The influence of biological sex, sexuality and gender role on ...

Copyright © The British Psychological Society
Reproduction in any form (including the internet) is prohibited without prior permission from the Society
579
The
British
Psychological
British Journal of Social Psychology (2006), 45, 579–597
Society
q 2006 The British Psychological Society
www.bpsjournals.co.uk
The influence of biological sex, sexuality
and gender role on interpersonal distance
David Uzzell1* and Nathalie Horne2
1Department of Psychology, University of Surrey, UK
2Research Development Centre, Greenwich, Lambeth, Lewisham & Southwark
PCTs, London, UK
This research reports on a conceptually and methodologically innovative study, which
sought to measure the influence of gender on interpersonal distance. In so doing, we
argue for an important distinction to be made between biological sex, gender role, and
sexuality. To date, however, progress in the study of interpersonal distance (IPD) has
been inhibited by poor operational definitions and inadequate measurement
methodologies. For our own investigation, we innovated on methodology by devising
the digital video-recording IPD method (DiVRID) that records interpersonal spatial
relationships using high quality digital video equipment. The findings highlighted not only
the validity of our innovative method of investigation, but also that a more sophisticated
conceptualization of the impact of gender on IPD is warranted than can be accounted
for by biological sex differences. In this study, we found that gender role accounts for
more of the variation in IPD than the conventionally reported gender variable, sex.
Of more than 1,200 papers on personal space between 1964 and 2003 recorded in the
PsycInfo database, two-thirds (67.6%) were published before 1983. Since that date,
there has been a marked decline in published studies and presumably research (1964 to
1973: 95; 1974 to 1983: 725; 1984 to 1993: 211; 1994 to 2003: 182). Such a decline
cannot be attributable to a lessening relevance of the research or its findings. The
importance of the socio-environmental context and contingencies that affect social
interaction means that the study of personal space is highly relevant to our
understanding of processes in social psychology (e.g. personal attraction, prejudice).
Current concern with quality of life, housing standards, and working conditions also
raises questions concerning the degree and quality of personal space available to
individuals and groups.
Why then the decrease in research interest in interpersonal space? We believe that
after several decades of studies there was a gradual realization that research on personal
space suffered from two principal shortcomings. First, and most fundamental to the
* Correspondence should be addressed to Professor David Uzzell, Department of Psychology, University of Surrey, Guildford,
Surrey, GU2 7XH, UK (e-mail: d.uzzell@surrey.ac.uk).
DOI:10.1348/014466605X58384

---

Copyright © The British Psychological Society
Reproduction in any form (including the internet) is prohibited without prior permission from the Society
580
David Uzzell and Nathalie Horne
conclusions and inferences we draw, are weaknesses in the methodologies employed to
measure interpersonal distance. Charges can be made on ecological grounds
(i.e. differences obtained in controlled laboratory settings are not a reflection of real
life differences) and on grounds of unreliability or inaccuracy.
Second one of the principal factors investigated to identify individual and intergroup
differences in the use of personal space was gender. Research sought to show that there
were significant gender differences in interpersonal distance (IPD) behaviour. The
theoretical formulation and operationalization of gender differences, however, has been
poor. In particular, changes in the definition and interpretation of gender over the last
decade invite us to revisit and question assumptions that were taken for granted in the
early years of research on personal space (e.g. increased dissociation between biological
sex and learned gender role). Moreover, the study of gender and the use of space has
become an increasingly litigious area. For example, in legal cases of sexual harassment,
spatial proximity can be a critical issue. Hern (1991) cites examples of judges specifying
minimum distances between anti-abortion protesters and clients in demonstrations
outside clinics.
In the current study, we address these two shortcomings by, firstly, devising a new
methodology drawing on recent technological developments, which enable a more
accurate and ecologically valid recording of personal space relationships, and, secondly,
by arguing for a more sophisticated conceptualization of sex differences using the
gender role concept. In particular, we look at gender differences in spatial behaviour
using the operational concept of IPD (as opposed to personal space, proxemics,
individual distance, interpersonal spacing, body-buffer zone, etc.) as this describes
concisely the focus of investigation in this paper.
Describing sex differences in IPD behaviour
In the study of IPD, it has been conventional to investigate gender differences simply by
comparing male and female behaviour against various benchmark indicators. On this
purely descriptive level, the general consensus among researchers (notwithstanding the
many inconsistencies and exceptions in the findings) is that male pairs maintain larger
distances than female pairs at all ages (Aiello, 1987; Barnard & Bell, 1982; Sussman &
Rosenfeld, 1982). Moreover, many studies have found that sex differences have most
impact on IPD in combination with other factors, such as race and age (Heckel & Hiers,
1977; Severty, Forsyth, & Wagner, 1979). However, these findings must be located in the
context of some quite serious methodological shortcomings in the study of IPD. Three
principal methodologies have been employed to examine variation in personal
distances; projective, laboratory, and observation. Projective techniques involve asking
participants to hypothetically imagine a situation and indicate, either with pencil or
paper or using dolls or figures, how they believe they or another individual would
respond spatially in that scenario (e.g. Duke & Nowicki, 1972; Gifford & Price, 1979;
Little, Ulehla, & Henderson, 1968; Summit et al., 1992). Reviewing the studies using this
technique, Hayduk (1983) concluded that it is simply not credible. It has several obvious
flaws such as requiring complex cognitive skills like reconstruction, imagination,
empathy, memory demands, and most difficult of all, re-scaling from life-size to the
perceived scale of the figures.
Laboratory measures have been also widely used (e.g. Dosey & Meisels, 1969; King,
1966; Kunzendorf & Denney, 1982). The most common laboratory method is the ‘stop-
distance’ method in which the experimenter asks one participant to enter a room and

---

Copyright © The British Psychological Society
Reproduction in any form (including the internet) is prohibited without prior permission from the Society
Sex, sexuality and gender role and interpersonal distance
581
approach another subject until the point when they start to feel uncomfortable with the
other’s proximity. Alternatively, ‘approach distance’ is used in which subjects are asked
to move towards another person(s) or person surrogate (e.g. a photograph of different
types of people) and indicate at which point they cease to feel comfortable. Laboratory
studies have the advantage that they employ human beings rather than dolls, they are
easily administered, and by contriving the setting to resemble, for example, an office,
the experiment can have some degree of ecological validity. However, accurate
measurements are difficult as participants are asked to use their own bodies to position
themselves according to hypothetical situations.
The third methodology, observation, has most ecological validity as it involves direct
observation of people interacting with each other in real situations and preferably by
unobtrusive means; it also gives rise to the most practical difficulty (e.g. Jorgenson,
1975; White, 1975), principally in the accurate measurement of interpersonal distances.
There are two types of observation which strive to be unobtrusive and field-based:
(a) unstructured, which is largely a naturalistic, unobtrusive, and uncontrolled
observation, and reflects people interacting in a real-world setting; (b) staged invasions
or blocked access in natural settings. In both situations, either unwitting subjects are
approached by a confederate or the paths of people are blocked by confederates and the
reactions monitored.
Two shortcomings are crucial to each of these techniques – ecological validity and
accuracy of measurement, with some resulting trade off between the two. While
observation studies and even laboratory studies can have ecological validity, to date,
they have not been able to permit accurate measurements of interpersonal distance.
This is likely to be a critical factor as variations in interpersonal distances are probably
extremely small and subtle. In defence of field experiments, it is only fair to note that
their goal has also been to identify variations in behavioural response, not to measure
the actual IPD between subjects. Likewise, projective techniques lack ecological validity
and it is doubtful whether they can provide accuracy of measurement either.
If the issue of gender differences is to be explored and explained with any degree of
confidence, then studies will require the employment of a highly accurate IPD recording
methodology that can be used within an ecologically valid setting.
Explanations of gender differences in IPD
Gender is not a biological fact, but a social and cultural construction prescribing how
men and women should behave. In recent years, however, there has been an increasing
dissociation between sex and gender, insofar as it is becoming more acceptable for
women to endorse at least some masculine traits and characteristics and for men to
endorse at least some feminine traits and characteristics. The perceived acceptability of
these gender role deviations are nonetheless highly person and context-dependent, and
in part a function of the extent to which there is deviation. However, the fact remains
that dissociation is possible; thus to automatically conflate sex with gender is now
considered inappropriate.
Explanations for variations in IPD can be characterized as having either focused on
the function of space in mediating interpersonal relationships and/or the higher order
factors that moderate variations in the functional use of space, with gender in particular
being one of the key explanatory variables. On a functional level, explanations have
centred on the role played by IPD in communication self-protection, and arousal
regulation. The communication function of IPD (Hall, 1959, 1963, 1966) tells another

---

Copyright © The British Psychological Society
Reproduction in any form (including the internet) is prohibited without prior permission from the Society
582
David Uzzell and Nathalie Horne
person and onlookers information concerning the nature of the relationship between
the individuals (e.g. it can distinguish between intimate and public relations, but also
closeness can be used to demonstrate affection for another). The self-protective
function (Dosey & Meisels, 1969, 1971) pertains to the idea that the greater the
individual perceives risk to their emotions, physical being or privacy, the greater the IPD
required to enable escape (Altman, 1975; Edney, Walker, & Jordan, 1976). Finally, the
arousal regulation function (Evans, 1974) uses IPD to control the amount of sensory
information they are receiving and avoid sensory overload.
Arguably, none of the above ‘functions’ of IPD are mutually exclusive and can be
considered to be lower order explanations (or manifestations) of higher order level
concepts like gender, age, attractiveness (Gifford, 1997), culture (Aiello, 1987),
personality (Gifford, 1997), disability (Eaton, Fuchs, & Snook-Hill, 1998; Rapp &
Gutzmann, 2000; Sommer, 1669, 2002), and other contextual factors (Sinha & Navyar,
2000). This study focuses on one especially powerful ‘moderating’ factor – gender.
Previous research has generally concluded that same-sex female pairs maintain closer
distances than same-sex male pairs (Aiello, 1987; Gifford, 1997). It has been argued that
basic sex differences are due to either a female predisposition to be more affiliative than
males and/or a stronger female socialization to be affiliative. One way to test this
hypothesis is to distinguish between the concepts of biological sex (i.e. male, female)
and gender role (i.e. masculine, feminine). To date, these concepts have been conflated
(i.e. used interchangeably) and although, in practice, the two concepts are likely to be
highly correlated in Western societies, there is no necessary association between the
two. In other words, men can be feminine and women can be masculine. Moreover, it is
not a simple case of assuming one or other gender role. Most individuals of either sex in
contemporary Western culture are likely to subscribe to a greater or lesser degree and in
varying complex combinations to both gender role concepts. Thus, while a female may
self-endorse a predominantly feminine gender role, she may also endorse some
masculine characteristics too. In short, applications of gender-related terms to the study
of human spatial behaviour have been confused and operationally inappropriate,
confounding basic biology with gender.
Explanations of sex differences with reference to social learning concepts require
some acknowledgement and appreciation of the socially constructed nature of the
gender roles, culturally known by the terms ‘masculinity’ and ‘femininity’ (Gilligan,
1982). Masculinity refers largely to instrumental traits and characteristics (e.g.
competitive, means-end, power-distant) while femininity refers to more relational traits
and characteristics (e.g. expression, collaborative, caring, affiliative). The concept of
gender role is permeated with normative and evaluative judgments about how men and
women should behave in accordance with masculine and feminine stereotypes. People
either comply with gender role stereotypes (which create cultural expectations and
obligations) or internalize them (in which case, the stereotype will be assimilated to self-
identity; Pulling & Stark, 2000).
Whether externally or internally generated, behaviour under the regulatory impact
of gender role stereotypes will exhibit predictable patterns. For example, subscribing
strongly to a feminine gender role will manifest itself in largely affiliative attitudes and
behaviours. By contrast, subscribing to a masculine gender role will manifest itself in
largely instrumental goal-oriented attitudes and behaviours, involving distance, and
means-end control (e.g. Rees, 2003).
Early work on gender roles emphasized the correlation between sex and gender
(i.e. males are and should be mostly masculine, while females are and should be mostly

---

Copyright © The British Psychological Society
Reproduction in any form (including the internet) is prohibited without prior permission from the Society
Sex, sexuality and gender role and interpersonal distance
583
feminine), and assumed that masculinity and femininity were polarized (see Bem, 1993
for a review). Masculinity and femininity were treated as ‘states’ occupying opposite
ends of a bipolar masculinity-femininity scale. However, in contemporary conceptu-
alizations of gender, it is appreciated that not only are the concepts ‘fully independent’
(which means that any one individual can have attributes of both), but that the
relationship between sex and gender is far more complex than originally assumed, with
the added dimension of sexuality (or sexual preference) also to be taken into
consideration.
For instance, homosexuality has been related to gender role ‘deviations’ or gender
role reversal (Al-Issa, 1987). An influential study by Thompson, Schartz, McCandless,
and Edwards (1973), and similar research by Evans (1971) and Heilbrun and Thompson
(1977) found that lesbians scored higher on masculinity compared with their ‘straight’1
controls, and gay men were less masculine and sometimes more feminine than their
straight controls. Others have consistently found likewise (Spence & Helmreich, 1978;
Oldham, Farnill, & Ball, 1982). Conversely, Bernard and Epstein (1978) found that gay
men scored higher than straight men on femininity.
The most well-established means of investigating gender role is known as the Bem
Sex Role Inventory (BSRI). It includes two scales: a Masculinity scale and a Femininity
scale constructed specifically for the purpose of identifying individuals’ gender roles as
masculine, feminine or androgynous, achieved by assessing the differences between the
subjects’ endorsements of masculine or feminine characteristics. If the difference
between the masculine/feminine endorsement is high, then the person is categorized as
the gender-type with the highest score (i.e. masculine or feminine); if the difference is
low, they are categorized as androgynous.
The BSRI has been used extensively, and hence its reliability and validity have been
subject to considerable scrutiny yielding both confirmatory (e.g. Ballardreisch & Elton,
1992; Blanchardfields, Suhrerroussel, & Hertzog, 1994; Campbell, Gillaspy, &
Thompson, 1997) and more sceptical results (Hoffman & Borders, 2001; Wilcox &
Francis, 1997). Much of the criticism of the scale has stemmed from disagreement with
the principle of androgyny. As this has slowly become accepted, critics have become
more interested in the internal validity of the measure (e.g. Wilcox & Francis, 1997).
A factor-analytical study by Campbell et al. (1997) found the short form of the inventory
(i.e. the one used in this study) was more reliable than the longer form. Chung (1995)
examined the construct validity of the BSRI for straight and gay men and found that it
was valid for use with both groups of men. More recently attention has focused on
assessing if and how the roles of men and women have changed since 1974 and whether
Bem’s measure is still valid. Holt (1998) repeated Bem’s method to categorize various
adjectives and concluded that it is still a valid measure for assessing present gender roles
although frequent revalidations would be warranted in the future.
Research question and hypotheses
The study reported in this paper investigated the degree to which interpersonal
distances between men and women are a function of the three different gender
variables: sex, gender role, and sexuality. The potential for dissociation between sex,
sexuality, and gender role affords an opportunity to test hypotheses regarding the
1 Following the growing convention in social psychology, we use the term ‘straight’, rather than heterosexual, by way of contrast
to gay/lesbian/bisexual.

---

Copyright © The British Psychological Society
Reproduction in any form (including the internet) is prohibited without prior permission from the Society
584
David Uzzell and Nathalie Horne
relative viability of biological versus social learning explanations of IPD behaviour.
Previous research has examined straight people’s choice of IPD when interacting with a
known lesbian or gay man (Gentry, 1988; Mooney, Cohn, & Swift, 1993), but there
appears to be no research concerning the IPDs that lesbians and gay men use between
themselves, i.e. research on IPD and sexuality and also gender role. Lesbians, gay men,
and bisexuals are ideal groups to investigate in this area due to their increased frequency
of gender role reversal. Consequently, a sample was selected for this study with the
specific intention of measuring the impact of sexual preferences on interpersonal
distances.
The principal hypotheses investigated in this paper are:
(1)
H1 – gender role (masculine/feminine) will account for more of the variance in IPD
than biological sex (male/female).
(2)
H2 – the IPD of masculine/masculine dyads will be greater than the IPD of
feminine/feminine dyads.
(3)
H3 – the IPD of intermediate/intermediate (i.e. middle of the sexuality range)
dyads is expected to be different from the masculine/masculine dyads or
feminine/feminine dyads.
Gender role reversal is also of interest in this research since the design is partly based
on the assumption that this phenomenon will be less common in straight than in
lesbian, gay, and bisexual people.
Methodology
In the introduction, we identified what we believe are essential features of a more
efficacious methodology for understanding the effect of biological sex and gender on
interpersonal distance. Balancing ecological validity with accuracy of measurement is
probably the most problematic issue. Therefore, we argue that an accurate and
ecologically valid measure needs to minimize intrusion, permit direct observation, and
enable accuracy of measurement of IPD as a subconscious behaviour.
Digital video-recording IPD method (DiVRID)
Since the majority of research on interpersonal distances was undertaken in the 1970s
and 1980s, there have been considerable technological advances in digital technology,
especially in the use of digital video-recording equipment, which permits highly
accurate measurements of recorded distances. In this experiment, the observation of
subjects was structured, conducted in a laboratory setting, and employed high quality
digital video-recording equipment so that consent could be obtained and interactions
could be controlled and accurately recorded. Two measures were taken; the IPD of each
dyad, and a pre-tested measure of each participant’s gender role.
The laboratory was laid out as in Figure 1. Three equally spread measurement mats
known to the participants as ‘Stations A, B, and C’ were laid on the floor. These were an
integral part of the means by which accurate measurements of interpersonal distances
could be recorded. Each mat was 0.89 £ 1.60 m and made from cream coloured,
durable plastic sheeting marked in black with a grid of 1 £ 100 squares. Distances were
subsequently converted into centimetres for data analysis and presentation. A video-
camera on a tripod and at a height of 2.5 meters was placed three meters from the near
edge of each mat. The cameras were positioned so that only the measurement mats filled
the frame.

---

Copyright © The British Psychological Society
Reproduction in any form (including the internet) is prohibited without prior permission from the Society
Sex, sexuality and gender role and interpersonal distance
585
Figure 1. Illustration of the layout of the laboratory.
In order to test the influence of biological sex, sexuality, and gender role on IPD, the
study focused on dyads as the unit of analysis. This was achieved by conducting the
experiment in groups of six participants so that each participant interacted with each of
the other five participants in the group of varying gender characteristics. Therefore,
each participant was associated with five IPD measurements for interactions with five
different people. Twelve sessions with six participants in each yielded 180 dyad trials in
total.
The grid on the mat enabled highly accurate measures of IPD between dyads to be
recorded using a high quality digital video-cameras. Measurements were taken from the
mid-point between Participant A’s feet to the mid-point between Participant B’s feet as
illustrated in Figure 2. The videotapes were analysed using a 15-second interval
observational recording method.
Participants
The authors were very alert to the various sampling problems through heterosexual bias
as identified by Herek, Kimmel, Amaro, and Melton (1991). These included not
confusing sexual orientation with gender identity and gender role conformity, as well as
perhaps one of the principal problems in this area of work, ensuring that the sample is
representative. Lesbians, gays, and bisexuals, notwithstanding cultural and societal
changes over the last decade or so, are still stigmatized groups and so, as with any social
stigma, an unknown proportion of the population will not disclose their sexuality.
Consequently, it is difficult to know the size of the population from which one is hoping
to draw a representative sample.
The sample comprised 72 university students from the University of Sussex, England.
A university sample was seen as appropriate, and not just a matter of convenience.
Figure 2. Illustration of measurement method between Participant 1’s and Participant 2’s feet.

---

Copyright © The British Psychological Society
Reproduction in any form (including the internet) is prohibited without prior permission from the Society
586
David Uzzell and Nathalie Horne
First, a university offered a better opportunity for recruiting participants of varying
sexualities and therefore increased the chance of achieving more diversity in gender role
reversal. Second, it was thought that students at university may be more accustomed to
the non-heterosexual sexualities and therefore would be more at ease, and that
homophobia may be less influential as a confounding variable. The sample was drawn by
two means. Participants were randomly recruited by approaching students outside the
Students’ Union on the university campus. The University of Sussex has an active
Lesbian, Gay, Bisexual, and Trans-sexual Society (LGBT), and members were
approached, with the cooperation of its officers, by e-mail. LGBT members were
asked to participate in a study as a favour for ‘a friend’ so that they did not believe they
were selected because of their membership of the society or their sexuality. All other
participants were recruited simply by asking them as they walked past the Students’
Union. Students were only approached if they appeared to be approximately 18 to 26
years of age, of white ethnicity, and UK nationals.
Of the 72 participants, 34 (47%) were male and 38 (53%) female, with ages ranging
from 17 to 31 with a mean age of 21.2 years. Just under 90% of the female sample
(N ¼ 33) said they were straight, four were lesbians, and one was bisexual. Of the male
sample, 25 (74%) said that they were straight, 8 were gay, and 1 was bisexual. The
majority of the participants (89%) were from the UK and all participants were white
adults.
Measures
The BSRI (Bem, 1974) comprises 60 adjectives against which participants indicate on a
7-point Likert scale how well each item describes them (1 never/almost never true; 2
usually not true; 3 sometimes but infrequently true; 4 occasionally true; 5 often true;
6 usually true; 7 always or almost always true). Of the adjectives, 20 constitute the
masculine measure and a second 20 constitute the feminine measure. The remaining 20
adjectives are neutral and are not included in the final calculations. The resulting BSRI
score identifies the participant on a scale from very masculine to very feminine, with
androgynous as a narrow band of scores in the middle.
In the same questionnaire, there were four demographic questions on the
participants’ sex, age, nationality, and sexuality (straight/gay/bisexual). They were also
asked whether they knew any of the other participants within the experimental session.
Procedure
Each session of the experiment consisted of six randomly selected participants of mixed
sexes, sexualities, and gender roles. Participants were taken to the laboratory situated
within the Students’ Union. The laboratory was approximately 11.5 £ 10.5 m with a
ceiling height of 4 m. Initially, participants were led to understand that their
participation was required for a psychology research project concerning memory and
their ability to remember facts about other people to whom they had spoken.
On entering the laboratory, participants were briefed that the purpose of the study
was to talk to each of the other five people, and ask prompt questions displayed on a
poster at each end of the room (these were positioned to encourage the interacting
participants to orientate themselves at 908 to the camera). The questions were their
‘partner’s’ name, age, mother’s name, father’s name, what time they woke up, what they
had for breakfast, and brothers’ and sisters’ names and ages. They were asked to

---

Copyright © The British Psychological Society
Reproduction in any form (including the internet) is prohibited without prior permission from the Society
Sex, sexuality and gender role and interpersonal distance
587
remember what they could about each person in the 2 minutes they were allocated,
using whatever mechanisms they felt necessary. They were also informed that they
would be asked to complete a short questionnaire afterwards. They were told that the
experiment would be video-recorded for future analysis, should it be required.
Participants were given a badge randomly numbered between 1 and 6 and asked to
go to their first station as instructed by the experimenter. They were reminded that they
should remain at each station for the full 2 minutes, rehearsing the information if
necessary. After the first interaction trial, the experimenter indicated that they should
stop talking and move on to the next interaction, which was indicated by the researcher.
This was repeated until all participants had spoken to each other (for order of dyads,
see Table 1).
Table 1. Order of dyad interaction
Station
A
B
C
Participants
1/2
3/4
5/6
3/5
2/6
1/4
1/3
4/6
2/5
1/6
5/4
2/3
2/4
3/6
1/5
Participants were then given a questionnaire marked with their badge number and
session number, and were reminded of the confidentiality of the data collected. Once all
the questionnaires were completed, participants were debriefed as to the true purpose
of the study and given the opportunity to ask questions; they were also given an
opportunity to withdraw their contribution to the experiment should they wish. No
participants indicated that they wished to withdraw their participation. Each group was
also asked if they had any suspicions that their IPD was being measured; none had.
Results
The analysis examined IPD and its relationship with three sex-related variables (sex,
sexuality, and BSRI score). Importantly, there were no significant differences in IPD
between dyads that knew each other and those that did not, tð178Þ ¼ 21:62, p ¼ ns;
thus, it was not necessary to account for this factor in any further analysis. Descriptive
statistics of the IPD measurements according to the categorization of the two variables
can be found in Table 2.
Categorization of BSRI scores
The BSRI scores are typically categorized into three groups. Bem’s scoring requires that
the androgynous group is defined ‘on the basis of their Femininity Minus Masculinity
Difference Score, with small difference scores indicating androgyny and large different
scores indicating either sex typing or cross sex typing’ (Bem, 1993, p. 120). The BSRI
defines androgynous people as having an index score of between 2 1 and þ 1, while
masculine and feminine individuals fall outside this range (i.e. masculine , 2 1; feminine
. 1). We accept Bem’s argument that an individual can have attributes of both
masculinity and femininity and therefore masculinity-femininity is not a continuous

---

Copyright © The British Psychological Society
Reproduction in any form (including the internet) is prohibited without prior permission from the Society
588
David Uzzell and Nathalie Horne
a
y
-
G
7.92
Lesbian
60.57
55.58
72.39
o
y
y-Ga
–
53.40
11.74
33.50
73.66
59.69
59.69
59.69
54.02
10.95
32.08
74.63
Ga
F
emale-Heter
F
emale
o
F
eminine-Feminine
Heter
o
o
minine
5.08
54.79
11.39
29.36
76.20
Male-Heter
63.55
11.22
35.08
93.35
female
60.20
53.19
67.16
o
F
emale
Bisexual-Heter
Intermediate-Fe
Heter
o
oup
gr
7.73
1.91
5.26
62.25
47.63
73.99
Male-Bisexual
64.69
62.23
66.68
emale
F
65.88
59.69
71.91
variable
o
Intermediate-
Intermediate
Lesbian-Heter
each
Heter
within
emale
Male-
o
–
dyads
56.33
11.18
32.08
76.20
eminine
61.57
10.38
32.08
93.35
Lesbian
71.30
10.23
51.92
82.07
76.20
76.20
76.20
Masculine-
F
for
emale-F
F
Heter
Lesbian-Lesbian
y
ments
o
emale
Male-Ga
measure
62.80
12.20
29.36
93.35
64.90
11.61
41.91
86.06
o
63.30
12.01
45.42
84.61
y-Heter
female
57.11
15.68
29.36
87.78
Masculine-
IPD
Male-F
Intermediate
Ga
Heter
the
o
for
statistics
9.66
9.75
8.88
–
63.41
45.42
84.61
70.00
53.04
87.78
Male-Heter
Male
63.56
46.35
76.20
66.19
66.19
66.19
e
y-Bisexual
Male-Male
o
Ga
Masculine-Masculine
Heter
Descriptiv
2.
able
T
Gender
Mean
SD
Minimum
Maximum
BSRI
Mean
SD
Minimum
Maximum
Sexuality
Mean
SD
Minimum
Maximum
Sexuality
Cont
Mean
SD
Minimum
Maximum

---