Content uploaded by Terri Fisher
Author content
All content in this area was uploaded by Terri Fisher
Content may be subject to copyright.
Arch Sex Behav (2007) 36:89–100
DOI 10.1007/s10508-006-9094-7
ORIGINAL PAPER
Sex of Experimenter and Social Norm Effects on Reports
of Sexual Behavior in Young Men and Women
Terri D. Fisher
Received: 13 July 2005 / Revised: 11 April 2006 and 21 July 2006 / Accepted: 28 July 2006 / Published online: 23 December 2006
C
Springer Science+Business Media, Inc. 2006
Abstract Past studies indicate that men generally report
having had more sexual experience and sexual partners than
women, as well as an earlier age at first intercourse. At least
some of these findings may partially reflect different re-
sponses to certain contextual variables in research. College
students (266 men and 463 women) were asked to anony-
mously report their sexual attitudes and behavior after read-
ing one of three fictitious statements about research findings
regarding gender differences in sexuality. Some past find-
ings were replicated, with men reporting somewhat more
sexual experience and more permissive sexual attitudes than
women. However, women reported a significantly younger
age at first intercourse than did men. While there was no sig-
nificant sex difference for total number of sexual partners,
there was a significant interaction. With female research as-
sistants (but not with male assistants), men reported more
sexual partners when they were told that women are now
more sexually permissive than men. This finding appeared
to be largely a function of the men who scored higher on mea-
sures of hypermasculinity and ambivalent sexism. Women’s
reports were not significantly affected by the wording of the
cover sheet, regardless of the sex of the research assistant.
Even in this anonymous survey, the sex of the experimenter
and the nature of the statement about research findings had
an impact on the sex differences that were found. In light of
these results, some previous conclusions about male-female
T. D. Fisher (
)
Department of Psychology,
The Ohio State University at Mansfield,
1680 University Drive,
Mansfield, Ohio 44906
e-mail: fisher.16@osu.edu
differencesin sexual behaviormay need to be examined more
closely.
Keywords Sexual behavior
.
Sex differences
.
Experimenter effects
.
Social norms
.
Social dominance
Introduction
Past research has consistently indicated that men report
more sexual experience and sexual partners than do women
(Baumeister, Catanese, & Vohs, 2001). Various explanations
have been proposed to account for this difference, with two
major approaches being the evolutionary psychology and the
social role perspectives. These gender differences in sexual
behavior are viewed by evolutionary psychologists as largely
innate and biologically-based due to divergent evolved mat-
ing strategies on the part of men and women as a function
of their differential parental investment (Buss, 1998;Buss
& Schmitt, 1993). In contrast, social role theorists (Eagly
& Wood, 1991, 1999) have argued that gender differences,
including those pertaining to sexuality, are generally the re-
sult of the different social roles prescribed for women and
men. Recently, Wood and Eagly (2002) proposed a bioso-
cial explanation of gender differences that takes into account
biological differences and constraints as well as socializa-
tion and gender roles. These theories differ in the degree to
which they acknowledge the role of social influences and
contextual factors on gender differences, with evolutionary
psychologists downplaying the role of socialization factors
and social role or biosocial theorists stressing the function
of the environment in influencing the ways in which sexual
behavior is expressed by men and by women.
The research which has established this gender difference
has been based on self-reported sexual behavior by means
Springer
90 Arch Sex Behav (2007) 36:89–100
of interviews and surveys. These gender differences have
generally been taken at face value
1
even though researchers
have long been aware that social desirability concerns and
degree of anonymity can effect a participant’s willingness to
admit to sensitive behaviors (Catania, 1999; Durant, Carey,
& Schroder, 2002; Meston, Heiman, Trapnell, & Paulhus,
1998; Schroder, Carey, & Vanable, 2003). It is possible that
being asked to report on past sexual activity poses different
perceived levels of threats and benefits on the part of the
two sexes, resulting in differences in willingness to report
on the full extent of one’s sexual behavior. Several studies
have demonstrated that gender differences in certain areas of
sexual behavior can be enhanced or diminished as a function
of context effects, such as the structure of the testing situa-
tion or the wording of questions (Alexander & Fisher, 2003;
Catania et al., 1996; Catania, Binson, Peterson, & Canchola,
1997; Durant et al., 2002; Tourangeau & Smith, 1996).
Sex of experimenter
Researchers usually take great care to try to eliminate any
influence on behavior that might be a function of the research
situation. One contextual variable that has not generally been
thought to be of great concern is the sex of the person admin-
istering the questionnaire in an anonymous, self-report sur-
vey (Johnson & Moore, 1993; Tourangeau & Smith, 1996).
Indeed, Winer, Makowski, Alpert, and Collins (1988)have
demonstrated that college student responses on an anony-
mous sex survey were generally not affected by the nature
of the researcher even if the questionnaire had been admin-
istered by a religious leader. Therefore, sex researchers have
not been particularly concerned with the sex of the ques-
tionnaire administrator, although Ehrhardt (1997) indicated
that, with respect to face to face interviews, researchers are
reaching a consensus that female interviewers are able to
elicit more responses from participants than male interview-
ers and therefore “may be preferred to men” (p. 361).
Perhaps the potential effect of the research assistant has
been too easily dismissed. Sex of the experimenter has been
clearly demonstrated to have an impact on behaviors as di-
verse as expression of aggression (Shope, Hedrick, & Geen,
1978), reported pain (Levine & DeSimone, 1991), extent
of cardiovascular reaction to mental stress (Larkin, Ciano-
Federoff,& Hammel, 1998), relativeleft frontal cerebral acti-
vation during EEG recording in defensive individuals (Kline,
1
The one exception has been in the area of reports of number of sexual
partners where it is apparent that the partner counts provided by het-
erosexual men and women should be comparable (McConaghy, 1999).
Various explanations have been posed to account for the mismatch in
reported partners, including dishonesty, inaccurate recall, the exclusion
of prostitutes from the data sample, sex differences in the methods
used to determine the number of partners, potential sampling bias, and
possible response bias (Wiederman, 1997).
Blackhart, & Joiner, 2002), and discussions between married
couples (Weisfeld & Stack, 2002). Leary et al. (1994) deter-
mined by means of the use of daily interaction records that
college students tend to be most concerned about convey-
ing an impression of being likable, competent, ethical, and
attractive when interacting with those of the other sex. In
addition to this general sex of experimenter effect, when a
research assistant is a college student peer, participants may
feel even greater potential threat to their reputational stand-
ing. In support of this idea, Mensch and Kandel (1988) found
that adolescents and young adults underreported their illicit
drug use if they were being interviewed by someone with
whom they were familiar or who they thought they might be
encountering in the future.
Sex of participant
Deaux and Major (1987) developed an interactive model of
gender-related behavior in which they proposed that gender-
role related behaviors are influenced by expectations and
situational cues. Applying the research cited above to this
interactive model, it would appear that, for heterosexual col-
lege students, the presence of a student research assistant
of the other sex in a sexuality study could easily activate a
gender-related schema, resulting in a bias in their reporting
of sexual activity.
Women generally feel more threatened than men when
asked to report sensitive behavior (Durant et al., 2002) and
still perceive a double-standard with regard to the acceptabil-
ity of permissive sexual behavior (Crawford & Popp, 2003;
Hyde & Durik, 2000; Tolman, 2002). Therefore, they might
be less likely to report casual sexual encounters, fearing a
potential reputational cost. In a study in which participants
were asked to report on their sexual behavior in one of three
testing conditions (while attached to a polygraph, alone in
a private room, or in a situation with less apparent privacy),
women were more likely than men to provide different re-
ports of sexual activity as a function of the different testing
conditions (Alexander & Fisher, 2003). Women reported the
lowest number of sexual partners, the least involvement with
impersonal sexuality (masturbation and viewing of erotica),
and the oldest age at first intercourse when in the least private
situation. This suggests that women feel that they have more
to lose than do men when it comes to reporting the full extent
of their sexual behavior. Women who are most accepting of
traditional roles and expectations might be particularly af-
fected by the contextual factors surrounding queries about
their sexual behavior and under certain circumstances be
more hesitant to reveal sexual behavior that is inconsistent
with the traditional female gender role.
On the other hand, men might have a motive to exag-
gerate their sexual experience. It has been argued by some
that sexuality is the primary arena in which men exercise
Springer
Arch Sex Behav (2007) 36:89–100 91
their dominance (Pratto, 1996;Segal,2001; Wood & Eagly,
2002). Dallos and Dallos (1997) suggested that a belief in
innate differences in male and female behavior is a form
of ideological power which serves to constrain the freedom
of women and is reflected in the sexual domain. Men who
feel that they have less than the typical amount of sexual
experience may feel that their status or masculine identity
is threatened and might be more likely to exaggerate their
expertise. Indeed, men whose masculinity has recently been
questioned appear to overcompensate by reporting more ex-
treme masculine behaviors than do men whose masculinity
has not been threatened (Willer, 2005).
Social dominance theorists have discussed the various
behavioral and attitudinal mechanisms by which those who
are motivated by the need to dominate try to maintain their
power (Sidanius & Pratto, 1999). Danso and Esses (2001)
found that white participants performed better on a test of
cognitive abilities when the research assistant was African-
American than when the assistant was white, an effect that
was magnified among those who scored higher on a measure
of social dominance orientation. They concluded that this
was due to a need to maintain social dominance, explaining
that “when these White test takers are put in a situation that
makes salient the perception that Blacks are making progress
in the academic domain .... they are especially motivated to
perform well in order to prove their superiority and maintain
their group dominance” (p. 163).
Being told that women are now more sexually experi-
enced than men while having close proximity to a female
research assistant (and therefore a female authority figure)
could result in men experiencing a threat to their status, lead-
ing to a level of defensiveness that might not occur with a
male research assistant. Because men who are sexist or hy-
permasculine are particularly sensitive to power threats from
women, they might be especially prone to social dominance
motives (Sidanius & Pratto, 1999).
The present study
In the present study, the focus was not initially on exper-
imenter effects but rather on the potential impact of so-
cial norms on students’ reports of sexual activities. In order
to further explore the role that social norms might play in
reporting sexual behavior, college students’ perceptions of
normative behavior were manipulated by exposing the par-
ticipants to one of three brief statements summarizing the
fictitious results of recent research on sexual behavior and
attitudes. The participants were then asked to report on their
own sexual behaviors and attitudes by means of an anony-
mous questionnaire. Because this was a questionnaire study
with little interaction between the research assistants and the
participants, only female research assistants were initially
used.
When the data from the study were analyzed, it was clear
that the males reported larger numbers of sexual partners
when they were told that women are now more permissive
and sexually experienced then men. In order to eliminate the
possibility of a sex of experimenter effect, the data collection
period was extended using male research assistants in place
of the female assistants who had been used previously. Thus,
this research became a study of experimenter effects in addi-
tion to an examination of the manipulation of social norms.
If this was an effect of the sex of the research assistant, the
apparent male “enhancement” of sexual experience that had
been seen in the first half of the study would be absent from
data collected by male research assistants.
Hypotheses
A social role explanation of gender differences in sexual
behavior (Eagly & Wood, 1991, 1999) would lead to the
prediction that such differences should be diminished when
participants are exposed to social norms which indicate that
women are no longer less sexually experienced and permis-
sive than men. Although this idea has not been empirically
tested with regard to sexual behavior, Grossman and Wood
(1993) found that when emotional responsiveness was pro-
moted as normative and desirable for both sexes, no signif-
icant gender differences were found in self-reports of emo-
tional state. It was, therefore, hypothesized that if permissive
sexual behavior was presented as normative for women, fe-
male participants would be more willing to admit such behav-
ior and male participants might report less of it. Specifically,
it was hypothesized that when participants were presented
with statements that indicated that men were no longer more
sexually permissive and experienced than women, women
would report a greater number of sexual partners, an earlier
age at first intercourse, and greater sexual experience than
when presented with the traditional social norm that men are
more experienced. Further, it was predicted that men who
were higher in hypermasculine traits or ambivalent sexism
would report more sexual partners, greater sexual experi-
ence, and earlier age at first intercourse when they were told
that women are now more sexually experienced than men,
but only when they had a female research assistant.
With regard to female participants, it was predicted that
those women who were higher in hyperfemininity or am-
bivalent sexism would be more affected by the statements on
the cover sheet such that they would report less permissive
sexual behavior when told that men are more sexually active
than women than in the other two conditions.
Sexual attitudes were not expected to be particularly af-
fected by the social norm condition. Alexander and Fisher
(2003) found that while different testing situations influenced
reporting of behavior, there were no significant differences
in reported sexual attitudes as a function of the condition. A
Springer
92 Arch Sex Behav (2007) 36:89–100
main effect of sex of participant was anticipated, with men
reporting more permissive sexual attitudes than women.
Method
Participants
Participants were students attending an open admissions re-
gional campus of a large midwestern university who selected
this study in order to partially fulfill a research requirement
for a general psychology class. To reduce heterogeneity in
sexual experience, only self-identified never-married hetero-
sexual students between the ages of 18 and 25 were included
in the sample to be analyzed, yielding a total of 266 men
and 463 women. This skewed sex ratio reflects the student
enrollment at the campus. The first 307 (111 male and 196
female) participants (M age = 19.1) were tested by female
research assistants and the next 422 participants (158 men
and 264 women) (M age = 19.2) were tested by male re-
search assistants. The percentage of men and women and the
mean age of the participants was equivalent for each portion
of the study.
The research assistants (three female and two male) were
undergraduate psychology majors in the same age range as
the participants. They were provided with a script from which
they were told not to deviate. There was very little interaction
between the research assistant and the participants. There
were no significant differences on the major variables as a
function of the individual research assistant.
Measures
The questionnaires administered to the participants had a
cover sheet that defined social norms in the form of one of
three statements written in boldface as part of a few “Fre-
quently Asked Questions” that were presented on the cover
sheet: “Recent research has indicated that males are more
sexually experienced and permissive than females and we
wanted to see if this is true at our campus” (MSEP), “Recent
research has indicated that females are more sexually expe-
rienced and permissive than males ...” (FSEP), or “Recent
research has indicated that there are now no differences in
the sexual behavior and permissiveness of males and females
...” (ND).
The 20-item version of the Marlowe–Crowne Social De-
sirability Scale (Strahan & Gerbasi, 1972) was used to mea-
sure participants’ tendency to choose responses that would
make them look like good people. Questions were answered
in a true-false format, and possible scores range from 0 to
20, with higher scores corresponding to more of a tendency
to give socially desirable responses. Sample items include
“I’m always willing to admit it when I make a mistake” and
“I sometimes feel resentful when I don’t get my way.” For
this sample, Cronbach’s alpha coefficient (α) was .72 and the
actual range of scores was 2 to 20.
The Cowart–Pollack Scale (Cowart-Steckler & Pollack,
1998) was used to measure extent of sexual experience. Par-
ticipants indicated in which of 30 sexual activities they had
engaged, ranging from light petting to various forms of sex-
ual intercourse. The original dichotomous response format
was modified to a 4-point scale reflecting the number of
times in which each activity was engaged, with possible to-
tal scores ranging from 0 to 90 (for this sample, α =.97 and
range = 0–90). Participants were also asked to indicate the
number of partners with whom they had engaged in sexual
intercourse and the age at which they had first engaged in
consensual sexual intercourse.
Attitudinal measures
Sexist beliefs were measured with the Glick and Fiske (1996)
Ambivalent Sexism Inventory, which consists of 22 items
measuring either benevolent sexism (“No matter how ac-
complished he is, a man is not truly complete as a person un-
less he has the love of a woman”) or hostile sexism (“Many
women are actually seeking special favors, such as hiring
policies that favor them over men, under the guise of asking
for ‘equality’”). The respondent is asked to agree or disagree
with these statements by means of a five-point Likert scale,
with higher total scores indicating a greater degree of am-
bivalent sexism toward women (for this sample, α =.85 and
range = .4–4.7).
The Hypermasculinity Inventory (Mosher, 1988) was ad-
ministered to male participants. This instrument measures
a macho-type personality pattern and consists of 30 forced-
choice items comprising three subscales (callous sexual at-
titudes toward women, violence as manly, and danger as ex-
citing). Higher scores reflect greater hypermasculinity (for
this sample, α =.83 and range = 0–27). Sample item pairs
include “I only want to have sex with women who are in
total agreement” or “I never feel bad about my tactics when I
have sex” and “When I’m bored I watch TV or read a book”
or “When I’m bored I look for excitement.”
Female participants completed the Hyperfemininity Scale
(Murnen & Byrne,
1991) which measures the degree to
which a woman internalizes a stereotypically feminine gen-
der role for herself. Hyperfeminine women view their pri-
mary value in terms of their sexual appeal and their attrac-
tiveness to men. This scale consists of 26 pairs of items
from which the participant must choose one as best reflect-
ing her attitudes about various aspects of traditional femi-
ninity. Higher scores indicate greater degrees of hyperfem-
ininity. Sample items include “I never use my sexuality to
manipulate men” or “I like to have a man ‘wrapped around
my finger.’” The possible scores range from 0 to 30, and
Springer
Arch Sex Behav (2007) 36:89–100 93
internal consistency was reported to be .76. For this sample,
the scores ranged from 0 to 21 and α =.49.
Two measures of sexual attitudes were used. The Attitudes
Toward Sexuality Scale (ATSS) (Fisher & Hall, 1988), a mea-
sure of general sexual attitudes, consists of 13 items with a
5-point Likert response scale yielding potential scores of 13
to 65 (for this sample α =.83 and range = 16–63). Sample
items include “Sexual intercourse for unmarried young peo-
ple is acceptable without affection existing if both partners
agree” and “Our government should try harder to prevent the
distribution of pornography.”
The Sexual Opinion Survey (SOS) (Fisher, Byrne, &
White, 1983) was used to measure the dimension of
erotophobia-erotophilia. This instrument consists of 21 state-
ments with which participants express their degree of agree-
ment by means of a 7-point Likert response scale. Possible
scores range from 0 to 126, with higher scores indicating
more positive emotional responses to sexuality (for this sam-
ple, α =.82 and range = 0–120). Sample items include “If
people thought I was interested in oral sex, I would be em-
barrassed” and “Engaging in group sex is an entertaining
idea.”
The Sociosexual Orientation Inventory (SOI) (Simpson &
Gangestad, 1991) measures the degree to which an individual
has an unrestricted sociosexual orientation (i.e., willingness
of an individual to participate in casual sexual encounters,
to have multiple sexual partners, and to have sex sooner in
a relationship). It consists of 11 attitudinal (“Sex without
love is ok”) and behavioral (“How often do you fantasize
about having sex with someone other than your current dating
partner?”) items, including a question about the number of
partners with whom one has had sex on only one occasion.
The typical range of scores for college students is 10–250
(Simpson, 1998; α =.97 and range = 10–217 for the current
sample).
Just prior to the measures of sexual experience and socio-
sexuality, a statement was printed in boldface which stressed
the importance of honest responses and restated the social
norm manipulation in the form of the research motive de-
scribed on the cover sheet. A manipulation check in the
form of posttest questions designed to determine whether
participants correctly recalled the wording of the cover sheet
was administered to most participants. Participants were pre-
sented with all three of the statements and were asked to
identify which one had appeared on their cover sheet. In ad-
dition, participants were asked to indicate on a 7-point scale
the statement that best reflected their opinion regarding dif-
ferences in the sexual behavior and permissiveness of males
and females, with 1 corresponding to “males are much more
sexually experienced and permissive than females,” 4 cor-
responding to “no difference,” and 7 corresponding to “fe-
males are much more sexually experienced and permissive
than males.” The manipulation check was not implemented
until after data collection had already begun, and occasion-
ally the research assistant neglected to administer the posttest
or the participant declined to answer it. Therefore, of the 729
participants whose data were used in the study, only 535
completed the posttest.
Procedure
Groups of between 3 and 15 participants reported to a large
classroom and were asked to leave as much space as possible
(at least three desks) between each student. An undergrad-
uate research assistant distributed the questionnaire packets
with the three different versions presented in random order.
Participants were asked to read carefully the cover sheet be-
fore beginning the survey. Participants were told that when
they had completed their questionnaires they were to place
them in a locked wooden box at the front of the room and
then complete the posttest form. Data collection sessions
were overseen by female research assistants for the first year
of data collection and male research assistants thereafter.
The questionnaires were not coded to provide information
regarding the number of participants in each testing ses-
sion, so it is unknown whether there was any effect of group
size.
Results
Manipulation check
Of the 535 participants who responded to the posttest manip-
ulation check, 75.3% of the male participants and 78.8% of
the female participants correctly recalled the wording of the
cover sheet, χ
2
(1) =.87, ns. Equivalent numbers of partici-
pants in each condition correctly recalled the wording of the
cover sheet, χ
2
(2) =4.46, ns. When the definition of correct
memory was expanded to include either of the nontraditional
wordings (no gender differences or women more permissive)
in the FSEP and ND conditions, the percentage remembering
correctly increased to 80.1% of the men and 85.1% of the
women, χ
2
(1) =2.18, ns.
A one-way analysis of variance (ANOVA) was performed
in order to determine if the participants’ opinions regard-
ing differences in the sexual behavior and permissiveness of
males and females were affected by the wording of the cover
sheet. There was a significant difference, F(2,553) =7.82,
p < .001, η
2
p
= .03, with participants in the FSEP (M =3.5,
SD =1.1) and ND (M =3.4, SD =1.0) conditions indicat-
ing opinions significantly closer to the “no difference” mid-
point of 4 than participants in the MSEP condition (M =3.1,
SD =1.1), thus indicating that the cover statement had an
influence on participants’ perceptions of social norms.
Springer
94 Arch Sex Behav (2007) 36:89–100
Table 1 Pearson correlation
coefficients for social
desirability responding and
sexual measures for male and
female participants
Males Females
Scale Nr Nr Fisher’s r to z
Sexual experience 263 −.12 448 −.31
∗∗∗
2.52
∗∗
Age at first coitus 190 .03 325 .10 −.75
Sexual partners 190 .11 326 −.04 1.56
Sexual partners (log) 190 .08 325 −.06 1.52
SOI 157 −.04 277 −.12
∗
.80
SOI (log) 157 −.07 277 −.16
∗∗
.88
SOS 269 −.25
∗∗∗
454 −.27
∗∗∗
.29
ATSS 269 −.09 454 −.20
∗∗∗
1.40
Note. SOI: Sociosexual Orientation Inventory; SOS: Sexual Opinion Survey; ATSS: Attitudes Toward
Sexuality Scale.
∗
p < .05.
∗∗
p < .01.
∗∗∗
p < .001.
Social desirability
Table 1 shows the correlations between social desirability
responding and the sexual measures for men and for women.
Only scores on the Sexual Opinion Survey were significantly
correlated with social desirability for men, with participants
having higher social desirability scores being more likely to
have lower SOS scores, hence indicating less erotophilia. For
women, social desirability responding was significantly cor-
related with extent of sexual experience as well as each of the
attitudinal measures. However, a series of Fisher’s r to z trans-
formations indicated that only the correlation with sexual ex-
perience was significantly different for men and women, with
social desirability responding being more highly correlated
with reported sexual experience for women than for men.
For ANOVAs on variables for which there was a significant
correlation with social desirability responding, social desir-
ability scores were included in the analysis as a covariate.
Gender differences in sexual behavior and attitudes
Means and SDs for the major measures in each testing condi-
tionareshowninTables 2–5 as a function of sex of participant
and sex of research assistant.
Sexual partners
Participants who had experienced sexual intercourse were
asked to report the number of partners with whom they had
engaged in intercourse. This question was left blank by those
who had never had sexual intercourse as well as by some who
chose not to indicate their number of partners, yielding a total
of 213 responses from those with a female research assistant
and 299 responses from those with a male research assistant,
with a range from 1 to 200. In order to reduce the effect
of outliers, a winsorizing procedure was used such that the
score of any student reporting more than 20 sexual partners
in their lifetime (8 men and 9 women) was set at a ceiling
score of 20 (the 95th percentile), yielding a total of 9 men
and 11 women with a score of 20. Even with this adjustment,
the data were still positively skewed, so a log transformation
was performed.
A 2 (Sex of Participant) × 2 (Sex of Research Assistant)
× 3 (Norm Condition) ANOVA revealed a significant three-
way interaction, F(2,502) =3.97, p =.02, η
2
p
= .02. When
the research assistant was female, there was a significant
Sex × Norm Condition simple interaction, F(2,502) =3.38,
p =.035, η
2
p
= .013. An analysis of the simple simple main
effects revealed that in the FSEP condition, men reported sig-
nificantly higher numbers of sexual partners than did women,
F(1,502) =7.11, p =.008, η
2
p
= .014, but the gender differ-
ences were not significant in either of the other two con-
ditions (see Tables 2 and 3). With male research assistants,
there were no significant effects for this variable.
For men with female research assistants, there was a sig-
nificant simple main effect of condition, F(2,502) =3.44,
p =.03, η
2
p
= .014. Pairwise comparisons with a Least
Significant Difference (LSD) test indicated a signifi-
cant (p =.011) difference between the MSEP (M =.32,
SD =.31) and the FSEP (M =.61, SD =.49) conditions and
a marginally significant (p =.071) difference between the
MSEP condition and the ND (M=.41, SD =.33) condition.
This simple main effect was not significant for women or for
men with male research assistants.
In order to explore whether or not traditional gender role
attitudes were moderators of these findings, two sets of
analyses were done using median splits to divide the par-
ticipants into groups that were either higher or lower on
these traits. The first set of analyses involved a median split
using the hypermasculinity and the hyperfemininity mea-
sures. A 2 (Sex of Participant) × 2 (Sex of Research As-
sistant) × 3 (Norm Condition) ANOVA indicated that there
Springer
Arch Sex Behav (2007) 36:89–100 95
Table 2 Means and SDs on behavior and attitudes measures for men with female research assistants as a function of testing condition
MSEP FSEP ND
Range MSDNMSDNMSDN
Age at first coitus 12–22 17.1 1.0 25 16.3 2.0 24 16.8 1.3 27
Sexual partners 0–20 2.7 2.2 25 7.1 7.2 25 3.4 2.8 27
Sexual partners (log) 0–1.3 0.3 0.3 25 0.6 0.5 25 0.4 0.3 27
Sexual experience 0–120 51.8 27.1 37 53.9 28.7 35 50.4 27.7 39
SOI 10–217 59.2 23.8 24 72.8 46.7 23 62.0 25.7 25
SOI (log) 1.8–5.4 4.0 0.5 24 4.1 0.8 23 4.0 0.5 25
SOS 0–120 73.4 17.2 37 74.6 14.4 35 70.6 22.0 39
ATSS 16–63 42.0 10.0 37 44.3 8.2 35 41.2 11.6 39
Hypermasculinity 0–27 11.1 6.3 36 11.4 4.9 32 11.0 5.2 37
Ambivalent sexism 4–4.7 3.0 0.6 36 3.0 0.6 35 3.0 0.5 39
Social desirability 2–20 10.0 3.4 37 9.4 3.0 35 9.6 3.4 39
Note. SOI: Sociosexual Orientation Inventory; SOS: Sexual Opinion Survey; ATSS: Attitudes Toward Sexuality Scale.
was a significant three-way interaction for those participants
who were higher in hypermasculinity or hyperfemininity,
F(2,256) =4.40, p =.013, η
2
p
= .03, but not for those lower
in these traits, F(2,202) =.61, ns.
Further analyses, reported in Table 6, indicated that there
was a significant interaction between testing condition and
sex of participant only among those who were above the me-
dian in hypermasculinity or hyperfemininity and who also
encountered a female research assistant, F(2,256) =3.89,
p =.02, η
2
p
= .03. Among the male participants in this
group, there was a significant effect of testing condition,
F(2,256) =3.2, p =.041, η
2
p
= .03. Pairwise comparisons
with an LSD test revealed that the transformed mean for
the FSEP condition (M =.80, SD =.49) was significantly
higher than that for the MSEP condition (M=.40, SD =.33)
and the ND condition (M =.50, SD =.32).
Similarly, after splitting the group into those who were
higher or lower with regard to ambivalent sexism, a 2
(Sex of Participant) × 2 (Sex of Research Assistant) × 3
(Norm Condition) ANOVA indicated that there was a sig-
nificant three-way interaction only for those participants
who were higher in ambivalent sexism, F(2,252) =5.06,
p =.007, η
2
p
= .04 but not for those lower on this trait,
F(2,232) =.46, ns. Further analysis (as shown in Table 7)
indicated that there a significant interaction between testing
condition and sex of participant only among those who were
above the median in ambivalent sexism and who encoun-
tered a female research assistant, F(2,252) =4.79, p =.009,
η
2
p
= .04. However, pairwise comparisons failed to reveal
a significant effect of testing condition for any subgroup.
Nonetheless, the men who were high in ambivalent sex-
ism with a female research assistant showed the same pat-
tern of means discussed above (MSEP =.33; FSEP =.66;
ND =.41), F(2,252) =2.57, p =.079, η
2
p
= .02.
Sexual experience
A 2 (Sex of Participant) × 2 (Sex of Researcher) × 3(Norm
Condition) ANOVA with social desirability as a covariate
revealed only a significant main effect of sex of participant.
Table 3 Means and SDs on behavior and attitudes measures for women with female research assistants as a function
of testing condition
MSEP FSEP ND
MSDNMSDNMSDN
Age at first coitus 16.3 1.1 46 16.3 1.5 47 16.2 1.2 45
Sexual partners 3.4 3.8 45 3.4 4.2 47 5.0 5.3 45
Sexual partners (log) 0.3 0.4 45 0.3 0.4 47 0.5 0.4 45
Sexual experience 44.6 26.9 62 48.5 26.1 65 47.5 25.9 66
SOI 38.4 20.4 41 40.9 23.7 45 44.8 28.6 41
SOI (log) 3.5 0.6 41 3.5 0.7 45 3.6 0.6 41
SOS 57.7 25.6 62 56.5 25.5 65 58.8 22.0 65
ATSS 42.0 9.6 64 40.8 9.4 65 39.7 10.0 67
Hyperfemininity 8.1 4.0 62 8.8 3.6 63 8.4 3.2 66
Ambivalent sexism 2.4 0.7 63 2.5 0.7 64 2.5 0.7 65
Social desirability 10.4 3.5 64 9.2 3.8 65 9.8 3.7 67
Note. SOI: Sociosexual Orientation Inventory; SOS: Sexual Opinion Survey; ATSS: Attitudes Toward Sexuality Scale.
Springer
96 Arch Sex Behav (2007) 36:89–100
Table 4 Means and SDs on behavior and attitudes measures for men with male research assistants as a function of
testing condition
MSEP FSEP ND
MSDNMSDNMSDN
Age at first coitus 16.5 1.7 30 16.8 1.6 54 16.5 1.6 30
Sexual partners 4.6 4.6 30 4.2 5.1 55 4.4 4.9 28
Sexual partners (log) 0.5 0.4 30 0.4 0.4 55 0.4 0.4 28
Sexual experience 53.5 28.2 40 58.4 22.5 71 58.8 22.3 40
SOI 79.1 33.2 23 61.9 32.7 40 70.5 48.3 22
SOI (log) 4.3 0.5 23 4.0 0.5 40 4.0 0.7 22
SOS 74.2 19.7 42 71.6 16.2 73 77.4 17.2 42
ATSS 44.1 10.6 42 42.0 9.9 73 45.4 9.0 42
Hypermasculinity 11.8 5.8 37 11.7 5.3 63 12.3 6.5 38
Ambivalent sexism 2.9 0.6 42 3.1 0.7 73 2.8 0.7 42
Social desirability 8.8 3.9 42 10.1 2.9 73 8.2 2.9 42
Note. SOI: Sociosexual Orientation Inventory; SOS: Sexual Opinion Survey; ATSS: Attitudes Toward Sexuality Scale.
Males (M =55.0, SD =25.6) reported more sexual experi-
ence than females (M =48.1, SD =25.9), F(1,695) =9.30,
p =.002, η
2
p
= .01.
Age at first intercourse
Of those participants who reported having engaged in sexual
intercourse, their age at first coitus ranged from 12 to 22. A
2 (Sex of Participant) × 2 (Sex of Research Assistant) × 3
(Norm Condition) ANOVA revealed only a significant main
effect of sex of participant. Women reported a younger age at
first intercourse (M =16.3, SD =1.5) than men (M =16.7,
SD =1.6), F(1,502) =5.54, p =.019, η
2
p
= .01.
Attitudinal measures
A log transformation was done for the SOI data because it
did not approximate a normal distribution. For the SOI, the
SOS, and the ATSS, 2 (Sex of Participant) × 2 (Sex of Re-
searcher) × 3 (Norm Condition) ANOVAs were calculated,
with social desirability scores used as covariates. There were
no significant effects of condition or sex of research assistant
and no significant interactions. For all three measures, there
was a significant main effect of sex of participant, with males
reporting significantly more permissive sexual attitudes than
females (see Table 8).
Discussion
This study began as an examination of the effects of social
norms on reports of sexual behavior. The wording of the
cover sheet was intended to provide three distinctive social
norms regarding gender differences in sexual behavior and
was predicted to affect participants’ reports of sexual be-
havior. The study was then extended to include sex of the
research assistant as a contextual variable. For most sexual
behaviors and attitudes measured, neither sex of researcher
Table 5 Means and SDs on behavior and attitudes measures for women with male research assistants as a function of
testing condition
MSEP FSEP ND
MSDNMSDNMSDN
Age at first coitus 16.6 1.5 68 16.3 1.8 59 16.2 1.8 59
Sexual partners 3.9 4.6 69 4.4 4.2 59 4.1 3.7 59
Sexual partners (log) 0.4 0.4 69 0.5 0.4 59 0.5 0.4 59
Sexual experience 51.2 24.3 88 50.0 25.9 80 45.4 26.9 86
SOI 41.4 30.4 56 46.0 30.2 44 44.1 31.9 50
SOI (log) 3.5 0.7 56 3.6 0.6 44 3.6 0.6 50
SOS 58.5 26.2 91 55.9 25.1 82 57.8 25.5 88
ATSS 41.5 9.3 92 41.0 9.3 81 40.8 9.3 88
Hyperfemininity 9.1 3.9 84 8.9 3.6 77 9.5 3.8 80
Ambivalent sexism 2.5 0.8 92 2.5 0.7 80 2.6 0.7 88
Social desirability 9.0 3.2 92 9.1 3.5 82 10.3 3.3 89
Note. SOI: Sociosexual Orientation Inventory; SOS: Sexual Opinion Survey; ATSS: Attitudes Toward Sexuality Scale.
Springer
Arch Sex Behav (2007) 36:89–100 97
Table 6 Means and SDs on sexual partner measures for men as a function of hypermasculinity, sex of research assistant,
and testing condition
MSEP FSEP ND
MSDN MSDN MSDN F η
2
p
High Hypermasculinity/Female Research Assistant
Sexual partners 3.3 2.7 10 9.9 7.8 13 4.0 2.9 15
Sexual partners (log) 0.4 0.3 10 0.8 0.5 13 0.5 0.3 15 3.2
∗
.03
High Hypermasculinity/Male Research Assistant
Sexual partners 4.5 5.1 13 5.5 6.5 28 6.1 5.9 16
Sexual partners (log) 0.5 0.4 13 0.5 0.5 28 0.6 0.5 16 0.3 .00
Low Hypermasculinity/Female Research Assistant
Sexual partners 2.4 1.7 14 4.4 5.8 10 2.7 2.5 12
Sexual partners (log) 0.3 0.3 14 0.4 0.4 10 0.3 0.3 12 0.4 .00
Low Hypermasculinity/Male Research Assistant
Sexual partners 5.0 4.6 14 2.6 2.1 20 2.0 1.6 10
Sexual partners (log) 0.5 0.4 14 0.3 0.3 20 0.2 0.3 10 2.3 .02
Note. According to Cohen (1988), when using η
2
p
as a measure of effect size, .01 is small, .06 is medium, and .14 is large.
Sexual partner range is 0–20; Sexual partner (log) range is 0–1.3.
∗
p < .05.
nor wording of the cover sheet had a significant effect. The
traditional gender differences in sexual experience and sex-
ual attitudes were found, with men reporting slightly higher
levels of sexual experience and much more permissive sex-
ual attitudes than women. On the other hand, women actually
indicated a younger age at first intercourse than did men.
However, for participants’ reports of number of sexual
partners, an interesting pattern was revealed. Results indi-
cated a significant three-way interaction between sex of par-
ticipant, sex of research assistant, and wording of the cover
sheet. Men who encountered a female research assistant and
whose cover sheets stated that women were now more sex-
ually permissive and experienced appear to have provided
inflated reports of numbers of sexual partners, possibly in a
competitive attempt to make sure that the men at their cam-
pus would still be considered more sexually experienced than
women. When the research assistant was a man, this apparent
reactance was not triggered.
This unexpected “enhancement” of male sexual experi-
ence could have been a defensive reaction to a perceived
feminized research environment stemming from a need for
social dominance or a desire to perpetuate hegemonic mas-
culinity (Connell, 2001). Further analyses in the present
study revealed that this was very likely the case because
the significant results appeared to be due to those men high
in hypermasculinity and ambivalent sexism. Men who were
not above the norm in these qualities did not respond in
the same way. The hypermasculine men were the ones who
Table 7 Means and SDs on sexual partner measures for men as a function of ambivalent sexism, sex of research assistant
and testing condition
MSEP FSEP ND
MSDN MSDN MSDN F η
2
p
High Sexism/Female Research Assistant
Sexual partners 3.0 3.0 9 7.9 7.4 15 3.4 2.9 15
Sexual Partners (log) 0.3 0.4 9 0.7 0.5 15 0.4 0.3 15 2.6 .02
High Sexism/Male Research Assistant
Sexual partners 2.9 2.0 13 4.5 5.7 33 4.7 4.2 10
Sexual partners (log) 0.4 0.3 13 0.4 0.4 33 0.5 0.3 10 0.6 .00
Low Sexism/Female Research Assistant
Sexual partners 2.6 1.7 15 6.0 7.2 10 3.4 2.8 13
Sexual partners (log) 0.3 0.3 15 0.5 0.5 10 0.4 0.3 13 0.8 .01
Low Sexism/Male Research Assistant
Sexual partners 5.9 5.5 17 3.9 4.2 22 4.3 5.4 18
Sexual partners (log) 0.6 0.4 17 0.4 0.4 22 0.4 0.4 18 1.6 .01
Note. According to Cohen (1988), when using η
2
p
as a measure of effect size, .01 is small, .06 is medium, and .14 is large.
Sexual partner range is 0–20; Sexual partner (log) range is 0–1.3.
Springer
98 Arch Sex Behav (2007) 36:89–100
Table 8 Means and SDs on attitudes measures for men and women
Males Females
Scale Range MSDNMSDNF p η
2
p
SOI 10–217 67.6 35.8 157 42.6 27.9 277 63.4 .00 .13
SOI (log) 1.8–5.4 4.1 0.6 157 3.6 0.6 277 66.3 .00 .14
SOS 0–120 73.3 17.7 268 57.7 25.0 453 81.7 .00 .10
ATSS 16–63 43.1 10.0 269 41.0 9.5 458 7.6 .01 .01
Note. SOI: Sociosexual Orientation Inventory; SOS: Sexual Opinion Survey; ATSS: Attitudes Toward Sexuality Scale;
According to Cohen (1988), when using η
2
p
as a measure of effect size, .01 is small, .06 is medium, and .14 is large.
seemed to react to the female presence in the study by over-
stating aspects of their sexual behavior. These findings are
reminiscent of the “backfire effect” found by Tourangeau
and Smith (1996) when participants exposed to restrictive
context questions reported more sexual partners than those
exposed to permissive context questions.
While it is possible that the high number of sexual partners
reported by men in the FSEP condition with female research
assistants was closer to the actual truth than those of men in
the other two situational contexts, it seems much more likely
that these reports were less accurate than those in the other
two conditions. It is difficult to come up with an explanation
(theoretical or otherwise) as to why being led to believe that
men are more sexually active and permissive than women
or that there are no gender differences in this realm would
lead the participants to underreport their sexual partners.
However, even if this unlikely scenario is what happened,
the results still provide evidence that reporting of sexual
partners can be influenced by social norms.
Why was the report of number of sexual partners most
influenced by the social norm context? There are several
possible reasons. First, research has indicated that there are
various means of calculating numbers of sexual partners,
including estimation, an actual count, a wild guess, and ap-
proximation (Brown & Sinclair, 1999; Wiederman, 1997).
Perhaps when faced with the motivation to maintain the tra-
ditional dominance of men in the sexual arena, hypermascu-
line men who relied on any strategy other than an actual tally
found it easier to “round up” more often than is typically
done. In addition, a popular conception, based on evolution-
ary psychology, is that men are naturally promiscuous (with
multiple sexual partners) whereas women are more prone to
monogamy. Perhaps upping the count of sexual partners was
seen as an easy way of maintaining the status quo. Finally, as
McConaghy (1999) has indicated, “the discrepancy in self-
reported number of sex partners of men and women is one of
the most troublesome examples in relation to the potential for
bias and unreliability in self-reported sexual experience...”
(p. 314). In today’s society, perhaps partner count is the most
commonly fudged aspect of sexual behavior reporting.
Contrary to expectations, it appears that the reports of men
were more influenced by the manipulation of social norms
than were those of women. The responses of the female
participants were not significantly influenced by the word-
ing of the cover sheet. Even an analysis of the subgroup of
women who were higher on hyperfemininity or ambivalent
sexism failed to yield a significant result. It must be noted that
the hyperfemininity measure had low internal consistency,
though that was certainly not a problem with the measure of
ambivalent sexism.
As expected, the impact of the social norm manipulation
did not extend to the various measures of sexual attitudes.
The endorsement of sexual attitudes appeared to be less de-
pendent on the vagaries of the situation and might, therefore,
provide more reliable data than the reporting of sexual be-
haviors. However, sexual attitudes are not great predictors of
sexual behavior, especially for women (Baumeister, 2000).
Nonetheless, it appears that women and men are further apart
with regard to their sexual attitudes than they are in their sex-
ual behavior.
Limitations
The most serious limitation of this study is due to the
serendipitous revelation of the effect of research assistant’s
sex. Sex of the research assistant was not initially an inde-
pendent variable and is thus confounded with time of data
collection. Therefore it is possible (though extremely un-
likely) that the reported results are due to a dramatic social
change that occurred in the time period immediately subse-
quent to that in which female research assistants were uti-
lized. Even if that were the case, however, it does not change
the reaction of the men in the first part of the study to being
told that women have surpassed them in the realm of sexual
permissiveness. It should also be noted that in most cases,
the effect sizes were small to moderate, with the exception
of the effect sizes for sex differences in the various sexual
attitude measures. This likely reflects a combination of fac-
tors, including weakness of the effect itself as well as the
unequal sample sizes. An additional limitation of the study
is its college student sample, although one could argue that
for this sort of study, this is precisely the sample needed,
since much of the literature on gender differences in sexual
behavior is based on research with this population.
Springer
Arch Sex Behav (2007) 36:89–100 99
Further research
Future research designed to provide a more complete expla-
nation of these findings is clearly needed. A study directly
designed to test the social dominance motivation behind the
apparent overinflation of some men’s reports of sexual be-
havior would indicate whether this specific motivation is be-
hind the distortion of reporting as is suggested by the present
study. A study in which anonymity appears less certain to
the participants might provoke even stronger effects of the
social norm manipulation as well as the sex of the exper-
imenter. It would also be informative if future researchers
actually interviewed the participants immediately after the
study in an attempt to ascertain the level of veracity contained
within reports of various sexual behaviors and participants’
explanations and justifications as to why they may have been
dishonest.
Implications
The present study has implications for research on gender
differences in sexual behavior. It is apparent that two rel-
atively minor manipulations (a statement of social norms
and sex of research assistant) resulted in notably different
reporting of number of sexual partners even though this was
an anonymous survey. This suggests that there are mecha-
nisms that go beyond simple self-protection or impression
formation with regard to how participants sometimes answer
certain questions related to their sexuality. Until investigators
find a means to fully control for influences that have differ-
ent effects on men and women, we will not be able to know
with certainty which gender differences in reported sexual
behaviors reflect true differences in behavior and which may
be, at least in part, a function of method of testing, social ex-
pectations, or social environment. If we do not know which
differences are real, we certainly cannot yet adequately ex-
plain their origins.
Acknowledgements Special thanks to Joey Billotte, Nathan Mollette,
Tasha Stumpf, Vicki Summers, and Carla Wilkinson, who served as
research assistants for this study. Michele Alexander, Dan Lehman,
and Paul Trapnell made helpful suggestions and Jim McNulty provided
valuable comments on an earlier draft of this paper.
References
Alexander, M. G., & Fisher, T. D. (2003). Truth and consequences: Us-
ing the bogus pipeline to examine sex differences in self-reported
sexuality. Journal of Sex Research, 40, 27–35.
Baumeister, R. F. (2000). Gender differences in erotic plasticity: The
female sex drive as socially flexible and responsive. Psychological
Bulletin, 126, 347–374.
Baumeister, R. F., Catanese, K. R., & Vohs, K. D. (2001). Is there a
gender difference in strength of sex drive? Theoretical views, con-
ceptual distinctions, and a review of relevant evidence. Personality
and Social Psychology Review, 5, 242–273.
Brown, N. R., & Sinclair, R. C. (1999). Estimating number of lifetime
sexual partners: Men and women do it differently. Journal of Sex
Research, 36, 292–297.
Buss, D. M. (1998). Sexual strategies theory: Historical origins and
current status. Journal of Sex Research, 35, 19–31.
Buss, D. M., & Schmitt, D. P. (1993). Sexual strategies theory: An
evolutionary perspective on human mating. Psychological Review,
100, 204–232.
Catania, J. A. (1999). A framework for conceptualizing reporting bias
and its antecedents in interviews assessing human sexuality. Jour-
nal of Sex Research, 36, 25–38.
Catania, J. A., Binson, D., Canchola, J., Pollack, L. M., Hauck, W.,
& Coates, T. J. (1996). Effects of interviewer gender, interviewer
choice, and item wording on responses to questions concerning
sexual behavior. Public Opinion Quarterly, 60, 345–375.
Catania, J. A., Binson, D., Peterson, J., & Canchola, J. (1997).
The effects of question wording, interviewer gender, and con-
trol on item response by African American respondents. In
J. Bancroft (Ed.), Researching sexual behavior: Methodological
issues (pp. 110–113). Bloomington: Indiana University Press.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences.
Hillsdale, NJ: Lawrence Erlbaum Associates.
Connell, R. W. (2001). The social organization of masculinity. In S.
M. Whitehead & F. J. Barrett (Eds.), The masculinities reader
(pp. 30–50). Malden, MA: Blackwell Publishers, Inc.
Cowart-Steckler, D., & Pollack, R. H. (1998).TheCowart-Pollack Scale
of Sexual Experience. In C. M. Davis, W. L. Yarber, R. Bauserman,
G. Schreer, & S. L. Davis (Eds.), Handbook of sexuality-related
measures (pp. 104–105). Thousand Oaks, CA: Sage Publications.
Crawford, M., & Popp, D. (2003). Sexual double standards: A review
and methodological critique of two decades of research. Journal
of Sex Research, 40, 13–26.
Dallos, S., & Dallos, R. (1997). Couples, sex and power: The politics
of desire. Philadelphia: Open University Press.
Danso, H. A., & Esses, V. M. (2001). Black experimenters and the
intellectual test performance of white participants: The tables
are turned. Journal of Experimental Social Psychology, 37, 158–
165.
Deaux, K., & Major, B. (1987). Putting gender into context: An interac-
tive model of gender-related behavior. Psychological Review, 94,
369–389.
Durant, L. E., Carey, M. P., & Schroder, K. E. E. (2002). Effects of
anonymity, gender, and erotophilia on the quality of data obtained
from self-reports of socially sensitive behaviors. Journal of Be-
havioral Medicine, 25,
439–467.
Eagly, A. H., & Wood, W. (1991). Explaining sex differences in so-
cial behavior: A meta-analytic perspective. Personality and Social
Psychology Bulletin, 17, 306–315.
Eagly, A. H., & Wood, W. (1999). The origins of sex differences in hu-
man behavior: Evolved dispositions versus social roles. American
Psychologist, 54, 408–423.
Ehrhardt, A. A. (1997). Gender. In J. Bancroft (Ed.), Researching sex-
ual behavior: Methodological issues (pp. 361–362). Bloomington:
Indiana University Press.
Fisher, T. D., & Hall, R. G. (1988). A scale for the comparison of sexual
attitudes of adolescents and their parents. Journal of Sex Research,
24, 90–100.
Fisher, W. A., Byrne, D., & White, L. A. (1983). Emotional barriers
to contraception. In D. Byrne & W. A. Fisher (Eds.), Adolescents,
sex, and contraception (pp. 207–239). Hillsdale, NJ: Lawrence
Erlbaum Associates.
Glick, P., & Fiske, S. T. (1996). The Ambivalent Sexism Inventory: Dif-
ferentiating hostile and benevolent sexism. Journal of Personality
and Social Psychology, 70, 491–512.
Springer
100 Arch Sex Behav (2007) 36:89–100
Grossman, M., & Wood, W. (1993). Sex differences in intensity of
emotional experience: A social role interpretation. Journal of Per-
sonality and Social Psychology, 65, 1010–1022.
Hyde, J. S., & Durik, A. M. (2000). Gender differences in erotic
plasticity—Evolutionary or sociocultural forces? Comment on
Baumeister (2000). Psychological Bulletin, 126, 375–379.
Johnson, T. P., & Moore, R. W. (1993). Gender interactions between
interviewer and survey respondents: Issues of pornography and
community standards. Sex Roles, 28, 1–19.
Kline, J. P., Blackhart, G. C., & Joiner, T. E. (2002). Sex, lie scales,
and electrode caps: An interpersonal context for defensiveness
and anterior electroencephalographic asymmetry. Personality and
Individual Differences, 33, 459–478.
Larkin, K. T., Ciano-Federoff, L. M., & Hammel, D. (1998). Effects
of gender of observer and fear of negative evaluation on cardio-
vascular reactivity to mental stress in college men. International
Journal of Psychophysiology, 29, 311–318.
Leary, M. R., Nezlek, J. B., Downs, D., Radford-Davenport, J., Martin,
J., & McMullen, A. (1994). Self-presentation in everyday interac-
tions: Effects of target familiarity and gender composition. Journal
of Personality and Social Psychology, 67, 664–673.
Levine, F. M., & DeSimone, L. L. (1991). The effects of experimenter
gender on pain reports in male and female subjects. Pain, 44,
69–72.
McConaghy, N. (1999). Unresolved issues in scientific sexology.
Archives of Sexual Behavior, 28, 285–318.
Mensch, B. S., & Kandel, D. B. (1988). Underreporting of substance use
in a national longitudinal youth cohort: Individual and interviewer
effects. Public Opinion Quarterly, 52, 100–124.
Meston, C. M., Heiman, J. R., Trapnell, P. D., & Paulhus, D. L. (1998).
Socially desirable responding and sexuality self-reports. Journal
of Sex Research, 35, 148–157.
Mosher, D. L. (1988). Hypermasculinity inventory. In C. M. Davis, W.
L. Yarber, & S. L. Davis (Eds.), Sexuality-related measures: A
compendium (pp. 225–227). Lake Mills, IA: Graphic Publishing
Company.
Murnen, S. K., & Byrne, D. (1991). Hyperfemininity: Measurement
and initial validation of the construct. Journal of Sex Research,
28, 85–106.
Pratto, F. (1996). Sexual politics: The gender gap in the bedroom,
the cupboard, and the cabinet. In D. M. Buss & N. M. Malamuth
(Eds.), Sex,power, conflict: Evolutionary and feminist perspectives
(pp. 179–230). New York: Oxford University Press.
Schroder, K. E. E., Carey, M. P., & Vanable, P. A. (2003). Methodolog-
ical challenges in research on sexual risk behavior: II. Accuracy
of self-reports. Annals of Behavioral Medicine, 26, 104–123.
Segal, L. (2001). The belly of the beast: Sex as male domination? In
S. M. Whitehead & F. J. Barrett (Eds.), The masculinities reader
(pp. 100–111). Malden, MA: Blackwell Publishers.
Shope, G. L, Hedrick, T. E., & Geen, R. G. (1978). Physical/verbal
aggression: Sex differences in style. Journal of Personality, 46,
23–42.
Sidanius, J., & Pratto, F. (1999). Social dominance: An intergroup
theory of social hierarchy and oppression. New York: Cambridge
University Press.
Simpson, J. A. (1998). Sociosexual Orientation Inventory. In C. M.
Davis, W. L. Yarber, R. Bauserman, G. Schreer, & S. L. Davis
(Eds.), Handbook of sexuality-related measures (pp. 565–567).
Thousand Oaks, CA: Sage Publications.
Simpson, J. A., & Gangestad, S. W. (1991). Individual differences in
sociosexuality: Evidence for convergent and discriminant validity.
Journal of Personality and Social Psychology, 60, 870–883.
Strahan, R., & Gerbasi, K. C. (1972). Short, homogeneous versions of
the Marlowe-Crowne Social Desirability Scale. JournalofClinical
Psychology, 28, 191–193.
Tolman, D. L. (2002). Dilemmas of desire: Teenage girls talk about
sexuality. Cambridge, MA: Harvard University Press.
Tourangeau, R., & Smith, T. W. (1996). Asking sensitive questions:
The impact of data collection mode, question format, and question
context. Public Opinion Quarterly, 60, 275–304.
Weisfeld, C. C., & Stack, M. A. (2002). When I look into your eyes:
An ethological analysis of gender differences in married couples’
nonverbal behaviors. Psychology, Evolution and Gender, 4, 125–
147.
Wiederman, M. W. (1997). The truth must be in here somewhere: Ex-
amining the gender discrepancy in self-reported lifetime number
of sex partners. Journal of Sex Research, 34, 375–386.
Willer, R. (2005, August). Overdoing gender: A test of the masculine
overcompensation thesis. Paper presented at the meeting of the
American Sociological Association, Philadelphia.
Winer, G. A., Makowski, D., Alpert, H., & Collins, J. (1988). An anal-
ysis of experimenter effects on responses to a sex questionnaire.
Archives of Sexual Behavior, 17, 257–263.
Wood, W., & Eagly, A. H. (2002). A cross-cultural analysis of the
behavior of women and men: Implications for the origins of sex
differences. Psychological Bulletin, 128, 699–727.
Springer
