What are the odds of a first marriage ending in divorce?
Evidence quality 4.63/5
Eight-dimension review score against the quality rubric . Each dimension scored 1–5.
- D1 Source grounding
- 5/5
- D2 Source authority
- 5/5
- D3 Arithmetic
- 4/5
- D4 Uncertainty
- 4/5
- D5 Scope
- 5/5
- D6 Prose
- 5/5
- D7 Perception honesty
- 4/5
- D8 Caveat completeness
- 5/5
Lifetime probability · lifetime, subgroup
1 in 2.4
42% lifetime chance
Most people overestimate this.
range 1 in 2.9 to 1 in 2.0
● your factors — click this risk ▾ to reveal
≈ As likely as
Perceived
The "50% of marriages end in divorce" claim is one of the most durable statistical myths in American culture. It has been repeated so often — in news articles, self-help books, wedding speeches, and casual conversation — that most adults treat it as settled fact. The figure was never a cohort estimate; it came from comparing annual marriages to annual divorces in a calendar year, a method that confuses flows with stocks and overstates the risk for any individual marriage.
Rough estimate: ~50% — the folk statistic most people cite
Source: editorial intuition, not polled
Actual
~14.2 divorces per 1,000 married women per year (2024)
US married women age 15+
Show derivation
The native rate is a refined divorce rate (divorces per 1,000 married women per year) from the BGSU National Center for Family & Marriage Research using 2024 ACS data. To estimate a lifetime probability for first marriages, we use life-table methods rather than naive compounding, because the hazard is not constant — it peaks around years 5-8 and declines thereafter. The ~42% central estimate draws on the consensus range in family demography literature (roughly 40-45% for all US first marriages, lower for post-2000 and college-educated cohorts). This is consistent with the observed decline from the peak-divorce era: the refined rate fell from ~22.6 per 1,000 in 1980 to 14.2 in 2024, and younger cohorts are divorcing at lower rates than their parents did. The 42% figure is a population average that masks sharp demographic gradients by education, age at marriage, and birth cohort.
Caveats: This is not a death risk; the normalized figure represents the probability that …
This is not a death risk; the normalized figure represents the probability that a first marriage will end in divorce at some point, not the probability of dying. The 42% central estimate is a population average for all US first marriages and masks large demographic gradients. For college-educated women who married after 2000, the best available estimates put the figure at roughly 25-30%. For women without a college degree, it is closer to 55-60%. Age at marriage, cohabitation history, and religious participation all shift the number further. The "50% of marriages end in divorce" figure was never a cohort estimate — it came from dividing annual divorces by annual marriages in the same calendar year, a method that confuses flows with stocks and has been repeatedly criticized in the demography literature. The rate has been declining for decades, and the decline is concentrated among younger, more educated cohorts.
Risks at similar odds
Other risks with roughly the same likelihood — useful for calibration.
Career obsolescence
What are the odds of needing to change careers due to technological disruption?
Forced job exit before retirement
How likely is a worker over 50 to be pushed out of their job before they planned to retire?
Frontline soldier casualty
What is the probability of a front-line infantry soldier being killed or seriously and permanently wounded over five years in a Ukraine-scale conventional conflict?
Recently viewed on this device
Stored locally — clear anytime.
Pick challenger
The “50% of marriages end in divorce” figure is arguably the most cited statistic in American domestic life, and it has been wrong for decades. The number came from comparing annual marriages to annual divorces in a calendar year — a method roughly as sensible as dividing annual births by annual deaths to estimate the odds of dying in a given year. Cohort analysis tells a different story: about 42% of US first marriages will eventually end in divorce, down from a peak near 50% for marriages that began in the late 1970s. The refined divorce rate has fallen from roughly 22.6 per 1,000 married women in 1980 to 14.2 in 2024, a 50-year low.
The decline is not uniform, and that is the interesting part. Kennedy and Ruggles showed in 2014 that the falling headline rate masked a bifurcated trend: divorce rates among people over 35 had actually doubled between 1990 and 2008, while rates among younger couples were stable or falling. The explanation is selection. Marriage has become increasingly concentrated among the educated and economically stable, and those marriages are more durable. The cultural perception — half of all marriages fail — lags the data by a generation.
The demographic gradient is steep enough to matter for any individual reader. For college-educated women who married after 2000, the best estimates put lifetime divorce risk at roughly 25-30%, about half the population average. For women without a college degree, it is closer to 55-60%. Age at marriage is the second-strongest predictor: marrying before 25 raises the risk substantially, while marriages in the late twenties show the lowest hazard in recent ACS data. The 42% headline is a real number, but it describes almost nobody in particular — and the 50% it replaced described nobody at all.
Related tidbits
Chronic back pain affects 80% of adults over a lifetime. Divorce hits about 42%. Your spine is a less reliable partner than your spouse.
17.5% of couples experience infertility. 42% of marriages end in divorce. One derails the family plan. The other derails the family. Both are more common than people admit at dinner parties.
Claim ledger
Every number below is what each source reported, with the verbatim quote we relied on and how we arrived at our figure. Click any link to verify directly.
-
[1] National Center for Family & Marriage Research, Bowling Green State University — Refined Divorce Rate in the U.S.: Geographic Variation, 2024
Refined Divorce Rate in the U.S.: Geographic Variation, 2024- Statistic
14.2 women per 1,000 married women aged 15+ divorced in the past 12 months (2024 ACS)- Excerpt
“"With 14.2 women divorcing per 1,000 married women, the U.S refined divorce rate decreased just slightly in 2024 from 14.4 in 2023." ”
- Source data from
- 2025-01-01
- Accessed
- 2026-04-11 · archived copy
- Calculation
- The refined divorce rate (per 1,000 married women) is the standard measure in family demography because it uses the at-risk population as the denominator, unlike the crude rate (per 1,000 total population). The 2024 figure of 14.2 continues a long decline from the 1980 peak of ~22.6. Used as the native annual rate. The lifetime estimate requires life-table methods because the divorce hazard is non-constant across marriage duration; the 42% central estimate comes from the demography consensus described in the assumptions field rather than from naive compounding of this annual rate.
- Independence
- BGSU NCFMR profiles are derived from ACS microdata. They are methodologically independent from CDC NCHS vital statistics, which use state-reported counts.
-
[2] Demography (Kennedy & Ruggles) — Breaking Up Is Hard to Count: The Rise of Divorce in the United States, 1980-2010
Breaking Up Is Hard to Count: The Rise of Divorce in the United States, 1980-2010- Statistic
Age-standardized divorce rates doubled among persons over 35 between 1990 and 2008; rates among youngest couples stable or declining- Excerpt
“"Using new data from the American Community Survey and controlling for changes in the age composition of the married population, we conclude that there was actually a substantial increase in age-standardized divorce rates between 1990 and 2008. Divorce rates have doubled over the past two decades among persons over age 35. Among the youngest couples, however, divorce rates are stable or declining." ”
- Source data from
- 2014-04-01
- Accessed
- 2026-04-11 · archived copy
- Calculation
- Kennedy & Ruggles demonstrated that crude divorce rate trends are misleading because they fail to account for changes in the age composition of the married population. Their age-standardized analysis showed that the apparent post-1980 decline in divorce masked rising rates among older adults and genuinely declining rates among younger cohorts. This bifurcated trend is critical to interpreting the lifetime probability: the 42% average blends a lower rate for post-2000 marriages (~35%) with higher rates for older cohorts who married in the peak-divorce era.
- Independence
- Kennedy & Ruggles used ACS microdata, the same upstream source as BGSU NCFMR profiles, so the two sources are partially dependent. However, the analytical contribution (age-standardization methodology and cohort decomposition) is independent.
-
[3] CDC National Center for Health Statistics — Marriage and Divorce
Marriage and Divorce- Statistic
Crude divorce rate: 2.4 per 1,000 population (2023 provisional, 45 reporting states and D.C.)- Excerpt
“"Divorce rate: 2.4 per 1,000 population (45 reporting States and D.C.)" ”
- Source data from
- 2025-03-17
- Accessed
- 2026-04-11 · archived copy
- Calculation
- The crude divorce rate (per 1,000 total population) is the figure most commonly cited in news media, but it is the wrong denominator for estimating individual risk — it includes children, never-married adults, and the already-divorced. The crude rate has fallen from 5.0 in 1985 to 2.4 in 2023, partly reflecting genuine decline and partly reflecting the falling share of the population that is married. Used here as a corroborating check and to illustrate why the "50% myth" arose: dividing annual divorces by annual marriages in a calendar year is algebraically similar to the crude rate and overstates the lifetime risk for any cohort.
- Independence
- CDC NCHS uses state vital statistics reports (administrative records), while ACS-based sources use household survey responses. Different collection pipelines.