The Separation of Church and Mate:
How Does Church Attendance Impact Marital Satisfaction and Why

Michael Briscoe
Brigham Young University - Idaho

Keywords: Marital, Satisfaction, Happiness, Church, Religion, Attendance, Marriage

Abstract

This study seeks to assess the relationship between church attendance and marital satisfaction. This study uses the 2000 wave of the General Social Survey. It was hypothesized by the author that more frequent church attendance would increase levels of marital satisfaction in respondents. This hypothesis can be explained by the theory of symbolic interactionism, the idea that individuals assign different meanings and symbols to different parts of their life and then act according to those beliefs. It is theorized that people who attend church more frequently view marriage with more symbolism and meaning than those who do not, and therefore those who attend church frequently put forth more effort to maintain that commitment. The data showed that there was a highly statistically significant relationship between church attendance and marital satisfaction, and that the more frequently an individual attends church, the more satisfied they will be with their marriage. However, although this relationship was shown to be highly statistically significant, the amount of influence church attendance has on marital satisfaction is minimal. The results of this study show that further research may need to be conducted, particularly between different religions and religious denominations, and the effect that each of these individually may have on marital satisfaction.

Introduction

For years, religion has preached the importance of marriage, and following right behind the churches have been social scientists, analyzing this relationship. Many researchers study multiple factors, including cohabitation, while discussing religiosity and its impact on marital quality (Arland, Axinn, and Hill 1992; Brown et al. 2006). Others focus on religious involvement and its impact on marital satisfaction in specific sects such as Mormonism (Kunz 1964; Loser et al. 2009) or Catholicism (Joanides, Mayhew, and Mamalakis 2002; Shehan, Bock, and Lee 1990). Some studies focusing on religious involvement and its impact on marriage have studied specific social groups rather than religious denominations, focusing their studies on groups such as undergraduate women (Hui, Lindsey, and Elliott 2007), or heterosexuals and lesbians (Schumm, Akagi, and Bosch 2008). Yet the question remains, what is the impact of simple church attendance on marital satisfaction?

Among low-income couples, it was hypothesized that religious involvement would help couples better deal with economic stresses. Research finds that religious activity may be related to the high level of marital satisfaction among low-income couples (Lichter and Carmalt 2009). Brown’s (2006) research, which studies the effect of cohabitation on marital quality, seems at first to be unrelated the issue of church attendance and marital satisfaction. Brown’s research found that cohabitation led to lower marital satisfaction and a higher chance of divorce. However, he found that couples who met in church were less likely to cohabit. Clearly, church attendance would play a role in how many couples met in church, and would therefore have an impact on subsequent cohabitation and marital satisfaction.

Studying the impact of religiosity or spirituality on marital quality presents some problems because of the way in which religiosity is measured. Is it measured by church attendance, tithes paid, prayers given, or someone’s personal view of their spirituality? When asking these questions the matter of religious homogamy becomes an important factor because different religions have different ways of showing and measuring spirituality (Heaton and Pratt 1990; Shehan, Bock, and Lee 1990). It is also possible that heavy religious involvement could harm a marriage if the parties have different religious affiliation. Williams and Lawler (2013) suggest that “Interchurch couples are an important group to study not only because of their numbers but also because research indicates that they are at risk for negative marital outcomes such as divorce” (2013:1070). Williams and Lawler continue to point out that “Findings have been less consistent in showing a relationship between religious heterogamy and marital satisfaction or happiness” (2013:1071).

The study of religion’s impact on a couple goes beyond the study of the couple themselves and their marital satisfaction, and has become an issue for therapists and counselors to consider as they look at how they counsel couples. Sociologists have questioned how this issue should be addressed in therapy, and the possible benefits it could add in treating couples wanting to improve their overall marital satisfaction (Schumm 2003). As Nelson, Kirk, and PedraSerres (2011) point out in the article, 7.9% of participants believed that marital satisfaction was not strongly connected to religious commitment or religious values. However, when asked to explain further, the participants revealed that the religious aspects that impacted their relationships were more connected with spirituality and morality than with religion or religious participation.

Buss (2002) concluded that while religion may have a relationship with marital satisfaction, it is not the direct cause, but rather “parasitize(s) existing mechanisms” (2002:203) in its effect on marriage. Green and Elliot (2010) find similarly that “religious affiliation was unrelated to health or happiness” (2010:159). However, religious affiliation does not necessarily determine the frequency of an individual’s church attendance, and is therefore inconclusive to my study.

In beginning this study, it was theorized that individuals who attend church more frequently place a higher importance on the institution of marriage. One study found that in order to stay committed in a marital relationship, there had to be satisfaction with that relationship, and that religion had a positive relationship with marital commitment. This could suggest that religion leads to higher marital satisfaction, and therefore higher marital commitment and vice versa, but that was not made clear in the research (Larson and Goltz 1989). Other research found that couples who regarded the sanctity of marriage reported higher levels of marital satisfaction (Stafford et al. 2014).

Another study on the relationship between religious activity and marital quality came to the conclusion that “Little support is found for the idea that an increase in religious activity leads to improved marital relations” (Booth et al. 1995:669). More recent research on the issue of church attendance and marital satisfaction has led to more positive conclusions, such as Lee G. Burchinal’s study (2009), in which the author stated, “The marital satisfaction scores for both husbands and wives who were church members or who were regular or occasional in their church attendance were consistently higher than the scores for the husbands and wives who were not church members or who did not attend church” (2009:308). However, Burchinal continued to explain that, “only one in six tests of mean difference were significant” (2009:310). While his findings were a good start, this shows the need for more research on the subject. Some other social scientists attempted to add to Burchinal’s research, testing the relationship between church attendance and marital satisfaction, finding that there is a positive correlation between these two variables (Atkins and Kessel 2008; Call and Heaton 1997; Dudley and Kosinski 1990; Fard, Shahabi, and Zardknahaneh 2013; Hui, Linsdey, and Elliott 2007; Hunt and Morton 1978, Lichter and Carmalt 2009; Marks 2006, Wilcox and Wolfindger 2007).

In fact, one piece of research has found that there may be a positive relationship between church attendance and overall well-being, community and marital satisfaction. Research found that not just individuals’ church attendance as adults had an impact on their marital satisfaction, but that church attendance as youth as well as father’s church attendance “evidenced long term conditional relationships with job and marital satisfaction” (Willits 1988:291).

Perhaps the most thorough research on the topic of church attendance and marital satisfaction was conducted by Wilcox and Wolfinger (2008), whose study concluded that “religious attendance (…) is associated with higher reports of relationship quality for both married and unmarried couples with children in urban America” (2008:841). The only downside to this research is that it is only applicable to Americans residing in urban areas, which make up 79.219 percent of the population, according to the 2000 US census, leaving the results potentially inapplicable to the 20.871 percent who live in rural areas (Federal Highway Administration 2011).

Some research seems to indicate that the general population does think that religion has an impact on marriage, and certainly on family. Research shows that a higher number of people from divorced families change or reject their religion or church affiliation. Interestingly though, children raised in single parent households are more likely than children from a two parent household to increase their church attendance as they grow older (Lau and Wolfinger 2011). I believe this is because these people see church attendance as a way to increase their marital satisfaction, and reduce the chance of possible divorce.

With so much research on the subject, one must ask the question, “What is the impact of church attendance on marital satisfaction, and why?” This issue is still relevant and worth studying because despite some previous agreement from sociologists that church may play a positive role in a marriage, church attendance has declined in the 20th and 21st centuries (Franck and Iannaccone 2014). The aim of this study is to study whether or not church attendance still plays a critical role in marital satisfaction, and should be given greater consideration as couples try to wade through the flood of marital advice and research.

FIGURE 1

Methods

Data

The unit of analysis in this study is adults; people over the age of 18. I have used the data from the 2000 wave of the General Social Survey. The data was collected by the National Opinion Research Center at the University of Chicago. In collecting this data, the researchers conducted personal interviews with the participants.

The population of the General Social Survey is people who live in houses in the United States, which “covered about 97.3% of the resident population in the United States in 1985” (2013:2096). The survey in the 2000 wave was administered only in English. So, the population is English speaking adults who live in houses.

Sample

The researchers for the General Social Survey used a multi-stage cluster sample. In stage one, the sampling units were divided and categorized as either “Standard Metropolitan Statistical Areas” or “Non-metropolitan counties”. In stage two, the researchers divided the sample by “block groups” and “enumeration districts”. In stage three, they were divided into blocks. Finally, at this third level of the cluster sampling, the interviewer would begin the interview process, beginning at the northwest corner of the block and walk in one direction and trying to contact every household he passed until the quota was filled. The General Social Survey states, “The average cluster size is five respondents per cluster. This provides a suitable balance of precision and economy” (2013:2097). The response rate for the 2000 wave of the General Social Survey was .700.

Measures

Dependent Variable

In this study, the dependent variable is marital satisfaction. This variable was measured by asking, “Taking things all together, how would you describe your marriage? Would you say that your marriage is very happy, pretty happy, or not too happy?” Participants were then allowed to answer using the following measures: “Very happy, Pretty happy, Not too happy.”

Table 1: Descriptive Statistics for Respondents’ Marital Happiness

Independent Variable

The independent variable being studied is church attendance. To measure this variable, the question was asked, “How often do you attend religious services?” Respondents could then answer: “Never, Less than once a year, About once or twice a year, Several times a year, About once a month, 2-3 times a month, Nearly every week, Every week, or Several times a week.”

Table 2: Descriptive Statistics for “How Often Do You Attend Religious Services?”

Control Variables

Five control variables are used in this study: sex, opinion of premarital sex, spouse labor force status, socio-economic index, and age. Sex was coded with “male” as 1, and “female” as 2. This variable was recoded with “male” as 1 and “female” as 0. To measure respondents’ opinions of premarital sex, the question was asked, “There's been a lot of discussion about the way morals and attitudes about sex are changing in this country. If a man and woman have sex relations before marriage, do you think it is always wrong, almost always wrong, wrong only sometimes or not wrong at all?” Responses were coded with “Always Wrong” as 1, “Almost Always Wrong” as 2, “Sometimes Wrong” as 3, “Not Wrong at All” as 4, and “Other” as 5. Responses of “Always Wrong”, “Almost Always Wrong”, and “Sometimes Wrong” were recoded as 0. The response of “Not Wrong at All” was recoded as 1. “Other” was recoded as missing.

In the third control variable, spouse labor force status, responses were coded: “Working Full Time” as 1, “working part time” as 2, “with a job, but not at work because of temporary illness, vacation strike” as 3, “unemployed, laid off, looking for work” as 4, “retired” as 5, “in school” as 6, “keeping house” as 7, “other” as 8. “Working full time” and “working part time” were recoded as 1. “with a job, but not at work because of temporary illness, vacation strike,” “unemployed, laid off, looking for work,” “in school,” and “keeping house” were recoded as 0. “Retired” and “other” were recoded as missing.

Table 3: Control Variables Descriptive Statistics

Hypotheses

H_o = There is no relationship between church attendance and marital satisfaction.

H_a = There is a relationship between church attendance and marital satisfaction.

H1 = People who attend church more frequently will report higher levels of marital satisfaction.

H2 = People with a higher socio-economic index will be more satisfied with their marriage.

H3 = Older people will be more satisfied with their marriage.

H4 = People who have a spouse with a labor status of “working” will be more satisfied with their marriage.

Procedure

Because the dependent variable is ordinal, both a logistic or linear regression could be used to test the relationship between church attendance and marital satisfaction. I have used a linear regression, which reveals more about the data, and is more appropriate for this study.

Results

In model 1, there is a highly statistically significant relationship between church attendance and marital satisfaction. As an individual increases one category in church attendance, they increase .023 in categories of marital satisfaction. The R square for this model was .014, meaning we can account for 1.4% of the variability of marital satisfaction of the respondents.

Model 2 adds the variable of SEI (socio-economic index). There is a relationship between SEI and marital satisfaction, statistically significant on the .001 level. In this model, as well as all other models in this table, the data shows that each increase in SEI leads to an increase in marital satisfaction category by .004. R2 rose by .021 by adding SEI to the regression, which means that we can account for 2.1% of the factors that determine marital satisfaction. The focal relationship remains statistically significant, but changes to a .021 increase in marital satisfaction categories.

The variable of sex is added in model 3, testing whether there is a relationship between being male and marital satisfaction. This relationship is not statistically significant. The R2 remains the same, but the focal relationship changes to a .022 increase in marital satisfaction categories.

Model 4 tests the relationship between the first three variables as well as the variable of whether the individual thinks premarital sex is not wrong at all. This relationship is not significant. However, the R2 is up to .049 from .035 in model 3. The focal relationship in model 4 now shows that an increase in church attendance increases marital satisfaction category by .033. The relationships between church attendance and marital satisfaction, as well as SEI and marital satisfaction remain highly statistically significant.

In model 5, the variable of age is added. The relationship between age and marital satisfaction is not significant. Adding age does not change the R2, the coefficients, or significance levels of the focal relationship, or of SEI.

The final model, model 6, adds the variable of individuals who have a spouse with a labor force status of “working”. This relationship was not statistically significant. With this variable added to the regression the R2 drops to .042. The coefficient for church attendance also drops to .030, although the relationship between church attendance and marital satisfaction remains highly statistically significant.

Discussion

H1: People who attend church more frequently will report higher levels of marital satisfaction. The data supports this hypothesis. This relationship was highly statistically significant, suggesting that my theory may be correct that attending church more frequently gives more symbolic meaning to marriage, and therefore those individuals who attend church frequently are more committed to maintaining a satisfying marriage.

Although this relationship was found to be highly statistically significant, the R2 and coefficients were very small. In model 1 the R2 was .014, and the coefficient for church attendance was .023. This shows that although there is a definite relationship between church attendance and marital satisfaction, church attendance is not a good variable to use to predict marital satisfaction because it accounts for such a small percentage of variability in marital satisfaction. Similar results were found in model 2, testing the relationship between socio-economic index and marital satisfaction.

H2: People with a higher socio-economic index will be more satisfied with their marriage. This hypothesis was also supported by the data. Many problems in marriage are caused by money. Some research does indicate that even low income couples report high levels of marital satisfaction, especially if those low income couples are religious (Lichter and Carmalt 2009). Still, I hypothesized that a higher socio-economic index would lead to a more satisfied marriage. This hypothesis was proven correct, but there are almost certainly other factors involved here, such as the fact that people with a higher socio-economic index can afford counseling, etc.

H3: Older people will be more satisfied with their marriage. I hypothesized that older individuals would be more satisfied with their marriage because of increased maturity, and that if they have made it longer in their marriage, they must have learned to make it work, otherwise they would have divorced when younger. In connection with my original hypothesis, I knew that older couples attend church more frequently, a factor that I believed increased marital satisfaction. The data did not support this hypothesis.

H4: People who have a spouse with a labor status of “working” will be more satisfied with their marriage. As mentioned in H2, many marital problems are caused by money, and one of the largest causes of stress in a marriage can be unemployment. The data did not support this hypothesis. Further research or a revision could be required here as stay at home spouses were coded together with unemployed or laid off spouses as “not working”. If this were coded differently, the results may change.

Conclusion

Church attendance does impact marital satisfaction. However, it is not a large factor when it comes to predicting marital satisfaction. Other variables can be used to more effectively predict marital satisfaction of an individual. My theory is that this relationship is caused by the increased symbolic meaning given to marriage by churches. This research added value to the general discussion of church and marriage as the research was gathered and analyzed in a way that could be applied to all individuals who attend church in the United States, rather than individual sects or religions. Further research is needed to determine if this symbolic meaning for marriage can be given to individuals through a secular source, so as to help increase marital satisfaction of all people, not only religious individuals. The final consideration for improving this study would be to study church attendance and marital satisfaction of couples rather than individuals, since one individual only makes up half of the marriage.

References