Content uploaded by Anthony F Bogaert
Author content
All content in this area was uploaded by Anthony F Bogaert on Mar 07, 2015
Content may be subject to copyright.
Archives of Sexual Behavior, Vol. 28, No. 3, 1999
The Relation Between Sexual Orientation
and Penile Size
Anthony F. Bogaert, Ph.D.,
1
,
3
and Scott Hershberger, Ph.D.
2
The relation between sexual orientation and penile dimensions in a large sample
of men was studied. Subjects were 5122 men interviewed by the Kinsey Institute
for Research in Sex, Gender, and Reproduction from 1938 to 1963. They were di-
chotomously cl assi ed as either homosexual (n
=
935) or heterosexual (n
=
4187).
Penile dimensions were assessed using ve measures of penile length and circum-
ference from K insey’s original protocol. On all ve measures, homosexual men
reported larger penises than did heterosexual men. Explanations for these differ-
ences are discussed, including the possibility that these ndings provide additional
evidence that variations in prenatal hormonal levels (or other biological mecha-
nisms affecting reproductive structures) affect sexual orientation development.
KEY WORDS:
sexual orientation; penile dimensions; hormones; homosexuality; Kinsey; men.
INTRODUCTION
Biologica l factors, such as prenatal hormonal conditions (e.g., testosterone
levels), have been argued to be important in the development of sexual orientation
(e.g., Ellis and Ames, 1987). However, one criticism of these approaches (e.g.,
Byne and Parsons, 1993) is that there is little evidence that homosexual men and
women differ from heterosexual men and women in genital morphology, a plausible
consequence if biological factors affecting reproduction, such as prenatal hormonal
levels, underlie sexual orientation. It is clear, for example, that alterations in typical
levels of prenatal hormones can affect genital size and morphology, as evidenced by
girls with congenital adrenal hyperplasia (CAH) who have masculinized genitalia
1
Health Studies, Brock University, St. Catharines, Canada L2S 3A1; e-mail: tbogaert@spartan.ac.
brocku.ca.
2
California State University, Department of Psychology, 1250 Bell ower Boulevard, Long Beach,
California 90840; e-mail: scotth@csulb.edu.
3
To whom correspondence should be addressed.
213
0004-0002/99/0600-0213$16.00/0
C
°
1999 Plenum Publishing Corporation
214 Bogaert and Hershberger
as a re sult of exposure to higher-than-typical levels of androgens prenatally or by
boys who have demasculinized genitalia as a result of androgen insensitivity syn-
drome (e.g., Conte and Grumbach, 1995). However, although women affected
by C AH have been found to have elevated rates of same-sex attraction (e.g.,
Zucker et al., 1996), these women make up only a very small percentage of the
people with same-sex attraction.
A number of factors may ac count for why heterosexuals and the vast majority
of homosexuals do not seem to differ in genital morphology. One explanation con-
cerns the timing of prenatal hormonal uctuations responsible for alterations in
sexual orientation on the one hand and the timing of possible alterations in genital
morphology on the other. For example , there is evidence that the beginning of
genital differentiation (e.g., development of the pe nis) oc curs earlier in gestation
(i.e., at 8 weeks) than the beginning of the differentiation of brain structures puta-
tively relevant for sexual identity and sexual orientation (Ellis and Ames, 1987).
Thus, a small alteration in testosterone levels in one phase of pregnancy may not
alter genital development in any detectable way, but it may have effects on brain
structures (e.g., hypothalamus) relevant for sexual orientation.
A related explanation is that alterations in typical prenatal hormones do not
always affect genital growth. For example, two recent studies of CAH boys, who
have been exposed to higher-than-typical levels of prenatal and prepubertal andro-
gens, did not nd different lengths of penises in these boys relative to controls,
although the samples of the CAH boys were small (n
=
9 and n
=
12) and 2 of the
boys in one CAH sample had micropenis (Levy and Husmann, 1996; Sutherland
et al., 1996). In addition, researchers administering higher-than-typical dosages of
prenatal testosterone did not induce more-than-typical growth in a small sample
of human fetal penile specimens (Baskin et al., 1997). This recent research sug-
gests that individual differences in adult phallic size may be largely the result of
differences in the inherent genetic structure for organoge nesis when stimulated by
appropriate hormonal conditions. However, these research st udies had, as men-
tioned, small samples, and it is clear that the mechanisms, including hormonal
ones, responsible for the growth (and the cessation of growth) of the genitalia are
not fully understood (Baskin et al., 1997).
A third explanation is that this issue has not been adequately studied in large
samples. A large sample may be needed to detect small variations in genital size
that have occurred because of alterations in t ypical levels of prenatal hormones
or other relevant growth and development mechanisms. One early and relatively
unknown study using a reasonably large sample did nd that homosexuals differ,
on average, from heterosexuals in genital morphology (s ee Nedoma and Freund,
1961). These investigators found that homosexual men (n
=
126) had larger penises
than a comparison sample of heterosexual men (n
=
86). More data are needed
to establish the reliability of this nding, however. A very large data set that
contains information on sexual orientation and genital size in men is the ori ginal
Sexual Orientation 215
Kinsey sample (Gebhard and Johnson, 1979; Kinsey et al., 1948). With over 5000
relevant cases, these data provide a rare opportunity to conduct powerful tests of
the relation between sexual orientation and genital size. These tests are the focus
of the present study. Speci cally, we present analyses on some of the original
Kinsey data to investigate whether homosexual and heterosexual men differ in
penile size. If homosexual men differ from heterosexual men, then this nding may
provide additional evidence that variations in prenatal hormonal levels (or other
biological mechanism s affecting reproductive structures) affect sexual orientation
development.
METHOD
Probands
From 1938 to 1963, 17,502 case histories were recorded by the Kinsey Insti-
tute for Sex Research using the i nterview schedule devised by Alfred C. Ki nsey
(Gebhard and Johnson, 1979). These data are currently stored in several les. The
les containing adult (i.e., 18 or over) White and non-White males with no convic-
tions for felonies or misdemeanors (other than traf c violati ons) comprise 6013
cases.
Sexual Orientation
Sexual orie ntation was classi ed according to the following criteria. Men
were classi ed as homosexual if they reported “extensive” homosexual experience,
de ned by Gebhard and Johnson (1979) as more than 20 male sexual partners or
more than 50 homosexual experiences (with one or more partners). Men were
classi ed as heterosexual if they met two criteria: (i) t hey reported either “no” or
“rare” homosexual experi ence, the latter de ned by Gebhard and Johnson (1979)
as 1 male sexual partner and/or 1–5 homosexual experiences, and (ii) they did not
respond that they experienced “much” or “some” sexual arousal to questions about
sexual arousal from s eeing and thinking of other males.
Using these criteria, 891 cases could not be classi ed as either heterosexual
or homosexual. Of the remaining 5122 cases, 935 were classi ed as homosexual
and 4187 were cl assi ed as heterosexual.
Penile Size Measures
In Kinsey’s original interview protocol, ve measures of penis s ize were
included: estimated erect penis length; measured accid penis length; measured
216 Bogaert and Hershberger
erect penis length; measured accid penis circumference; and measured erect penis
circumference. Penis length was estimated or measured along the top of the penis
from the belly to the tip. For penis circumference, the men were told to measure
at the point of maximum circumference. During the interviews, Kinsey and his
colleagues often slowly slid a nger along a standard ruler with the numerals
not visible and t old the men to indicate when the length of their erect penis had
been reached. For the measured sizes, the men were given speci c instructions
on how to measure their penises, and precision of measurement was stressed.
These measured sizes were performed after the interview, and the participants
mailed their measurements to the Kinsey Institute using standard response cards
and preaddressed stamped envelopes (Gebhard and Johnson, 1979). The Kinsey
researchers recorded the sizes to the nearest quarter inch. The data derived from
these measures have provided normative information on penis size and have been
shown to support ndings from other samples (Jamison and Gebhard, 1988).
Additional Demographics
The two groups were also assessed on a number of demographic variables,
including age, educat ion level, race/ethnicity, and parental socioeconomic status
(SES). Educational level could ra nge from 1
=
1st grade or illiterate to 20
=
Ph.D.,
MD., or LL.D. Parental SES refers to the nancial status of the proband’s parents
when he was 14 to 17 years of age. It was coded on an 8-point scale from 1 (extreme
poverty) to 8 ( extreme wealth). Parental SES is preferred to proband’s education as
a measure of social class because many of the probands were still in school (i.e.,
college and university) when interviewed and their current educa tional level would
not re e ct their ultimate education level. Finally, the two groups we re assessed on
height and weight. Height and we ight were reported by the respondent, not by
the interviewer. Height was recorded to the nearest inch. Current weight was not
recorded in the computer les in exact pounds, but instead in 10-lb. ranges, where,
for exa mple, 11
=
110–120 lb., 12
=
120–129 lb., and so on.
RESULTS
Similar percentages of homosexual and heterosexual men had valid responses
(i.e., nonmissing data) for the estimated penis length measure a sked during the
interview (n
=
813 or 87% vs. n
=
3147, or 82%, respectively). In addition,
the two groups had similar pe rcentages of those who returned the cards with the
measured penis si zes, either assessed as a ratio of returned cards to total possible
respondents (both groups at approximately 44%) or as a rat io of returned ca rds to
those who had a valid response for the estimated penis length measure during the
interview (both groups at approximately 54%).
Sexual Orientation 217
Table I.
Comparisons of Homosexual and Heterosexual Men on Penile Dimensions
Homosexuals Heterosexuals
(n
=
813)
a
(n
=
3417)
a
Variable ¯x
b
SD ¯x
b
SD t
c
p
Estimated erect size 6.32 0.95 5.99 0.91 9.16
<
0.0001
Measured accid penis size 4.10 0.87 3.87 0.71 5.06
<
0.0001
Measured erect penis size 6.46 0.82 6.14 0.74 7.29
<
0.0001
Measured circumference 3.84 0.65 3.70 0.66 3.92
<
0.0001
of accid penis
Measured circumference 4.95 0.69 4.80 0.72 3.86
<
0.0001
of erect penis
a
Sample size varies on some measures because of missing cases.
b
Measurements are inches.
c
t-test based on separate variance estimates when variances differ.
Table II.
Comparisons of Homosexual and Heterosexual Men on Demographic Variables
Homosexuals Heterosexuals
(n
=
813)
a
(n
=
3417)
a
Variable ¯x SD ¯x SD t p
Age 30.77 10.65 30.45 11.27 0.77 ns
Parental SES
b
4.75 1.59 4.73 1.40 0.23 ns
Education 14.17 3.81 15.83 3.52 11.03
<
0.0001
Height
c
69.50 2.82 69.70 2.76 1.93 0.06
Weight
d
15.35 2.17 15.94 2.24 6.90
<
0.0001
a
Sample size varies for some variables because of missing cases.
b
Parental SES varies from 1
=
(extreme poverty) to 8
=
(extreme wealth).
c
Height is in inches.
d
Weight is 10-lb, ranges, where, for example, 15
=
150–159 lbs.
On all ve measures, the homosexual men reported larger penises than did
the heterosexual men (all ps
<
0
.
0001, two-tailed). Thus, differences occurred on
both estimated and measured penis size (Table I).
4
The homosexual and heterosexual men did not differ in the proportion of
non-Whites comprising the groups (i.e ., both 8% non-White), nor did they differ
in age or parental SES (Table II). However, the two groups did differ in weight and
marginally in height, where the homosexual men were smaller than the heterosex-
ual men (see also Blanchard and Bogaert, 1996). The two groups also differed in
educational level, although, as mentioned, this measure is probably less meaningful
4
We also tested whether differences occurred between homosexual and heterosexual men using a more
restricted de nition of heterosexuality, where “heterosexual” men were those who did not have any
(even “rare”) same-sex sexual contact. As mentioned, Kinsey’s de nition of “rare” meant 1 partner
and/or 1–5 same-sex sexual experiences. All penile differences remained signi cant (all ps
<
0
.
0001).
218 Bogaert and Hershberger
than parental SES. To eval uate any possible confounding effects these three vari-
ables might have in contributing to the penis size differences between the two
groups, we entered education, height, and weight, along with one of the ve penile
measures, into ve logistic regression a nalyses to predict sexual orientation. All
ve penile measures remained signi cant (all ps
<
0
.
0001).
DISCUSSION
In the largest sample of its kind in the world, homosexual men were found
to report signi cantly larger penises (in both length and circumference) than did
heterosexual men.
5
Moreover, the effects were not due to possible confounding
factors, such as height, weight, or education. Indeed, the fact that these differences
occurred, even though homosexual men may be physically smaller than heterosex-
ual men (Blanchard and Bogaert, 1996; Bogaert and Blanchard, 1996a) suggests
these differences are robust, given the small but signi cant relationship between
penis size and physical size (e.g., r
=
0
.
15 between erect penis length and height
in these data; see also Siminoski and Bain, 1993).
Alterati ons of typical levels of prenatal hormones in homosexual men may
account for these ndings in one of two ways. First, a lower l evel of prenatal
masculiniz ing hormones (e.g., testosterone), particularly at critical periods, may
reduce the loss of androgen receptors in the developing penis which allows it to
grow to a larger size. Such down-regulation of androgen receptors has been argued
to occur in the cessation of genital growth in rodents (Takane et al., 1990), although
to date there is little evidence of such mechanisms in humans (Baskin et al., 1997;
Levy a nd Husmann, 1996; Sutherland et al., 1996).
Second, the pattern of prenatal horm onal uctuations found to occur in preg-
nant female rats under conditions of stress may be rel evant (see Ward, 1984, for
a review). Prenatal stress in pregnant female rats reduces the level of prenatal
androgens, which alters the size of the sexually dimorphic nucleus of the preop-
tic a rea and sex-typical behavior in male offspring. However, it is also clear that
prior to this eventual and permanent decline in testosterone in pregnant female
rats, there is a signi cant initial surge of this hormone (see, for example, Ward
and Weisz, 1980, p. 328). If similar prenatal hormonal patterns occur i n pregnant
human mothers under stress (e.g., D¨orner et al., 1980; but also see Bailey et al.,
1991) or other conditions, then one might expec t homosexual men to have, along
with more “feminine” characteristics, which re ect the late but permanent decline
of pre natal testosterone, a number of more “masculine” characteristics, which re-
ects the early testosterone surge. One of these masculine characteristics may be
larger genitalia.
5
In a footnote, Tripp (1975, p. 83) made reference to the fact that the Kinsey data show differences in
penis size between homosexual and heterosexual men, but he evidently never published any analysis
to support his assertion.
Sexual Orientation 219
Some research indicates that homosexual men may have more incidence
of non-right-handedness (e.g., Lindesay, 1987), a characteristic more likely to
occur in men than in women. This handedeness pattern has been interpreted to be
consistent with Ward’s nding of an early surge of testosterone prior to an eventual
and permanent decline of this hormone in pregnant rats under conditions of stress
(Geschwind and Galaburda, 1985; James, 1989). However, it should be noted
that the sexual orientation/handedness relationship is inconsistent (see Zucker
and Bradley, 1995), with a number of l arge-scale studies showing no elevated
rates of non-right-handedness in homosexual men relative to heterosexual men
(e.g., Bogaert and Blanchard, 1996b). It should also be noted that Ward and her
colleagues have not detected evidence of altered genital size (e.g., smaller or larger
penises) in the male offspring of the stressed pregnant rat s (e.g., Ward, 1992).
Alternative biological explanations for the penis size difference c oncern other
(i.e., nonprenatal androgenic) growth and development mechanisms relevant to re-
production/sex differentiation. For example, it is clear that penile growth occurs
in various phases of fetal, childhood, and adolescent life and that other hormones
aside from androgens, such as growth and thyroid hormone, are partially responsi-
ble for human penile growth (Baskin et al., 1997). If homosexual and heterosexual
men differ on one or more of these other biological mechanisms, then these factors,
alone or in combination with prenatal androgens, may account for the difference
between homosexual and heterosexual men in penis size. It is of note that growth
and devel opment characteristics, such as physical size and age of puberty onset
(e.g., Blanchard and Bogaert, 1996; Bogaert and Blanchard, 1996a) and elements
of the pit uitary-hypothalamic-gonadal axis, in particular sites of the hypothalamus
(LeVay, 1991; Swaab and Hofman, 1990), may differ between homosexual and
heterosexual men.
Researc h on possi ble genetic differences between homosexual and heterosex-
ual men (e.g., Bailey a nd Pillard, 1991; Hamer et al., 1993) is also of note, given
recent research suggesting that individual differences in adult phallic size may be
largely the result of differences in the inherent genetic structure for organogenesis
under appropriate hormonal conditions (Baskin et al., 1997; Levy and Husmann,
1996; Sutherland et al., 1996). If so, the possible genetic differences between ho-
mosexual and heterosexual men ma y affect, not surprisingly, a number of elements
of t he reproductive/sex differentiation system.
Another explanation for these differences concerns systematic error from
reporting bias. Although random or unsystematic error from, for example, self-
report measures is less likely (and not more likely) to lead to reliable/signi cant
diffe rences, there is a possibility of syst ematic error in these data. In particular,
homosexual me n may be more likely than heterosexual men to exaggerate the size
of their penises to conform to an ideal standard of sexual attractiveness. There is
some evidence, for example, that gay men may be more prone t han het erosexual
men to body image dissatisfaction (e.g., Siever, 1994). However, there are a number
reasons why reporting bias may be an unlikely explanation for these differences.
220 Bogaert and Hershberger
First, the differences in penis size exist across different measures and methods of
measurements (i.e., estimating, measuring, circumference, and length), suggesting
reliabili ty of measurement. Second, it is clear that, despite the fact t hat gay men
may be more concerned than heterosexual men with some elements of their body,
such as the need for a thin physique (e.g., Siever, 1994), heterosexual men are in
fact concerned—even preoccupied—about the size of their penises and often de-
sire a larger one (e.g., Zilbergeld, 1978). Third, heterosexual men are often much
more conforming than homosexual men to stereotypically “masculine” gender
roles (e.g., Bailey and Zucker, 1995). Fourth, there is evidence tha t homosexual
men may be less likely than heterosexual men to falsify/exaggerate personality
and attitudinal information on questionnaires where social desirability is a n issue
(Freedman, 1975). Finally, these ndings replicate a smaller but independent study
by Nedoma and Freund (1961) in which a physician conducted the penile mea-
surements. Given that this issue remains unresolved, however, collecting additional
data to assess the possibility of self-report bias is worthwhile.
REFERENCE S
Bailey, J. M., and Pillard, R. C. (1991). A genetic study of male sexual orientation. Arch. Gen. Psychiat.
48: 1089–1096.
Bailey, J. M., and Zucker, K. J. (1995). Childhood sex-typed behavior and sexual orientation: A
conceptual analysis and quantitative review. Dev. Psychol. 31: 43–55.
Bailey, J. M., Willerman, L., and Parks, C. (1991). A test of the maternal stress theory of human male
homosexuality. Arch. Sex. Behav. 20: 277–294.
Baskin, L. S., Sutherland, R. S., DiSandro, M. J., Hayward, S. W., Lipschutz, J., and Cunha, G.
(1997). The effect of testosterone on androgen receptors and human penile growth. J. Urol. 158:
1113–1118.
Blanchard, R., and Bogaert, A. F. (1996). Biodemographic comparisons of homosexual and heterosex-
ual men in the Kinsey interview data. Arch. Sex. Behav. 25: 551–579.
Bogaert, A. F., and Blanchard, R. (1996a). Physical development and sexual orientation in men:
Height, weight, and onset of puberty differences. Pers. Indiv. Diff. 21: 77–84.
Bogaert, A. F., and Blanchard, R. (1996b). Handedness in homosexual and heterosexual men in the
Kinsey interview data. Arch. Sex. Behav. 25: 373–378.
Byne, W., and Parsons, B. (1993). Human sexual orientation: The biologic theories reappraised. Arch.
Gen. Psychiat. 50: 228–239.
Conte, F., and Grumbach, M. (1995). Pathogenesis, classi cation, diagnosis, and treatment of anomalies
of sex. In Degroot, L. J. (ed.), Endocrinology, W. B. Saunders, Philadelphia.
D¨orner, G., Geier, T., Ahrens, L., Krell, L, M¨unx, G., Sieler, H., Kitter, E., and M¨uller, H. (1980).
Prenatal st ress as possible aetiogenic factor of homosexuality in human males. Endokrinologie
75: 365–368.
Ellis, L., and Ames, M. A. (1987). Neurohormonal functioning and sexual orientation: A theory of
homosexuality-heterosexuality. Psychol. Bull. 101: 233–258.
Freedman, M. (1975). Homosexuals may be healthier than straights. Psychol. Today 8: 28–32.
Gebhard, P. H., and Johnson, A. B. (1979). The Kinsey Data: Marginal Tabulations of the 1938–1963
Interviews Conducted by the Institute for Sex Research, W. B. Saunders, Philadelphia.
Geschwind, N., and Galaburda, A. M. (1985). Cerebral lateralization. Biological mechanisms, associ-
ations, and pathology: II. A hypothesis and a program of research. Arch. Neurol. 42: 634– 654.
Hamer, D. H., Hu, S., Magnuson, V. L., Hu, N., and Pattatucci, A. (1993). A linkage of DNA markers
on the X chromosomes and male sexual orientation. Science 261: 321–327.
Sexual Orientation 221
James, W. J. (1989). Foetal testosterone levels, homosexuality, and handedness: A research proposal
for jointly testing Geschwind’s and D¨orner’s hypothesis. J. Theoret. Biol. 136: 177–180.
Jamison, P., and Gebhard, P. (1988). Penis size increase between accid and erect states: An analysis
of the Kinsey data. J. Sex Res. 24: 177–183.
Kinsey, A. C., Pomeroy, W. B., and Martin, C. E. (1948). Sexual Behavior in the Human Male, W. B.
Saunders, Philadelphia.
LeVay, S. (1991). A difference in hypothalamic structure between heterosexual and homosexual men.
Science 253: 1034–1037.
Levy, J. B., and Husmann, D. (1996). Congenital adrenal hyperplasia: Is there an effect on penile
growth? J. Urol. 156: 780–782.
Lindesay, J. (1987). Laterality shift in homosexual men. Neuropsychologia 25: 965–969.
Nedoma, K., and Freund, K. (1961). Somatosexuln´
õ
n´alezy u homosexu´aln´
õ
ch mu
Ï
zu
±
[Somatosexual
ndings in homosexual men]. Ceskoslovenska Psychiatre 57: 100–103.
Siever, M. D. (1994). Sexual orientation and gender as factors in socioculturally acquired vulnerability
to body dissatisfaction and eating disorders. J. Consult. Clin. Psychol. 62: 252–260.
Siminoski, K., and Bain, J. (1993). The relationship among height, penile length, and foot size. Ann.
Sex Res. 6: 231–235.
Sutherland, R. S., Kogan, B. A., Baskin, L. S., Mevorach, R. A., Conte, F., Kaplan, S., and Grumbach,
M. M. (1996). The effect of prepubertal androgen exposure on adult penile length. J. Urol. 156:
783–787.
Swaab, D. F., and Hofman, M. A. (1990). An enlarged suprachiasmatic nucleus in homosexual men.
Brain Res. 537: 141–148.
Takane, K. K., George, F., and Wilson, J. D. (1990). Androgen receptor of rat penis is down-regulated
by androgen. Am. J. Physiol. 258: E46.
Tripp, C. A. (1975). The Homosexual Matrix, McGraw-Hill, New York.
Ward, I. L. (1984). The prenatal stress syndrome: Current status. Psychoneuroendocrinology 9: 3–11.
Ward, I. L . (1992). Sexual behavior: The product of perinatal hormonal and prepubertal social factors.
In Gerail, A. A., Moltz, H., and Ward, I. L. (eds.), Handbook of Behavioral Neurobiology: Vol. 11.
Sexual Differentiation, Plenum Press, New York.
Ward, I. L., and Weisz, J. (1980). Maternal stress alters plasma testosterone in fetal males. Science 207:
328–329.
Zilbergeld, B. (1978). Male Sexuality, Little, Brown, Boston.
Zucker, K. J., and Bradley, S. J. (1995). Gender Identity Disorder and Psychosexual Problems in
Children and Adolescents, Guilford, New York.
Zucker, K. J., Bradley, S. J., Oliver, G., Blake, J., Fleming, S., and Hood, J. (1996). Psychosexual
development of women with congenital adrenal hyperplasia. Horm. Behav. 30: 300–318.
