Gender-Typed Colour Preference

What it is

The cleanest early test is a 2011 experiment by Vanessa LoBue and Judy DeLoache, published in the British Journal of Developmental Psychology. It tested 192 children from 7 months to 5 years. Each child saw pairs of small identical objects, such as coasters and plastic clips, that differed only in colour. One object in every pair was pink, and the other was green, blue, yellow or orange. The child reached for one.

The results, as reported in the British Psychological Society's summary of the paper, have a clear shape. Under about two and a half, boys and girls picked pink about half the time, which is chance. Girls chose it more than boys at age two but not earlier, clearly by two and a half, and just under 80 per cent of the time by age four, a bias that weakened by age five. Boys chose pink less often than chance at ages two, four and five, reaching about 20 per cent by age five. A second experiment with 64 children aged two to three narrowed the window and found the split opening between two and a half and three. The authors argued that an innate preference should be visible whatever the child's grasp of gender, and read the timing, which coincides with the age at which children begin to understand gender, as pointing to learning.

A 2010 study by Vasanti Jadva, Melissa Hines and Susan Golombok used a preferential looking task with 120 infants aged 12, 18 or 24 months. It found no significant sex differences in preferences for colours or shapes, and girls and boys alike preferred reddish colours over blue. The authors suggested girls may learn to prefer pink because the toys they enjoy are often pink. This page read the paper's abstract, not its full text.

In effect

Wong and Hines followed 126 toddlers in a study published in Archives of Sexual Behavior in 2015. The children averaged 29 months at the start, with a range of 20 to 40 months, and were tested twice, six to eight months apart. Colour preferences were measured with colour cards and with neutral toys recoloured in gender-typed colours. Toy and activity preferences were measured by parent report on the Preschool Activities Inventory.

A sex difference in colour preference appeared between ages two and three and grew stronger close to the third birthday, when it was large, with an effect size above 1. An effect size above 1 means the two group averages sat more than one standard deviation apart. The toddlers' colour preferences were stronger and less stable than their parents', emerged later than their toy and activity preferences, and were largely uncorrelated with them. The authors read that last result as suggesting colour preference and toy preference are shaped partly by different factors, and proposed social and cognitive influences, together with play with toys that happen to come in gender-typed colours.

A companion paper by the same authors, on gender colour-coding and toddlers' play with a train and a doll, found that gender differences in toy preference were larger when the toys came in gender-typical colours and, at the second testing, appeared only in those colours. Boys and girls combined played more with the toy typical of the other sex when its colour matched their own sex.

By adulthood the preference has mostly gone. In Swiss samples reported in 2019 by Domicele Jonauskaite and colleagues, girls aged 10 to 14 chose pink or purple more often than boys, but almost no adult of either sex named pink as a favourite. The authors link the pattern to gender stereotypes and status differences between women and men. The wider adult picture, including a laboratory lean towards red in women that a later comparison did not repeat, is on the page for sex differences in colour preference.

The mechanism most often offered for how a colour comes to belong to a group is ecological valence: liking tracks the things a colour is attached to. In the 2011 Berkeley and Stanford study, each university's students rated their own university's colours higher, and more so the greater their school spirit. For toddlers, the step from a colour's attachments to a child's sense of belonging is the authors' proposal.

What it does not say

It does not say the preference is purely learned. The toddler studies show when the split appears and what it correlates with. They do not rule out a biological contribution, and the authors of the 2015 study describe their explanation as a suggestion.

It does not say the effect is large in general. The large effect size belongs to toddlers near their third birthday, measured with colour cards and recoloured toys, at an age when the preference is described as unstable. It is smaller or absent in the other samples reviewed here.

It does not say pink does anything to anyone. Whether pink calms people is a separate question, and the tests with a comparison group are described on the page for Baker-Miller pink.

---