Interpersonal Sexual Desire by Age
About This Chart
Data Visualization Method
- Data: Model-predicted trends for two Sexual Desire Inventory-2 (SDI-2) measures — partner-directed desire and desire toward an attractive other — shown separately for men and women. The values are predictions from the study's age model, not raw age-group averages.
- Sample and period: A cross-sectional online self-report study of 8,150 adults (men n=2,618, women n=5,532), collected between 2007 and 2018 across an age range of roughly 15 to 80.
- Model: Within each sex, desire scores were fit to age with a third-degree polynomial regression, following the original study. The plotted lines are those fitted curves, converted back into each measure's original score units.
- Rescaling: The two measures have different maximum scores (54 for partner desire, 16 for attractive-other desire), so each was divided by its own theoretical maximum and placed on a shared 0–100 display scale.
- Age steps and 60+: Ages run in two-year steps up to a final 60+ point. That last point is not a single-age prediction: it is an N-weighted summary of model predictions for ages 60 through 67, weighted by each sex's age-specific sample size.
- Chart form: An animated line chart where four series move together over age. Each line ends with its current 100-point score and a sex symbol; the numbers sit at the line ends rather than on a Y axis.
Creator's Note
Most charts in this archive are built from public or open datasets. This one works differently — it sits directly on a published study.
The aim was to keep the research record intact, rather than flatten it into a fast visual.
What stands out in the author's reading is not which group scores higher, but the timing. In the model, women's curves begin turning downward earlier — around age 27 for desire toward an attractive other, and around age 32 for partner-directed desire. The men's curves stay flatter for longer.
The interest is in the shape of each curve across age, not a ranking between people.
Data Sources
- Wieczorek et al. (2022) — the Archives of Sexual Behavior study these desire measures and the age model are drawn from.
- OSF data and code repository — participant-level data and analysis code used to reproduce the model predictions.
The values were reproduced from these open files, not digitized from the published figure.
Limitations
These are model-predicted trends — not observed averages, and not individual forecasts. The lines describe the average age pattern the study's model fits across many different people, each measured once. They do not track how a single person's desire changes over a lifetime.
The study is also cross-sectional: different people at different ages, compared at one moment. It does not follow one group over time.
The model's explanatory power for age is very small (adjusted R² of roughly .002 to .017), so age accounts for only a sliver of the variation between people.
The sample was a self-selected online group, not a representative population. Older ages are thinly covered — participants over 60 number only a few dozen — so any late-life turn in the curves should be read with care.
Finally, the 0–100 figures are a display rescaling, not percentages. Partner desire and attractive-other desire come from different sets of items, so the gap between those two lines is not an exact, like-for-like difference in the same unit.
What This Chart Doesn't Show
"Other" here means desire toward an attractive person, measured by two self-report items. It does not mean infidelity, actual behavior, or desire toward strangers in general.
Both measures capture reported desire — not what anyone did.
A score like 73 out of 100 is not a percentage. It does not mean 73% of people, a 73% probability, or a share of respondents. It is one rescaled instrument score on a common display range.
The chart also leaves out the study's solitary desire measure, and it does not merge the two desire types into a single index.