A blind, randomized preference test. Rate each image on gut feeling; your position is calculated from the ratings, never self-reported.
Loading…
Three parts: faces, figures, then full pictures. Rate 1 to 7, from 1 (“no”) to 7 (“wow”). Judge against the others in this set, not against people in general, and use the whole scale: a 4 should be your typical rating, with plenty of 1s, 2s, 6s and 7s. A few images come back twice to check consistency. Don't discuss your answers until everyone's finished.
No images yet. The test opens once the image set is ready.
Done.
Your ratings couldn't be saved (connection problem). Copy this code and send it to the organizer:
No completed tests yet.
Face compass
Figure compass
Soft ↔ muscular
Add a result code
For friends whose results couldn't save directly.
No completed tests yet.
Face vs. figure
How much each person's rating of the full picture followed the face (teal) versus the figure.
Taste twins
Everyone's ratings
Every woman, every person. Switch between face, figure and full picture; tap a column to sort, tap a photo to see her larger.
Use AI-generated adult women or a licensed research set, never photos of real people you know or found online. Keep everything standardized: neutral expression, plain background, same fitted activewear, same pose.
Add a subject
One subject = one woman. Face crop for part 1, figure cropped at the neck for part 2, full picture for part 3. Skip any image to leave her out of that part.
Place her on each axis. Readers never see these.
SoftAngular
ConventionalDistinctive
Top-heavyBottom-heavy
SlimCurvy
SoftMuscular
Balance check
Aim for similar counts in every cell. Subjects scored 0 on an axis sit on the line and aren't counted.
All ratings
Every woman's ratings for face, figure and the full picture. Tap a column to sort.