A team of researchers used a machine learning algorithm to examine the best predictors of sexual satisfaction across two samples. Romantic love, dyadic sexual desire, the importance of sex in a relationship, and relationship satisfaction were among the top predictors of sexual satisfaction. The findings were published in the Journal of Social and Personal Relationships.
Psychology studies suggest that couples who enjoy satisfying sex lives have happier and more stable relationships. Researchers Laura M. Vowels and her colleagues set out to examine the extent that a scientific model might be able to predict sexual satisfaction. While previous research suggests that factors like intimacy and commitment contribute to a positive sexual experience, Vowels and her team say no study has compared multiple relationship variables side by side. Elucidating which variables best predict sexual satisfaction may help researchers identify relationship factors to target during interventions.
“There is a lot of research out there on predictors of sexual satisfaction but all of this research is conducted using traditional statistical analyses which cannot reliably tell us about which factors are the most important in predicting sexual satisfaction. Sexual satisfaction, however, is among the most important predictors of relationship success and thus an important area of study,” explained Vowels, a principal researcher for Blueheart.io and postdoctoral researcher at the University of Lausanne.
The study authors turned to machine learning — a scientific approach to data that allows a comparison of the relative predictability of a set of variables. The researchers applied this machine learning algorithm across two separate samples. The first sample was made up of 891 participants with an average age of 32, of whom 54% were straight, 21% were bisexual, and 18% were gay. The second sample consisted of 955 participants with an average age of 30, of whom 56% were straight and 41% were bisexual.
Across both studies, participants completed online questionnaires that addressed various demographic and relationship factors. In the second study, the romantic partners of 754 of the participants also answered the survey. Measures included sexual desire, sexual behavior, contraceptive use, relationship satisfaction, and sexual satisfaction.
The study authors ran analyses to determine which variables explained the most variance in sexual satisfaction. Overall, sexual satisfaction was remarkably predictable, with the assessed variables explaining between 48% and 62% of the variance. The researchers next identified the top ten predictors among 94 variables in the first sample and 71 variables in the second sample.
For both samples, relationship satisfaction was the number one predictor of sexual satisfaction. “Participants who scored high in relationship satisfaction scored up to five-points higher in sexual satisfaction compared to those with average relationship satisfaction,” the researchers report. “In contrast, individuals who scored low in relationship satisfaction, scored up to 15-points lower in sexual satisfaction compared to those with average level of relationship satisfaction.” Desire for partnered sexual activity and for solo sexual activity were also among the top predictors in both samples.
People’s perceptions of sex and love were particularly important among the first sample. Believing that sex is a demonstration of love, but not believing that love comes before sex or that love is most important predicted higher sexual satisfaction. Romantic love was especially important among the second sample, with more romantic love toward one’s partner predicting higher sexual satisfaction. Being in a longer relationship and engaging in diverse sexual behaviors like oral sex and mutual masturbation also predicted greater sexual satisfaction among the second sample. Contrary to some previous data, gender did not significantly predict sexual satisfaction.
As far as partner effects, partner’s sexual satisfaction, romantic love, dyadic desire, and relationship satisfaction were all among the top ten predictors of own sexual satisfaction. “For women, partner’s sexual satisfaction was almost as predictive of women’s sexual satisfaction than their own relationship satisfaction. For men, this association was much smaller,” Vowels and her team report, noting that this finding may reflect the societal view that men’s sexual pleasure is more important than women’s.
The study authors acknowledge the limitations of their study, including that they were unable to investigate all predictors of sexual satisfaction and may have missed out on key variables like self-esteem and sociocultural characteristics.
“The study is purely predictive and we cannot make any causal claims based on the study. We also used two existing datasets so we were limited to the variables that were measured in these datasets and thus there may be other factors we weren’t able to consider,” Vowels said.
The study, “Identifying the strongest self-report predictors of sexual satisfaction using machine learning”, was authored by Laura M. Vowels, Matthew J. Vowels, and Kristen P. Mark.