Page 1

Journal of Applied Psychology 2009, Vol. 94, No. 2, 511–523

© 2009 American Psychological Association 0021-9010/09/$12.00 DOI: 10.1037/a0013451

Healing the Wounds of Organizational Injustice: Examining the Benefits of Expressive Writing Laurie J. Barclay and Daniel P. Skarlicki University of British Columbia Clinical and health psychology research has shown that expressive writing interventions— expressing one’s experience through writing— can have physical and psychological benefits for individuals dealing with traumatic experiences. In the present study, the authors examined whether these benefits generalize to experiences of workplace injustice. Participants (N ⫽ 100) were randomly assigned to write on 4 consecutive days about (a) their emotions, (b) their thoughts, (c) both their emotions and their thoughts surrounding an injustice, or (d) a trivial topic (control). Post-intervention, participants in the emotions and thoughts condition reported higher psychological well-being, fewer intentions to retaliate, and higher levels of personal resolution than did participants in the other conditions. Participants in the emotions and thoughts condition also reported less anger than did participants who wrote only about their emotions. Keywords: justice, emotion, retaliation, expressive writing

To date, research examining how to recover from fairness violations has focused on the manager’s or the organization’s perspective. Studies, have focused on, for example, changing attributions (i.e., shaping the victim’s attributions with explanations and apologies; e.g., Kim, Ferrin, Cooper, & Dirks, 2004; Shaw, Wild, & Colquitt, 2003), creating social equilibrium (i.e., engaging in social rituals, such as penance or punishment; e.g., Reb, Goldman, Kray, & Cropanzano, 2006), and instituting structural strategies (i.e., implementing structures such as monitoring or incentives to increase positive exchanges and decrease negative exchanges between parties; see Dirks, Lewicki, & Zaheer, in press, for a discussion). Studies have also examined the effectiveness of training managers in fairness principles to prevent violations of fairness and/or to manage employees’ perceptions (Greenberg, 2006; Skarlicki & Latham, 1996). Although these strategies can be effective tools in preventing violations of fairness or the escalation of conflict, they tend to focus on organizational outcomes (e.g., legal claiming, retaliation) rather than helping the victim recover (e.g., mitigating physical and psychological health consequences). They also relegate the victim to a passive rather than an active role. Thus, these strategies might be limited in their ability to address the victim’s experience of injustice. In the present study, we argue for victim-centered interventions that go beyond outcomes that are relevant primarily to organizations and include outcomes that are particularly important to the aggrieved employee, such as physical health and psychological well-being. These interventions are important in light of evidence that organizations and managers tend to underestimate the emotional and cognitive toll of injustice on the employee. Victims, for instance, can be preoccupied with the unfairness to a larger degree than the transgressor (Kramer & Neale, 1998) and can ruminate about a violation, sometimes for years (Matthews, 1988). Bateson (1977) argued that individuals continuously search for closure after negative events occur and that unresolved negative experiences can influence future perceptions and reactions. Hansson, Jones, and Fletcher (1990) found that experiences of injustice or

Individuals who have experienced unfairness in the workplace often report significant, painful, and enduring consequences. Cognitively, individuals are motivated to make sense of the experience and attribute responsibility for the unfairness (Folger & Cropanzano, 1998, 2001). Emotionally, individuals often experience injustice as “hot and burning,” which engenders emotions, such as anger, rage, shame, and guilt (Bies & Tripp, 1996, 2002; Mikula, 1986; Weiss, Suckow, & Cropanzano, 1999). Behaviorally, individuals often seek to “right the wrong” by engaging in retaliation (Skarlicki & Folger, 1997). Experiencing unfairness in the workplace can also take a toll on employees’ physical and psychological health. Workplace unfairness, for instance, is associated with increased anxiety (Harlos & Pinder, 2000), insomnia (Greenberg, 2006), depression (Tepper, 2001), exhaustion (Elovainio, Kivima¨ki, & Helkama, 2001; Elovainio, Kivima¨ki, & Vahtera, 2002), psychiatric disorders (Kivima¨ki, Elovainio, Vahtera, & Ferrie, 2003; Kivima¨ki, Elovainio, Vahtera, Virtanen, & Stansfeld, 2003), and coronary heart disease (Kivima¨ki et al., 2005).

Laurie J. Barclay and Daniel P. Skarlicki, Sauder School of Business, University of British Columbia, Vancouver, British Columbia, Canada. This research was conducted as part of Laurie J. Barclay’s doctoral dissertation. An earlier version of this article was presented at the 2005 Academy of Management conference, Honolulu, Hawaii. The research was supported by grants from the Social Sciences and Humanities Research Council of Canada awarded to Laurie J. Barclay and Daniel P. Skarlicki as well as the Wilfrid Laurier University Grace Anderson Fellowship awarded to Laurie J. Barclay. We thank Peter Darke, Rob Folger, Peter Frost (since deceased), Lisa Keeping, Sally Maitlis, Chet Robie, Sandra Robinson, Marc-David Seidel, Lorne Sulsky, and Tom Tripp for their helpful suggestions as well as comments on drafts of this article. Correspondence concerning this article should be addressed to Laurie J. Barclay, who is now at the School of Business and Economics, Wilfrid Laurier University, 75 University Avenue, Waterloo, Ontario, Canada N2L 3C5. E-mail: lbarclay@wlu.ca 511


512

RESEARCH REPORTS

betrayal retain their symbolic impact across a victim’s lifetime. Victims can also be left on their own to deal with unfairness, such as when managers do not realize that a violation has occurred or are unable or unwilling to repair the violation. Despite previous calls in the literature for tools and interventions that can be used to help victims of workplace injustice manage their experiences (e.g., Shapiro, 2001), the victim’s perspective has received little study. Expressive writing—a victim-centered intervention developed in the clinical and health psychology literature for managing trauma— has been found to have positive physical and psychological benefits for its participants (Pennebaker, 1993, 1997b; Pennebaker & Beall, 1986). In the present study, we investigated whether the expressive writing intervention can mitigate the negative consequences of experiencing organizational injustice. First, we explored whether the benefits observed in expressive writing research (i.e., increased physical health and psychological well-being) generalize to experiences of organizational injustice. Second, we extended previous research by investigating whether the intervention impacts justice-related outcomes, including emotions (i.e., anger), behavioral intentions (i.e., intentions to retaliate), and perceived resolution.

An Overview of the Expressive Writing Intervention Although variations exist, the typical expressive writing intervention involves participants writing about a negative experience for 20 min per day over 4 days (Pennebaker, 1994). Participants are randomly assigned to conditions in which they write about (a) a trivial topic (control condition), (b) only their emotions, (c) only their thoughts, or (d) both their emotions and their thoughts regarding a negative or traumatic experience. In general, individuals who write about both their emotions and their thoughts report better physical and psychological health at the conclusion of the intervention than do individuals in the other conditions (Smyth, 1998). At least three explanations have been advanced to explain these results (see Sloan & Marx, 2004a, for a detailed summary). First, the expressive writing intervention repeatedly exposes individuals to the negative experience and allows them to address the fear or anxiety that can accompany the experience (Lepore, Greenberg, Bruno, & Smyth, 2002; Sloan & Marx, 2004b). Second, the expressive writing intervention can help individuals confront their experience, reducing “inhibition” effects. This explanation proposes that individuals often cope with negative situations by inhibiting or suppressing their thoughts and emotions (Pennebaker, 1989, 1993, 1997b). Although inhibition can help individuals avoid the negative effects of the situation, the act of inhibition consumes physical and mental energy, which can place considerable stress on the body and have damaging psychological and physiological consequences. Psychologically, inhibition can limit individuals’ ability to cognitively process the situation, which not only blocks understanding but is also associated with a tendency for increased rumination, dreams, and thought disturbances (Pennebaker, 1993, 1997b). Suppressing these thoughts and emotions can act as a low-grade stressor that undermines the body’s defenses and increases the risk of illness (e.g., Pennebaker, 1989; Selye, 1976). From this perspective, expressive writing is effective because it allows individuals to actively think about the experience and acknowledge their emotions, which reduces the physiological

and psychological work associated with inhibition and decreases overall stress (Pennebaker, 1997a). Third, the expressive writing intervention allows individuals to vent their emotions. Once emotions are less potent, individuals are better able to cognitively process the situation (e.g., actively think about the experience and engage in sensemaking—a process in which individuals create understanding so that they can act in a principled and informed manner; Weick, 1993, 1995). Sensemaking can help individuals resolve ambiguities about what happened and why (Roberson & Stevens, 2006), better understand their reality by creating order and making retrospective sense (Weick, 1993), and manage uncertainty by creating rational accounts of the world that enable subsequent action (Maitlis, 2005). The creation of a narrative or an account allows individuals to structure and organize the event in their mind, which can help individuals (a) highlight and analyze the central features of the situation rather than focusing on details or irrelevant issues, (b) gain additional insights (e.g., connect the issues to other experiences) and assimilate the event into their broader experiences, and (c) increase understanding and perceived meaning of the experience. Thus, cognitive processing can make the situation seem less overwhelming, more meaningful, and more resolved (Pennebaker, 1997a, 1997b, 1997c).

Relating Expressive Writing to Experiences of Organizational Injustice The expressive writing intervention was initially developed in the context of traumatic life experiences (e.g., Pennebaker, 1993, 1997b; Pennebaker & Beall, 1986) and has been found to have positive benefits for both current (e.g., death of a loved one) and past (e.g., childhood sexual abuse) traumatic experiences (e.g., Sloan & Marx, 2004b). The intervention has also been expanded to other experiences, including physical illnesses such as cancer (Zakowski, Ramati, & Morton, 2004) and stressful life events such as entering college (e.g., Pennebaker, Colder, & Sharp, 1990). Few studies, however, have explored whether the expressive writing intervention can be applied to workplace experiences. A notable exception is a study by Spera, Buhrfeind, and Pennebaker (1994), which found that unemployed professionals who wrote about their emotions and thoughts surrounding job loss were reemployed more quickly than those who wrote about nontraumatic topics or did not write at all. The authors argued that writing allowed participants to work through their negative emotions as well as attain closure regarding the job loss, which reduced the tendency for their negative emotions to resurface in job interviews. This study, however, did not find any significant results for physical health, did not measure psychological well-being or perceived resolution, and only explored one expressive writing condition (i.e., emotions and thoughts). We considered that the expressive writing intervention could be effectively applied to experiences of organizational injustice because of the similarities between many traumatic experiences and organizational justice violations. Specifically, the extant research shows that the expressive writing intervention can be relevant for experiences that (a) impact individuals’ physical health and psychological well-being, (b) involve emotion, and (c) require cognitive processing. Experiences of organizational injustice often contain each of these components. As noted above, experiencing


RESEARCH REPORTS

workplace injustice can impact individuals’ physical and psychological health (Kivima¨ki et al., 2005). Moreover, experiences of workplace unfairness can be highly emotional—they are often described as hot and volatile and are characterized by expressions of pain, anger, and rage (Bies & Tripp, 1996, 2002). Finally, justice theories highlight the importance of cognitive processing, including the role of sensemaking in determining why an injustice occurred and who is to blame (Folger & Cropanzano, 1998, 2001).

Physical and Psychological Outcomes As noted above, research examining the expressive writing intervention has tended to focus on its benefits for physical health and psychological well-being (Smyth, 1998). Writing about emotions and thoughts has been found to be more effective than expressing emotion or thoughts alone because individuals not only vent emotion but also cognitively process the situation (Pennebaker, 1997a, 1997b, 1997c). These two mechanisms in combination facilitate individuals’ understanding of the event, which, in turn, can make the experience seem less overwhelming and can also serve to release the physical and psychological resources that are consumed when individuals inhibit the negative experience. Consistent with this logic, we predicted that writing about emotions and thoughts regarding workplace injustice would be associated with fewer physical symptoms and higher levels of psychological well-being at the conclusion of the intervention compared with the other three conditions. We made this prediction because the emotions and thoughts condition can reduce demands on physical and psychological resources, facilitate the venting of negative emotions, and encourage the assimilation of the event into the individual’s larger experiences (Smyth, 1998). Hypothesis 1: Writing about both emotions and thoughts regarding an unfair work experience is associated with fewer physical symptoms (Hypothesis 1a) and higher psychological well-being (Hypothesis 1b) than writing about only emotions, only thoughts, or topics unrelated to the experience of workplace injustice (i.e., control condition).

Anger We expected that individuals who write about both their emotions and thoughts surrounding an unfair workplace experience would report less anger at the conclusion of the intervention than individuals in the other three conditions. Although individuals in the emotions only condition also have the opportunity to vent their anger, this condition was expected to be less beneficial than when the venting is combined with cognitive processing. When individuals only vent their emotion, it can be analogous to taking the lid off a pressure cooker— once the emotion is released, it becomes less intense but the emotion itself remains and continues to simmer. In contrast, when individuals are given the opportunity to express both emotions and thoughts, they not only vent emotions but the presence of less potent emotions can allow individuals to then cognitively process the situation. This, in turn, increases the likelihood that the individual can view the situation more objectively, gain insights from it, and integrate it into their broader experiences.

513

Hypothesis 2: Writing about both emotions and thoughts regarding an unfair work experience is associated with less anger than writing about only emotions, only thoughts, or topics unrelated to the experience of workplace injustice (i.e., control condition).

Retaliation Previous research on the expressive writing intervention has not considered retaliation intentions as an outcome variable. When the intervention is examined within the context of organizational justice, however, intentions to retaliate can be particularly relevant because retaliation is a common reaction to experiences of injustice (e.g., Bies & Tripp, 2002; Skarlicki & Folger, 1997). It is also important from a victim’s perspective because engaging in these acts can entail personal risk and potential costs, particularly if he or she gets caught (Skarlicki & Folger, 2005). We proposed that writing about emotions and thoughts would be associated with fewer intentions to retaliate at the conclusion of the intervention as compared with the other three conditions. Research has demonstrated that retaliation behaviors can arise from both affective and cognitive antecedents. Allred (1999, 2000), for example, argued that individuals can be motivated to retaliate because they need an outlet for their negative emotions. Other researchers have argued that individuals can be compelled to retaliate to “even the score,” restore a balance of justice, or bring a sense of resolution to the event (Bies & Tripp, 2002). As argued above, the expressive writing intervention can target both affective and cognitive mechanisms, through venting emotion and sensemaking, respectively. Specifically, when individuals release negative emotions in their writing, they are less likely to inhibit their emotion, allow it to escalate, or direct attention toward suppressing it. This, in turn, can reduce individuals’ desire to retaliate because the direct relationship between emotion and behaviors has been alleviated. Engaging in sensemaking, on the other hand, can decrease retaliation tendencies by helping individuals gain a broader perspective, and become more objective, and can allow them to distance themselves from the situation (Pennebaker, 1997b). Moreover, after cognitively processing the situation, the individual may decide that it is not in his or her best interests to retaliate. Taken together, individuals who write about both their emotions and their thoughts were expected to report fewer intentions to retaliate at the conclusion of the intervention than those in the other three conditions because both affective and cognitive mechanisms are activated in this type of writing. Hypothesis 3: Writing about both emotions and thoughts regarding an unfair work experience is associated with fewer retaliation intentions than writing about only emotions, only thoughts, or topics unrelated to the experience of workplace injustice (i.e., control condition).

Perceived Resolution An underlying assumption within the expressive writing intervention is that by providing individuals with the opportunity to vent their emotions and cognitively process the negative experience, individuals can achieve a sense of closure or resolution (Pennebaker, 1997b; Spera et al., 1994). To date, however, few


(—) ⫺.05 (—) .17 ⫺.05 (—) .26ⴱ ⫺.11 ⫺.04 (—) ⫺.03 ⫺.01 ⫺.24ⴱ .24ⴱ (—) ⫺.04 ⫺.02 .29ⴱⴱ .29ⴱⴱ ⫺.11 (—) .46ⴱⴱ ⫺.13 ⫺.04 .30ⴱⴱ .32ⴱⴱ ⫺.25ⴱ Note. Reliabilities are presented along the diagonal within parentheses. pre ⫽ pre-intervention; post ⫽ post-intervention. Gender was dummy coded: 0 ⫽ male; 1 ⫽ female. p ⬍ .05. ⴱⴱ p ⬍ .01 (two-tailed). ⴱ

(.83) .67ⴱⴱ ⫺.19 ⫺.02 .13 ⫺.06 ⫺.19 ⫺.14 .26ⴱⴱ (.95) .09 .49ⴱⴱ ⫺.21ⴱ ⫺.46ⴱⴱ .03 ⫺.02 ⫺.15 ⫺.30ⴱⴱ .26ⴱ (.93) .46ⴱⴱ .42ⴱⴱ .36ⴱⴱ ⫺.30ⴱⴱ ⫺.23ⴱ .25ⴱ ⫺.09 ⫺.20 ⫺.36ⴱⴱ .39ⴱⴱ (.92) .02 ⫺.14 ⫺.17 ⫺.33ⴱⴱ .27ⴱⴱ .33ⴱⴱ ⫺.03 ⫺.15 ⫺.07 .21ⴱ ⫺.10 (.89) .81ⴱⴱ ⫺.03 ⫺.03 ⫺.23ⴱ ⫺.25ⴱ .22ⴱ .25ⴱ ⫺.09 ⫺.10 ⫺.06 .15 ⫺.15 (.89) ⫺.21ⴱ ⫺.22ⴱ .27ⴱⴱ .39ⴱⴱ .10 .24ⴱ ⫺.10 ⫺.21ⴱ .15 ⫺.17 ⫺.12 ⫺.08 .27ⴱⴱ (.83) .71ⴱⴱ ⫺.27ⴱⴱ ⫺.18 .21ⴱ .10 .11 .08 ⫺.07 ⫺.07 .10 ⫺.18 ⫺.01 .01 .36ⴱⴱ 1.05 1.05 1.34 1.34 1.35 1.59 1.42 1.36 1.94 2.06 0.44 6.29 16.37 2.11 1.11

7 6 5 4 3 2 1 SD M

(.91) ⫺.30ⴱⴱ ⫺.29ⴱⴱ .00 .02 ⫺.18 ⫺.22ⴱ .21ⴱ a

Unless otherwise noted, the response set for the measures consisted of a 7-point Likert-type scale, with responses ranging from strongly disagree (1) to strongly agree (7). The items were averaged to form the scale, with greater values signifying higher levels of each measure than lower values. Reliabilities are provided in Table 1.

Variable

Measures

Table 1 Means, Standard Deviations, Reliabilities, and Correlations Among Dependent and Control Variables

Participants were recruited from a large northwestern university in North America. They were paid $40 for their participation. All participants indicated that they had experienced unfair treatment from a previous manager. A total of 101 individuals volunteered to participate. One participant was dropped from the analyses because her schedule changed after starting the study and she was no longer able to complete the sessions. Thus, the final sample consisted of 100 individuals, all of whom completed the entire experimental protocol. Participants were an average of 23 years of age, with 5 years of work experience. Among the sample, 75% were women and 97% were students (of whom 84% were undergraduates, 6% were pursuing a master’s degree, 4% were in a master of business administration program, 2% were unclassified, and 1% was earning a doctorate). A variety of occupations were represented in the sample: 47% worked in the service industry (e.g., meter readers, cashiers), 27% were employed in sales (e.g., call center employees, retail sales), and 16% worked in professional offices (e.g., college instructors, research analysts). On average, participants had been employed at the job where the injustice occurred for 16 months. Participants wrote about a range of unfair experiences (e.g., denial of promotion, being demeaned, being subject to racism, sexual harassment), however, there were no significant differences across the conditions regarding how unfair individuals perceived their treatment. Following procedures by Pennebaker (1994), participants were randomly assigned to one of four conditions (25 participants per condition) in which they wrote about (a) their emotions, (b) their thoughts, (c) both their emotions and their thoughts concerning the unfairness, or (d) a control condition, in which participants wrote about a trivial topic (i.e., objective descriptions of schedules, rooms). A detailed description of the procedure is provided in Appendix A.

8

9

Participants and Procedure

2.90 2.46 4.28 4.88 4.57 3.25 2.87 2.15 3.90 4.37 0.75 22.74 16.25 3.58 2.93

Method

Physical symptoms (pre) Physical symptoms (post) Psychological well-being (pre) Psychological well-being (post) Anger (pre) Anger (post) Retaliation intentions (pre) Retaliation intentions (post) Perceived resolution (pre) Perceived resolution (post) Gendera Age Time lapse (months) Situation resolution Negative affectivity

10

11

12

Hypothesis 4: Writing about both emotions and thoughts regarding an unfair work experience is associated with a greater sense of personal resolution than writing about only emotions, only thoughts, or topics unrelated to the experience of workplace injustice (i.e., control condition).

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

13

14

15

studies have tested whether expressive writing is associated with feelings of resolution. Following the arguments presented above, we expected that writing about emotions and thoughts would help individuals achieve a sense of resolution because it repeatedly exposes them to the experience, allows them to vent emotion and cognitively process their thoughts, and can make the experience seem less overwhelming.

(.84)

RESEARCH REPORTS

514


RESEARCH REPORTS

Dependent variables. Physical symptoms were measured with Pennebaker’s (1982) 13-item physical symptom scale. Responses ranged from 1 (not at all) to 7 (a great deal). Sample items included “headache,” “fatigue,” and “tense muscles.” Psychological well-being was measured with a 5-item Satisfaction With Life Scale (Diener, Emmons, Larsen, & Griffen, 1985). Sample items are as follows: “In most ways my life is close to my ideal” and “The conditions of my life are excellent.” Anger was assessed with a 9-item subscale of the State–Trait Anger Expression Inventory (STAXI) focusing on anger toward the transgressor (Spielberger, 1996). Responses ranged from 1 (not at all) to 7 (extremely). Sample items include “I feel angry” and “I feel aggravated.” Retaliation intentions were measured with McCullough et al.’s (1998) 5-item scale focusing on retaliation toward their previous manager. Sample items include “I’ll make him/her pay,” “I’m going to get even,” and “I want to see him/her hurt and miserable.” Perceived resolution was measured with 1 item developed for this study, “I feel like I have closure.” Control variables. We controlled for gender, age, and time since the unfair experience because previous research shows that these variables can influence the effectiveness of the expressive writing intervention (Smyth, 1998). Within organizational settings, the effectiveness of the intervention might also be influenced by whether the situation was resolved before the intervention began. Thus, we controlled for situation resolution with one item measured pre-intervention, “This situation has been settled.” Finally, research has demonstrated that negative affectivity can influence evaluations of emotions and outcomes (e.g., well-being) as well as inflate relationships between stressors and outcomes (e.g., Brief, Butcher, & Roberson, 1995; Burke, Brief, & George, 1993). Accordingly, we controlled for pre-intervention negative affectivity in all analyses using the negative items of the Positive and Negative Affect Schedule (PANAS) scale (Watson, Clark, & Tellegen, 1988). Responses for this measure ranged from not at all (1) to a great deal (7).

515

cluded that the participants had followed the instructions related to their specific condition. The means, standard deviations, reliabilities, and correlations among the variables are presented in Table 1. Because of the focused and directional nature of the hypotheses, the results were analyzed with planned comparisons, with one-tailed significance levels (cf. Glass & Hopkins, 1996). To test the hypotheses, we used t tests to compare the means among the conditions at the conclusion of the intervention, controlling for the control variables and the baseline (pre-intervention) measure.1 In Hypothesis 1, we predicted that participants in the combined emotions and thoughts condition would report fewer physical symptoms (Hypothesis 1a) and higher psychological well-being (Hypothesis 1b) at the conclusion of the intervention than participants in the other three conditions. As shown in Table 4, no significant differences in physical symptoms were observed. Hypothesis 1a was not supported. Psychological well-being, in contrast, was significantly higher in the emotions and thoughts group than in the three other conditions at the conclusion of the intervention, supporting Hypothesis 1b. In Hypotheses 2, 3, and 4, we predicted that, at the conclusion of the intervention, writing involving both emotions and thoughts would be associated with less anger (Hypothesis 2), fewer intentions to retaliate (Hypothesis 3), and a higher sense of perceived resolution (Hypothesis 4) relative to the other three writing conditions. The results shown in Table 5 indicate that participants in the emotions and thoughts condition reported significantly lower levels of anger than did participants in the emotions only condition, but the levels did not significantly differ from participants in the thoughts only or control conditions. Thus, Hypothesis 2 was only partially supported. Participants in the emotions and thoughts condition reported fewer intentions to retaliate and higher levels of personal resolution than did participants in the other three conditions, supporting Hypotheses 3 and 4.

Discussion Results To determine whether participants adhered to the writing directions, we analyzed the writing samples with Linguistic Inquiry and Word Count (LIWC2007) software (see Appendix B and Pennebaker, Booth, and Francis, 2007, for a detailed explanation of this program). This software program identifies the proportion of words in the writing samples that fall into categories related to emotion (e.g., hate, nervous, sad) and cognition (e.g., because, effect, should). The results, including confidence intervals and effect sizes, are summarized in Tables 2 and 3. Our analysis shows that the participants in the emotions only condition used a significantly higher proportion of words related to negative emotions, anger, anxiety, and sadness than those in the other three conditions. Individuals in the thoughts only condition used a significantly higher proportion of words related to discrepancy (e.g., should, would, could) than participants in the other three conditions. Participants in the thoughts only condition also used a significantly higher proportion of words related to causation (e.g., because, effect, hence) than participants in the emotions only and control conditions but did not significantly differ from participants in the emotions and thoughts condition. We con-

Although organizational justice research has demonstrated that unfair experiences can influence employee behaviors and wellbeing, few studies have examined how victims might manage their own experiences and mitigate the negative consequences associated with experiencing workplace unfairness. We investigated whether the benefits of expressive writing observed in clinical and health psychology research generalize to organizational injustice experiences. We extended previous research by investigating whether expressive writing influences fairness-related outcomes, including anger, retaliation intentions, and perceived resolution. 1 Considerable debate exists regarding the appropriateness of difference scores for measuring effects (see Edwards, 1994, 1995; Tisak & Smith, 1994 for discussion). Difference scores, in which the pre-intervention score is subtracted from the post-intervention score, can result in “fallacious conclusions” because these scores are often subject to poor reliability, ambiguous interpretation, confounding of component measures, and failure to explain variance beyond their components (Cronbach & Furby, 1970, p. 68; Edwards, 1994). Instead of using difference scores, we controlled for the pre-intervention measure in our analyses following procedures recommended by Edwards (1994), as this technique has been empirically demonstrated to prevent and/or reduce the abovementioned issues.


RESEARCH REPORTS

516

Table 2 Manipulation Check: Analysis of Emotion Words by Condition 95% CI

I condition

J condition

M

Mean difference (I ⫺ J)

SE

Lower bound

Upper bound

95% CI Effect size (d)

M

Mean difference (I ⫺ J)

Negative emotionsa (e.g., worthless) Emotions only Thoughts only Emotions and thoughts Control Thoughts only Emotions only Emotions and thoughts Control Emotions and thoughts Emotions only Thoughts only Control Control Emotions only Thoughts only Emotions and thoughts Emotions only Thoughts only Emotions and thoughts Control Thoughts only Emotions only Emotions and thoughts Control Emotions and thoughts Emotions only Thoughts only Control Control Emotions only Thoughts only Emotions and thoughts a ⴱ

4.61 2.29 3.07 0.52 2.29 4.61 3.07 0.52 3.07 4.61 2.29 0.52 0.52 4.61 2.29 3.07 1.16 0.34 0.57 0.12 0.34 1.16 0.57 0.12 0.57 1.16 0.34 0.12 0.12 1.16 0.34 0.57

Partial eta squared for condition ␩2p ⫽ .68; p ⬍ .05 (two-tailed).

b

.29 .29 .29

1.74 0.96 3.51

2.90 2.12 4.67

1.70 1.20 3.45

⫺2.32ⴱ ⫺0.78ⴱ 1.77ⴱ

.29 .29 .29

⫺2.90 ⫺1.36 1.19

⫺1.74 ⫺0.20 2.35

⫺1.70 ⫺0.91 2.49

⫺1.54ⴱ 0.78ⴱ 2.55ⴱ

.29 .29 .29

⫺2.12 0.20 1.97

⫺0.96 1.36 3.13

⫺1.20 0.91 4.79

⫺4.09ⴱ .29 ⫺1.77ⴱ .29 ⫺2.55ⴱ .29 Anxietyc

⫺4.67 ⫺2.35 ⫺3.13

⫺3.51 ⫺1.19 ⫺1.97

⫺3.45 ⫺2.49 ⫺4.79

1.49 0.63 1.10 0.11 0.63 1.49 1.10 0.11

0.58 0.36 0.81

1.05 0.82 1.28

1.48 1.05 1.95

⫺0.82ⴱ ⫺0.22 0.23

.12 .12 .12

⫺1.05 ⫺0.46 ⫺0.01

⫺0.58 0.01 0.46

⫺1.48 ⫺0.93 1.35

⫺0.59ⴱ 0.22 0.45ⴱ

.12 .12 .12

⫺0.82 ⫺0.01 0.22

⫺0.36 0.46 0.68

⫺1.05 0.93 2.28

⫺1.04ⴱ ⫺0.23 ⫺0.45ⴱ

.12 .12 .12

⫺1.28 ⫺0.46 ⫺0.68

⫺0.81 0.01 ⫺0.22

⫺1.95 ⫺1.35 ⫺2.28

Mixed results were found regarding the physical and psychological benefits of the expressive writing intervention. No significant differences in physical symptoms were found between conditions at the conclusion of the intervention. At least two reasons might help explain these findings. First, previous research has demonstrated that expressive writing has a stronger relationship with physical health for individuals with a pre-existing physical health issue (e.g., cancer, rheumatoid arthritis) and a weak relationship or no relationship for generally healthy individuals (Meads & Nouwen, 2005). Thus, it is possible that health improvements are most likely to occur for individuals who suffer health ailments, which was not the case for our sample. Second, our physical symptoms measure might not have been sufficiently sensitive to detect differences. Previous studies conducted with the same measure have shown that ratings of this measure tend to exhibit relatively little decline and can even increase over the course of time (Greenberg & Stone, 1992).

d

Effect size (d)

1.10 1.49 0.63 0.11 0.11 1.49 0.63 1.10

.16 .16 .16

0.55 0.08 1.06

1.19 0.72 1.70

1.18 0.52 2.06

⫺0.87ⴱ ⫺0.47ⴱ 0.52ⴱ

.16 .16 .16

⫺1.19 ⫺0.79 0.20

⫺0.55 ⫺0.15 0.84

⫺1.18 ⫺1.04 1.70

⫺0.40ⴱ 0.47ⴱ 0.99ⴱ

.16 .16 .16

⫺0.72 0.15 0.67

⫺0.08 0.79 1.31

⫺0.52 1.04 2.86

⫺1.38ⴱ ⫺0.52ⴱ ⫺0.99ⴱ

.16 .16 .16

⫺1.70 ⫺0.84 ⫺1.30

⫺1.06 ⫺0.20 ⫺0.67

⫺2.06 ⫺1.70 ⫺2.86

Sadnessd (e.g., sad)

.12 .12 .12

␩2p ⫽ .48;

Upper bound

0.87ⴱ 0.40ⴱ 1.38ⴱ

(e.g., afraid)

0.82ⴱ 0.59ⴱ 1.04ⴱ

c

Lower bound

Angerb (e.g., hate)

2.32ⴱ 1.54ⴱ 4.09ⴱ

␩2p ⫽ .46;

SE

0.95 0.33 0.56 0.13 0.33 0.95 0.56 0.13 0.56 0.95 0.33 0.13 0.13 0.95 0.33 0.56

0.63ⴱ 0.39ⴱ 0.82ⴱ

.12 .12 .12

0.40 0.17 0.60

0.86 0.62 1.05

1.23 0.75 1.82

⫺0.63ⴱ ⫺0.23ⴱ 0.20

.12 .12 .12

⫺0.86 ⫺0.46 ⫺0.03

⫺0.40 ⫺0.01 0.43

⫺1.23 ⫺0.65 0.83

⫺0.39ⴱ 0.23ⴱ 0.43ⴱ

.12 .12 .12

⫺0.62 0.01 0.20

⫺0.17 0.46 0.66

⫺0.75 0.65 1.56

⫺0.82ⴱ ⫺0.20 ⫺0.43ⴱ

.12 .12 .12

⫺1.05 ⫺0.43 ⫺0.66

⫺0.60 0.30 ⫺0.20

⫺1.82 ⫺0.83 ⫺1.56

␩2p ⫽ .37.

Results indicated that individuals who wrote about their emotions and thoughts reported significantly higher psychological well-being at the conclusion of the intervention relative to the emotions only (d ⫽ .69), thoughts only (d ⫽ .79), and control (d ⫽ .81) conditions. According to Cohen (1988), effect sizes (ds) of .20, .50, and .80 represent small, medium, and large effect sizes, respectively. Thus, our results show that the intervention had medium to large effects on psychological well-being. These findings are significant because this is the first study to empirically demonstrate the benefits of the expressive writing intervention for experiences of workplace injustice. Second, previous research has been criticized for using indirect and/or poorly validated measures of well-being (Meads & Nouwen, 2005; Smyth, 1998). In the present study, we used a well-established and validated measure of psychological well-being. Third, although the importance of emotions and sensemaking as a reaction to injustice is highlighted in justice theories (e.g., Folger & Cropanzano, 2001), our study


RESEARCH REPORTS

517

Table 3 Manipulation Check: Analysis of Cognition Words by Condition Discrepancya (e.g., should)

Causationb (e.g., because)

95% CI

I condition

J condition

M

Thoughts only Emotions only Emotions and thoughts Control Emotions only Thoughts only Emotions and thoughts Control Emotions and thoughts Emotions only Thoughts only Control Control Emotions only Thoughts only Emotions and thoughts a ⴱ

2.74 2.11 2.34 0.63 2.11 2.74 2.34 0.63 2.34 2.11 2.74 0.63 0.63 2.11 2.74 2.34

Partial eta squared for condition ␩2p ⫽ .56; p ⬍ .05 (two-tailed).

b

Mean difference (I ⫺ J)

95% CI

SE

Lower bound

Upper bound

Effect size (d)

0.63ⴱ 0.40ⴱ 2.11ⴱ

.20 .20 .20

0.23 0.01 1.71

1.04 0.81 2.52

0.76 0.46 3.00

⫺0.63ⴱ ⫺0.23 1.48ⴱ

.20 .20 .20

⫺1.04 ⫺0.63 1.07

⫺0.23 0.17 1.88

⫺0.76 ⫺0.31 2.78

0.23 ⫺0.41ⴱ 1.71

.20 .20 .20

⫺0.17 ⫺0.81 1.30

0.63 ⫺0.01 2.11

0.31 ⫺0.46 2.93

⫺1.48ⴱ ⫺2.11ⴱ ⫺1.71ⴱ

.20 .20 .20

⫺1.88 ⫺2.52 ⫺2.11

⫺1.07 ⫺1.71 ⫺1.30

⫺2.78 ⫺3.00 ⫺2.93

M 2.47 2.06 2.30 1.13 2.06 2.47 2.30 1.13 2.30 2.06 2.47 1.13 1.13 2.06 2.47 2.30

Mean difference (I ⫺ J)

SE

Lower bound

Upper bound

Effect size (d)

0.41ⴱ 0.17 1.34ⴱ

.16 .16 .16

0.09 ⫺0.15 1.02

0.73 0.48 1.65

0.63 0.26 2.45

⫺0.41ⴱ ⫺0.24 0.93ⴱ

.16 .16 .16

⫺0.73 ⫺0.56 0.61

⫺0.09 0.08 1.25

⫺0.63 ⫺0.42 2.06

0.24 ⫺0.17 1.17ⴱ

.16 .16 .16

⫺0.08 ⫺0.48 0.85

0.56 0.15 1.48

0.42 ⫺0.26 2.64

⫺0.93ⴱ ⫺1.34ⴱ ⫺1.17ⴱ

.16 .16 .16

⫺1.25 ⫺1.65 ⫺1.48

⫺0.61 ⫺1.02 ⫺0.85

⫺2.06 ⫺2.45 ⫺2.64

␩2p ⫽ .47.

provides empirical evidence that releasing negative emotion and engaging in sensemaking can help mitigate the negative effects of experiencing injustice on employee well-being. We also investigated whether the expressive writing intervention influences anger, retaliation intentions, or perceived resolution. Participants who wrote about their emotions and thoughts reported significantly less anger at the conclusion of the intervention relative to individuals in the emotions only condition (d ⫽ ⫺.51) but did not differ significantly from participants in the thoughts only condition (d ⫽ ⫺.05), and small effects were observed for the control condition (d ⫽ ⫺.25). Interestingly, at the

end of the intervention, individuals in the emotions only condition reported the highest level of anger among all of the conditions. Considerable debate exists in the literature regarding whether venting emotion, and in particular anger, is beneficial for decreasing negative emotions. Some writers argue that catharsis, defined as the relieving of tension by expressing emotions, can reduce anger (Freud, 1930/1979), whereas others suggest that expressing emotion can actually increase or exaggerate negative emotions (Bushman, Baumeister, & Stack, 1999; Geen & Quanty, 1977). Our results support the latter perspective—venting emotion on its own has fewer benefits in terms of reducing anger than when

Table 4 Planned Comparisons Between Conditions at the Conclusion of the Intervention: Physical Symptoms and Psychological Well-Being Physical symptoms

Psychological well-being

CIa

I condition

J condition

M

Mean difference (I ⫺ J)

Emotions only Thoughts only Control

2.31 2.60 2.57 2.51

⫺0.29 ⫺0.26 ⫺0.20

Emotions and thoughts

CIa

SE

Lower bound

Upper bound

Effect size (d)

.24 .24 .24

⫺⬁ ⫺⬁ ⫺⬁

0.11 0.14 0.21

⫺0.36 ⫺0.33 ⫺0.24

M

Mean difference (I ⫺ J)

SE

Lower bound

Upper bound

Effect size (d)

5.39 4.84 4.76 4.74

0.55ⴱⴱ 0.63ⴱⴱ 0.65ⴱⴱ

.23 .24 .24

0.17 0.24 0.25

⬁ ⬁ ⬁

0.69 0.79 0.81

Note. Means and standard errors were obtained after controlling for the pre-intervention/baseline measure and the control variables (i.e., gender, age, time lapse since unfairness, situation resolution, and negative affectivity). a A 90% unidirectional confidence interval (CI) is provided to reflect the one-tailed, directional nature of the analyses (cf. Shavelson, 1988). ⴱⴱ p ⬍ .01 (one-tailed).


RESEARCH REPORTS

518

Table 5 Planned Comparisons Between Conditions at the Conclusion of the Intervention: Anger, Retaliation Intentions, and Perceived Resolution CIa

I condition

J condition

M

Mean difference (I ⫺ J)

Lower bound

SE

CIa Upper bound

Effect size (d)

M

Mean difference (I ⫺ J)

Anger Emotions and thoughts Emotions only Thoughts only Control

2.94 3.68 3.01 3.31

⫺0.74ⴱ ⫺0.07 ⫺0.37

Emotions only Thoughts only Control

5.11 4.21 3.96 4.19

0.90ⴱ 1.15ⴱ 0.92ⴱ

Emotions and thoughts

.43 .43 .45 Perceived

.51 .52 .53

Lower bound

Upper bound

Effect size (d)

⫺0.10 ⫺0.05 ⫺0.01

⫺0.59 ⫺0.55 ⫺0.50

Retaliation intentions

⫺⬁ ⫺0.02 ⫺⬁ 0.65 ⫺⬁ 0.38 resolution

0.06 0.28 0.04

SE

⬁ ⬁ ⬁

⫺0.51 ⫺0.05 ⫺0.25

1.71 2.29 2.25 2.22

⫺0.58ⴱ ⫺0.54ⴱ ⫺0.51ⴱ

.29 .30 .30

⫺⬁ ⫺⬁ ⫺⬁

0.52 0.67 0.53

Note. Means and standard errors were obtained after controlling for the pre-intervention/baseline measure and the control variables (i.e., gender, age, time lapse since unfairness, situation resolution, and negative affectivity). a A 90% unidirectional confidence interval (CI) is provided to reflect the one-tailed, directional nature of the analyses (cf. Shavelson, 1988). ⴱ p ⬍ .05 (one-tailed).

emotional release is combined with cognitive processing. Our results are also consistent with the perspective that the process of venting emotions facilitates individuals’ ability to engage in sensemaking, which in turn allows individuals to attach meaning to the experience, gain insights, and put the experience behind them. Expressive writing also had benefits in terms of fewer intentions to retaliate. Individuals who wrote about both their emotions and their thoughts reported lower intentions to retaliate than did participants in the emotions only (d ⫽ ⫺.59), thoughts only (d ⫽ ⫺.55), or control (d ⫽ ⫺.50) conditions. Victims can be motivated to retaliate for a variety of reasons, including the belief that retaliation will “balance the scales,” reestablish order, and help manage negative emotions (Allred, 1999, 2000; Skarlicki & Folger, 2005). Although retaliation can be functional in some cases (e.g., to prevent future violations; Bies & Tripp, 2002), it can come at a cost for the individual and the organization. To date, research has focused on how retaliation can be diminished through actions taken by the organization or manager. Our findings suggest that intentions to retaliate can also be diminished by victim-centered interventions. Although an underlying theme in the expressive writing literature is that writing facilitates the development of a sense of closure or personal resolution with the experience (e.g., Spera et al., 1994), few studies have empirically tested this proposition. Our results show that participants who wrote about both their emotions and thoughts reported significantly higher levels of perceived resolution at the conclusion of the intervention than did participants in the emotions only (d ⫽ .52), thoughts only (d ⫽ .67), or control (d ⫽ .53) condition. This is important because previous research on recovering from organizational injustice has deemed recovery as a response to others’ attempts to remedy the situation. That is, recovery attempts do not generally actively engage victims in the recovery process. For instance, individuals might receive monetary

compensation for the violation, but the remedy is aimed at mitigating damages rather than helping the victim recover, and it is often unclear how compensation can help victims manage their personal sense of violation (Darley & Pittman, 2003). Our findings suggest that victims can take their own steps to manage the effects of experiencing workplace unfairness. Taken together, following Cohen’s (1988) standards, the expressive writing intervention, and in particular, the combined emotions and thoughts condition, had small effects for anger, medium effects for retaliation intentions and sense of resolution, and large effects for psychological well-being. These results should be viewed in light of several potential limitations. First, although our sample size was sufficient to find significant results among most of our hypotheses, a larger sample might have resulted in more hypotheses being supported. Second, although our participants came from a variety of backgrounds, our sample consisted of individuals attending a university. Although an examination of the participants’ essays showed that the violations were nontrivial and spanned a range of justice issues, future research needs to explore whether the results generalize beyond individuals in higher education. Third, our measures of resolution (situation and perceived) were assessed with one-item scales. We deemed single-item scales to be appropriate in the current context because of concerns that lengthy scales could influence the results. Research has found that thinking or responding to questions related to issues that have positive or negative valence can polarize individuals’ perceptions and attitudes, even if they have no new information (Tesser, 1978). Our approach reduced the likelihood that perceived resolution was affected by the design of our questionnaire. Fourth, we asked participants to write about their general experiences of unfairness. Post-intervention interviews (see Appendix A) indicated that for the majority (79%) of participants, the event triggered multiple dimensions of unfairness. Moreover, participants who only


RESEARCH REPORTS

identified a violation on one dimension in the interview often wrote about multiple justice dimensions. Denials of promotions (typically a distributive violation), for example, were also interpreted as reflecting a lack of respect (interpersonal violation) and voice (procedural violation). As a result, we were unable to untangle whether the intervention was more effective for some aspects of justice than others. Our open-ended approach is consistent with Shapiro’s (2001) contention that victims are more likely to be concerned about how the experience as a whole impacts them rather than how each dimension affects their experience. Future research needs to explore whether the intervention is more effective for some aspects of justice than for others. Our findings also have an important research implication. Roberson and Stevens (2006), for instance, had their participants write out critical incidents pertaining to workplace diversity and called for additional studies utilizing methodologies that capitalize on open-ended accounts of fairness-related experiences. We speculate that although it was not the authors’ intention, this research strategy could, in some instances, serve as an intervention in itself. This possibility warrants future study. Participants in our sample focused on fairness violations that had occurred in the past; the average time lapse since the unfair experience was 16 months. Although time lapse has not been associated with differences in overall effect size, individuals who write about current traumas tend to report significantly higher psychological well-being than participants who write about past traumas (Smyth, 1998). No studies have examined whether there is an optimal time for the administration of the intervention. Pennebaker and Chung (2007), however, argued that It is likely that defenses such as denial, detachment, distraction, and distancing may, in fact, be quite healthy in the hours and days after an upheaval. A technique such as expressive writing may be inappropriate until several weeks or months later.” (p. 270)

Future research should examine when (i.e., how soon after the unfair experience) the intervention is most effective. Finally, although laboratory studies provide control of extraneous variables, they can be limited by problems associated with demand characteristics (Gordon, Slade, & Schmitt, 1986, 1987). Thus, it is important to investigate whether the results generalize to field settings. One practical implication of the present study is that expressive writing interventions can help individuals who have experienced workplace injustice manage their reactions. Employee assistance professionals, coworkers, family, and friends hear about unfair treatment all of the time (DeGoey, 2000), but they might not know how to help the victim with his or her struggle. One option is to encourage victims of workplace injustice to write a private journal to help them emotionally and cognitively work through the experience. As noted above, however, the expressive writing intervention might be more appropriate for addressing the residual emotions and thoughts that occur in the weeks and months following the incident rather than following the initial upheaval (Pennebaker & Chung, 2007). We also note that the expressive writing intervention is a highly personal exercise that is typically done in a private manner. This raises the question of when and how the intervention should be used in organizations, as it can entail practical and ethical implications in these settings. If individuals believe that their managers might have access to their writing, they might be less willing to

519

engage in the very type of writing that has been found to be most effective in this paradigm. Other than this, the intervention might be effectively applied in a host of contexts both inside and outside the organization. Within the organization, the intervention might be useful when companies hire employees who have experienced unfairness at their previous place of employment or when employees change departments or work teams after experiencing an injustice. Outside agencies, such as outplacement services, might also find the intervention helpful, particularly with clients who feel that that they have been unfairly laid off. These examples represent only some of the possibilities. Future research should explore when and who can most effectively use the intervention in organizations. In summary, our study illustrates the potential benefits associated with providing victims of organizational injustice with the opportunity to take an active role in the recovery process. Importantly, this intervention is not intended as a first line of defense when an unfair situation occurs, nor do its beneficial effects absolve the organization or managers of their responsibilities to treat employees fairly in the first place.

References Allred, K. G. (1999). Anger and retaliation: Toward an understanding of impassioned conflict in organizations. In R. J. Bies, R. J. Lewicki, & B. H. Sheppard (Eds.), Research on negotiations in organization (Vol. 7, pp. 27–58). Greenwich: JAI Press. Allred, K. G. (2000). Anger and retaliation in conflict: The role of attribution. In M. Deutsch & P. T. Coleman (Eds.), The handbook of conflict resolution: Theory and practice (pp. 236 –255). San Francisco: Jossey-Bass. Bateson, G. (1977). Afterword. In J. Brockman (Ed.), About Bateson (pp. 235–247). New York: Dutton. Bies, R. J., & Tripp, T. M. (1996). Beyond distrust: “Getting even” and the need for revenge. In R. M. Kramer & T. R. Tyler (Eds.), Trust in organizations: Frontiers of theory and research (pp. 246 –260). Thousand Oaks: Sage. Bies, R. J., & Tripp, T. M. (2002). “Hot flashes and open wounds”: Injustice and the tyranny of its emotions. In S. W. Gilliland, D. D. Steiner, & D. P. Skarlicki (Eds.), Emerging perspectives on managing organizational justice (pp. 203–221). Greenwich, CT: Information Age. Brief, A. P., Butcher, A. H., & Roberson, L. (1995). Cookies, disposition, and job attitudes: The effects of positive mood-inducing events and negative affectivity on job satisfaction in a field experiment. Organizational Behavior and Human Decision Processes, 62, 55– 62. Burke, M. J., Brief, A. P., & George, J. M. (1993). The role of negative affectivity in understanding relations between self-reports of stressors and strains: A comment on the applied psychology literature. Journal of Applied Psychology, 78, 402– 412. Bushman, B. J., Baumeister, R. F., & Stack, A. D. (1999). Catharsis, aggression, and persuasive influence: Self-fulfilling or self-defeating prophecies? Journal of Personality and Social Psychology, 76, 367–376. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum.0Cronbach, L. J., & Furby, L. (1970). How we should measure “change”—Or should we? Psychological Bulletin, 74, 68 – 80. Darley, J. M., & Pittman, T. S. (2003). The psychology of compensatory and retributive justice. Personality and Social Psychology Review, 7, 324 –336. DeGoey, P. (2000). Contagious justice: Exploring the social construction of justice in organizations. Research in Organizational Behavior, 22, 51–102. Diener, E., Emmons, R. A., Larsen, R. J., & Griffen, S. (1985). The Satisfaction With Life Scale. Journal of Personality Assessment, 49, 71–75.


520

RESEARCH REPORTS

Dirks, K. T., Lewicki, R. J., & Zaheer, A. (in press). Repairing relationships within and between organizations: Building a conceptual foundation. Academy of Management Review. Edwards, J. R. (1994). Regression analysis as an alternative to difference scores. Journal of Management, 20, 683– 689. Edwards, J. R. (1995). Alternatives to difference scores as dependent variables in the study of congruence in organizational research. Organizational Behavior and Human Decision Processes, 64, 307–324. Elovainio, M., Kivima¨ki, M., & Helkama, K. (2001). Organizational justice evaluations, job control, and occupational strain. Journal of Applied Psychology, 86, 418 – 424. Elovainio, M., Kivima¨ki, M., & Vahtera, J. (2002). Organizational justice: Evidence of a new psychosocial predictor of health. American Journal of Public Health, 92, 105–108. Folger, R., & Cropanzano, R. (1998). Organizational justice and human resource management. Thousand Oaks, CA: Sage. Folger, R., & Cropanzano, R. (2001). Fairness theory: Justice as accountability. In J. Greenberg & R. Cropanzano (Eds.), Advances in organizational justice (pp. 1–55). Stanford, CA: Stanford University Press. Freud, S. (1979). Civilization and its discontents (J. Riviere, Trans.). London: Hogarth Press. (Original work published 1930). Geen, R. G., & Quanty, M. B. (1977). The catharsis of aggression: An evaluation of a hypothesis. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 10, pp. 1–37). New York: Academic Press. Glass, G. V., & Hopkins, K. D. (1996). Statistical methods in education and psychology (3rd ed.). Boston: Allyn & Bacon. Gordon, M. E., Slade, L. A., & Schmitt, N. (1986). The “science of the sophomore” revisited: From conjecture to empiricism. Academy of Management Review, 11, 191–207. Gordon, M. E., Slade, L. A., & Schmitt, N. (1987). Student guinea pigs: Porcine predictors and particularistic phenomena. Academy of Management Review, 12, 160 –163. Graybeal, A., Sexton, J. D., & Pennebaker, J. W. (2002). The role of story-making in disclosure writing: The psychometrics of narrative. Psychology and Health, 17, 571–581. Greenberg, J. (2006). Losing sleep over organizational injustice: Attenuating insomniac reactions to underpayment inequity with supervisory training in interactional justice. Journal of Applied Psychology, 91, 58 – 69. Greenberg, M. A., & Stone, A. A. (1992). Emotional disclosure about traumas and its relation to health: Effects of previous disclosure and trauma severity. Journal of Personality and Social Psychology, 63, 75– 84. Hansson, R. O., Jones, W. H., & Fletcher, W. L. (1990). Troubled relationships in later life: Implications for social support. Journal of Social and Personal Relationships, 7, 451– 463. Harlos, K. P., & Pinder, C. C. (2000). Emotions and injustice in the workplace. In S. Fineman (Ed.), Emotion in organizations (Vol. 2, pp. 255–276). London: Sage. Kim, P. H., Ferrin, D. L., Cooper, C. D., & Dirks, K. T. (2004). Removing the shadow of suspicion: The effects of apology versus denial for repairing competence- versus integrity-based trust violations. Journal of Applied Psychology, 89, 104 –118. Kivima¨ki, M., Elovainio, M., Vahtera, J., & Ferrie, J. E. (2003). Organizational justice and health of employees: Prospective cohort study. Occupational Environmental Medicine, 60, 27–34. Kivima¨ki, M., Elovainio, M., Vahtera, J., Virtanen, M., & Stansfeld, S. A. (2003). Association between organizational inequity and incidence of psychiatric disorders in female employees. Psychological Medicine, 33, 319–326. Kivima¨ki, M., Ferrie, J. E., Brunner, E., Head, J., Shipley, M. J., Vahtera, J., & Marmot, M. G. (2005). Justice at work and reduced risk of coronary heart disease among employees: The Whitehall II study. Archives of Internal Medicine, 165, 2245–2251. Kramer, R. M., & Neale, M. A. (1998). Power and influence in organizations. Thousand Oaks, CA: Sage. Lepore, S. J., Greenberg, M. A., Bruno, M., & Smyth, J. M. (2002).

Expressive writing and health: Self-regulation of emotion-related experience, physiology, and behavior. In S. J. Lepore & J. M. Smyth (Eds.), The writing cure: How expressive writing promotes health and emotional wellbeing (pp. 99 –117). Washington, DC: American Psychological Press. Maitlis, S. (2005). The social processes of organizational sensemaking. Academy of Management Journal, 48, 21– 49. Matthews, C. (1988). Hardball: How politics is played—Told by one who knows the game. New York: Summit Books. McCullough, M. E., Rachal, K. C., Sandage, S. J., Worthington, E. L., Jr., Brown, S. W., & Hight, T. L. (1998). Interpersonal forgiving in close relationships: II. Theoretical elaboration and measurement. Journal of Personality and Social Psychology, 75, 1586 –1603. Meads, C., & Nouwen, A. (2005). Does emotional disclosure have any effects? A systematic review of the literature with meta-analyses. International Journal of Technology Assessment in Health Care, 21, 153–164. Mikula, G. (1986). The experience of injustice: Toward a better understanding of its phenomenology. In H. W. Bierhoff, R. L. Cohen, & J. Greenberg (Eds.), Justice in interpersonal relations (pp. 103–123). New York: Plenum Press. Pennebaker, J. W. (1982). The psychology of physical symptoms. New York: Springer-Verlag. Pennebaker, J. W. (1989). Confession, inhibition, and disease. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 22, pp. 211–244). New York: Academic Press. Pennebaker, J. W. (1993). Putting stress into words: Health, linguistic, and therapeutic implications. Behavior Research and Therapy, 31, 539 –548. Pennebaker, J. W. (1994). Hints on running a writing experiment. Unpublished manual. Retrieved January 26, 2004, from http://homepage.psy .utexas.edu/homepage/faculty/pennebaker/reprints/ Pennebaker, J. W. (1997a). Confession, inhibition and disease. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 22, pp. 211–244). New York: Academic Press. Pennebaker, J. W. (1997b). Opening up: The healing power of expressing emotions. New York: Guilford Press. Pennebaker, J. W. (1997c). Writing about emotional experiences as a therapeutic process. Psychological Science, 8, 162–166. Pennebaker, J. W., & Beall, S. K. (1986). Confronting a traumatic event: Toward an understanding of inhibition and disease. Journal of Abnormal Psychology, 95, 274 –281. Pennebaker, J. W., Booth, R. J., & Francis, M. E. (2007). LIWC2007: Linguistic inquiry and word count. Austin, TX: liwc.net. Pennebaker, J. W., & Chung, C. K. (2007). Expressive writing, emotional upheavals, and health. In H. Friedman & R. Silver (Eds.), Handbook of health psychology (pp. 263–284). New York: Oxford University Press. Pennebaker, J. W., Chung, C. K., Ireland, M., Gonzales, A., & Booth, R. J. (2007). The LICW 2007 dictionary. Austin, TX: licw.net. Pennebaker, J. W., Colder, M., & Sharp, L. K. (1990). Accelerating the coping process. Journal of Personality and Social Psychology, 58, 528 –537. Pennebaker, J. W., Francis, M. E., & Booth, R. J. (2001). Linguistic inquiry and word count. Mahwah, NJ: Erlbaum. Reb, J., Goldman, B. M., Kray, L. J., & Cropanzano, R. (2006). Different wrongs, different remedies? Reactions to organizational remedies after procedural and interactional injustice. Personnel Psychology, 59, 31– 64. Roberson, Q. M., & Stevens, C. K. (2006). Making sense of diversity in the workplace: Organizational justice and language abstraction in employees’ accounts of diversity related incidents. Journal of Applied Psychology, 91, 379 –391. Selye, H. (1976). The stress of life. New York: McGraw-Hill. Shapiro, D. L. (2001). The death of justice theory is likely if theorists neglect the “wheels” already invented and the voices of the injustice victims. Journal of Vocational Behavior, 58, 235–242. Shavelson, R. J. (1988). Statistical reasoning for the behavioral sciences (2nd ed.). Boston: Allyn & Bacon. Shaw, J. C., Wild, E., & Colquitt, J. A. (2003). To justify or excuse? A


RESEARCH REPORTS meta-analytic review of the effects of explanations. Journal of Applied Psychology, 88, 444 – 458. Skarlicki, D. P., & Folger, R. (1997). Retaliation in the workplace: The roles of distributive, procedural, and interactional justice. Journal of Applied Psychology, 82, 434 – 443. Skarlicki, D. P., & Folger, R. (2005). Broadening our understanding of organizational retaliatory behavior. In R. W. Griffin & A. M. O’LearyKelly (Eds.), The dark side of organizational behavior (pp. 373– 402). San Francisco, CA: Jossey-Bass. Skarlicki, D. P., & Latham, G. P. (1996). Increasing citizenship behavior within a labor union: A test of organizational justice theory. Journal of Applied Psychology, 81, 161–169. Sloan, D. M., & Marx, B. P. (2004a). A closer examination of the written disclosure paradigm. Journal of Consulting and Clinical Psychology, 72, 165–175. Sloan, D. M., & Marx, B. P. (2004b). Taking pen to hand: Evaluating theories underlying the written disclosure paradigm. Clinical Psychology: Science and Practice, 11, 121–137. Smyth, J. M. (1998). Written emotional expression: Effect sizes, outcome types, and moderating variables. Journal of Consulting and Clinical Psychology, 66, 174 –184. Spera, S. P., Buhrfeind, E. D., & Pennebaker, J. W. (1994). Expressive writing and coping with job loss. Academy of Management Journal, 37, 722–733.

521

Spielberger, C. D. (1996). State–Trait Anger Expression Inventory. Tampa, FL: Psychological Assessment Resources. Tepper, B. J. (2001). Health consequences of organizational injustice: Tests of main and interactive effects. Organizational Behavior and Human Decision Processes, 86, 197–215. Tesser, A. (1978). Self-generated attitude change. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 11, pp. 290 –339). New York: Academic Press. Tisak, J., & Smith, C. S. (1994). Defending and extending difference score methods. Journal of Management, 20, 675– 682. Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54, 1063–1070. Weick, K. E. (1993). The collapse of sensemaking in organizations: The Mann Gulch disaster. Administrative Science Quarterly, 38, 628 – 652. Weick, K. E. (1995). Sensemaking in organizations. Thousand Oaks, CA: Sage. Weiss, H. M., Suckow, K., & Cropanzano, R. (1999). Effects of justice conditions on discrete emotions. Journal of Applied Psychology, 84, 786 –794. Zakowski, S. G., Ramati, A., & Morton, C. (2004). Written emotional disclosure buffers the effects of social constraints on distress among caner patients. Health Psychology, 23, 555–556.

Appendix A Experimental Protocol Our experimental protocol was based on previous research on the expressive writing intervention (e.g., Graybeal, Sexton, & Pennebaker, 2002) and procedures recommended by Pennebaker (1994). Participants attended five laboratory sessions, once per day, on consecutive days (Monday through Friday). Participants were tested individually to ensure an environment in which they would feel safe writing about their experience.

Confidentiality Participants were informed that information provided would be kept confidential and were assigned a numeric identifier that was used for all of their data. The numeric identifier was kept separate from identifying information, and only the lead author had access to the master list. Participants were informed that confidentiality could be broken in exceptional cases, such as a suicidal threat or threat against another individual. They were assured that in no other instances would confidentiality be broken. Participants were also provided with contact information for the lead author as well as counseling services in case they became distressed. No participant required these services as a result of their participation in the study.

Procedures At the first session, participants completed a questionnaire that included the preintervention measures of the dependent variables

(e.g., physical symptoms, psychological well-being, anger associated with the unfairness, retaliation intentions, and perceived resolution) as well as demographic information. Before completing the questionnaire, participants were asked to select one unfair workplace experience that they had experienced from a supervisor and to use this experience for the entire study. In order to prevent inflation of ratings, we did not ask participants to reflect on this experience before completing the questionnaire nor did we ask them to describe the experience (see rationale in the Debriefing section). They were simply given the questionnaire to complete. All participants were given the same questionnaire regardless of their condition. Thus, participants in the control condition responded to questions about the unfair incident (i.e., anger, retaliation intentions, perceived resolution), but their writing was not connected to the unfair experience (see instructions below). Participants were asked to write for 20 min about the same unfair experience during sessions 2–5 (20 min/4 writing sessions for a total of 80 min of writing). In the fifth/final session, participants wrote about their assigned topic for 20 min and then completed postintervention measures (e.g., physical symptoms, psychological well-being, anger, retaliation intentions, and perceived resolution). Standard instructions from previous research were used (e.g., Graybeal et al., 2002; Pennebaker, 1994); however, they were altered to fit an organizational justice context. Specifically,

(Appendixes continue)


RESEARCH REPORTS

522

whereas in previous research, individuals were asked to write about “an extremely emotional issue,” participants in the present study were instructed to write about an unfair workplace experience. As noted above, participants were informed that they should use the same unfair workplace experience throughout the entire study. The same instructions were provided for every writing session (with the exception of the control condition; see below). The instructions were as follows.

Control Condition Consistent with previous research, participants were instructed to write about four different topics. The instructions were the same except for the italicized topic. In this session, I would like you to write about how you managed your time for the last 24 hours. Do not explore your emotions or feelings; please try to be completely objective and descriptive. Go into as much detail as possible. All of your writing will be completely confidential. Don’t worry about spelling, sentence structure, or grammar. The only rule is that once you begin writing, you continue to do so until 20 minutes is up.

Emotions Only Condition In this session, I would like you to write about your emotions and feelings surrounding an unfair workplace experience that has affected you and your life. In your writing, I’d like you to explore your deepest emotions. It is important that you do not explore your thoughts (i.e., “I think that . . .”; “I believe that . . .”). Please write only about what you feel about the situation (i.e., “I feel that . . .”). All of your writing will be completely confidential. Don’t worry about spelling, sentence structure, or grammar. The only rule is that once you begin writing, you continue to do so until 20 minutes is up.

Thoughts Only Condition In this session, I would like you to write about your thoughts surrounding an unfair workplace experience that has affected you and your life. In your writing, I’d like you to explore your deepest thoughts. It is important that you do not explore your emotions or feelings (i.e., “I feel that . . .”). Please write only about what you think about the situation (i.e., “I think that . . .”; “I believe that . . .”). All of your writing will be completely confidential. Don’t worry about spelling, sentence structure or grammar. The only rule is that once you begin writing, you continue to do so until 20 minutes is up.

Emotions and Thoughts Condition In this session, I would like you to write about your emotions, feelings, and thoughts surrounding an unfair workplace experience that has affected you and your life. In your writing, I’d like you to explore your deepest emotions (i.e., I feel . . .) and thoughts (i.e., “I think that . . .”; “I believe that . . .”). All of your writing will be completely confidential. Don’t worry about spelling, sentence structure, or grammar. The only rule is that once you begin writing, you continue to do so until 20 minutes is up.

Participants wrote about “everything you have done since you got up this morning” in the second writing session, “the room you are currently sitting in” in the third writing session, and “your bedroom at home” in the final writing session.

Debriefing During the final session, after participants were finished with the experimental protocol (i.e., writing and questionnaire), they were interviewed to determine the nature of the unfairness that they had been writing about. This was done at the end of the study instead of the beginning in order to prevent (a) contamination with the manipulation (e.g., participants in the thoughts only condition might have expressed emotions during the interview, which would have contaminated the manipulation), (b) participants from anchoring themselves in the story that they might have initially told, justifying why it was unfair, and/or focusing on details of the situation rather than their emotional/cognitive reaction to it, (c) control participants from venting emotion or cognitively processing the injustice before the post-intervention measures were taken or inflating pre-intervention scores by having them reflect on the unfairness, and (d) participants from perceiving or benefiting from social support associated with expressing the unfairness to another person. Participants were debriefed in accordance with American Psychological Association guidelines. In addition to providing information about the type of injustice they had experienced, participants were asked about their impressions of the study and whether their participation resulted in any distress or harm (no harm was identified for any participant). In order to ensure the integrity of the experimental design, participants were mailed information about the nature of the study, results, and conclusions after data collection from all participants was complete.

Appendix B Description of the Linguistic Inquiry and Word Count (LIWC) Software The LIWC program was developed in order to efficiently and effectively process text samples, including identification of emotional and cognitive components (Pennebaker et al., 2007; Pennebaker, Francis, & Booth, 2001). In the present study, participants handwrote their writing samples. These samples were then transcribed into electronic text files, which were processed with the LIWC2007 software program. The program analyzed each writing sample, one word at a time, and cross-checked these words with

the software’s dictionary. The program is designed such that if a target word is found in the dictionary, the appropriate word category (e.g., negative emotions, causation) is incremented. The LIWC2007 dictionary is composed of 4,500 words and word stems that were identified and validated through an extensive process (see Pennebaker et al., 2007). Both the reliability and validity of the LIWC software have been examined across a number of studies (e.g., Pennebaker et al.,


RESEARCH REPORTS

2001). Validity was assessed by examining the relationship between individuals’ questionnaire responses and writing samples as well as by judges, who evaluated essays on the basis of their emotional and cognitive content and compared these evaluations with the LIWC output. Support for external validity was found— LIWC scales and judges’ ratings, for instance, were highly correlated (e.g., negative emotions, .31; causation, .44; see Pennebaker et al., 2007, for a detailed description and full results). In terms of reliability, the LIWC software has been evaluated by

523

examining the correlation between the occurrences of each word in a category with the sum of other words in the same category. The dimensions used in the present study have been found to have reliabilities ranging from .80 to .97 (see Pennebaker et al., 2007, for full results). Received November 19, 2007 Revision received June 13, 2008 Accepted June 26, 2008 䡲

Healing the Wounds of Organizational Injustice Examining the Benefits of Expressive Writing  

JournalofAppliedPsychology ©2009AmericanPsychologicalAssociation 2009,Vol.94,No.2,511–523 0021-9010/09/$12.00 DOI:10.1037/a0013451 Universit...

Advertisement