Living together before or without marriage

Three numbers, three questions

Cohabitation means living with a romantic partner without being married to that partner. A current-status question asks whether someone lives with an unmarried partner today. A lifetime-history question asks whether they have ever done so. Those are different denominators and time windows. Pew's 2019 report cited a Current Population Survey estimate that 7% of all U.S. adults were then living with an unmarried partner. The same report's separate relationship-history analysis found that 59% of adults ages 18 to 44 had ever lived with an unmarried opposite-sex partner.

Pew published a newer analysis in 2026 using a study that has followed the same birth cohort since the late 1990s. Among adults ages 38 to 43 in 2023, 77% had lived with an unmarried romantic partner at some point. That group had more years to have the experience than many adults in an 18-to-44 group, and the studies use different designs. Do not read 59% and 77% as a 18-percentage-point rise in national cohabitation. They are snapshots of different people, ages, and measurement histories.

Picture one person who lived with a partner at 25, separated at 30, and lives alone at 40. They count in an 'ever cohabited' measure but not in a current cohabitation measure at 40. Another person may live with a partner now and also count in a lifetime measure. The distinction explains how the 7% current figure can be much smaller than a lifetime figure without either survey contradicting the other.

What the long-running cohort adds

The newer Pew study analyzed the National Longitudinal Survey of Youth 1997, which has returned to the same participants every year or two. The respondents were 38 to 43 in 2023. Following people lets researchers collect living-arrangement history as it develops instead of asking everyone in middle age to reconstruct every earlier relationship at one interview. It remains one birth cohort, so its 77% should not be presented as the rate among all U.S. adults in 2023.

Of those cohort members, 40% had lived with one unmarried partner, 22% with two, and 15% with three or more by the observed age. These categories add to the 77% who had ever cohabited. The remaining group had not reported that experience by this point in the study. The number of partners is a count of shared-home relationships, not a count of everyone someone dated or everyone they considered marrying. It also does not tell us whether any individual relationship was healthy or wanted.

Education differences appear in this cohort: 82% of participants with a high school diploma or less had ever cohabited, compared with 70% of those with a bachelor's degree. The education categories describe the highest level reached by ages 38 to 43, which may differ from the level someone had when they first moved in with a partner. These figures do not show that education causes or prevents cohabitation. They show that the lifetime experience was common in both groups, with a measurable difference in this cohort.

When people first shared a home

Pew also described age at first cohabitation among cohort members who had ever cohabited. Twenty-seven percent first lived with an unmarried partner before age 20; 39% first did so at 20 to 24; 21% at 25 to 29; and 13% at 30 or older. The base for these percentages is the ever-cohabited group, not every 38-to-43-year-old. A reader cannot multiply a single row by the full U.S. adult population and claim a national count.

The timing differed by education. Among ever-cohabited adults with a high school diploma or less by their early 40s, 40% first cohabited before 20; among those with a bachelor's degree, 12% did. The pattern describes when the experience occurred in these observed groups. It does not explain why an individual moved in, whether they could afford another household, or whether they later married. The survey can place an event in a life history without assigning its motive.

If your question is whether moving in during your 30s is unusual, the 13% figure is a poor personal benchmark. It includes people first cohabiting at any age from 30 onward but only through the cohort's early 40s. People who first move in with a partner after that age have not yet had a chance to enter this group. The study's window ends at the available interview, not at the end of each person's relationship life.

Living together and later marriage

Among the cohort members who had ever cohabited, about two-thirds had married someone they lived with by the time Pew examined their records. That says nothing about whether every cohabiting relationship led to marriage. One person could live with several partners and marry one of them, while another could remain unmarried in a long relationship. The measure is about whether they ever married a cohabiting partner by their early 40s.

Pew also found that 77% of ever-cohabited adults with at least a bachelor's degree had married a cohabiting partner, compared with 59% of those with a high school diploma or less. These are descriptive shares within this cohort. They should not be turned into a claim that living together raises or lowers a couple's chance of marriage. The study observes choices and outcomes; it does not randomly assign people to share a home.

The 2019 report asked a different question about expectations. Among cohabiters who were not engaged when they moved in, 44% said they saw the move as a step toward marriage. That is what respondents said they had thought at move-in; it is not a marriage-outcome rate. The 2019 expectation and the later cohort outcome use different samples, so they cannot be joined into a single conversion funnel.

Has cohabitation become more common?

In the earlier National Survey of Family Growth analysis that Pew summarized, the share of adults ages 18 to 44 who had ever lived with an unmarried opposite-sex partner rose from 54% in 2002 to 59% in the later survey. Over that span, the share who had ever married fell from 60% to 50%. These are changes in separate lifetime measures. People can contribute to both because living together and marriage can happen in sequence.

That earlier relationship-history analysis covered opposite-sex relationships. The newer cohort study uses its own definition of an unmarried romantic partner. A change between 2002 and the later survey should be read within the earlier series, while the 2023 cohort result stands as a separate view of one generation's accumulated experience. Mixing the two into a single three-point trend would hide the definition change and the cohort's older age at measurement.

A current cohabitation rate answers a different trend question. Pew's 2019 report cited a rise from 3% of U.S. adults living with an unmarried partner in 1995 to 7% in its later estimate. That comparison describes who was sharing a home at each point. It does not count everyone who had ever tried cohabitation. Use it when the question concerns present household arrangements, and use a history measure when the question concerns whether someone has ever done it.

Choose the estimate that fits your question

If you want to know how many adults were living with an unmarried partner at a point in time, cite the 7% current estimate and its year. If you want to know whether younger adults had ever lived with one, cite the 59% among ages 18 to 44 and say that the analysis covered opposite-sex relationships. If you want to describe experience by the early 40s in one followed cohort, cite the 77% in 2023. Each answer is useful once the group and time window travel with it.

None of these estimates tells a couple whether they should share a home or marry. They also cannot place partners who live separately in a cohabiting household. The numbers describe living arrangements and histories in the observed groups. A personal decision needs a conversation about finances, expectations, privacy, and the practical terms of sharing a home; a national share cannot settle those questions.