Ilana M. Horwitz’s new book focuses on the one in four American high school students who are “raised with religious restraint”—they orient their lives around God and try to behave in ways that they believe will please God. Her book is based on 10 years of survey data and 200 interviews.
She finds that these religious students excel in high school and college. Generally, these students are more likely to graduate from college; boys from lower-middle-class families especially benefit. But girls—especially from middle- and upper-class families—question the value of attending religious colleges and “undermatch” in their choices.
Horwitz shares the results of her work in God, Grades & Graduation: Religion’s Surprising Impact on Academic Success (just published by Oxford University Press). Horwitz is the Fields-Rayant Chair in Contemporary Jewish Life at Tulane University, but her book is primarily (as she explains below) about Christian students. She responded via email to questions about the book.
Q: How did you first stumble on your idea about the role of religious belief in academic performance?
A: In 2015, I was a doctoral student at Stanford University studying sociology and education, and I spent much of my time thinking about how race, class and gender shaped educational outcomes. One day, I came across the Pew Religious Landscape Study. The headline was that Americans were becoming less religious, but what struck me was that the U.S. was by far the most devout country compared to other advanced industrial countries.
Frankly, I was surprised that religion still had such a strong hold on Americans, and it piqued my curiosity. At the time, I was living in Stanford’s family graduate housing, which is very communal. Parents spend a lot of time in shared playgrounds, and as I got to know my neighbors, I noticed that religion played a central role for many of them. Most of them were Protestant or Latter-day Saints, and as I listened to them talk about Bible study, going to church and child rearing, I started to wonder—does Americans’ religious upbringing influence their educational trajectories?
I also noticed that in all the religious families I was hanging out with, there was a consistent gender pattern: dads were students and moms stayed home with the kids. When I thought about my immediate social network, which is mostly filled with religiously liberal Jews and Catholics, I could only think of one of about 30 families in which the mother stayed home with children while the father went to work. I became fascinated by the differences in my social worlds and [that] led to more questions: How might religious expectations around education and work explain this gender pattern, especially at such a selective university? And what role might social class play in the relationship between religion and academics?
I dug around to see what the research said about religious upbringing and academic outcomes but surprisingly came up empty-handed. I found a few dozen articles, all of which were based on survey data and were mostly written in the early 2000s, but not a single book about how religion shapes people’s academic outcomes throughout the life course. So, I decided to write one.
Q: Were there differences among religious groups?
A: Yes. There are significant differences in the educational outcomes of girls raised by Jewish parents versus girls raised by non-Jewish parents, which I detail in a forthcoming article in the American Sociological Review. However, the research in my book centers on Christianity because it is the most prevalent religious group in America. I am especially curious about Christians who are intensely religious— those who display high degrees of “religiosity” as measured by how they say they behave and what they say they believe. I foreground the role of religious intensity because this is where the most profound polarization exists in the current American landscape.
Here are the three main findings.
- First, more intensely religious teens see an educational attainment bump. In the book, I describe a study I conducted with Ben Domingue and Kathleen Mullan Harris where we used sibling differences to estimate the associations between religiosity on short- and long-term academic success. We found that more religious adolescents completed more years of education 14 years after their religiosity was measured, even in models with family fixed effects.
- Second, the extent to which religious teens see an educational attainment bump varies by social class. In my book, I explain why religious teens from working-class and middle-class families see the biggest educational attainment bump. However, religious teens from poor families and from professional-class families do not see a strong educational attainment bump.
- Third, while religious teens complete more years of education, they often do so at lower-quality colleges (as measured by selectivity based on the institution’s incoming SAT score). This process is called “undermatching” and is most prevalent among religious teens from the professional class. Scholars tend to think of undermatching as a class phenomenon, but my research suggests undermatching is also a religious phenomenon that particularly affects professional-class teens.
Q: Why are these students more likely to graduate from college?
A: To understand how a religious upbringing affects one’s education, I looked at the entire road from secondary school to college and decomposed teens’ academic trajectories into performance effects and choice effects. Performance effects reflect how students perform academically, with grade point average being the most common measure of performance. Choice effects reflect the decisions that students make conditional on their performance. These decisions are most common at educational transition points, such as the transition between grade levels or the transition after high school.
Religious teens are more likely to graduate from college because they earn better grades in middle and high school than less religious teens. In the sibling study I mentioned above, high school GPA mediated 68 percent of the relationship between religiosity and educational attainment. This makes sense given that high school GPA is among the strongest predictors of academic success after high school, including college completion.
This obviously begs the question: Why do religious adolescents earn better grades in secondary school than their less religious peers? In my book, I argue that schools and churches promote similar ideals—both institutions value kids who abide by the rules and respect authority figures. Intensely religious teens are precisely these types of kids—they are deeply conscientious and cooperative. As it turns out, the very dispositions that teens adopt to please God are also the dispositions that help them earn good grades.
Nonreligious readers who were academically successful might wonder where they fit into the story. Indeed, if religious teens are getting such good grades, why are the most selective colleges overrepresented by atheists? My book explains that atheists are academically successful but for a different reason. Rather than being motivated to please God by being well-behaved, atheists are intrinsically motivated to pursue knowledge, think critically, and are open to new experiences. This turns out to be even more important for academic performance than being conscientious and cooperative. Disavowing a belief in God is not what causes teens to do well academically. Instead, it’s a selection effect—the kinds of people who are exceptionally curious and therefore engage in self-directed behavior tend to be the kinds of people who are willing to go against the grain and take the unpopular religious view that God doesn’t exist. In fact, some of the most academically accomplished adolescents were those who grew up religious but moved away from religion by their mid-20s.
Q: Why do religious girls who are academic achievers tend to distrust attending competitive colleges?
A: This comes back to choice effects that I mentioned above. Since college is voluntary, students can choose whether to apply to college and which colleges to apply to. While the decisions that students make about higher education when faced with an educational transition are shaped by their previous academic performance, the decision is also influenced by other factors because students consider the costs and benefits of the different choices. The choices one makes about where to attend college have downstream consequences on the kinds of jobs they can get, how much they earn and their health. Since religious students have better academic performance in high school, we would expect them to make more ambitious choices about higher education. This is generally the case, except in one social class group: religious teens from the professional class.
When it comes to the transition to college, religious teens from the professional class make less ambitious choices about where to attend college than we would expect given their stellar report cards. This is especially the case for girls. They undermatch in the college selection process because educational decisions are social decisions that highlight the effect of the home environment on norms and values surrounding education. Religious teens, especially girls, make choices that reflect their familial and social ties, rather than making a choice to optimize their social class standing by getting a prestigious career. There are millions of young men and women who do not live to impress college admissions counselors. For them, it is God who matters.
Q: What should colleges and their admissions officers do with the information in your book?
A: I wrote this book because I want readers to, in Arlie Hochschild’s words, “scale the empathy wall.” We live in an increasingly polarized America, with religion being a central axis of division. Avowedly secular and deeply religious Americans don’t trust or think very highly of each other. Academia is overwhelmingly liberal (and secular) and tends to dismiss religion as a legitimate feature of life. Despite my own liberal politics, I agree with New York Times opinion writer Ross Douthat, who points out that the intelligentsia doesn’t get religion. While I don’t have a stake in the future of secularization, I do think it’s vital for higher education faculty and administrators to be more cognizant of how religion structures the lives of millions of American teens and their families. If you are a college admissions officer, you might not feel compelled by applicants who allude to overtly religious themes in their essays, or applicants whose moral foundations celebrate authority, loyalty and sanctity as opposed to helping the oppressed. Yet these students have important perspectives that are important for diversifying the academy.
But there is something even bigger at stake: religious polarization threatens the future of higher education. A 2017 Pew study found that among Republicans (many of whom would describe themselves as religious), only 30 percent felt warmly toward college professors. Here is another grim indicator of how unpopular higher education is among right-leaning Americans: in 2017, a majority of Republicans and Republican-leaning independents (58 percent) said that colleges and universities have a negative effect on the country.
This is a problem. We shouldn’t be so divided on the value of higher education institutions, especially since so much of what happens in life depends on a college degree. Colleges are ideal contexts for emerging adults to develop ties across religious lines, because universities are one of the only institutions left that bring Americans together who possess a diversity of ideas, perspectives and cultures. Religious students and atheists don’t need to agree on the truth about religion to share a common commitment to valuing civic engagement. Forging ties across religious lines is critically important because in civil society, members of different faiths need to get along. In other words, colleges need not be seen as places where the truth on religious matters is settled but rather as gathering points where civil comity is learned.