Case Study 7.1: The Waist-to-Hip Ratio — A Finding That Got More Complicated

Case Study 7.1: The Waist-to-Hip Ratio — A Finding That Got More Complicated

Overview

In 1993, psychologist Devendra Singh published a paper in the Journal of Personality and Social Psychology proposing that men's attraction to women's body shape is not primarily driven by overall weight or body mass but by a specific geometric ratio: the waist-to-hip ratio (WHR). Specifically, Singh argued that a WHR of approximately 0.70 — a waist circumference that is 70% of hip circumference, the classic "hourglass" figure — is universally preferred by men across cultures because it serves as an honest signal of reproductive fitness.

The evolutionary argument was clean and memorable. Low WHR in women correlates with favorable estrogen-to-androgen balance, which in turn correlates with cardiovascular health, reproductive hormone profiles, and lower risk for several diseases including diabetes. A waist that is distinctly narrower than the hips may therefore signal "good genes" and current reproductive health. Singh's initial studies, using line drawings of female figures with WHR systematically varied from 0.70 to 1.0, found consistent preference for lower WHR figures — and he interpreted this as evidence of a universal evolved male preference.

The paper has been cited tens of thousands of times and remains one of the most-referenced findings in the evolutionary psychology of physical attractiveness.

The Evolutionary Argument in Detail

Singh grounded his prediction in parental investment theory: males, as the lower-investing sex in the standard evolutionary framework, have more to gain by attending to cues of female reproductive fitness when selecting mates. Among such cues, WHR has an appealing property as an honest signal — it reflects hormonal milieu during development and is difficult to substantially alter (short of modern surgical intervention). A woman with a naturally low WHR is demonstrating, involuntarily, something real about her endocrine profile.

Singh's findings seemed to hold across age groups and across cultures — at least as reported in his initial studies. Subsequent research added apparent cross-cultural support, with studies in the United States, Germany, and several other countries finding consistent preference for WHR around 0.7.

📊 Research Spotlight: Singh (1993) — The Original Study Singh used nine female figure drawings, divided into three weight categories (underweight, normal, overweight) and three WHR values (0.70, 0.80, 0.90 within each weight category). He found that normal-weight figures with WHR = 0.70 were consistently rated most attractive, healthiest-looking, and most desirable as long-term partners. Figures with WHR = 0.90 were rated lower on all dimensions regardless of weight category.

How the Evidence Got Complicated

More recent cross-cultural research, using more methodologically rigorous designs and populations that genuinely diverge from Western norms, has substantially complicated Singh's original picture.

The weight-versus-ratio confound. Critics quickly pointed out that Singh's line drawings varied both WHR and overall body shape simultaneously. When researchers tested WHR while controlling more carefully for body mass index (BMI), the findings were mixed. Several studies found that BMI (overall body weight) was actually a stronger predictor of attractiveness ratings than WHR — suggesting that what Singh's participants were responding to was total body fatness rather than the specific ratio. The ratio may be a correlate of the real cue rather than the cue itself.

Non-Western populations. The Hadza of Tanzania — hunter-gatherers whose way of life has some resemblance to ancestral conditions that evo psych invokes — were studied by Frank Marlowe and colleagues specifically on WHR preferences. Hadza men, unlike Western men, did not show consistent preference for low WHR. They preferred figures in a higher WHR range than the 0.70 standard and showed stronger sensitivity to overall weight. This finding is striking precisely because the Hadza represent the kind of population that evolutionary accounts of ancestral preference should predict — and they diverged substantially from the prediction.

Historical variation. Artistic representations of the female body across historical periods do not support a universal preference for 0.70 WHR. Pre-Columbian Mesoamerican figurines, European Renaissance painting, and the "Venus" figurines of the Upper Paleolithic all depict female ideal bodies that diverge considerably from the hourglass standard. The considerable variation in "ideal" female body shape across historical periods and artistic traditions is difficult to reconcile with a strong universalist claim.

⚠️ Critical Caveat: What Sample Are You Studying? Most cross-cultural replication attempts for WHR preference have used populations that share more with Singh's original American samples than with genuine cultural diversity. Studies using university undergraduates in Germany, Australia, and Japan are not the same as studying the full range of human cultural variation. The WEIRD-sample problem is acute in the attractiveness literature: if the populations studied share substantial cultural exposure to Western media, finding similar preferences does not demonstrate universality — it may demonstrate the reach of Western beauty norms.

What This Case Teaches Us

The WHR case is instructive not because Singh's hypothesis was entirely wrong — there may well be some evolved sensitivity to body shape indicators of reproductive health — but because of what happened to it as researchers tested it more carefully.

The original finding was specific (0.70 is the magic number), universalist (true across cultures), and evolutionary in framing. Each of those features has eroded under scrutiny. The ratio may matter, but so does weight; the preference is not consistent across genuinely diverse populations; and the evolutionary interpretation competes with cultural learning accounts.

⚖️ Debate Point: Does Erosion Invalidate the Insight? An evolutionary psychologist would reasonably argue that the failure of any specific WHR number to hold up universally does not disprove the broader claim that body-shape cues signal reproductive health and attract male attention. A cultural constructionist would respond that the variation is so substantial, and the non-Western divergences so pronounced, that the universality claim has been abandoned. Both have a point: the underlying logic of honest-signal attraction may be real even if the specific parameter values are culturally inflected.

Discussion Questions

Singh's study used line drawings, not real people. What are the advantages and disadvantages of this methodological choice? How might findings differ using photographs or three-dimensional stimuli?
The Hadza data is one of the most often-cited challenges to WHR universalism. Does one non-Western population's divergence from the predicted preference disprove the evolutionary hypothesis? What would you need to see to conclude the hypothesis is falsified?
Beauty standards for female bodies in Western culture have changed substantially within living memory (compare 1950s Hollywood ideals with 1990s ideals with contemporary ideals). How does this rate of change challenge evolutionary accounts that invoke stable ancestral preferences?

This case study relates to Chapter 7 themes: good-genes hypotheses, cross-cultural universals and their limits, WEIRD-bias in the attractiveness literature, and the just-so story problem.