An Update to the GSS Sexlessness Graph

Sexual inactivity and singleness rates among young adults in the 2024 GSS data

The 2024 General Social Survey data is out, marking the third wave since the 2018 survey that the Washington Post’s infamous sex recession graph was based on. There’s not much to go through here, partly because they unfortunately appear to have axed a good chunk of the sexual behaviour module; no partner counts, for example. So let’s get this out of the way, starting with another update to the graph. Was this poorly photoshopped graph nonetheless on the money?

Incels: so back?

Basically, after a brief dip, the male sexual inactivity rate is back to where it was in 2018, with 27% of men aged 18–29 reporting no sex in the past year, compared to 20% of women.¹²

Still no big singleness gap

Contrary to the megaviral Pew survey which reported a 29% singleness gap among under-30s in 2022, the GSS has shown no evidence for a large or widening singleness gap, in line with other surveys. This continues to be the case: about 56% of men and 54% of women aged 18–29 reported being single in 2024.³⁴

Conclusion

Though the data might not make for quite as much of a sexy line-go-up graph as the fake one (until they photoshop 2025/50% onto this graph at least), similar to the NSFG we nonetheless see some evidence for a continuation of the ‘sex recession’, with sexual inactivity rates among under-30 men higher than in any prior year, and among under-30 women higher than in any prior year outside the COVID year. The data has been more clear and consistent when it comes to adolescent sexual activity, but it makes sense that those trends would carry over to young adulthood to some extent.

The 7% gap in sexual inactivity isn’t anything too out of the ordinary, and can be reasonably explained by men tending to be slightly older than their partners, as well as a somewhat larger number of young men than women (also the CIs are obviously quite wide). The adolescent data hasn’t shown a gendered pattern either. It’s possible that in the case of singleness the apparent trend partly reflects a real trend of shrinking relational age gaps, but more than anything it’s another reminder of how data with relatively shaky reliability—the Pew survey included—has the potential to be misleading.

Disclaimers

Consistent with other post-COVID surveys, there was a record low response rate of 44.6%. I used the weighting variable designed to adjust for nonresponse bias, but it likely won’t have fully captured psychological differences between those who did and didn’t participate.

Like the 2022–23 NSFG, this survey used a multimodal survey design, which could complicate comparisons with earlier waves. The codebook recommended including this disclaimer:

Changes in opinions, attitudes, and behaviors observed in 2021, 2022 and 2024 relative to historical trends may be due to actual change in concept over time and/or may have resulted from methodological changes made to the survey methodology during the COVID-19 global pandemic. Research and interpretation done using the 2021, 2022, and 2024 GSS data should take extra care to ensure the analysis reflects actual changes in public opinion and is not unduly influenced by the change in data collection methods. For more information on the 2021, 2022, and 2024 GSS methodology and its implications, please visit https://gss.norc.org/Get-The-Data.

Unlike the NSFG however, survey mode didn’t appear to meaningfully affect the results (even after increasing the age range to up the sample size).⁵

1. Those who answered ‘NOT AT ALL’ to SEXFREQ: ‘About how often did you have sex during the last 12 months?’. In 2012, PARTNERS was used (a variable excluded from the most recent survey), as SEXFREQ was only asked to people who had had one or more sexual partners on this year.

2. The sample size for 18–29s whose ballot included the sexual behaviour module and who answered this question was 249.

3. Those who answered ‘I DON'T HAVE A STEADY PARTNER’ to POSSLQ/POSSLQY.

4. The sample size here was 453.

5. This may be partly because the NSFG used a web-first approach, following up later with a face-to-face interview if they didn’t respond, whereas this survey used both modes initially, following up with the alternative later. It could also be because unlike the NSFG survey, saying one hadn’t had sex didn’t allow one to skip a bunch of other questions for a small monetary reward. I couldn’t say.