Relationship statistics: what the data actually shows

What is the average age to get married?

The US Census Bureau's Current Population Survey has tracked median age at first marriage since 1890. The trend over the past seventy years is one of the most consistent in all of demographic research: both men and women are marrying substantially later than previous generations, and the pace of that shift has accelerated since the 1990s.

In 1950, the median age at first marriage was 22.8 for men and 20.3 for women. By 2024 those figures stand at 30.5 and 28.6 respectively, increases of nearly eight years for both genders. The underlying drivers include rising educational attainment, the normalisation of cohabitation before marriage, and shifting economic expectations around homeownership and financial stability before formalising a relationship.

UK patterns follow a very similar trajectory, with the Office for National Statistics reporting a median age at first marriage of approximately 33.4 for men and 31.5 for women in recent years, roughly three years later than US figures. For a personalised view of where your relationship timeline sits relative to these norms, use the relationship timeline calculator or the when to propose calculator.

How common is divorce?

The oft-cited "50% of marriages end in divorce" figure is an oversimplification of a more nuanced picture. CDC/NCHS data and Philip Cohen's 2019 analysis in Socius suggest that approximately 40 to 45% of first marriages end in divorce, and that figure has been declining since the early 1980s peak. The crude divorce rate, which peaked at 5.3 per 1,000 people in 1981, had fallen to approximately 2.5 by the early 2020s.

The decline is concentrated among college-educated married couples, whose divorce rates have fallen substantially since the 1980s. Divorce rates among non-college-educated couples have remained higher and declined more slowly. This educational divergence is now one of the defining features of US marriage patterns.

Marriage order matters significantly. Second marriages have a divorce rate of approximately 60%, and third marriages approach 73%. These higher rates reflect selection effects: people who have already divorced are statistically more likely to divorce again. Use the divorce probability calculator for personalised estimates, or see signs your marriage is over for qualitative research.

How often do couples argue?

John Gottman's research at the University of Washington, carried out over more than three decades with thousands of couples, provides the most rigorous available data on relationship conflict. Most couples in stable relationships report arguing 1 to 3 times per month. The frequency of conflict is less predictive of relationship outcomes than how couples manage it.

Gottman's findings on perpetual conflict are particularly striking: approximately 69% of relationship conflicts are what he terms "perpetual problems," meaning they are never fully resolved and tend to resurface throughout the relationship. These include differences in personality, spending habits, household preferences, and parenting philosophy. The couples who fare best are those who can manage these perpetual issues with humour and acceptance rather than contempt or stonewalling.

The 5:1 ratio, sometimes called the "magic ratio," describes the relationship between positive and negative interactions. Couples heading toward divorce average roughly 0.8 positive interactions for every negative one. Stable couples average approximately 5 positive interactions for every negative one. Use the relationship communication quiz to assess your own conflict patterns.

How common is infidelity?

The General Social Survey (NORC) has asked respondents about extramarital sex since 1991 using the variable EVSTRAY, making it the longest-running nationally representative dataset on infidelity in the US. Across multiple survey waves, 20 to 25% of married men and 10 to 15% of married women report having had sex with someone other than their spouse while married.

These figures are widely considered underestimates given the social desirability bias involved in self-reporting infidelity. Some anonymous surveys produce higher figures, particularly for women in younger cohorts, suggesting that the gender gap may be narrowing. Rates also vary by age bracket: older cohorts, who have had longer exposures to marriage, report higher lifetime rates.

Infidelity measurement is sensitive to question framing. Studies that ask about "emotional" rather than only "sexual" infidelity report substantially higher rates across all groups. The GSS figures specifically capture sexual infidelity. For a personalised comparison and context on how infidelity rates relate to relationship risk, use the infidelity calculator.

What percentage of couples meet online?

A landmark 2019 study by Michael Rosenfeld, Reuben Thomas, and Sonia Hausen, published in the Proceedings of the National Academy of Sciences, tracked how US couples met from the 1990s through to the late 2010s. Online meeting overtook meeting through friends as the most common way to find a romantic partner around 2013, and the gap has widened considerably since then.

By 2017, approximately 39% of heterosexual couples and more than 65% of same-sex couples in the US had met online. Pew Research Center's 2023 Online Dating Survey found that over 30% of US adults had used a dating app, rising to 53% among adults under 30. The share of couples who met at bars or through mutual friends has declined in parallel, though meeting at work and through social settings remains significant.

The implications extend beyond how people find partners: Rosenfeld's data also suggests that couples who meet online are slightly more likely to be ethnically mixed and to span larger geographic distances than those who met offline. For context on your own dating pool, see the dating pool calculator or explore data on how many dates before sex.

What are the most common attachment styles?

Attachment theory, originally developed by John Bowlby and extended to adult romantic relationships by Cindy Hazan and Phillip Shaver in their 1987 Journal of Personality and Social Psychology paper, proposes that adults develop characteristic styles of relating to intimate partners based on early caregiving experiences. Four main styles are now widely recognised in the research literature.

Hazan and Shaver's original sample found approximately 56% secure, 25% avoidant, and 19% anxious attachment. Later nationally representative data from Mickelson, Kessler, and Shaver (1997) in a sample of over 8,000 adults found similar proportions, with roughly 50% secure. Across multiple studies, approximately 50% of adults have a predominantly insecure attachment style when measured using validated instruments.

Attachment styles are moderately stable across the lifespan but can shift with significant relationship experiences, therapy, and life events. They predict a range of relationship outcomes including relationship satisfaction, conflict behaviour, and likelihood of seeking support from a partner. Use the attachment style quiz to identify your own profile, or explore how attachment intersects with love languages.

How long do people date before getting engaged?

The Knot's 2023 Real Weddings Study, which surveyed over 10,000 recently married couples in the US, found that the average couple dated for 2 to 3 years before getting engaged. The average engagement itself lasted approximately 15 months, making the total time from first date to wedding roughly 4 to 5 years for most couples. Both figures have increased compared to equivalent surveys from the early 2000s.

Cohabitation before marriage has become the statistical norm rather than the exception. Over 70% of couples who married in the past decade had lived together before the wedding. Research on whether cohabitation predicts better or worse outcomes is mixed: earlier studies found a "cohabitation effect" linked to higher divorce rates, but more recent analyses controlling for the timing of cohabitation (before vs after engagement) find weaker or no effect.

Milestone timing varies considerably by education level, age, and geography. College-educated adults tend to have longer pre-engagement relationships and longer engagements than non-college-educated adults. Use the when to propose calculator, the moving in together timeline, and the relationship timeline calculator for personalised context.

How many close friends does the average person have?

The Survey Center on American Life's 2021 report on friendship in the US found that the average American has 3 to 5 close friends, a figure that has remained relatively stable at the population level. However, the distribution has shifted significantly over the past three decades. In 1990, 3% of Americans reported having no close friends at all. By 2021, that figure had risen to 12%, a fourfold increase.

Friendship counts decline systematically with age, particularly after the major life transitions of the early 30s: starting a career, entering a long-term relationship, and having children all reduce the time and energy available for maintaining friendships. Men have experienced steeper declines in friendship counts than women over this period, with the proportion of men reporting zero close friends rising from 3% in 1990 to 15% in 2021.

Robin Dunbar's social brain research suggests that humans can maintain meaningful contact with approximately 150 people in total, with closer inner layers of roughly 15 good friends and 5 intimate friends. The 3 to 5 figure in the American Life survey corresponds to this innermost circle. For more detailed data on adult friendship patterns and how they change across the lifespan, see adult friendship statistics.

What is the average age gap in relationships?

Analysis of the Census Bureau's American Community Survey data on married couples shows that the most common age gap is 0 to 2 years, accounting for approximately 33% of all US married couples. When this category is extended to gaps of 0 to 3 years, it covers roughly half of all marriages. Age gaps of 10 years or more are present in approximately 8% of marriages.

Male-older pairings outnumber female-older pairings across all age groups, though the ratio varies by the size of the gap. For gaps of 1 to 5 years, male-older pairings account for roughly 64% of couples. For gaps of 10 years or more, male-older pairings account for approximately 76%. This reflects both long-established social norms and the structure of dating markets.

Younger cohorts show a narrowing of this asymmetry. Among adults under 35, female-older pairings are more common relative to the historical baseline, and attitudes toward age gaps in relationships have become more permissive across the board. For context on where your own relationship fits within the population distribution, use the age gap calculator.

How long does it take to get over a breakup?

Research on breakup recovery is complicated by the enormous variation between individuals and relationships, making population averages less informative than for most of the other topics on this page. Meta-analytic reviews of the available studies suggest that most people experience significant emotional recovery within 3 to 6 months following the end of a relationship that lasted under two years. Longer relationships take longer to recover from, and divorce recovery follows a distinct trajectory.

Studies on divorce adjustment, including longitudinal work by Mavis Hetherington and others, consistently find that significant emotional and functional recovery averages 12 to 18 months, with some individuals taking two years or more. Three variables consistently predict faster recovery: being the person who initiated the separation, the absence of infidelity, and having strong social support.

The model of discrete "stages" of grief following a breakup has limited empirical support. Recovery is more typically described in the research as non-linear, with periods of improvement interrupted by setbacks triggered by anniversaries, chance encounters, or social media exposure. Use the breakup recovery calculator for context on your timeline, or explore the should I break up quiz if you are still in the decision phase.