Scandinavian Journal of Psychology, 2008, 49, 429–437 DOI: 10.1111/j.1467-9450.2008.00651.x

© 2008 The Authors. Journal compilation © 2008 The Scandinavian Psychological Associations. Published by Blackwell Publishing Ltd., 9600
Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA. ISSN 0036-5564.

Blackwell Publishing Ltd Personality and Social Sciences

The influence of skin tone, hair length, and hair colour on ratings of
women’s physical attractiveness, health and fertility

VIREN SWAMI,

1

ADRIAN FURNHAM

2

and KIRAN JOSHI

2
1

Department of Psychology, University of Westminster, UK

2

Department of Psychology, University College London, UK

Swami, V., Furnham, A. & Joshi, K. (2008). The influence of skin tone, hair length, and hair colour on ratings of women’s physical
attractiveness, health and fertility.

Scandinavian Journal of Psychology

,

49

, 429–437.
The present study investigated the role of skin tone, hair colour and hair length in perceptions of women’s physical attractiveness, health and
fertility. One-hundred and thirty men and 112 women rated a series of 12 line drawings that varied in three levels of skin tone, two levels of
hair colour and two levels of hair length. Results showed a number of interactions between the three variables, suggesting that these phenotypes
are highly intercorrelated. However, there were also significant main effects of each of the variables, with hair colour generally explaining the
greatest amount of variance. In general, light-toned figures were rated the most positively. Contrary to expectations, however, brunettes were
rated more positively than blondes, and hair length had only a weak effect on ratings of attractiveness. Implications of these findings and the
limitations of the use of line drawings are discussed in conclusion.

Key words:

Physical attractiveness, skin tone, hair colour, hairstyle, health, fertility.

Viren Swami, Department of Psychology, University of Westminster, 309 Regent Street, London W1B 2UW, UK.

E-mail: virenswami@hotmail.com

INTRODUCTION
Recent years have witnessed an explosion in research
examining the attributes that women and men find attractive
in potential partners (Swami, 2007). This body of work has
tended to focus on facial characteristics and body size (for
a review, see Swami & Furnham, 2008: ch. 5–6), which is
surprising given the multitude of bodily features that could
potentially be quantified (Pokrywka,

Ç

abri

h

& Krakowiak,
2006). The present study is a preliminary attempt to contribute
to filling this gap in the literature by focusing on three bodily
characteristics that have received relatively little focus,
namely skin tone, hair style and hair colour. In the following
sections, we briefly review the relevant literature relating to
these three characteristics (for a fuller discussion, see Swami
& Furnham, 2008: ch. 4).

Skin tone

The available research on judgements of skin tone suggests
that, within ethnic groups, there is a general preference for
women with skin tones lighter than the local average and
men with skin tones darker than the local average (Aoki,
2002; Frost, 2005). In a widely-cited study, van den Berghe
and Frost (1986) reported that, of the 51 societies for which
data were then available, 47 preferred lighter-toned skin in
women. This mirrors a natural difference in skin tone
among women and men in almost all cultures, with women
tending to be the fairer sex as a result of both natural (Jablonski
& Chaplin, 2000) and sexual selection (Frost, 2006).
Across ethnic groups, however, the picture is complicated
by the legacy of colonialism and racism in some societies,
which has typically meant a denigration of darker skin tones
and an association between fairness of skin with goodness
and power (see Hall, 1995; Russell, Wilson & Hall, 1992;
Swami, 2007). Thus, a number of studies have shown that
dark-toned individuals face more discrimination than
individuals with light-toned skin (Frisby, 2006; Keenan, 1996;
Leslie, 1995), and that European (Caucasoid) features
are accorded higher status, even within some minority ethnic
groups (Maddox, 2004; Thompson & Keith, 2001).
Finally, there is a strong association between skin tone
and social status (Keith & Herring, 1991). It has been suggested
that, in most socioeconomically developed societies, higher
incomes and ease of travel to warmer climates during the
winter has allowed skin tone to become a ubiquitous symbol
of a person’s socioeconomic status or occupation. In developed
societies in the West, being able to get a tan serves as a
symbol of status, a message relayed to individuals through
the mass media and social interactions (Jackson & Aicken,
2000; Wichstrom, 1994). Indeed, many studies have found
that a tan appearance has a positive effect on the appearance
of both women and men (Bond & Cash, 1992; Broadstock,
Borland & Gason, 1992; Hill, 2002; Jackson & Aicken,
2000; Miller

et al.

, 1990; Sahay & Piran, 1997; Smith,
Cornelissen & Tovée, 2007).

430

V. Swami

et al.

Scand J Psychol 49 (2008)
© 2008 The Authors. Journal compilation © 2008 The Scandinavian Psychological Associations.

Hair colour and length

Although there are many interesting features of head hair,
including the quality and quantity of hair, hair length, hair
colour and lustre (Graham & Jouhar, 1981), and despite
hair being an important individual marker (Bull & Hawkes,
1982), relatively little research has examined the impact of
cranial hair on perceptions of physical attractiveness. One
recent exception is Frost’s (2006) application of the “rare-colour
advantage” hypothesis to explain the evolution of blonde
hair. More generally, a number of studies have found evidence
for a rare-colour advantage in terms of hair colour. Thelen
(1983), for instance, showed that the preference for a
brunette increased in proportion to the rarity of her hair
colour. Corroborative evidence suggests that women tend to
change their hair colour and style to a type that is less
common in the general population (Schweder, 1994).
On the other hand, there is also evidence that perceptions
of hair colour vary with changing fashions and media icons.
Rich and Cash (1993) examined the frequency with which
different hair colours were represented in women’s maga-
zines over four decades and concluded that the percentage
of blondes was generally over-represented. The one excep-
tion was the 1960s, when Jackie Kennedy (a brunette)
became

the

style icon in much of North America, if not
internationally. More generally, in the West blondes tend to
be perceived as more attractive, seductive and out-going,
although they may also be perceived as untrustworthy,
manipulative and promiscuous (Guthrie, 1976; Lawson,
1971).
Fewer studies have dealt with hair length and its effect on
perceptions of attractiveness. Early studies suggested that
men with long hair are judged to be more feminine, non-
conforming and youthful, and less potent, educated and
intelligent than men with short hair (Pancer & Meindl, 1978;
Peterson & Curran, 1976). However, these studies were con-
ducted at a time in Western political history when the length
of a man’s hair was equated with sociopolitical activism. For
women, Terry and Krantz (1993) reported that long hair was
associated with decreased social forcefulness, a finding
attributed to long female hair softening the contour of the
face and giving the appearance of youthfulness.

The present study

In the present study, we sought to examine the relative
contributions, and possible interactions, of skin tone, hair
length and hair colour to judgements of women’s attractive-
ness, health and fertility. These variables were selected
primarily on the basis of their relative neglect within the
literature on physical attractiveness, but also due to their
phenotypic relation. To achieve this aim, we made use of line
drawings of the female figure modified from those used by
Swami

et al.

(2007). It should be noted that studies of this
kind have very rarely introduced colour to their stimuli,
which makes this a novel application in itself. On the other
hand, line drawings lack the ecological validity of photographic
or three-dimensional stimuli (Bateson, Cornelissen & Tovée,
2007; Smith

et al.

, 2007; Treleaven, Furnham & Swami, 2006),
although they do allow each characteristic to be manipulated
independently and systematically.
Based on the above review of the literature, a number of
hypotheses were formulated prior to the investigation.
Specifically, we predicted that our British participants would
show a preference for long (as opposed to short) hair, light-
toned or tan (as opposed to dark-toned) skin and blonde
hair (as opposed to brunettes) when making judgements of
attractiveness. However, we also predicted that there would
be significant interactions between these characteristics,
particularly between skin tone and hair colour (after Swami

et al.

, 2007). We further hypothesised that ratings of health
and fertility would generally mirror those of attractiveness.
METHODS

Participants

The participants in this study were 130 men and 112 women
recruited opportunistically from a campus setting in Greater
London. The sample had a mean age of 22.73 years (

SD

= 6.01)
and a mean body mass index (BMI) of 22.31 kg/m

2

(

SD

= 3.01),
which falls within the normal BMI range. To ensure that there was
no variance in participants’ ethnic backgrounds, only participants
who self-reported as being of European Caucasian descent were
invited to take part in the study. Although we did not collect further
demographics, we are fairly certain that the sample is representative
of relatively affluent members of British society, certainly with
access to print and electronic media. All participants took part in
the study on a voluntary basis and were not remunerated for their
participation.

Stimuli

The stimuli were modified versions of the line drawings developed
by Swami

et al.

(2007), which themselves had been modified from
greyscale images developed and used by Furnham, Swami and Shah
(2006). For their study, Swami

et al.

(2007) introduced colour to the
stimuli using Microsoft Paint, and developed a set of 18 stimuli that
varied in three levels of skin tone (pale, tan and dark), three levels
of body weight (underweight, normal weight and overweight) and
two levels of hair colour (blonde, black-haired). In the present
study, we followed the same technique, but rather than including
body weight as a bodily manipulation, we instead included two levels
of hair length (short, long). Thus the final set of stimuli consisted
of 12 stimuli that varied in three levels of skin tone, two levels of
hair colour and two levels of hair length (see Fig. 1).

Procedure

Participants were approached and invited to take part in a study on
human beauty in various campus settings, such as a cafeteria,
communal spaces and libraries. Testing was conducted in small
group settings (typically between 10 and 20 participants per group).
The stimuli were printed as high-quality images on cards measuring
210

×

297 mm

2

, each with a unique identifying letter (see Fig. 1).