The importance of femininity to beauty in women

One possible explanation of the typical heavy masculinization seen among the female models in mainstream fashion magazines -- such as Elle -- is that these magazines do not cater to men.  The implication is that women -- at least the ones who patronize the magazines -- are apparently interested in masculinized models, and the publishers are simply using models with looks that appease magazine buyers and maximize sales.  Is this the case?

This question is best answered in light of research on what constitutes attractiveness in women.  As documented below, there is overwhelming evidence that in the absence of physical defects, somewhat above average feminine looks happen to be the most important correlate of beauty in women, as rated by both men and women.  This has been shown regardless of whether facial femininity is actually measured, facial photographs are simply rated, people are allowed to construct a female face they find very attractive using a computer, people are allowed to manipulate the femininity of a face on a computer, or facial composites based on very attractive women and ordinary women are compared.  It is also the case that in these studies inter-rater agreement is usually excellent, and men and women rate attractiveness similarly.

Therefore, it is certainly not the case that the typical heavy masculinization seen among the female models in mainstream fashion magazines is a response to public demand, especially that of women.  If the female models were feminine and attractive, one would expect fashion magazine sales to increase because more men will browse/buy the magazines, just to look at the women.  So why are feminine models uncommon in mainstream fashion magazines?  Well, the editors of these magazines have to cater to the preferences of their unofficial bosses, i.e., the fashion designers who happen to be the major reason for the existence of these magazines, and the top-ranked fashion designers, who happen to be disproportionately gay, undoubtedly care more about what they find aesthetically appealing than magazine sales, and it is well known that what these gays find aesthetically appealing are looks approximating those of adolescent boys.

It could be pointed out that using feminine and attractive women to model fashion products will diminish sales among feminists, but feminists are not the women targeted by fashion magazines.  Additionally, as in the case of the excessive skinniness of fashion models in general, something that most people, including women, find aesthetically unacceptable, the aesthetically unacceptable level of masculinization [for most people] in most high-fashion models does not diminish fashion merchandize sales among women because of the high desirability of designer clothing, which gives fashion designers a broad license as to what kind of models they can use.

Now, on to the studies.  

Table 1 shows the effect of facial femininity on attractiveness ratings of women in individual studies used by Gillian Rhodes in a meta-analysis.(1, pdf)  The effect of femininity on attractiveness is reported in terms of r, a correlation statistic.  The interpretation of r is described below.

The correlation coefficient, r, can range from -1 (perfectly negative) to 0 (none) to 1 (perfectly positive).

Let there be a large sample of women that are rated as either attractive or plain on the one hand, and either feminine or normal on the other hand.  If r = 0.65 (can also be described as r = 65%), then if n% of women rated as normal are also rated as attractive, then n+65% of women rated as feminine will also be rated as attractive.  Similarly, if r = 0.71 and 20% of women rated as plain are also rated as feminine, then 20+71% = 91% of women rated as attractive will also be rated as feminine.       
 
The average r value summed over these studies, and also including the effect of femininity on attractiveness ratings of men, was about 0.64 (Table 2; details for men shown here), which is a high value and shows that femininity is a strong correlate of attractiveness in women.

In Gillian Rhodes’ meta-analysis, femininity emerged as the most powerful correlate of beauty (Table 2).  The r values in Table 2 are based on Zr values, as in the last column of Table 1; Zr values refer to the z-transformed effect size correlations, which are obtained using the following formula.

The Z-transformation is useful for computing the average effect over several studies (meta-analysis) because the more the r value of a population is away from zero, the more the distribution of r values sampled from that population becomes skewed; the Z-transformation helps normalize the distribution.

Let us consider two studies where facial femininity was manipulated on a computer to see what effect it had on the attractiveness ratings.

In the following manipulation by Rhodes et al. (2000), the femininity of the average face on the left has been exaggerated by 50% on the right, but neither of these faces was rated as better looking.

Fig 1. From Rhodes et al. (2000).

Rhodes et al. (2000) then came up with the following series of manipulations of an average face, and in both cases, feminine exaggeration corresponded to highest attractiveness ratings (raters were 48 white males and 48 white females).

The highest attractiveness rating in the two series below was for 36% feminine exaggeration in women.  Women assigned the highest attractiveness rating to a more feminine face than men (42% exaggeration vs. 29% exaggeration).  The most attractive Chinese face corresponded to a greater feminine exaggeration (52% vs. 19%) than the most attractive white face.  Higher attractiveness ratings were also assigned to somewhat more feminized male faces.

Click the two images below for larger views.

Fig 2. Manipulation of an average white female face.  From Rhodes et al. (2000).

Fig 3. Manipulation of an average Chinese female face.  From Rhodes et al. (2000).

Fig 4 shows average (composite) white and Japanese faces used by Perrett et al. (1998).

Fig 4. Composite white and Japanese faces.  From Perrett et al. (1998).

Fig 5 shows average (composite) white and Japanese faces that were 50% feminized (left) and 50% masculinized (right) by Perrett et al. (1998).  Raters were shown faces that continuously ranged between the two extremes and asked to pick the most attractive face.

Fig 5. Feminized and masculinized white and Japanese faces.  From Perrett et al. (1998).

Fig 6 shows that more feminine faces were preferred for both men and women.

Fig 6. The effect of feminization of face shape on judgments of female and male attractiveness.  a, Female stimuli; b, male stimuli. Overall, subjects preferred a feminine face shape to an average shape both within and between populations.  The degree of feminization preferred was greater within than between populations for female faces.  From Perrett et al. (1998).

There are newer studies that have also shown the importance of femininity to attractiveness in women, as assessed by both men and women.  One of these studies showing a higher attractiveness rating of more feminine faces in women has been mentioned at this site previously. Another such study is briefly described below.

In a study investigating the relation between voice pitch, femininity of face and attractiveness rating, women with a more feminine voice (high pitch) also had a more feminine face and were rated as better looking (Fig 7).(2)

Fig 7. Facial prototypes constructed from women with high- and low-pitched voices. Top row: faces from Canadian sample; bottom row: faces from UK sample; left column: faces constructed from women with high-pitched voices; right column: faces constructed from women with low-pitched voices.(2)

And, one should not forget the body, too.  Assume that the following 12 series of images were shown to a large number of people selected at random, and the people were asked to pick the most attractive woman (labeled 1 to 12, starting from top).  Which one would it be?

All images shown below can be clicked for larger versions.

Most people would undoubtedly pick woman number 7 as the most attractive.  Now, which of these 12 women also happens to be the most feminine? 

More can be said about the importance of femininity to beauty in women, but this will be in future additions to this site.

References 

1. Rhodes, G., The evolutionary psychology of facial beauty, Annu Rev Psychol, 57, 199 (2006).
2. Feinberg, D. R., Jones, B. C., DeBruine, L. M., Moore, F. R., Smith, M. J. L., Cornwell, R. E., Tiddeman, B. P., Boothroyd, L. G., and Perrett, D. I., The voice and face of woman: one ornament that signals quality?, Evol Hum Behav, 25, 398 (2005).

List of all studies used by Gillian Rhodes to examine the relationship between masculinity-femininity and facial attractiveness in men and women:

• Bruce V, Burton A, Dench N. 1994. What's distinctive about a distinctive face? Q. J. Exp. Psychol. A. 47A: 119 - 41
  Cunningham MR, Barbee AP, Pike CL. 1990.
• What do women want? Facialmetric assessment of multiple motives in the perception of male facial physical attractiveness. J. Pers. Soc. Psychol. 59: 61 - 72.
• Dunkle JH, Francis PL. 1990. The role of facial masculinity/femininity in the attribution of homosexuality. Sex Roles 23: 157 - 67.
• Ishi H, Gyoba J, Kamachi M, Mukaida S, Akamatsu S. 2004. Analyses of facial attractiveness on feminised and juvenilised faces. Perception 33: 135-45.
• Johnston VS, Franklin M. 1993. Is beauty in the eye of the beholder? Ethol. Sociobiol. 14: 183 - 99.
• Johnston VS, Hagel R, Franklin M, Fink B, Grammer K. 2001. Male facial attractiveness: evidence for hormone-mediated adaptive design. Evol. Hum. Behav. 22: 251 - 67.
• Koehler N, Simmons LW, Rhodes G, Peters M. 2004. The relationship between sexual dimorphism in human faces and fluctuating asymmetry. Proc. R. Soc. Ser. B - Bio. (Suppl.) 271: S233 - S6.
• Little AC, Burt DM, Penton-Voak IS, Perrett DI. 2001. Self-perceived attractiveness influences human female preferences for sexual dimorphism and symmetry in male faces. Proc. R. Soc. Ser. B - Bio. 268: 39-44.
• Little AC, Hancock PJB. 2002. The role of masculinity and distinctiveness in judgments of human male facial attractiveness. Br. J. Psychol. 93: 451-64.
• Neave N, Laing S, Fink B, Manning JT. 2003. Second to fourth digit ratio, testosterone and perceived male dominance. Proc. R. Soc. Ser. B - Bio. 270: 2167-72.
• O'Toole AJ, Deffenbacher KA, Valentin D, McKee K, Huff D, Abdi H. 1998. The perception of face gender: the role of stimulus structure in recognition and classification. Mem. Cognit. 26: 146-60.
• Penton-Voak IS, Jones BC, Little AC, Baker S, Tiddeman B, et al.2001. Symmetry, sexual dimorphism in facial proportions and male facial attractiveness. Proc. R. Soc. Ser. B - Bio. 268: 1617-23.
• Perrett DI, Lee KJ, Penton-Voak I, Rowland D, Yoshikawa S, et al. 1998. Effects of sexual dimorphism on facial attractiveness. Nature 394: 884-7.
• Rhodes G, Chan J, Zebrowitz LA, Simmons LW. 2003. Does sexual dimorphism in human faces signal health? Proc. R. Soc. Ser. B - Bio. 270: S93-S5.
• Rhodes G, Hickford C, Jeffery L. 2000. Sex-typicality and attractiveness: are supermale and superfemale faces super-attractive. Br. J. Psychol. 91: 125-40.
• Scheib JE, Gangestad SW, Thornhill R. 1999. Facial attractiveness, symmetry and cues of good genes. Proc. R. Soc. Ser. B - Bio. 266: 1913-7.
• Swaddle JP, Reierson GW. 2002. Testosterone increases perceived dominance but not attractiveness in human males. Proc. R. Soc. Ser. B - Bio. 269: 2285-9