Makeup and male attractiveness
Motivation
Makeup is often used by women, and indeed society often expects women to wear makeup whether they want to or not. Men, however, are not usually expected to wear makeup. Because it is unusual, researchers wanted to know if men who wear “subtle cosmetics” would be rated as more attractive by viewers.
The basic experiment was simple, as described in the abstract of the paper:
Twenty men were photographed facing forward, under constant camera and lighting conditions, with neutral expressions, and closed mouths. Each man was photographed twice: once without any cosmetics applied and another time with subtle cosmetics applied by a professional makeup artist. Two hundred participants then rated those 40 images on attractiveness.
Each of the 200 raters rated all 40 images, in random order, so they saw each male participant both with and without makeup. The twenty men were recruited from the area of Franklin and Marshall College through “word-of-mouth, social media, and paper flyers” and paid $10 for their time, while the raters were recruited through Amazon Mechanical Turk and paid $0.80 for their time.
Data
The dataset is provided in two forms.
The first form presents the average rating for each face (of 20) with and without makeup, and can be used for simple analyses of the effect of makeup. The second form presents every rating as a row, producing 200 × 40 = 8,000 observations, each annotated with information about the participant and the rater.
Data preview
male-makeup-faces.csv
male-makeup-ratings.csv
Variable descriptions
| Variable | Description |
|---|---|
| FaceID | Number identifying the participant whose face was rated |
| Age | Age of the participant whose face was rated |
| Race | Race of the participant whose face was rated |
| NoMakeup | Average rating, by the 200 raters, of the attractiveness of the participant’s face without makeup. The scale is from 1 to 7, where 1 = ‘Not at all attractive’ and 7 = ‘Very attractive’. |
| Makeup | Average rating, by the 200 raters, of the attractiveness of the participant’s face with makeup. The scale is from 1 to 7, where 1 = ‘Not at all attractive’ and 7 = ‘Very attractive’. |
| Variable | Description |
|---|---|
| Rater_Sex | Sex of this rater |
| Rater_Age | Age of this rater |
| Rater_Race | Race of this rater |
| Rater | Number identifying the rater |
| Participant | Number identifying the participant whose face was rated |
| Makeup | MM if the participant wore makeup in the photo, NM otherwise |
| rating | Rating of the attractiveness of the participant’s face. The scale is from 1 to 7, where 1 = ‘Not at all attractive’ and 7 = ‘Very attractive’. |
| Participant_Age | Age of the participant whose face was rated |
| Participant_Race | Race of the participant whose face was rated |
Questions
- Conduct a t test or ANOVA analysis on the face dataset. Is there evidence that makeup increases the attractiveness of male faces? How large is the effect, if any?
- Of course, different raters may have different standards, tending to rate higher or lower than other raters. Using the full ratings data, build a model that allows ratings to depend both on the person being rated and on the rater, as well as on whether the participant was wearing makeup. Summarize your results.
- We could think of raters as merely a sample from the population of possible raters who could rate these faces. Adapt your model into a hierarchical model and summarize your results.
References
Batres C, Robinson H (2022). Makeup increases attractiveness in male faces. PLoS ONE 17(11): e0275662. https://doi.org/10.1371/journal.pone.0275662
Data available from the Open Science Framework: https://osf.io/j6zvy/