An Institute for Family Studies analysis of pooled 1991 to 2014 General Social Survey answers reported that about 17 in 100 ever-married U.S. men had sex with someone other than a wife while married. The result was 16.5% for men and 10.7% for women. It is a historical self-report about sexual infidelity in marriage, not a current rate or a measure of emotional cheating.
See how U.S. sources measure marriage duration and divorce timing→
On this page
What does the 16.5% estimate measure?
The short answer is about 17 in 100 ever-married U.S. men in one Institute for Family Studies analysis of pooled General Social Survey answers. The analysis reported 16.5% for men and 10.7% for women in survey years from 1991 through 2014.
The question was narrow. It asked whether someone had ever had sex with a person other than their husband or wife while married. It does not count emotional cheating, dating-app use, or a current relationship outside marriage.
That wording matters because “cheating” can cover several different experiences. A person can mean sex outside a marriage, sexual activity without a partner's consent, an emotional attachment, or conduct that a couple had agreed was acceptable.
Who got counted in the marriage-only estimate?
The General Social Survey is run by NORC at the University of Chicago. An Institute for Family Studies analysis pooled answers from 1991 through 2014. It reported that 16.5% of ever-married men and 10.7% of ever-married women said they had sex outside marriage. We did not independently reproduce that tabulation from the public GSS file.
Read that as a historical self-report from people who had ever married, rather than a current annual rate. Older respondents have had more years in which an event could have happened, and the answer depends on what respondents chose to disclose.
The latest public GSS codebook keeps the same basic question. In the 2022 file, `EVSTRAY` asks whether a respondent ever had sex with someone other than a husband or wife while married. Its unweighted response table includes never-married respondents, so it cannot replace the earlier sex-specific estimate.
Why a broader relationship survey gives a different result
YouGov asked 1,125 U.S. adult citizens from an opt-in panel in April 2025 whether they had engaged in sexual activities with someone else without their main romantic partner's consent. Twenty-nine percent of men and 21% of women said yes.
The sample was not limited to married people, and the wording did not require a marriage. That survey therefore measures a broader relationship experience than the GSS question.
YouGov also reported in 2022 that 33% of adults who had ever been in a monogamous relationship said they had cheated physically, emotionally, or both. In that report, 13% of married adults who were currently in monogamous relationships said they had cheated on their current partner.
U.S. survey results use different questions and groups
The table may be wider than the screen. Focus it and use the arrow keys to reach every column.
| Source and period | Who was asked | Question scope | Men | Women |
|---|---|---|---|---|
| GSS analysis, 1991 to 2014 | Ever-married respondents | Ever had sex with someone other than a spouse while married | 16.5% | 10.7% |
| YouGov, 2025 | U.S. adults | Sexual activity with someone else without a main partner's consent | 29% | 21% |
| YouGov, 2022 | Adults currently married and in monogamous relationships | Cheated physically, emotionally, or both on a current partner | Not reported in the article's summary by sex | Not reported in the article's summary by sex |
Do not turn these into a trend The source groups and question wording differ. These results describe separate survey measures.
How to read it: Self-reported survey responses can miss behavior people do not disclose.
Download the table (CSV) ↓Why the numbers do not line up
The GSS estimate starts with people who had ever married and asks about sex outside marriage. The newer YouGov results include people in relationships outside marriage and use broader descriptions of behavior.
Time period matters too. One figure pools 1991 through 2014, while the YouGov results are single surveys from 2022 and 2025. The groups, years, and questions changed, so the difference cannot prove that cheating rose or fell.
A useful answer keeps those parts attached to the number. If a page gives a percentage without the relationship type, question, and years, you cannot tell who was in the room.
What the surveys cannot tell you
Neither survey can establish a personal risk. They do not describe whether a particular partner has cheated, whether a relationship is healthy, or what a couple has agreed about monogamy.
They also cannot measure undisclosed behavior. The surveys describe what people chose to report, in the groups and years that each source asked.
If you are trying to decide what a partner's actions mean, a national percentage is only background. The clearer question is what you and your partner agreed to, and whether the behavior matched that agreement. For related context, see U.S. marriage statistics, average length of marriage, and what the U.S. divorce rate measures. They answer different questions from infidelity.
