INTRODUCTION Religious coping has been studied extensively in many groups in the United States (Pargament, Koenig, & Perez, 2000), with the exception of Buddhists. However research is needed in this area because it is estimated that there are anywhere from 600,000 to 1.8 million Buddhists in the United States (Smith, 2000). In a qualitative study done by Phillips et al. (2007), 24 Buddhist participants in the US reported that lovingkindness was one of seven major beneficial coping mechanisms that they used. Lovingkindness means having compassion for one’s self and others (Chen 2006) and includes three components: doing nice acts for others (morality), viewing things from another’s perspective (wisdom), and using nonjudgmental techniques of awareness (Phillips et al., 2007). Most studies have examined lovingkindness in nonBuddhist samples. Carson et al (2004) found that couples who went through a relationship enhancement program featuring various forms of lovingkindness meditations had significant improvements in relationship happiness, relationship stress, stress coping efficacy, and overall stress. In another study, Neff (2004) found that compassion, as measured by the Self Compassion Scale, was linked to more adaptive coping strategies, greater life satisfaction, social connectedness and emotional intelligence. Participants who had higher compassion scores reported considerably less maladaptive coping strategies, depression, and anxiety. Though lovingkindness appears to be positively correlated with healthy outcomes, currently there is no measure of lovingkindness as a form of Buddhist coping. The present study‘s goal is to create a reliable and valid measure for Buddhist coping. It is hypothesized that items dealing with lovingkindness will form their own subscale, and that this coping strategy will be positively correlated with healthier outcomes from stress and general life satisfaction.
The participants in this study consisted of 550 Buddhists 43% with 87% originally from the United States. The 13% participants not originally from the United States had been in the country on average 17.4 years (SD=15.2). Participants ranged in age from 18 to 75 (M= 45; SD=13.6). The majority of the sample 86% was European-American, with the largest minority Asian-American 7%. The most commonly identified Buddhist sect to which participants belonged was Vajrayana 40%, followed by Mahayana Buddhism 28%, then Theraveda Buddhism 24%. Participants had practiced Buddhism for an average of 12.4 years (SD=10.6). Participants reported they were ‘moderately spiritual’ (M=3.42, SD=.716) on a four-point rating scale.
Participants completed a Buddhist coping measure that was created from a review of the literature and participant responses from a qualitative study which examined how Buddhists use their religion to cope with stress (Phillips et al., under review). The Buddhist coping measure was comprised of 18 subscales with 5-8 items per subscale for a total of 95 items, including five items dealing with lovingkindness (e.g., “tried to be compassionate towards others”). Originally each of the 18 subscales had 10 items but after being reviewed and categorized by 10 experienced Buddhists, only items that had been placed in the correct subscale by at least 8 of the 10 Buddhists were kept. Participants in the present study were asked to consider a stressful life event they were currently experiencing, and answer how much they were presently using each item to deal with that stressor. Participants completed the General Outcomes Scale (Lazarus & Folkman, 1984), which has demonstrated good reliability and validity (Lazarus & Folkman, 1984). This scale contains five items, rated on a five-point Likert scale from 1 (“strongly disagree”) to 5 (“strongly agree”).
After receiving IRB approval participants were recruited through the online social website Facebook or by email. Participants found through Facebook were sent a message through their membership to a Buddhist Group. The email addresses for sangha leaders across the United States were obtained through www.buddhanet.net, and the leaders were asked to mention the survey to their fellow Buddhists. An online survey was created on surveymonkey.com, using the measures noted previously, as well as basic demographic questions. Participants who wanted to be eligible for a drawing were directed to a second survey on surveymonkey.com, where they provided their email. Three participants were randomly chosen, and sent a $50 gift card to a place of their choice.
RESULTS A Principal Components Analysis was conducted with a Promax rotation because the factors were expected to correlate (Pett, Lackey, & Sullivan, 2003). Given the high number of cases that had at least one Buddhist coping item missing, cases were excluded by replacing missing values with the mean item score. Participants who skipped five or more of the 95 Buddhist coping items were eliminated (9 cases were dropped). The Kaiser-Meyer-Olkin statistic for sampling adequacy was adequate at .928, and Bartlett’s Test of Sphericity was significant (÷2 = 25046.5, p < .01). The scree test revealed a cutoff of 3 factors, and the K-G rule would leave a total of 21 factors with eigenvalues over 1. Given the limitations of these two tests, statisticians have noted the importance of using theory to determine where to cut off the factor structure (Pett, Lackey, & Sullivan, 2003). The first 14 factors were interpretable when using the factor loadings minimum standard for oblique rotations of 0.45 and therefore kept. One item cross-loaded on the first 14 factors, and was eliminated. The final seven factors with eigenvalues over 1.0 were uninterpretable (i.e., multiple cross-loadings with previous factors, high load items with no particular pattern), or similar theoretically to the first 14 factors. Factor 10 of the factor analysis appeared to represent lovingkindness, containing three items with factor loadings greater than 0.45 (three of the five lovingkindness items in the survey). Table 1 lists the factor loadings for the these items. Five hundred thirty-eight of the 550 participants completed the three items in Factor 10, with a mean of 9.5 (SD=1.9) for the total subscore and 3.2 for each individual item, falling between ratings of ‘Sometimes’ and ‘Quite a Bit.’ Descriptive statistics for all major variables can be found in Table 2. Hierarchical regressions were conducted to determine the ability of lovingkindness to predict adjustment to stress over and above demographic and general religious variables (See Table 3 for results). To correct for Type 1 error, a Bonferroni correction was conducted, meaning for ∆R² to be significant, p had to be less than .008 (.05/6). This was because there were originally six dependent variables – general outcomes and five other measures of adjustment to life stress. However, this study will only focus on the results of general outcomes from the stressor. Lovingkindness predicted general outcomes from a stressful life event (∆R²= .103, p<.008), over and above demographic (e.g., age, gender) and general religious variables (e.g., years as a practicing Buddhist). Those participants who reported engaging in more lovingkindness to deal with a stressor reported better outcomes from the stressful event (â = 0.34, p < .008).
DISCUSSION The goal of the present study was to create a reliable and valid measure of Buddhist coping, specifically to examine whether lovingkindness was a unique form of religious coping, and whether it helped Buddhists cope with stress. Lovingkindness items did form their own factor in factor analysis, and showed evidence of adequate internal reliability. Lovingkindness was a far better predictor of general outcomes from stress than general religious variables and demographic variables. Individuals who spent more time using lovingkindness techniques when facing a particular stressor reported better outcomes from the event. The present study did not use a true experimental design therefore a causal relationship cannot be determined, but a positive correlation was found. Another limitation of this study is that due to the fact that the survey was only offered online our participants only include people who are computer literate and who have access to a computer. Also, the study only used self-report instruments, therefore a monomethod bias could exist. Despite these limitations, this study significantly adds to the research on religious coping, providing the first known survey of Buddhist coping. Clinicians could use such a measure to help them understand how their Buddhist clients use religion to cope with stress. For instance, this might help patients understand the importance of using lovingkindness towards themselves and others to better their outcomes from a stressor. In terms of research implications, a longitudinal study incorporating the Buddhist coping measure could demonstrate that religious coping in Buddhists predicts outcomes at a later point in time.
REFERENCES Carson, J., Carson, M., Gil, K., & Baucom, D. (2004). Mindfulness-based relationship enhancement. Behavior Therapy, 35, 471-494.Chen, Y-H. (2006). Coping with suffering: The Buddhist perspective. Handbook of Multicultural Perspectives on Stress and Coping (73-89). New York, NY: Springer.Lazarus, R.S., & Folkman, S. (1984). Stress, appraisal, and coping. New York: Springer.Neff, K. (2004). Self-compassion and psychological well-being. Constructivism in the Human Sciences, 9, 27-37. Pargament, K.I., Koenig, H.G., & Perez, L. (2000). The many methods of religious coping. Journal of Clinical Psychology, 56, 519-543.Phillips, R.E., Colvin, S., Abarr, A., Dunn, M, & Reed, A. (under review). A qualitative study of Buddhist forms of religious coping. Journal for the Scientific Study of Religion. Smith, T. W. 2002. Religious diversity in America. Journal for the Scientific Study of Religion, 41, 577-585.