Intersectionality interrupted

When I was in college a lot of people were reading Black Feminist Thought, by the sociologist Patricia Hill Collins, which came out in 1990 (it’s now pushing 11,000 citations in Google Scholar, and Prof. Collins is a colleague in my department). That book helped popularize intersectionality, from the work of Kimberlé Crenshaw, a legal scholar.

Thinking about and acting on intersecting inequalities was a big issue in the 1990s. It motivated me to do my only research on social movements (the women’s suffrage movement), as well as my dissertation, which included an article on the intersection of race, class and gender in U.S. labor markets (related to work my advisor Reeve Vanneman and colleagues were doing back then).

Signs

Anyways, long story short: I was interested to see that the latest edition of the journal Signs (paywalled) is devoted to intersectionality, or the critical analysis of how different kinds of inequality and identity occur simultaneously. I haven’t kept up with the theoretical side of this work, which has drifted away from the statistical modeling vein we were mining.

First I read the essay by Catharine MacKinnon, whose work I’ve been teaching for years in courses on gender, theory, and inequality. Since I last paid attention, she did a lot of work on women and international law, and in the essay here she discusses rape and genocide in the Balkan wars. Just as she once asked some feminists (paraphrasing), “if rape is about violence and not sex, why doesn’t he just hit her?”, she now asks (paraphrasing), “if genocide is about wiping people out, why do they commit mass rape against women instead of just killing them?” Thanks in part to her legal and theoretical work, the idea of genocide as a national, racial or ethnic crime is linked to sex-based atrocities such as forced prostitution and impregnation.

Although it’s hard to read, I am a sucker for MacKinnon’s wordplay (and always hear her phrase, “Man fucks woman; subject verb object,” when I talk about subjects and objects). So I was drawn in by her introduction to intersectionality as a method, which included, “Talking about thinking about the way one thinks is complicated, in that one is doing what one is talking about doing at the same time one is talking about doing it.”

For example, as we think about how we think, she wants us to avoid confusing the products of inequality for their causes. She writes,

No question about it, categories and stereotypes and classifications are authentic instruments of inequality. And they are static and hard to move. But they are the ossified outcomes of the dynamic intersection of multiple hierarchies, not the dynamic that creates them. They are there, but they are not the reason they are there.

Anyway, I recommend the essay, which, in addition to rape and genocide, also discusses the conundrums of intersecting inequality in U.S. law, where race and gender discrimination each are illegal, but discrimination by race-and-gender simultaneously is somehow sometimes left out.

That last point draws heavily off the work of Kimberlé Crenshaw, who, along with Sumi Cho and Leslie McCall, guest-edited the special issue.

Interruption

And here, after recommending the issue and praising its authors, I offer a criticism: intersectionality has a writing situation. Everyone in academia has their jargon. But in this area there is a common aesthetic preference for extra words and clauses — including long words and clauses — that is a real barrier to entry for those who don’t spend a lot of their time reading it.

Here is the opening paragraph of the introductory essay, by the guest editors:

As intersectionality has emerged in a number of discursive spaces, the projects and debates that have accompanied its travel have converged into a burgeoning field of intersectional studies. This field can be usefully framed as representing three loosely defined sets of engagements: the first consisting of applications of an intersectional framework or investigations of intersectional dynamics, the second consisting of discursive debates about the scope and content of intersectionality as a theoretical and methodological paradigm, and the third consisting of political interventions employing an intersectional lens.

That’s 86 words. I think not much would be lost cutting it down to 42 words, like this:

In the growing field of intersectional studies, we identify three categories of work. First, there are applications of the intersectional framework and studies of intersecting inequalities. Second, there are debates about intersectionality itself as a paradigm. And third, there are intersectional politics.

That’s just an example chosen for convenience — there are worse and better passages in the various essays, and I don’t want to belabor it. I suspect that to many outsiders this problem seems obvious, but I don’t know how these writers see it. I think academics should try to say what they want to say as clearly and directly as possible. If this principle were directly weighed against the loss of nuance — and aesthetic satisfaction — it might entail, I hope the balance would tip in the direction of readability.

What is ‘nationally representative,’ and did Regnerus have it?

I’m off to Minneapolis to present a talk tomorrow on “The Regnerus Affair” at the Minnesota Population Center, subtitle: “Gay Marriage, the Supreme Court, and the Politics of Sociology.”

In my preparation, I was putting together notes from previous posts, the critique I co-authored with Andrew Perrin and Neal Caren, the infamous paper itself, and the media coverage of the scandal. And one piece of it I never really questioned got me thinking: his insistence that his dataset was a “a random, nationally-representative sample of the American population.” The news media repeated this assertion routinely, but what does it mean?

The data, collected by Knowledge Networks, are definitely not truly random. But not much is. They have standing panel of participants who get rewards for participating in a certain number of online surveys. The recruitment of the original panel is where the randomness comes in, with dialing (more or less) random phone numbers. But who chooses to be in it is not random, of course. What the firm does, then, is apply weights to the sample. That is, you don’t count each person as one person, you count them as a certain multiple of a person, so that the weighted total sample looks like the target population — in this case all noninstitutionalized American adults ages 18-39.

In the paper, Regnerus offers an appendix which compares his New Family Structures Study to the national population as represented in better, larger samples, such as the Current Population Survey (CPS). He writes:

Appendix A presents a comparison of age-appropriate summary statistics from a variety of socio-demographic variables in the NFSS, alongside the most recent iterations of the Current Population Survey, the National Longitudinal Study of Adolescent Health (Add Health), the National Survey of Family Growth, and the National Study of Youth and Religion—all recent nationally-representative survey efforts. The estimates reported there suggest the NFSS compares very favorably with other nationally-representative datasets.

So, he eyeballs the comparisons and determines the result is “very favorable.” I had previously eyeballed the first few rows of that table and reached the same conclusion. This is the distribution of age, race/ethnicity, region and sex from that table:

So, it looks very similar to the national population as counted by the benchmark CPS. But both of these surveys are weighted on these factors. That is, after the sample is drawn, they change the counts of people to make them match what we know from Census data (which are weighted, too, incidentally). So the fact that NFSS matches CPS on this characteristics just means they did the weights right, so far.

Think about it this way. If I collect data on 6 men and 4 women, it’s easy to call my data “representative” if I weight those 6 men by .83 and the 4 women by 1.25. The more variables you try to match on the harder the math gets, but the principle is the same.

But now I looked further down the table, and Regnerus’s data don’t compare “very favorably” to the national data on some other variables. Here are household income (from CPS) and self-reported health (from the National Survey of Family Growth):

This means that, when you apply the weights to the NFSS data, which produces comparable distributions on age, sex, race/ethnicity and region, you get a sample that is quite a bit poorer and less healthy than the national average as represented by the better surveys.

I was confused by this partly because according to the Knowledge Networks documentation on the NFSS, income was one of the weighting variables.

I don’t know how big an issue this is. Do you? And do you know of a standard by which a researcher or research firm can declare data “nationally representative” in this age of small, fast, low-response, online surveys?

 

Women’s Employment and the Decline in Marriage Are No Longer Related

Originally published on TheAtlantic.com.

For a few decades, women’s rising share of the workforce probably led to fewer women getting married. But that’s not the case anymore.

CBS

It is common knowledge—and true—that marriage rates are falling and unmarried parenting is becoming more common (nicely illustrated here). On the other hand, it is also common knowledge—but not true—that women’s employment rates have continued to rise in the last two decades (as illustrated here.)

In the long run of history, there is little doubt these trends are related: As women’s economic independence increased with better job opportunities, marriage became more optional and fewer women got (or stayed) married. But in the medium run, on the scale of a few decades rather than long eras, it’s not that simple.

Here are the trends in marriage and labor force participation for women using U.S. Census data going back to 1900.

Source: My analysis of Census data from IPUMS.

In the long run of the past 111 years, there certainly are more employed women and more single women. But the trends only moved strongly in the same direction for the three decades from 1960 to 1990, when the percent of women not married more than doubled from 18 percent to 43 percent and the percent in the labor force almost doubled from 41 percent to 76 percent. In the last two decades labor force participation has frozen while the percent not married has jumped another 7 points.

Here is the trick: Despite the real connection between non-marriage and employment—in which women don’t feel as strong a need to be married if they are employed—the lion’s share of rising employment has been among married women. Women’s employment opportunities made non-marriage more viable but also changed marriage. As the employment rates of married and non-married women grew more similar, the decline of marriage has made less of a difference to the total employment rate. Moving women from married to single doesn’t do much anymore. Here are the employment trends:

The American Stall
So we need to understand the stalled rise in employment because it may be the key to understanding progress toward gender equality generally.

In a previous post I suggested that stalled progress resulted from feeble work-family policy, anti-feminist backlash, and weak anti-discrimination enforcement. A recent analysis by economists Francine Blau and Lawrence Kahn lends support to the first: work-family policy. Economix writer Catherine Rampbell highlighted the paper, which tracked employment rates over 22 wealthy countries for two decades. During that time, U.S. women fell from sixth to 17th in labor force participation rates—rising just one percentage point while women in the average country increased 12 points.

Here are the labor force participation rates for the 22 countries for 1990 and 2010. Dots to the left of the blue line show countries where women’s labor force presence increased; dots to the right show decreases. At the extreme, for example, Ireland saw a jump from 45 percent to 72 percent.

Source: My chart from the Blau and Kahn paper.

What happened? One big change was the advance of several work-family policies. The average number of weeks of guaranteed parental leave increased from 37 to 57 in these countries, with the U.S. adding only a 12-week rule under the Family Medical Leave Act (covering only half the workforce). The average country on this list now provides a guaranteed 38 percent of parents’ wages while they’re on leave, while the U.S. provides none. Seven of the countries now protect a right to part-time work, and three-quarters guarantee equal treatment for part-time workers. Public spending on child care as a proportion of GDP increased by more than a third outside the U.S., and the average country now spends more than four-times as much as the U.S.

Together, based on the experience of these countries, Blau and Kahn estimate these changes account for more than a quarter of U.S. women’s slippage relative to other countries. That’s not everything, but it’s a substantial bite. If we had kept up with the average country’s policies, U.S. women would have had an 82 percent labor force participation rate, putting them at 11th on the list instead of 17th.

On the Other Hand
Not all work-family policies are the same. One way to divide them is between those that protect time out of paid work (parental leave, part-time protections) and those that protect time in paid work (especially state-supported childcare). As Blau and Kahn note, U.S. women have much lower rates of part-time work than those in most other rich countries, but we also have higher rates of women in professional and managerial jobs. That might be because employers in those countries are reluctant to hire or promote women who are expected to take time out of the labor force when they have children—which is exactly the goal of some of our low-fertility peer countries. How, and whether, such policies can improve family life while also promoting gender equality is the subject of a rich debate—which unfortunately remains in the realm of the hypothetical here in the U.S.

Married women learning that paid work pays

The economist Raquel Fernández has a new paper out called, “Cultural Change as Learning: The Evolution of Female Labor Force Participation over a Century” (published version here, free version here). If I understand it, though, “female” labor force participation really only refers to married women. Correct me if you understand this better than I do and I’m wrong, but I think that’s a problem for the theory.

The basic point is that married women learn from the experience of others, producing a generational change in employment rates as positive experiences transmit to younger cohorts. As it became more culturally acceptable for married women to have jobs, the cultural effect accelerated, but it reached a saturation point resulting in the stalled progress toward higher employment rates among (married) women. Here are the trends she uses:

The normative survey question she relies on is about whether it’s OK for a woman to work “if her husband can support her.” The S-shape of labor force participation rates is supposed to be consistent with the cultural transmission theory (rather than being caused by, for example, anemic work-family policy, anti-feminist backlash, or hollow anti-discrimination enforcement).

But I don’t see anything in the paper about increasing non-marriage (now about twice as common as in 1960), or about labor force participation rates for single women. Shouldn’t economists be concerned about that kind of selection issue? In fact, labor force participation rates for single women have stalled, too, as my figure shows:

I don’t think attitudes toward married women’s work — or anything about marriage alone — are going bear the burden of explaining two decades of stalled progress into the labor force for both single and married women. I’m happy to have cultural explanations as part of the mix here, but I don’t think this one will do it.

 

 

Gender and family time: change and stall visualized

The Pew Research Center put out a report this month titled, “Modern Parenthood: Roles of Moms and Dads Converge as They Balance Work and Family.” It analyzes trends in time use among men and women in families, showing the big changes since the 1960s, and adds Pew’s own survey data on attitudes and perceptions. Lots of interesting information.

But what jumped out at me was that the stall in progress did not feature much in Pew’s narrative, written by Kim Parker and Wendy Wang. I really noticed that when the Joy Cardin show featured the report on Wisconsin Public Radio, and Cardin’s intro was this:

Family gender roles are converging, according to a new survey from the Pew Research Center. Father’s have more than doubled the time they spend on housework. More moms are paid to work outside the home. (The audio is here.)

Those facts are true, but old news – older than the new news, which is that nothing much has happened since the early 1990s. Here are the trends, in Pew’s nice graphics. See if you can find the stall point in each figure.

The last one, parents’ child care time, is the only one that shows continued real progress, albeit slower, in the last decade.

I favor three explanations for this gender stall:

• Work-family policy, as described by Stephanie Coontz here.
• Cultural trends toward “egalitarian essentialism,” which “blends aspects of feminist equality and traditional motherhood roles” (e.g., intensive parenting mania), as described by David Cotter, Joan Hermsen and Reeve Vanneman here.
• Weaker government enforcement of anti-discrimination law, as described in the new book Documenting Desegregation, by Don Tomaskovic-Devey and Kevin Stainback.

These explanations do not exclude others.

“No differences” survives the Regnerus paper

Coming soon (or at least sometime in the future): An article by Andrew Perrin, Neal Caren and myself, now accepted by the Journal of Gay and Lesbian Mental Health, ”Are Children of Parents Who Had Same-Sex Relationships Disadvantaged? A Scientific Evaluation of the No-Differences Hypothesis.”

Here is the abstract:

In a widely publicized and controversial article, Regnerus seeks to evaluate what he calls the “‘no-differences’ paradigm” with respect to outcomes for children of same-sex parents. We consider the scientific claims in Regnerus in light of extant evidence and flaws in the article’s evidence and analytical strategy. We find that the evidence presented does not support rejecting the “no-differences” claim, and therefore the study does not constitute evidence for disadvantages suffered by children of same-sex couples. The state of scientific knowledge on same-sex parenting remains as it was prior to the publication of Regnerus.

I have posted a preprint of the article here.

Book review: The Rise of Women, by DiPrete and Buchmann

Originally published on TheAtlantic.com.

(Charles Dharapak/AP Images)

The Rise of Women: The Growing Gender Gap in Education and What it Means for American Schools is both ambitious and modest in its goals: Sociologists Thomas DiPrete and Claudia Buchmann provide an ambitious analysis of why and how girls are outperforming boys in high school and going on to get a disproportionate share of college degrees. However, the authors modestly remain within their subject matter and avoid the unsupported claims about women’s looming social dominance that have inflated much of the conversation about gender dynamics today.

This allows us to have a reasonable, valuable conversation about an important problem: the failure of the education system to help a majority of students to reach their academic potential. We clearly do not have a problem of over-education among women. Even among Whites alone, women as well as men are graduating college at rates lower than those in the most educationally advanced societies (which used to include the United States). Rather, we have a dysfunctional system that underperforms for men more than for women.

Rather than focusing on the full range of educational failures, DiPrete and Buchmann focus on a low-hanging fruit policy question: How can we improve college degree attainment for the approximately one-third of students who are ready to graduate college but do not, because they do not have the resources, they change their minds for some reason, or they are not adequately supported in the endeavor?

Women up

Since the 1980s, women have gotten the majority of bachelor’s degrees. That’s mostly because they also perform better in high school, getting better grades and taking more advanced courses. DiPrete and Buchmann set aside the issue of the potential cognitive advantages of girls, which may or may not be “innate.” Such differences are too small and stable to account for the rapid change and large advantage in educational attainment women now hold. The reasons we do not have more people completing college—and gaining more skills and knowledge to enrich their lives—are not genetic or biological, but rather social and economic. We can do better, for both men and women.

While women have continued their upward historical educational trajectory since World War II, men’s achievement of college degrees stagnated—coinciding historically with the growing necessity of having higher education for economic security. If you ever needed proof that majorities of people do not respond in predictably self-interested ways to economic incentives, it is the stagnation of male college graduation rates even as the returns to a college degree spiked upward.

DiPrete and Buchmann’s sensible policy suggestions draw from this key insight: The difference between men and women, and how it has changed, can best be understood by studying differencesamong men and women—within genders. That means we don’t just study what family, school, and environmental effects matter, but who is most strongly affected by such differences in the social context.

One important lesson: Schools with high overall performance have a smaller female advantage. That leads to the straightforward conclusion that we can address the gender gap partly by increasing the quality of schools across the board. Easier said than done, but no less important—or less true—for it.

Men up

It is important to connect women’s educational rise with the other trends that have upended gender relations in the U.S., and the authors do an admirable job of tying these in. In particular, the rise in women’s employment opportunities, the decline or delay in marriage, and falling fertility rates have all increased the incentives for (and ability of) women to complete college. And, of course, the rise in education has in turn fueled these other developments as well. For example, college graduate women as well as men are more likely to get (and stay) married than those who completed high school only. Maybe by getting a college degree they improve their marriage-market options—and reduce the odds that they will divorce by increasing the educational parity in their marriages.

While the title of DiPrete and Buchmann’s book is overly dramatic, the subtitle is appropriately limited: The Growing Gender Gap in Education and What it Means for American Schools. Because although women are more likely to graduate college and get some advanced degrees than men are today, there is nothing in this trend that implies women will surpass men in overall earnings or economic (much less political) power in the foreseeable future.

Education, especially measured at the bachelor’s degree level, is merely one indicator in a whole suite of gender dynamics in which men overwhelmingly dominate. Further, women’s educational advantage is not so great that they will overcome the labor-market advantages that men have at all educational levels, the imbalances within families that persist today, or the tendency of women to end up in less lucrative fields of study and thus occupations.

The biggest problem for gender inequality among the college-educated remains the lack of gender integration across fields of study, which stalled in the 1980s. Men and women still largely educate themselves in different fields, with dramatic implications for their career trajectories and earnings throughout their lives. Segregation in fields of study is closely related to the issue of occupational segregation in the labor market. Both reflect a complex combination of choices and constraints made in varying social contexts—with decisions made early in life producing irreversible effects. In the latest reports, women are just 26 percent of workers in computer and math-related professional occupations and 14 percent of those in architectural and engineering professions.

And DiPrete and Buchmann’s analysis helps understand this stubborn problem. They report that high school is the key location to understand major-field segregation. Among high school boys and girls with strong interest in science and technology fields, there is no gender gap in the likelihood of completing such a major. The difference is in the rates of intention to major in those areas. Between 8th and 12th grade, girls lose interest in science, technology, engineering, and mathematics (STEM fields, for short) much more than boys do.

Women’s desire for people-oriented work, for work that is intrinsically interesting, and for occupations that permit work-family balance cannot fully explain their lower rates of majoring in STEM-related fields. Rather, the major source of the difference is that women do not express interest in STEM-related careers while in high school—and that is not because high school girls are not as good at math and science. Instead, the difference may be that boys believe they are better at math and science, especially math. The key policy insight in this area is that science-intensive high school environments greatly increase girls’ interests in physical science and engineering-related careers.

This is an important book, and although somewhat technical in its analysis sections it deserves a wide readership.

I have two minor complaints about The Rise of Women. The first is over its insistent focus on the four-year college degree and the economic benefits it brings. The fact that women receive more bachelor’s degrees than men but continue to earn less money confirms that a bachelor’s degree is not a first-class ticket to labor-market success. Although this helps to focus the book, it also distracts from the more universal problems we have, including an obsession with the material benefits of education.

DiPrete and Buchman conclude that we need to find ways to motivate students in middle and high school to devote more energy to their studies, by improving the quality of education as well as the quality of information students have to make the connection between what they learn in school and their future career ambitions. Too many boys don’t cognitively grasp that the difference between merely making it versus excelling through high school is measured in higher education success and potential career satisfaction. Finding ways to get this across might really help their motivation to work harder, the authors argue. But truly high-quality education takes students beyond such material calculations into the realm of the intrinsic beauty of discovery, the power of wonder, and the search for knowledge as a key to life, the universe and everything.

My second knock is that the authors seem not to notice the broad trend of slowing advances for women. For example, even though their charts show it, they don’t mention that the share of law and medical degrees earned by women slowed and then peaked in the early 2000s—and has declined since. Naturally, that is not the central concern of a study devoted to understanding women’s advantages. But in the context of the general gender stall, it’s important to realize that women’s progress across many areas is highly interrelated.

Why taller-wife couples are so rare

Originally published on TheAtlantic.com.

It’s not just because women are, on average, shorter than men.

Keith Urban and his wife Nicole Kidman arrive at the 2009 American Music Awards. (Chris Pizzello/AP Images)

Men are bigger and stronger than women. That generalization, although true, doesn’t adequately describe how sex affects our modern lives. In the first place, men’s and women’s size and strength are distributions. Strong women are stronger than weak men, so sex doesn’t tell you all you need to know. Otherwise, as retired colonel Martha McSally put it with regard to the ban on women in combat positions, “Pee Wee Herman is OK to be in combat but Serena and Venus Williams are not going to meet the standard.”

Second, how we handle that average difference is a matter of social construction: We can ignore it, minimize it, or exaggerate it. In the realm of love and marriage, we so far have chosen exaggeration.

Consider height. The height difference between men and women in the U.S. is about 6 inches on average. But Michael J. Fox, at five feet, five inches, is shorter than almost half of all U.S. women today. On the other hand, at five-foot-ten, Michelle Obama is taller than half of American men. So how do people match up romantically, and why does it matter?

Because everyone knows men are taller on average, straight couples in which the man is shorter raise a problem of gender performance. That is, the man might not be seen as a real man, the woman as a real woman, if they don’t (together) display the normal pattern. To prevent this embarrassment, some couples in which the wife is taller might choose to be photographed with the man standing on a step behind the woman, or they might have their wedding celebrated with a commemorative stamp showing her practically on her knees—as the British royals did with Charles and Diana, who were both the same height: five foot ten.

 But the safer bet is just to match up according to the height norm. A new study from Britain—which I learned of from the blogger Neuroskeptic—measured the height of the parents of about 19,000 babies born in 2000. They found that the woman was taller in 4.1 percent of cases. Then they compared the couples in the data to the pattern found if you scrambled up those same men and women and matched them together at random. In that random set, the woman was taller in 6.5 percent of cases. That means couples are more often man-taller, woman-shorter than would be expected by chance. Is that a big difference? I can explain.

For illustration, and to compare the pattern with the U.S., I downloaded the 2009 Panel Study of Income Dynamics, a U.S. survey that includes height reported for 4,600 married couples.* These are the height distributions for those spouses, showing a median difference of 6 inches.

Clearly, if these people married (and didn’t divorce) at random we would expect the husband to be taller most of the time. And that is what we find. Here is the distribution of height differences from those same couples:

The most common arrangement is the husband five to six inches taller, and a small minority of couples—3.8 percent—are on the left side of the red line, indicating a taller wife.

But does that mean people are seeking out taller-husband-shorter-wife pairings? To answer that, we compare the actual distribution with a randomized outcome. I made 10 copies of all the men and women in the data, scrambled them up, and paired them at random. This is the result:

Most couples are still husband taller, but now 7.8 percent have a taller wife—more than twice as many.

Here are the two distributions superimposed, which allows us to see which arrangements are more or less common in the actual pairings than we would expect by chance:

Now we can see that from same-height up to “man 7 to 8 inches taller”, there are more couples than we would expect by chance. And below same-height—where the wife is taller—we see fewer in the population than we would expect by chance. (There also are relatively few couples at the man-much-taller end of the spectrum—at 9 inches or greater—where the difference apparently becomes awkward, a pattern also seen in the British study.)

Humans could couple up differently, if they wanted to. If it were desirable to have a taller-woman-shorter-man relationship, it could be much more common. In these data, we could find shorter husbands for 28 percent of the wives. Instead, people exaggerate the difference by seeking out taller-man-shorter-woman pairings for marriage (or maybe the odd taller-woman couples are more likely to divorce, which would produce the same result).

What difference does it make? When people—and here I’m thinking especially of children—see men and women together, they form impressions about their relative sizes and abilities. Because people’s current matching process cuts in half the number of woman-taller pairings, our thinking is skewed that much more toward assuming men are bigger.

* I must note that Dalton Conley and Abigail Weitzman have a forthcoming paper for the 2013 Population Association of America conference on height differences, which also uses the PSID data, as well as the National Longitudinal Survey of Youth. I haven’t seen the paper, but the abstract is here.

Poverty Poses a Bigger Risk to Pregnancy Than Age

Originally published in TheAtlantic.com.

The problem of income inequality often gets forgotten in conversations about biological clocks.

The dilemma that couples face as they consider having children at older ages is worth dwelling on, and I wouldn’t take that away from Judith Shulevitz’s essay in the New Republic, “How Older Parenthood Will Upend American Society,” which has sparked commentary from Katie Roiphe,Hanna Rosin, Ross Douthat, and Parade, among many others.
The story is an old one—about the health risks of older parenting and the implications of falling fertility rates for an aging population—even though some of the facts are new. But two points need more attention. First, the overall consequences of the trend toward older parenting are on balance positive, both for women’s equality and for children’s health. And second, social-class inequality is a pressing—and growing—problem in children’s health, and one that is too easily lost in the biological-clock debate.

Older mothers

First, we need to distinguish between the average age of birth parents on the one hand versus the number born at advanced parental ages on the other. As Shulevitz notes, the average age of a first-time mother in the U.S. is now 25. Health-wise, assuming she births the rest of her (small) brood before about age 35, that’s perfect.

Consider two measures of child well-being according to their mothers’ age at birth. First, infant mortality:

Source: Centers for Disease Control.

Health prospects for children improve as women (and their partners) increase their education and incomes, and improve their health behaviors, into their 30s. Beyond that, the health risks start accumulating, weighing against the socioeconomic factors, and the danger increases.

Second, here is the rate of cognitive disability among children according to the age of their mothers at birth, showing a very similar pattern:

Source: Calculations made for my working paper, available here. To match up children with their birth parents in the Census, I had to limit the sample to children living with two married parents, where both are in their first marriage, so it’s a pretty select group.

Again, the lowest risks are to those born when their parents are in their early 30s, a pattern that holds when I control for education, income, race/ethnicity, gender, and child’s age.

When mothers older than age 40 give birth, which accounted for 3 percent of births in 2011, the risks clearly are increased, and Shulevitz’s story is highly relevant. But, at least in terms of mortality and cognitive disability, an average parental age in the late 20s and early 30s is not only not a problem, it’s ideal.

Unequal health

But the second figure above hints at another problem—inequality in the health of parents and children. On that purple chart, a college graduate in her early 40s has the same risk as a non-graduate in her late 20s. And the social-class gap increases with age. Why is the rate of cognitive disabilities so much higher for the children of older mothers who did not finish college? It’s not because of their biological clocks or genetic mutations, but because of the health of the women giving birth.

For healthy, wealthy older women, the issue of aging eggs and genetic mutations from fathers’ run-down sperm factories is more pressing than it is for the majority of parents, who have not graduated college.

If you look at the distribution of women having babies by age and education, it’s clear that the older-parent phenomenon is disproportionately about more-educated women. (I calculated these from the American Community Survey, because age-by-education is not available in the CDC numbers, so they are a little different.)

Most of the less-educated mothers are giving birth in their 20s, and a bigger share of the high-age births are to women who’ve graduated college—most of them married and financially better off. But women without college degrees still make up more than half of those having babies after age 35, and the risks their children face have more to do with high blood pressure, obesity, diabetes, and other health conditions than with genetic or epigenetic mutations. Preterm births, low birth-weight and birth complications are major causes of developmental disabilities, and they occur most often among mothers with their own health problems.

Most distressing, the effects of educational (and income) inequality on children’s health have been increasing. Here are the relative odds of infant mortality by maternal education, from 1986 to 2001, from a study in Pediatrics. (This compares the odds to college graduates within each year, so anything over 1.0 means the group has a higher risk than college graduates.)

This inequality is absent from Shulevitz’s essay and most of the commentary about it. She writes, of the social pressure mothers like her feel as they age, “Once again, technology has given us the chance to lead our lives in the proper sequence: education, then work, then financial stability, then children”—with no consideration of the 66 percent of people who have reached their early 30s with less than a four-year college degree. For the vast majority of that group, the sequence Shulevitz describes is not relevant.

In fact, if Shulevitz had considered economic inequality, she might not have been quite as worried about advancing parental age. When she worries that a 35-year-old mother has a life expectancy of just 46 more years—years to be a mother to her child—the table she consulted applies to the whole population. She should breathe a little bit easier: Among 40-year-old white college graduateswomen are expected to live an average extra five years compared with those who have a high school education only.

When it comes to parents’ age versus social class, the challenges are not either/or. We should be concerned about both. But addressing the health problems of parents—especially mothers—with less than a college degree and below-average incomes is the more pressing issue—both for potential lives saved or improved and for social equality.

Conservatives doubling down on single mothers and crime

The reaction from the right-wing family folks to my post about single mothers and crime.

In a post that appeared here and on TheAtlantic.com — but got extra attention when the figure appeared on Jezebel, Mother Jones, Slate, Andrew Sullivan’s Dish, Family Scholars, and Alas! — my point was that conservatives had been quick to attribute the rise in crime in the 1980s to the “breakdown of the family” and single parents. But then they didn’t revise their stories when crime dropped drastically without a reversal of the family structure trends. I thought single mothers deserved an apology.

They’ll have to keep waiting, I guess.

Within a few hours on Tuesday, we heard from some of the leading lights on the issue. Their conclusion was that if I don’t want more marriage then I must want more incarceration.

Maggie (let’s “drive a wedge between blacks and gays!”) Gallagher, on Family Scholars:

The ending of the violent crime wave was a great policy achievement. It seems however to involve massively incarcerating millions of young men we have not succeeded in civilizing. There is never only one way to skin a cat. But I’m not sure our souls should rest too comfortably on the solution we found.

Ross (Charles Murray is “brilliant!”) Douthat tweeted:

Fortunately, family breakdown doesn’t create any problems that mass incarceration of young men can’t solve.

(One good turn deserves another, I guess: Murray himself tweeted back: “Lovely. Menckenesque. Maybe even Wildeian (if Oscar had been a social scientist).”

Soon Elizabeth Marquardt, who is always committed to “excellent arguments and accurate data,” chimed in:

Well for pete’s sake, one thing we now do is lock up a lot more of those fatherless boys and throw away the key. Which means less crime.

Then Brad (recession is good news for marriage!) Wilcox produced a 10-year old graph of incarceration rates under the title, “Who Needs An Intact Family? Jail Will Do Just Fine,” suggesting that we might be “forced to choose between a stronger marriage culture and mass incarceration.”

What followed on all these blogs was a debate about all the interesting possible explanations for the decline of crime in America. Lost in all of it was that none of the old “family values” folks backed off the original story that crime rose in the first place because of family “breakdown.” Just to clarify: It is one thing to say that children who grow up with one parent present are more likely to commit violent crime. As a statistical association that is true, though the causal story is muddied by important confounds. But it is another — unjustified — thing to say that the crime wave of the 1980s and 1990s was substantially driven by the rise of single parents, which was a common claim, as I illustrated in the post.

Ballpark-acts of violence

(Here is the part where I spend more time thinking about the actual evidence than the pundits I’m complaining about. What follows isn’t scientific analysis: this is data ballparking and rumination.)

As Stephen Demuth and Susan Brown wrote in a 2004 article, identifying the causal effect of family structure itself on whether kids become violent is very difficult. You need to consider parental monitoring and supervision, the quality of the relationship between parents and children, the level of conflict in the home, as well as poverty, education, family transitions, housing, neighborhood factors, and so on.

In that article, however, they offered some numbers we can look at to ballpark the relationship between single-parents and violent crime.

Using the National Adolescent Survey of Adolescent Health, they added up the self-reported violent acts of students in grades 7 to 12 in 1995. The kids were asked how many times in the last year they (1) “hurt someone badly enough to need bandages or care from a doctor or nurse,” (2) “use or threaten to use a weapon to get something from someone,” and (3) “take part in a fight where a group of your friends is against another group.” They didn’t ask the specific number of times, but rather asked kids to report “0 (never), 1 (once or twice), 2 (three or four times), or 3 (five or more times).”

That’s enough to get a sense of what was going on in 1995, when the survey was done. If you take the scores on the three items added together, and take the mean of that sum for kids in different family structures, you can ballpark how many of these acts of violence kids in each family type committed. The idea is just to get a sense of the magnitude of the family structure difference and the relative contribution of kids in each group.

I’ve turned the scale into the average number of acts to make the figure below. For example, the mean for children living with single mothers was 1.2 on the scale. I gave them 1.5 (for “once or twice”), and then, for the additional 0.2, added 0.2*2 (with 2 representing the difference between 1.5 and 3.5 for “three or four times”). That’s a total of 1.9 “ballpark-acts”:

To show the relative contribution of the different groups of kids, I set the width of the bars to their share of the sample, in which 26% were living with either a single mother or single father, which is about right for that period. You can see from the estimate, for example, that kids living with single mothers admitted to 0.7 more ballpark-acts of violence per year, and those with single fathers admitted to 1.1 more.

With the amount of violence committed by each kid, and the relative size of the groups, it’s easy to calculate that those 26% of kids living with single parents committed 35% of the violent acts — more than their share, but not most of the total.

Remember, there are no controls in that analysis – it’s just compositional. There are a host of real reasons children with single parents commit more violence. This is just to see how big is the difference we are trying to understand.

Still, I have no trouble believing the decline of married-couple living arrangements contributed to the rise in violent crime rates since the 1960s, especially if you set aside the huge spike in violence from the mid-1980s to the mid-1990s. Single-parent families have lots of challenges and shortages — mostly of money and time — that make it harder on average to keep their kids in line. (And, of course, one of the causes of single-parenthood is the rise of incarceration.) In the absence of sufficient big-government nanny-state infrastructure to help them get by, it’s not surprising that the problems they experience include a higher risk of violence.

But I see nothing to justify the apocalyptic end-of-civilization associations that were commonly served up then — and which apparently are still palatable to many of today’s family conservatives.

Aside: scale

In that Demuth and Brown article, they have multivariate models that allow us to compare effect sizes on violence. They show, for example, that having a parent who graduated college is associated with odds of committing violent acts 3.5-times as much as living with a single mother, once other factors are controlled (-.42 versus .12). It’s always easier to say there is an association than to ascertain how important it is relative to other things — and especially for how it contributes to a trend over time.

Consider some other findings in this literature that allow us to compare the scale of effects:

• A study of religion and family structure effects on delinquency (an amalgamated concept) reported that living with a single parent increased the odds of delinquency about as much as 6 years of education reduced the odds.
• A study from one school district found that, for boys (but not girls), having a single mother increased the odds of being referred for delinquency by 13%, compared with 9% for being poor.
• Many studies — like this, this, and this, find no effect of family structure on delinquency once aspects of the parent-child relationship are controlled.

A common finding is that the level of attachment children feel with their mothers swamps the family structure effect, as was found in this study, in which there was no effect of family structure on serious or non-serious delinquency. Some, like this one, find that living with a single mother has a significant effect on delinquency, but it is smaller than the effect of maternal closeness.

Given the importance of factors such as poverty level, parents’ education, and peer and network effects – as well as family relationships – the more important and useful questions have to do with how we can improve the environment and outcomes for children with single parents (and for their parents). The causal impact of family structure itself is less pressing.