Roots Run Deep: Investigating Psychological Mechanisms Between History of Family Aggression and Abusive Supervision

Abstract

In this article, we examine the relationships between supervisor-level factors and abusive supervision. Drawing from social learning theory (Bandura, 1973), we argue that supervisors' history of family aggression indirectly impacts abusive supervision via both hostile cognitions and hostile affect, with angry rumination functioning as a first-stage moderator. Using multisource data, we tested the proposed relationships in a series of 4 studies, each providing evidence of constructive replication. In Study 1, we found positive relationships between supervisors' history of family aggression, hostile affect, explicit hostile cognitions, and abusive supervision. We obtained the same pattern of results in Studies 2, 3, and 4 using an implicit measure of hostile cognitions and controlling for previously established antecedents of abusive supervision. Angry rumination moderated the indirect relationship between supervisors' history of family aggression and abusive supervision via hostile affect only. Overall, the results highlight the important role of supervisor-level factors in the abusive supervision dynamics. (PsycINFO Database Record (c) 2014 APA, all rights reserved).

Full text

Roots Run Deep: Investigating Psychological Mechanisms between History of Family
Aggression and Abusive Supervision
Patrick Raymund James M. Garcia
The Australian National University
Simon Lloyd D. Restubog
The Australian National University
Christian Kiewitz
University of Dayton
Kristin L. Scott
Clemson University
Robert L. Tang
De La Salle – College of Saint Benilde
Authors’ notes:
We are grateful to Prashant Bordia, Arthur Brief, Alessandra Capezio, George Chen, Tom
Denson, Sandy Hershcovis, Beryl Hesketh, Marie Mitchell, and Bennett Tepper for their
insightful comments on an earlier version of this manuscript. We also thank Craig Anderson and
Lawrence James for providing us access to their measures. This research was supported by the
Australian Research Council grant (DP1094023) awarded to the second author. Please address
correspondence to Patrick Raymund James M. Garcia, Patrick.Garcia@anu.edu.au.

Roots Run Deep: Investigating Psychological Mechanisms between History of Family
Aggression and Abusive Supervision
ABSTRACT
In this paper, we examine the relationships between supervisor-level factors and abusive
supervision. Drawing from social learning theory (Bandura, 1973), we argue that supervisors’
history of family aggression indirectly impacts abusive supervision via both hostile cognitions
and hostile affect, with angry rumination functioning as a first-stage moderator. Using multi-
source data, we tested the proposed relationships in a series of four studies, each providing
evidence of constructive replication. In Study 1, we found positive relationships between
supervisors’ history of family aggression, hostile affect, explicit hostile cognitions, and abusive
supervision. We obtained the same pattern of results in Studies 2, 3, and 4 using an implicit
measure of hostile cognitions and controlling for previously established antecedents of abusive
supervision. Angry rumination moderated the indirect relationship between supervisors’ history
of family aggression and abusive supervision via hostile affect only. Overall, the results highlight
the important role of supervisor-level factors in the abusive supervision dynamics.
Keywords: abusive supervision, family aggression, implicit measure, hostile affect, hostile
cognitions, rumination, social learning, workplace aggression

History of family aggression and abusive supervision 1
Roots Run Deep: Investigating Psychological Mechanisms between History of Family
Aggression and Abusive Supervision
Abusive supervision is generally understood as “subordinates’ perceptions of the extent
to which supervisors engage in the sustained display of hostile verbal and nonverbal behaviors,
excluding physical contact” (Tepper, 2000, p. 178). Because the phenomenon involves behaviors
such as repeated scapegoating, angry outbursts, and public criticism towards subordinates (Bies,
2001), it has negative implications for a wide variety of work and non-work attitudes (e.g.,
meaning of work, job and life satisfaction; Rafferty & Restubog, 2011; Tepper, 2007) and
behaviors (e.g., in- and extra-role job performance, workplace deviance, spousal undermining;
Lian, Ferris, & Brown, 2012; Mitchell & Ambrose, 2007; Restubog, Scott, & Zagenczyk, 2011;
Shoss, Eisenberger, Restubog, & Zagenczyk, 2013). While research to date has yielded
important insights, we know only little about supervisor-specific factors that engender perceived
abuse (a notable exception is work on deep-level dissimilarity by Tepper, Moss, and Duffy,
2011). In reaction, Tepper’s (2007) critical appraisal of the abusive supervision literature urges
researchers to consider supervisors’ characteristics in their studies because leaders’ attitudes and
behaviors are profoundly influenced by their personality and prior experiences (House, Shane, &
Herold, 1996). Indeed, empirical evidence suggests that individual differences directly influence
the occurrence of workplace aggression in general (Aquino & Bradfield, 2000; Douglas &
Martinko, 2001; Garcia, Restubog, & Denson, 2010; Inness, Barling, & Turner, 2005) and of
abusive supervision in particular (Kiazad, Restubog, Zagenczyk, Kiewitz, & Tang, 2010).
Undoubtedly, a better understanding of the supervisor-specific factors that enable abusive
supervision is crucial for the development of more targeted policies and interventions aimed at
reducing its occurrence.

History of family aggression and abusive supervision 2
To address this issue, we adopt a social learning perspective in investigating the
psychological mechanisms associated with abusive supervision (Bandura, 1973). Given the
highly influential role of parents in their child’s social development, we specifically investigate
the impact of supervisors’ history of family aggression — operationalized as observed inter-
parental aggression — on abusive supervision. Based on social learning theory (Bandura, 1973)
and ancillary research (Eisenberg, Cumberland, & Spinrad, 1998; Huesmann, 1988), we propose
that a history of family aggression predisposes supervisors to engage in abusive supervision
because repeated exposure to family aggression facilitates the development of and chronic
accessibility to hostile cognitions and affect (i.e., the retention of aggressive concepts in memory
as well as the expressivity of hostile affect). Moreover, we posit that the relationships between
supervisors’ history of family aggression, hostile cognitions, hostile affect, and abusive
supervision are strengthened by supervisors’ predisposition to engage in angry rumination.
We make three important contributions to the abusive supervision literature. First, by
adopting the social learning theory (Bandura, 1973), we offer an alternative lens by which to
examine the factors that influence abusive supervision. By framing it as a form of learned
aggressive behavior, we are able to extend the known antecedents of abusive supervision to
include important non-work related experiences (e.g., history of family aggression). Indeed, most
studies have utilized a justice-based approach focusing on the role of organizational mistreatment
as antecedents of abuse (Tepper, 2007). It should be noted that we are not proposing that a social
learning approach is better compared to displaced aggression or equity theories. However, we
believe that utilizing an alternative theoretical approach may lead to additional insights as to the
possible reasons why supervisors abuse their subordinates. Second, our paper examines
psychological factors that influence abusive supervision. Unlike prior work, it links previous

History of family aggression and abusive supervision 3
experience (i.e., history of family aggression) and abusive supervision by implicating hostility-
related cognitions (i.e., explicit and implicit hostile cognitions, angry rumination) and hostility-
related affect. We argue that a distal antecedent such as an individual’s history of family
aggression influences how that individual thinks and feels about hostility. This in turn is
associated with a predisposition towards aggressive behavior (e.g., abusive supervision). Lastly,
to our knowledge, our study is the first to examine the role of implicit social cognitions in
abusive supervision research. The inclusion of implicit hostile cognitions extends the
nomological net of abusive supervision as well as takes into account mounting evidence that
aggressive cognitions are composed of both explicit and implicit components (Bing et al., 2007;
Frost, Ko, & James, 2007). In the sections that follow, we further elaborate on these theoretical
linkages.
Supervisors’ History of Family Aggression, Explicit and Implicit Hostile Cognitions
Individuals learn aggressive behaviors not only via direct experience (Bandura, 1973;
Kiewitz et al., 2012) but also vicariously by observing and modeling authority figures (e.g.,
parents). One context that has been identified to provide ample opportunity for exposure to
aggression and to be highly influential in learning aggressive behavior is the family environment
(Cappell & Heiner, 1990; Chermack & Walton, 1999). Parents, in particular, are highly
influential agents in social learning processes because children regard them as authority figures
and depend on them for care and nurturance (Bandura, 1973). In this study, we focus on
observed inter-parental aggression (i.e., the extent to which supervisors observed their parents
aggress towards each other during childhood) as an indicator of family aggression history.
Empirical studies support the salience of observational learning in acquiring aggressive
tendencies. In a series of experiments, Bandura and colleagues (Bandura, 1965; Bandura, Ross,

History of family aggression and abusive supervision 4
& Ross, 1963) found that exposure to human and filmed aggressive models doubled aggressive
behavior in children relative to those who were not exposed to aggressive models. Furthermore,
children are more likely to imitate the behavior of a model who had been nurturing and had
power over resources important to a child (e.g., parents).
According to social learning theory (Bandura, 1973), observational learning contributes
to long-term aggressiveness through the retention and symbolic representation of stimuli in
memory. Specifically, humans have the ability to encode and transform information into
functional verbal symbols that subsequently guide social behavior. In the case of aggression,
verbal representations in the form of aggressive concepts are later synthesized to form memory
codes that reflect specific behavioral patterns. Huesmann (1988) further elaborated on these
propositions through his work on aggressive scripts, which refer to sets of particularly well-
rehearsed, highly associated concepts in memory, often involving causal links, goals, and action
plans. Notably, aggressive scripts guide how individuals approach social interactions, as they
comprise information about what environmental cues to process, types of behavioral strategy to
use (e.g., in conflicts), and expected outcomes of chosen behaviors (Huesmann, 1988).
Besides contributing to the development of aggressive concepts in memory, social
learning theory also suggests that repeated exposure to family aggression strengthens
individuals’ beliefs about the acceptability of aggressive behavior (Bandura, 1973; Huesmann,
1988). That is, exposure to aggression during childhood fosters the notion that aggressive
behavior is acceptable and useful in settling interpersonal conflicts. This belief is further
reinforced when a child observes that aggressive behaviors result in positive consequences (e.g.,
obedience and control) because the child learns to expect similar positive outcomes by enacting
aggressive behaviors (Bandura, 1973). Indeed, empirical evidence in the form of a longitudinal

History of family aggression and abusive supervision 5
survey-based study suggests that exposure to aggression during childhood is positively related to
long-term increases in hostile cognitions (i.e., aggressive scripts and normative beliefs about
aggression; Guerra, Huesmann, & Spindler, 2003). Based on these arguments, we suggest that
individuals who previously and repeatedly observed inter-parental aggression may have learned
that aggression constitutes acceptable and useful behavior in relationships and hence may be
more inclined to behave aggressively in other contexts as well (e.g., at work with subordinates).
This claim is supported by research demonstrating that social learning of aggression generalizes
across contexts and is not limited to where it was learned (e.g., the family setting; Huesmann,
Eron, Lefkowitz, & Walder, 1984; Straus, 1990).
Complementing the above notions is the realization in recent social learning research that
human cognition involves both conscious and unconscious processes (Anderson & Bushman,
2002). On one hand, individuals are consciously aware of their beliefs in the appropriateness of
and preference for aggressive behavior (Huesmann, 1988). On the other hand, they are unaware
of how accessible and strongly associated such aggressive concepts are in memory (Anderson &
Bushman, 2002). As a consequence, there is growing consensus among personality theorists that
researchers need to tap into both explicit and implicit social cognitions in order to fully account
for the influence of social cognitions on behavior (Bing et al., 2007; Greenwald & Banaji, 1995).
Notably, research has shown that implicit cognitions account for a significant amount of variance
over and above explicit cognitions in predicting experimental task performance (e.g., word-maze
and word-recall tasks; McClelland, Koestner, & Weinberger, 1989). In light of these arguments,
we propose that supervisors’ observation of inter-parental aggression (i.e., their history of family
aggression) is associated with both explicit and implicit hostile cognitions.

History of family aggression and abusive supervision 6
Hypotheses 1a: History of family aggression is positively related to explicit hostile
cognitions.
Hypothesis 1b: History of family aggression is positively related to implicit hostile
cognitions.
Supervisors’ History of Family Aggression and Hostile Affect
In our research model, we further propose that supervisors’ history of family aggression
increases the likelihood of developing highly accessible hostile affect. According to social
learning theory (Bandura, 1973), emotional learning can occur when people, places, and events
acquire emotion-eliciting properties through repeated associations with painful experiences (e.g.,
family aggression). These conditioned associations increase the variety of environmental cues
that can trigger the experience of hostile emotions. That is, hostile affect witnessed during inter-
parental aggression becomes symbolically encoded as part of a child’s aggressive scripts
(Huesmann, 1988; Abelson, 1981). Given the psychological parallels that exist between parent-
child and supervisor-subordinate relationships (Game, 2008), it is plausible that conditioned
negative emotions learned in the family context will also be triggered in the workplace.
Further support for the relationship between history of family aggression and hostile
affect comes from ancillary research on emotional socialization, particularly on the influence of
parental expressivity of emotion (Eisenberg et al., 1998; Fredrickson, 1998; Kim, Conger,
Lorenz, & Elder, 2001). According to this line of work, exposure to emotional expressions of
hostility may influence children’s arousal through contagion, vicarious processes, or through the
meaning associated with the emotional displays (Eisenberg et al., 1998). For instance, when
asked to report what they felt after observing live expressions of hostile affect, participants aged
4-5 years old described having negative feelings similar to the emotions exhibited by the adult

History of family aggression and abusive supervision 7
models (Cummings, 1987). Moreover, repeated exposure to hostile emotional displays during
family interactions may signal to children that such behavior is not only appropriate and
functional in social interactions (Dusek & Danko, 1994) but also justified (Eisenberg et al.,
1998). These propositions are supported by studies examining the relationship between parental
emotional expressivity and emotional regulation among children and adults (Burrowes &
Halberstadt, 1987; Eisenberg et al., 2001). For example, in a cross-sectional study, college
students from negatively expressive families reported less control over feelings of anger
compared to their peers even after controlling for anger intensity (Burrowes & Halberstadt,
1987). Similarly, a lab study showed that exposure to negative parental expressivity has been
found to increase children’s emotional expressivity and externalizing of behavioral problems
(Eisenberg et al., 2001). To this end, we expect history of family aggression to be associated with
greater hostile affect. That is, observation of inter-parental aggression increases the likelihood
that environmental cues and ambiguous situations trigger hostile affect due to conditioned
emotions. The more anger-eliciting situations the child is exposed to, the higher the chances that
aggressive concepts in memory contain hostility-related affect. We also expect that supervisors
who have observed family aggression are less likely to regulate hostile affect, thereby increasing
its occurrence at work. In line with this reasoning, we predict that:
Hypothesis 2: History of family aggression is positively related to hostile affect.
Hostile Cognitions, Hostile Affect, and Subordinates’ Perceptions of Abusive Supervision
We further propose that both supervisors’ hostile cognitions and hostile affect will be
positively associated with subordinates’ perceptions of abusive supervision. Consistent with
social learning theory (Bandura, 1973), we argue that highly accessible encoded hostile
cognitions and affect influence one’s behavioral tendencies towards aggression. As an example

History of family aggression and abusive supervision 8
for such a link involving hostile cognitions, Guerra and colleagues (2003) found that the
relationship between community violence exposure and subsequent aggressive behavior 12
months later was partially mediated by children’s beliefs about the acceptability of aggression.
Notably, the level of aggressive behavior remained stable across Grades 1 to 6 irrespective of
whether teacher or parent reports were used. Similarly, Anderson and Bushman (2002) argue that
hostile affect in the form of anger provides a justification for aggressive retaliation and
potentially interferes with higher-level cognitive processes, including those used in moral
reasoning. In other words, feeling angry may serve as a cue to react and engage in aggressive
behaviors (e.g., yelling, public ridicule).
Supervisors’ Angry Rumination as a First-Stage Moderator
Thus far we have argued that history of family aggression fosters the development of
hostile cognitions and affect, which in turn predisposes supervisors to engage in abusive
supervision. We further qualify this prediction by arguing that not all supervisors who have
observed family aggression will develop hostility-related cognitions and affect. Dispositional
characteristics are likely to influence these processes. An important personality factor that
explains and predicts the occurrence of anger and aggressive behavior is angry rumination
(Denson, Pedersen, Ronquillo, & Nandy, 2009). It refers to the tendency to focus on negative
self-evaluations or negative interpretations of one’s life causing the amplification of negative
emotion. In the context of our research model, we treat angry rumination as a first-stage
moderator because of its capability to increase the accessibility of aggressive thoughts and anger
experiences (Pedersen et al., 2011). Consistent with social learning theory, angry rumination
represents a form of rehearsal for encoded behavioral scripts (Bandura, 1973). That is, when
individuals replay what they have observed in the past (e.g., family aggression), it becomes more

History of family aggression and abusive supervision 9
readily accessible in memory (Huesmann, 1988). Rumination also provides the opportunity to
revise and cognitively rehearse aggressive scripts, thus enhancing its appropriateness in various
situations (Bandura, 1973). Angry rumination can also amplify the experience of negative
emotion because it prolongs the activation of the emotional experience (Clark & Isen, 1981). For
instance, a recent neuroimaging study showed that areas of the brain associated with negative
affect were active during the time participants engaged in angry rumination, thus corroborating
the positive relationship between rumination and increased experience of activated emotions
(Denson et al., 2009). Importantly, empirical evidence suggests that those individuals who are
predisposed to ruminate about past offenses are not only more susceptible to feeling extremely
hostile but also to behaving more aggressively (Bushman & Geen, 1990; Ray et al., 2005).
Collectively, we expect that supervisors’ rumination about their past experiences (i.e., history of
family aggression) will increase the strength of the indirect effect between history of family
aggression and abusive supervision via hostile cognitions and hostile affect:
Hypotheses 3: The conditional indirect effect of history of family aggression on abusive
supervision via hostile cognitions will be stronger when angry rumination is high
compared to when angry rumination is low.
Hypotheses 4: The conditional indirect effect of history of family aggression on abusive
supervision via hostile affect will be stronger when angry rumination is high compared to
when angry rumination is low.
Overview of Studies
We tested our hypotheses in four studies, each providing evidence of constructive
replication (Lykken, 1968). As shown in Table 1, Study 1 tests Hypothesis 1a (explicit hostile
cognitions) and Hypothesis 2 (hostile affect) while Studies 2 to 4 assess Hypothesis 1b (implicit

History of family aggression and abusive supervision 10
hostile cognitions) and Hypotheses 2 to 4 (hostile affect and angry rumination). In Study 3, we
obtained parent ratings of family aggression history to minimize response distortion and memory
recall problems associated with retrospective accounts (Hardt & Rutter, 2004). We included a
temporal element in Studies 3 and 4 to help reduce concerns associated with common method
variance (Podsakoff, MacKenzie, & Podsakoff, 2012). To rule out alternative explanations for
our findings, we included previously established antecedents of abusive supervision as control
variables in Studies 2, 3, and 4.
Study 1
Participants and Procedure
As part of an unrelated project, we distributed survey kits to 258 supervisors enrolled in a
part-time MBA program in the Philippines. All surveys were in English. Supervisors completed
a supervisor questionnaire and generated an anonymous code to allow matching of surveys from
his/her corresponding subordinate. Upon completion, supervisors received a sealed envelope
containing demographic and behavioral questions (i.e., abusive supervision items) to be given to
one of their subordinates. Surveys were returned to the research team in a pre-paid envelope via
postal mail.
1
Altogether, we were able to match 154 of the 258 supervisor-subordinate surveys.
Among supervisors, 55.80% of participants were males; 59.70% were above 30 years old; and
average organizational tenure was 2.69 years. Supervisors had supervised their subordinate for
an average of 2.79 years. Supervisors worked in various business sectors, such as food and
beverage (26%), airline and transportation (25%), government service (22%), and hotels (22%).
The subordinate sample consisted of 54.50% males; approximately 81.90% were between 19-30
years old; and average organizational tenure was 2.12 years.

History of family aggression and abusive supervision 11
Measures
History of family aggression (α = .86). Supervisors’ history of family aggression was
assessed with a modified version of the Conflict Tactics Scale (CTS; Straus, 1979). Supervisors
were instructed to recall the “worst” year of their childhood (i.e., the time when their parents
fought the most) and indicate the frequency (1 = never to 7 = always) with which they witnessed
their parents use aggressive tactics toward each other during that year. Example items include:
“One of my parents yelled at the other” and “One of my parents threw something at the other.”
The validity of the CTS was corroborated by prior work that examined the level of
agreement between CTS scores reported by more than one family member (Straus, 1979; Straus,
Hamby, Boney-McCoy, & Sugarman, 1996). To provide additional validity evidence for the
CTS, we collected data from an independent sample of 267 student and parent dyads in the
Philippines. First, we correlated the student-reported observed inter-parental aggression with the
parent-rated measure. Results suggest that the two measures were positively correlated (r = .43,
p<.01). Second, we examined the correlation between parent-reported inter-parental aggression
with the student-reported hostile cognitions and hostile affect. Bivariate correlations revealed
that parent ratings of history of family aggression were positively related to both hostile
cognitions (r =.17, p<.01) and hostile affect (r = .13, p<.05). Overall, these results support the
validity of the CTS as a measure of history of family aggression.
Explicit hostile cognitions (α = .83). In this study, explicit hostile cognitions were
measured using eight items from the general approval of aggression subscale of the normative
beliefs about aggression scale (NOBAGS; Huesmann & Guerra, 1997). The subscale reflects the
endorsement of aggressive scripts through self-reported beliefs about the acceptability of

History of family aggression and abusive supervision 12
aggression. Supervisors were asked to rate (1 = strongly disagree to 7 = strongly agree) the
acceptability of statements such as: “If you’re angry, it is okay to say mean things to people.”
Hostile affect (α = .98). Hostile affect was measured using the state hostility scale
developed by Anderson, Deuser, and DeNeve (1995). It consists of 19 self-relevant statements
containing anger- and hostility-related adjectives. Supervisors were requested to rate the extent
to which they agreed or disagreed with statements such as: “I feel furious” and “I feel angry.”
Abusive supervision (α = .97). Using Tepper’s (2000) 15-item scale, subordinates were
asked to rate the extent to which their supervisor engaged in abusive behaviors (1 = I cannot
remember him/her using this behavior with me to 7 = S/he always uses this behavior towards
me). An example item is: “My immediate supervisor puts me down in front of others.”
Results
Descriptive statistics, correlations, and reliability coefficients are presented in Table 2.
Supervisors’ history of family aggression was positively associated with explicit hostile
cognitions (r = .57, p<.001) and hostile affect (r = .56, p<.001), thus supporting Hypotheses 1a
and 2. In turn, explicit hostile cognitions and hostile affect were positively correlated with
subordinates’ perceptions of abusive supervision (r = .60 and r = .54, both p<.001). While we
found support for the hypothesized relationships, the current study only examined explicit hostile
cognitions. Having proposed that history of family aggression is also associated with highly
automatic (unconscious) aggressive concepts in memory (Anderson & Bushman, 2002), we
account for implicit hostile cognitions in the subsequent studies, while also testing for the
moderating role of angry rumination. In addition, we aimed at providing a more stringent test of
the proposed relationships by controlling for previously established antecedents of abusive
supervision.

History of family aggression and abusive supervision 13
Study 2
Participants and Procedure
Data were gathered in ten call centers in the Philippines. Surveys were administered to
supervisors during their designated, pre-allocated breaks. A member of the research team was
present during questionnaire completion and made sure the word completion task (WCT) was
completed within three minutes. As in Study 1, supervisors were asked to generate an
anonymous code for matching both supervisor and subordinate questionnaires. Supervisors were
given a sealed envelope containing the subordinate questionnaire and asked to give it to a
subordinate with whom they interacted on a regular basis. Surveys were returned to the research
team via postal mail in a pre-paid envelope. Out of 530 supervisors, we were able to match 199
supervisor-subordinate dyads. The supervisor sample was 57% male. Supervisors were on
average 33.26 years old, 84% had been working in their company for ten years or less, and had
supervised their subordinate for an average of 3.18 years. The subordinate sample comprised
54% females; 80% were between 19-30 years of age; and 81% had been working in their
respective companies for five years or less.
Measures
History of family aggression (α = .96). As in Study 1, supervisors’ history of family
aggression was assessed with the Conflict Tactics Scale (Straus, 1979).
Implicit hostile cognitions. Hostile cognitions were measured using a word completion
task (WCT) developed by Anderson, Carnagey, and Eubanks (2003). The WCT is an implicit
measure assessing the accessibility of aggressive cognitions that appears as a simple word
completion task. Supervisors were asked to complete as many of 98 word fragments as possible
in three minutes by filling in missing letters. Half of the items could be completed to form either

History of family aggression and abusive supervision 14
aggressive or non-aggressive words. For instance, the item “expl_ _e” may be completed as
“explore” or “explode.” An accessibility of aggressive thoughts score was then calculated by
dividing the number of aggressive word completions by the total number of word completions.
We conducted two validation studies to examine the construct validity of the word
completion task. In the first validation study, using a sample of 191 full-time employees in the
Philippines, we correlated the WCT scores with other established measures of hostile cognitions
such as the NOBAGS (Normative Beliefs About Aggression Scale; Huesmann & Guerra, 1997)
and the Hostile Automatic Thoughts Questionnaire (HAT; Snyder, Crowson, Houston, Kurylo,
& Poirier, 1997). Results showed that both the NOBAGS (r = .16, p<.05) and the HAT (r = .18,
p<.05) correlated positively with WCT scores. It should be noted that obtaining such low but
significant correlations was expected because the WCT is an implicit measure of hostile
cognitions (Bing et al., 2007; McClelland et al., 1989), whereas the NOBAGS and the HAT are
explicit self-report measures of hostile cognitions. Indeed, implicit measures have been noted to
measure additional and oftentimes unique information in comparison to explicit self-reports
(James & McIntyre, 2000). Next, we assessed the association of WCT scores with constructs that
have been found to be theoretically related to hostile cognitions. For instance, one personality
trait that has shown significant associations with increased aggression is trait anger (Spielberger,
Jacobs, Russell, & Crane, 1983). Thus, we expected WCT scores to be positively associated with
trait anger. Using the same sample of 191 employees, we found a significant positive relationship
between trait anger and WCT scores (r = .25, p<.01). Finally, we expected WCT scores to be
negatively related to self-control because the latter has been found to override automatic
cognitive tendencies to aggress (Stucke & Baumeister, 2006). Correlational analysis suggests
that WCT scores were significantly negatively related to self-control (r = -.23, p<.01).

History of family aggression and abusive supervision 15
In the second validation study, we collected data from an independent sample consisting of 126
full-time employees in the Philippines. We correlated the WCT scores with the Conditional
Reasoning Test of Aggression (CRTA; James & McIntyre, 2000), an implicit measure of
aggression tapping into an individual’s proclivity to use certain implicit biases in reasoning that
enhance the rational appeal of aggression. Correlational analysis revealed a significant positive
relationship between the CRTA and WCT (r = .41, p<.001). Overall, these results provide
validity evidence for the word completion task as a measure of hostile cognitions.
Hostile affect (α = .97). As with Study 1, hostile affect was assessed using the State
Hostility Scale developed by Anderson and colleagues (1995).
Angry rumination (α = .90). Angry rumination was measured using the 10-item Angry
Rumination scale developed by Denson, Pedersen, and Miller (2006). Supervisors were asked to
think about their experiences with their respective families while growing up and to rate the
extent to which they ruminate about these experiences. An example item is: “I keep thinking
about events that angered me for a long time.”
Abusive supervision (α = .96). As in Study 1, subordinates’ perceptions of abusive
supervision were measured using Tepper’s (2000) 15-item scale.
Control variables. We controlled for subordinates’ neuroticism (α = .88), which was
assessed with a 10-item scale developed by John and Srivastava (1999). Individuals high in
neuroticism may perceive greater abusive supervision compared to their low neuroticism
counterparts because of their low emotional stability (John & Srivastava, 1999). Based on Aryee,
Chen, Sun, and Debrah (2007), we also controlled for previously established antecedents of
abusive supervision such as interactional justice (α = .95) and procedural justice (α = .97),
measured with Niehoff and Moorman’s (1993) nine-item and six-item scale, respectively.

History of family aggression and abusive supervision 16
Psychological contract violation (α = .90) was assessed via Robinson and Morrison’s (2000)
four-item scale. The theoretical basis for controlling these variables is derived from social
exchange theory (Blau, 1964). Supervisors who have been unfairly treated by organizational
agents are more likely to engage in abusive supervision because of the need to restore equity and
instigate retribution (Tepper, 2007). In addition, we accounted for supervisors’ gender, age, and
duration of working relationship with the supervisor in our analyses. Evidence suggests that
males engage in more workplace aggression than females (Hershcovis et al., 2007). However, it
is also likely that females may engage in higher levels of abusive supervision than males because
of its nonphysical and verbal nature (Tepper, 2000). We controlled for age due to its positive
relationship to aggression (Tremblay et al., 1999), with older supervisors having been found to
engage in greater abusive supervision compared to their younger counterparts (Hoobler & Brass,
2006). Finally, we controlled for duration of working relationship with the supervisor based on
research suggesting that surface-level demographic dissimilarities (e.g., sex and ethnic
dissimilarity between supervisor and subordinate) diminish over time as individuals learn more
about each other (Harrison, Price, & Bell, 1998). Supervisor’s gender was dummy coded as 0 =
male and 1 = female. Both supervisor’s age and duration of working relationship with the
supervisor were assessed in years.
2
Results
Descriptive statistics are reported in Table 3. Consistent with Hypotheses 1b and 2,
supervisors’ history of family aggression was positively related to implicit hostile cognitions (r =
.32, p<.01) and hostile affect (r = .67, p<.01), respectively
3
. Hypotheses 3 and 4 proposed that
the strength of the indirect effect between history of family aggression and abusive supervision
via implicit hostile cognitions and hostile affect is conditional on the level of the supervisors’

History of family aggression and abusive supervision 17
angry rumination. Following Preacher, Rucker, and Hayes (2007), we modeled angry rumination
as a first-stage moderator of the paths leading from history of family aggression to implicit
hostile cognitions and hostile affect, respectively. We tested our moderated mediation model
using the SPSS macro developed by Preacher et al. (2007), which allowed us to implement
bootstrapping methods and probe the significance of conditional indirect effects without
assuming normality in the sampling distribution. Specifically, we examined the significance of
the conditional indirect effects through 5,000 bootstrap samples using the second-order delta
method to obtain standard errors (SE) and 95% confidence intervals (cf. Preacher et al., 2007).
The cross-product term (history of family aggression × angry rumination) was not
significantly associated with implicit hostile cognitions (β = .00, ns). Furthermore, the
conditional indirect effect between supervisors’ history of family aggression and subordinates’
abusive supervision perceptions via implicit hostile cognitions was not significant for both high
(indirect effect = .04, SE = .02, z = 1.79, 95% CI: .00 to .09) and low angry rumination (indirect
effect = .03, SE = .03, z = 1.29, 95% CI: .00 to .09). Thus, Hypothesis 3 was not supported. In
contrast, the cross-product term (history of family aggression × angry rumination) was
significantly related to hostile affect (β = .14, R
2
∆ = .02, p<.001). Results revealed that the
conditional indirect effect between supervisors’ history of family aggression and subordinates’
abusive supervision via hostile affect was significant and stronger for high (indirect effect = .16,
SE = .05, z = 2.96, 95% CI: .05 to .30) as opposed to low angry rumination (indirect effect = .05
SE = .03, z = 1.96, 95% CI: .02 to .13; see Table 6). Using the moderated mediation macro, we
also calculated several conditional indirect effects at different values of the moderator to
determine the region of significance based on the Johnson-Neyman technique (Preacher et al.,
2007). As Figure 1 shows, the conditional indirect effect of supervisors’ history of family

History of family aggression and abusive supervision 18
aggression on abusive supervision (through hostile affect) was significant at .05 alpha level for
angry rumination values above 1.16. Thus, Hypothesis 4 was supported.
A major limitation of Studies 1 and 2 is our reliance on retrospective data, specifically on
how history of family aggression was measured (i.e., asking supervisors to recall their
experiences). Research evidence suggests that memories of past events may be influenced by
inaccurate recall and response distortion (Hardt & Rutter, 2004). For this reason, we conducted a
third study that utilized parent ratings of history of family aggression. Although parent ratings
are also retrospective in nature, we believe that we can obtain a more accurate account of family
aggression history since parents were either the victim/s or perpetrator/s of the aggressive act
(Hardt & Rutter, 2004). Thus, they may be more aware of the extent to which their child
observed inter-parental aggression. In addition, a three-month time-lag was introduced to further
alleviate problems associated with common method variance.
Study 3
Participants and Procedure
Participants were recruited from four customer service organizations, two call centers and
two retail service organizations in the Philippines. In consultation with the HR division of each
participating organization, a list of supervisors with key responsibilities in managing work units
and supervising employees were obtained. A total of 277 supervisors were initially invited via
email to participate in a leadership survey, with 236 supervisors responding with interest in the
study. As a first step, these supervisors generated an anonymous code to facilitate matching of all
study questionnaires. At Time 1, supervisors were asked to give a sealed envelope containing the
parent questionnaire to one of their parents. In this questionnaire, parents were requested to
provide information concerning the focal supervisor’s history of family aggression (i.e., the

History of family aggression and abusive supervision 19
extent to which the supervisor observed their parents engaging in aggressive tactics towards each
other before the age of 18). Next, each supervisor received a supervisor questionnaire containing
items that assessed supervisor demographics, hostile rumination, hostile affect, and control
variables. Moreover, the word completion task assessing hostile cognitions was administered to
supervisors by groups. As with Study 2, supervisors were asked to complete as many of 98 word
fragments as possible within three minutes. At Time 2, three months after Time 1 data collection,
one of the direct reports to the focal supervisor completed an employee survey. These reports
were identified by asking their respective HR divisions to nominate a subordinate with whom the
supervisor frequently interacted in the previous three months. In the survey, the subordinates
were asked to provide information concerning their perceptions of abusive supervision.
Altogether, we were able to match 134 parent-supervisor-subordinate questionnaires. The
supervisor sample consisted of 60% males, with a mean age of 31.25 years. 66% of the
supervisors had been working in their respective companies for 1-5 years and had supervised
their respective subordinate for 2.63 years on average. The subordinate sample comprised 54%
females; approximately 79% were 21-30 years old; and 83% had been working in their
respective companies for five years or less. The parent sample consisted of 45% mothers, and the
average age was 55.60 years.
Measures and Results
We retained the Study 2 measures for Study 3. In line with previous work (O’Keefe,
1998), history of family aggression was measured by asking parents to recall their child’s
“worst” year in the family; that is, the time when the parent fought with his/her spouse the most
(prior to the age of 18). Subordinates’ neuroticism was assessed with Goldberg’s (1992) scale (α
= .89).

History of family aggression and abusive supervision 20
Descriptive statistics for Study 3 are presented in Table 4. Paralleling the results of
Studies 1 and 2, parent ratings of history of family aggression were positively related to implicit
hostile cognitions (r = .27, p<.01) and hostile affect (r = .48, p<.01), respectively, thus
supporting Hypotheses 1b and 2. The interaction term (parent ratings of history of family
aggression × angry rumination) was not significantly associated with implicit hostile cognitions
(β = .00, ns). As in the previous studies, we then tested the significance of the conditional
indirect effect using 5,000 bootstrap samples. The conditional indirect effect between parent
ratings of history of family aggression and abusive supervision via implicit hostile cognitions
was not significant for both high (indirect effect = .02, SE = .03, z = .73, 95% CI: -.02 to .10) and
low angry rumination (indirect effect = .05, SE = .04, z = 1.22, 95% CI: -.01 to .14). Hypothesis
3 was thus not supported. Conversely, the interaction term (parent ratings of history of family
aggression × angry rumination) was significantly related to hostile affect (β = .14, R
2
∆ = .02,
p<.05). The conditional indirect effect between parent ratings of history of family aggression and
abusive supervision perceptions via hostile affect was statistically significant for high angry
rumination (indirect relationship = .18, SE = .06, z = 3.09, 95% CI: .08 to .31) but not for low
angry rumination (indirect relationship = .05, SE = .05, z = 1.05, 95% CI: -.03 to .16; see Table
6). As depicted in Figure 2, the conditional indirect effect becomes significant at the .05 alpha
level for angry rumination values above 2.31. Overall, Hypothesis 4 was supported.
A limitation of Study 3 is that the main study variables were all measured at the same
time. This might pose a problem because respondents’ answers may be influenced by how the
items in the surveys are presented and how they perceive the items are related to one another
(Podsakoff et al.,2012). Studies have shown that correlations between variables can change
depending on item order (Schwarz, Strack, & Mai, 1991). In addition, measuring variables of

History of family aggression and abusive supervision 21
interest at the same time poses the risk that mood state at the time of survey response may
influence the predicted relationships (Podsakoff et al., 2012). These issues were addressed in
Study 4 by introducing three measurement periods over a period of 6.5 months.
Study 4
Participants and Procedure
Data were obtained from full-time employees enrolled in various postgraduate programs
in business, science, and education at a large university in the Philippines. Lecturers were
approached to obtain permission to distribute questionnaires to participants. Research team
members visited the classes to explain research goals and to assure participant anonymity and
voluntary participation. At Time 1, a survey assessing history of family aggression, angry
rumination and control variables were administered to 331 participants. They returned completed
questionnaires in sealed envelopes the following week in class, where research team members
collected a total of 274 surveys. Survey forms with a large number of missing responses (n=17)
and from participants with no direct reports (n=10) were excluded. Two weeks after the Time 1
data collection, we administered a second survey assessing hostile affect and hostile cognitions
during class time. A total of 301 surveys were returned. A third wave of data (Time 3) was
collected approximately 6 months after the second survey. At this point, each participant
received a sealed envelope containing the subordinate questionnaire (assessing neuroticism and
abusive supervision) to be passed along to one of their subordinates. Completed questionnaires
were returned via postal mail in a pre-paid envelope. We received 198 subordinate surveys. After
removal of surveys with a large number of missing responses or incorrect identity codes, we
were able to match 162 supervisor-subordinate dyads across the three measurement periods. The
supervisor sample consisted of 46.90% males; with a mean age of 33.47 years. 63% had been

History of family aggression and abusive supervision 22
working in their companies between 1-5 years and had supervised their subordinate for an
average of 3.44 years. The subordinate sample comprised 52.50% males; approximately 84%
were between 21-30 years of age; and 82.10% had been working in their respective companies
for five years or less.
Measures and Results
The same measures and controls used in Study 2 were retained for Study 4, the only
exception being the assessment of subordinates’ neuroticism via Goldberg’s (1992) scale.
Descriptive statistics are reported in Table 5. Consistent with Hypotheses 1b and 2, there was a
significant positive relationship of supervisors’ history of family aggression with implicit hostile
cognitions (r = .43, p<.001) and hostile affect (r = .65, p<.001), respectively. The interaction
term (supervisors’ history of family aggression × angry rumination) was not significantly
associated with implicit hostile cognitions (β = .05, R
2
∆ = .00, ns). The conditional indirect effect
between supervisors’ family aggression history and abusive supervision via implicit hostile
cognitions was also non-significant for both high (indirect effect = .07, SE = .04, z = 1.87, 95%
CI: .00 to .13) and low angry rumination (indirect effect = .05, SE = .04, z = 1.60, 95% CI: .00 to
.11). Thus, Hypothesis 3 was not supported. In contrast, our results indicated that the interaction
term (supervisors’ history of family aggression × angry rumination) was significantly related to
hostile affect (β = .14, R
2
∆ = .02, p<.05) over and above the control variables. Moreover, the
conditional indirect effect of supervisors’ history of family aggression on abusive supervision via
hostile cognitions was significantly stronger for those supervisors with high (indirect effect = .14,
SE = .05, z = 2.79, 95% CI: .04 to .26) as opposed to low angry rumination (indirect effect = .07,
SE = .03, z = 1.99, 95% CI: .02 to .17; see Table 6). Figure 3 shows that the conditional indirect

History of family aggression and abusive supervision 23
effect becomes significant at the .05 alpha level for angry rumination values above 1.46.
Overall, Hypothesis 4 was supported.
General Discussion
Theoretical Contributions
The main goal of our paper was to investigate the relationship between supervisors’
history of family aggression and subordinates’ perceptions of abusive supervision. Our results
speak to several gaps in the literature. First, we presented empirical evidence suggesting that
history of family aggression is associated with abusive supervision. Our findings respond to
Tepper’s (2007) call to focus more empirical attention on understanding the antecedents and
psychological factors that may influence the occurrence of abusive supervision. Notably, we
were able to demonstrate that history of family aggression increases the likelihood of abusive
supervisory behavior over and above the influence of organizational mistreatment (e.g.,
supervisors’ procedural justice, supervisors’ interactional justice, and supervisors’ psychological
contract violation) and other person variables (e.g., supervisors’ demographic characteristics and
subordinates’ neuroticism). This contributes to the growing number of findings that assign
individual differences (i.e., previous experiences and personality) a more central role in
explaining workplace aggression (Douglas & Martinko, 2001; Greenberg & Barling, 1999). Like
all employees, supervisors bring their predispositions and experiences from non-work
environments to the workplace, which in turn impacts how they interact with their constituents.
Our results are also in line with social learning theory (Bandura, 1973), which emphasizes the
important role of observational learning in acquiring aggressive social behavior. Modeling the
aggressive behaviors of their parents, abusive supervisors may have learned the utility of
aggression as a means of settling interpersonal conflicts in the workplace in general and of

History of family aggression and abusive supervision 24
managing subordinates in particular. To this effect, abusive supervision could be conceived as a
form of aggressive behavior that is not only displaced but also socially learned from influential
socialization agents.
Second, there has been limited research on psychological factors that influence the
abusive supervision phenomenon. In this study, we found support for the proposition that
supervisors’ history of family aggression increases the likelihood of hostility-related cognitions
and affect to develop, thus corroborating our theoretical arguments based on social learning
theory (Bandura, 1973). Accordingly, abusive supervision can be understood as being influenced
by accessible hostility-related thoughts and feelings in the form of aggressive scripts and
concepts, which had been stored and strengthened in memory through repeated observations of
inter-parental aggression. To this end, our proposed model actually provides an explanation of
the factors underlying the sustained characteristic of abusive supervision, as abusive supervisors
may engage in repeated aggression towards their subordinates because of their highly accessible
hostile cognitions and hostile affect. In essence, the frequent experience of hostile thoughts and
feelings increases abusive supervisors’ tendencies to react with hostility in social situations and
especially towards their subordinates due to the inherent power asymmetry (cf. Keltner,
Gruenfeld, & Anderson, 2003).
It is also important to note that the hypothesized positive relationship between history of
family aggression and hostile cognitions was significant for both explicit and implicit cognitive
measures, with hostile cognitions being measured using the NOBAGS (explicit; Huesmann &
Guerra, 1997) in Study 1 and with the WCT (implicit; Anderson et al., 2003) in Studies 2, 3, and
4. As such, our empirical findings provide preliminary evidence that previous learning
experiences in the family environment influence cognitions both at the explicit and implicit level.

History of family aggression and abusive supervision 25
Specifically, history of family aggression predisposes individuals to think in an aggressive
manner not only through conscious beliefs about its acceptability but also via increasing
aggressive concepts in memory. This view is in line with the proposition that hostile cognitions
can include both beliefs about the acceptability of aggression (Bandura, 1973) as well as
aggressive scripts (Huesmann & Guerra, 1997). We should note at this point, however, that we
did not find any support for Hypothesis 3 positing a conditional indirect effect via hostile
cognitions. Recall that supervisors’ history of family aggression and hostile affect were both
measured using explicit measures, whereas hostile cognitions were assessed using an implicit
measure in Studies 2 to 4. Given that explicit and implicit measures tap into different aspects of a
phenomenon (James & McIntyre, 2000), it is possible that the non-significant interactions
resulted from the differences in measurement approaches.
Finally, regarding the study’s contributions, we deem three issues especially notable.
First, we were able to constructively replicate our results in four studies, thereby providing
evidence for their generalizability. Second, to our knowledge, we are the first to use an implicit
measure of social cognition in the context of abusive supervision. The study takes into account
the implicit aspect of cognitive processing that was at one time consciously controlled but had
become automatic (Greenwald & Banaji, 1995). We were able to show that even the more
automatic and unconscious aspects of hostile cognitions can increase the likelihood of engaging
in abusive behaviors. Indeed, Greenwald and Banaji (1995) noted that the use of implicit or
indirect measures of cognition can extend the scope of constructs under investigation because
“much of social cognition occurs in an implicit mode” (p. 20). Lastly, we were able to provide
preliminary evidence that angry rumination may increase the occurrence of abusive supervision.
That is, when supervisors ruminate about their previous experiences such as family aggression,

History of family aggression and abusive supervision 26
they also relive the accompanying emotions associated with the remembered event. Rumination
then increases the frequent activation of hostile emotions, such as anger, which also activates
associated emotions and memories in the semantic network. Indeed, rumination has been
recognized as a strategy that can be utilized either to manage or exacerbate negative emotions
(Gross & Thompson, 2007).
Practical Implications
Our study results have several implications for managerial practice. First, executive
coaching through cognitive-behavioral coaching (CBC; Ducharme, 2004) may prove useful by
exposing abusive supervisors to their aggressive beliefs and helping them dispute their hostile
thoughts. To the extent that organizations find themselves ill-equipped to provide such
interventions, we suggest referrals to leadership training programs that train supervisors to deal
with subordinates in non-aggressive ways with a particular focus on better managing hostile
thoughts and emotions. One example for such an intervention would be emotional intelligence
training (EI; Clarke, 2006), which could help managers and employees better deal with their own
emotions and those of others (Salovey & Mayer, 1990). The value of EI training in the present
context lies with increasing awareness of one’s emotions and emotional triggers as well as the
feelings and needs of subordinates (e.g., subordinates needing feedback and support; Palmer,
Walls, Burgess, & Stough, 2001). Given that angry rumination may exacerbate the experience of
negative emotions, EI training may also be beneficial by offering alternative emotional
regulation strategies such as attentional deployment in the form of distraction. Abusive
supervisors may be taught how to change their internal focus from their unpleasant experiences
in the past to evoking thoughts and memories that are inconsistent with their hostile emotions.

History of family aggression and abusive supervision 27
Second, our findings support the notion that implicit social cognitions tend to increase
abusive behavior in the workplace. We believe that this finding has particular importance in
employee recruitment and selection. Specifically, organizations may better identify and screen
out aggression-prone individuals through the use of implicit measures – such as the conditional
reasoning test of aggression (CRTA; James & McIntyre, 2000) – during the selection process
(Bing, Burroughs, Whanger, Green, & James, 2000). Indeed, studies have shown that
aggressiveness has an implicit component that can be reliably assessed and measured in the
workplace (Bing et al., 2007; James & Mazerolle, 2002).
Limitations
Although the results of this study supported most of the hypothesized relationships, our
study is not without limitations. First, arousal was not included in the proposed model due to
practical considerations. Increased psychological (e.g., perceptions of excitement or relaxation)
and physiological (e.g., heart rate or blood pressure) arousal leads to heightened aggressive
behavior, similar to the influence of hostile cognitions and affect (Anderson & Bushman, 2002).
Our study focused on the role of cognitions and affect because these variables can be easily
measured within an organizational context. In contrast, assessing arousal requires precise and
controlled measurements.
Second, it is also important to note the large cross-source effects we found between
supervisor reports of hostile affect and subordinate reports of abusive supervision (r < .50),
which exceed levels typically reported in the literature (Berry, Carpenter, & Barratt, 2012)
4
. We
offer two plausible explanations for this finding. First, supervisors who experienced higher levels
of hostile affect may have expressed these in the workplace which in turn influenced
subordinates’ emotions and perceptions of abusive supervision similar to emotional contagion

History of family aggression and abusive supervision 28
(Wu & Hu, 2009). Second, the large correlations may be due to cultural factors, specifically the
concept of familism (i.e., belief system characterized by loyalty, reciprocity, and solidarity;
Triandis, 1995). In collectivist countries such as the Philippines, the supervisor may be construed
as a symbolic representation of a parent who is expected to provide material and emotional
support (Restubog & Bordia, 2006). Thus, it is possible that subordinates in our sample were
more sensitive to abusive supervision because it might be interpreted as a form of betrayal in
societies high in collectivism and familism. However, it should be noted that we were able to
consistently obtain significant correlations of similar magnitude across four independent samples
suggesting that results are not due to Type 1 error.
Third, our research was cross-sectional in nature (despite introducing time lags in Studies
3 and 4). As such, it precludes us from drawing conclusions about cause-and-effect relationships.
In reflecting about alternatives to our theorizing, we point to the theoretical work by Douglas and
colleagues (2008). In modeling how situational and personality factors result in workplace
aggression, these authors outlined three alternative processes involving affect/emotions,
attributions, and attitudes. These processing routes vary in terms of the level of deliberate or
mindful processing (i.e., automatic versus conscious processing of stimuli), which alludes to the
salience of temporal processes. Moreover, with the current research design, we cannot rule out
the possible role of genetic influences in the occurrence of abusive supervision. Rowe (1994)
proposed that the transmission of aggressive behavior may be inherited rather than learned.
Genetic influences may also interact with the environment (i.e., socially learned behavior)
resulting in increased aggressive tendencies. Indeed, a number of studies have shown that
individuals possessing a low transcription variant of the monoamine oxidase-A gene are at risk
for aggression only when exposed to an abusive home environment (Caspi, et al., 2002; Kim-

History of family aggression and abusive supervision 29
Cohen et al., 2006). Moreover, in a longitudinal sample of monozygotic and dizygotic twins,
Jafee and colleagues (2005) found that the positive relationship between child maltreatment and
future conduct problems becomes stronger as heritable risk increases (e.g., children whose
monozygotic twin has conduct disorder). These limitations emphasize the need to employ
experimental and longitudinal research designs that would not only assess the temporal ordering
of the relationships presented here but also account for a more complete picture of when
exposure to aggression relates to future abusive behavior.
The fourth limitation concerns the measures used to assess history of family aggression
and hostile affect. With regard to history of family aggression, a major concern lies in its
retrospective nature. For example, individuals may not accurately remember what had happened
in the past, or they may reconstruct their previous experiences to suit their current needs or
circumstances (Squire, 1989). To minimize these issues, we implemented several procedures (as
recommended in family aggression research (Hardt & Rutter, 2004; Straus, 1979), such as a)
specifying a one-year referent period to aid recall of family aggression, b) asking participants to
rate specific and concrete aggressive acts to reduce subjectivity, and c) cross-validating results
obtained in Studies 1, 2, and 4 by obtaining parent ratings of history of family aggression in
Study 3. Despite these steps, social desirability and memory recall problems cannot be entirely
ruled out. Thus, future research may benefit from obtaining more objective measures of family
aggression (e.g., archival records) if feasible. Another measurement issue arises with how hostile
affect was assessed across the four studies. Hostile affect was measured using self-reports of the
extent to which the respondents experienced specific emotions such as being “angry” and
“aggravated” (Anderson et al., 1995). Although this measure captures affective reactions, it is
also self-ascribed and assesses conscious discrete emotions, which may not necessarily tap the

History of family aggression and abusive supervision 30
automatic aspects of hostile affect. In this regard, it would have been ideal to measure hostile
affect using an implicit measure similar to the word completion task used for hostile cognitions.
In summary, our research takes a first step towards understanding the role of supervisor-
level factors such as history of family aggression and angry rumination in the abusive
supervision dynamics. Results from the four studies indicated that supervisors’ history of family
aggression was associated with increased hostile affect, explicit and implicit hostile cognitions,
which in turn were related to increased perceptions of abusive supervision. Furthermore, angry
rumination further strengthened these relationships by amplifying hostile affect. The current
study provided evidence that abusive supervision can be conceptualized not only as a reaction to
workplace injustice but also as a form of socially learned behavior.

History of family aggression and abusive supervision 31
References
Abelson, R. P. (1981). Psychological status of the script concept. American Psychologist, 36,
715-729.
Anderson, C. A., & Bushman, B. J. (2002). Human aggression. Annual Review of Psychology,
53, 27-51.
Anderson, C. A., Carnagey, N. L., & Eubanks, J. (2003). Exposure to violent media: The effects
of songs with violent lyrics on aggressive thoughts and feelings. Journal of Personality
and Social Psychology, 84, 960-971.
Anderson, C. A., Deuser, W. E., & DeNeve, K. M. (1995). Hot temperatures, hostile affect,
hostile cognition, and arousal: Tests of a general model of affective aggression.
Personality and Social Psychology Bulletin, 21, 434-448.
Aryee, S., Chen, Z. X., Sun, L., & Debrah, Y. A. (2007). Antecedents and outcomes of abusive
supervision: Test of a trickle-down model. Journal of Applied Psychology, 92, 191-201.
Aquino, K., & Bradfield, M. (2000). Perceived victimization in the workplace: The role of
situational factors and victim characteristics. Organization Science, 11, 525-537.
Bandura, A. (1965). Influence of models’ reinforcement contingencies on the acquisition of
imitative responses. Journal of Personality and Social Psychology, 1, 589-595.
Bandura, A. (1973). Aggression: A social learning analysis. Englewood Cliffs, NJ: Prentice-
Hall.
Bandura, A., Ross, D., & Ross, S. A. (1963). Imitation of film-mediated aggressive models. The
Journal of Abnormal and Social Psychology, 66, 3-11.

History of family aggression and abusive supervision 32
Becker, T. E. (2005). Potential problems in the statistical control of variables in organizational
research: A qualitative analysis with recommendations. Organizational Research
Methods, 8, 274-289.
Berry, C. M., Carpenter, N. C., & Barratt, C. L. (2012). Do other-reports of counterproductive
work behavior provide an incremental contribution over self-reports? A meta-analytic
comparison. Journal of Applied Psychology, 97, 613-636.
Bies, R. J. (2001). Interactional (in)justice: The sacred and the profane. In J. Greenberg & R.
Cropanzano (Eds.), Advances in organizational justice. Stanford, CA: Stanford
University Press.
Bing, M. N., Stewart, S. M., Davison, H. K., Green, P. D., McIntyre, M. D., & James, L. R.
(2007). An integrative typology of personality assessment for aggression: Implications
for predicting counterproductive workplace behavior. Journal of Applied Psychology, 92,
722-744.
Blau, P. M. (1964). Exchange and power in social life. New York: John Wiley.
Burrowes, B. D., & Halberstadt, A. G. (1987). Self- and family-expressiveness styles in the
experience and expression of anger. Journal of Nonverbal Behavior, 11, 254-268.
Bushman, B. J., & Geen, R. G. (1990). Role of cognitive-emotional mediators and individual
differences in the effects of media violence on aggression. Journal of Personality and
Social Psychology, 58, 156-163.
Cappell, C., & Heiner, R. B. (1990). The intergenerational transmission of family aggression.
Journal of Family Violence, 5, 135-152.
Caspi, A., McClay, J., Moffitt, T. E., Mill, J., Martin, J., Craig, I. W., et al. (2002). Role of
genotype in the cycle of violence in maltreated children. Science, 297, 851-854.

History of family aggression and abusive supervision 33
Chermack, S. T., & Walton, M. A. (1999). The relationship between family aggression history
and expressed aggression among college males. Aggressive Behavior, 25, 255-267.
Clark, M. S., & Isen, A. M. (1981). Toward understanding the relationship between feeling states
and social behavior. In A. H. Hastorf & A. M. Isen (Eds.), Cognitive social psychology
(pp. 73-108). New York: Elsevier.
Clarke, N. (2006). Emotional intelligence training: A case of caveat emptor. Human Resource
Development Review, 5, 422-441.
Cummings, E. M. (1987). Coping with background anger in early childhood. Child Development,
58, 976-984.
Denson, T. F., Pedersen, W. C., & Miller, N. (2006). The displaced aggression questionnaire.
Journal of Personality and Social Psychology 90, 1032-1051.
Denson, T. F., Pedersen, W. C., Ronquillo, J., & Nandy, A. S. (2009). The angry brain: Neural
correlates of anger, angry rumination, and aggressive personality. Journal of Cognitive
Neuroscience, 21, 734-744.
Douglas, S. C., Kiewitz, C., Martinko, M. J., Harvey, P., Younhee, K. I. M., & Jae Uk, C.
(2008). Cognitions, emotions, and evaluations: An elaboration likelihood model for
workplace aggression. Academy of Management Review, 33, 425-451.
Douglas, S. C., & Martinko, M. J. (2001). Exploring the role of individual differences in the
prediction of workplace aggression. Journal of Applied Psychology, 86, 547-559.
Ducharme, M.J. (2004). The cognitive-behavioral approach to executive coaching. Consulting
Psychology Journal: Practice and Research, 56, 214-224.
Dusek, J. B., & Danko, M. (1994). Adolescent coping styles and perceptions of parental child
rearing. Journal of Adolescent Research, 9, 412-426.

History of family aggression and abusive supervision 34
Eisenberg, N., Cumberland, A., & Spinrad, T. L. (1998). Parental socialization of emotion.
Psychological Inquiry, 9, 241-273.
Eisenberg, N., Gershoff, E. T., Fabes, R. A., Shepard, S. A., Cumberland, A. J., Losoya, S. H., et
al. (2001). Mothers’ emotional expressivity and children’s behavior problems and social
competence: Mediation through children’s regulation. Developmental Psychology, 37,
475-490.
Fredrickson, B. L. (1998). Cultivated emotions: Parental socialization of positive emotions and
self-conscious emotions. Psychological Inquiry, 9, 279-281.
Frost, B. C., Ko, C.-H. E., & James, L. R. (2007). Implicit and explicit personality: A test of a
channeling hypothesis for aggressive behavior. Journal of Applied Psychology, 92, 1299-
1319.
Game, A. M. (2008). Negative emotions in supervisory relationships: The role of relational
models. Human Relations, 61, 355-393.
Garcia, P. R. J. M., Restubog, S. L. D., & Denson, T. F. (2010). The moderating role of prior
exposure to aggressive home culture in the relationship between negative reciprocity and
aggression. Journal of Research in Personality, 44, 380-385.
Goldberg, L.R. (1992). The development of markers for the Big-Five factor structure.
Psychological Assessment, 4, 26-42.
Greenberg, L., & Barling, J. (1999). Predicting employee aggression against coworkers,
subordinates and supervisors: The roles of person behaviors and perceived workplace
factors. Journal of Organizational Behavior, 20, 897-913.
Greenwald, A. G., & Banaji, M. R. (1995). Implicit social cognition: Attitudes, self-esteem, and
stereotypes. Psychological Review, 102, 4-27.

History of family aggression and abusive supervision 35
Gross, J. J., & Thompson, R. A. (2007). Emotion regulation: Conceptual foundations. In J. J.
Gross (Ed.), Handbook of emotion regulation (pp. 3-24). New York: The Guilford Press.
Guerra, N. G., Huesmann, L. R., & Spindler, A. (2003). Community violence exposure, social
cognition, and aggression among urban elementary school children. Child Development,
74, 1561-1576.
Hardt, J., & Rutter, M. (2004). Validity of adult retrospective reports of adverse childhood
experiences: Review of the evidence. Journal of Child Psychology and Psychiatry, 45,
260-273.
Harrison, D. A., Price, K. H., & Bell, M. P. (1998). Beyond relational demography: Time and the
effects of surface- and deep-level diversity on work group cohesion. Academy of
Management Journal, 41, 96-107.
Hershcovis, M. S., Turner, N., Barling, J., Arnold, K. A., Dupré, K. E., Inness, M., ... &
Sivanathan, N. (2007). Predicting workplace aggression: A meta-analysis. Journal of
Applied Psychology, 92, 228.
Hoobler, J. M., & Brass, D. J. (2006). Abusive supervision and family undermining as displaced
aggression. Journal of Applied Psychology, 91, 1125-1133.
House, R. J., Shane, S. A., & Herold, D. M. (1996). Rumors of the death of dispositional
research are vastly exaggerated. Academy of Management Review, 21, 203-224.
Huesmann, L. R. (1988). An information processing model for the development of aggression.
Aggressive Behavior, 14, 13-24.
Huesmann, L. R., Eron, L. D., Lefkowitz, M. M., & Walder, L. O. (1984). The stability of
aggression over time and generations. Developmental Psychology, 20, 1120-1134.

History of family aggression and abusive supervision 36
Huesmann, L. R., & Guerra, N. G. (1997). Children’s normative beliefs about aggression and
aggressive behavior. Journal of Personality and Social Psychology, 72, 408-419.
Inness, M., Barling, J., & Turner, N. (2005). Understanding supervisor-targeted aggression: A
within-person, between-jobs design. Journal of Applied Psychology, 90, 731-739.
Jaffee, S. R., Caspi, A., Moffitt, T. E., Dodge, K. A., Rutter, M., Taylor, A., et al. (2005). Nature
× nurture: Genetic vulnerabilities interact with physical maltreatment to promote conduct
problems. Development and Psychopathology, 17, 67-84.
James, L. R., & Mazerolle, M. D. (2002). Personality in work organizations. Thousand Oaks,
CA: Sage.
James, L. R., & McIntyre, M. D. (2000). Conditional Reasoning Test of Aggression Test Manual.
Knoxville, TN: Innovative Assessment Technology.
John, O. P., & Srivastava, S. (1999). The Big Five Trait Taxonomy: History, measurement, and
theoretical perspectives. In L. A. Pervin & O. P. John (Eds.), Handbook of personality:
Theory and research (2nd ed., pp. 102-138). New York: Guilford Press.
Keltner, D., Gruenfeld, D. H., & Anderson, C. (2003). Power, approach, and inhibition.
Psychological Review, 110, 265-284.
Kiazad, K., Restubog, S. L. D., Zagenczyk, T. J., Kiewitz, C., & Tang, R. L. (2010). In pursuit of
power: The role of authoritarian leadership in the relationship between supervisors'
machiavellianism and subordinates' perceptions of abusive supervisory behavior. Journal
of Research in Personality, 44, 512-519.
Kiewitz, C., Restubog, S. L. D., Zagenczyk, T. J., Scott, K. D., Garcia, P. R. J. M., & Tang, R. L.
(2012). Sins of the parents: Self-control as a buffer between supervisors’ previous
experience of family undermining and subordinates’ perceptions of abusive supervision.

History of family aggression and abusive supervision 37
The Leadership Quarterly, 23, 869-882.
Kim, K. J., Conger, R. D., Lorenz, F. O., & Elder, G. H. J. (2001). Parent-adolescent reciprocity
in negative affect and its relation to early adult social development. Developmental
Psychology, 36, 366-380.
Kim-Cohen, J., Caspi, A., Taylor, A., Williams, B., Newcombe, R., Craig, I. W., et al. (2006).
MAOA, maltreatment, and gene-environment interaction predicting children’s mental
health: New evidence and a meta-analysis. Molecular Psychiatry, 11, 903-913.
Lian, H., Lance Ferris, D., & Brown, D. J. (2012). Does taking the good with the bad make
things worse? How abusive supervision and leader–member exchange interact to impact
need satisfaction and organizational deviance. Organizational Behavior and Human
Decision Processes, 117, 41-52.
Lykken, D. T. (1968). Statistical significance in psychological research. Psychological Bulletin,
70, 151-159.
McClelland, D. C., Koestner, R., & Weinberger, J. (1989). How do self-attributed and implicit
motives differ? Psychological Review, 96, 690-702.
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.
Niehoff, B. P., & Moorman, R. H. (1993). Justice as a mediator of the relationship between
methods of monitoring and organizational citizenship behavior. Academy of Management
Journal, 36, 527-556.
O’Keefe, M. (1998). Factors mediating the link between witnessing interparental violence and
dating violence. Journal of Family Violence, 13, 39-57.

History of family aggression and abusive supervision 38
Palmer, B., Walls, M., Burgess, Z., & Stough, C. (2001). Emotional intelligence and effective
leadership. Leadership & Organization Development Journal, 22, 5-10.
Pedersen, W. C., Denson, T. F., Goss, R. J., Vasquez, E. A., Kelley, N. J., & Miller, N. (2011).
The impact of rumination on aggressive thoughts, feelings, arousal, and behaviour.
British Journal of Social Psychology, 50, 281-301.
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.
Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling strategies for assessing and
comparing indirect effects in multiple mediator models. Behavior Research Methods, 40,
879-891.
Preacher, K. J., Rucker, D. D., & Hayes, A. F. (2007). Addressing moderated mediation
hypotheses: Theory, methods, and prescriptions. Multivariate Behavioral Research, 42,
185-227.
Rafferty, A. E., & Restubog, S. L. D. (2011). The influence of abusive supervisors on followers'
organizational citizenship behaviours: The hidden costs of abusive supervision. British
Journal of Management, 22, 270-285.
Ray, R., Ochsner, K., Cooper, J., Robertson, E., Gabrieli, J., & Gross, J. (2005). Individual
differences in trait rumination and the neural systems supporting cognitive reappraisal.
Cognitive, Affective, & Behavioral Neuroscience, 5, 156-168.
Restubog, S. L. D., & Bordia, P. (2006). Workplace familism and psychological contract breach
in the Philippines. Applied Psychology: An International Review, 55, 563-585.

History of family aggression and abusive supervision 39
Restubog, S. L. D., Scott, K. L., & Zagencyzk, T. J. (2011). When distress hits home: The role of
contextual factors and psychological distress in predicting employees' responses to
abusive supervision. Journal of Applied Psychology, 96, 713-729.
Robinson, S.L., & Morrison, E.W. (2000). The development of psychological contract breach
and violation: A longitudinal study. Journal of Organizational Behavior, 21, 525-546.
Rowe, D. C. (1994). The limits of family influence: Genes, experience, and behavior. New York:
Guilford Press.
Salovey, P., & Mayer, J. D. (1990). Emotional intelligence. Imagination, Cognition, &
Personality, 9, 185-211.
Schwarz, N., Strack, F., & Mai, H. P. (1991). Assimilation and contrast effects in part-whole
question sequences: A controversial logic analysis. Public Opinion Quarterly, 55, 3-23.
Shoss, M. K., Eisenberger, R., Restubog, S. L. D., & Zagenczyk, T. J. (2013). Blaming the
organization for abusive supervision: The roles of perceived organizational support and
supervisor's organizational embodiment. Journal of Applied Psychology, 98, 158.
Squire, L. (1989). On the course of forgetting very long-term memory. Journal of Experimental
Psychology: Learning, Memory, and Cognition, 15, 241-245.
Straus, M. A. (1979). Measuring intrafamily conflict and violence: The Conflict Tactics (CT)
scales. Journal of Marriage and the Family, 41, 75-88.
Straus, M. A. (1990). Ordinary violence, child abuse, and wife beating: What do they have in
common? New Brunswick, NJ: Transaction Publishers.
Straus, M. A., Hamby, S. L., Boney-McCoy, S., & Sugarman, D. B. (1996). The revised Conflict
Tactics Scales (CTS2): development and preliminary psychometric data. Journal of
Family Issues 17, 283-316.

History of family aggression and abusive supervision 40
Stucke, T. S., & Baumeister, R. F. (2006). Ego depletion and aggressive behavior: Is the
inhibition of aggression a limited resource? European Journal of Social Psychology, 36,
1-13.
Tepper, B. J. (2000). Consequences of abusive supervision. Academy of Management Journal,
43, 178-190.
Tepper, B. J. (2007). Abusive supervision in work organizations: Review, synthesis, and research
agenda. Journal of Management, 33, 261-289.
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.
Tremblay, R. E., Japel, C., Perusse, D., McDuff, P., Boivin, M., Zoccolillo, M., et al. (1999). The
search for the age of ‘onset’ of physical aggression: Rousseau and Bandura revisited.
Criminal Behaviour and Mental Health, 9, 8-23.
Triandis, H. C. (1995). Individualism-collectivism: New directions in social psychology. Boulder,
CO: Westview Press.
Wu, T., & Hu, C. (2009). Abusive supervision and employee emotional exhaustion:
Dispositional antecedents and boundaries. Group and Organization Management, 34,
143-169.

History of family aggression and abusive supervision 41
Footnotes
1
Both supervisors and subordinates (Study 1 to 4) and the supervisors’ parent (Study 3)
received coffee vouchers as an incentive for participation.
2
The same set of control variables were included in the analyses for Studies 3 and 4.
Following the recommendation of Becker (2005), analyses for Studies 2 to 4 were
repeated without the controls. The pattern of results were essentially identical (with or
without control variables), indicating that the significant relationships among the study
variables were not due to the presence of the covariates.
3
We also assessed the indirect effects of hostile cognitions and hostile affect using
Preacher and Hayes’ (2008) bootstrapping technique for multiple mediator models for
Studies 2, 3, and 4. The total indirect effect of history of family aggression through both
hostile cognitions and hostile affect was significant for all Studies [Study 2, point
estimate = .20, 95% CI: .12 to .31; Study 3, point estimate = .17, 95% CI: .08 to .30; and
Study 4, point estimate = .20 95% CI: .11 to .30]. Examination of the specific indirect
effects revealed that hostile cognitions was a significant mediator [Study 2, point estimate
= .05, 95% CI: .02 to .10; Study 3, point estimate = .04, 95% CI: .01 to .10; and Study 4,
point estimate = .09, 95% CI: .04 to .15] as was hostile affect [Study 2, point estimate =
.15, 95% CI: .06 to .25; Study 3, point estimate = .13, 95% CI: .05 to .24; and Study 4,
point estimate = .12, 95% CI: .03 to .21]. Overall, these results supported both
Hypotheses 1a and 1b.
4
We would like to thank the anonymous reviewer for this insightful comment.

History of family aggression and abusive supervision 42
Table 1
Summary of Research Design for Each Study
Study Hypotheses
tested Sample History of
family
aggression
Rumination
(conditional
indirect
effects)
Explicit or
implicit
cognitions Controls Time lag
1 H1a and H2
154 supervisor-
subordinate
dyads Self-reported No Explicit
(NOBAGS) No No
2 H1b, H2,
H3, and H4
199 supervisor-
subordinate
dyads Self-reported Yes Implicit
(WCT)
Procedural and
interactional justice,
psychological
contract violation,
supervisor age,
gender, and duration
of working
relationship,
subordinates’
neuroticism
No
3 H1b, H2,
H3, and H4
134 parent-
supervisor-
subordinate triads
Parent-rated Yes Implicit
(WCT) Same as Study 2 Yes
(3 months)
4 H1b, H2,
H3, and H4
162 supervisor-
subordinate
dyads Self-reported Yes Implicit
(WCT) Same as Study 2
Yes
(6.5 months with
3 measurement
periods)
Note: NOBAGS refers to the General Approval of Aggression subscale of Huesmann and Guerra’s (1997) Normative Beliefs about
Aggression Scale. WCT refers to the Word Completion Task (WCT) by Anderson, Carnagey, and Eubanks (2003).

History of family aggression and abusive supervision 43
Table 2
Means, Standard Deviations, and Intercorrelations among Variables in Study 1
Variables M SD 1 2 3 4
1. Supervisors’ history of family aggression 3.60 1.18 (.86)
2. Supervisors’ hostile cognitions (explicit;
NOBAGS) 3.71 1.05 .57*** (.83)
3. Supervisors’ hostile affect 3.63 1.44 .56*** .47*** (.98)
4. Subordinates’ perceptions of abusive
supervision 3.70 1.36 .60*** .54*** .72*** (.97)
Note: Study 1, N = 154; * p < .05, ** p < .01, *** p < .001. Cronbach alpha reliabilities are reported along the diagonal.

History of family aggression and abusive supervision 44
Table 3
Means, Standard Deviations, and Intercorrelations among Variables in Study 2
Variables M SD 1 2 3 4 5 6 7 8 9 10 11 12
1. Supervisor gender .43 .50
2. Supervisor age 33.26 7.34 .07
3. Procedural justice 5.27 1.38 -.01 .15*
(.97)
4. Interactional
justice 5.50 1.45 .14 -.05 .50***
(.95)
5. Psychological
contract violation 2.33 1.34 -.23** .01 -.24** -.32** (.90)
6. Duration of work
relationship with
supervisor 3.18 3.83 .12 .40**
.01 -.03 .09 --
7. Subordinate
neuroticism 3.16 1.11 -.06 .13 .13 -.37** .32**
-.01 (.88)
8. Supervisors’
history of family
aggression 1.39 1.34 -.20** -.09 -.08 -.14 .33**
-.09 .20**
(.96)
9. Supervisors’ angry
rumination 3.00 1.50 -.09 -.07 -.16* -.17* .49**
.02 .34**
.41**
(.90)
10. Hostile cognitions
(implicit; WCT) .19 .08 -.08 -.04 -.07 -.35** .17* -.10 .26**
.32**
.31**
11. Hostile affect 2.37 1.30 -.09 -.04 -.19** -.31** .36**
-.07 .35**
.67**
.46**
.38**
(.97)
12. Subordinates’
perceptions of
abusive
supervision
2.09 1.20 -.26** -.02 -.25** .56** .34**
-.15*
.47**
.35**
.34**
.49**
.52**
(.96)
Note: Study 2, N = 199; * p < .05, ** p < .01, *** p < .001. Cronbach alpha reliabilities are reported along the diagonal.

History of family aggression and abusive supervision 45
Table 4
Means, Standard Deviations, and Intercorrelations among Variables in Study 3
Variables M SD 1 2 3 4 5 6 7 8 9 10 11 12
1. Supervisor gender .40 .49
2. Supervisor age 31.25 3.40 -.04
3. Procedural justice 5.10 1.43 -.02 .13 (.96)
4. Interactional
justice 5.33 1.49 .07 .02 .57** (.98)
5. Psychological
contract violation 2.37 1.32 -.22 .01 -.30** -.30** (.91)
6. Duration of work
relationship with
supervisor 2.63 2.26 .05 .21*
-.07 -.13 .20*
7. Subordinate
neuroticism 3.21 1.09 -.06 .00 -.24** -.32** .26**
.06 (.89)
8. Parent ratings of
history of family
aggression 1.56 1.08 -.15 -.12 .04 -.05 .13 -.05 .24**
(.93)
9. Supervisors’ angry
rumination 3.10 1.51 -.13 -.05 -.12 -.19* .45**
.10 .38**
.41**
(.95)
10. Hostile cognitions
(implicit; WCT) .19 .08 -.03 .02 .01 -.34** .12 -.12 .27**
.27**
.31**
11. Hostile affect 2.47 1.26 -.03 -.02 -.21* -.37** .21* -.06 .48**
.48**
.40**
.42**
(.97)
12. Subordinates’
perceptions of
abusive
supervision
2.27 1.28 -.18 -.06 -.28** -.51** .35**
-.13 .54**
.30**
.37**
.48**
.60**
(.96)
Note: Study 3, N = 134; * p < .05, ** p < .01, *** p < .001. Cronbach alpha reliabilities are reported along the diagonal.

History of family aggression and abusive supervision 46
Table 5
Means, Standard Deviations, and Intercorrelations among Variables in Study 4
Variables M SD 1 2 3 4 5 6 7 8 9 10 11 12
1. T1 Supervisor
gender .53 .50
2. T1 Supervisor age 33.47 7.72
.08
3.
T1 Procedural
justice 5.23 1.34
-.02 .10 (.95)
4.
T1 Interactional
justice 5.45 1.47
.05 -.07 .48*** (.98)
5.
T1 Psychological
contract violation 2.34 1.31
-.02 -.01 -.24** -.30*** (.90)
6.
T1 Duration work
relationship with
supervisor 3.44 2.85
.07 .37*** -.09 -.15 .16*
7.
T1 Subordinate
neuroticism 3.12 1.04
-.11 .13 -.12 -.42*** .34*** .11 (.96)
8.
T1 Super
visors’
history of family
aggression 1.46 1.31
-.02 -.11 -.04 -.16 .30*** -.06 .24** (.96)
9.
T1 Supervisors’
angry rumination 2.99 1.51
-.04 -.12 -.19* -.13 .46*** .01 .38*** .38***
(.95)
10.
T2 Hostile
cognitions
(implicit; WCT) .18 .08
.02 -.08 -.01 -.37*** .23** -.06 .24** .43***
.34***
11. T2 Hostile affect 2.43 1.30
.01 -.05 -.17* -.32*** .31*** -.01 .33*** .65***
.44***
.49***
(.98)
12. T3 Subordinates’
perceptions of
abusive
supervision
2.08 1.11
-.15 -.03 -.12 -.53*** .37*** -.03 .42*** .45***
.35***
.55***
.54***
(.95)
Note: Study 4, N = 162; * p < .05, ** p < .01, *** p < .001. Cronbach alpha reliabilities are reported along the diagonal.

History of family aggression and abusive supervision 47
Table 6
Summary of Conditional Indirect Effects at low and high Levels of Angry Rumination in Studies 2, 3, and 4
Indirect Relationship via
Implicit Hostile Cognitions Indirect Relationship via
Hostile Affect
Level of
angry
rumination Estimate Standard
Error Z score Confidence
Interval Estimate Standard
Error Z score Confidence
Interval
Study 2:
High .04 .02 1.79 .00 to .09 .16 .05 2.96 .05 to .30
Low .03 .03 1.29 .00 to .09 .05 .03 1.96 .02 to .13
Study 3:
High .02 .03 .73 -.02 to .10 .18 .06 3.09 .08 to .31
Low .05 .04 1.22 -.01 to .14 .05 .05 1.05 -.03 to .16
Study 4:
High .07 .04 1.87 .00 to .13 .14 .05 2.79 .04 to .26
Low .05 .04 1.60 .00 to .11 .07 .03 1.99 .02 to .17

History of family aggression and abusive supervision 48
Figure 1. The conditional indirect effect of supervisors’ history of family aggression on
subordinates’ abusive supervision via hostile affect with angry rumination as a first-stage
moderator (Study 2). Dashed lines represent upper and lower confidence limits (95% bootstrap
CIs).

History of family aggression and abusive supervision 49
Figure 2. The conditional indirect effect of parent ratings of history of family aggression on
subordinates’ abusive supervision via hostile affect with angry rumination as a first-stage
moderator (Study 3). Dashed lines represent upper and lower confidence limits (95% bootstrap
CIs).

History of family aggression and abusive supervision 50
Figure 3. The conditional indirect effect of supervisors’ history of family aggression on
subordinates’ abusive supervision via hostile affect with angry rumination as a first-stage
moderator (Study 4). Dashed lines represent upper and lower confidence limits (95% bootstrap
CIs).