Why do marriages really end?
When people are asked to predict the number one reason for divorce, most say money problems or communication issues. The research tells a different story. Guess first, then see the data.
What do you think is the number one reason for divorce?
Gender differences
Women were more likely to cite infidelity (25.2% vs 15.6%), drinking or drug use (12.4% vs 7.5%), and physical or mental abuse (9.0% vs 0.0%). Men were more likely to cite growing apart (13.1% vs 7.5%). Incompatibility and communication issues showed no significant gender difference. (Amato & Previti, 2003)
Top reasons for divorce: what the research actually shows
When people are asked to predict the leading cause of divorce, money problems and communication breakdown consistently top the list. The peer-reviewed research tells a different story. Amato and Previti's landmark 2003 study (Journal of Marriage and Family, n=208 divorced adults) asked participants to describe in their own words why their marriage ended. Infidelity ranked first at 21.6%, followed by incompatibility or growing apart at 19.2%, drinking or drug use at 10.6%, and growing apart or falling out of love at 9.6%. Financial problems ranked 10th — cited by just 2.9% of respondents as the primary reason. This divergence between what people predict and what the data shows is the central insight of this page.
A larger 2012 study by Hawkins, Willoughby, and Doherty (Family Relations, n=886) using a checklist format found that when participants could select multiple contributing reasons, "communication problems" and "growing apart" scored higher. This suggests financial and communication problems are background stressors that contribute to marital decline, while infidelity and incompatibility tend to be the acute precipitating events that trigger the actual decision to divorce. The top reasons for divorce therefore depend on whether you are measuring slow-building contributing factors or the specific event that prompted the final decision. Both answers are correct in context.
The divorce rate itself provides further context. The US crude divorce rate fell from 4.7 per 1,000 population in 1990 to 2.4 per 1,000 in 2021 (CDC NVSS). This halving of the crude rate is frequently cited as evidence that marriage has become more stable, but it partly reflects a declining marriage rate — fewer people entering marriage means fewer people who can divorce. Among couples who do marry, first-marriage dissolution rates remain approximately 40-50% over a lifetime. The reasons for divorce statistics in this calculator are drawn from self-reported data from divorced adults, which captures the subjective experience of marriage ending more accurately than demographic data alone.
Why do couples grow apart? The psychology of gradual disconnection
Growing apart — variously called "incompatibility," "loss of love," or "gradual disconnection" — is consistently one of the top two or three cited reasons for divorce across all major studies. Unlike infidelity, which is a discrete event, growing apart is a process that unfolds over years. Gottman's longitudinal research, following thousands of couples over decades, identified the mechanisms most predictive of this outcome: the erosion of positive sentiment (partners stop interpreting each other's behaviour charitably), the decline of repair attempts (bids to reconnect after conflict that go unreciprocated), and the replacement of genuine curiosity about the partner's inner life with indifference. Growing apart is not a passive process — it is the cumulative result of small disconnections that are not repaired.
Why do money problems divorce statistics consistently underperform expectations? Financial stress is highly salient and chronic — it is something couples argue about visibly and repeatedly, making it easy to identify as a marriage problem. But Amato and Previti's data suggests that financial arguments are rarely the reason people ultimately leave. They are a context within which deeper incompatibilities become apparent. Couples with strong foundations — shared values, genuine friendship, mutual respect — typically navigate significant financial stress without divorcing. Couples whose foundations have already weakened find that financial conflict becomes the arena where pre-existing disconnection plays out. Communication breakdown divorce statistics show the same pattern: poor communication is almost always a symptom of something deeper rather than the underlying cause. The research on reasons for divorce consistently points to infidelity and the loss of the relational foundation as the actual drivers, while money and communication difficulties are better understood as contributing factors that accelerate pre-existing decline.
Why do people get divorced? The data
Amato and Previti (2003) asked 208 divorced individuals to describe in their own words why their marriage ended. The results challenge common assumptions: infidelity ranked first at 21.6%, followed by incompatibility at 19.2%. Financial problems, which the general public consistently expects to top the list, ranked 10th at just 2.9%.
A larger follow-up study by Hawkins et al. (2012) using a checklist format with 886 respondents found similar patterns, with "growing apart" and "too much conflict" also featuring prominently when multiple reasons were permitted.
The US crude divorce rate has declined from 4.7 per 1,000 population in 1990 to 2.4 in 2021 (CDC NVSS). However, this decline partly reflects a falling marriage rate: fewer people are entering marriage in the first place, particularly among younger cohorts.
Frequently asked questions
In the most widely cited open-ended study, Amato and Previti (2003) found infidelity was the most commonly reported primary cause at 21.6%, followed by incompatibility at 19.2%. Financial problems ranked 10th at 2.9%, far below what most people expect.
Yes. Women were significantly more likely to cite infidelity (25.2% vs 15.6%), drinking or drug use (12.4% vs 7.5%), and physical or mental abuse (9.0% vs 0.0%). Men were more likely to cite growing apart (13.1% vs 7.5%). The abuse finding is particularly stark: 9% of women cited it as the primary reason, while no men did.
The crude divorce rate has fallen from 4.7 per 1,000 in 1990 to 2.4 in 2021, but this partly reflects a declining marriage rate. Among couples who do marry, the first-marriage divorce rate is approximately 40–50%. Younger generations are marrying later and less often, which reduces the pool of people who can divorce.
Financial stress is a chronic, visible stressor that most couples experience, making it easy to imagine as a marriage-ender. Research does find that financial arguments predict divorce risk at a population level. But when individual divorced people are asked what actually ended their marriage, they point to infidelity and incompatibility far more often. Money may be background pressure; the acute breaking point is typically something more personal.
US Census Bureau and ACS data consistently show that certain occupations are associated with substantially higher divorce rates than the national average of approximately 40-50%. Gaming managers, bartenders, flight attendants, dancers, and gaming service workers show rates of 50-60%. The common thread is irregular hours, high-stress environments, and frequent contact with strangers outside the family unit. By contrast, actuaries, software engineers, clergy, and medical scientists show some of the lowest divorce rates, typically under 20-25%. Occupation is a proxy for broader lifestyle factors including income stability, work schedule, and social environment rather than a direct causal force. (Source: US Census Bureau ACS; Belt 2015)
The relationship between children and divorce is complex. Having children does reduce the probability of divorce in the short term, as couples with dependent children are less likely to file. This is often called the "child protection effect." However, it also means some unhappy marriages persist longer than they otherwise would. Longitudinal studies show that relationship satisfaction typically declines after the birth of a first child, particularly for couples who had the child within two years of marriage. The timing matters: couples who have children after five or more years of marriage show smaller satisfaction declines than those who have children early. (Source: Doss et al. 2009, JFAM; Twenge JM et al. 2003)
Gottman's longitudinal research identified two peak risk periods for divorce. The first is around 7 years of marriage, driven by unresolved conflict, diverging life goals, and the decline of early romantic passion. The second peak is around 16-20 years, often coinciding with empty nest transitions when children leave home and couples discover they have grown apart or never built a strong friendship foundation. The "7-year itch" is culturally well known and has modest empirical support: a disproportionate share of divorces does occur around this mark. However, marriages that survive 20+ years have a substantially lower risk of subsequent divorce. (Source: Gottman JM & Levenson RW. 2000; CDC NVSS)
In the US, the average contested divorce takes 12-18 months from filing to finalisation, while an uncontested divorce can be completed in 3-6 months in most states. Costs vary enormously: uncontested divorces where both parties agree on all terms can cost $1,500-$3,000 in total legal and filing fees. Contested divorces involving disputes over children, property, or spousal support average $15,000-$30,000 per party in attorney fees. High-conflict divorces with trials can exceed $100,000 per party. Mediation significantly reduces both time and cost, with the average mediated divorce costing $3,000-$8,000 total. (Source: Forbes Advisor; American Academy of Matrimonial Lawyers)
- Amato PR, Previti D (2003). People's reasons for divorcing. Journal of Family Issues, 24(5):602–626. N=208 divorced individuals.
- Hawkins AJ, Willoughby BJ, Doherty WJ (2012). Reasons for divorce and openness to marital reconciliation. Journal of Divorce & Remarriage, 53(6):453–463. N=886.
- CDC National Center for Health Statistics / NVSS. Crude divorce rate 2021: 2.4 per 1,000 population.
- ONS. Divorces in England and Wales. Annual divorce statistics and grounds data.
This page presents research findings on reasons for divorce. It is not legal advice, marriage counselling, or a substitute for professional support. The Amato & Previti study (2003) used a US sample of 208 individuals, which may not fully represent all demographics or cultures.