Men once had a higher tendency than women to derive satisfaction from paid work. Now that gender gap has nearly flipped—but only among the least religious.

Reading Time: 6 minutes

If you had enough money to live comfortably without working, would you do so? Researchers often use this type of question to measure “work ethic.” When Americans are asked this question, most say they would keep working anyway. 

But a fascinating difference has developed in the answers people give—a difference involving not only gender, but religiosity. 

The “gender revolution” over the last five decades has featured dramatic shifts in patterns of gender inequality, largely driven by women’s mass entry into the labor force. Women’s employment reached record highs, the gender pay gap declined, and women’s educational attainment reached and surpassed men’s. 

But recent research shows progress toward gender equality has slowed or even stalled, and its benefits remain unevenly distributed, largely because of the continued precarity of women’s paid work.

Societal beliefs about women’s work have long been a measure of gender equality, with recent scholarship focusing on trends in these attitudes to assess the progress (or lack thereof) of the gender revolution where we often see persistent inequities. For example, while women joined the workforce in rapidly rising numbers over the last half-century, men continue to work in the paid labor force more than women, and women’s work seems to remain more vulnerable to personal or societal shocks.

Cyrus Schleifer, Eman Abdelhadi, Samuel Perry, and I moved beyond the usual gender attitude questions used to measure change over time, considering women’s and men’s views toward their own work over the last half-century and how religion may complicate those views. 

Could different preferences help explain the persistent gender pay gap where men work more than women? And how might religion influence different work preferences among women and men so that the intersection of the two creates different trends over time?

In a study recently published in Sociological Science, we use a half-century of data from the nationally-representative General Social Survey to explore these questions. We found that men used to be more likely than women to report working for its own sake. But over time, this gender gap has declined to the point where it is now virtually nonexistent. Women became more likely over time to desire work for its own sake and men became less so, meeting in the middle.

In many arenas of social life, movement toward gender parity results primarily from women changing. But in this case, men actually changed more than women, with their declining desire to work even greater than women’s increase.

Men used to be more likely than women to derive satisfaction from work. But this gender gap is now virtually nonexistent. And that’s mostly a secular phenomenon.

One might have assumed that religion simply makes everyone feel like they should work even if they did not have to, but that is not the case. Religious involvement acts as a countervailing influence, bolstering the gap such that frequently-attending men and women have not yet converged in their desire to work. They remain in the patterns and preferences of the past, with men continuing to desire to work more than women.

In short, the meeting in the middle is largely a secular phenomenon, with the least religious Americans flipping the script and actually moving toward a reversal of the gender gap.

Traditional scripts frame women in the workplace as less than ideal, something to do if financially necessary but not because the work is intrinsically rewarding. Historically, this gender frame was reinforced by religion. Combining patterns for all years, we found a clear trend where men who attend religious services are much more likely to report working because it is satisfying, while women who attend are less likely to say the same. 

Again, it is the men who seem most impacted by both time and religion. Women are more likely to express satisfaction in working if they are less religious. But men’s declining desire to work over time and religion’s reinforcing role in gender differentiation appeared to be more pronounced among men.

We were interested not only in general trends over time or by religion, but also whether religion changed how things play out over time. Specifically, we wanted to know whether the gender gap had converged more for secular than religious people.

We found clear differences in whether and when gender convergence occurred depending on how religious people are. Among the most secular Americans, those who never attend services, gender parity was reached in the 1990s. In fact, among this group, women are now more likely than men to report working for its own sake, in large part because of secular men’s rapidly falling desire to work if it was not financially necessary. 

If we could imagine a world where everyone was secular and work was not a financial necessity, we might see a trend toward women being more likely to work than men.

Among those in the middle who attend services about once a month, convergence is happening at a similar rate as for the population as a whole—though a very small gap seems to persist. We were less interested in the middle than in the extremes—the least and the most religious.

Unlike nonreligious Americans (among whom the gender gap in work attachment has actually reversed) and moderately religious Americans (among whom the gender gap is virtually nonexistent), among the most religious the gender gap clearly persists with men being substantially more likely to work for its own sake.

Religion reinforces familism and gender complementarianism, which assigns women and men “complementary” roles linking masculinity with earning and providing, while femininity is associated with motherhood and care work. So we had already expected that the most religious Americans would have the most persistent traditional gender gap.

Although the most religious Americans have not yet converged, men’s dropping desire to work and women’s rising desire to work are society-wide trends, and even the most religious Americans could be expected to converge at some point in the future.

The gender gap in desire to work for pay has closed among all but the most religious Americans. At a moment when women disproportionately had their labor force participation curtailed in response to the COVID-19 pandemic, our results demonstrate that women do not simply have a preference for “opting out” of employment. Women desire to work for work’s sake at the same rate men do. Neither explicit gender attitudes considered in past research nor the gendered orientations toward work considered here can explain why women are more likely to leave work than men.

Yet women are, in fact, more likely to leave the workforce.

It appears that the structural barriers women face—such as a lack of free or affordable childcare and a lack of workplace flexibility and support, alongside men’s avoidance of care work—are pushing and pulling women out of the labor force and perpetuating gender inequality. At the same time, by examining how religious participation is implicated in the gender gap in work orientations, we demonstrated one of the ways the gender revolution’s effects have been uneven. Pockets of Americans, specifically the most religiously-active Americans, continue to have more gender-polarized orientations toward their employment. By attending to cultural institutions such as religion, we can trace the ways the gender revolution has been uneven and stalled both structurally and ideologically.

Religious participation clearly matters, but we might expect that the impact of religious participation varies by the type of religious institution in which it occurs. While we control for religious tradition in our primary results, in other analyses we looked at potential variation across some key religious subgroups. 

Overall, we found that never attending conservative and mainline Protestants have converged over time, with women trending toward more desire to work than men. But among the most frequently attending conservative and mainline Protestants, there is less convergence—especially among mainline Protestants, where instead of women trending toward more desire to work they are actually trending toward less desire to work. 

Among Catholics, never-attenders were already converged in the 1970s and remain similar today, whereas frequent attenders started out diverged but have converged over time. Among Black Protestants there is general gender similarity in desire to work across both time and attendance: in other words, Black Protestant women and men have been and remain similar in their desire to work regardless of time period or how religious they are.

These results suggest that the overall patterns were driven primarily by conservative and especially mainline Protestants (among whom regular attendance is more distinctive rather than just expected as it is among conservative Protestants). And although it would appear that religious involvement in historically white Protestant churches has reinforced traditional gendered orientations toward work, that has not been the case for historically Black Protestant churches.

Our study is not without limitations. As our focus was on social change over time and the GSS collects general population samples, there are only a handful of respondents from minority religious traditions in any given survey year, so by looking at on-average patterns, we end up focusing on the impact of religion for the dominant religious group (Christianity). We could, as discussed above, consider variation across various categories of Christians, but we couldn’t consider whether and how the patterns would be different for minority religious groups. This is an especially important point, as research has shown that religious service attendance can mean different things for different religious groups. Among Muslims in the United States—where attendance is not required of women and can be an indicator of greater engagement in public life—women who attend more frequently are actually more likely to work for pay

We would also like to highlight that the question about people’s orientation toward their work was only asked of those working. Additional analyses we conducted led us to suspect the convergence pattern might have been even more pronounced if the question were asked of everyone, but as it was not, we can’t know for sure how shifts in who was asked the question could have affected the patterns.Traditionalist institutions like religion contribute to unevenness in the gender revolution, but preferences cannot explain the persistent society-wide precarity of women’s work: Women now prefer to work for work’s sake at the same rate men do. And among the most secular Americans, women actually prefer to work more than men do.

Avatar photo

Landon Schnabel

Landon Schnabel is the Robert and Ann Rosenthal Assistant Professor of Sociology at Cornell University. A former minister, much of his research focuses on (non)religion, gender, sexuality, politics, social...