Abstract
In this paper, we find that a proxy of prenatal testosterone exposure (i.e., digit ratio) is a significant predictor of preferences for products that differ in perceived masculinity vs. femininity. A more masculine (feminine) digit ratio predicts choice of products that have an increasingly masculine (feminine) image. This relationship is statistically significant for male consumers, but not for females.
Introduction
People's sex-typed, or gender-related, characteristics and behaviors are influenced by sex hormones that are present during prenatal development (for a review, see Cohen-Bendahan, van de Beek, & Berenbaum, 2005). For instance, studies in psychology and biomedicine show that prenatal testosterone exposure influences one's sexual orientation and gender identity, as well as other physical, cognitive, and personality characteristics that are gender-related (for a review, see Hines, 2010). In this paper, we replicate these studies in the marketing/consumer domain and examine the link between prenatal testosterone exposure and preferences for gender-imaged products.
In particular, we investigate the link between consumer preferences and the “2D:4D digit ratio”—which refers to the ratio of the length of the index (2D) and ring finger (4D) and which is an established biomarker indicating the level of testosterone to which individuals were exposed before birth (McIntyre, 2006). The digit ratio reliably differs by sex, with males having a lower digit ratio. Moreover, studies find that individuals with lower (vs. higher) digit ratios display more pronounced masculine and less pronounced feminine behaviors than members of their sex typically display. Most relevantly for marketing, the digit ratio has also been shown to predict visual preferences for masculine vs. feminine toys within both genders in a laboratory setting (i.e., eye fixations on masculine vs. feminine toys) (Alexander, 2006). In the present conceptual replication, we investigate whether the digit ratio predicts actual product choices (beyond eye-tracking) in a real consumption environment (beyond laboratory).
Conclusion
Our results are consistent with research showing that sex hormones which are present during prenatal development predict sexually differentiated dispositions and behaviors later in life (Cohen-Bendahan et al., 2005). In particular, we replicate the finding that prenatal testosterone exposure affects product preferences (Alexander, 2006), yet conclude that digit ratio affects not only visual preferences for toys in laboratory, but also actual product and clothing color choices in real markets. We anticipate that prenatal testosterone exposure may affect preferences in other product categories than these as well, because the digit ratio is a relatively permanent, innate characteristic that may exert its influence on a wide range of economic preferences (for a review, see Millet, 2011).
However, we failed to replicate the association between prenatal hormone exposure and product preferences in females. It is possible that males' gendered traits and behaviors are generally more driven by biological factors, whereas women may be more likely to be influenced by situational, social, and cultural factors (see Lippa, 2003). Alternatively, the pattern could also be explained by the fact that women are more likely to experience cyclical shifts in sex hormones that might overshadow the effects of prenatal testosterone. For example, we know that women's clothing choices and color preferences are affected by ovulation (Beall and Tracy, 2013 and Durante et al., 2011).5
In broader terms, our results are consistent with research documenting hormonal (Durante et al., 2011) and even genetic (Simonson & Sela, 2011) influences on consumer behavior. Nevertheless, more future research is needed to better document and understand the effect of biological variables on consumption decisions more broadly.
Regarding the practical implications of the present study, our results suggest that the digit ratio could serve as a useful segmentation variable for marketers who sell distinctly positioned products and brands along the masculinity–femininity continuum. Indeed, marketers are increasingly interested in more subtle gender image-based marketing strategies than to simply target masculine products at males and feminine products at females (cf. Grohmann, 2009). For example, products that have less masculine or more feminine images (e.g., pink shirts; decorated watches) are increasingly sold to male consumers. The fact that we found similar patterns for very different products – (i) a soda beverage, which is a relatively non-conspicuous, inexpensive, and hedonic product, versus (ii) clothing, which is more conspicuous and expensive, and both hedonic and utilitarian – gives confidence that our results may generalize to multiple product categories and consumption settings. Thus, we anticipate that similar results could be found for both markedly cheap and expensive product categories (e.g., soap vs. cars), and for both conspicuous/hedonic and mundane/utilitarian product categories (e.g., entertainment vs. gasoline). These are only speculations, however, and need further replication studies.
