What percentage of women report usually swallowing versus spitting in sexual encounters according to national surveys?
Executive summary
National, probability-based surveys that ask about oral sex (e.g., NHANES and NSFG) document how common oral sex is but do not ask a standard, nationally representative question about whether women “usually swallow” or “usually spit” after oral sex, so there is no definitive national percentage from high-quality government surveys [1] [2]. The figures most often cited in popular reporting come from convenience polls, website surveys, and community threads that produce widely varying estimates — commonly reporting a majority who swallow in those samples (e.g., 58%–79%) but with important methodological limitations [3] [4].
1. What the rigorous national surveys actually ask — and what they don’t
Large, government-sponsored sexual behavior studies such as NHANES and the National Survey of Family Growth (NSFG) are the best sources for nationally representative sexual behavior data in the U.S.; they measure prevalence and timing of oral sex and related exposures, but neither NHANES nor NSFG include standardized, nationally representative items that identify whether respondents usually swallow or spit after cunnilingus/fellatio, so they cannot supply a reliable national percentage for that specific behavior [1] [2].
2. What convenience and website surveys report — a majority but inconsistent numbers
In the absence of a national survey item, most published figures come from convenience samples: a YourTango social-media poll reported 79% of responding women said they swallow (a self-selecting sample on a dating/advice site and explicitly non‑scientific) [4], an UncoveringIntimacy site survey reported 58% among women who allow oral ejaculation [3], and various sex‑advice aggregators and blogs reproduce similar majority‑swallow findings while citing small or informal surveys [5] [6]. These numbers consistently suggest that in non‑representative online samples a majority report swallowing at least some of the time, but the exact percentage varies by sample and wording.
3. Smaller or older informal studies give different slices — frequency matters
Some informal or localized studies report different patterns when the question is framed as “every time” versus “sometimes”: an informal Sacramento study cited on an advice forum reported that 24% of married women swallow every time they engage in oral sex, illustrating how frequency thresholds change estimates [7]. Community polls (e.g., BabyCenter threads) and magazine interviews produce anecdotal responses that further highlight large within‑group variation and the influence of question wording and sample population [8] [9].
4. Why estimates vary — methodological and social factors
Differences across sources reflect clear methodological problems: convenience samples over‑represent certain demographics (readers of sex blogs, forum participants), self‑selection and social desirability bias can push people toward reporting what they think is normative, question phrasing (ever, usually, sometimes, every time) matters enormously, and many reports conflate “allowing ejaculation in the mouth” with “swallowing” rather than distinguishing spit, swallow, or spitting later, so published percentages are not comparable across studies [4] [3] [5].
5. A cautious summary: what can be said with confidence
What can be stated with confidence from the provided reporting is this: nationally representative surveys document that oral sex is common but do not yield a national “swallow vs spit” percentage [1] [2]; multiple convenience and online surveys report that a plurality or majority of respondents say they swallow at least sometimes, with estimates commonly reported in the 50–80% range in those samples [3] [4] [5]; and when the question is narrowed to “every time” the reported share can be much lower (e.g., 24% in one informal report) [7]. Because of these limitations, any single percentage cited from blogs or polls should be treated as indicative of that sample only, not as a nationally representative statistic.