Stress, March 2009; 12(2): 152166 q Informa Healthcare USA, Inc.

ISSN 1025-3890 print/ISSN 1607-8888 online DOI: 10.1080/10253890802228178

Forgiveness and the appraisal-coping process in response to relationship conicts: Implications for depressive symptoms

RENATE YSSELDYK, KIMBERLY MATHESON, & HYMIE ANISMAN

Department of Psychology, Carleton University, Ottawa, Ontario, K1S 5B6, Canada
(Received 14 January 2008; revised 24 April 2008; accepted 26 May 2008 )

Abstract The present investigation assessed the mediating role of appraisal-coping processes in the relation between forgiveness and depressive symptoms associated with intimate relationship conicts. Study 1 assessed the role of forgiveness in the context of a severe relationship stressor, namely women experiencing dating abuse, along with the appraisal-coping responses and depressive symptoms associated with such a stressor. Study 2 evaluated the function served by forgiveness among men and women in response to non-abusive relationship stressors, including the dissolution of the relationship, and also assessed the relations among forgiveness, appraisal-coping processes, and depressive symptoms. Women who encountered dating abuse were less likely to forgive their partners, and this was linked to higher levels of depressive symptoms. The relation between forgiveness and lower depressive symptoms was partially mediated by lower threat appraisals, secondary appraisals of the effectiveness of emotion-focused coping, and the reduced endorsements of this coping strategy (Study 1). Appraisal-coping processes similarly mediated the relation between forgiveness and depressive symptoms among men and women reporting conict in an ongoing (non-abusive) relationship or a relationship break-up (Study 2). It is suggested that the relation between forgiveness and diminished distress operates primarily by guiding individuals appraisals of the conict and by diminishing the reliance on emotion-focused coping.

Keywords: Appraisals, coping, depressive symptoms, forgiveness, relationship conicts, stress

Introduction Interpersonal conicts are often stressful, and forgiveness might function like a coping strategy to diminish distress. In this regard, forgiving reactions may attenuate the negative impacts of relationship stressors, thereby limiting psychological disturbances such as depression that might otherwise arise, particularly when the conict is severe or persistent (McCullough 2000; Maltby et al. 2001; Brown 2003). Although various theoretical perspectives have been advanced regarding forgiveness in relation to stress and coping processes (Worthington and Scherer 2004; Strelan and Covic 2006; Worthington 2006), few studies have empirically examined how forgiveness operates within such a framework (Maltby et al. 2007; Rhoades et al. 2007). The present study examined the role of forgiveness in the context of relatively severe relationship

transgressions that are likely distressing, namely dating abuse (Study 1), and in the context of relationship dissolution (Study 2). The extent to which forgiveness was linked to appraisals and coping was assessed to determine whether these processes provided a mechanism through which forgiving reactions might serve to reduce stress-related depressive symptoms.

Forgiveness and the appraisal-coping process Despite denitional inconsistencies (McCullough et al. 2000; Macaskill 2005), most conceptualizations of forgiveness share the notion that it renders perceptions of interpersonal interactions more positive (i.e., more prosocial) and less negative (i.e., reduced negative thoughts, feelings, or behaviors) (McCullough et al. 2000). Although some individuals might have a more

Correspondence: H. Anisman, Department of Psychology, Carleton University, Ottawa, Ontario, K1S 5B6, Canada. Tel: 1 613 520 2699. Fax: 1 613 520 3667. E-mail: hanisman@ccs.carleton.ca

Forgiveness and the appraisal-coping process forgiving personality disposition (Berry et al. 2001; Brown 2003), forgiveness has also been viewed as a process (Strelan and Covic 2006; Worthington 2006; Worthington et al. 2007) that varies in response to specic transgressions and unfolds over time (McCullough et al. 2003; Tsang et al. 2006). In this regard, forgiveness may inuence appraisals and coping strategies used to deal with transgressions, thereby limiting the depressive symptoms that may be linked to those stressful interactions (Maltby et al. 2001; Brown 2003; Brown and Phillips 2005).

153

(Folkman and Lazarus 1980; Carver and Scheier 1994), vary over time (Tennen et al. 2000), and serve different functions or goals (Mattlin et al. 1990).

A brief overview of stress and coping. A greater understanding of the relations among forgiveness, appraisal-coping processes, and psychological health might benet from a theoretical perspective such as that advanced by Lazarus and Folkman (1984). Ordinarily, when individuals encounter a stressor, they engage in appraisals regarding the signicance or meaning of the event. These primary appraisals involve an evaluation of whether the event represents a threat, and if so, they are followed by secondary appraisals of whether the individual is equipped to cope (Lazarus and Folkman 1984; Carver and Scheier 1994; Peacock and Wong 1996; Chang 1998). Appraisals reecting greater threat or harm and a perceived inability to contend effectively with a stressor are typically associated with greater distress, including depressive symptoms (Lawson et al. 2002). Stressor appraisals may inuence the coping strategies individuals adopt, which serve to buffer against negative psychological outcomes (Lazarus and Folkman 1984). Generally, coping strategies have been grouped into superordinate categories, and in particular, problem-focused and emotion-focused coping (Folkman and Lazarus 1980). Problemfocused coping entails cognitive or behavioral efforts (cognitive restructuring or problem-solving techniques) to deal with the stressor (Carver et al. 1989; Endler and Parker 1994; Matheson and Anisman 2003). Emotion-focused strategies involve attempts to manage emotional aspects of the stressor, and are often divided into those that entail emotional approach (engaging with emotions evoked through rumination, expression, or seeking support) and avoidance (disengaging or denying the relevance of the event) (Endler and Parker 1994; Snyder and Pulvers 2001). It has been suggested that problemfocused coping is adopted when the individual appraises the situation as controllable and is typically related to more positive psychological outcomes, whereas emotion-focused coping might predominate when the situation is appraised as uncontrollable or threatening and has been linked to poorer well-being (Carver et al. 1989; Lazarus 1999; Matheson and Anisman 2003). However, the effectiveness of particular coping strategies may be situation-specic

Stress and coping in relation to forgiveness. It has been suggested (Strelan and Covic 2006) that forgiveness may operate in a fashion consistent with other coping processes delineated by Lazarus and Folkman (1984), in that individuals forgive interpersonal transgressions in order to diminish stress-related consequences (Worthington 2006). Given previous research demonstrating the positive link between forgiveness and well-being (Witvliet et al. 2001; Brown 2003), it might be expected that forgiveness would be positively associated with problem-focused coping, but negatively associated with emotion-focused coping strategies (e.g., rumination, emotional suppression; Witvliet and McCullough 2007). Others, however, have suggested that forgiveness may serve as a religious (Pargament et al. 2000), emotion-focused (Worthington and Scherer 2004), or proactive (i.e., self-improving) coping strategy (Strelan and Covic 2006). Indeed, ndings regarding the link between forgiveness and coping processes have been inconsistent. For example, forgiveness of a transgression has been associated with high levels of emotion-focused coping (Konstam et al. 2003), especially among women (Maltby et al. 2007). Conversely, diminished emotion-focused coping styles have been associated with dispositional forgiveness (Seybold et al. 2001; Ysseldyk et al. 2007) and with forgiveness of perpetrators of distal events (e.g., terrorism; Rhoades et al. 2007). These variations in the relations between forgiveness and coping could reect any number of factors, including a lack of consideration of specic forms of emotion-focused coping (e.g., engagement vs. avoidance) (Konstam et al. 2003; Maltby et al. 2007). In addition, these coping differences might be derived from variations in appraisals of the situation. For example, appraisals of an interpersonal stressor as a challenge may lead to forgiveness, whereas threat appraisals might be associated with unforgiveness (i.e., emotions reecting resentment, bitterness, hostility, hatred, anger, and fear; Worthington 2006). Appraisals of control may also be important in determining the coping responses that are used by forgiving individuals (Maltby et al. 2007). However, empirical research linking forgiveness to appraisal processes has been limited, although there is reason to believe that greater forgiveness is associated with lower appraisals of loss (Maltby et al. 2007) and with reduced stress appraisals in general, which in turn were related to better health outcomes (Lawler et al. 2005).

154 R. Ysseldyk et al. Forgiveness and appraisal-coping processes across relationship stressors The effectiveness of forgiveness in promoting adaptive coping responses and maintaining positive psychological states may vary across different relationship conicts. In particular, the benets of forgiveness might be less evident when transgressions are severe (Fincham et al. 2005), including those entailing abuse. Forgiveness in the context of an abusive relationship may be a double-edged sword. On the one side, it may buffer women against negative affect (Gordon et al. 2004; Reed and Enright 2006), perhaps by minimizing appraisals of the magnitude of the transgression or increasing perceptions of control over the situation (Boon and Holmes 1999). On the other side, forgiving an abusive partner might reect an avoidant strategy, with the understanding that attempts to confront or negotiate with the partner could increase the threat of violent confrontation (Follingstad et al. 1999; Matheson et al. 2007). In effect, forgiveness may serve to reduce womens appraisals of the severity of the situation or may be linked to secondary appraisals regarding the limited effectiveness of the various coping options available. These appraisals, in turn, may result in avoidant coping efforts that render women more likely to remain in their abusive relationships, and hence more vulnerable to depression. The function served by forgiveness may also differ depending on whether the conict occurs in the context of an ongoing relationship versus relationship termination. It has been suggested that, in an ongoing relationship, forgiveness may be tied to reconciliation (Worthington and Wade 1999; Waldron and Kelley 2005). To the extent that forgiveness is forthcoming in a relationship that both partners wish to maintain, it may well facilitate a resolution (perhaps by enhancing perceived control and minimizing the use of emotion-focused coping strategies), and hence promote well-being. Although relationship break-ups are often considered as a type of relationship transgression (Rye and Pargament 2002; Exline et al. 2004), it would seem reasonable to expect that forgiveness in this instance may be less motivated by a desire for reconciliation. Indeed, in the context of a break-up, forgiving responses may differ depending on which partner initiated the relationship dissolution (Hall and Fincham 2006). For the initiator of the break-up, forgiveness may serve to attenuate negative emotional exchanges or feelings. However, for the individual who is being left, a lack of forgiveness may paradoxically promote well-being precisely because it might facilitate the use of emotion-focused engagement coping strategies that allow individuals to vent their feelings. Although emotion-focused coping is often regarded as counterproductive, there has been increasing recognition that emotional approach or engagement strategies can facilitate the ability to come to terms with feelings in emotionally charged situations and, in doing so, may reduce distress (Stanton and Franz 1999; Stanton et al. 2000; Austenfeld and Stanton 2004). Thus, although forgiveness may be offered to halt the continuation of negative emotional exchanges, such a benet may be limited to the individual who initiated the relationship dissolution. For the jilted party, forgiveness of such conicts might have deleterious effects, including depressive symptoms, as it would prematurely close opportunities to express or vent their emotions. Overview of the present research The goal of the present research was to assess the relations among forgiveness, appraisal-coping processes, and depressive symptoms in the face of intimate relationship stressors. In this regard, two studies were conducted that differed in the characteristics of the relationship transgressions that were examined. In Study 1, the severity of the transgression varied, whereas in Study 2 the role of forgiveness in the context of an ongoing or dissolved relationship was assessed. Study 1 examined womens willingness to grant forgiveness for a transgression occurring in an ongoing dating relationship, including conicts that were characterised as psychologically or physically abusive. Psychological disturbances may be pronounced in association with a relationship that entails abuse (Golding 1999; Katz and Arias 1999; Campbell 2002; Ackard et al. 2003). In this regard, violence against women by their male partners exists in several forms, including psychological, physical, and sexual aggression, and has frequently been documented within the dating relationships of college-aged women (Follingstad et al. 1992; Neufeld et al. 1999; Matheson et al. 2007). Although men also experience abuse (Nicholls and Dutton 2001; Archer 2000), for the sake of greater conceptual clarity, Study 1 focused on the role of forgiveness in relation to womens experiences of abuse from their male partners. Study 2 assessed the role of the appraisal-coping process in the relation between forgiveness and depressive symptoms in association with a nonabusive transgression. In this respect, the transgression occurred in the context of an ongoing relationship, or entailed the relationship dissolution itself. As gender differences have been noted in the willingness to forgive (Macaskill et al. 2002; Mullet et al. 2003; Shackelford et al. 2002; Exline 2005) and in moderating appraisal-coping processes related to forgiveness (Maltby et al. 2007), the role of gender was also assessed in Study 2. The appraisal-coping processes that were expected to link forgiveness and depressive symptoms are

Forgiveness and the appraisal-coping process

155

Figure 1. Conceptual model of the relations between forgiveness and depressive symptoms, depicting the mediating role of primary stressor appraisals, secondary appraisals of coping effectiveness, and coping strategies endorsed.

depicted in Figure 1. Stemming from this model, it was hypothesized that: (1) forgiveness would be related to primary appraisals (greater perceived control and lower threat/impact of the conict), as well as secondary appraisals that emotion-focused coping (engagement and avoidant) would be ineffective. Moreover, forgiveness would be associated with greater problem-focused and lower endorsements of emotion-focused coping strategies. (2) the relations between forgiveness and coping strategies would be mediated by appraisals. (3) forgiveness would be associated with lower depressive symptoms, and this relation would be mediated by appraisal and coping processes. (4) the relations between forgiveness, appraisalcoping processes, and depressive symptoms would differ as a function of the severity of the conict. In this regard, women in abusive dating relationships (Study 1) would be less likely to forgive their partner, and the benets of forgiveness on appraisals, coping, and depressive symptoms would be less evident among these women. (5) the relations between forgiveness, appraisalcoping processes, and depressive symptoms would differ as a function of the nature (i.e., within ongoing vs. dissolution) of the conict (Study 2). The processes associated with relationship dissolution, in turn, might be dependent on whether the individual had been the initiator or victim of the break-up. It was expected that forgiveness would provide benets in the context of stable relationships involving relatively non-severe transgressions, but that these benets would be less evident among those who had not initiated the ending of their relationship.

Methods Study 1: Participants, procedure, and measures Female undergraduate students (N 85; Mean age 19.8 years, SD 2.2) were recruited to participate in a study assessing the impact of intimate relationships on psychological health. To achieve statistical reliability, women experiencing abuse in their dating relationships were over-sampled by recruiting women who responded afrmatively to a pre-measure indicating whether they perceived their current relationships to be characterized by emotional or physical abuse. The majority of women who provided racial/ethnic information indicated that they were Euro-Caucasian (n 64, 78.0%), whereas the remainder self-identied as Black (n 5, 6.1%), Asian (n 5, 6.1%), South Asian (n 6, 7.3%), Hispanic (n 1), or Native Canadian (n 1). After providing written informed consent, women completed self-report measures of abuse within their current relationships, appraisals and coping strategies in relation to their most troublesome recent conict with their intimate partner, forgiveness, and depressive symptoms. Women were debriefed and provided with a list of contact numbers that included counselling services should they experience any distress. This study was approved by the Carleton University Ethics Committee for Psychological Research. Abuse. The Revised Conict Tactics Scale (CTS-2; Straus et al. 1996) evaluated the behavioral tactics womens partners used toward them to resolve conicts. The tactics involving physical assault and psychological aggression were examined to establish the presence of abuse targeted at women in their dating relationships within the past month. Responses were made on a 6-point scale ranging from 0 (never) to 5 (more than 10 times). Internal consistencies for physical assault (a 0.81) and psychological aggression (a 0.70) subscales were adequate. The most common tactic reported was psychological aggression

156 R. Ysseldyk et al. from male partners, with 31 women scoring greater than 4 (i.e., reported more than 6 10 instances within the past month). We determined a cut-off score of 4 to dene psychological abuse, which was a relatively conservative criterion that maximized the likelihood that aggressive acts (e.g., my partner shouted at me) were considered in the context of broader conict (Kuffel and Katz 2002). Indeed, the mean number of events constituting psychological aggression reported by women who were categorized as being in psychologically abusive relationships (M 8.58, SD 3.39) was not considerably different from the number of such events reported by women in other studies who self-dened their relationships as psychologically abusive (Pipes and LeBov-Keeler 1997). Women in the present study who experienced psychological aggression were more likely to report physical assault, r 0.52, p , 0.001; 21 women reported an incident of physical assault, and 12 of these women also indicated high levels of psychological aggression. Thus, women were identied as involved in an abusive relationship if they indicated any instance of physical assault or if they scored 4 or higher on the psychological aggressiveness subscale (n 40).1 Stressor appraisals. To assess appraisals of conict stressors within their current relationships, women were asked to consider the disagreement that you and your partner had in the past month that you found the most troublesome. They then responded to three questions, namely, how much control did you feel that you had over the resolution to this disagreement? (control), was resolving this disagreement important to you? (centrality), and did this disagreement only affect you in a minor way, or do you feel that it affected almost everything you did? (impact). Responses ranged from 1 (hardly/not at all) to 7 (extremely). Coping strategies and appraisals of coping effectiveness. To assess the multidimensional nature of coping responses, we employed a scale (SCOPE: Survey of Coping Prole Endorsement; Matheson and Anisman 2003) comprising 44 items that reliably assess 12 strategies. These coping strategies include a range of cognitive/behavioral (problem-solving, cognitive restructuring, active and cognitive distraction, rumination) and socio-emotional methods (humor, social-support seeking, emotional expression, otherand self-blame, emotional containment, passive resignation). Respondents rated each item using a 5-point scale ranging from 0 (never) to 4 (almost always) in terms of whether they used each strategy to deal with their conict. In addition to the endorsement of coping behaviors, women rated the perceived effectiveness of each behavior using the same 5-point scale, in terms of whether the behavior could have improved their situation either psychologically or in reality, irrespective of whether or not they had actually used that behavior in response to their relationship conict. The psychometric characteristics of this scale have previously been described (Matheson and Anisman 2003; Mantler et al. 2005), including its use in the context of abusive dating relationships (Matheson et al. 2007). In order to reduce the number of coping variables and facilitate more parsimonious analyses, reduction of strategies to a subset of more stable styles is commonly considered. Given the modest sample size in the present investigation, we adopted the coping structure from a larger study (N 473; Matheson et al. 2007), which also included women experiencing abuse from their intimate partners. Thus, three coping styles were included in our analyses. The rst coping style reected problem-focused efforts (problemsolving, cognitive restructuring, active distraction, social support seeking, and humor; a 0.83 and 0.90 for endorsements and effectiveness, respectively). The second style reected emotional-avoidant coping efforts (self-blame, emotional containment, and passive resignation; a 0.84 and 0.85), and the third style reected emotional-engagement coping

Table I. Pearson zero-order correlations among appraisals and coping strategies in Study 1. 1 Stressor appraisals 1. Control 2. Centrality 3. Extent of impact Coping effectiveness 4. Engagement 5. Avoidant 6. Problem-focused Coping strategies endorsed 7. Engagement 8. Avoidant 9. Problem-focused 20.52*** 20.46*** 0.04 0.18 20.01 0.14 0.60*** 0.41*** 20.01 0.55*** 0.38*** 0.11 0.40*** 0.53*** 0.18 0.09 0.11 0.69*** 0.67*** 0.28* 20.24* 20.18 0.10 0.14 0.12 0.06 0.29** 0.20 20.06 0.72*** 0.02 2 3 4 5 6 7 8

0.08 20.37***

0.18

0.03

0.15

*p , 0.05; **p , 0.01; ***p , 0.001.

Forgiveness and the appraisal-coping process behaviors (rumination, emotional expression, and other-blame; a 0.87 and 0.85). Coping dimensions were moderately correlated with one another (see Table I). Forgiveness. The State Forgiveness Scale (SFS; Brown and Phillips 2005) comprises 7 items reecting responses toward ones current partner, including a lack of avoidant or vengeful feelings, but primarily in terms of more prosocial attitudes toward him. Women rated forgiveness of their partners in relation to their relationship conict using a scale ranging from 1 (strongly disagree) to 7 (strongly agree). Higher mean scores indicated greater forgiveness (a 0.88). Depressive symptoms. The 21-item Beck Depression Inventory (BDI; Beck et al. 1961) was used to assess depressive symptoms. Responses were summed to provide a single index, such that higher scores reected greater depressive symptoms (a 0.78). Study 2: Participants, procedure, and measures Male (n 35) and female (n 64) undergraduate students (M age 19.82 years; SD 2.31) participated in a study described as assessing the impact of intimate relationships on psychological health. Based on responses to a pre-measure, participants were contacted if they were currently in a heterosexual dating relationship (and indicated that their current relationship was not characterized by emotional, physical, or sexual abuse) (n 61), or had experienced a relationship break-up within the last six months (n 38). Participants who reported their ethnic status indicated that they were Euro-Caucasian (n 68, 68.7%), Asian (n 16, 16.1%), Black (n 7, 7.1%), Middle Eastern (n 3, 3.0%), or Hispanic (n 1). After providing written informed consent, participants described, in writing, a recent relationship conict or break-up. When the conict entailed a relationship break-up, they also indicated whether they had been the initiator (sole or mutual) or victim of the break-up. As in Study 1, responses to this conict were then assessed in terms of coping strategies endorsed and appraisals of the likely effectiveness of these strategies (SCOPE; Matheson and Anisman 2003; as . 0.80), willingness to forgive ones partner (SFS; Brown and Phillips 2005; a 0.85), and depressive symptoms (BDI; Beck et al. 1961; a 0.91). In addition, a detailed measure of primary stressor appraisals (SAM: Stress Appraisal Measure; Peacock and Wong 1990) was administered in Study 2. All participants were debriefed and provided with a list of contacts that included counselling services. This study was approved by the Carleton University Ethics Committee for Psychological Research. Conict type. The severity of participants relationship conicts was coded by the researcher (Ysseldyk) and by three independent judges. Each

157

conict description was read aloud and coded by each individual as minor (e.g., acute disagreements that were quickly resolved, for example, how to spend leisure time on weekend), moderate (e.g., chronic arguments that remained unresolved, for example, differing views on importance of religion), or entailed the break-up of the relationship. Having pre-selected participants on the basis of self-reporting a history that did not include abuse (i.e., which would have constituted a severe conict), not surprisingly, no such conicts were described. Based on this, 35 participants described a minor conict, 19 conveyed a moderate conict, and 45 reported a break-up (31 as sole or mutual initiator, 14 as victim). There were no differences in the nature of the conicts reported by men and women, x2 (2) 4.28, ns. Stressor appraisals. Appraisals of the conict or breakup were assessed using the SAM (Peacock and Wong 1990), a 28-item scale that assesses cognitive appraisals of a given stressor situation. The SAM taps into several primary appraisal dimensions including perceived threat (a 0.85), challenge (a 0.61), and centrality (a 0.87). Appraisals of control were assessed along three subscales reecting the extent to which the event was perceived as controllable by oneself (a 0.77), by others (a 0.86), or as uncontrollable by anyone (a 0.73). Respondents rated each item from 1 (not at all) to 5 (extremely). Correlations among the appraisal dimensions were moderate (Table II). Statistical analyses Multiple regression analyses were conducted to assess the relations among forgiveness (criterion variable) and each set of appraisal and coping indices (in separate analyses), including primary appraisals (control, centrality, and extent of impact), secondary appraisals of coping effectiveness, and coping endorsements (problem-focused, emotional-engagement, and avoidant). Likewise, multiple regression analyses were conducted to examine the relations among depressive symptoms and the appraisal-coping variables. To assess the potential mediating roles of appraisal-coping processes, Baron and Kennys (1986) procedures were followed. To examine the possible moderating effects of abuse on the relations between forgiveness and appraisal-coping processes and depressive symptoms in Study 1, a series of hierarchical regressions was conducted, wherein abuse (coded 0 for no abuse, and 1 for abuse) and forgiveness were entered on the rst step, followed by their interaction on the second. Simple slope analyses were conducted to further examine signicant interactions. In Study 2, a series of hierarchical regressions was conducted wherein forgiveness was regressed onto gender on the rst step, each set of appraisal and coping variables on the second step (in separate

158 R. Ysseldyk et al.

Table II. Pearson zero-order correlations among appraisals and coping strategies in Study 2. 1 Stressor appraisals 1. Centrality 2. Threat 3. Challenge 4. Controllable (self) 5. Controllable (other) 6. Uncontrollable Coping effectiveness 7. Engagement 8. Avoidant 9. Problem-focused Coping strategies endorsed 10. Engagement 11. Avoidant 12. Problem-focused

10

11

0.60*** 0.31*** 20.04 0.12 0.03

20.01 20.32*** 0.04 0.24***

0.40*** 0.30*** 20.07

0.38*** 20.23*** 20.11 20.15 0.25* 20.18 20.10 0.32***

20.04 0.18 0.26* 0.14

0.11 0.02 0.03

0.37*** 0.25* 0.05

0.01 20.09 0.18 0.18 0.03 0.37***

20.05 20.03 0.39***

0.67*** 0.25*

0.28**

0.21* 0.18 0.00

0.49*** 0.34*** 0.02

0.15 0.04 0.48***

0.34*** 0.21* 0.17

0.52*** 0.36*** 0.27**

0.37*** 0.40*** 0.31**

0.31** 0.31** 0.68***

0.63*** 0.39***

0.39***

p , 0.10; *p , 0.05; **p , 0.01; ***p , 0.001.

Forgiveness and the appraisal-coping process analyses), and the interactions between gender and the relevant appraisal or coping variables on the third. Hierarchical regressions were also conducted to assess the relations between depressive symptoms and both forgiveness and appraisal-coping processes that included gender as a potential moderating variable. Finally, to assess whether the relation between forgiveness and depressive symptoms differed as a function of the severity or nature of the conict, a hierarchical regression analysis was conducted wherein depressive symptoms were regressed onto three orthogonally coded dummy variables (participants who were in a relationship vs. those who reported a break-up; those who described a mild vs. a moderately severe conict within an ongoing relationship; those who had initiated vs. did not initiate the break-up) on the rst step. Forgiveness scores were entered on the second step, and the interactions between the dummy variables and forgiveness on the third. A comparable series of hierarchical regressions was conducted to assess possible interactive effects of the severity or nature of the conict and forgiveness on the appraisal and coping dimensions. Power analyses were conducted post-hoc for both studies to ensure that sample sizes were adequate (based on power 0.80, medium effect sizes (h 2 0.10), signicant at p , 0.05; Cohen 1992). Results Study 1: The inuence of forgiveness and appraisal-coping processes on depressive symptoms in response to relationship stressors including abuse Forgiveness and the appraisal-coping process. Examination of the zero-order correlations and

159

multiple regression results (Table III) indicated that, as hypothesized, womens willingness to forgive their partners was not only related to appraisals of the impact of the conict as being limited, but was also associated with perceiving greater control over the resolution of the conict. Likewise, forgiveness was related to secondary appraisals, in that greater forgiveness was associated with perceiving both emotion-focused engagement and avoidant strategies as less effective. Finally, forgiveness was related to womens coping endorsements, and particularly with a lower likelihood of adopting emotion-focused engagement or avoidant strategies (Table III). The appraisal dimensions associated with both forgiveness and emotion-focused coping were assessed as potential mediators. The relation between forgiveness and lower emotional engagement coping was partially mediated by primary appraisals of control, Sobels z 22.98, p , 0.001, and impact, Sobels z 23.10, p , 0.001, as well as secondary appraisals of the effectiveness of this coping strategy, Sobels z 22.47, p , 0.05. However, in all instances forgiveness continued to have a direct effect (direct b 2 0.27, 2 0.26, and 2 0.31, respectively; all p , 0.01). Similarly, the relation between forgiveness and avoidant coping was partially mediated by control appraisals, Sobels z 2 2.54, p , 0.01, impact appraisals, Sobels z 22.13, p , 0.05, and perceived effectiveness of this strategy, Sobels z 2 2.63, p , 0.001, but once again, forgiveness remained a signicant direct predictor (b 20.32, 20.36, and 20.33, respectively; all p , 0.05). Thus, forgiveness was directly related to lower endorsements of engagement and avoidant coping strategies, as well as indirectly via its relations to primary (control and extent of impact) and secondary appraisals.

Table III. Multiple regression analyses predicting forgiveness and depressive symptoms from appraisals and coping strategies to contend with a relationship conict in Study 1. Forgiveness r Stressor appraisals Control Centrality Extent of impact Coping effectiveness Engagement Avoidant Problem-focused Coping strategies endorsed Engagement Avoidant Problem-focused

Depressive symptoms R2 0.289*** r

b
20.19 20.13 0.30** 0.26 0.12 0.15

R2 0.178***

pra

Sobelsa

0.48*** 0.10 20.38*** 20.30** 20.33*** 0.10 20.45*** 20.47*** 20.08

0.37*** 0.11 20.27** 0.129** 20.11 20.26 0.11 0.254*** 20.25 20.31* 0.03

20.32** 20.09 0.35*** 0.35*** 0.31** 20.16 0.47*** 0.51*** 20.09

20.33** 20.34** 0.153** 20.37** 20.37** 0.342***

21.30 21.83 21.85 21.54

0.31* 0.34** 20.23*

20.28* 20.25*

22.71** 23.01**

p , 0.10; *p , 0.05; **p , 0.01; ***p , 0.001; a The partial correlations are between forgiveness and depressive symptoms controlling for the respective appraisal and coping variables; partial correlations are provided instead of standardized regression coefcients, as they can be compared directly to the zero-order correlation between forgiveness and symptoms to evaluate changes in the magnitude of the relation. The Sobels values are for the indirect mediated path linking forgiveness and depressive symptoms.

160 R. Ysseldyk et al. Processes linking forgiveness and depressive symptoms. As expected, women who were more willing to forgive their partners reported lower depressive symptoms, r 2 0.43, p , 0.001. The extent to which appraisals and coping processes were linked to depressive symptoms, and hence potentially mediated this relation, was therefore evaluated. As seen in Table III, multiple regression analyses revealed that primary stressor appraisals were predictive of depressive symptoms, in that greater symptoms were related to perceiving the conict as less controllable and having a more threatening impact. In addition, appraisals of coping effectiveness and coping endorsements were signicantly related to depressive symptoms, such that symptoms were greater when emotion-focused strategies were appraised as more effective and were employed (Table III). The potential mediating roles of appraisals and coping in the relation between forgiveness and depressive symptoms were examined, focusing on those variables that were correlated with both constructs, namely appraisals of control, the impact of the conict, and the effectiveness and endorsement of emotional engagement and avoidant coping strategies. Although several of the appraisal and coping styles served a partial mediating role, in no instance did they render the relation between forgiveness and depressive symptoms non-signicant, suggesting that forgiveness also had a direct effect on well-being, even after considering the mediated relations through appraisals and coping (Table III). examined, these too were found to be non-signicant, with the exception of avoidant coping endorsements, R 2 0.041, F(1,81) 4.60, p , 0.05. A simple cha slope analysis revealed that forgiveness was especially related to lower endorsements of avoidant coping among non-abused, b 2 0.53, p , 0.001, relative to abused women, b 2 0.35, p , 0.05. Study 2: The inuence of forgiveness and appraisal-coping processes on depressive symptoms as a function of the severity and nature of relationship stressors Forgiveness and the appraisal-coping process. Men were more forgiving of their partners (M 6.08, SD 0.91) than were women (M 5.53, SD 1.36), R 2 0.041, F(1,95) 4.09, p , 0.05. However, gender did not signicantly moderate the relations between forgiveness and any of the appraisal or coping variables. As expected, primary appraisals were related to forgiveness. In particular, greater forgiveness was associated with perceiving the conict as less threatening, as less uncontrollable in general, and as more controllable by oneself (Table IV). As in Study 1, forgiveness was related to appraisals of coping effectiveness, although in this instance, forgiveness was only linked to appraising avoidant coping as less effective. In relation to coping endorsements, forgiveness was related to lower endorsements of both emotional engagement and avoidant coping (Table IV). Given that forgiving individuals had a lower inclination to endorse emotional engagement and avoidant coping styles, it was of interest to assess whether these relations were mediated by primary appraisals of the conict that were related to both forgiveness and emotional coping (i.e., threat and uncontrollability), or by the perceived effectiveness of avoidant coping strategies. Appraisals of threat and uncontrollability were found to account for the relation between forgiveness and lower endorsements of emotion-focused coping. Specically, primary appraisals of the situation as threatening, Sobels z 2 1.93, p 0.053, or as uncontrollable, Sobels z 2 2.25, p , 0.05, reduced the direct relation between forgiveness and engagement coping to non-signicance ( b 2 0.11, 2 0.13, ns, respectively). Likewise, when threat appraisals were considered in the relation between forgiveness and avoidant coping, they again served a mediating role, Sobels z 2 1.73, p 0.08, reducing the direct relation between forgiveness and coping to non-signicance ( b 2 0.17, ns). However, neither appraisals of the uncontrollability of the situation, Sobels z 2 1.24, ns, nor the effectiveness of avoidant coping, Sobels z 2 1.50, ns, served as signicant mediators. Thus, forgiveness following the conicts reported in Study 2 was directly related to a lower likelihood of using emotional engagement and avoidant coping. As in Study 1, the associations with threat and control

Forgiveness in the context of abuse. As hypothesized, women in abusive dating relationships were less willing to forgive their partners (M 5.73, SD 1.18), and reported greater depressive symptoms (M 11.33, SD 5.72) than women who were not in such relationships (forgiveness M 6.52, SD 0.63; depressive symptoms M 8.57, SD 5.26). In addition, although primary appraisals did not differ between abused and non-abused women, those in abusive relationships were more likely to appraise avoidant strategies as effective (M 1.33, SD 0.71), and to endorse avoidant (M 1.94, SD 0.69) and engagement coping (M 1.82, SD 0.78) relative to non-abused women (avoidant effectiveness M 0.94, SD 0.60; avoidant endorsements M 1.48, SD 0.68; engagement endorsements M 1.29, SD 0.70), all p , 0.05. Contrary to expectations, the interaction between forgiveness and abuse was not signicant, R 2 0.006, F , 1, indicating that forgiveness was cha associated with lower depressive symptoms, irrespective of the abusive nature of womens relationships. When the interactive effects of abuse and forgiveness on each of the appraisal and coping variables were

Forgiveness and the appraisal-coping process

161

Table IV. Multiple regression analyses predicting forgiveness and depressive symptoms from appraisals and coping strategies to contend with a relationship conict in Study 2. Forgiveness r Stressor appraisals Centrality Threat Challenge Controllable (self) Controllable (others) Uncontrollable Coping effectiveness Engagement Avoidant Problem-focused Coping strategies Engagement Avoidant Problem-focused Depressive symptoms R2 0.177** 0.10 20.22* 0.13 0.21* 0.08 20.34*** 20.12 20.17* 0.13 20.23* 20.24** 0.06 0.27* 20.30* 0.04 0.02 0.02 20.25* 0.084* 20.01 20.24 0.25* 0.110* 20.11 20.27* 0.24* 0.36*** 0.37*** 20.10 0.04 0.00 0.10 0.04 20.03 0.06 0.23 0.37** 20.36*** 0.257*** 20.29** 20.29** 21.81 2 2.02* 0.08 0.34*** 20.15 20.30** 20.14 0.23* 20.08 0.34* 20.05 0.09 20.09 0.07 0.005 r

R2 0.175**

pra

Sobelsa

20.30** 20.31** 20.29**

21.71 21.55 21.18

p , 0.10; *p , 0.05; **p , 0.01;***p , 0.001; a The partial correlations are between forgiveness and depressive symptoms controlling for the respective appraisal and coping variables, and the Sobels values are for the indirect mediated path linking forgiveness and depressive symptoms.

appraisals (the latter in the case of engagement coping only) partially accounted for these relations; however, unlike Study 1, secondary appraisals did not.

as in Study 1, in all instances forgiveness continued to have a direct effect on well-being even after considering the mediated relations through appraisals and coping (Table IV).

Processes linking forgiveness and depressive symptoms. Men reported marginally fewer depressive symptoms (M 7.60, SD 7.36) than did women (M 10.85, SD 9.15), R 2 0.033, F(1,96) 3.26, p 0.074. As in Study 1, forgiveness was associated with lower depressive symptoms, b 2 0.32, p , 0.001, and gender did not moderate this relation, F , 1. Depressive symptoms were related to appraising the conict as more threatening and as uncontrollable by oneself or anyone (Table IV). In contrast to Study 1, appraisals of coping effectiveness were not related to depressive symptoms, whereas coping endorsements were, particularly with respect to the increased use of emotional engagement and avoidant strategies. Gender did not moderate the relations between appraisals or coping and depressive symptoms. The potential mediating roles of appraisals and coping in the relation between forgiveness and depressive symptoms were examined, focusing on those dimensions that were signicantly related to both constructs, namely appraisals of threat, of how much control one had over the conict, of how controllable the conict was by anyone, and endorsements of both emotional engagement and avoidant coping. Although only primary appraisals of threat played a partial mediating role in the relation between forgiveness and depressive symptoms, both emotional engagement and avoidant coping endorsements partially mediated this relation. However,

Forgiveness as a function of severity and nature of relationship conicts. It was expected that the relation between forgiveness and depressive symptoms might differ as a function of the severity (i.e., mild, n 35 vs. moderate, n 19) or nature (i.e., within ongoing vs. dissolution) of the relationship conict. The processes associated with the dissolution of the relationship, in turn, might be dependent on whether the individual perceived him or herself to have been an initiator (solely or mutually, n 31) or a victim of the breakup (n 14). A one-way ANOVA (collapsed across sex, given sample sizes and prior ndings that sex did not moderate processes linked to forgiveness) conducted to examine differences in forgiveness was signicant, F(3,92) 4.69, p , 0.01, h 2 0.131. As expected, those who reported a conict within an ongoing relationship were more forgiving of their partner (M 6.02, SD 1.11) than those responding to the dissolution of their relationship (M 5.24, SD 1.32), p , 0.001. There were no differences in forgiveness between those describing a mild versus moderate conict, or as a function of who initiated the break-up. No differences in levels of depressive symptoms were found between the groups, R 2 0.020, Fcha(3,92) 2.44, ns, nor was the cha interaction between type of conict and forgiveness signicant, R 2 0.020, Fcha , 1. Evidently, the cha severity or nature of the conict did not moderate the

162 R. Ysseldyk et al. relation between forgiveness and lower depressive symptoms. The type of conict did not moderate the relations between forgiveness and primary stressor appraisals. However, the interaction between the type of conict and forgiveness was signicant in relation to secondary appraisals of coping effectiveness, and in particular, the effectiveness of emotional engagement, R 2 0.200, Fcha(3,87) 8.26, p , 0.001, and cha avoidant coping, R 2 0.115, Fcha(3,87) 4.46, cha p , 0.01. Specically, the interaction contrasts between participants who had or had not initiated the dissolution of their relationship were signicant, t(87) 4.78, p , 0.001 (engagement), t(87) 3.41, p , 0.01 (avoidant). Among individuals who had been the victim of the break-up, forgiveness was associated with appraising emotion-focused strategies as less effective, b 2 0.84, p , 0.001 (engagement), b 2 0.63, p , 0.05 (avoidant); however, among individuals who had been the sole or mutual initiator of the break-up, forgiveness and appraisals of the effectiveness of such strategies were not related, bs 0.26, 0.11, ns, (engagement and avoidant coping, respectively). Lastly, the relations between forgiveness and endorsements of emotional engagement coping also varied as a function of the nature of the conict, R 2 0.083, Fcha(3,87) 2.94, p , 0.05, and once cha again, this was signicant with respect to the interaction contrast between individuals who had or had not been initiators of their break-ups, t(87) 2.88, p , 0.01. Among those who had been the victim of the break-up, forgiveness was related to lower levels of endorsing emotional engagement coping, b 2 0.65, p , 0.05, whereas among individuals who had been the initiator of the breakup, forgiveness and endorsements of this type of coping were not related, b 0.12, ns. mechanism linking forgiveness and lower depressive symptoms. Links among forgiveness, appraisal-coping processes, and depressive symptoms. Despite several theoretical considerations regarding forgiveness within a stress and coping framework (Worthington and Scherer 2004; Strelan and Covic 2006; Worthington 2006), the role of primary and secondary appraisals in relation to forgiveness has not been systematically examined in past research. Thus, the ndings of the present investigation were novel, as a willingness to forgive was consistently related to both primary and secondary appraisals. In both studies, primary appraisals of threat (dened as the extent of impact of the transgression in Study 1) partially mediated the relation between forgiveness and depressive symptoms, suggesting that lower distress among forgiving individuals might be accounted for by reductions in the perceived impact of the situation. Indeed, these ndings are also consistent with the suggestion that forgiveness may promote positive reappraisals of stressful interpersonal events, thereby decreasing negative and increasing positive emotions, along with the associated health outcomes (Witvliet and McCullough 2007). In addition, control appraisals partially mediated the relation between forgiveness and both emotion-focused engagement and avoidant coping endorsements in Study 1, supporting the view that forgiveness in the context of situations that are perceived as controllable should be related to reduced emotion-focused coping (Maltby et al. 2007). Although control appraisals also mediated the relation between forgiveness and engagement coping in Study 2, these appraisals did not mediate the relation with avoidant coping efforts. Given that the conicts considered in Study 1 were likely more severe (i.e., abusive) than in Study 2, it may be that under normal circumstances of dating conicts, forgiving individuals inclination to reject avoidant coping strategies occurs relatively independently of their need to take control over the situation. Beyond the importance of primary appraisals, forgiveness was associated with secondary appraisals, and in particular, secondary appraisals of the ineffectiveness of emotion-focused coping efforts. As indicated earlier, although research on stress and coping has often considered the role of primary appraisals (Lazarus and Folkman 1984; Carver and Scheier 1994; Peacock and Wong 1996; Chang 1998), much less attention has been devoted to the role of secondary appraisals (i.e., appraisals of the effectiveness of particular coping strategies) in determining coping responses. Indeed, as expected, secondary appraisals played an integral role in mediating the relations between forgiveness and the coping efforts endorsed. This was particularly notable with respect to engagement coping efforts, and less so in relation to avoidant coping (at least in the context of the

Discussion The ndings of the present investigation are consistent with past research linking forgiveness of relationship transgressions with positive psychological health (Maltby et al. 2001; Witvliet et al. 2001; Brown and Phillips 2005). This was the case irrespective of gender and the severity or nature of the relationship transgression, although forgiveness was less forthcoming under conditions of abuse (Study 1), or in response to a partner with whom a relationship had just ended (Study 2). Also in line with previous research (Lazarus and Folkman 1984; Carver et al. 1989; Matheson and Anisman 2003), appraisals and coping strategies were predictive of depressive symptomatology across the two studies. Most notably, consistent with expectations, appraisal-coping processes evoked by stressful experiences served as a mediating

Forgiveness and the appraisal-coping process non-abusive transgressions reported in Study 2). Previous research concerning coping effectiveness has focused mainly on the success of outcomes stemming from coping strategies that have already been used (Ntoumanis and Biddle 1998; Gottlieb and Rooney 2004), whereas in the present investigation, participants considered each coping behavior in terms of its potential effectiveness, irrespective of whether or not they had used this strategy. In this sense, our approach may have been more true to the dynamic aspects of the appraisal-coping process, reecting individuals perceptions of themselves as having the resources necessary to contend with the stressor (Lazarus and Folkman 1984). Of course, participants in the present investigation were also responding to a conict that had already occurred, and so it remains possible that their appraisals were based on outcomes generated by their actual coping strategies. Although this possibility cannot be dismissed, consideration of this alternative causal direction (i.e., coping endorsements preceding effectiveness appraisals) in our mediational analyses was not signicant. Thus, the reduced likelihood of endorsing emotion-focused coping strategies among forgiving individuals was likely derived from their perceptions that such strategies would not be constructive. Taken together, these ndings suggest that an important function of forgiveness is to inuence how people come to understand or appraise their situations, rather than operating primarily as a coping strategy itself (Pargament et al. 2000; Worthington and Scherer 2004; Strelan and Covic 2006). This said, in both studies, forgiveness was negatively related to both types of emotion-focused coping and not at all related to problem-focused coping. It had been suggested that some of the inconsistent ndings in previous research, and particularly in relation to emotion-focused coping, were due to the lack of distinction between engagement and avoidant strategies (Seybold et al. 2001; Konstam et al. 2003). This may be true under certain circumstances, in that when individuals contending with a relationship break-up were considered, forgiveness was associated with lower emotional engagement coping among the jilted partners. Although these ndings are suggestive of variations in the role of forgiveness and coping across transgressions, the differences appeared to be more a matter of degree than in fundamentally altering (i.e., reversing) the relation between forgiveness and coping2. Moreover, in line with previous suggestions that forgiveness may regulate or even replace negative emotions as a pathway to improved health (Worthington 2006; Witvliet and McCullough 2007), reduced endorsements of both types of emotion-focused coping partially mediated the relation between forgiveness and lower depressive symptoms. In effect, forgiveness did not seem to facilitate the ability to solve problems (given the lack of a relationship with problem-focused

163

coping), but rather was more closely aligned with a diminished need to rely on emotion-focused coping strategies that, in turn, reduced the occurrence of stress-related psychological disturbances (i.e., depressive symptoms). Such ndings may also have implications for clinical practice and prevention. For example, if forgiving individuals are less likely to engage in typically maladaptive emotion-focused coping styles, forgiveness might promote a letting go of negative emotions more broadly. In this sense, forgiveness may buffer against negative evaluations of the self and ones relationships in general, thus increasing autonomy and decreasing sociotropy in line with Becks (1983) diathesis-stress model of depression (Bieling et al. 2004). Forgiveness, appraisal-coping processes, and depressive symptoms across relationship stressors. It had originally been expected that the severity or nature of the relationship stressors would moderate the role of appraisal-coping processes in the relation between forgiveness and depressive symptomatology. Specically, in Study 1, it was anticipated that if women forgave their currently abusive partners, this might undermine their well-being by rendering them more vulnerable to an unsafe environment (McCullough 2000), in part through inappropriate appraisals and the endorsement of ineffective coping strategies. Although women in abusive relationships were less likely to forgive, such a moderating effect of abuse was not borne out. Even with a transgression as severe as abuse, when forgiveness was not forthcoming, the individual was more vulnerable to compromised psychological health. However, had the role of forgiveness been examined in response to severe domestic violence, the positive relation between forgiveness and greater depressive symptoms might have been observed. Forgiving attitudes were also linked to the nature of the relationship conicts in Study 2 in that, not surprisingly, those who reported a conict within an ongoing relationship were more forgiving of their partners than those responding to the dissolution of their relationship. Interestingly, the tendency to forgive was not signicantly different between those who had or had not initiated the break-up. However, the relations between forgiveness and emotional engagement coping, as well as appraisals of the effectiveness of emotional engagement and avoidant coping, varied as a function of individuals role in the break-up. As expected, among the jilted parties, withholding forgiveness was associated with appraisals of emotion-focused coping as more effective and with a greater likelihood of endorsing emotionally engaging coping strategies (i.e., rumination, emotional expression, other-blame; Rhoades et al. 2007; Ysseldyk et al. 2007). It seems that those who are especially hurt by a partners choice to terminate the relationship, and thus unwilling to forgive, may be more likely to rely on such emotionally expressive techniques in an attempt to come to terms with their feelings

164 R. Ysseldyk et al. (Austenfeld and Stanton 2004; Stanton et al. 2000; Stanton and Franz 1999). Conclusions and caveats. Despite the consistent relations observed between forgiveness and depressive symptoms and the mediating role of the appraisalcoping process in the present investigation, there may be limits to the generalizability of the ndings. It is possible that the consistency in responses across transgressions reected characteristics of the sample itself, which comprised young, relatively highfunctioning students whose relationships may have involved low levels of investment (e.g., no children, limited shared assets). Indeed, it has been suggested that older people tend to be more forgiving than younger people (Mullet et al. 1998), and the nature of the transgressions as well as the appraisals and coping strategies linked to forgiveness might also vary with age (Hunt et al. 2003). Thus, although dating relationship conicts may predict future relationship distress (Wekerle and Wolfe 1999), variations in the relations among forgiveness, appraisal-coping processes, and depressive symptoms might be evident among older populations experiencing severe domestic abuse or divorce. A second limitation of the present investigation was the reliance on retrospective self-report measures. Although there has been ample criticism of self-report questionnaire measures of coping (Stone et al. 1998; Coyne and Racioppo 2000), this does not belie the fact that the self-report approach was predictive of individuals behavioral propensities and of well-being. Moreover, across the two studies, the differential relations between forgiveness and emotion- versus problem-focused coping appraisals and endorsements were consistent. Nevertheless, a disjunction may exist between what a person says they would do and what they actually do in a given situation (Stone et al. 1998). Indeed, it is possible that participants reports may have been biased by their current distress. Thus, a longitudinal analysis concerning the impact of forgiveness on day-to-day cognitions and coping behaviors in response to relationship conicts may provide important insights (McCullough et al. 2003; Tsang et al. 2006). Moreover, although a broad array of coping strategies was assessed, still other means of coping (e.g., alcohol and substance use) may be informative in future research, especially with regard to depressive symptoms in the context of abuse (Swan and Snow 2006). Finally, it is important to emphasize that the data of the present investigation are correlational, and so causal conclusions must be considered cautiously (e.g., appraisals of control may lead to a greater willingness to forgive). In this regard, it has been suggested that both forgiveness and coping may be bidirectional processes (Strelan and Covic 2006). However, all mediated models considering alternative directional pathways were tested, but none were signicant. Thus, in the present study, forgiveness likely acted on depressive symptoms by affecting appraisal and coping processes. In conclusion, the present investigation points to forgiveness as being more forthcoming in response to some relationship transgressions, and when forgiveness was granted, it was associated with benets to psychological health (i.e., lower depressive symptoms). Most notably however, in line with Lazarus and Folkmans (1984) stress and coping theory, one of the mechanisms by which such benets were derived was through the appraisal-coping process. In this regard, forgiveness appeared to be primarily associated with the diminished likelihood that individuals would turn to emotion-focused coping strategies, in part because forgiving individuals interpreted the transgression as less threatening and more controllable, and believed that emotional coping would be ineffective. Thus, the present investigation brings us closer to answering some of the questions regarding the role of stress and coping processes in the direct and indirect effects of forgiveness on psychological health (Worthington et al. 2007). Rather than operating as a coping strategy itself, forgiveness primarily served to limit cognitions that might otherwise result in an escalation of the emotional aspects of the situation.

Acknowledgements This research was funded by research grants from the Canadian Institutes of Health Research and the Social Sciences and Humanities Research Council of Canada. Renate Ysseldyk was supported by a Social Sciences and Humanities Research Council of Canada Fellowship. Hymie Anisman holds a Canada Research Chair in Behavioral Neuroscience. Our appreciation to Catherine Brunelle, Anna-Marie Danielson, Neil McVicar, Anilise Otten, Alla Skomorovksy, and Kerry Sudom for their assistance. Declaration of interest: The authors report no conicts of interest. The authors alone are responsible for the content and writing of the paper. Notes
1. Although there is some evidence that psychological and physical aggression may be differentially related to forgiveness (Gauche and Mullet 2005), when the analyses were conducted with physical and psychological abuse considered as separate variables (and also retaining psychological abuse as a continuous variable, which could not be done with physical assault given its high skew), the pattern of ndings was identical to those described. 2. It has also been argued that forgiveness may be differentially related to other constructs based on its operationalization (e.g., prosocial attitudes vs. absence of revenge or avoidance) (Fincham and Beach 2002; Brown 2003). Thus, it should be noted that this possibility was assessed in both an independent sample of women in abusive dating relationships (N 138) and in Study 2 using a measure of unforgiveness (TRIM; McCullough et al. 1998); however, forgiveness was, in all cases, negatively associated with

Forgiveness and the appraisal-coping process

emotion-focused coping, threat appraisals, and depressive symptoms. The only different result was found in the relations among abuse, forgiveness, and appraisals of control. Specically, in non-abusive relationships, such forgiveness was not associated with control appraisals, b 0.13, ns; in contrast, when abuse was present in womens relationships, forgiveness was related to lower appraisals of control, b 20.32, p , 0.05. In effect, these data suggest that women who forgave (i.e., did not avoid or seek revenge against) their abusive partners may have perceived their situations as beyond their control.

165

References
Ackard DM, Neumark-Sztainer D, Hannan P. 2003. Dating violence among a nationally representative sample of adolescent girls and boys: Associations with behavioral and mental health. J Gend Spec Med 6:3948. Archer J. 2000. Sex differences in aggression between heterosexual partners: A meta-analytic review. Psychol Bull 126:697 702. Austenfeld JL, Stanton AL. 2004. Coping through emotional approach: A new look at emotion, coping, and health-related outcomes. J Personal 72:1335 1363. Baron RM, Kenny DA. 1986. The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. J Pers Soc Psychol 51: 11731182. Beck AT. 1983. Cognitive therapy of depression: New perspectives. In: Clayton PJ, Barnett JE, editors. Treatment of depression: Old controversies and new approaches. New York: Raven Press. p 265 290. Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J. 1961. An inventory for measuring depression. Arch Gen Psychiat 4: 561 569. Berry JW, Worthington Jr. EL, Parrott L, OConnor LE, Wade NG. 2001. Dispositional forgivingness: Development and construct validity of the Transgression Narrative Test of Forgivingness. Pers Soc Psychol Bull 27:1277 1290. Bieling PJ, Beck AT, Brown GK. 2004. Stability and change of sociotropy and autonomy subscales in cognitive therapy of depression. J Cogn Psychother 18:135148. Boon SD, Holmes JG. 1999. Interpersonal risk and the evaluation of transgressions in close relationships. Pers Rel 6:151 168. Brown RP. 2003. Measuring individual differences in the tendency to forgive: Construct validity and links with depression. Pers Soc Psychol Bull 29:759771. Brown RP, Phillips A. 2005. Letting bygones be bygones: Further evidence for the validity of the Tendency to Forgive scale. Pers Indiv Diff 38:627638. Campbell JC. 2002. Health consequences of intimate partner violence. Lancet 359:13311336. Carver CS, Scheier MF. 1994. Situational coping and coping dispositions in a stressful transaction. J Pers Soc Psychol 66: 184 195. Carver CS, Scheier MF, Weintraub JK. 1989. Assessing coping strategies: A theoretical based approach. J Pers Soc Psychol 56: 267 283. Chang EC. 1998. Dispositional optimism and primary and secondary appraisal of a stressor: Controlling for confounding inuences and relations to coping and psychological and physical adjustment. J Pers Soc Psychol 74:11091120. Cohen J. 1992. A power primer. Psychol Bull 112:155159. Coyne JC, Racioppo MW. 2000. Never the twain shall meet? Closing the gap between coping research and clinical intervention research. Am Psychologist 55:655664. Endler NS, Parker JDA. 1994. Assessment of multidimensional coping: Task, emotion, and avoidance strategies. Psychol Assess 6:5060. Exline JJ. 2005. Not so innocent: Can recalling our own offences promote forgiveness toward others? Presented at the annual meeting of the Soc Pers Soc Psychol New Orleans, LA.

Exline JJ, Baumeister RF, Bushman BJ, Campbell WK, Finkel EJ. 2004. Too proud to let go: Narcissistic entitlement as a barrier to forgiveness. J Pers Soc Psychol 87:894 912. Fincham FD, Beach SRH. 2002. Forgiveness in marriage: Implications for psychological aggression and constructive communication. Pers Rel 9:239251. Fincham FD, Jackson H, Beach SRH. 2005. Transgression severity and forgiveness: Different moderators for objective and subjective severity. J Soc Clin Psychol 24:860875. Folkman S, Lazarus RS. 1980. An analysis of coping in a middleaged community sample. J Health Soc Behav 21:219239. Follingstad DR, Rutledge LL, McNeill-Harkins K, Polek DS. 1992. Factors related to physical violence in dating relationships. In: Viano ED, editor. Intimate violence: Interdisciplinary perspectives. Washington, DC: Hemisphere Publishing. p 121 135. Follingstad DR, Bradley RG, Laughlin JE, Burke L. 1999. Risk factors and correlates of dating violence: The relevance of examining frequency and severity levels in a college sample. Viol Vict 14:365380. Gauche M, Mullet E. 2005. Do we forgive physical aggression in the same way that we forgiven psychological aggression? Aggress Behav 31:559570. Golding JM. 1999. Intimate partner violence as a risk factor for mental disorders: A meta-analysis. J Fam Viol 14:99132. Gordon KC, Burton S, Porter L. 2004. Predicting the intentions of women in domestic violence shelters to return to partners: Does forgiveness play a role? J Fam Psychol 18:331338. Gottlieb BH, Rooney JA. 2004. Coping effectiveness: Determinants and relevance to the mental health and affect of family caregivers of persons with dementia. Aging Men Health 8:364 373. Hall JH, Fincham FD. 2006. Relationship dissolution following indelity: The roles of attributions and forgiveness. J Soc Clin Psychol 25:508522. Hunt S, Wisocki P, Yanko J. 2003. Worry and use of coping strategies among older and younger adults. J Anx Disord 17: 547 560. Katz J, Arias I. 1999. Psychological abuse and depressive symptoms in dating women: Do different types of abuse have differential effects? J Fam Viol 14:281295. Konstam V, Holmes W, Levine B. 2003. Empathy, selsm, and coping as elements of the psychology of forgiveness: A preliminary study. Counsel Val 47:172 183. Kuffel S, Katz J. 2002. Preventing physical, psychological, and sexual aggression in college dating relationships. J Prim Prevent 22:361374. Lawler KA, Younger JW, Piferi RL, Jobe R, Edmondson K, Jones WH. 2005. The unique effects of forgiveness on health: An exploration of pathways. J Behav Med 28:157167. Lawson C, MacLeod C, Hammond G. 2002. Interpretation revealed in the blink of an eye: Depressive bias in the resolution of ambiguity. J Abn Psychol 111:321328. Lazarus RS. 1999. Stress and emotion. A new synthesis. New York: Springer. Lazarus RS, Folkman S. 1984. Stress appraisal, and coping. New York: Springer. Macaskill A. 2005. Dening forgiveness: Christian clergy and general population perspectives. J Pers 73:12371265. Macaskill A, Maltby J, Day L. 2002. Forgiveness of self and others and emotional empathy. J Soc Psychol 142:663665. Maltby J, Macaskill A, Day L. 2001. Failure to forgive self and others: A replication and extension of the relation between forgiveness, personality, social desirability and general health. Pers Indiv Diff 30:881885. Maltby J, Macaskill A, Gillett R. 2007. The cognitive nature of forgiveness: Using cognitive strategies of primary appraisal and coping to describe the process of forgiving. J Clin Psychol 63: 555 566.

166 R. Ysseldyk et al.

Mantler J, Matejicek A, Matheson K, Anisman H. 2005. Coping with employment uncertainty: Strategies of employed and unemployed workers. J Occup Health Psychol 10:200209. Matheson K, Anisman H. 2003. Systems of coping associated with dysphoria, anxiety, and depressive illness: A multivariate prole perspective. Stress 6:223234. Matheson K, Skomorovsky A, Fiocco A, Anisman H. 2007. The limits of adaptive coping: Well-being and mood reactions to stressors among women in abusive dating relationships. Stress 10:7591. Mattlin JA, Wethington E, Kessler RC. 1990. Situational determinants of coping and coping effectiveness. J Health Soc Behav 31:103122. McCullough ME. 2000. Forgiveness as human strength: Theory, measurement, and links to well-being. J Soc Clin Psychol 19: 4355. McCullough ME, Rachal KC, Sandage SJ, Worthington Jr. EL, Brown SW, Hight TL. 1998. Interpersonal forgiving in close relations II: Theoretical elaboration and measurement. J Pers Soc Psychol 75:1586 1603. McCullough ME, Pargament KI, Thoresen CE. 2000. The psychology of forgiveness: History, conceptual issues, and overview. In: McCullough ME, Pargament KI, Thoresen CE, editors. Forgiveness: Theory, Research, and Practice. New York: Guilford Press. p 1 14. McCullough ME, Fincham FD, Tsang J. 2003. Forgiveness, forbearance, and time: The temporal unfolding of transgressionrelated interpersonal motivations. J Pers Soc Psychol 84: 540 557. Mullet E, Houdbine A, Laumonier S, Girard M. 1998. Forgivingness: Factor structure in a sample of young, middle-aged, and elderly adults. Eur Psychol 3:289297. Mullet E, Barros J, Frongia L, Usaie V, Shaghi SR. 2003. Religious involvement and the forgiving personality. J Pers 71:119. Neufeld J, McNamara J, Ertl M. 1999. Incidence and prevalence of dating partner abuse and its relationship to dating practices. J Interpers Viol 14:125137. Nicholls T, Dutton D. 2001. Abused committed by women against male intimates. In: Brothers BJ, editor. The abuse of men: Trauma begets trauma. New York: Haworth. p 41. Ntoumanis N, Biddle S. 1998. The relationship of coping and its perceived effectiveness to positive and negative affect in sport. Pers Indiv Diff 24:773788. Pargament KI, Koenig HG, Perez LM. 2000. The many methods of religious coping: Development and initial validation of the RCOPE. J Clin Psychol 56:519 543. Peacock EJ, Wong PT. 1990. The stress appraisal measure (SAM): A multidimensional approach to cognitive appraisal. Stress Med 6:227236. Peacock EJ, Wong PT. 1996. Anticipatory stress: The relation of locus of control, optimism, and control appraisals to coping. J Res Pers 30:204222. Pipes R, LeBov-Keeler K. 1997. Psychological abuse among college women in exclusive heterosexual dating relationships. Sex Roles 36:585603. Reed GL, Enright RD. 2006. The effects of forgiveness therapy on depression, anxiety, and posttraumatic stress for women after spousal emotional abuse. J Consult Clin Psychol 74:920929. Rhoades GK, McIntosh D, Wadsworth M, Ahlkvist J, Burwell R, Gudmunsen G, et al. 2007. Forgiving the September 11th terrorists: Associations with coping, psychological distress, and religiosity. Anx Stress Coping 20:109 128. Rye MS, Pargament KI. 2002. Forgiveness and romantic relationships in college: Can it heal the wounded heart? J Clin Psychol 58:419 441. Seybold KS, Hill PC, Neumann JK, Chi DS. 2001. Physiological and psychological correlates of forgiveness. J Psychol Christian 20:250259. Shackelford TK, Buss DM, Bennett K. 2002. Forgiveness or break-up: Sex differences in responses to a partners indelity. Cog Emot 16:299307. Snyder CR, Pulvers K. 2001. Dr. Seuss, the coping machine, and Oh, the places youll go. In: Snyder C, editor. Coping with stress. Effective people and processes. Oxford: Oxford University Press. p 3 29. Stanton AL, Franz R. 1999. Focusing on emotion: An adaptive coping strategy? In: Snyder CR, editor. The psychology of what works. New York: Oxford University Press. p 90. Stanton AL, Danoff-Burg S, Cameron CL, Bishop M, Collins CA, Kirk SB, et al. 2000. Emotionally expressive coping predicts psychological and physical adjustment to breast cancer. J Consult Clin Psychol 68:875882. Stone AA, Schwartz JE, Neale JM, Shiffman S, Marco CA, Hickcox M, et al. 1998. A comparison of coping assessed by ecological momentary assessment and retrospective recall. J Pers Soc Psychol 74:16701680. Straus MA, Hamby SL, Boney-McCoy S, Sugarman DB. 1996. The revised conict tactics scale (CTS2). J Fam Iss 17:283316. Strelan P, Covic T. 2006. A review of forgiveness process models and a coping framework to guide future research. J Soc Clin Psychol 25:10591085. Swan SC, Snow DL. 2006. The development of a theory of womens use of violence in intimate relationships. Viol Against Wom 12:10261045. Tennen H, Afeck G, Armeli S, Carney MA. 2000. A daily process approach to coping: Linking theory, research, and practice. Am Psychologist 55:626636. Tsang J, McCullough ME, Fincham FD. 2006. The longitudinal association between forgiveness and relationship closeness and commitment. J Soc Clin Psychol 25:448472. Waldron VR, Kelley DL. 2005. Forgiving communication as a response to relational transgressions. J Soc Pers Rel 22:723742. Wekerle C, Wolfe DA. 1999. Dating violence in mid-adolescence: Theory, signicance, and emerging prevention initiatives. Clin Psychol Rev 19:435 456. Witvliet CVO, McCullough ME. 2007. Forgiveness and health: A review and theoretical exploration of emotion pathways. In: Post SG, editor. Altruism and health: Perspectives from empirical research. Oxford: Oxford University Press. p 259276. Witvliet CVO, Ludwig TE, VanderLaan KL. 2001. Granting forgiveness or harboring grudges: Implications for emotions, physiology, and health. Psychol Sci 12:117123. Worthington Jr. EL, 2006. Forgiveness and reconciliation: Theory and application. New York: Brunner-Routledge. Worthington Jr. EL, Scherer M. 2004. Forgiveness is an emotionfocused coping strategy that can reduce health risks and promote health resilience: Theory, review, and hypotheses. Psychol Health 19:385405. Worthington Jr. EL, Wade NG. 1999. The psychology of unforgiveness and forgiveness and implications for clinical practice. J Soc Clin Psychol 18:385418. Worthington Jr. EL, Witvliet CVO, Pietrini P, Miller AJ. 2007. Forgiveness, health, and well-being: A review of evidence for emotional versus decisional forgiveness, dispositional forgivingness, and reduced unforgiveness. J Behav Med 30:291302. Ysseldyk R, Matheson K, Anisman H. 2007. Rumination: Bridging a gap between forgivingness, vengefulness, and psychological health. Pers Indiv Diff 42:15731584.