Higher Education as Moral Community: Institutional Influences on Religious Participation During College

Author(s): Jonathan P. Hill

Source: Hill, Jonathan P. 2009. “Higher Education as Moral Community: Institutional Influences on Religious Participation During College.” Journal for the Scientific Study of Religion 48(3):515-534.

Link: http://onlinelibrary.wiley.com/doi/10.1111/j.1468-5906.2009.01463.x/abstract

Research on the religiosity of college students and graduates has produced differing views on how college attendance affects student religiosity. Some studies find that attending college has a unilaterally secularizing effect on students, but more recent research has found that students actually become more religious during college. Jonathan Hill adds to this work by examining how institutional-level influences determine a student’s religious participation both during and after college. Using longitudinal data to study the impact of college attendance on religious service attendance over an individual’s lifetime, Hill finds that “college campuses do not engender any long-term secularization at the individual level” (529), and that the short-term and long-term effects of institutional influences vary somewhat by type of institution. 

Background and Research Questions
In the past, most research on the relationship between religion and higher education focused itself through the lens of secularization theory, arguing that the pluralistic and anti-authoritarian culture of college campuses leads to decreases in religious beliefs and practice among students. However, more recent research contends that this may not be the case, positing that college students are focused on succeeding socially and working toward desirable career prospects, which leaves little time for reflection on one’s religious life, and leads to sustained or even increased levels of belief and practice. Hill argues that changes in religious participation during college are most likely a result of campus culture and institutional constraints more than any changes in belief. This argument raises the first research question of how college students differ from their non-college-attending peers in terms of religious participation both during and after college. 

Hill also asserts that previous research neglects the role of the college as an institutional force which organizes student life, which could be especially important in some religiously-affiliated institutions which hold students to strict behavioral standards, create homogeneous environments, and generally produce a culture where deviation from a set of religious practices and beliefs is frowned upon. Hill hypothesizes that this environment de-privatizes college students’ religious experiences, which will lead to students from “strongly religious” institutions maintaining their pre-college levels of religious practice. On the other hand, Hill suggests that religious students who do not attend religious schools may develop what is called a “subcultural identity” as they create social groups that provide insulation between the student and the university’s secularizing influence. If this hypothesis holds, Hill argues, students in non-religious schools will actually show less of a decline in religious participation than their peers at religious schools. Institutional effects may also vary according to the religious affiliation of the institution, depending on the level of student religious activity and involvement required by the school. As such, Hill believes that the rate of decline in religious participation will be steeper in mainline Protestant and Catholic colleges than in evangelical Protestant or non-religious colleges. 

Hill also examines how the relationship between a student’s personal religious beliefs and the beliefs held by the institution affect religious participation during college. It is possible that students whose personal religious tradition matches that of their institution will receive the most religious “benefit,” and will maintain their level of religious participation more effectively than will their peers from other traditions. It is possible, however, that students will adjust their religious participation to match that which is considered normal in the faith community of the institution’s affiliated religion, which may lead to a net change in either direction. 

Data and Analysis
This study utilizes data from two national datasets. Individual-level data comes from the National Longitudinal Survey of Youth 1997 (NLSY97), a nationally representative panel survey which collected information on just under 9,000 American youths annually from 1997 through 2004. Information on the institutions attended by the NLSY97 participants was gleaned from the Integrated Post-Secondary Data Service, a census of postsecondary educational institutions in the United States. Because the NLSY97 data follows the same people over a period of time, Hill’s causal arguments are much stronger than previous work on the religious outcomes of college graduates based on cross-sectional data. 

The main dependent variable in Hill’s study is how often the respondent reports attending religious services, ranging from “never” to “every day.” The primary independent variable captures the religious affiliation of the respondent’s college or university. Other pertinent institutional-level data includes the college’s status as a public, private for profit, or private not for profit institution; whether or not the institution requires chapel attendance; student body size; status as a four-year institution; status as an elite institution; racial composition; and gender composition. On the individual level, predictor variables include the respondent’s religious tradition, whether they earned a bachelor’s degree or an associate’s degree, their living situation, whether they are married or cohabitating, and whether they have children. 

The use of panel data allows Hill to utilize both fixed and random effects models. Fixed effects models naturally account for the effect of any unchanging characteristics such as gender and race and allow Hill to focus on the relationships between his variables of interest, net of everything else. This method reduces the issue of selection effects which might otherwise confound the study’s findings. Random effects models fill in the gaps, so to speak, left by the fixed effects models’ inability to estimate the effect characteristics which are constant over time, making these two analytical methods well-suited to tandem use.

In terms of the net effect of college attendance on religious participation, Hill finds that there is a very strong positive relationship between college attendance and subsequent religious service participation. Students do generally experience a downturn in participation levels during college, but this downturn is not large, amounting to just a few services per year for the average student. 

Hill also examines the effect of institutional type of religious participation. Unsurprisingly, students from very religiously conservative schools (i.e. Bible colleges) have the highest levels of religious participation after college, followed by those from evangelical colleges and historically black colleges and universities.  People who attended Catholic, mainline Protestant, and nonreligious public and private schools have the lowest religious service attendance. Indeed, by the third year of college, students at Catholic, mainline Protestant, and nonreligious public and private schools are basically identical to one another in terms of church attendance, despite the fact that these students’ levels of religious participation varied widely before college. It is important to note, however, that selection effects are strongly in play here; students who opt in to a religiously conservative college are going to differ from students who opt in to a public college, and as such differences in church attendance between those two groups of students is due more to individual differences than institutional effects. Hill compensates for this by using fixed effects models, and finds that institutional type generally has no significant effect. However, when exposure to college and institutional type are considered together, Hill finds that college type matters in terms of rates of decline and increase in religious participation over time. Students at Catholic and mainline Protestant schools decline in religious participation about 2.5 times the rate of public college students, supporting his theory that students at mainline religious institutions do not feel the need to participate very much religiously, as opposed to students at public institutions who “dig in” to their faith as a way of forming a subcultural identity in a large mostly secular community. 

Finally, Hill tests the relationship between a student’s own religious tradition and that with which the student’s college affiliates. Few of the findings in this analysis are significant, but generally there is some support for the idea that a student will adjust his or her religious participation to match that of the generally population of the college. Interestingly, a possible race gap appears here between mainline Protestant and black Protestant college graduates in that black Protestant students get more of a “boost” to their religious participation than their white counterparts. 

Overall, Hill’s study is a thoughtful and thorough exploration of the relationship between college attendance and religious participation. Hill’s causal arguments are sound and logical, and the project’s focus on the importance of institutional-level forces in the formation of students’ religious participation fills a void in the literature on this topic. As mentioned above, the use of random and fixed effects models is also particularly apt for this research question. That being said, it bears mentioning that Hill’s definition of “religious participation” only includes service attendance, excluding several other usual indicators of religious behavior such as prayer and scripture reading. Although this restriction may have been necessary for the sake of parsimony, this difference of definitions should be held in mind when considering Hill’s work in relation to other studies on religious participation among college students and graduates. On the topic of parsimony, one wonders if all the analyses Hill presents are strictly necessary. He offers seven initial hypotheses—some of which, he admits, deal with forces that “are sometimes, but not always, in conflict with one another” (519)—and runs multiple analyses for each, leaving the reader with a very dense results section to wade through. However, Hill’s final product, while complex, is a worthwhile study which addresses a gap in research on this topic and will hopefully serve as a springboard for future research on the relationship between religion and higher education. 

Keywords: higher education, moral communities, religious participation, college students

Sector: Catholic,  Evangelical Protestant,  Muslim,  Other Protestant,  Private,  Public

Outcomes: Cultural Impact,  Peers and Social Networks,  Religious and Spiritual,  School Choice

Date Posted: 2016-08-05