Introduction

Although traditional religious practices have drastically declined over the years, religiosity still remains a topic of interest for many researchers, particularly in the field of sociology and psychology. Increasing secularisation of western societies throughout the 20th century has raised questions about the relevance of religion and family life (Sherkat & Ellison, 1999). Much of the research on the relationship between religiosity and marriage was on the idea that couples, who are more religious are more likely than other couples to have happy and stable marriages (Sullivan, 2001; Glenn & Supancic, 1984; Call & Heaton, 1997). Compared to other couples, couples who attend church more frequently seemed to have higher marital satisfaction (Kunz & Albrecht, 1977; Wilson & Musick, 1996). More extensive evidences also exist on ties between involvement in religion and marital satisfaction, for instance, Mahoney, Pargament, Tarakeshwar, and Swank (2001); Glenn (1982); Bahr and Chadwick (1985); Hatch, James, and Schumm (1986); Larson and Goltz (1989); and Booth, Johnson, Branaman, and Sica (1995). Compared to other couples, couples who attend church services more frequently have been shown to be less likely to divorce and to have more stable marriages (Albrecht & Kunz, 1980; Nye, White, & Frideres, 1973; White & Booth, 1991).

In a comprehensive review of all studies published from the 1930s to the 1990s, Jenkins (1991) found conflicting evidence for propositions such as high religiosity promotes marital satisfaction and increased church attendance increases marital satisfaction. Regarding stability, Jenkins (1991) concluded that there is a ‘moderate’ support for the proposition that ‘high religiosity promotes marital stability. Booth et al. (1995) reported a reciprocal relationship between religiosity and marital satisfaction, such that changes in marital satisfaction predicts changes in religiosity over time. But, most of these studies were reported from the United States of America, Canada, or New Zealand. One of the recent findings from Turkey, published in this journal, also showed that religiousness had a major effect on marital satisfaction (Hünder & Gencöz, 2005).

Given the somewhat consistent findings of positive association of religiosity with marital stability and satisfaction, the present study was expected to further validate the association between religiosity and marital satisfaction among the married adults and thereby contributing to the already existing literature. This study differed from other studies by taking the first-married and remarried adults as participants from a European context. Considering the increase in the rate of divorce, separation, remarriage and cohabitation, we presumed that a comparative study between these two groups of married adults (first-married and remarried) could give us some insight whether religiosity contribute, to marital satisfaction and play a role in the stability or longevity of a marital relationship. Therefore, the findings from this study were expected to have clinical and therapeutic significance by helping therapists and health workers to draw conclusions and implications for marital and couple therapy.

For the purpose of this research, we defined marital stability in terms of the status of the couple’s relationship, whether the persons still remain in their first marriage or not. Those who still remain in their first marriage are referred as ‘first-married adults’ (marital status 1) and those who were divorced or separated and remarried or living with a partner as ‘remarried adults’ (marital status 2).

Research Questions and Hypotheses

The present study primarily examined the possibility whether people who are more religiously oriented, frequent church goers, or those who perform religious duties more faithfully have higher rate of marital satisfaction. On the basis of the previous studies cited, we expected a positive correlation between religiosity and marital satisfaction. Since the current study was cross-sectional in nature, we also presumed that there would be a significant effect of marital status and gender on religiosity and marital satisfaction.

Method

Participants

The total sample consisted of 787 respondents from the Flanders region in Belgium. They were non-randomly selected heterosexually married adults, of which 424 were first-married adults and 363 were remarried. There were 396 women and 391 men. The mean age of women was 44.80 years (SD = 9.12) and the mean age of men was 47.31 years (SD = 9.09). The respondents had been married for a mean of 15.23 years (SD = 11.44), with a range of 18 months to 50 years.

Measures

Religiosity was measured using a 4-item scale. This scale is a brief measure of religiosity constructed by Rohrbaugh and Jessor (1975), which attempts to capture important dimensions of religiosity, including ritual, consequence, and experience. It also provides an overall rating of religiosity. The following four questions were asked: “How often do you attend religious services?”; “In general, how important are religious or spiritual beliefs in your day-to-day life?”; “When you do have problems or difficulties in your work, family, or personal life, how often do you seek spiritual comfort?”; and “In general, would you say you are a religious person?”. All statements were measured on a 5-point scale. Total religiosity scores ranged from 1 to 20, and were classified as: (1) Weak religiosity (1 to 9); (2) Medium religiosity (10 to 14); and (3) Strong religiosity (15 to 20).

The Maudsley Marital Questionnaire (MMQ) is a short 20-item scale relating to Marital (M) (10 items; theoretical scale score range: 0–80), Sexual (S) (5 items; range: 0–40) and General Life (GL) (5 items, range: 0–40) maladjustment. In the MMQ scale, a higher score implies greater adjustment problems, and greater adjustment problems mean, lower marital satisfaction. The validity of the Dutch version of MMQ was established in 1983 by Arrindell, Emmelkamp, and Bast (1983).

Procedure

We employed a snowball sampling technique (snowball sampling consists of identifying respondents who then refer researchers to other respondents) and sought the help of regular college students for data collection. As part of student’s study requirement they were asked to take one each questionnaire and get it duly filled-in by a heterosexually married or remarried adult of their choice. The intention of this research was made clear by a covering letter and the confidentiality of the respondents was assured. Participation in the study was done merely out of goodwill. A stamped addressed envelope was provided with those that were distributed externally.

Method of Analysis

All statistical analyses were conducted using the SAS package. The ANOVA was used to examine the effect of the independent variables on the dependent variables. Pearson’s Product Moment Correlation examined the association between the two dependent variables.

Results

In total, 988 questionnaires were distributed and 787 (80%) were returned. In this study, the Cronbach’s alpha for Religiosity Scale was = 0.90 (Mean = 9.66, SD = 4.42) and this scale had a highly significant (p < .001) inter-item correlation (R1-R2 = 0.65; R1-R3 = 0.58; R1-R4 = 0.64; R2-R3 = 0.74; R4-R2 = 0.77; R3-R4 = 0.72) with one another. The Cronbach’s alpha for the different MMQ scales were: MMQ-M = .90 (M = 13.58, SD = 10.79); MMQ-S = 0.80 (M = 8.67, SD = 7.21); and MMQ-GL = .66 (M = 9.66, SD = 4.89; N = 787).

Effect of Marital Status and Gender on Marital Satisfaction and Religiosity

As shown in Table 1, the effect of marital status was statistically significant on religiosity, F (1, 785) = 72.12, p  < .0001, est η2 = 0.084. For the MMQ scales the effect of marital status was highly significant on MMQ-S, F (1, 785) = 52.34 p < .0001, est η2 = 0.062, but for other two subscales a marginally significant effect was reported, MMQ-M: F (1, 785) = 4.65, p < .05, est η2 = 0.005, MMQ-GL: F (1, 785) = 4.57, p < .05, est η2 = 0.006.

Table 1 Analysis of Variance for religiosity and MMQ-S by marital status
Full size table

As shown in Table 2, the effect of gender was statistically significant on religiosity, F (1, 785) = 15.94, p  < .0001, est η2 = 0.020. For the MMQ scales the effect of gender was highly significant only on MMQ-M, F (1, 785) = 12.23 p  < .0005, est η2 = 0.015. There was no indication of a significant effect of gender on other two subscales (MMQ-S and MMQ-GL) (see Tables 1 and 2 for details).

Table 2 Analysis of Variance for religiosity
Full size table

Relationship between Religiosity and Marital Satisfaction

As shown in Table 3, religiosity has a negative correlation with MMQ scales with an exception of MMQ-S. Religiosity indicated a significant (r = .19, p = .0001) positive correlation with (MMQ-S) sexual-adjustment scale (a higher score in the MMQ scales implies greater adjustment problems and greater adjustment problems are interpreted as lower marital satisfaction).

Table 3 Correlation Coefficients between MMQ scales and Religiosity for different subgroups
Full size table

Summary and Discussion

The primary goal of the current study was to examine the association between religiosity and marital satisfaction among the first-married and remarried people. Since it was a comparative study, first, we examined the possible effect of gender and marital status (whether the respondents are first-married or remarried) on religiosity and marital satisfaction. As revealed by ANOVA, the effect of gender and marital status was statistically significant on religiosity. The effect of gender and marital status on marital satisfaction was partial, significantly effecting only on MMQ-S and MMQ-M by gender and marital status respectively. However, it should be noted that the effect size (the measure of the strength of the effect) of marital status on marital satisfaction scales, for instance, on marital adjustment (MMQ-M) and general life adjustment (MMQ-GL) was only five in 1000 (η2 = 0.0058 for MMQ-M and 0.0057 for MMQ-GL). But in the case of sexual adjustment the measure of strength of the effect was around 6% (η2 = 0.062), again considerably a low effect size. Therefore, perhaps as previous researches had shown a number of other variables, such as, age (Anderson, Russell & Schumm, 1983; Schumm & Bugaighis, 1986); education (Janssen, Poortman, De Graff, & Kalmijn, 1998; and Kalmijn, 1999); number of children (White & Edwards, 1990), and others might have effected the sexual adjustment and marital happiness of the married people. Obviously, the present study did not attempt to establish the causes and reasons for any such effect of the independent variables. Perhaps, this could be attributed to other variables. Therefore, logically the next step would be to look into the possibility of the impact of other variables, such as demographic variables age, and education, which are, for instance, major determinants in explaining church commitment (religiosity) Billiet et al. (2003).

In the current study, one of the subscales of the MMQ (MMQ-S, the sexual adjustment scale) showed a highly significant positive correlation with religiosity—(it should be remembered that a higher score in the MMQ scales implies greater adjustment problems and greater adjustment problems are interpreted as lower marital satisfaction). This finding would imply that adults with high religiosity report have higher rate of sexual adjustments problems. Interestingly, such a correlation is observed among both first-married and remarried adults ruling out the effect of marital status. The other two subscales (MMQ-M and MMQ-GL) showed a positive correlation though not significant with religiosity. Therefore, in a broad sense, with a very moderate positive correlation between religiosity and general life adjustment, it is suggestive that there is a tendency towards a positive correlation between religiosity and marital satisfaction. Therefore, the findings of the present study, to some extent, further confirmed the claims of some of the prior studies (e.g., Albrecht & Kunz, 1980; White & Booth, 1991) that there is a positive association between marital stability and religiosity.

Implications of the Findings

In the light of the findings of this study, therapists and health professionals might consider reminding couples or partners who are on the verge of separation or divorce that stability in a (marital) relationship is not necessarily characterized by ‘problem-less’ or ‘problem free’ as this study has indicated, particularly, of sexual-adjustment problems.

In the same manner, taking into consideration the tendency towards a positive correlation of religiosity with marital satisfaction found in this study, the therapists could suggest or discuss the role of a transcendental element or a religious or spiritual aspect of the couples or of the partners for marital problems. Several studies (Rose, Westefeld, & Ansley, 2001; Patterson, Hayworth, Turner, & Raskin, 2000; Johnson & Sandage, 1999) have shown that considering religious beliefs in therapeutic context can be useful for couples and families in their relationships.

Limitations of the Present Study

There are a number of limitations in this study. Due to non-random nature of the sample, generalization of the findings of this study has to be done in caution. Similarly, since the study was limited to non-clinical sample generalizing the results to a wider clinical population could also be done with caution. Since this study was only a cross-sectional and correlational, which identify only associations and does not lend itself to causal interpretations.

Religiosity on this study was only a general measure, perhaps use of a multidimensional religiosity scale could serve to illuminate the multifaceted aspects of religiosity. Future researches might employ a scale, for instance, using the Post-Critical Belief Scale, a new multidimensional instrument for measuring religious attitudes in a secularized context, Hutsebaut (1996, 1997) that can measure multidimensionality of religiosity.

Finally, since the current study relied entirely on self-report measures, it is subject to common critiques of self-report measurement such as, socially desirable responding, retrospective reconstruction, and so forth.

Conclusions

Three major conclusions to be drawn from the present study were, first of all, the effect of gender and marital status was statistically significant on religiosity. Secondly, the effect of gender and marital status on marital satisfaction was only partial. However, in relation to marital status and religiosity, further research is required to establish a reciprocal link between these two variables (that is, whether religiosity enhances the marital stability and stability promotes religiosity). Another aspect that could be investigated in future is the influence of other potential confounding variables such as, strict imposition of religious rules, for instance, certain divorced persons are not given full participation or communion in certain churches. In other words, did divorce or separation become a further cause for the decline of religious practices as far as separated or remarried adults are concerned?