The Family, Measured SOC 1113 · Trend Lab · Midwestern State University

The most personal institution in your life is also the most measured. Marriage, divorce, who meets whom, who scrubs the pan, and what family structure does or does not do to children: all of it has published numbers, and most of what people confidently believe about those numbers is wrong. Every figure here is hidden until you commit a guess or a prediction. The last tab performs an autopsy on the most politically charged family claim in America, using the method this course has been building since Chapter 2.

The family since 1960

In 1960 the median first-time bride was 20 years old. Start with a guess about now.

What is the median age at first marriage for American women today? Drag your guess, then lock it in.
24.0

Median age at first marriage

Women, 1960
20.3
Women, 2024
28.6
Men, 1960
22.8
Men, 2024
30.2
Second call. Compared with 1980, is the U.S. divorce rate today higher, about the same, or lower?

Divorces per 1,000 people

1979, the record year
5.3
2023
2.4
72% → 51%
of adults married, 1960 to 2010, and it has hovered near half since
18% → 40%
of births to unmarried mothers, 1980 to 2023
13% → 28%
of households that are one person, 1960 to 2021
18%
of Americans live in multigenerational households, more than double the 1971 share
Read the direction carefully. Marriage is later and rarer, divorce has fallen by more than half from its peak, and household forms have multiplied. The family is not disappearing. It is diversifying, and each form fits an economy, exactly as the nuclear family fit the 1950s. The divorce miss is the one to remember: most rooms guess higher, and the myth that half of marriages end has outlived the data by decades.

Sources: Census Bureau, Table MS-2, median age at first marriage (1960 and 2024). Pew Research Center, Barely Half of U.S. Adults Are Married (2011), and Census data since. CDC/NCHS: divorce record 5.3 per 1,000 (1979), 2.4 per 1,000 in provisional 2023 data. NCHS: births to unmarried women 18.4% (1980), 40.0% (2023). Census: one-person households 13% (1960), 28% (2021). Pew: multigenerational households 18% of the population (2021).

How couples actually meet

Stanford's How Couples Meet and Stay Together survey has asked American couples the same question for years: how did you two meet? Before the chart appears, make a call about the most recent data. The options are alphabetical.

By 2017, what was the number one way heterosexual couples met?

Share of couples who met each way, 1995 and 2017

Couples can name more than one way, so the columns do not sum to 100.

Every traditional matchmaker is in decline. Friends, family, church, school, and the workplace all introduced fewer couples in 2017 than in 1995, and one channel replaced them. Sociologists call the old pattern intermediation: your network vetted your partners before you met them. The apps removed the intermediaries. Here is the part worth sitting with: homogamy survived. People on the apps still overwhelmingly match with people of similar education and class. The matchmaker changed. The marriage market's rules mostly did not.

Source: Rosenfeld, Thomas, and Hausen, Disintermediating your friends (PNAS 2019), Table S3, heterosexual couples. Bar or restaurant figures include couples who first connected elsewhere, including online, before meeting there.

Who does the dishes

Arlie Hochschild named the second shift in 1989: the unpaid workday that starts when the paid one ends. The Bureau of Labor Statistics measures it every year by having Americans keep time diaries. On days when women do household activities, they average 2.8 hours.

On days when men do household activities, what do they average? Drag your guess in hours, then lock it in.
2.5 h

Time diaries, average day, 2024

Did any household activities: women
87%
same: men
75%
Housework hours: women
0.81
same: men
0.30
Cooking and cleanup: women
0.91
same: men
0.51
Childcare, kids under 6: women
2.8
same: men
1.7
The home is a workplace with a division of labor. These are national time diaries, not opinions, and every row splits the same direction. The family does not sit outside society's power structures. It reproduces them, hour by hour, at the kitchen sink. Hold this tab in mind for the class and gender chapters: both start exactly here.

Source: Bureau of Labor Statistics, American Time Use Survey, 2024 results (released June 2025). Hours are per day: housework and cooking rows are averages across all adults, the 2.8 and 2.1 figures are averages on days the person did household activities, and childcare rows are adults living with children under 6.

The autopsy: family structure and crime

You have heard the claim. Kids from single-parent homes turn to crime, so family breakdown explains the crime rate. It is one of the most repeated family claims in American politics, and it is exactly the kind of claim this course taught you to cut open in Chapter 2. Work the steps in order. No skipping ahead.

First, predict: when researchers compare adolescents from single-parent and two-parent families, is there a raw statistical association with criminal involvement?
1The association is real

A systematic review of 48 studies (Kroese and colleagues, 2021) finds that growing up in a single-parent family is associated with an elevated risk of criminal involvement in adolescence. Honest sociology does not deny an inconvenient correlation. It asks the next question.

2What else differs between these families?
Poverty rate, single-mother families
21%
Poverty rate, married-couple families
~4%

A roughly five-fold poverty gap (Census, 2024), before you count neighborhood, school quality, parental time, and the conflict that usually precedes a divorce. Comparing single-parent and married-couple households is never a comparison of family structure alone. Chapter 2 vocabulary: the claim is loaded with confounders, and families are not randomly assigned to their structures. That is a selection effect.

3Control for what parents actually do, and watch the gap move

Demuth and Brown (2004) tested this on a national adolescent sample. Delinquency is higher in single-parent families, but the differences shrink once you account for family processes: closeness, involvement, supervision, and monitoring. And the difference between single-father and single-mother families disappears entirely once processes are controlled. What predicts delinquency is what parents do and what their circumstances let them do, far more than how many adults are on the lease.

4The verdict

The correlation is real and modest. The causal story told on cable news is not established: the Kroese review itself concludes that more research is needed to separate the effects of divorce, parental death, and being born to a single parent. Family structure in these studies stands in for money, time, conflict, neighborhood, and stress. Blaming the structure lets everything it stands in for off the hook.

And one Chapter 8 question to carry out the door: which families get labeled as the crime problem, and which confounders, poverty, neighborhood, policing, never make it into the headline? A charged correlation plus a ready-made villain is how moral entrepreneurs work. Now you can take one apart.

Sources: Kroese, Bernasco, Liefbroer, and Rouwendal, Growing up in single-parent families and the criminal involvement of adolescents: a systematic review (Psychology, Crime and Law, 2021, 48 studies). Demuth and Brown, Family Structure, Family Processes, and Adolescent Delinquency (Journal of Research in Crime and Delinquency, 2004). Poverty rates by family type from Census Bureau data as published for 2024. This tab presents published findings and does not report any new analysis.

Questions to take with you

  1. Which guess did you miss by the most across the tabs, and in which direction? Most rooms miss divorce high and men's housework high. What keeps those myths alive?
  2. Marriage at 28 instead of 20 changes what a marriage is for. Name two things a 1960 marriage provided that people now assemble before marrying, and what that does to the institution.
  3. The apps removed friends, family, and church from matchmaking, yet homogamy survived. What does that tell you about where the marriage market's rules actually live?
  4. Run the second shift through Chapter 5: which agents of socialization teach children who does the dishes, and how would you expect the 2054 time diaries to look?
  5. Reproduce the autopsy from memory in four sentences: association, confounders, controls, verdict. This is the exam skill, and it works on any charged correlation.
  6. Find one claim about families in your own feed this week and name the step of the autopsy it skips.