An oldie but goldie:

Mazur, A., Halpern, C., & Udry, J. R. (1994). Dominant looking male teenagers copulate earlier. Ethology and Sociobiology, 15(2), 87-94. DOI: 10.1016/0162-3095(94)90019-1 http://psycnet.apa.org/record/1994-44767-001

Quick summary:

  • Study shows evidence that assessments of facial dominance predicts timing of first sexual encounter
  • Physical attractiveness also contributes, but questionable methodology for its scoring precludes reliably and precisely inferring how it compares with facial dominance as a predictor (see limitations below)

https://i.imgur.com/nDbpsx3.png

https://i.imgur.com/E4BhYSv.png

Original Authors' comments (excerpt from full-text):

Attractiveness and dominant appearance each account for variance in sexual experience beyond that explained by pubertal development, but dominance is the better predictor. This result is consistent with our expectation that dominant looking men have earlier coital opportunities than submissive looking men.

Lacking data on female choice, we cannot say if dominant looking men have more sexual access because women give it to them, or because the men obtain it for themselves, or for both reasons (Small 1992). Possibly women are especially attracted to dominant looking males, assuming from facial cues that these men are capable of protecting them or of providing resources if they have children; or, women may enjoy the popularity of consorting with a man who is salient in the status hierarchy. Alternatively. dominant looking males may obtain sexual experience through their own superior efforts to exploit sexual opportunities, as in the prosaic situation of a high school dance, where the assertive boy immediately seeks a female partner while the shy boy is too bashful to ask for a dance, much less a greater intimacy.

Dominant appearance and attractiveness are imperfectly correlated, so it is worth asking what makes one boy look dominant while another appears submissive? Qualitative inspection of portraits, especially comparing boys judged attractive but not dominant, and vice versa, supports the generalization by Mazur et al. (1984): Dominant faces are likely to be handsome or muscular, oval or rectangular in shape, and with prominent as opposed to weak brow and chin. Submissive faces are often round (pudgy) or narrow (skinny), less attractive, and have glasses.

Methodology

Sample description

  • Sample Size: 58
  • Demographics: males randomly selected for a 3-year panel study of hormones and sexual behavior from a list of all white males in grades 7 - 8 of a mixed urban, suburban and rural school district in a southeastern state. Study completed when the participants were in the 10-11th grade
  • Dates: ~1987

Questionnaire Measures

  • administered in-home with interviewer present; the study used only the final questionnaire at the end of the 3-year study period
  • heavy petting variable: sum of six items inquiring about heavy petting activities
  • two coitus variables: one dichotomy (whether one had ever had sex) and the other cumulative; 43% of the participants had intercourse at least once by the final questionnaire
  • pubertal development: indexed by a factor score based on subjects self-ratings of 8 items related to Tanner stage and other indicators

Measuring Dominance and Physical attractiveness

  • Used high school yearbook portraits, usually senior portraits, copied on slides for projection in random order in front of college classes of 17-28 students who acted as raters
  • raters told to rate each face on a seven point scale of dominance-submissiveness
  • raters given descriptions of behavioral patterns embodied by a dominant male
  • Median score for each portrait used as the participant's facial dominance score
  • Dominance scores subsequently regressed on a number of attributes, only wearing glasses had a significant or sizable effect with wearing glasses scoring on average 1 point less in dominance
  • Physical attractiveness rated by the in-home questionnaire interviewer on a 5-point scale

Limitations

  • Single-rater scoring for physical attractiveness is highly problematic
  • Median scoring for Dominance is similarly problematic, though (arguably) less serious; no indication of the agreement between the various raters; no methods employed to standardize or account for interrater mean and variance differences