Secularization is a general process whereby religion becomes less salient in social institutions such as education, government, and medicine, and as a framework for understanding the world around us. But the phenomenon is multi-dimensional and is manifest differentially in different cultural contexts (Davie, 2013). To the degree that education is an indicator of secularization, evidence indicates that the process is complex. Data from the General Social surveys (GSS) in the U.S. conducted between 1972 and 2018 show a slight negative relationship between education and religious attendance (attendance drops by .01 times per month on average for each additional year of schooling completed). This negative relationship exists because religious groups with lower average education have higher average attendance (the correlation between average attendance and average education across Christian religious groups is -.21). But the education effect—based on fixed effect regression using major Christian denominations—is positive when averaged across groups. Attendance increases by .03 times a month on average for each year of schooling completed within Christian denominations. Moreover, getting a bachelor’s degree is associated with a decline in religious belief, but there is a still an overall positive relationship between education and church attendance (Schwadel, 2016). Education may facilitate participation in formal religious activities. Hungerman (2014) reviews conflicting literature and finds that higher levels of education lead to lower levels of religious affiliation in Canada. This finding is consistent with GSS data showing that education has a larger positive relationship with non-affiliation (r=.090) than with church attendance (r=.001). Also using Canadian data, Dilmaghani (2019) finds that higher education is associated with non-affiliation and lower attendance.

The positive relationship observed between education and church attendance does not extend to other aspects of religious experience. People with more education distinguish between formal participation in religious organizations and acceptance of religious worldviews in other aspects of their lives (Schieman, 2011). Glaeser and Sacerdote (2008) conclude that education increases social skills that enhance the utility of social activities such as going to church, but also emphasizes secular beliefs at odds with a religious worldview. The positive relationship between education and church attendance on the one hand, and the negative relationship between education and other aspects of religion such as belief and use of religious views in everyday decision making on the other, suggests compartmentalization. But the relationship between education and religious characteristics varies by religious tradition (McFarland, Wright, and Weakliem, 2011). For example, the positive relationship between religiosity and education is present among evangelical Protestants, Black Protestants, and Catholics, but not among mainline Protestants.

Another relevant line of research demonstrates that religious affiliation is associated with a wide variety of social characteristics including marriage, childbearing, women’s labor force participation, and earnings. Lehrer (2004) argues that religion alters the perceived costs and benefits of engaging in different types of social activities. Moreover, religiosity accentuates the effects of religious affiliation. Theological emphasis on family and socioeconomic achievement alter the importance people place on choices they make. Scholars have paid less attention to the relationship between education and these social behaviors within religious communities. If the relationship between education and religiosity is shifting, then the relationship between education and related social characteristics may also change.

This paper examines education, church attendance, and several other social characteristics, comparing Mormons (members of the Church of Jesus Christ of Latter-day Saints) with the U.S. population. Mormons are of interest because the relationship between education and church attendance is particularly strong among Mormons (Heaton and Albrecht, 1984). In addition, ample evidence indicates that Mormons differ from the national average in several social characteristics that are associated with religion (Heaton, Bahr, and Jacobson, 2004). The compartmentalization hypothesis implies that education might be positively associated with church attendance, but not with strength of religious identity or with other characteristics related to religion such as attitudes toward abortion or homosexuality. However, if the relationship between education and church attendance is stronger for Mormons, education may also have a different influence on these social characteristics among Mormons.

First, I compare educational attainment for Mormons and the nation. I then reexamine the relationship between education and church attendance. Third, in light of the compartmentalization hypothesis, I compare the relationship between education and formal participation with the relationship between education and strength of religious affiliation. Finally, I explore the relationship between education and several other social behaviors and attitudes. Where possible, the paper will also assess whether relationships are changing over the last four decades. The religious landscape has changed dramatically as mainline Protestant membership declines and the number with no formal religious affiliation grows.

Data Source

Analysis is based on the General Social Survey (GSS) conducted by the National Opinion Research Center at the University of Chicago (https://gss.norc.org/about-the-gss). The GSS asks questions about standard demographic characteristics, attitudes, behaviors, and attributes in the United States since 1972 using a national probability sample. All surveys from 1972 to 2018 are pooled. It is one of the best sources covering social and attitudinal trends. For purposes of this analysis, it is one of the best sources for comparing Mormons to the national population. GSS data also enables us to see if there are trends over time. Comparison with three other data sets including the National Election Surveys (NES), the Social Capital Community Benchmark Survey (SCCBS), and the Pew U.S. Religious Landscape Survey (PEW) shows that three of the four surveys show similar relationships among attendance, education, and conservatism for Mormons and the national population (see Table A in the Appendix). The PEW survey is an outlier. This analysis uses the GSS data because it includes a broader set of social variables and allows us to examine trends over more than four decades. Table 1 shows the years the survey was taken and the numbers of respondents.

Analysis proceeds in three steps. First, national/LDS comparisons of the relationship between education and other social characteristics of interest are graphed. Second, statistical tests for the relationships nationally and among Mormons, along with the difference between these groups on education effects, are calculated (see Table 2). Statistical tests are derived from OLS regression if the outcomes are measured at the ordinal or interval level and with logistic regression if outcomes are nominal.

The first two numeric columns in Table 2 show the statistical significance level (p value) for the relationship between each characteristic and education. The first numeric column shows the p value for the entire nation. The second numeric column shows the p value for the LDS subsample. Relationships between variables that aren’t statistically significant are labeled “n.s.”. Since “conversion” and “defection” are specific to the LDS subsample, p values for these relationships appear only in the second numeric column. The third numeric column demonstrates that the relationship between these various characteristics and education is significantly different for Mormons than for the rest of the nation in every instance.

I also test for trends over time using OLS regression or logistic regression depending on the distribution of the characteristics of interest. Statistical tests for trends over time are reported in Table B in the Appendix. In addition, I evaluate whether relationships between education and social characteristics can be explained by frequency of church attendance again using OLS regression or logistic regression.

Are Mormons More Educated?

LDS men and women are more likely to enter college than is the case nationally. They are also slightly more likely to go beyond 16 years (LDS men are 4.5% higher than the national average and LDS women 2.2% more likely to have postgraduate education). On average Mormon men have .75 years more schooling and Mormon women have .6 years more schooling than the gender specific national averages. The largest difference is that 10% more LDS members have gone to college.

Nationwide, women have gained some ground in educational attainment since the 1970s, and are now essentially equal with men. Gains for LDS women have been slower than for other groups, so their educational attainment is comparable to men and women nationally, but lower than for LDS men. However, these differences aren’t statistically significant.

Education and Religious Participation

GSS respondents report that they attend church 1.8 times per month on average. Overall, church attendance has a small positive correlation with education in the national population, but there is a substantial positive and statistically significant correlation among Mormons. Note that the lowest-attending group is LDS males without a high school degree (Figure 3). Female attendance tends to be more frequent than men’s nationally and among Mormons, but this is not the case for LDS females with college degrees, and there is even a slight drop-off in attendance among the most educated females. Therefore, LDS congregations have a higher percentage who have gone to college and attend regularly (40% compared to 13% nationally). Among regular church attenders, education is 1.3 years higher in Mormon congregations than is the case nationally and this difference is statistically significant.

Although average church attendance has declined as average education has increased, the relationship between education and attendance is relatively stable over time for Mormons and the nation at large (see statistical tests in Table 2).

Compared to other groups, education of weekly church attendees is relatively high in the LDS Church (Figure 4). Only Episcopalians, Presbyterians, and Jews are higher.

Strength of Affiliation

The compartmentalization hypothesis posits that education has a positive relationship with religious attendance, but not with more intrinsic aspects of religiosity such as faith. I test this by comparing attendance with the strength of religious affiliation (measured by the question “would you call yourself a strong [whatever their religious preference] or not strong” member of your religion). Thirty-eight percent of the respondents in the nation say their affiliation is very strong, and 63% of Mormons say their affiliation is very strong. Because attendance and strength of affiliation are measured on different scales, I standardized these measures for Figure 5. The compartmentalization hypothesis is supported in the national population where education has a slight positive relationship with attendance and a small negative relationship with affiliation strength. But compartmentalization is not evident for Mormons, as education has a positive relationship with both attendance and affiliation strength. Examination of change over time suggests a slight convergence because the negative relationship between education and strength of affiliation is becoming weaker in the national population and the positive relationship among Mormons is also becoming weaker.

Conversion and Defection

Thirty-nine percent of current Mormons say they converted (since age 16). The proportion of members who are converts declines with education, with the interesting exception of females with post-graduate education (Figure 6a). The patterns for defection look very similar (Figure 6b). Thirty-two percent of those who were LDS at age 16 no longer are. More educated members are less likely to defect, with the exception of women with postgraduate education. Even though the conversion rate into the LDS Church is relatively stable over time and the defection rate is increasing, the relationship between education and religious switching is stable.

Education and Family Life

LDS doctrine emphasizes the importance of marriage in God’s plan for eternal progression. Mormons are more likely to marry (86% of the LDS sample has ever been married compared to 79% nationally) than the U.S. population. Moreover, unlike the U.S. population, education is positively associated with getting married (Figure 7). The relationship between education and marriage is reduced by 22 percent, but not eliminated, when controlling for the positive relationship between church attendance and marriage. Analysis of temporal shifts indicate that even though the marriage rate is declining, the positive relationship between education and marriage is relatively stable over time among Mormons.

Divorce and marital separation are equally common among Mormons and the national population: one fifth of those who have ever married have been divorced or separated. The relationship between education and divorce is slightly negative in the nation, but more negative among Mormons (Figure 8). The negative effect of education among Mormons is reduced by 40%, but not eliminated, when church attendance is statistically controlled. Even though divorce is increasing, the relationship between education and divorce is relatively stable over time. Mormon women report having almost one child more than the national average (2.73 compared to 1.93). Unlike the national pattern of smaller families as education increases, more educated Mormons have larger families (Figure 9). The effect of education on children is moderated somewhat when church attendance is statistically controlled, but it is not eliminated. Also, the religious difference between Mormons and others in number of children among those with a college education does not appear to be declining over time. Even though family size is declining over time and the negative impact of education on family size is becoming slightly more negative in the nation, the positive relationship between education and family size is not declining among Mormons.

Education and Social Attitudes

LDS teachings highlight motherhood as the most important role for women, and the responsibility of men to preside at Church functions and in the home. A scale for gender equality was created by combining questions on women’s role in politics, maternal employment, and putting the husband’s career first (alpha=.757). Mormons score lower than the national average on this scale. For example, Mormons are more likely to think that a mother’s working hurt children (47% compared to 33% nationally), and that it’s better for men to work and women to tend the home (61% compared to 38% nationally). Although the relationship between education and gender equality is positive for the nation and for Mormons, the positive relationship is weaker among Mormons (Figure 10). Statistically adjusting for church attendance increases the positive education effect among Mormons by 30%, but this effect is still smaller than in the nation. Overall, women are more supportive of gender equality than men, and the gender difference is comparable for Mormons and the nation. As support for gender equality has increased, the relationship between education and support for gender equality has weakened slightly for Mormons and the nation as a whole.

The LDS Church teaches that homosexual relations are wrong. Eighty-three percent of Mormons who have been surveyed since 1972 agree that homosexual relations are always wrong compared to 62% nationally. Figure 11 shows average scores on a scale from 1 (homosexual relations always wrong) to 4 (homosexuality not wrong at all). Nationally, more educated people are more likely to say homosexuality is acceptable, but not among Mormons where the relationship with education is flat. When church attendance is controlled for, the education effect becomes somewhat positive among Mormons but is still much smaller than in the nation. As moral judgment against homosexuality has declined, the relationship between education has shifted in the positive direction for Mormons and the national population but the relationship is still relatively flat for Mormons.

The LDS Church policy is that elective abortion is a sin that could lead to excommunication. Nineteen percent of Mormons say it is permissible to get an abortion for any reason compared to 42% nationally. Favorability toward abortion follows a similar pattern to that toward homosexuality: a positive relationship with education in the nation, but not among Mormons (Figure 12). When church attendance is accounted for, the relationship between abortion and education becomes positive for Mormons, but is still much smaller than the effect nationally. Acceptance of abortion has increased somewhat in the nation (0.2 percent per year), but not among Mormons. However, the relationship between education and support for abortion has been stable over time for both groups.

There is a slight tendency for more educated people to think the U.S. spends too much on social programs in the nation. I created a scale including spending for the environment, health care, welfare, social security, and assistance for childcare (alpha=.62). The scale ranges from 1 for spending too little to 3 for spending too much. Compared to the nation, Mormons are more likely to say too much is spent on protecting the environment (14% compared to 9% nationally), healthcare (11% compared to 6% nationally), social security (11% compared to 6%), and assistance for childcare (12% compared to 7%). This tendency is much more pronounced among more educated Mormons (Figure 13). As with other factors considered, statistically controlling for attendance reduces but does not eliminate the education effect among Mormons. The positive relationship between education and opposition to social programs is declining in the national population but not among Mormons.

Mormons are more conservative than the nation as a whole, with 53% saying they are conservative compared to 34% nationally. Figure 14 shows average values on a scale that ranges from 1 for very liberal to 7 for very conservative; the modal response is moderate (value 4 on the scale). Given the findings so far, it should be no surprise that educated Mormons are more conservative politically, even though educated people in the U.S. have a slight tendency to be more liberal. About a third of the relationship between conservatism and education among Mormons can be explained by church attendance. Over time, the negative correlation between education and conservatism has shifted for Americans in general, becoming slightly more pronounced, but the positive correlation among Mormons has not changed. The intersection of education, church attendance, and political conservatism creates a unique demographic in Mormonism. One-third of the membership attends church regularly (two or more times a month), has gone to college, and is politically conservative. No other major religious group in the United States comes close to this (Figure 15). Presbyterians and Episcopalians do not reach 15%, and the national average is 10%. The percentage of the population who are college educated, frequent religious attenders, and politically conservative has increased over time both nationally and among Mormons, but the correlation between education and fitting in this group has declined slightly.

Decker (2019) attributes conservatism among Mormons in Utah to resistance against the nation’s growing approval of nonmarital relationships, abortion, and homosexuality. In Table 3, I consider whether traditional morality explains the relationship between education and conservatism among Mormons. The first row of numbers shows the effect of education on being conservative for the nation and for Mormons (using regression analysis). The first column reports results for the nation and the second shows the same statistical model applied to Mormons. The effect of education is negative in the nation and positive for Mormons. Subsequent rows show the effect of education when other factors are added to the statistical model. Church attendance does account for about 40% of the relationship among Mormons, but other factors are less relevant. Statistical control for being married, attitudes toward abortion and homosexual relations, gender norms, and attitudes toward government spending on social programs do not account for the relationship. In short, the relationship between education and conservatism extends beyond traditional family values. The correlation between education and conservatism among Mormons is higher in the Mountain West than in other areas, but is positive in the remainder of the country.

Conclusions and Prospects

This paper documents a systematic pattern of differential educational influence. Among Mormons, higher educational attainment has a more positive relationship with religious involvement, marriage and child-bearing, and political conservatism; and a less positive relationship with acceptance of homosexuality, abortion, and gender equality when compared with the rest of the nation. Conservatism among educated Mormons appears to be a broader phenomenon than a response to any specific issue. Moreover, the differential impact of education in Mormonism cannot be explained solely by the high correlation between education and church attendance. Clearly, the compartmentalization hypothesis does not apply to Mormons. They do not evidence a pattern where education is associated with church attendance, but not with strength of religious identity or social attitudes.

The data suggests no single explanation for the unusual role of education among Mormons. The Church of Jesus Christ of Latter-day Saints emphasizes individual morality above social justice, family above community, tradition above adaptation, and individual responsibility above public policy. Leaders dress and speak conservatively. The LDS Church has three universities, and administers religious programs for high school and college students that emphasize the superiority of religious perspectives over secular or scientific perspectives. A lay ministry integrates members into the organization and more educated people are more likely to have the skills to fill these positions. Because of these interrelated phenomena, social and political conservativism tends to become linked with an overriding ideology closely linked with education and religious participation.

Analysis suggests that the intersection between education, religious involvement, and conservatism is unlikely to change in the foreseeable future. The pattern of influence is relatively stable over time despite societal changes in family life and social attitudes. Senior leaders in the Church have given many years of service and been well integrated into the religious culture. All of the top fifteen leaders have postgraduate degrees and/or respectable careers in business. In recent years, the Church has announced a policy of “Home Centered, Church Supported” doctrinal curriculum (Nelson, 2018). This policy reduces the time members spend in official Church activities and emphasizes the responsibility of families to teach and provide wholesome activities for children. This change could lead to a more liberal approach among more-educated Mormons. On the other hand, and more likely, conservatism can increase, owing to the current disposition of more-educated Mormons.

The linkage among education, religious participation, and social conservatism provides several advantages. Inclusion of more educated people enhances social capital and financial resources to maintain and grow the organization. The conservative image promotes a sense of stability that fosters confidence.

There are also several disadvantages. Segments of the population are more likely to be excluded (note the drop in attendance and retention among women with postgraduate education and low attendance of men without a high school degree). Conservative values also can obstruct adjustments to social trends. Elimination of formal racial barriers to full religious participation did not occur until 1978, long after integration occurred in other major social institutions. Inclusion of women in leadership and acceptance of LGBTQ members also lag behind national trends. Interpretation of scripture is also filtered by social conservatism to legitimize this perspective at the expense of messages that are more egalitarian and inclusive.

Appendix

Table A displays the direction and strength of the relationship between (1) education and church attendance, and (2) education and conservatism for both LDS and non-LDS subsamples from four surveys.

Table B on the following page shows the direction of the trend over time for the relationship of various social characteristics with education for both the national sample and the LDS subsample. Since conversion and defection are limited to the LDS subsample, these cells are left empty for the national sample. A “+” denotes a positive direction. A “-” denotes a negative direction. Relationships between variables that are not statistically significant are labeled “n.s.”.

References

Albrecht, Stan L. and Tim B. Heaton, 1984. “Secularization, Higher Education and Religiosity.” Review of Religious Research 26, no. 1: 43–58.

Davie, Grace. 2013.  The Sociology of Religion . Los Angeles: SAGE Publications.

Decker, Rod. 2019.  Utah Politics: The Elephant in the Room . Salt Lake City: Signature Books.

Dilmaghani, Maryam. 2019. “Education and Religious Decline: Evidence from the Canadian Compulsory Schooling Laws,”  Education Economics  27, no. 3: 287–307.

Glaeser, Edward and Bruce L. Sacerdote. 2008. “Education and Religion.”  Journal of Human Capital  2, no. 2: 188–215.

Heaton, Tim B., Stephen J. Bahr, and Cardell K. Jacobson. 2004.  A Statistical Profile of Mormons: Health, Wealth, and Social Life . Lewiston, NY: Edwin Mellen Press.

Hungerman, Daniel M., 2014. “The Effect of Education on Religion: Evidence from Compulsory Schooling Laws.”  Journal of Economic Behavior and Organization  104: 52–63.

Lehrer, Evelyn L. 2004. “Religion as a Determinant of Economic and Demographic Behavior in the United States.”  Population and Development Review  30, no. 4: 707–726.

McFarland, Michael J., Bradley R. E. Wright, and David L. Weakliem. 2011. “Educational Attainment and Religiosity: Exploring Variations by Religious Tradition.”  Sociology of Religion  72, no. 2: 166–188.

Nelson, Russell M. 2018. “Opening Remarks.” General Conference of the Church of Jesus Christ of Latter-day Saints. https://www.churchofjesuschrist.org/study/general-conference/2018/10/opening-remarks?lang=eng.

Schieman, Scott. 2011. “Education and the Importance of Religion in Decision Making.”  Journal for the Scientific Study of Religion,  50, no 3: 570–587.

Schwadel, Philip. 2017. “Does Higher Education Cause Religious Decline?: A Longitudinal Analysis of the Within- and Between Person Effects of Higher Education on Religiosity.” The Sociological Quarterly 57, no. 4: 759–786.