When Leaving Your Ex,Love Yourself

Abstract

Divorce is a highly stressful event, and much remains to be learned about the factors that promote psychological resilience when marriages come to an end. In this study, divorcing adults (N = 109) completed a 4-min stream-of-consciousness recording about their marital separation at an initial laboratory visit. Four judges rated the degree to which participants exhibited self-compassion (defined by self-kindness, an awareness of one’s place in shared humanity, and emotional equanimity) in their recordings. Judges evidenced considerable agreement in their ratings of participants’ self-compassion, and these ratings demonstrated strong predictive utility: Higher levels of self-compassion at the initial visit were associated with less divorce-related emotional intrusion into daily life at the start of the study, and this effect persisted up to 9 months later. These effects held when we accounted for a number of competing predictors. Self-compassion is a modifiable variable, and if our findings can be replicated, they may have implications for improving the lives of divorcing adults.

Keywords

divorce emotions psychological stress

Imagine the following situation: Your spouse decides your marriage is over. You knew your relationship was not going well, but now it is really over. The prospect of a divorce is not so bad for some people, but you are distraught and find yourself wondering how to put your life back together. What steps can you take to start the healing process?

Unfortunately, this situation is all too common: Nearly 2 million adults in the United States become divorced each year (Tejada-Vera & Sutton, 2010). Most adults handle the transition well (Amato, 2010; Mancini, Bonanno, & Clark, 2011), but for a small percentage of divorcees, marital separation is associated with lasting decreases in psychological well-being (Lucas, 2005) and an increased risk for physical health problems (Sbarra, Law, & Portley, 2011). For anyone experiencing divorce, a critical—perhaps the critical—question is how to recover over time.

When their marriage ends in divorce, many people turn a harsh light on themselves: “It was my fault—I should have acted differently”; “I wasn't good enough”; “I am not attractive enough.” These recriminations are essentially perceptions of causality, and it is well known that the more people focus their thoughts on regret and longing, the worse their outcomes (Emery, 1994; Hetherington & Kelly, 2002; Sbarra & Emery, 2008). Although researchers know a good deal about what happens when people become mired in negative appraisals of their marital separation, little is known about the correlates of adopting a different stance with respect to the end of marriage—one characterized by an empathetic and tender view of the self. In the study reported here, we examined whether self-compassion influences the course of psychological recovery following marital separation.

Self-compassion is an integrative construct that encompasses self-kindness (i.e., treating oneself with understanding and forgiveness), recognition of one’s place in shared humanity (i.e., acknowledgment that people are not perfect and that personal experiences are part of the larger human experience), and mindfulness (i.e., emotional equanimity and avoidance of overidentification with painful emotions; Neff, 2003a, 2003b). In distressing situations, people who are high in self-compassion tend to experience negative affect without becoming overwhelmed or mired in negative thoughts about their experiences; they view themselves and their actions empathetically and are able to see both the highs and the lows of life as part of the human experience. Self-compassion appears to be distinct from self-esteem (Leary, Tate, Adams, Batts Allen, & Hancock, 2007; Neff, 2003a); the latter construct is more highly correlated with hubris and inflated views of self-worth.

It is well established that positive emotions have an “undoing” effect on negative affect by promoting a greater range of behavioral action and broadening people’s perspective on available coping resources (Fredrickson, 2001). There is increasing interest in how self-compassion can reduce negative affect associated with stressful life events (Allen & Leary, 2010; Leary et al., 2007; Tugade & Fredrickson, 2004). Allen and Leary (2010) recently proposed that one reason people high in self-compassion cope well with stressful events is that they tend to rely on positive cognitive restructuring when thinking about difficult experiences. For example, using an Internet-based experience-sampling procedure, Leary et al. (2007) found that participants who were higher in self- compassion were significantly less likely to endorse negative thoughts (e.g., to think that their life was more “screwed up” than other people’s lives) with regard to their recent negative experiences. Compared with people low in self-compassion, people high in self-compassion tend to use more emotion-focused strategies and fewer avoidance-oriented strategies when dealing with personal setbacks (Neff, Hsieh, & Dejitterat, 2005).

Although self-compassion has emerged as an important construct for understanding well-being, most of what is known about individual differences in self-compassion has come from self-reports on the Self-Compassion Scale (SCS; Neff, 2003b). It is clear that self-compassion is a reliably measured subjective state; however, as are all evaluative self-reports, subjective reports of self-compassion are vulnerable to a variety of response biases (Paulhus & Vazire, 2007). Using observational assessments instead of self-reports may obviate these biases, and observational assessments have proven useful in a variety of social and clinical contexts (e.g., Mehl, 2006). In the current study, we examined how well judges agreed about the degree to which recently separated adults exhibited self-compassion when discussing their relationship history and divorce-related experiences.

An important task in research on divorce is identifying moderating variables that predict positive outcomes (Amato, 2010); we believe that the yet-unstudied construct of self-compassion may be one such variable. Asking people to speak directly about their relationship history and experiences with marital separation can provide rich data for examining variability in self-compassion, specifically with respect to how people view the end of their marriage. The ability to be kind to oneself in distressing circumstances, to view one’s divorce as part of the slings and arrows that challenge all people, and to experience the emotional pain of divorce without becoming entrenched in it appear to be vital for promoting well-being and protecting against lasting distress following divorce. But is self-compassion uniquely associated with positive outcomes following marital separation, or is the construct merely a correlate of feeling good in general?

Drawing on the literature we have just reviewed, we hypothesized (a) that judges would be able to reliably identify behavioral indicators of self-compassion in recently separated adults and (b) that self-compassion would moderate the course of these adults’ emotional adjustment over time, such that adults judged to be higher in self-compassion at the beginning of the study would exhibit greater declines in divorce-related distress over a 9-month follow-up period. We expected that this effect would hold after we accounted for self-reported indices of well-being (e.g., general mood, self-esteem, and optimism), traitlike variables that are generally associated with adjustment to divorce (e.g., attachment anxiety and avoidance), participants’ use of positive and negative emotion words when describing their divorce-related experiences, and the degree to which participants found talking about their separation to be emotionally difficult.

Method

Participants

The sample and procedures used in this study are described in detail elsewhere (Mason, Sbarra, & Mehl, 2010). Participants were 105 community-dwelling adults (38 men, 67 women; mean age = 40.4 years, SD = 10 years), who reported having been in a relationship with their former partner for more than 13.5 years (SD = 103.10 months). On average, participants reported having experienced a marital separation 3.8 months before entering the study (SD = 2.1 months).

Procedure

During an initial laboratory visit, participants were asked to mentally recall a detailed image of their former partner for 30 s; then, for 4 min, they spoke into a digital recorder in a stream-of-consciousness manner about their thoughts and feelings regarding their separation experience. Four trained judges who were unacquainted with the participants and all aspects of the study coded the audio files and rated each participant on modified versions of all items from the Self-Compassion Scale–Short Form (SCS-SF; Raes, Pommier, Neff, & Van Gucht, 2011). All judges were undergraduate research assistants who were blind to the our hypotheses but familiar with the literature on self-compassion; they were provided with detailed training and a scoring manual to aid them in scoring the items.

After the initial visit, participants returned to the laboratory for a 3-month follow-up assessment, followed by either a 6-month or a 9-month follow-up assessment. To decrease the burden on participants and reduce attrition, we used a planned-missingness design (see McArdle & Nesselroade, 1994). Eighty-five percent (n = 89) of the initial participants returned for the 3-month assessment, and 76% of participants were retained over the entire study period (6-month follow-up: n = 37; 9-month follow-up: n = 43). There were no significant differences in initial divorce-related distress among participants who completed the final (i.e., 6-month or 9-month) assessment and those who did not; however, participants who did not complete both follow-up assessments were judged to have evidenced significantly less self-compassion at the initial visit than did participants who completed all three assessments, t(103) = 2.25, p = .03.

Measures

All self-report measures demonstrated adequate internal consistency (αs = .75–.93), and all but the task rated-emotional-difficulty (TRED) scale were completed at study entry. Divorce-related psychological adjustment served as the outcome variable; observer-rated self-compassion served as the focal predictor variable; divorce-related demographics and emotional difficulty during the stream-of-consciousness task were entered into the model as covariates; and the remaining measures were entered into the model as competing predictors.

Outcome variable: divorce-related psychological adjustment

During the initial laboratory visit, we assessed participants’ psychological adjustment to marital separation using the 22-item Impact of Event Scale–Revised (IES-R; Weiss & Marmar, 1997); example items from this scale include “Any reminders brought back feelings about [the divorce]” and “I was jumpy and easily startled.” Higher scores on this scale reflect greater emotional intrusion of the separation into everyday experience, greater somatic hyperarousal in the aftermath of the separation, and a higher level of avoidance behaviors following the separation.

Focal predictor variable: observer-rated self-compassion

The four judges listened to each participant's 4-min audio recording about his or her relationship history and separation experience. Judges rated each participant's stream-of- consciousness recording using a modified version of the SCS-SF (Raes et al., 2011). The SCS-SF was developed using multiple samples and includes a single, higher-order factor of self-compassion. Scores on the SCS-SF are highly correlated with scores on the full SCS.

In the present study, all SCS-SF items were reworded from a first-person perspective to a third-person perspective; when possible, items were made to apply specifically to divorce-related experiences. For example, the SCS-SF item tapping mindfulness originally read, “When something painful happens, I try to take a balanced view of the situation”; it was reworded to read, “When describing something painful about the divorce, this person seems to take a balanced view of the situation.” One modified (reworded) self-kindness item read, “When describing hard times during their divorce, this person tries to give themselves the caring and tenderness they need,” and a modified common-humanity item read, “This person seems to accept their failings as part of the human condition.” Table 1 shows example excerpts from transcripts of stream-of-consciousness recordings that received high and low scores for each subscale of the SCS-SF.

Table 1.

Excerpts From Stream-of-Consciousness Recordings That Received High and Low Scores on Each Subscale of the Self-Compassion Scale–Short Form (Raes, Pommier, Neff, & Van Gucht, 2011)

Subscale	High score	Low score
Self-kindness	“Looking back you have to, um, take that in—take the best out of it and move on from there, for the best of, um, both parties. . . just forgive yourself and him for everything you did or didn’t do.”	“I don’t know how I managed to do this. It was all my fault, I pushed him away for some reason. I needed him so much, still need him. . . . What did I do? I know I did it all wrong.”
Mindfulness	“You feel guilty and hurt, but you just have to deal with reality. There are strong emotions, sure, but there’s just reality in the end.”	“I, uh, can’t stand that the thought of my selfishness drove her to that. I don’t know, uh, if we can ever get back together. . . it seems like there is too much hurt, there is too much water under the bridge, too many hurtful things that were said. I wish, wish, wish I could go back but I can’t.”
Common humanity	“. . . it is just something that, uh, happens these days, and I guess it is happening more often than not these days so, uh, that is what the situation is. . . and you tell yourself you’re not the only person to experience this.”	“I am so alone in this. I spent more time in my life as an adult with my husband than I have without. So you feel so alone without that person. I know I, um, I know I could have done things better, but I don’t understand why he doesn’t care.”

Judges scored each of the 12 modified SCS-SF items using the same 5-point Likert-type scale as used in the self-report inventory (from 1, almost never, to 5, almost always). The scores for all items were summed to create a composite score for each participant (M = 3.43, SD = 0.35). Higher composite scores indicated greater self-compassion.

We determined intercoder agreement by computing the intraclass correlation coefficient (ICC) for each of the 12 items and averaging across items. Judges demonstrated considerable agreement in identifying variability in the degree to which participants exhibited self-compassion in their stream-of-consciousness recordings, ICC(2, k) = .77. Judge-rated self-compassion did not differ significantly between men and women, t(103) = 1.62, p = .21.

Divorce-related demographics

Participants reported the length of the relationship prior to the separation, as well as the time (in months) since they had separated from their spouse.

Task rated emotional difficulty during the stream-of- consciousness recording

Immediately after completing the stream-of-consciousness recording, participants completed a TRED scale that consisted of four items (e.g., “How upsetting did you find this task?”). A high score indicated that the participant found the stream-of-consciousness task emotionally difficult and exerted considerable effort trying to control his or her emotions while completing the recording.

Beck Depression Inventory–II

We used the Beck Depression Inventory–II (BDI; Beck, Steer, & Brown, 1996) to assess depressed mood; higher scores on the inventory reflect greater mood disturbance (including sustained sadness, loss of appetite, irritability, and poor sleep) in the 2 weeks before the initial visit.

Emotion Regulation Questionnaire

Participants completed the 10-item Emotion Regulation Questionnaire (ERQ; Gross & John, 2003), a widely used measure designed to assess people’s propensity to use two emotion-regulation strategies: cognitive reappraisal and emotional suppression.

Experiences in Close Relationships–Revised Questionnaire

We used the Experiences in Close Relationships–Revised Questionnaire (ECR-R; Fraley, Waller, & Brennan, 2000) to assess attachment-related anxiety and avoidance (two dimensions that underlie individual differences in adult attachment patterns). The Anxiety scale captures hypervigilance to attachment-related themes within relationships (e.g., anxiety about rejection or abandonment in relationships). The Avoidance scale captures the degree to which people tend to minimize attachment-related behaviors, thoughts, and feelings.

Self-esteem and optimism

At the initial laboratory visit, participants completed divorce-specific, self-report versions of Rosenberg's Self-Esteem Scale (Rosenberg, 1965) and the Life Orientation Test–Revised (Scheier, Carver, & Bridges, 1994) as indices of divorce-related self-esteem and optimism. Each item began with “Since this separation. . .” (e.g., “Since this separation, I take a positive attitude toward myself and life”). High scores reflect high levels of self-esteem and optimism with respect to the respondent's marital-separation experience.

Emotion words in stream-of-consciousness recordings

All stream-of-consciousness recordings were transcribed by independent transcribers (not judges), and the transcribed text was analyzed using the Linguistic Inquiry and Word Count (LIWC; Pennebaker, Francis, & Booth, 2001) text-analysis program. The LIWC program uses an internal dictionary consisting of 74 standardized linguistic categories (e.g., personal pronouns and words reflecting positive and negative affect) to analyze each word in a text file; it then calculates the percentage of words in the text that fall into specific linguistic categories. For our purposes, we focused on the linguistic categories of positive emotion, which includes words reflecting positive feelings (e.g., happy, good, love, joy) and optimism (e.g., certainty, pride, win), and negative emotion, which includes words reflecting anxiety (e.g., nervous), anger (e.g., hate), and sadness (e.g., sad).

Results

Bivariate correlations among the study variables are displayed in Table 2. Participants’ composite self-compassion scores were significantly negatively correlated with both IES-R scores at study entry and measures assessing negative affect.

Table 2.

Bivariate Correlations Among Variables Assessed in the Study

Variable	1	2	3	4	5	6	7	8	9	10	11	12	13
1. IES-R score at initial assessment	1.0
2. Judge- rated self-compassion	−.57**	1.0
3. Time since marital separation	.05	−.13	1.0
4. Relationship length	.03	−.11	.09	1.0
5. Self-reported self-esteem	−.61**	.52**	−.04	−.10	1.0
6. Self-reported optimism	−.58**	.47**	−.12	−.14	.72**	1.0
7. BDI score	.68**	−.59**	.05	.04	.78**	.69**	1.0
8. Attachment anxiety	.49**	−.32**	.06	−.00	.51**	.51**	.48**	1.0
9. Attachment avoidance	.03	−.01	.03	.09	.13	.16	.09	.40**	1.0
10. ERQ: reappraisal	−.26**	.17	−.25**	.04	−.33**	.39**	−.35**	−.25**	−.10	1.0
11. ERQ: suppression	.02	.01	.00	−.03	.18	.26**	.20*	.11	.27**	−.23*	1.0
12. TRED scale	.43**	−.51**	.14	.07	.45**	.39**	.48**	.40*	.11	−.26**	.18	1.0
13. LIWC: positive- emotion words	.00	.01	−.00	.16	.02	−.04	.05	.08	−.07	.07	−.16	.26**	1.0
14. LIWC: negative- emotion words	.27**	−.41**	.10	−.11	.12	.17	.34*	.05	−.12	−.16	−.08	.33**	.13

Note: IES-R = Impact of Event Scale–Revised (Weiss & Marmar, 1997); BDI = Beck Depression Inventory–II (Beck, Steer, & Brown, 1996); ERQ = Emotion Regulation Questionnaire (Gross & John, 2003); TRED = task rated emotional difficulty; LIWC = Linguistic Inquiry Word Count text-analysis program (Pennebaker, Francis, & Booth, 2001).

*p < .05. **p < .01.

Using multilevel modeling, we found systematic declines in IES-R scores over time, including a significant linear decrease, b = −0.87, SE = 0.09, p < .001, as well as a significant quadratic increase, b = 0.21, SE = 0.03, p < .001. Together, the linear and quadratic effects reflect a systematic decrease in IES-R scores that slowed over the course of the 9-month assessment period; as time went on, the quadratic component of change increased IES-R scores at the final assessment.

In our first set of conditional models, we examined the effect of self-compassion on IES-R scores at the start of the study and whether change in IES-R scores was predicted by the interactions of self-compassion with time (i.e., initial assessment, 3-month follow-up, 6-month follow-up, and 9-month follow-up) and time-squared; these models included no control variables. Consistent with the cross-sectional correlation observed at the first assessment, results showed that participants judged to be higher in self-compassion reported less divorce-related emotional distress at study entry, b = −1.35, SE = 0.15, p < .001. Both two-way interactions (Self-Compassion × Time and Self-Compassion × Time²) were significant: Adults lower in self-compassion evidenced faster linear decreases in IES-R scores over time, b = 0.85, SE = 0.30, p = .005, and faster quadratic increases in IES-R scores over time, b = −0.22, SE = 0.10, p = .03. Thus, as time went on, participants’ IES-R scores increased, and this effect was stronger for participants judged to be lower in self-compassion (see Fig. 1).

Fig. 1.

Score on the Impact of Event Scale–Revised (IES-R; Weiss & Marmar, 1997) as a function of time of assessment and judge-rated self-compassion. Self-compassion was measured using the Self-Compassion Scale–Short Form (Raes, Pommier, Neff, & Van Gucht, 2011). Results are shown for participants judged to be high in self-compassion (1 SD above the mean), average in self-compassion, and low in self-compassion (1 SD below the mean).

We next sought to determine whether accounting for relevant covariates and competing predictors eliminated the effects of interest (see Table 3). These effects changed very little after we accounted for the 12 covariates and competing predictors. Thus, although judge-rated self-compassion correlated positively with self-reported self-esteem and optimism and correlated negatively with many of the variables associated with affective distress (see Table 2), none of these variables accounted for the initial association between IES-R scores and self-compassion or the interactive effect of self-compassion and time on IES-R scores.

Table 3.

Multilevel Model Results Predicting Changes in IES-R Scores Over the 9-Month Follow-Up Period

Parameter	b	SE b	p
Intercept	1.48	0.05	< .001
Time	−0.84	0.10	< .001
Time²	0.20	0.03	< .001
Judge-rated self-compassion	−0.61	0.17	.001
Time since marital separation	−0.033	0.02	.05
Relationship length	−0.00	0.00	.52
Self-reported self-esteem	−0.03	0.12	.79
Self-reported optimism	−0.11	0.06	.04
BDI score	0.02	0.007	.008
Attachment anxiety	0.08	0.04	.06
Attachment avoidance	−0.04	0.05	.35
ERQ: reappraisal	0.01	0.03	.70
ERQ: suppression	−0.02	0.04	.42
TRED scale	0.05	0.03	.10
LIWC: positive-emotion words	−0.05	0.03	.14
LIWC: negative-emotion words	0.01	0.04	.74
Self-Compassion × Time interaction	0.76	0.27	.006
Self-Compassion × Time² interaction	−0.19	0.09	.03

Note: All predictor variables were grand-mean centered prior to being entered in the model. IES-R = Impact of Event Scale–Revised (Weiss & Marmar, 1997); BDI = Beck Depression Inventory–II (Beck, Steer, & Brown, 1996); ERQ = Emotion Regulation Questionnaire (Gross & John, 2003); TRED = task rated emotional difficulty; LIWC = Linguistic Inquiry Word Count text-analysis program (Pennebaker, Francis, & Booth, 2001). Akaike’s information criterion for this model was 488.55.

Because the interactions between self-compassion and time are focal aspects of the model presented in Table 3, we explored the possibility that the covariates and competing predictors interacted with time in a similar way, thus reducing the significant interaction between self-compassion and time. Because of the large number of covariates and competing predictors, we used a forward-selection stepwise procedure to determine if any of their interactions with time or time-squared would eliminate the significant Self-Compassion × Time and Self-Compassion × Time² effects. If the interaction between a covariate or competing predictor and time or time-squared was significant, it was retained in the model. Of these 24 interactions, only the Attachment Anxiety × Time interaction was significant, b = −0.06, SE = 0.03, p = .01. Participants who reported higher levels of attachment anxiety evidenced slower linear decreases in IES-R scores over time. This effect did not eliminate the significance of the Self-Compassion × Time and Self-Compassion × Time² interactions; therefore, our final model included three significant interaction terms.

Finally, we conducted an end-point analysis by recentering time around the 9-month assessment. The main effect of self-compassion remained significant at the final assessment, b = −0.32, SE = 0.15, p < .03.¹ Nine months after their initial assessment, participants whose stream-of-consciousness recordings were judged to exhibit higher levels of self-compassion evidenced significantly less divorce-related emotional distress.

Discussion

Because the incidence of divorce is high, it is critical to understand the factors that promote resilience and enhance well-being when marriage dissolves. We found that self-compassion can be assessed observationally (with high interrater agreement) using adults’ verbal accounts of their relationship history and separation experience. Higher levels of self-compassion at the initial laboratory visit were associated with less divorce-related distress up to 9 months later. Over time, participants judged to be lower in self-compassion evidenced significantly faster rates of linear decline in their IES scores, but these participants also evidenced significantly faster rates of quadratic increase in these scores toward the end of the study period. We tested whether the effects of self-compassion would be eliminated when a wide range of covariates and competing predictors was included in the analysis (cf. Miller & Chapman, 2001), but the effects remained robust after we accounted for relationship-specific demographics, self-reported mood states, habitual patterns of emotion regulation, attachment styles, self-reported emotional difficulty during the stream-of-consciousness task, and the positive and negative emotion words participants used during the stream-of-consciousness task. Thus, self-compassion evidenced a high degree of utility for predicting positive adjustment to marital separation.

From these findings, a critical question emerges: To what degree can adults experiencing marital separation or divorce become self-compassionate? Self-compassion is believed to be a teachable skill (see Neff, 2011), and the literature on loving-kindness interventions (e.g., Fredrickson, Cohn, Coffey, Pek, & Finkel, 2008), compassion-focused interventions (e.g., Gilbert & Procter, 2006), and mindfulness-based interventions (e.g., Baer, 2003; Hofmann, Sawyer, Witt, & Oh, 2010; Shapiro, Brown, & Biegel, 2007) is growing quickly. Together with experimental research demonstrating that self- and other-forgiveness can speed the course of recovery from depressive symptoms after divorce (Rye et al., 2005), our findings suggest that it may be advisable to encourage divorcing adults to cultivate self-compassion. Basic approaches to cultivating self-compassion could include noticing and accepting negative thoughts about the end of one’s marriage (without becoming mired in a cycle of self-recrimination), cultivating self-kindness even while experiencing emotional pain (e.g., moving from regret and self-recrimination to self-forgiveness), and acknowledging that difficult experiences, including divorce, are part of the ups and downs of the human experience.

To the extent that these processes induce positive mood states, divorcing adults may find opportunities to grow and even flourish from the experiences surrounding the end of their marriage (see Tashiro, Frazier, & Berman, 2006). Thus, self-compassion may serve to disarm potentially noxious emotional states and, like other positive emotions, to exert an undoing effect (Fredrickson, 2001) on the negative mood states that can arise in the aftermath of difficult social transitions. People high in self-compassion may feel the pain of marital separation, but they avoid ruminating about their negative mental states, punishing themselves for real or perceived transgressions, and wallowing in their isolation and loneliness. Although future research is needed to determine the precise mechanisms linking self-compassion and positive adjustment to divorce, our results indicate that self-compassion serves a powerful protective function when marriage comes to an end.

In evaluations of the role of positive emotions in the context of adjustment to a stressful life event, care must be taken to avoid potentially tautological conclusions. For example, one possible criticism of the current research is that people who felt better at the start of the study would be expected to do better over time regardless of how the predictor or outcome variables were measured. Our approach to data analysis accounted for this concern: If judge-rated self-compassion merely reflected positive affect or the lack of negative affect, the effects of self-compassion would have been eliminated by the covariates and competing predictors, but including competing predictors in our models did not eliminate the effects of self-compassion at the start of the study or either of the interactions between self-compassion and time. Of course, care must be taken in interpreting the self-compassion variable and its interactions with time; it is most accurate to conclude that in the fully controlled model, participants who were judged to have more self- compassion than would be expected after accounting for relevant covariates and competing predictors evidenced the least distress at study entry, the slowest rate of linear decline, and the slowest rate of quadratic increase over time.

The durability of the effect of self-compassion was especially notable. At the 9-month assessment, the IES-R scores of participants whose judge-rated self-compassion scores were 1 standard deviation below the mean increased by 0.32 points, on average, relative to the initial assessment; this was an increase of nearly 0.5 standard deviation on the IES-R. Considering the role of attachment anxiety in our findings illustrates the significance of the effects of self-compassion. Attachment anxiety was strongly negatively associated with adjustment to divorce (for similar findings, see Birnbaum, Orr, Mikulincer, & Florian, 1997; Davis, Shaver, & Vernon, 2003) but was not uniquely associated with the IES-R intercept at the initial assessment (p = .06) or at the end of the study (p = .12). Thus, compared with a construct that has important implications for understanding well-being following social separations, self-compassion appears to be at least as useful for predicting future adjustment.

The findings from this study should be considered in light of its limitations. A primary limitation of this study is that it did not include a self-report measure of self-compassion, and it is difficult to determine how judge-rated self-compassion compares with subjective reports of the same construct. The planned-missingness design precluded assessments from all participants at all time points; although this design feature may have attenuated attrition over the 9-month follow-up period, it would have been ideal to include measurements of all participants at all assessments. Finally, the relatively small number of men in the sample relative to women prevented us from testing for whether the association between self-compassion and adjustment to divorce differed by sex.

This study is among the first to assess self-compassion observationally, and our findings contribute to a deeper understanding of the factors that may promote resilience in the face of marital separation and divorce.

Footnotes

Acknowledgements

The authors are grateful to Aida DeJonghe, Lauren Lee, Lindsey Bupp, Ashley Mason, and Rita Law for their assistance with this study, and to Sara Algoe, who provided consultation on the assessment of positive emotions in this study.

The authors declared that they had no conflicts of interest with respect to their authorship or the publication of this article.

This research was supported by Grant No. 028454 from the National Institute of Aging, Grant No. 074637 from the National Institute of Mental Health, and Grant No. 0919525 from the National Science Foundation to David A. Sbarra, and by a grant from the University of Arizona Honors College to Hillary L. Smith.

Notes

References

Allen

Leary

M.R.

(2010). Self-compassion, stress, and coping. Social & Personality Psychology Compass, 4, 107–118.

Amato

P. R.

(2010). Research on divorce: Continuing trends and new developments. Journal of Marriage and Family, 72, 650–666.

Baer

R. A.

(2003). Mindfulness training as a clinical intervention: A conceptual and empirical review. Clinical Psychology: Science and Practice, 10, 125–143.

Beck

A. T.

Steer

R. A.

Brown

G. K.

(1996). The Beck Depression Inventory–II. San Antonio, TX: Harcourt Assessment.

Birnbaum

G. E.

Orr

Mikulincer

Florian

(1997). When marriage breaks up: Does attachment contribute to coping and mental health? Journal of Social and Personal Relationships, 14, 643–654.

Davis

Shaver

P. R.

Vernon

M. L.

(2003). Physical, emotional, and behavioral reactions to breaking up: The roles of gender, age, environmental involvement, and attachment style. Personality and Social Psychology Bulletin, 29, 871–884.

Emery

R. E.

(1994). Renegotiating family relationships: Divorce, child custody, and mediation. New York, NY: Guilford Press.

Fraley

R. C.

Waller

N. G.

Brennan

K. A.

(2000). An item response theory analysis of self-report measures of adult attachment. Journal of Personality and Social Psychology, 78, 350–365.

Fredrickson

B. L.

(2001). The role of positive emotions in positive psychology: The broaden-and-build theory of positive emotions. American Psychologist, 56, 218–226.

10.

Fredrickson

B. L.

Cohn

M. A.

Coffey

K. A.

Pek

Finkel

S. M.

(2008). Open hearts build lives: Positive emotions, induced through loving-kindness meditation, build consequential personal resources. Journal of Personality and Social Psychology, 95, 1045–1062.

11.

Gilbert

Procter

(2006). Compassionate mind training for people with high shame and self-criticism: Overview and pilot study of a group therapy approach. Clinical Psychology & Psychotherapy, 13, 353–379.

12.

Gross

J. J.

John

O. P.

(2003). Individual differences in two emotion regulation processes: Implications for affect, relationships, and well-being. Journal of Personality and Social Psychology, 85, 348–362.

13.

Hetherington

E. M.

Kelly

(2002). For better or for worse: Divorce reconsidered. New York, NY: Norton.

14.

Hofmann

S. G.

Sawyer

A. T.

Witt

A. A.

(2010). The effect of mindfulness-based therapy on anxiety and depression: A meta-analytic review. Journal of Consulting and Clinical Psychology, 78, 169–183.

15.

Leary

M. R.

Tate

E. B.

Adams

C. E.

Batts Allen

Hancock

(2007). Self-compassion and reactions to unpleasant self- relevant events: The implications of treating oneself kindly. Journal of Personality and Social Psychology, 92, 887–904.

16.

Lucas

R. E.

(2005). Time does not heal all wounds: A longitudinal study of reaction and adaptation to divorce. Psychological Science, 16, 945–950.

17.

Mancini

A. D.

Bonanno

G. A.

Clark

G. E.

(2011). Stepping off the hedonic treadmill: Individual differences in response to major life events. Journal of Individual Differences, 32, 144–152.

18.

Mason

A. E.

Sbarra

D. A.

Mehl

M. R.

(2010). Thin-slicing divorce: Thirty seconds of information predict changes in psychological adjustment over 90 days. Psychological Science, 21, 1420–1422.

19.

McArdle

J. J.

Nesselroade

J. R.

(1994). Using multivariate data to structure developmental change. In Cohen

S. H.

Reese

H. W.

(Eds.), Life-span developmental psychology: Methodological contributions (pp. 223–267). Hillsdale, NJ: Erlbaum.

20.

Mehl

M. R.

(2006). The lay assessment of subclinical depression in daily life. Psychological Assessment, 18, 340–345.

21.

Miller

G. A.

Chapman

J. P.

(2001). Misunderstanding analysis of covariance. Journal of Abnormal Psychology, 110, 40–48.

22.

Neff

K. D.

(2003a). The development and validation of a scale to measure self-compassion. Self and Identity, 2, 223–250.

23.

Neff

K. D.

(2003b). Self-compassion: An alternative conceptualization of a healthy attitude toward oneself. Self and Identity, 2, 85–101.

24.

Neff

K. D.

(2011). Self-compassion. Stop beating yourself up and leave insecurity behind. New York, NY: William Morrow.

25.

Neff

K. D.

Hsieh

Y.-P.

Dejitterat

(2005). Self-compassion, achievement goals, and coping with academic failure. Self and Identity, 4, 263–287.

26.

Paulhus

D. L.

Vazire

(2007). The self-report method. In Robins

R. W.

Fraley

R. C.

Krueger

R. F.

(Eds.), Handbook of research methods in personality psychology (pp. 224–239). New York, NY: Guilford Press.

27.

Pennebaker

J. W.

Francis

M. E.

Booth

R. J.

(2001). Linguistic Inquiry and Word Count (LIWC): LIWC 2001. Mahwah, NJ: Erlbaum.

28.

Raes

Pommier

Neff

Van Gucht

(2011). Construction and factorial validation of a short form of the Self-Compassion Scale. Clinical Psychology & Psychotherapy, 18, 250–255.

29.

Rosenberg

(1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press.

30.

Rye

M. S.

Pargament

K. I.

Pan

Yingling

D. W.

Shogren

K. A.

Ito

(2005). Can group interventions facilitate forgiveness of an ex-spouse? A randomized clinical trial. Journal of Consulting and Clinical Psychology, 73, 880–892.

31.

Sbarra

D. A.

Emery

R. E.

(2008). Deeper into divorce: Using actor-partner analyses to explore systemic differences in coparenting conflict following custody dispute resolution. Journal of Family Psychology, 22, 144–152.

32.

Sbarra

D. A.

Law

R. W.

Portley

R. M.

(2011). Divorce and death: A meta-analysis and research agenda for clinical, social, and health psychology. Perspectives on Psychological Science, 6, 454–474.

33.

Scheier

M. F.

Carver

C. S.

Bridges

M. W.

(1994). Distinguishing optimism from neuroticism (and trait anxiety, self-mastery, and self-esteem): A reevaluation of the life orientation test. Journal of Personality and Social Psychology, 67, 1063–1078.

34.

Shapiro

S. L.

Brown

K. W.

Biegel

G. M.

(2007). Teaching self-care to caregivers: Effects of mindfulness-based stress reduction on the mental health of therapists in training. Training and Education in Professional Psychology, 1, 105–115.

35.

Tashiro

Frazier

Berman

(2006). Stress-related growth following divorce and relationship dissolution. In Fine

M. A.

Harvey

J. H.

(Eds.), Handbook of divorce and relationship dissolution (pp. 361–384). Mahwah, NJ: Erlbaum.

36.

Tejada-Vera

Sutton

P. D.

(2010). Births, marriages, divorces, and deaths: Provisional data for 2009 (National Vital Statistics Reports No. 58). Hyattsville, MD: National Center for Health Statistics.

37.

Tugade

M. M.

Fredrickson

B. L.

(2004). Resilient individuals use positive emotions to bounce back from negative emotional experiences. Journal of Personality and Social Psychology, 86, 320–333.

38.

Weiss

D. S.

Marmar

C. R.

(1997). The Impact of Event Scale–Revised. In Wilson

J. P.

Keane

T. M.

(Eds.), Assessing psychological trauma and PTSD (pp. 399–411). New York, NY: Guilford Press.