Religiosity and the Impact of Religious Secondary Schooling

Author(s): Allyssa A. Wadsworth and Jay K. Walker

Source: Wadsworth, Allyssa A. and Jay K. Walker. 2017. "Religiosity and the impact of Religious Secondary Schooling." Journal of School Choice 11(1):131-147.



Several studies have established that religious education, particularly Catholic education, provides young people with a slight boost to academic outcomes like school performance and college attendance. Little has been said, however, about how religious school attendance influences religious participation later in life. To fill this gap in religious school scholarship, Wadsworth and Walker use longitudinal data about public and religious school graduates to track how religious school attendance affects church attendance and volunteerism throughout young adulthood. They find that graduates of religious schools do exhibit a higher rate of religious participation two years after finishing high school as compared to their public school peers, but eight years after graduation this effect weakens and becomes more irregular.

Background and Research Questions

Parents consider a wide variety of factors when making decisions about their child’s education, including cost, safety, and quality of instruction. Some parents also want their child’s school to inculcate students with a particular religious worldview, with the hope that their child will benefit morally and spiritually from attendance. While students in religious schools do generally experience greater exposure to religious ideologies and religious services than do their peers in nonreligious schools, there has been little research into whether or not these effects are robust over time. To address this concern, Wadsworth and Walker examine the long-term effects of religious schooling on religious participation after a student leaves school. In an era of growing numbers of options for parents seeking alternatives to traditional public schooling, as well as decreasing attendance in Catholic schools, the authors of this article hope to provide parents with more information as to whether or not investing in religious education during their child’s youth will pay off spiritually later in the child’s life.

Data and Analysis

The authors use data from the National Educational Longitudinal Survey (NELS), a nationally representative survey that followed a cohort of students from their eighth grade year in 1988 through young adulthood, with the final wave of data collection occurring in 2000. The authors include students from all public schools, Catholic schools, and religious private schools in their sample. They measure religious activity at two time points after students finish secondary school—two years post-graduation and eight years post-graduation. Religious participation was measured differently in the two post-high school data collections.  Two years after high school graduation, respondents were asked about time spent on religious activities and about volunteer activities with religious organizations.  Religious activity was measured at eight years post-graduation by the number of days a respondent reports attending religious services per month and whether or not respondents report attending any religious service per month.

To effectively compare students across school sectors, Wadsworth and Walker use propensity score matching to ensure that they are comparing students who are as similar as possible in all measured factor aside from the type of school they attended. Students are assigned a propensity score based on several factors relating to their own demographic characteristics and family backgrounds. Students with similar scores but different educational backgrounds are then compared to one another. As a brief example, a white public school student who attends youth group activities whose parents are married and make between $50,000 and $75,000 per year would be compared to a Catholic school student and a religious private school student with the same characteristics. This method helps the researchers make their claim that any long-term differences they find in religious activity between school sectors are the result of the type of school the student attended and not any other background characteristics.


Wadsworth and Walker find that religious participation varies by school sector and gender. Two years after finishing high school, both male and female Catholic and (non-Cathaolic) religious school graduates are more likely to report spending time doing religious activities than their public school graduate peers but are not more or less likely to report volunteering for a religious charity. Eight years post-graduation, however, effects are less consistent. At this point, male Catholic school graduates are more likely than public school graduates to attend religious services, but they do not report statistically higher frequencies of religious attendance. Female Catholic school graduates do not differ significantly from their public school peers on either religious activity measure. Results for non-Catholic religious school graduates are just the opposite.  Male non-Catholic religious school graduates have significantly higher frequencies when they attend religious services compared with public school graduates, and female graduates report both higher incidence and frequency of attendance.  Therefore, the effect of Catholic school attendance on young adult religious participation as measured in this paper is strongest for non-Catholic religious school graduates, and is marginal for Catholic school graduates.


The authors of this article make a valid argument that long-term religious activity outcomes of religious school graduates have been under-studied, and this article makes a solid first step in filling in that gap in the literature. The use of propensity score matching lends credence to the authors’ findings; some issues, however, make the results presented in the study less than fully convincing. The authors do not control for factors in respondents’ lives after graduating high school, despite the fact that factors like attendance at a religious college might influence post-graduation religious activity. Beyond this issue, the incongruence between measures of religious activity at two years post-graduation and eight years post-graduation make comparisons between the two problematic. This second concern could be addressed by an explanation within the article about why the different measures of religious activity should be linked or how one could serve as a proxy for the other, but no such explanation is given. In sum, although the authors make a good effort at explaining religious outcomes of religious school graduates, the numerous unaccounted-for factors that are left out of their models detract from the conclusiveness of their findings.

Keywords: long term effects, religious participation, church attendance, volunteerism, propensity score matching

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

Outcomes: Morality and Character Formation,  Religious and Spiritual,  School Choice

Date Posted: 2017-06-13