Introduction

During the last 15 years there has been a proliferation of interest and research in the area of religion and psychology. Despite the fact that the operational definition of the construct of religion has proven extremely difficult, there has emerged a consistent body of results which indicate a strong relationship between religion and positive adjustment (Crawford et al. 1989). Furthermore, Myers and Diener (1995) reported a positive relationship between religion and well-being across cross-cultural samples. Additionally, the relationship between increased religious involvement and positive psychological adjustment has also been observed in African-Americans (Handal et al. 1989) and adolescents (Mosher and Handal 1997). Furthermore, a review of studies conducted by Koeing (2001) found positive associations between religious beliefs or practices and indices of psychological well-being.

In addition to the relationship between religion and positive adjustment, there is a line of research that has consistently demonstrated a statistically significant and clinically meaningful relationship between religion and negative psychological adjustment as measured by epidemiological measures, such as the General History Questionnaire (GHQ) and the Langner Symptom Survey (LSS); (Fabricatore et al. 2004; Handal et al. 1989; Mosher and Handal 1997). Specifically, participants who were low on a measure of religion obtained a mean score above the cutoff score (4) used for the identification of individuals in distress and in need of treatment, while participants who were high on a measure of religion obtained a mean score below the cutoff score used for the identification of individuals in distress and in need of treatment.

More recently, considerable research has emerged on the relationship between religious coping and psychological adjustment. Fabricatore (2002) stated that the role of religious coping in understanding the relationship between religion and mental health, according to Pargament (1997), is as an essential one. Pargament hypothesized that the use of religious coping in stressful situations will facilitate positive adjustment. Pargament et al. (1988) developed the Religious Problem Solving Scale, which contains three major approaches to religious coping with adversity: collaborative, deferring, and self-directing. The collaborative style reflects the joint responsibility for problem solving by God and the individual, while the deferring style implies placing all responsibility for problem solving on God while passively waiting to receive solutions. The self-directing approach emphasizes the individual’s personal responsibility and active role in problem solving with God playing a passive role. Hathaway and Pargament (1990) demonstrated that both self-directing and collaborative problem-solving styles have been linked to greater general psychological competence, while the deferring religious coping method has been related to lower levels of psychological resourcefulness. However, in several studies, the self-directing approach has also been associated with negative outcomes, such as anxiety and depression (Bickle et al. 1998; Schaefer and Gorsuch 1991). Specifically, Bickle et al. (1998) found an increase in depressive affect under conditions of high stress with the reported use of the self-directing religious coping style. The use of the collaborative coping style, on the other hand, produced a decrease in depression under the same conditions.

Research involving the construct of religious coping has typically focused on the use of religious coping in a specific situation; for example, tragic events such as the Oklahoma City Bombing and September 11th (Pargament et al. 1998). The results of these studies demonstrated a positive relationship between religious coping and adjustment in a specific situational context. However, the current study differs in that it examines religious coping in a dispositional context.

While both religion and religious coping have been investigated in terms of their direct relationship to psychological adjustment, the role of religious coping as a moderator between religion and both positive and negative psychological adjustment has been investigated only recently. As cited by Fabricatore (2000), there are distinct religious variables that have been found to serve as a buffer against the detrimental effects of stressors on well-being such as faith (Ellison 1991) and spiritual life integration (Fabricatore et al. 2000).

The purpose of the current study was to examine the relationships between and among religion, religious coping, and positive and negative aspects of psychological adjustment. The second purpose of this study was to ascertain empirically the role of religion and religious coping on psychological adjustment. More specifically, this study investigated whether the four religious coping styles of Self-Directing, Deferring, Collaborative, and Turning to Religion would significantly moderate the relationship between religion and psychological adjustment.

Methods

Participants

Participants in this study were 189 (45 males and 144 females) college undergraduates from an urban, Midwestern Catholic University. Five percent were age 18, 32.3% were 19, 29.1% were 20, 21.2% were 21, 6.9% were 22, and 5.3% were 23 or older. Participant mean age was 20 (SD = 1.24) and they were mostly Caucasian (83.6%). Other ethnicities represented were as follows: 5.8% African-Americans, 3.2% Asian, 2.6% Hispanic, and 2.6% other. Participants were primarily of middle to upper-middle socioeconomic status as determined by their mother and father’s reported income (M = $50,001–60,000; $60,001–70,000, respectively) and education levels (M = College Graduate; College, respectively). Participants represent college undergraduates of all year levels: 21.7% freshman, 32.3% were sophomore, 21.7 were juniors, 21.7% were seniors, and 2.1% were fifth year seniors.

In terms of religion, 60.8% were Catholic, 12.7% were Protestant, 3.2% were Hindu, 3.2% were Atheist, 3.2% were Agnostic, 2.1% were Muslim, 0.5% were Jewish, 7.9% were other, and 5.8% reported no preferences. In terms of self-reported religiousness, 5.8% reported being extremely religious, 16.4% reported being quite religious, 40.7% reported being moderately religious, 23.8% reported being a little religious, and 12.2% reported being not at all religious. With regard to self-reported spirituality, 7.9% reported being extremely spiritual, 24.9% reported being quite spiritual, 41.3% reported being moderately spiritual, 16.9% reported being a little spiritual, and 7.9% reported being not at all spiritual. In terms of the participant’s description of their own religiousness and spirituality, the following was reported: 67.2% were spiritual and religious, 16.4% were spiritual but not religious, 6.3% were religious but not spiritual, and 9.5% were neither religious nor spiritual.

Measures

Personal Religiosity Inventory (PRI)

The PRI (Lipsmeyer 1984) is a 45-item, nine scale, multidimensional measure of religiosity. The scales measure personal prayer (PRP); ritual attendance (RA); non-ritual, church-related activity (NRA); belief in God (BLFGOD); belief in an afterlife (AFTLIFE); perceived congruence of a person’s religious beliefs with their attitudes on social and moral issues (RSM); the extent to which an individual’s ideas about religion guide their philosophy or way of life (IDEO); the subjective experience of feeling close to God (CLOSEGOD); and integration or the extent to which persons perceive that their relationship with God influences their cognition, affect, and behavior (INT). Most of the items use a 6-point Likert response format; however, others use a multiple choice or yes/no format.

According to Lipsmeyer, test–retest reliability coefficients over a 1-week period were between .83 and .97 for the nine scales in an adult population. Additionally, Lipsmeyer found that the PRI had high concurrent validity; religious professionals (e.g., priests, ministers, nuns) scored significantly higher on all scales than the general public. Also, Lipsmeyer reported that atheists, agnostics, and those with no religious preference scored significantly lower than other major religious groups. Lipsmeyer reported that each subscale of the PRI correlated highest with integration (INT), and that it had the highest stability coefficient and was the best single measure of religion. Consequently, INT was used as the overall measure of religion in this study.

The Religious Problem Solving Scales (RPSS)

The RPSS (Pargament et al. 1988) is a 12-item, brief self-report measure that assesses how individuals incorporate their relationship with God into their attempts to cope with stressors. Each of the three theoretically derived and empirically supported scales (Collaborative, Deferring, and Self-Directing) contains 12 items, to which individuals respond using a 5-point Likert scale ranging from Never to Always. Each scale, therefore, has a possible range of 12–60, with higher scores indicating more frequent use of the respective approach to coping. The Collaborative, Deferring, and Self-Directing scales have alpha reliability coefficient of .94, .91, and 94, respectively, and 1-week stability coefficients of .93, .87, and .94, respectively. In regard to validity, the scales are correlated in expected directions, with Collaborative and Deferring correlated at .47, Collaborative and Self-Directing at −.61, and Self-Directing and Deferring correlated at −.37.

The RCOPE Scale

The RCOPE (Carver et al. 1989) is a 4-item subscale of the larger COPE measure, a multidimensional coping inventory to assess the different ways in which people respond to stress. The Cronbach’s alpha reliability coefficient was .92. The 8-week test–retest reliability was .86.

The Satisfaction with Life Scale (SWLS)

The SWLS (Diener et al. 1985) is a 5-item measure developed to assess positive psychological adjustment. Participants rate their overall satisfaction on the five items using a 7-point Likert scale. Scores range from 5 to 35, with higher scores indicating higher life satisfaction. Diener et al. (1985) conducted a study at a Midwestern public university and reported a coefficient alpha of .87. A two-month stability coefficient of .82 was reported by the authors. Each of the five items loaded on a single factor, with loadings ranging from .61 to .84.

The Langner Symptom Survey (LSS)

The LSS (Langner 1962) was developed to measure overall current psychological maladjustment. The LSS consists of 22 items that are either scored 0 or 1, indicating the presence or absence of a psychiatric symptom. Langner reported that a cutoff score of 4 or greater differentiated patients from non-patients and correctly identified 84% of those with psychological difficulties. The LSS has an overall internal constancy (Cronbach’s alpha) of .80.

Last, a demographic questionnaire was also included in order to gather information about participant’s age, gender, religion, race, year-level in college, residence, parental marital status, parental education, and parental income.

Procedure

Participants received extra credit as compensation for their involvement in the study. The packets included the following instruments: demographic questionnaire, Personal Religious Inventory (Lipsmeyer 1984), Religious Problem Solving Scale (Pargament et al. 1988), Langner Symptom Survey (LSS: Langner 1962), Satisfaction with Life Scale (Diener et al. 1985), and the COPE scale (Carver et al. 1989). For all instruments given, the participants entered self-report responses onto scantron sheets. Participants received packets in class and completed the questionnaires at home and returned sealed envelopes containing the packets to the experimenter before each class.

Data Analyses

In order to determine if religious coping served as a moderator between religion and psychological adjustment, a series of eight moderator regression analyses were conducted; four using the Satisfaction with Life Scale measuring positive adjustment and four using the Langner Symptom Survey measuring negative adjustment. The general procedure for conducting these analyses was to regress the predictor variable religion (INT) in the first block, the moderator variable(s) (COLLAB, DEFER, DIR, RCOPE) in the second block, and the interaction terms (COLLAB × INT, DEFER × INT and DIR × INT, RCOPE × INT) in the third block on to the outcome variables (satisfaction with life, maladjustment). Thus, there are three paths that lead to the outcome variable (see Fig. 1). Path 1 is religion, path 2 is the effect of religious coping, and path 3 is the interaction of the two. After the entry of each block of terms in the equation, the proportion of variance gained was tested for significance (p < .05). Individual terms within the block were assessed for significance. If the interaction term was statistically significant, the moderator hypothesis was supported. If supported, in order to determine the manner to which each religious coping style moderated the relationship between religion and satisfaction with life and maladjustment, participants’ scores were used to categorize them as high and low religion as well as high and low religious coping, which was then graphed against life satisfaction and maladjustment. These results are presented in Figs. 2–9.

Fig. 1
figure 1

Moderator model depicting three causal paths that feed into the outcome variable of interest. Adapted from Baron and Kenny (1986)

Full size image
Fig. 2
figure 2

Integration and self-directing religious coping as a moderator of satisfaction with life

Full size image
Fig. 3
figure 3

Integration and deferring religious coping as a moderator of satisfaction with life

Full size image
Fig. 4
figure 4

Integration and collaborative religious coping as a moderator of satisfaction with life

Full size image
Fig. 5
figure 5

Integration and turning to religion (RCOPE) religious coping as a moderator of satisfaction with life

Full size image
Fig. 6
figure 6

Integration and self-directing religious coping as a moderator of maladjustment

Full size image
Fig. 7
figure 7

Integration and deferring religious coping as a moderator of maladjustment

Full size image
Fig. 8
figure 8

Integration and collaborative religious coping as a moderator of maladjustment

Full size image
Fig. 9
figure 9

Integration and turning to religion (RCOPE) religious coping as a moderator of maladjustment

Full size image

Results

The correlation coefficients between religion and religious coping variables are presented in Table 1. In general, measures of religion and religious coping are significantly correlated at the p < .01 level and ranged from .46 to .79. In terms of religion and religious coping, strong positive correlations were found between religion as measured by Integration and Collaborative and Turning to Religion, while a strong negative correlation was found between Integration and Self-Directing and a moderate relationship was found between Integration and Deferring. In terms of the relationship among measures of religious coping, all of the measures were strongly correlated with one another. It is notable that Self-Directing was negatively correlated with Collaborative, Deferring, and Turning to Religion.

Table 1 Pearson correlations among religion, religious coping, and adjustment
Full size table

The first series of regression analyses were performed to determine the degree to which each of the four religious coping variable (Self-Directing, Collaborative, Deferring, Turning to Religion) moderated the relationship between religion as measured by integration and positive adjustment as measured by satisfaction with life. As can be seen in Table 2 all four coping styles moderate the relationship between religion and satisfaction with life. Furthermore, as can be seen in Figs. 3–5, which contain information for Deferring, Collaborative, and Turning to Religion, the same pattern of results was obtained for each religious coping variable. That is, the high integration participants for the high deferring, high collaborative, and high turning to religion groups were all more satisfied than were the other three groups for each of these religious coping styles. Conversely, a second pattern emerged with regard to the self-directing religious coping group, which is present in Fig. 2. In this group, the high integration participants who were also high self-directing were the least satisfied with life, as compared to the other three integration and self-directing groups.

Table 2 Results of hierarchical moderated regression for integration and religious coping as a moderator of satisfaction with life
Full size table

Results of Tests of Moderation on Maladjustment (LSS)

A second series of regression analyses were performed to determine the degree to which each of the four religious coping variable (Self-Directing, Collaborative, Deferring, and Turning to Religion) moderated the relationship between religion as measured by integration and maladjustment as measured by the Langner Symptom Survey (LSS). As can be seen in Table 3, all four coping styles moderated the relationship between religion and maladjustment. As can be seen in Figs. 7–9, which contain information for Deferring, Collaborative, and Turning to Religion, the same pattern of results was obtained for each religious coping variable. That is, the high integration participants for the high deferring, high collaborative, and high turning to religion groups were also high self-directing and were the most maladjusted compared to the other three integration and self-directing groups.

Table 3 Results of hierarchical moderated regression for integration and religious coping as a moderator of maladjustment
Full size table

Discussion

The purpose of this investigation was to determine whether the four religious coping styles of Self-Directing, Deferring, Collaborative, and Turning to Religion would significantly moderate the relationship between religion and psychological adjustment.

The tests of moderation yielded significant results, that is, the interaction terms between religion and religious coping significantly predicted psychological adjustment and maladjustment. The moderation regression results demonstrated that greater levels of religion and religious coping significantly predicted both positive and negative psychological adjustment.

In terms of collaborative religious coping, results indicated that those high on integration and high on collaborative coping were more satisfied with life and showed less distress than the other three groups. In contrast, the high integration and high self-directive group was significantly less satisfied and showed more distress than the other groups.

The finding that the high integration and high self-directing group was more distressed and less satisfied with life than the other groups raises the hypothesis that the self-directing approach may not be a measure of religious coping. Correlation coefficients in Table 1 appear to indicate that self-directing may not be a measure of religious coping. The correlation between self-directing and integration is significant and highly negative (−.68), indicating that those high in religion are low in self-directing. In addition, self-directing is significantly and negatively correlated with the other three measure of religious coping, and each of the negative correlation is strong.

From this pattern of correlation one could conclude that self-directing is not a measure of religious coping. If a person is high on religion, which influences their thoughts, feelings, and emotion, one would expect that an individual would include God in one way or another in their coping; however, the self-directing participants in this study do not. When employing the self-directing approach an individual is excluding God from the process and is taking an active role and personal responsibility for problem solving. Therefore, it appears that individuals high on religion and high on self-directing may experience cognitive dissonance or inconsistency. This is relevant to the point that although religion and religious coping are strongly related it does not mean that they are the same construct. In fact, the social psychology literature suggests that behavior is not necessarily determined by attitudes (Kraus 1995). It is a questionable assumption that people who identify as religious will employ their religion as part of their ways of coping.

Thus, it is likely that those individuals who identified as being highly religious and highly self-directing experience internal conflict because they do not incorporate God into their problem solving. In turn this contributes to their low life satisfaction and high distress when compared to other groups.

In general, results illustrate that Collaborative, Deferring and Turning to Religion (RCOPE) coping styles had a stronger relationship than the self-directing style to increased life satisfaction and less maladjustment. Kirkpatrick and Hood (1990) research on the Intrinsic-Extrinsic model of religious orientations showed that living and using religion are not opposites and are frequently occurring. In other words, it appears that those individuals who employ the Deferring, Collaborative and Turning to Religion religious coping approaches are person who “live” and “use” their religion. Therefore, it appears that these individuals who perceive their relationship with God as an important part of their daily functioning are more likely to actively engage in strategies that include God in problem solving rather than exclusively putting the responsibility of solving problem on themselves.

In sum, this study has illustrated that the interaction of religion and religious coping significantly predict psychological adjustment both positively and negatively. These results suggest that it is insufficient for clinicians to simply ask their clients “Are you a religious person?” It is imperative for clinicians to access how religion helps the client by asking questions such as “In times of stress do you utilize your religion?” “In times of stress how do you utilize your religion?” and “Does it help you?”

Future research may include comprehensively investigating religion along with religious coping, to determine how predictive they are to psychological adjustment using an adult population, having a breadth of sampling from different geographical regions from both church and non-church communities. Additionally, future research might include studying the concept of religion and religious coping among other religious groups. These recommendations are particularly warranted since our results are based on a sample of young adults who are primarily Christian and reside in the Midwest.