Hypothesis 5 (retested in Study 3) predicted that subordinate-attributed performance promotion motives would moderate the indirect effect of abusive supervision on OCB via subordinate forgiveness. We used the information from the regression results of Table 12 to conduct path moderation analyses at high and low levels of subordinate-attributed performance promotion motives. The moderated path analysis results are shown in Table 12. The path estimates revealed that the indirect effect of abusive supervision on OCB via subordinate forgiveness varied across low (P_MXP_YM = -.02, n.s.) and high (P_MXP_YM = .04, p < .05.) levels of subordinate-attributed performance promotion motives; the difference between the two effects was significant ([.04] - [-.02] =.06, p < .05). The total effect of abusive supervision on OCB via subordinate forgiveness varied across low (P_YX + P_MXP_YM = .05, p < .10) and high (P_YX + P_MXP_YM = .11, p < .01) levels of subordinate-attributed performance promotion motives; the difference between the two effects was significant ([.11] - [.05] =.06, p < .05). We note that this result (a significantly positive difference between the two effects) was consistent with our previous findings in Study 1 and 2, providing evidence for a moderated mediation effect, even though we found a positive (rather than a negative) association between abusive supervision and forgiveness. Thus, Hypothesis 5 was partially supported.

General Discussion

The focus of this study is subordinates’ ability to forgive supervisors’ abuse, and the ensuing impact that has on the willingness of subordinates to contribute to the organization beyond what is absolutely necessary (i.e., OCBs). Based on the predictions of organizational justice theory (Greenberg, 1987), we suggested that when subordinates perceive that supervisory abuse was motivated by a desire to enhance subordinate performance. In other words, the central issue here is whether there is a seemingly “good” intention behind supervisors’ abuses.

Theoretical Implications

We conducted three studies to test our hypotheses, two from samples in Taiwan’s military and one from MTurk platform in the U.S. Across three datasets, we found support for our prediction: subordinates’ perception that supervisor abuse was motivated by performance promotion motives moderated the relationship between abusive supervision and subordinate generalized forgiveness, and generalized forgiveness mediated the interactive effect of abusive supervision and supervisors’ performance promotion motives on OCB, indicating a beneficial impact by supervisory performance promotion motives on the effects of abusive supervision on OCB via generalized forgiveness. Our findings offer the following theoretical implications.

First, we discovered a pivotal role of subordinate generalized forgiveness when facing abusive supervision. Parallel to the findings of prior studies, victims’ forgiveness leads to altruistic behaviors (e.g., Karremans & Van Lange, 2008; Karremans et al., 2005). Subordinates’ generalized forgiveness can reverse the seemingly unavoidable negative consequences of supervisor abusiveness and instill victims with compassion, benevolence, and love (Enright et al., 1992). We suggest that such positive effects might be due to a reduction in subordinates’ perceived injustice at the workplace (Wenzel & Okimoto, 2010, 2012) or an increase in their positive emotions (Spector & Fox, 2002). Moreover, our results complement the existing literature, which suggests that abused subordinates may imitate supervisors’ offensive behaviors, as documented by either the cascade model of abusive supervision (Liu et al., 2012) or social learning explanations (Garcia et al., 2014).

Second, some may argue that, it is likely that supervisors might purposely exhibit an image of prosocial motives in order to influence subordinates’ perceptions (cf. Van Dijk et al., 2004). As shown in our study findings, as long as subordinates perceive a performance promotion motive behind supervisory abuse, it will result in a higher tendency of forgiveness. But how about when subordinates found out that it was only resulted from successful impression management of supervisors. According to the tenets of ethical leadership (e.g., Treviño, Hartman & Brown, 2000), supervisors’ behaviors are assessed by subordinates on some ethical grounds; however, there is likelihood that supervisory abuse be interpreted in a “normatively appropriate” view such as tough love or performance-driven discipline if they choose to believe that it is based on morally acceptable motives. In a similar vein, subordinate-attributed performance promotion motives may be used to justify supervisory abuse if the behaviors serve a higher-level purpose, a phenomenon termed moral disengagement by Bandura (1999). Future studies may want to examine subordinates’ reactions when “true” motives are revealed to the subordinates.

Thirdly, in addition to moral justification and attribution of blame, Bandura (1999) has proposed a variety of cognitive mechanisms that facilitate unethical behavior. For instance, the use of “dehumanization” is to frame the victims as undeserving of basic human consideration. That is, supervisors may dehumanize their subordinates before/during/after the abuse. We recommend a more complete examination of different cognitive mechanisms of moral disengagement as potential for moderating the link between supervisory abuse and subordinate forgiveness.

Finally, in Study 3, our results showed an overall positive relationship between abusive supervision and subordinate generalized forgiveness, and supervisory performance promotion motives showed an elevated moderating effect by abusive supervision on subordinate forgiveness making the relationship between the two becoming more positive (rather than less negative). Although this result seems to be contradictory to many existing findings on abusive supervision and to the predictions of organizational justice theory (Greenberg, 1987), some recent studies have started to examine the potential positive impact by abusive supervision on related outcome variables. For example, Tröster and Van Quaquebeke (2021), have found a positive effect by abusive supervision on subordinates’ helping behavior through the latter’s experiences of self-blame and felt guilt, which in turn motivated them to engage in more helping behaviors. We recommend that future studies may want to explore such a paradoxical relationship between abusive supervision and subordinate forgiveness to arrive at a more complete model of leader-subordinate relationship.

Practical Implications

Turning to the practical implications of our study, we note the negative impact of abusive supervision on subordinates’ cognitive, affective, and behavioral outcomes. Such destructive leadership style is fundamentally unethical in nature (Brown et al. 2005; Treviño et al. 2000). At the organization level, companies should ensure a zero-tolerance culture on abusive supervision and provide training programs to help managers equipped with better emotional management techniques and proper interpersonal attitudes of respect and empathy towards subordinates. For supervisors, if they really are concerned about their subordinates, they should make that concern more clear. Also, they should be refrained from actions of moral disengagement, shifting blames to others. We advise supervisors to care about the long-term welfare and well-being of their subordinates and treat them with respect and dignity, who will see and appreciate their genuine concern.

For subordinates, scholars have suggested various coping strategies to mitigate the stress caused by the environment (in this case, the abusive supervisors) (e.g., Lazarus & Folkman, 1984; Folkman & Moskowitz, 2004). It may help to consider the possibility that the abuse is actually motivated by pro-social goals. However, even when performance promotion motives of supervisors may result in beneficial outcomes for subordinates, we do not intend to encourage abusive supervision, even with a seemingly “positive” intention. Rather, we want to highlight the power of forgiveness. Our data vividly showed that forgiveness can turn a detrimental outcome into a less destructive one and that supervisors’ performance promotion motives (subordinate-centered) behind abusive supervision can have an enhancing impact.

Limitations and Future Research

This study has several limitations that need to be addressed. The first lies in the possibility of common source effects (Podsakoff et al., 2003), as we used self-report surveys to collect our data. It has been suggested that self-report data is the most valid approach for examining perceptual constructs and individuals’ internal states (Chan, 2009)—in our case, abusive supervision, generalized forgiveness, and attributions. Also, common method bias is less likely to be a concern when interaction effects are considered (Evans, 1985). As Podsakoff et al. (2012) argued, if interaction effects are found, they should be taken as strong evidence that common method bias was not a significant concern. Notably, our results showed significant interaction effects, and the mediated effects through forgiveness varied across levels of our moderating variable. Although self-report method has been used in other studies (e.g., Choi, 2007), having objective raters would greatly increase the validity of the measure. With that being said, future research will benefit from collections of multisource data.

Second, one may argue that subordinates’ perceptions of abusive supervision and attributions of supervisor intent are not equivalent to the actual scores of supervisors’ abusiveness and their true intent. Indeed, scholars have raised concerns about the objectivity of the assessment of abusive supervision (Martinko et al., 2013). We believe that our study results relied on what was in the minds and hearts of the abused subordinates, rather than being an “objective” measure of supervisor abusiveness or intent. Future studies may want to obtain an “average” level of supervisor abusiveness or supervisor intent rated by all subordinates in each unit and their relationship with subordinate generalized forgiveness and OCB to obtain a more comprehensive picture of the effects of abusive supervision.

Third, although our model emphasizes attribution and generalized forgiveness as key explanatory mechanisms linking abusive supervision to OCB, it is possible that dispositional differences in how individuals respond to mistreatment may partially account for the observed patterns. For instance, people dispositionally inclined to forgive others may find it easier to regulate negative emotions and adopt a forgiving stance in the face of adversity (Worthington & Scherer, 2004). In parallel, personality traits such as agreeableness and emotional stability have been shown to predict higher levels of organizational citizenship behavior, even under challenging interpersonal conditions (Ilies et al. 2009). These individuals may be more likely to forgive and persist in citizenship behaviors, regardless of how they interpret their supervisor’s motives. While we focused on cognitive attributions as a socially embedded sensemaking process, future research could incorporate trait-level moderators or controls to examine the relative contribution of personality-based versus attribution-based mechanisms.

Lastly, we conducted our study both in Taiwan and in the U.S. Future research should examine the effects of the research context on our theoretical model to determine whether the relationships found in this study can be generalized to other cultural contexts. In the end, still, we hope to have provided an additional perspective on supervisor-subordinate conflict and forgiveness.

Author Note

The data that support the findings of this study are available from the corresponding author upon reasonable request. We have no known conflicts of interest to disclose.

References

Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting interactions. Thousand Oaks, CA: Sage. https://doi.org/10.2307/2348581

Aquino, K., Tripp, T. M., & Bies, R. J. (2001). How employees respond to personal offense: the effects of blame attribution, victim status, and offender status on revenge and reconciliation in the workplace. Journal of Applied Psychology, 86(1), 52-59. https://doi.org/10.1037/0021-9010.86.1.52

Aquino, K., Tripp, T. M., & Bies, R. J. (2006). Getting even or moving on? Power, procedural justice, and types of offense as predictors of revenge, forgiveness, reconciliation, and avoidance in organizations. Journal of Applied Psychology, 91(3), 653-668. https://doi.org/10.1037/0021-9010.91.3.653

Aquino, K., Grover, S. L., Goldman, B., & Folger, R. (2003). When push doesn't come to shove interpersonal forgiveness in workplace relationships. Journal of Management Inquiry, 12(3), 209-216. https://doi.org/10.1177/1056492603256337

Aryee, S., Chen, Z. X., Sun, L. Y., & Debrah, Y. A. (2007). Antecedents and outcomes of abusive supervision: Test of a trickle-down model. Journal of Applied Psychology, 92(1), 191-201. https://doi.org/10.1037/0021-9010.92.1.191

Aryee, S., Sun, L. Y., Chen, Z. X. G., & Debrah, Y. A. (2008). Abusive supervision and contextual performance: The mediating role of emotional exhaustion and the moderating role of work unit structure. Management and Organization Review, 4(3), 393-411. https://doi.org/10.1111/j.1740-8784.2008.00118.x

Balliet, D., & Joireman, J. (2010). Ego depletion reduces proselfs’ concern with the well-being of others. Group Processes & Intergroup Relations, 13(2), 227-239. https://doi.org/10.1177/1368430209353634

Bandura, A. (1999). Moral disengagement in the perpetration of inhumanities. Personality and Social Psychology Review, 3(3), 193-209. https://doi.org/10.1207/s15327957pspr0303_3

Barclay, L. J., & Saldanha, M. F. (2016). Facilitating forgiveness in organizational contexts: Exploring the injustice gap, emotions, and expressive writing interventions. Journal of Business Ethics, 137, 699-720. https://doi.org/10.1007/s10551-015-2750-x

Bies, R. J., & Shapiro, D. L. (1987). Interactional fairness judgments: The influence of causal accounts. Social Justice Research, 1(2), 199-218. https://doi.org/10.1007/BF01048016

Bies, R. J., & Tripp, T. M. (1996). Beyond distrust: Getting even and the need for revenge. In R. M. Kramer & T. R. Tyler (Eds.), Trust in organizations (pp. 246-260). Thousand Oaks, CA: Sage. https://doi.org/10.4135/9781452243610.n12

Boulding, K. (1963). Conflict and defense. New York, NY: Harper & Row.

Bradfield, M., & Aquino, K. (1999). The effects of blame attributions and offender likableness on forgiveness and revenge in the workplace. Journal of Management, 25(5), 607-631. https://doi.org/10.1177/014920639902500501

Brislin, R. W. (1986). The wording and translation of research instruments. In W. J. Lonner & J. W. Berry (Eds.), Field methods in cross-cultural research (pp. 137-164). Thousand Oaks, CA: Sage.

Brown, M. E., Treviño, L. K., & Harrison, D. A. (2005). Ethical leadership: A social learning perspective for construct development and testing. Organizational Behavior and Human Decision Processes, 97(2), 117-134. https://doi.org/10.1016/j.obhdp.2005.03.002

Chan, D. (2009). So why ask me? Are self-report data really that bad? In C. E. Lance & R. J. Vandenberg (Eds.), Statistical and methodological myths and urban legends: Doctrine, verity and fable in the organizational and social sciences (pp. 309-336). New York, NY: Routledge. https://doi.org/10.1111/j.1744-6570.2009.01160_5.x

Chi, S.-C. S., & Liang, S.-G. (2013). When do subordinates' emotion-regulation strategies matter? Abusive supervision, subordinates' emotional exhaustion, and work withdrawal. The Leadership Quarterly, 24(1), 125-137. https://doi.org/10.1016/j.leaqua.2013.03.004

Choi, J. N. (2007). Change-oriented organizational citizenship behavior: Effects of work environment characteristics and intervening psychological processes. Journal of Organizational Behavior, 28(4), 467-484. https://doi.org/10.1002/job.433

Cox, S. S., Bennett, R. J., Tripp, T. M., & Aquino, K. (2012). An empirical test of forgiveness motives' effects on employees' health and well-being. Journal of Occupational Health Psychology, 17(3), 330-340. https://doi.org/10.1037/a0028314

Davidson, M., & Friedman, R. A. (1998). When excuses don't work: The persistent injustice effect among Black managers. Administrative Science Quarterly, 43(1), 154-183. https://doi.org/10.2307/2393594

De Cremer, D., & Van Lange, P. A. M. (2001). Why prosocials exhibit greater cooperation than proselfs: The roles of social responsibility and reciprocity. European Journal of Personality, 15(S1), 5-18. https://doi.org/10.1002/per.418

De Dreu, C. K., Weingart, L. R., & Kwon, S. (2000). Influence of social motives on integrative negotiation: A meta-analytic review and test of two theories. Journal of Personality and Social Psychology, 78(5), 889-905. https://doi.org/10.1037//0022-3514.78.5.889

Edwards, J. R., & Lambert, L. S. (2007). Methods for integrating moderation and mediation: A general analytical framework using moderated path analysis. Psychological Methods, 12(1), 1-22. https://doi.org/10.1037/1082-989X.12.1.1.supp

Enright, R. D., Gassin, E. A., & Wu, C. R. (1992). Forgiveness: A developmental view. Journal of Moral Education, 21(2), 99-114. https://doi.org/10.1080/0305724920210202

Einarsen, S., Aasland, M. S., & Skogstad, A. (2007). Destructive leadership behaviour: A definition and conceptual model. The Leadership Quarterly, 18, 207-216. https://doi.org/10.1016/j.leaqua.2007.03.002

Evans, M. G. (1985). A Monte Carlo study of the effects of correlated method variance in moderated multiple regression analysis. Organizational Behavior and Human Decision Processes, 36(3), 305-323. https://doi.org/10.1016/0749-5978(85)90002-0

Farh, C. I., & Chen, Z. (2014). Beyond the individual victim: Multilevel consequences of abusive supervision in teams. Journal of Applied Psychology, 99(6), 1074-1095. https://doi.org/10.1037/a0037636

Farmanara, P. (2021). Abusive supervisory behavior aimed at raising work group performance. Journal of Management Inquiry, 30(1), 40-58. https://doi.org/10.1177/1056492619889687

Fehr, R., Gelfand, M. J., & Nag, M. (2010). The road to forgiveness: a meta-analytic synthesis of its situational and dispositional correlates. Psychological Bulletin, 136(5), 894-914. https://doi.org/10.1037/a0019993.supp

Folkman, S., & Moskowitz, J. T. (2004). Coping: Pitfalls and promise. Annual Review of Psychology, 55, 745-774. https://doi.org/10.1146/annurev.psych.55.090902.141456

Foo, M. D., Uy, M. A., & Baron, R. A. (2009). How do feelings influence effort? An empirical study of entrepreneurs’ affect and venture effort. Journal of Applied Psychology, 94(4), 1086-1094. https://doi.org/10.1037/a0015599

Fredrickson, B. L. (2001). The role of positive emotions in positive psychology: The broaden-and-build theory of positive emotions. American Psychologist, 56(3), 218-226. https://doi.org/10.1037/0003-066X.56.3.218

Garcia, P. R. J. M., Restubog, S. L. D., Kiewitz, C., Scott, K. L., & Tang, R. L. (2014). Roots run deep: Investigating psychological mechanisms between history of family aggression and abusive supervision. Journal of Applied Psychology, 99(3), 883-897. https://doi.org/10.1037/a0036463

Grant, A. M., & Ashford, S. J. (2008). The dynamics of proactivity at work. In A. P. Brief & B. M. Staw (Eds.), Research in Organizational Behavior (Vol. 28, pp. 3-34). Elsevier. https://doi.org/10.1016/j.riob.2008.04.002

Greenberg, J. (1987). A taxonomy of organizational justice theories. Academy of Management Review, 12(1), 9-22. https://doi.org/10.5465/AMR.1987.4306437

Heider, F. (1958). The Psychology of Interpersonal Relations. New York: Wiley. https://doi.org/10.1037/10628-000

Ilies, R., Fulmer, I. S., Spitzmuller, M., & Johnson, M. D. (2009). Personality and citizenship behavior: The mediating role of job satisfaction. Journal of Applied Psychology, 94(4), 945-959. https://doi.org/10.1037/a0013329

Karremans, J. C., Van Lange, P. A., & Holland, R. W. (2005). Forgiveness and its associations with prosocial thinking, feeling, and doing beyond the relationship with the offender. Personality and Social Psychology Bulletin, 31(10), 1315-1326. https://doi.org/10.1177/0146167205274892

Karremans, J. C., & Van Lange, P. A. (2008). Forgiveness in personal relationships: Its malleability and powerful consequences. European Review of Social Psychology, 19(1), 202-241. https://doi.org/10.1080/10463280802402609

Kim, P. H., Dirks, K. T., Cooper, C. D., & Ferrin, D. L. (2006). When more blame is better than less: The implications of internal vs. external attributions for the repair of trust after a competence-vs. integrity-based trust violation. Organizational Behavior and Human Decision Processes, 99(1), 49-65. https://doi.org/10.1016/j.obhdp.2005.07.002

Kim, P. H., Ferrin, D. L., Cooper, C. D., & Dirks, K. T. (2004). Removing the shadow of suspicion: the effects of apology versus denial for repairing competence-versus integrity-based trust violations. Journal of Applied Psychology, 89(1), 104 -118. https://doi.org/10.1037/0021-9010.89.1.104

Krasikova, D. V., Green, S. G., & LeBreton, J. M. (2013). Destructive leadership: A theoretical review, integration, and future research agenda. Journal of Management, 39, 1308-1338. https://doi.org/10.1177/0149206312471388

Lazarus, R. S., & Folkman, S. (1984). Stress appraisal, and coping. New York, NY: Springer. https://doi.org/10.4135/9781412952576.n198

Lee, K., & Allen, N. J. (2002). Organizational citizenship behavior and workplace deviance: The role of affect and cognitions. Journal of Applied Psychology, 87(1), 131-142. https://doi.org/10.1037//0021-9010.87.1.131

Liang, Y., Yan, M., Chen, Y., & Miao, C. (2024). Abusive supervision differentiation and work outcomes: An integrative review and future research agenda. Applied Psychology: An International Review, 73(4), 1931-1959. https://doi.org/10.1111/apps.12532

Liu, D., Liao, H., & Loi, R. (2012). The dark side of leadership: A three-level investigation of the cascading effect of abusive supervision on employee creativity. Academy of Management Journal, 55(5), 1187-1212. https://doi.org/10.5465/amj.2010.0400

Liu, J., Kwan, H. K., Wu, L. Z., & Wu, W. (2010). Abusive supervision and subordinate supervisor-directed deviance: The moderating role of traditional values and the mediating role of revenge cognitions. Journal of Occupational & Organizational Psychology, 83(4), 835-856. https://doi.org/10.1348/096317909X485216

Malle, B. F., & Knobe, J. (1997). The folk concept of intentionality. Journal of Experimental Social Psychology, 33(2), 101-121. https://doi.org/10.1006/jesp.1996.1314

Martinko, M. J., Harvey, P., Brees, J. R., & Mackey, J. (2013). A review of abusive supervision research. Journal of Organizational Behavior, 34(S1), S120-S137. https://doi.org/10.1002/job.1888

Mawritz, M. B., Folger, R., & Latham, G. P. (2014). Supervisors' exceedingly difficult goals and abusive supervision: The mediating effects of hindrance stress, anger, and anxiety. Journal of Organizational Behavior, 35(3), 358-372. http://doi.org/10.1002/job.1879

McCullough, M. E., Rachal, K. C., Sandage, S. J., Worthington, E. L, Jr, Brown, S. W., & Hight, T. L. (1998). Interpersonal forgiving in close relationships II: Theoretical elaboration and measurement. Journal of Personality and Social Psychology, 75(6), 1586-1603. https://doi.org/10.1037//0022-3514.75.6.1586

McCullough, M. E., & Witvliet, C. V. (2001). The psychology of forgiveness. In S. J. Lopez & C. R. Snyder (Eds.), Oxford Handbook of Positive Psychology (Vol. 2, pp. 446-455). Oxford University Press, USA. https://doi.org/10.1093/oxfordhb/9780199396511.013.47

Michel, J. S., Newness, K., & Duniewicz, K. (2016). How abusive supervision affects workplace deviance: A moderated-mediation examination of aggressiveness and work-related negative affect. Journal of Business and Psychology, 31(1), 1-22. https://doi.org/10.1007/s10869-015-9400-2

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(4), 1159-1168. https://doi.org/10.1037/0021-9010.92.4.1159

Olekalns, M., & Smith, P. L. (2003). Social motives in negotiation: The relationships between dyad composition, negotiation processes and outcomes. International Journal of Conflict Management, 14(3/4), 233-254. https://doi.org/10.1108/eb022900

Organ, D. W. (1988). Organizational citizenship behavior: The good soldier syndrome. Toronto: Lexington Books. https://doi.org/ 10.1016/B978-0-08-097086-8.22031-X

Organ, D. W. (1997). Organizational citizenship behavior: It's construct clean-up time. Human Performance, 10(2), 85-97. https://doi.org/10.1207/s15327043hup1002_2

Podsakoff, P. M., MacKenzie, S. B., Lee, J. Y., & Podsakoff, N. P. (2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88(5), 879-903. https://doi.org/10.1037/0021-9010.88.5.879

Podsakoff, P. M., MacKenzie, S. B., & Podsakoff, N. P. (2012). Sources of method bias in social science research and recommendations on how to control it. Annual Review of Psychology, 63, 539-569. https://doi.org/10.1146/annurev-psych-120710-100452

Pruitt, D. G., & Lewis, S. A. (1975). Development of integrative solutions in bilateral negotiation. Journal of Personality and Social Psychology, 31(4), 621-633. https://doi.org/10.1037/0022-3514.31.4.621

Rahim, M. A. (2011). Managing conflict in organizations. Third Edition. Transaction Publishers. https://doi.org/10.5040/9798400682117.0004

Riordan, C. A., Marlin, N. A., & Kellogg, R. T. (1983). The effectiveness of accounts following transgression. Social Psychology Quarterly, 46(3), 213-219. https://doi.org/10.2307/3033792

Schyns, B., & Schilling, J. (2013). How bad are the effects of bad leaders? A meta-analysis of destructive leadership and its outcomes. The Leadership Quarterly, 24(1), 138-158. https://doi.org/10.1016/j.leaqua.2012.09.001

Sitkin, S. B., & Bies, R. J. (1993). Social accounts in conflict situations: Using explanations to manage conflict. Human Relations, 46(3), 349-370. https://doi.org/10.1177/001872679304600303

Shaw, J. C., Wild, E., & Colquitt, J. A. (2003). To justify or excuse? A meta-analytic
review of the effects of explanations. Journal of Applied Psychology, 88, 444-458. https://doi.org/10.1037/0021-9010.88.3.444

Sitkin, S. B., & Bies, R. J. (1993). Social accounts in conflict situations: Using explanations to manage conflict. Human Relations, 46(3), 349-370. https://doi.org/10.1177/001872679304600303Spector, P. E., & Fox, S. (2002). An emotion-centered model of voluntary work behavior: Some parallels between counterproductive work behavior and organizational citizenship behavior. Human Resource Management Review, 12(2), 269-292. https://doi.org/10.1111/1468-2389.00193

Struthers, C. W., Eaton, J., Santelli, A. G., Uchiyama, M., & Shirvani, N. (2008). The effects of attributions of intent and apology on forgiveness: When saying sorry may not help the story. Journal of Experimental Social Psychology, 44(4), 983-992. https://doi.org/10.1016/j.jesp.2008.02.006

Tepper, B. J. (2000). Consequences of abusive supervision. Academy of Management Journal, 43(2), 178-190. https://doi.org/10.2307/1556375

Treviño, L. K., Hartman, L. P., & Brown, M. (2000). Moral person and moral manager: How executives develop a reputation for ethical leadership. California Management Review, 42(4), 128-142. https://doi.org/10.2307/41166057

Tröster, C., & Van Quaquebeke, N. (2021). When victims help their abusive supervisors: The role of LMX, self-blame, and guilt. Academy of Management Journal, 64(6), 1793-1815. https://doi.org/10.5465/amj.2019.0559

Van Dijk, E., De Cremer, D., & Handgraaf, M. J. J. (2004). Social value orientations and the strategic use of fairness in ultimatum bargaining. Journal of Experimental Social Psychology, 40(6), 697-707. https://doi.org/10.1016/j.jesp.2004.03.002

Vogel, R. M., Mitchell, M. S., Tepper, B. J., Restubog, S. L., Hu, C., Hua, W., & Huang, J. C. (2015). A cross‐cultural examination of subordinates' perceptions of and reactions to abusive supervision. Journal of Organizational Behavior, 36(5), 720-745. https://doi.org/10.1002/job.1984

Weick, K. E. (1995). Sensemaking in Organizations. Thousand Oaks, CA: Sage. https://doi.org/10.3406/sotra.1996.2274

Weiner, B. (1995). Attribution theory in organizational behavior: A relationship of mutual benefit. In M. J. Martinko (Ed.), Attribution theory: An organizational perspective (pp.3-6). Delray Beach, FL: St. Lucie Press. https://doi.org/10.4324/9781315137926-1

Weiner, B., Amirkhan, J., Folkes, V. S., & Verette, J. A. (1987). An attributional analysis of excuse giving: studies of a naive theory of emotion. Journal of Personality and Social Psychology, 52(2), 316 - 324. https://doi.org/10.1037//0022-3514.52.2.316

Wenzel, M., & Okimoto, T. G. (2010). How acts of forgiveness restore a sense of justice: Addressing status/power and value concerns raised by transgressions. European Journal of Social Psychology, 40(3), 401- 417. https://doi.org/10.1002/ejsp.629

Wenzel, M., & Okimoto, T. G. (2012). The varying meaning of forgiveness: Relationship closeness moderates how forgiveness affects feelings of justice. European Journal of Social Psychology, 42(4), 420-431. https://doi.org/10.1002/ejsp.1850

Worthington, E. L., Jr., & Scherer, M. (2004). Forgiveness is an emotion-focused coping strategy that can reduce health risks and promote health resilience: Theory, review, and hypotheses. Psychology & Health, 19(3), 385-405. https://doi.org/10.1080/0887044042000196674

Xu, A. J., Loi, R., & Lam, L. W. (2015). The bad boss takes it all: How abusive supervision and leader-member exchange interact to influence employee silence. The Leadership Quarterly, 26(5), 763-774. https://doi.org/10.1016/j.leaqua.2015.03.002

Xu, E., Huang, X., Lam, C. K., & Miao, Q. (2012). Abusive supervision and work behaviors: The mediating role of LMX. Journal of Organizational Behavior, 33(4), 531-543. https://doi.org/10.1002/job.768

Zhang, Y., & Bednall, T. C. (2016). Antecedents of abusive supervision: A meta-analytic review. Journal of Business Ethics, 139(3), 455-471. https://doi.org/10.1007/s10551-015-2657-6