- Joined
- Sep 21, 2019
- Messages
- 548
- Time Online
- 1d 6h
- Reputation
- 673
According to a study using a General Linear Model (GLM), researchers examined how much variance in attractiveness could be attributed to facial and body characteristics for short-term and long-term mating preferences, across genders. The GLM modeled variables including short-term and long-term effects for both men and women, as well as the impact of body and facial attractiveness, both independently and combined.
For men, the GLM explained 50% of the variance in the dataset, meaning it accounted for half of the factors influencing attractiveness. The remaining 50% was not studied and could include non-physical factors like personality, or other physical attributes such as height, scent, or vocal pitch.
For women, the GLM explained 42% of the variance, which was surprisingly lower than for men. This difference might reflect the study’s limitations, such as not comparing highly muscular bodies to exceptionally attractive faces (e.g., a male model’s face). The study also didn’t explore how traits interact or compensate for one another—for example, whether a strong physique could offset an unattractive face, or how a combination of great height, body, and face might amplify attractiveness.
A significant limitation of this study is the lack of transparency: the researchers didn’t share the specific body or face images used, provide supplementary data, or include the GLM equations. Additionally, the study is 16 years old, which is nearly a generation. Since then, societal standards of attractiveness have evolved significantly due to the rise of dating apps, Instagram, and TikTok, which have heightened public perceptions of “average” and “handsome.”
Based on the study, of the 42% variance explained for women, 19% is attributed to the face and 13% to the body. To break this down into a pie chart: if 42% of the total variance is explained by face and body combined, the face in isolation accounts for 1942×100≈45% \frac{19}{42} \times 100 \approx 45\% 4219×100≈45% of the explained variance, while the body in isolation accounts for 1342×100≈31% \frac{13}{42} \times 100 \approx 31\% 4213×100≈31%. This suggests that, within the explained variance, the face is more influential than the body, though the body still plays a significant role.
The remaining 58% of the variance (100% - 42%) was not explained by the study, and the researchers didn’t provide equations to model these unknown factors. If I’ve misinterpreted the study, please clarify. I used grok to generate a pie chart but it seems to help visualize it massively