«UNDERSTANDING EMPLOYEES’ BEHAVIORAL REACTIONS TO AGGRESSION IN ORGANIZATIONS by MARIE S. MITCHELL B.S., George Mason University, 1993 M.A.H.R., ...»
The Influence of Individual Factors Trait anger. The notion that anger promotes aggressive reactions is commonly embraced by aggression researchers (Berkowitz, 1990; Geen, 1990). However, anger is a feeling that generally goes away, unlike trait anger, which is a stable characteristic. Trait anger is the tendency to perceive a wide-range of situations as anger-provoking (Fox & Spector, 1999). Not surprisingly, researchers have found that individuals with high trait anger experience anger more easily when they encounter annoying conditions (Spielberger, 1996; Spielberger, Krasner, & Solomon, 1988).
Trait anger also intensifies negative reactions, particularly when individuals feel they are personally attacked (Deffenbacher, 1992). For example, Fox and Spector (1999) found trait anger strongly predicted counter-productive workplace behaviors. Douglas and Martinko (2001) found trait anger heightened incidents of workplace aggression, particularly when individuals also held low levels of self-control. Further, a recent meta-analysis of workplace aggression demonstrated trait anger was a significant predictor of interpersonal and organizational aggression (Hershcovis, Turner et al., in press).
While trait anger strengthens aggressive reactions, it also appears to minimize nonaggressive ones. In a prisoner’s dilemma experiment using wartime conditions, Kassinove, Roth, Owens, and Fuller (2002) found individuals with high trait anger were less likely to engage in neutral or cooperative solutions. Because high trait anger intensifies negativity, these authors argued and found trait anger heightened competitive reactions. Similarly, Deffenbacher (1992) found high trait angry individuals were more likely to act out aggressively and less able to engage in constructive coping behaviors.
Based on this review, I predict individuals with high trait anger will be more likely to respond to supervisor aggression with aggression (whether retaliatory aggression or displaced aggression), and less likely to respond with non-aggression (i.e., constructive problem-solving or withdrawal). Rather, individuals who are low in trait anger are more likely to react to aggression
with non-aggression. Therefore, I predict:
Hypothesis 4: Trait anger will moderate the positive relationships between perceived supervisor aggression and (a) retaliatory aggression and (b) displaced aggression such that the relationships will be stronger when trait anger is high rather than low, and the positive relationships between perceived supervisor aggression and (c) constructive problem-solving and (d) withdrawal will be stronger when trait anger is low rather than
Need for social approval. Researchers have argued that individuals desire a certain level of social contact (e.g., Foa & Foa, 1980; Hill, 1987; Martin, 1984). Foa and Foa (1980) contend social contact can be an exchange resource; the value of that contact depends on the source of the interaction. According to Hill (1987), contact becomes most valuable when it provides individuals with (1) positive stimulation (i.e., a sense of belonging), (2) recognition or praise, and (3) emotional support. When contact provides these rewards, it can be a strong motivator of human behavior (Martin, 1984). However, not all individuals desire social contact to the same extent, meaning some individuals seek social contact and approval more than others (Armeli, Eisenberger, Fasolo, & Lynch, 1998). The need for social approval describes an individual tendency to seek social approval and favorable evaluations of others (Martin, 1984).
In general, research shows that individuals with high need for social approval engage in help-seeking behaviors and desire contact from individuals in more powerful positions to gain favorable evaluations (Martin, 1984). Individuals with a high need for social approval attempt to act in socially appropriate ways and avoid behavior that would be frowned upon by others.
Research has shown that individuals who have high need for social approval seek out support from others (e.g., Hill, 1991; Nadler, 1983), especially if they fail to resolve problems on their own (e.g., DePaulo, 1982; DePaulo, Dull, Greenberg, & Swaim, 1989; Rosen, 1983). Further, because negative interactions heighten feelings of negative affiliation and emotion, individuals with a high need for social approval avoid or retreat from threatening situations (e.g., Exline, 1963; Mehrabian & Ksionzky, 1974; Terhune, 1968).
In this way, individuals with a high need for social approval who are victim to supervisor aggression would be less likely to react aggressively (i.e., retaliatory aggression and displaced aggression) and more likely to react non-aggressively (i.e., constructive problem-solving and
withdrawal). Therefore, I predict:
Hypothesis 5: The need for social approval will moderate the positive relationships between perceived supervisor aggression and (a) constructive problem-solving and (b) withdrawal such that the relationship will be stronger when need for social approval is high rather than low, and the positive relationships between perceived supervisor aggression and (c) retaliatory aggression, (d) displaced aggression will be stronger when need for social approval is low rather than high.
Sample and Procedure Surveys were distributed to individuals called to jury duty by a county circuit court in the southeastern U.S. The jurors were addressed at the beginning of the day before they were called to serve. The researcher explained that the purpose of the survey had nothing to do with jury duty or court system, but instead was to investigate sensitive issues that affect individuals at work. Interested participants picked up surveys from and returned surveys to the researcher.
Over the course of four weeks, 321 individuals participated in the study (29% response rate).
The average age of the participants was 42.4 years old (SD = 12.12); the average company tenure was 8.1 years (SD = 7.82). Approximately 52.7% were currently working in non-management positions, 48.6% were female, and 71.8% were white (12.2% were Hispanic, 8.8% were Black).
Measures Pretest for new and adapted measures. Measures of fear of retaliation, constructive problem-solving and withdrawal were adapted and/or integrated from previously validated measures. As a consequence, these measures were pretested on a separate sample of 62 individuals called for jury duty. The average age of this sample was 43.0 (SD=12.57); the average company tenure was 8.34 years (SD=7.92). Approximately 53.1% were nonmanagement, 48.4% were female, and 71.9% were white (12.5% were Hispanic, 10.9% were Black). The purpose of the pretest was to conduct an exploratory factor analysis and reliability analysis on the adapted items. The results suggest the items of the measures fell on their representative constructs, and all measures held robust reliabilities. (See Appendix A for details of the pretest results.) Perceived supervisor aggression. Supervisor aggression was assessed with the shortened 6-item version of Tepper’s (2000) Abusive Supervision measure. 1 The instructions of the measure were adapted, asking participants to rate the frequency by which they experienced “intentional” behaviors by their immediate supervisor (1=never, 2=once a year, 3=twice a year, 4=several times a year, 5=monthly, 6=weekly, 7=daily) (alpha=.92). Example items are “My boss ridiculed me” and “My boss put me down in front of others.” Fear of retaliation. Fear of retaliation from the supervisor was assessed with an adapted version of Fox and Spector’s (1999) fear of future punishments measure. The 3-item measure asked respondents to indicate their level of agreement on a 7-point Likert-like scale (1=strongly disagree, 7=strongly agree) (alpha=.91). Example items are “I am afraid of reacting against my supervisor for fear of future punishments” and “I would not act out against my supervisor because he/she would retaliate against me.” Aggressive modeling. Consistent with Aquino and Douglas (2003), aggressive modeling was assessed by adapting Robinson and O’Leary-Kelly’s antisocial behavior scale. Two sources of modeling were assessed: supervisor aggressive modeling and, separately, coworker aggressive modeling. Each 9-item measure asks respondents to rate the number of times they observed supervisors (alpha=.91) and coworkers (alpha=.92) engage in the listed behaviors over the course of the past year (1=never, 2=1-3 times, 3=4-6 times, 4=7-9 times; 5=10 or more times). Example items are “Damage property belonging to the organization” and “Started an argument with someone at work.” Mitchell & Ambrose (2004) developed this shortened measure from two separate published data sets. They performed exploratory and confirmatory factor analyses on Tepper’s original 15-item measure, which revealed two distinct factors. One factor represented behaviors of proactive interpersonal abuse by the supervisor (e.g., “ridicules me” and “tells me my thoughts and feelings are stupid”), rather than passive acts disrespect (e.g., “doesn’t give me credit for jobs requiring a lot of effort”). Mitchell and Ambrose argue the shortened 6-item measure best reflects acts of supervisor aggression.
Absolute hierarchical status. Consistent with Aquino, Douglas, and Martinko (2004), respondents were asked to indicate their status in the organization (1=non-management, 2=linemanagement, 3=middle-management, and 4=senior/executive management). Non-management was described as a position that did not supervise other personnel. Line-management was described as a position that supervises non-management personnel. Middle-management was described as a position that supervises line-management personnel. Senior or executive management was described as a position that supervises middle-management personnel.
Trait anger. Trait Anger was assessed using the 7-item anger subscale of Buss and Perry’s (1992) Aggression Questionnaire. The measure assesses an individual’s dispositional tendency toward anger in everyday life. Participants rated their agreement on a 5-point scale (1= very slightly true of me, 5=very highly true of me) (alpha=.79). Sample items include “I have trouble controlling my temper” and “When frustrated, I let my irritation show.” Need for social approval. The 9-item Martin-Larsen Approval Motivation measure (MLAM; Martin, 1984) was used. The MLAM assesses approval seeking tendencies, and asks respondents to rate their agreement on a 7-point Likert-like scale (1=strongly disagree, 7=strongly agree) (alpha=.72). Sample items are “I find it difficult to talk about my ideas if they are contrary to group opinion” and “If there is any criticism or anyone says anything about me, I can take it” (reverse-coded).
Retaliatory aggression. Retaliatory aggression was assessed by adapting Bennett and Robinson’s (2000) and Aquino, Lewis, and Bradfield’s (1999) interpersonal deviance measures.
The 10-item measure asks respondents to rate the frequency they intentionally engaged in behaviors against their supervisor over the past year (1=never, 2=once a year, 3=twice a year, 4=several times a year, 5=monthly, 6=weekly, 7=daily) (alpha=.84). Example items are “Swore at my supervisor” and “Avoided my supervisor on purpose.” Displaced aggression. Three targets of displaced aggression were assessed: toward the organization, coworkers, and customers. Organization displaced aggression was assessed by adapting Robinson and Bennett’s (2000) 12-item organizational deviance measure. Based on a visual review of the organizational deviance items, three items were deleted due to conceptual overlap with other dependent variables. One item conflicted with retaliatory aggression (“Neglected to follow my boss’s instructions”) and two items conflicted with withdrawal behaviors (“Put little effort into my work,” and “Worked slower than I could have worked”).
The resulting 9-item measure asked respondents to rate the frequency they intentionally engaged in the behaviors listed against the organization (alpha=.77). Example items are “Taken property from work without permission” and “Falsified a receipt to get reimbursed for more money than I spent on business expenses.” Coworker and customer displaced aggression were assessed by adapting Robinson and Bennett’s 7-item interpersonal deviance measure to each representative source. Example coworker displaced aggression items are “Made an obscene comment or gesture toward a coworker” and “Publicly embarrassed a coworker” (alpha=.85). Example customer displaced aggression items are “Acted rudely toward a customer” and “Swore at a customer” (alpha=.82). All displaced aggression measures asked respondents to indicate the frequency they intentionally engaged in the stated behaviors over the course of the past year (1=never, 2=once a year, 3=twice a year, 4=several times a year, 5=monthly, 6=weekly, 7=daily).
Constructive problem-solving. Constructive problem-solving was assessed by adapting and integrating Rusbult, Farrell, Rogers and Mainous’ (1988) “voice” measure with Tepper and et al.’s (2001) constructive resistance scale. The 8-item measure assesses general work behavior that actively and constructively tries to improve conditions, and asked respondents to indicate the frequency they engaged in the stated behaviors over the course of the past year (1=never, 2=once a year, 3=twice a year, 4=several times a year, 5=monthly, 6=weekly, 7=daily) (alpha=.93).