New study fails to replicate the link between fertility in women and wearing red clothing

A new replication study published in the peer-reviewed journal Hormones and Behavior casts doubt on claims that women unconsciously signal their fertility by wearing red clothing.

Previous research had found that women were more likely to wear red or pink at peak fertility. In other female mammals, such as chimpanzees, red skin coloration is used to advertise the fertile window in an observable manner (estrus). Humans do not display observable physical changes associated with ovulation, but it was thought women may still be using the color red to attract mates at peak fertility.

The new replication study tested the association between ovarian hormones and wearing red clothing using an ethnically diverse sample of 164 women who were 18 to 36 years old. But the researchers found no compelling evidence that women who were more fertile were more likely to wear red than another color of clothing.

There was a weak relationship between fertility and wearing red only among 18– to 22-year-old women, but this relationship could be a statistical artifact. The relationship only existed when all non-red clothing was grouped into a single category. “We did not find compelling evidence of this effect when using other statistical techniques,” the study concludes.

PsyPost interviewed the study’s corresponding author, Khandis R. Blake of the University of New South Wales, about the research. Read her responses below:

PsyPost: Why were you interested in this topic?

Blake: I am interested in female intrasexual competition, and the hypothesis that women choose to wear red clothing when they are fertile has potentially profound implications for understanding the evolution of female mating strategies. We were also in a lucky position in that the data we analyzed was collected as part of a larger study and for another purpose. We had photographs of women’s clothing at two time points and corresponding estradiol and progesterone values already, so this meant it was easy for us to see if we could replicate recent findings.

What should the average person take away from your study?

Past work has shown that women’s tendency to wear red clothing tracks their potential fertility, including their estradiol to progesterone ratio. The E:P ratio is a hormonal marker of fertility in the menstrual cycle, where high values indicate a greater likelihood of conception from intercourse. We show that the relationship between the estradiol to progesterone ratio and the likelihood of wearing red only arises for women who are very young (18–22 years old). Because researchers have used the relationship between potential fertility and red clothing to claim that women exhibit estrus, this is an important finding. If only a subgroup of fertile women wear more red when they are fertile, and we have no reason to doubt their ability to conceive (which we didn’t)—then red clothing is probably not a cue of human ovulation.

The other, perhaps more important thing to take away is that we could only replicate the relationship between the E:P ratio and red clothing when using a specific kind of statistical technique. This technique involved grouping all possible color combinations par red (and pink) and comparing this amalgamated group to red. This is the technique other researchers who have found the red-fertility effect have used. The technique doesn’t distinguish between the frequency of other color combinations, and instead groups everything that isn’t red as one color (“not red”) and compares this to red (it’s a binary technique). In everyday life, the choice between choosing clothing colors is probably more complex than this grouping allows. Testing the hypothesis that women choose to wear one color when E:P ratios are high should also ideally include an investigation of all possible color choices as well.

In our study, we undertook this approach. We compared binary and multinomial analyses to see whether we could replicate the red-fertility effect using different techniques. When we investigated the association between the E:P ratio and red clothing when differentiating amongst the other colors worn by the sample (a multinomial technique), no compelling relationship between the variables arose. What does this mean? Together, these results show that the red-fertility effect only arises under particular statistical conditions and only for a subgroup of women (very young women).

Are there any major caveats? What questions still need to be addressed?

One unresolved question is why there was a positive association between E:P ratios and the probability of wearing red for younger women (albeit one we couldn’t replicate using multinomial techniques). It seems as if red clothing may be a mate attraction tactic employed by young potentially fertile women in some conditions, but future research is needed to replicate this effect and identify when and for whom the effect applies. My hunch is that the use of red clothing amongst younger, fertile women—if the effect replicates—is probably related to experience effects. Future work will tell.

Is there anything else you would like to add?

Signaling sexual proceptivity or enhancing attractiveness are just two of many possible reasons for women’s clothing color choices. The role of other factors that may potentially influence clothing color choice—for example, mood, clothing color availability, or stable personal preferences for particular clothing colors—have been neglected by empirical research. Future studies could investigate whether fertile women prefer ornamentation that captures visual attention rather than red ornamentation per se, or whether other signals of sexual proceptivity (e.g., sexualized clothing) are affected by ovarian hormones and mating-relevant goals. Women’s choice to wear red clothing is probably more nuanced than previously theorized.

The study, “No compelling positive association between ovarian hormones and wearing red clothing when using multinomial analyses“, was also co-authored by Barnaby J.W. Dixson, Siobhan M. O’Dean, and Thomas F. Denson.