This article was downloaded by: [McMaster University]

On: 06 March 2015, At: 08:48

Publisher: Routledge
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

International Journal of Sexual Health

Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/wijs20

Women’s Experiences of Sexual Attention: a Cross-

Sectional Study of U.S. University Students
a b b b
Samantha J. Gregus , Christina M. Rummell , Thomas J. Rankin & Ronald F. Levant
a
Department of Psychology, The University of Arkansas, Fayetteville, Arkansas, USA
b
Department of Psychology, The University of Akron, Akron, Ohio, USA
Accepted author version posted online: 25 Feb 2014.Published online: 07 Oct 2014.

Click for updates

To cite this article: Samantha J. Gregus, Christina M. Rummell, Thomas J. Rankin & Ronald F. Levant (2014) Women’s
Experiences of Sexual Attention: a Cross-Sectional Study of U.S. University Students, International Journal of Sexual Health,
26:4, 239-257, DOI: 10.1080/19317611.2014.885922

To link to this article: http://dx.doi.org/10.1080/19317611.2014.885922

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 International Journal of Sexual Health, 26:239–257, 2014
Copyright C Taylor & Francis Group, LLC

ISSN: 1931-7611 print / 1931-762X online

DOI: 10.1080/19317611.2014.885922

WOMEN’S EXPERIENCES OF SEXUAL ATTENTION: A CROSS-SECTIONAL STUDY

OF U.S. UNIVERSITY STUDENTS

Samantha J. Gregus1, Christina M. Rummell2, Thomas J. Rankin2, Ronald F. Levant2

1
Department of Psychology, The University of Arkansas, Fayetteville, Arkansas, USA
2
Department of Psychology, The University of Akron, Akron, Ohio, USA

ABSTRACT. Objectives: This cross-sectional investigation of women’s experiences of sexual

attention examined the role of dispositional and situational variables in how women experi-
enced sexual attention (as positive, negative, or neutral). Methods: Participants were 350 U.S.
college women recruited from undergraduate psychology courses. They completed question-
naires on objectified body consciousness, social physique anxiety, self-esteem, body esteem,
Downloaded by [McMaster University] at 08:48 06 March 2015

and social desirability. A subset (N = 275) also reported retrospectively on experiences with
sexual attention in 1 of 4 contexts: at a bar/club, at a gym, at school, or at work. It was
hypothesized that the context where sexual attention occurs would be associated with how
positive or negative the encounter was experienced. Results: The relationship between con-
text and experience of sexual attention approached but did not achieve significance. When
controlling for demographic variables and social desirability, self-esteem and body esteem
were negatively associated with self-objectification and social physique anxiety. White women
younger than the age of 25 with higher body mass index were most likely to engage in self-
objectifying behaviors. Themes identified from responses to open-ended questions describe
reasons for experiencing sexual attention to be uncomfortable or pleasurable. Conclusions:
There is wide variability in women’s experiences of sexual attention. Self-esteem and body
esteem may protect against self-objectification. Future studies should examine how contextual
factors influence women’s experiences of sexual attention.

KEYWORDS. Sexual attention, objectification, self-objectification, women

Sexual attention can be experienced any- Unwanted Sexual Attention and

where: on the street, at work, at school, in a Objectification
bar, at the gym, or on a date. For some peo- Several studies have examined women’s
ple, in certain circumstances, sexual attention is experiences of unwanted sexual attention
desirable, appreciated, and experienced as val- (Esacove, 1998; Fairchild & Rudman, 2008).
idating. For others, in particular circumstances, Unwanted sexual attention is considered a main
it can be unwanted and experienced as harmful component of sexual harassment and involves
or degrading (Esacove, 1998; Fairchild & Rud- the degradation of women at the individual
man, 2008). The purpose of the present study level (Gelfand, Fitzgerald, & Drasgow, 1995).
was to better understand how women experi- Unwanted sexual attention can take the form
ence the receipt of sexual attention in different of sexual harassment by either known perpetra-
contexts. Most research to date has focused on tors or strangers (stranger harassment) and can
unwanted sexual attention and the objectifica- range from catcalls, whistles, or stares to sexual
tion of women. assault (Cochrane, Frazier, & Olson, 1997).

Received 22 May 2013; revised 28 December 2013; accepted 17 January 2014.

Address correspondence to Samantha J. Gregus, Department of Psychology, The University of Arkansas, 216 Memorial Hall,
Fayetteville, AR 72701, USA. E-mail: sgregus@uark.edu

239
 240 INTERNATIONAL JOURNAL OF SEXUAL HEALTH

Unwanted sexual attention is often experi- Self-objectification has been found to be nega-
enced as sexually objectifying. Sexual objecti- tively related to women’s self-esteem and body
fication occurs when the behavior of one per- esteem and positively related to body shame
son, the agent, leaves another person feeling and body surveillance (i.e., habitual monitor-
that the agent has separated her or his body, ing of the body and appearance; Fredrick-
body parts, or sexuality from her or his identity son & Roberts, 1997). Simply anticipating
as an individual (Fredrickson & Roberts, 1997). the “male gaze” produced significantly greater
Self-objectification theory posits that the more body shame and social physique anxiety in a
exposure women have to being sexually objecti- group of women than did anticipation of a “fe-
fied, the more they are likely to internalize these male gaze” (Calogero, 2004).
messages of being appreciated only for their Interestingly, research has demonstrated
bodies or body parts (Fredrickson & Roberts, that women who receive compliments about
1997). Thus, they adopt an observer’s perspec- their external appearance in a positive manner
tive on their physical selves and engage in self- still show increases in body surveillance, body
Downloaded by [McMaster University] at 08:48 06 March 2015

objectification. Sexual objectification does not dissatisfaction, and body shame (Calogero,
invariably lead to self-objectification, but rather Herbozo, & Thompson, 2009; Tiggemann &
creates the potential for this outcome (Yoder, Boundy, 2008). These findings suggest that
2007). When women self-objectify, they define some women may come to expect to be ob-
their self-worth in terms of their appearance and jectified and that compliments may serve as
thus become motivated to appear attractive and reminders that their bodies are evaluated by
pleasing to others (Breines, Crocker, & Garcia, others (Calogero et al., 2009; Tiggemann &
2008). Boundy, 2008). Because many women regard
Negative physical and psychological being sexually objectified as a normative expe-
consequences have been associated with rience, it is possible that they have internalized
both unwanted sexual attention and self- the notion that their self-worth is related to their
objectification. In a qualitative study, Esacove physical attractiveness and that this physical at-
(1998) found that women who reported tractiveness can be measured by how much sex-
unwanted sexual attention by strangers ex- ual attention they receive from others.
perienced both physical symptoms (e.g.,
chest pains, sweating, sleep disturbances,
feeling shaky, and feeling flustered) and Variations in Experiences of Sexual
emotional effects (e.g., anger or rage, depres- Attention
sion, feeling violated, disconnected, guilty, Moffitt and Szymanski (2011) found that al-
and self-conscious) following the experience. though many women disliked unwanted sexual
Moffitt and Szymanski (2011) found additional attention and experienced negative conse-
negative sequelae, including feelings of disgust, quences from it, some women reported ben-
degradation, sadness, anxiety, and worsened efits such as finding it complimentary and
relationships with other women. Additionally, confidence-boosting. Liss, Erchull, and Ramsey
Fairchild and Rudman (2008) found harass- (2011) found that women can have a wide
ment by strangers to be linked to an increased range of responses to sexualized male attention,
fear and perceived risk for rape. from very positive to very negative. In a similar
Objectification theory posits multiple neg- vein, Fischer, Bettendorf, and Wang (2011) of-
ative psychological consequences of self- fered the perspective that women may have
objectification, including shame, appearance been socialized to view sexual objectification
anxiety, concern about physical safety, and as an experience that is validating and possibly
decreased awareness of internal body states empowering. Therefore, there are likely a num-
(Fredrickson & Roberts, 1997). It also posits ber of sociocultural, individual, and situational
a loss of “flow,” defined as “an optimal en- variables that influence how a woman experi-
gagement of one’s skills” (Myers, 2008, p. 393). ences sexual attention.
 S. J. GREGUS ET AL. 241

Sociocultural McKinley, 1998; McKinley & Hyde, 1996).

In U.S. and other Western cultures, physi- Additionally, higher levels of trait self-esteem
cal attractiveness is often the ticket to social and have been found to predict lower levels of
financial success. This may motivate women to the Body Surveillance subscale of the OBCS
objectify themselves and may motivate both (Breines et al., 2008). Several other studies
men and women to reinforce other women have examined the relationship between self-
for appearing attractive (Fairchild & Rudman, objectification, self-esteem, and body esteem
2008; Fredrickson & Roberts, 1997; Moffitt (Greenleaf & McGreer, 2006; Szymanski &
& Szymanski, 2011). Furthermore, the desire Henning, 2007). These studies looked at out-
to be thin and have the “ideal” body type comes from objectification but did not exam-
has led to wide acceptance and sometimes ine self-esteem and body-esteem as predictors
encouragement of self-objectification (Harper of, and therefore as potential protective factors
& Tiggemann, 2008). A recent review of re- for, objectification. Additionally, previous stud-
search has documented that the media depict ies have not controlled for demographic vari-
Downloaded by [McMaster University] at 08:48 06 March 2015

a narrow and often unattainable standard of ables such as age, race, and body mass index
women’s physical beauty and consequently link (BMI) or for social desirability in exploring these
this standard with a woman’s degree of sexiness relationships. BMI is an estimate of body fat
and worth (American Psychological Association that can be used to assess level of overweight
[APA], 2007). Indeed, exposure to sexually ob- or obesity in adult men and women (National
jectifying media has been shown to be related to Heart, Lung, and Blood Institute, 2013). In pre-
girls placing greater importance on beauty and vious research, BMI has been found to corre-
appearance in defining their own self-worth, late with body esteem and self-esteem (Jang
as well as in defining the value of women in & Lee, 2013; Shloim, Hetherington, Rudolf, &
general (Gordon, 2008). Given this sociocul- Feltbower, 2013; Wilson, Latner, & Hayashi,
tural context, sexual attention and objectifica- 2013). Thus, many researchers control for BMI
tion may be experienced as normal or validating when assessing body esteem’s or self-esteem’s
and therefore may provide some women the relationships with other variables (Pearl & Puhl,
opportunity to experience a sense of mastery 2013; Rudiger & Winstead, 2013; Schoene-
and control (Fischer et al., 2011). feld & Webb, 2013). Controlling for social de-
sirability is also important because those vari-
ables that are more valued in a culture (such
Individual as appearance-related variables for Western
Although all U.S. women live in a sexually women) are typically those that are most af-
objectifying environment, not all of them are fected by socially desirable responding (Fisher
similarly affected by it (Fredrickson & Roberts, & Katz, 2000; Hurt et al., 2007).
1997). This suggests that there might be fac-
tors that protect some women from the harm-
ful effects of objectification. Two individual Contextual
difference variables that may be associated Situational or contextual factors impact
with women’s differing experiences with sex- the perception of unwanted sexual attention.
ual attention are self-esteem and body esteem. Fairchild (2010) asked women to complete a
Higher self-esteem and higher body esteem checklist of specific contextual features that
may buffer some women from the harmful ef- would make an experience of stranger harass-
fects of sexual objectification. Therefore, these ment either more enjoyable or more frighten-
variables may be potential protective factors. ing. Consistent with prior sexual harassment
Previous research has shown that body es- research, participants reported that attractive,
teem is negatively related to both the Body younger perpetrators would create a more en-
Shame and Body Surveillance subscales of the joyable experience, while older, less attractive
Objectified Body Consciousness Scale (OBCS; men who were of a different race would make
 242 INTERNATIONAL JOURNAL OF SEXUAL HEALTH

the experience more frightening. Further, being McKinley & Hyde, 1996), such studies have
alone, at night, and in a public place tended not controlled for the influence of demographic
to be the most frequently checked items for variables, BMI, or social desirability.
increased fear, while being with girlfriends, at Finally, most research on objectification
a bar or restaurant, and during the daytime theory is quantitative in nature. Although quan-
tended to be the most frequently checked items titative research generates generalizable knowl-
for increased enjoyment (Fairchild, 2010). So- edge about sexual objectification and its respec-
cial context thus appears to influence whether tive psychosocial correlates, it is not well suited
sexual attention is received positively or nega- to provide a deeper understanding of the “mul-
tively and therefore should be a focus of further tifaceted nature of sexual objectification” (Mof-
research. fitt & Szymanski, 2011, pp. 68–69). Moreover,
no published studies have looked at women’s
phenomenological experience of sexual atten-
Limitations of Prior Studies and New tion as it occurs in the moment or shortly there-
Research Directions
Downloaded by [McMaster University] at 08:48 06 March 2015

after. Therefore, women’s open-ended reports

Most research on sexual attention has ex- of their experience of sexual attention shortly
amined unwanted sexual attention/advances. after it occurs would be useful in enriching our
Although this is clearly important, there has understanding of the phenomena of sexual at-
been scant investigation of wanted attention, tention and objectification in a larger sociocul-
or advances to which a given woman might tural context.
be indifferent. Sexual attention that is desired
by a particular woman would likely impact her Purpose and Hypotheses
differently than would attention that she does
The present study examined situational and
not desire. Therefore, research should explic-
individual difference variables associated with
itly examine whether the sexual attention being
whether women experienced sexual attention
studied was wanted, unwanted, or experienced
as positive or negative. Based on limitations in
in a neutral fashion. In addition, prior research
the extant literature, we formulated three hy-
has explored contextual factors in a very limited
potheses and one exploratory question.
manner. Further examination of the contextual
factors that may be associated with the differing
experiences that women have with sexual at-
Hypothesis 1
tention is needed. For example, research should
explore the nature of the environment in which The social context where sexual attention
the sexual attention occurs and whether sexu- occurs will be associated with how positively or
ality is heightened or experienced as intrusive negatively the encounter was experienced. In
in that environment. environments of heightened sexuality (i.e., on a
Although prior research has shown that night out at a club or bar), participants will be
women who received appearance-related com- more likely to experience sexual attention pos-
pliments showed increases in body shame and itively. In environments where a focus on sexu-
surveillance, research has not addressed the ality seems intrusive and distracting (i.e., either
broader question of whether participants who work or classroom settings), participants will be
desire sexual attention will still show elevated more likely to experience sexual attention neg-
levels of self-objectification. However, this may atively. In environments that are more ambigu-
be anticipated due to the normative nature ous with regard to whether sexual attention is
of objectification. Additionally, although previ- appropriate or not (i.e., the campus recreation
ous research has shown that self-esteem and center, which contains a coeducational gym),
body esteem are significantly negatively related participants’ responses to sexual attention will
to both the Body Shame and Body Surveil- be mixed, with some viewing it positively and
lance subscales of the OBCS (McKinley 1998; others negatively.
 S. J. GREGUS ET AL. 243

Hypothesis 2 being married, partnered, or engaged, 9.1% re-

Participants who experience sexual atten- ported being nonexclusively or casually dating,
tion positively will nonetheless show elevated 0.6% reported being either divorced or sepa-
levels of the body shame and body surveil- rated, and the remaining 23.1% indicated that
lance dimensions of self-objectification due to they were single and not currently dating any-
the normative nature of sexual objectification. one. Of those in a romantic relationship of some
kind, 85.1% reported that they were satisfied or
extremely satisfied with the relationship.
Hypothesis 3
Self-esteem and body esteem will predict, Procedure
with negative beta weights, self-objectification
(as measured by the Body Shame and Body This study was approved by the university
Surveillance subscales of the OBCS) and social institutional review board prior to data collec-
physique anxiety, after controlling for social de- tion. Participants were recruited from under-
Downloaded by [McMaster University] at 08:48 06 March 2015

sirability and demographic variables. This will graduate psychology courses and were offered
indicate that self-esteem and body esteem may course extra credit for their participation. Re-
be protective factors for self-objectification and search assistants visited individual classrooms
social physique anxiety. to describe the study, and participants who ex-
pressed an interest in the study provided their
e-mail addresses to the researchers. Participants
Exploratory question were then e-mailed a link to the online sur-
vey, which was conducted using SurveyMon-
To explore women’s phenomenological ex-
key, a commercially available survey tool. Par-
periences of sexual attention shortly after the
ticipants were first presented with the informed
experience occurs, we inquired as to whether
consent document, where they were asked to
women recount such an experience as either
click “Next” to indicate their consent to partici-
uncomfortable or pleasurable and what factors
pate in the study. To avoid duplicate responses,
made it so.
participants were required to create a unique
identifier (not their name) that would be linked
METHOD to their survey responses. Next, the participants
completed the questionnaires described in the
Participants next section in the order listed, as well as the
Participants were 365 women attending a demographic form.
large public Midwestern research university. Fif- Upon completion of these measures, the
teen participant cases were deleted from the next screen was randomly selected from one of
data file due to incomplete data, leaving a final the contexts: night out, classroom or work con-
sample size of 350 (95.9% completion rate). text,1 or the student recreation center (gym). A
Ages of the women ranged from 18 to 55 years, question was presented on the screen that read,
with a mean age of 20.37 years (SD = 3.91). “In the last 48 hours, did you [insert randomly
The majority of the participants were European selected context here], AND while you were
American/Caucasian (85.1%), with 8.9% identi- in [context], did you notice someone check-
fying as African American/Black and 6.0% iden- ing you out (e.g., eyeing you up and down);
tifying as other ethnicities. In regards to sex- did someone hit on you (e.g., whistle at you,
ual orientation, 93.1% of the women identified try to pick you up, or take you home); or did
as exclusively heterosexual, 1.4% as exclusively someone make sexually derogatory gestures or
lesbian, and 5.5% as falling somewhere in be-
tween. Approximately half the sample reported 1As described in the Results section, this was later divided
that they were currently in an exclusive dat- into two separate categories based on differences in participant
ing relationship (50.9%), while 16.3% reported responses.
 244 INTERNATIONAL JOURNAL OF SEXUAL HEALTH

comments to you?” (Example: “In the last Kurtz, Tarquini, and Iobst (2008) and Crowne
48 hours, did you have a ‘night out’ [e.g., went and Marlowe (1960). The current study found
to a bar, club, party, etc.?], AND while you were an α coefficient of .72.
on your night out . . .”). If the participant an-
swered yes, she was directed to a new screen
containing the Context-Specific Sexual Atten- Rosenberg self-esteem scale (RSES)
tion Questionnaire (CSSAQ). If the participant The RSES (Rosenberg, 1965) is a 10-item
answered no, she was randomly presented with questionnaire. Participants respond to ques-
the same question about one of the two re- tions about their self-worth and satisfaction with
maining contexts, continuing until either she themselves. A sample item is, “On the whole,
answered “yes” or all contexts had been pre- I am satisfied with myself.” Items are rated on
sented. Participants who answered “no” to all a 4-point Likert scale, ranging from 1 (strongly
three options were thanked for their participa- disagree) to 4 (strongly agree). Scores range
tion and exited from the survey. from 10 to 40, with higher scores indicating
Downloaded by [McMaster University] at 08:48 06 March 2015

Once participants completed the CSSAQ higher self-esteem. Scale developers reported
regarding their specific context, they were given an α of .81 and test–retest reliability of .85
the option to provide more information about (Rosenberg, 1965). Scores on the RSES have
what happened when they received sexual at- also strongly correlated with peer ratings of self-
tention, how it made them feel, and details esteem (Demo, 1985). The current study re-
about the encounter (“What about the expe- vealed an α of .84.
rience was uncomfortable?” and “What about
the experience was pleasurable?”). Participants
who completed this section were given an addi- Social physique anxiety scale (SPAS)
tional extra credit point. After entering their in- The SPAS measures a person’s anxiety re-
formation, the participants were then presented sulting from the thought of another person eval-
with a debriefing form and were asked to close uating her or his body (Hart, Leary, & Rejeski,
their Web browsers when finished. 1989). The scale contains 12 items, which are
rated on a 5-point Likert scale, ranging from 1
Measures (strongly disagree) to 5 (strongly agree). Scores
range from 12 to 60, with higher scores indicat-
Marlowe-Crowne social desirability ing greater social physique anxiety. An example
scale (MCSDS) item from this measure is, “It would make me
The MCSDS (Crowne & Marlowe, 1960) uncomfortable to know others were evaluating
consists of 33 true/false items that measure the my physique/figure.” Past studies have shown
tendency to respond in socially desirable ways. an α of .90 (Hart et al., 1989), test–retest reli-
Eighteen are keyed as “true” items (i.e., highly ability of .94 during a 2-week interval (Scott,
desirable behaviors but low probability of oc- Burke, Joyner, & Brand, 2004), and 8-week
currence) and 15 are keyed as “false” items (i.e., test–retest reliability of .82 (Hart et al., 1989).
socially disapproved behaviors but high proba- These latter authors also demonstrated con-
bility of occurrence). An example of a true item struct validity for the SPAS through moder-
is, “I never hesitate to go out of my way to help ate correlations with other measures that tap
someone in trouble.” Scores range from 0 to concerns with others’ evaluations. The current
33, with higher scores indicating more socially study found a full-scale α of .91.
desirable responding. The scale has been found
to have good internal consistency with an α of
.88 and test–retest reliability of .89 (Crowne Objectified body consciousness scale
& Marlowe, 1960). Nordholm (1974) reported The OBCS (McKinley & Hyde, 1996) was
split-half reliability to be .73 for the measure. used to measure three aspects of objectified
Evidence for construct validity was provided by body consciousness, reflected in the following
 S. J. GREGUS ET AL. 245

three subscales: Body Surveillance (“viewing for the purpose of calculating their BMI so that
the body as an outside observer”), Body Shame BMI could be used as a control variable.
(“feeling shame when the body does not
conform”), and Appearance Control Beliefs
(McKinley & Hyde, 1996, p. 181). A sample Context-specific sexual attention
item is, “When I’m not exercising enough, I questionnaire
question whether I am a good enough person.” The CSSAQ, developed specifically for the
Each subscale consists of 8 items, for a total current study, assessed women’s subjective ex-
of 24 items, rated on a 7-point Likert scale; re- periences of sexual attention. This measure
sponse choices ranged from 1 (strongly disagree) asked about participants’ experiences in the last
to 7 (strongly agree). Subscale scores range from 48 hr in one of four contexts likely to be en-
8 to 56, with higher scores indicative of greater countered by the typical college student: in the
endorsement. The total scores range from 24 student recreation center (i.e., a coeducational
to 168. Higher scores indicate increased self- gym), in an academic environment, in a work
objectification. Prior research has reported αs of
Downloaded by [McMaster University] at 08:48 06 March 2015

environment, or while on a night out (e.g., at

.89 for Body Surveillance, .75 for Body Shame, a bar or club). The questionnaire consisted of
and .72 for Appearance Control Beliefs (McKin- 16 statements that tapped participants’ positive
ley & Hyde, 1996). The current study showed and negative experiences when receiving sex-
an α of .84 for both Body Shame and Surveil- ual attention plus a number of background and
lance. Control Beliefs were not analyzed in the exploratory questions, which were not analyzed
present study. in the current study. The response choices for
these Likert-type items ranged from 1 (strongly
disagree) to 5 (strongly agree). The items broadly
Body esteem scale (BES) assessed the following domains: (a) feelings of
The BES (Franzoi & Shields, 1984) is a 35- attraction toward and interest in the person who
item scale that lists different appearance and provided sexual attention; (b) how the partici-
functional aspects of the body (i.e., “waist,” pant felt about her body and self-esteem dur-
“weight,” “physical stamina”) in which a partic- ing the experience of receiving sexual attention;
ipant rates on a scale of 1 (having strong nega- and (c) how pleasurable, impressionable, excit-
tive feelings) to 5 (having strong positive feelings) ing, appropriate, and comfortable her experi-
how positive or negative she feels toward each ence of sexual attention was. It should be noted
aspect of her appearance. Total scores range that the questions regarding experiences of ob-
from 35 to 175, with higher scores indicating jectification were sex-neutral (e.g., referring to
higher body esteem. The current study summed the person who gave sexual attention as “he
the results as a total score following Silberstein, or she”). However, because the majority of our
Striegel-Moore, Timko, and Rodin (1988), who participants identified as heterosexual and all
found α = .93. Convergent and discriminant but two of the respondents reported receiving
validity were also demonstrated by the scale sexual attention from a man, the main focus of
developers. In the present study, α = .92. our research was male-to-female sexual atten-
tion. Nonetheless, because they still provided
relevant data, responses concerning attention
Demographic questionnaire from a woman were not removed.
Participants were also asked to complete a
brief demographic questionnaire. Information
about participant gender identity, age, sexual Statistical Analyses
orientation, ethnicity, socioeconomic status, re- The data were first cleaned and missing
lationship status, and relationship satisfaction values were replaced. It was then assessed
was collected. In addition, participants were whether there were differences between those
also asked to provide their height and weight who did or did not indicate that they were
 246 INTERNATIONAL JOURNAL OF SEXUAL HEALTH

the recipients of sexual attention in the last age (t = 0.59, p = .55), relationship status (t =
48 hr and thus did or did not complete the 0.12, p = .91), or BMI (t = 0.63, p = .53). Of
CSSAQ. Next, the CSSAQ was subjected to the participants who completed the CSSAQ, 32
exploratory factor analysis. Descriptive statis- (11.6%) answered questions about being hit on
tics and bivariate correlations for the study or checked out in the student recreation center,
variables were then calculated. The three hy- 119 (43.2%) answered questions about being
potheses were analyzed using between-subjects hit on or checked out during a night out, and
multivariate analysis of variance, bivariate 124 (45.1%) answered questions about being
correlations, and multiple linear regression, hit on or checked out in a professional setting
respectively. Finally, in regard to the exploratory (defined as either work or an academic con-
question, qualitative responses to open-ended text).
questions were coded and analyzed for emer-
gent themes.
RESULTS
Downloaded by [McMaster University] at 08:48 06 March 2015

Data Cleaning and Missing Values Exploratory Factor Analysis of CSSAQ

Before conducting statistical analyses, the An exploratory factor analysis of the 16
data were thoroughly screened to ensure the CSSAQ items was carried out using principal
accuracy of the file. There was a small amount axis factoring (PAF). Given our goal of under-
(0.28%) of missing data, with the number of standing the latent structure of the items and
missing responses per item ranging from 0 to 8. subscales comprising the CSSAQ, PAF seemed
With a small percentage of data points missing to be the more appropriate technique. We also
from a large data set, the problem is less seri- chose PAF because it does not assume mul-
ous, according to Tabachnick and Fidell (2007). tivariate normality (Fabrigar, Wegener, Mac-
Nonetheless, we took a conservative approach, Callum, & Strahan, 1999). This was impor-
and replaced missing values using the Statistical tant given that responses to some CSSAQ items
Package for the Social Sciences (SPSS) Version were skewed, which would violate the assump-
19 Linear Trend at Point method. tion. Based on a parallel analysis performed
using O’Connor’s (2000) SPSS computer pro-
gram, we fixed the number of factors to extract
Preliminary Analyses at four and used Promax rotation with Kaiser
Three-hundred and fifty participants com- normalization. This oblique rotation criterion
pleted the initial set of measures (i.e., the was chosen following Preacher and MacCallum
MCSDS, RSES, SPAS, OBCS, BES, and the de- (2003), who emphasized that orthogonal cri-
mographic questionnaire). Following this, 68 of teria “are rarely defensible because factors are
the 350 participants indicated that they had rarely if ever uncorrelated in empirical studies”
not been hit on or checked out in any of the (p. 40).
three contexts presented in the CSSAQ, and Extraction with four factors accounted for
they were exited from the survey. The remain- 55.33% of the variance. We set the minimum
ing 282 participants went on to complete the loading allowable to .35, which resulted in the
CSSAQ. However, 7 of these participants’ data removal of two items. In addition, one item
were removed due to incomplete data, for a was removed because of cross-loading prob-
final total of 275 complete responses from the lems, following Tabachnick and Fidell’s (2007)
CSSAQ. suggestion that items that load .32 or greater
Results of independent-samples t tests indi- on a second factor should be removed. As a re-
cated that there were no significant differences sult, the third and fourth factors were not viable
between participants who were exited from the because each factor had only two items and at
survey (N = 68) and those who went on to least three items are required for factor stability
complete the CSSAQ (N = 275) in regard to (Kahn, 2006).
 S. J. GREGUS ET AL. 247

TABLE 1. Factor Loadings and Text of the Nine Items Retained in the Two Factors of the CSSAQ

Factor Loadings Item Text

Factor 1: “Feelings of Attraction and Pleasure”

.74 How interested were you in the person checking you out or hitting on you?
.73 Would you please rate how pleasurable the experience feels as you think about it right now?
.69 Would you please rate how pleasurable the experience felt at the time it occurred?
.65 How attractive did you find this person?
.50 Please rate the level of inappropriateness of the actions of the person who checked you out. (Reverse Scored)
Factor 2: “Feelings of Self-Esteem”
.87 While this person was checking you out or hitting on you, how would you rate your self-esteem?
.60 Afterward, how did you feel about your body?
.54 How exhilarating or exciting did you find this experience of being hit on or checked out?
.46 Please rate your overall experience [in the particular context].

Note. N = 275. CSSAQ = Context-Specific Sexual Attention Questionnaire. Factor 1 accounted for 26.7% of the variance, Cronbach’s
alpha = .79. Factor 2 accounted for 12.2% of the variance, Cronbach’s alpha = .67.
Downloaded by [McMaster University] at 08:48 06 March 2015

Therefore, the foregoing produced two fac- objectification. Social desirability was positively
tors. There were five items on the first factor. associated with body esteem and negatively
The highest two loading were, “How interested associated with social physique anxiety, body
were you in the person checking you out or hit- surveillance, and body shame. Assumptions of
ting on you?” and “Would you please rate how normality and absence of multicollinearity were
pleasurable the experience feels as you think checked and met. Normality was assessed by
about it right now?” We labeled this factor “feel- examining skewness and kurtosis of the ma-
ings of attraction and pleasure.” There were four jor outcome variables as well as by examining
items on the second factor. The highest loading histograms. Multicollinearity was assessed by
item was, “While this person was checking you examining correlations among major outcome
out or hitting on you, how would you rate your variables as well as by examining collinearity
self-esteem?” We labeled this factor “feelings of statistics (i.e., tolerance and variable inflation
self-esteem.” The alpha coefficients were evalu- factor).
ated using the criteria developed by Ponterotto
and Ruckdeschel (2007, p. 1003). Feelings of
attraction and pleasure (α = .79) would be con- Tests of Hypotheses
sidered “good,” whereas feelings of self-esteem Hypothesis 1: The context where sexual
(α = .67) would be considered “fair.” The text attention occurs will be associated with
of all nine items and their loadings on the re- how positively or negatively the
spective factors are displayed in Table 1. encounter was experienced, with the
most positive experiences expected in
Descriptive Statistics bars and clubs, the most negative in
Means, standard deviations, and intercorre- workplaces or classrooms, and mixed
lations of the study variables are shown in Ta- results at the campus gym
ble 2. All outcome variables were significantly First, it is noteworthy that for all contexts,
correlated with each other in the expected di- 80.5% of the 275 participants completing the
rections. Self-esteem and body esteem were CSSAQ responded that they felt complimented
positively associated with each other. Social by the sexual attention they received. The per-
physique anxiety and self-objectification (as centages of women in each of the contexts who
measured by the Body Shame and Body reported feeling complimented were: 86.5% at
Surveillance subscales of the OBCS) were also the gym, 82.6% during a night out, and 72.4%
positively associated with each other. Self- in class or work contexts.
esteem and body esteem were both negatively The positive or negative quality of the en-
correlated with social physique anxiety and self- counter was measured by the factors of feelings
 248 INTERNATIONAL JOURNAL OF SEXUAL HEALTH

TABLE 2. Means, Standard Deviations, and Intercorrelations Among Major Outcome Variables

Self- Social Social Physique Body Body Body Esteem

Variables Mean (SD) Esteem Desirability Anxiety Surveillance Shame Scale Factor 1 Factor 2

Self-Esteem 3.04 (0.43) 1 — — — — — — —

Social 15.71 (4.63) .295∗∗ 1 — — — — — —
Desirability
Social Physique 3.40 (0.74) −.460∗∗ −.265∗∗ 1 — — — — —
Anxiety
Body 4.47 (1.05) −.157∗∗ −.234∗∗ .513∗∗ 1 — — — —
Surveillance
Body Shame 3.49 (1.17) −.419∗∗ −.178∗∗ .649∗∗ .439∗∗ 1 — — —
Body Esteem 3.35 (0.58) .524∗∗ .273∗∗ −.703∗∗ −.272∗∗ −.528∗∗ 1 — —
Factor 1: 2.92 (0.89) −.077 −.049 .035 .087 .024 −.049 1 —
“Feelings of
Attractive-
ness”
Factor 2: 3.33 (0.63) .023 −.009 .101 .097 .074 −.017 .501∗∗ 1
Downloaded by [McMaster University] at 08:48 06 March 2015

“Feelings of
Self-Esteem”

Note. N = 275. Self-esteem was measured by the Rosenberg Self-Esteem Scale; social desirability was measured by the Marlowe-
Crowne Social Desirability Scale; social physique anxiety was measured by the Social Physique Anxiety Scale; body surveillance and body
shame were measured by the Objectified Body Consciousness Scale; body esteem was measured by the Body Esteem Scale; and feelings
of attractiveness and feelings of self-esteem were measured by the Context-Specific Sexual Attention Questionnaire. ∗∗ p < .01.

of attraction and pleasure and feelings of self- 2.63 (SD = 0.89), and 2.92 (SD = 0.91) for
Esteem from the CSSAQ. Thus, higher scores the gym, night-out, work, and class contexts,
on these factors indicate more positive experi- respectively. Means for feelings of self-esteem
ences, whereas lower scores on these factors in- were 3.22 (SD = 0.57), 3.43 (SD = 0.67),
dicate more negative experiences. A between- 3.18 (SD = 0.63), and 3.28 (SD = 0.56) for
subjects multivariate analysis of variance was the gym, night-out, work, and class contexts,
performed on both measures of the dependent respectively. The test of between-subject ef-
variable of self-objectification. The indepen- fects was not significant for feelings of attraction
dent variable was context. Although this was ini- and pleasure, F(3, 261) = 1.831, p = .14, but
tially categorized into three groups—gym, night it approached significance for feelings of self-
out, work/class—preliminary analyses indicated esteem, F(3, 261) = 2.275, p = .08. Tukey
differences in how participants rated the expe- follow-up tests revealed no significant differ-
rience of sexual attention in work settings as ences between contexts on either measure of
opposed to classroom settings, in which sex- positive experiences. Thus, Hypothesis 1 was
ual attention seemed to be considered more not supported.
appropriate. Thus, the work/class category was
divided into two separate categories: work and
class. Using Wilks’s criterion, differences in the Hypothesis 2: Participants who
two dependent variables attributable to differ- experience sexual attention positively
ences in these four contexts approached sig- will nonetheless show elevated levels on
nificance, F(6, 520) = 2.107, p = .051. These the body shame and body surveillance
results indicated a small effect size, partial η2 = dimensions of self-objectification due to
.024. This indicates that the experience of pos- the normative nature of sexual
itive feelings when receiving sexual attention objectification
may depend on the context in which the sexual This hypothesis was tested by analyzing a
attention occurs. series of bivariate correlations, which are shown
Means for feelings of attraction and plea- in Table 2. Participants’ positive experiences
sure were 3.08 (SD = 0.59), 2.95 (SD = 0.93), of sexual attention were measured with the
 S. J. GREGUS ET AL. 249

Feelings of Attraction and Pleasure and Feelings body esteem, and self-esteem were significant
of Self-Esteem subscales of the CSSAQ. Body predictors in the final model. BMI was posi-
shame scores (as measured by that subscale tively associated with body shame, while body
of the OBCS) were not significantly associated esteem and self-esteem were both negatively
with feelings of attraction and pleasure (r = .02, related to body shame. BMI and social desir-
p = .69) nor were they significantly associated ability were controlled for in these analyses.
with feelings of self-esteem (r = .07, p = .22). The second regression analysis was con-
Likewise, body surveillance scores (as measured ducted to assess the relative role of de-
by that subscale of the OBCS) were not signifi- mographic variables, social desirability, body
cantly associated with feelings of attraction and esteem, and self-esteem in predicting self-
pleasure (r = .09, p = .15) nor were they sig- objectification (as measured by Body Surveil-
nificantly associated with feelings of self-esteem lance subscale scores of the OBCS). Cor-
(r = .10, p = .11). Thus, Hypothesis 2 was not relations of the demographic variables with
supported. self-objectification scores revealed that age
Downloaded by [McMaster University] at 08:48 06 March 2015

and race/ethnicity were significant correlates.

Race/ethnicity and age were both dummy-
Hypothesis 3: Self-esteem and body coded so that White participants versus par-
esteem will predict, with negative beta ticipants of color were compared and those
weights, self-objectification (as measured aged 18 to 24 years old (of typical college age;
by body shame and body surveillance) Arnett, 2004) were compared to those aged
and social physique anxiety, after 25 years and older. Younger participants were
controlling for social desirability and more likely to have higher self-objectification
demographic variables, indicating that scores than were older participants, and White
self-esteem and body esteem may be participants were more likely to have higher
protective factors for self-objectification self-objectification scores than were partici-
and social physique anxiety pants who were members of racial and ethnic-
This hypothesis was tested through three minority groups. These were treated as control
multiple linear regression analyses using data variables along with social desirability scores
from all 350 participants. The first analy- and were entered in the first model. This re-
sis was conducted to assess the relative role sulted in an adjusted R2 of .087 and R2 change of
of demographic variables, social desirability .095, F(3, 342) = 11.920, p < .001. For the sec-
(as measured by the MCSDS), body esteem ond model, body esteem and self-esteem scores
(as measured by the BES), and self-esteem were entered, resulting in an adjusted R2 of .126
(as measured by the RSES) in predicting self- and R2 change of .044, F(2, 340) = 8.775, p <
objectification (as measured by the Body Shame .001. The results are shown in Table 3, where
subscale of the OBCS). Correlations of the one can see that age, social desirability, and
demographic variables with self-objectification body esteem were significant predictors in the
were first inspected. It was determined that final model, with scores on all scales being nega-
BMI was significantly positively correlated with tively related to self-objectification scores, while
self-objectification. Hence, BMI scores were controlling for age, race/ethnicity, and social de-
treated as a control variable along with social sirability.
desirability scores, and both were entered in The final regression analysis was conducted
the first model. This model resulted in an ad- to assess the relative role of demographic vari-
justed R2 of .097 and R2 change of .102, F(2, ables, social desirability, body esteem, and self-
344) = 19.638, p < .001. Body esteem and esteem in predicting social physique anxiety (as
self-esteem scores were entered in the second measured by the SPAS). Correlations of the de-
model, which resulted in an adjusted R2 of .317 mographic variables with the SPAS scores re-
and R2 change of .223, F(2, 342) = 56.499, p < vealed that BMI was significantly positively cor-
.001. Examination of Table 3 reveals that BMI, related with social physique anxiety. BMI and
 250 INTERNATIONAL JOURNAL OF SEXUAL HEALTH

TABLE 3. Regression Analysis of the Prediction of Body Shame, Body Surveillance, and Social Physique Anxiety

Predictor
Outcome Variable Variable B SEB β T

Body Shame Social Desirability −.00 .01 −.01 −0.12

BMI .03 .01 .15 3.14∗∗
Self-Esteem −.56 .14 −.21 −3.87∗∗∗
Body Esteem −.77 .11 −.38 −6.86∗∗∗
Body Surveillance Social Desirability −.04 .01 −.18 −3.31∗∗∗
Ethnicity .23 .15 .08 1.49
Age −.75 .21 −.18 −3.49∗∗∗
Self-Esteem .11 .15 .04 0.71
Body Esteem −.44 .11 −.24 −3.97∗∗∗
Social Physique Anxiety Social Desirability −.01 .01 −.06 −1.51
BMI .03 .01 .17 4.51∗∗∗
Self-Esteem −.22 .08 −.13 −2.88∗∗
Body Esteem −.74 .06 −.57 −12.61∗∗∗
Downloaded by [McMaster University] at 08:48 06 March 2015

Note. N = 350. BMI = body mass index. Body surveillance and body shame were measured by the Objectified Body Consciousness
Scale; social physique anxiety was measured by the Social Physique Anxiety Scale.
∗∗ p < .01. ∗∗∗ p < .001.

social desirability scores were thus treated as Some participants provided more than one
control variables and were entered in the first reason; thus, each reason was coded sepa-
model. This model resulted in an adjusted R2 rately. A total of 379 qualitative responses were
of .180 and R2 change of .185, F(2, 344) = coded. The interrater reliability analysis using
39.051, p < .001. For the second model, total the Kappa statistic was found to be .87 (p <
scores for body esteem and self-esteem were .001), which falls in the range of “almost per-
entered, resulting in an adjusted R2 of .533 and fect agreement” (Landis & Koch, 1977).
R2 change of .353, F(2, 342) = 99.614, p < Participants were not required to provide a
.001. The results are shown in Table 3, where response for the open-ended follow-up ques-
one can see that BMI, body esteem, and self- tions. Therefore, 156 participants provided a
esteem were significant predictors in the final total of 183 responses to the question assessing
model. BMI was positively associated with so- discomfort with the experience, and 182 partic-
cial physique anxiety, while body esteem and ipants provided a total of 196 responses to the
self-esteem were both negatively related to so- question assessing what about the experience
cial physique anxiety. BMI and social desirabil- was pleasurable. Note that participants were
ity were both controlled for. Taken collectively, able to answer either or both of the questions,
these results support Hypothesis 3. so there is some overlap between those who
responded to each question. Table 4 provides
a summary of the themes that emerged from
Exploratory Question these questions. In regard to the themes asso-
Thematic analysis was used to identify ciated with an uncomfortable experience, the
emerging themes from participant responses most frequent response (18.6%) was that par-
to the two open-ended follow-up questions ticipants did not feel uncomfortable in the situ-
(“What about the experience was uncomfort- ation. The next two most frequent themes were
able?” and “What about the experience was inappropriate staring or ogling and whether or
pleasurable?”). Specifically, each participant’s not the participant was in a romantic relation-
response was coded for its content indepen- ship with another person. Relatedly, a few par-
dently by the first and second authors of this ticipants noted they believed the person who
study. Coding was done by hand without uti- gave sexual attention (the “agent”) was in a ro-
lization of computer software. These codes mantic relationship with someone else, which
were then organized into overarching themes. also made the experience uncomfortable.
 S. J. GREGUS ET AL. 251

TABLE 4. Summary of Qualitative Themes in Response to the Because most women responded to both
Question ‘What About the Experience Was Uncomfortable?’
questions, the data show that in the majority of
Percentage of the cases (78.2%), those who felt the situation
Theme N Responses was uncomfortable were also the women who
Did not feel uncomfortable 34 18.6 reported there was nothing pleasurable about
Inappropriate staring or ogling 27 14.8 the situation. Likewise, those who felt the situ-
Committed romantic relationship 21 11.5 ation was pleasurable reported there was noth-
Found agent unappealing 16 8.7
Awkward gestures and/or sexual comments 15 8.2 ing uncomfortable about the situation. In an
Invasion of personal space 14 7.7 additional 13.8% of the cases, one of the ques-
Felt interrupted/bothered 12 6.6 tions (regarding either pleasure or discomfort)
Feeling objectified 11 6.0
Not knowing the agent 6 3.3
was left unanswered, making a comparison be-
Feeling unattractive or insecure with self 5 2.7 tween how they responded to the two questions
Age of agent (too old, too young) 5 2.7 impossible.
Miscellaneous 17 9.3
Total 183 100.0
Downloaded by [McMaster University] at 08:48 06 March 2015

Note. Some responses contained more than one theme. DISCUSSION

Summary and Discussion of Findings
Because this was reported on only a few oc- Hypothesis 1
casions, the two categories were collapsed into We hypothesized that the context where
a single theme. Comments relating to feeling sexual attention occurred would make a dif-
objectified emerged in 6% of the responses. In- ference in how positively or negatively the en-
terestingly, participants feeling unattractive or counter was experienced by participants. We
insecure with themselves emerged as a reason found no significant differences in the experi-
for discomfort in 2.7% of the cases. ence variables attributable to differences in con-
Table 5 provides a summary of the themes text. Although tests of between-subjects effects
associated with a pleasurable experience, the were not significant, it should be noted that the
most frequent theme (30.1% of responses) was lowest levels of pleasure occurred in the work
that nothing was pleasurable about the experi- context, whereas the highest levels of pleasure
ence. The following themes each represented occurred in the night-out and gym contexts.
approximately 15% of the responses: receiving Hence, receiving sexual attention in a context
attention in general, validating physical attrac- such as the workplace, where it is considered
tiveness, and an overall self-esteem boost. inappropriate, could potentially be associated
with greater displeasure with the experience
and with negative sequelae similar to those
TABLE 5. Summary of Qualitative Themes in Response to the
Question ‘What About the Experience Was Pleasurable?’ identified by Fredrickson and Roberts (1997).
Perhaps we failed to find significance due to
Percentage of
Theme N Responses
methodological errors, as will be discussed.
As noted, the participants were not evenly
Nothing was pleasurable 59 30.1 distributed across contexts. Less than 12%
Receiving attention 32 16.3
Validating attractiveness 30 15.3
answered questions about the gym context.
Self-esteem boost, general 30 15.3 Because the context questions were randomly
Appeal of agent 17 8.7 presented, this could be because more college
Receiving attention from someone other 5 2.6
undergraduate students engage in a night out
than significant other
Feeling wanted or desired 5 2.6 than in a session at the gym. However, it is
Having fun 4 2.0 also possible that women were less likely to ex-
Miscellaneous 14 7.1 perience sexual attention in the gym context.
Total 196 100.0
Perhaps if we had obtained larger numbers of
Note. Some responses contained more than one theme. participants in the gym context, we might have
 252 INTERNATIONAL JOURNAL OF SEXUAL HEALTH

found stronger evidence of differences in the esteem were both still significantly negatively
experience of receiving sexual attention based related to social physique anxiety and self-
on the context in which it occurred. objectification. In other words, participants with
higher self-esteem and body esteem were less
likely to engage in self-objectifying behaviors
Hypothesis 2 (e.g., they were less likely to feel bad about their
We hypothesized that participants who ex- bodies and monitor their appearance more of-
perienced sexual attention as positive would ten) and were less likely to feel anxious in
still show elevated levels of body shame and anticipation of their bodies being evaluated.
body surveillance due to the normative nature These findings extend prior research (Befort
of sexual objectification. This hypothesis was et al., 2001; Breines et al., 2008; McKinley,
not supported. Body shame and body surveil- 1998; Mercurio & Landry, 2008) by control-
lance were not related to pleasure experienced ling for the demographic variables of BMI, age,
as a result of the sexual attention. Prior research and race/ethnicity, as well as social desirabil-
Downloaded by [McMaster University] at 08:48 06 March 2015

has shown that women who viewed unwanted ity, while examining the roles of self-esteem
sexual attention as complimentary still showed and body esteem as potential protective fac-
increases in body dissatisfaction, body shame, tors for objectification. Importantly, this sug-
and self-surveillance (Calogero et al., 2009; gests that targeting young women’s self-esteem
Tiggemann & Boundy, 2008). Thus, it may be and body image may be important prevention
that the experience of receiving sexual atten- tools for managing future, unavoidable sexual
tion is more complex and multifaceted than was harassment.
captured by our data, and/or that methodolog-
ical errors prevented significant findings.
Exploratory question
There was much variation in how women
Hypothesis 3 experienced sexual attention, consistent with
We hypothesized that self-esteem and body findings from previous literature (Fairchild,
esteem would predict, with negative beta 2010; Liss et al., 2011). Agent characteris-
weights, self-objectification (as measured by tics of age, appeal (e.g., attractiveness), and
body shame and body surveillance) and so- whether the person was a stranger emerged
cial physique anxiety after controlling for as- as themes in response to both open-ended
sociated demographic variables and social de- questions. Additionally, inappropriate behavior
sirability. This hypothesis was supported. With (e.g., staring, gestures, sexual comments, invad-
regard to the associated demographic variables, ing personal space) were frequently endorsed
we found that BMI was significantly associated as characteristics that caused the experience to
with body shame and social physique anxiety, be more uncomfortable. Interestingly, partici-
such that participants with higher BMIs exhib- pants’ own feelings of unattractiveness or in-
ited greater amounts of body shame and so- security contributed to making the experience
cial physique anxiety. In addition, participant uncomfortable. These later findings are con-
age and race/ethnicity were associated with sistent with the results of Fuller-Tyszkiewicz,
body surveillance, such that younger partici- Reynard, Skouteris, and McCabe (2012), who
pants (those aged 18–24 years old) and those found that feeling unattractive in a situation
who identified as White had higher levels of as well as being in the presence of someone
body surveillance. Our results suggest that par- they did not know made women more self-
ticipants at the greatest risk for engaging in conscious.
self-objectifying behaviors are White women In regard to positive reactions, many
younger than the age of 25 years with higher women felt their physical attractiveness was
BMIs. When these differences were controlled validated by the sexual attention. Some par-
for, it was found that self-esteem and body ticipants indicated that one way to judge the
 S. J. GREGUS ET AL. 253

quality of their looks was by whether or not findings are not necessarily generalizable to
they received sexual attention. A number of women of color or to lesbian or bisexual
women indicated that receiving sexual atten- women. Relatedly, as this was a convenience
tion from another person, whether wanted or sample, we cannot generalize the results to the
unwanted, tended to make them feel good population at large, as would be the case with
about themselves. Although the quantitative a representative sample. Because our results
data indicated that 80.5% of the participants showed significant correlations between body
felt complimented by sexual attention (even if surveillance and some demographic variables
it was unwanted), the responses to the open- (age, race/ethnicity, BMI), it is possible that find-
ended questions suggest that a more complex ings may differ depending on the characteristics
process might have occurred. More than half of of the participant sample. Similarly, the con-
the participants who responded to open ended texts examined in our study were designed to be
questions indicated a myriad of reasons why the representative of a typical college woman’s ex-
experience was uncomfortable. One interpre- periences. Researchers using a sample of non-
Downloaded by [McMaster University] at 08:48 06 March 2015

tation of these apparently conflicting findings is college women may benefit from revising or
that participants have internalized society’s ob- expanding the contexts examined. Further, as
jectification of their bodies to such an extent previously mentioned, responses to the CSSAQ
that they actually do feel valued when they are were not evenly distributed among the con-
objectified; in other words, their self-worth is texts. Thus, future research with college women
tied up in how they look. This interpretation should be designed to obtain larger numbers of
is consistent with the results of Fischer et al. participants in those contexts that were not as
(2011) and Liss et al. (2011) and is concern- represented in this study—particularly the gym
ing for two reasons. First, even though many context. This will help determine whether there
women reported feeling confident or validated are in fact differences in the experience of re-
after this experience, many other women felt ceiving sexual attention based on the context
that there was nothing positive about it and felt in which it is received. Also, because the two
objectified. Second, it may thus be likely that open-ended questions were optional, there is
many women are self-objectifying to an extent the possibility of a response bias for those par-
where they perceive their self-worth to depend ticipants who chose to answer these questions
on whether or not they are “hot” enough to versus those who did not.
get a man’s attention (APA, 2007). Some pre- One strength of the current study is that
vious scholars have suggested that embracing to our knowledge, this is the first time that
a sexualized female image is an empowering women’s reactions to sexual attention have
component of a third wave of feminism (Baum- been studied within 48 hr of the receipt of
gardner & Richards, 2004). However, our data that sexual attention. This timeframe was cho-
as well as the findings of Liss et al. (2011) do sen to allow a weekend to pass by, given that
not support this notion. much sexual attention is received on weekend
outings. This has the benefit of shortening the
typical time period after which researchers ask
Limitations and Future Directions participants to recall their feelings and thoughts
The current study has a number of limita- about sexual attention, which may increase the
tions. First, we neglected to collect data that accuracy of recall.
would have enabled the calculation of a re- Limitations of the CSSAQ include only the
sponse rate. We also do not know how length “fair” level of reliability (using the Ponterotto &
of the questionnaires affected the response rate, Ruckdeschel, 2007, criteria) for the feelings of
other than the fact that seven participants did self-esteem factor and the need for confirma-
not complete the CSSAQ in its entirety. Sec- tory factor analysis of the two-factor structure.
ond, the sample included primarily White, het- Because this instrument was exploratory and
erosexual, college-age women; as such, our developed for the current study, it could also
 254 INTERNATIONAL JOURNAL OF SEXUAL HEALTH

benefit from investigation of its construct va- pleasure and a boost in self-esteem. In addi-
lidity, including convergent, divergent, concur- tion, although the context results did not at-
rent, and predictive validity by assessing cor- tain statistical significance in our sample, the
relations with other instruments. Alternatively, role of context in regard to how women expe-
given its limitations, other ways to assess the role rience sexual attention certainly deserves fur-
of context in the experience of sexual attention ther investigation. Using improved methods,
should be explored. investigators should assess whether in the
Because this research was cross-sectional, workplace context sexual attention may be
causal relationships cannot be assessed. A lon- experienced as personally disruptive and in-
gitudinal prospective design could be employed appropriate. This could be compared to social
to assess whether or not body esteem and self- contexts of heightened sexuality, such as in a
esteem do indeed confer protection against self- club or bar, where it might be experienced as
objectification. The data collected are also sub- pleasurable and affirming. The findings also sug-
ject to some retrospective bias, as we did not gest that women may have come to expect to be
Downloaded by [McMaster University] at 08:48 06 March 2015

collect ratings during the experience, but rather objectified. Hence, it is possible that they have
up to 48 hr after the experience took place. A internalized the notion that their self-worth is
more accurate assessment of women’s experi- related to their physical attractiveness and that
ences of sexual attention would occur immedi- this physical attractiveness can be measured by
ately after the event took place, using an expe- how much sexual attention they receive from
rience sampling procedure aided by personal others. Given the finding that 15.3% of the
digital assistants or cell phones. women across all four contexts who responded
Additional studies are needed to examine to the exploratory questions reported that their
contextual and individual difference variables self-esteem increased, it may be the case that
in a sample of lesbian and bisexual women, as even in situations where sexual attention was
the women in the current sample were predom- deemed inappropriate, a significant percentage
inantly heterosexual. Similarly, future research of women still reported receiving a self-esteem
should assess whether or not these findings hold boost.
if the agent of the sexual attention is another Additionally, our findings suggest that the
women (e.g., in a lesbian bar or club). Addi- women most likely to engage in self-objectifying
tionally, because the Enjoyment of Sexualiza- behaviors were White women younger than the
tion Scale (Liss et al., 2011) was not published age of 25 with higher BMIs. However, when
at the time of our data collection, future re- controlling for the variables of race and BMI,
search should incorporate this scale into the as- both self-esteem and body esteem emerged
sessment of individual difference variables and as potentially protective factors against self-
contextual factors relating to experiences of sex- objectifying behaviors (body surveillance and
ual objectification. For example, research could body shame). Though our research could not
examine whether or not differences exist in the establish causality, it may be that the more posi-
level of enjoyment of sexualization for partic- tively a woman feels about herself and her body,
ipant women depending on the demograph- the less likely she is to worry about what oth-
ics of the target. Further, relationships between ers think of her or to seek appearance valida-
enjoyment of sexualization, body esteem, self- tion from outside sources. Thus, these findings
esteem, body shame, and body surveillance may have implications for programs that target
should also be assessed. young women’s self-esteem and body image. At
the very least, such programs need to be pur-
sued and evaluated, because they might pro-
Summary and Conclusions tect young women from experiencing the neg-
Our results showed that there is variability ative effects from unwanted sexual attention as
in how individual women respond to sexual at- well as from culture’s standards for beauty and
tention, from feeling uncomfortable to feeling power.
 S. J. GREGUS ET AL. 255

ACKNOWLEDGEMENTS Psychology, 24, 349–354. doi:10.1037/h00

47358
The authors would like to extend a sincere
Demo, D. H. (1985). The measurement of
thank you to John Rummell who generously cre-
self-esteem: Refining our methods. Jour-
ated the Web site used for data collection.
nal of Personality and Social Psychol-
ogy, 48, 1490–1502. doi:10.1037/0022-
3514.48.6.1490
REFERENCES
Esacove, A. W. (1998). A diminishing of self:
American Psychological Association. (2007). Women’s experiences of unwanted sex-
Report of the APA Task Force on the Sexu- ual attention. Health Care for Women In-
alization of Girls. Washington, DC: Author. ternational, 19, 181–192. doi:10.1080/073
Arnett, J. J. (2004). Emerging adulthood: The 993398246359
winding road from late teens through the Fabrigar, L. R., Wegener, D. T., MacCallum,
twenties. New York, NY: Oxford University R. C., & Strahan, E. J. (1999). Evaluating the
Downloaded by [McMaster University] at 08:48 06 March 2015

Press. use of exploratory factor analysis in psycho-

Baumgardner, J., & Richards, A. (2004). Grass- logical research. Psychological Methods, 4,
roots: A field guide for feminist activism. New 272–299. doi:10.1037/1082-989X.4.3.272
York, NY: Farrar, Straus and Giroux. Fairchild, K. (2010). Context effects on
Befort, C., Robinson-Kurpius, S. E., Hull-Banks, women’s perceptions of stranger harass-
E., Nicpon, M. F., Huser, L., & Sollen- ment. Sexuality and Culture, 14, 191–216.
berger, S. (2001). Body image, self-esteem, doi:10.1007/s12119-010-9070-1
and weight-related criticism from romantic Fairchild, K., & Rudman, L. A. (2008). Ev-
partners. Journal of College Student Devel- eryday stranger harassment and women’s
opment, 42, 407–419. objectification. Social Justice Research,
Breines, J. G., Crocker, J., & Garcia, J. A. 21, 338–357. doi:10.1007/s11211-008-007
(2008). Self-objectification and well-being 3-0http://www.stopstreetharassment.org/wp-
in women’s daily lives. Personality and content/uploads/2011/04/KimberlyFairchild
Social Psychology Bulletin, 34, 583–598. StreetHarassarticle.pdf
doi:10.1177/0146167207313727 Fischer, A. R., Bettendorf, S. K., & Wang,
Calogero, R. M. (2004). A test of objec- Y. (2011). Contextualizing sexual objecti-
tification theory: The effect of the male fication. The Counseling Psychologist, 39,
gaze on appearance concerns in college 127–139. doi:10.1177/00110000103811
women. Psychology of Women Quarterly, 41
28, 16–21. doi:10.1111/j.1471-6402.2004. Fisher, R. J., & Katz, J. E. (2000). Social-
00118.x desirability bias and the validity of self-
Calogero, R. M., Herbozo, S., & Thompson, J. K. reported values. Psychology & Marketing, 17,
(2009). Complimentary weightism: The po- 105–120. doi:0742-6046/00/020105-16
tential costs of appearance-related commen- Franzoi, S. L., & Shields, S. A. (1984).
tary for women’s self-objectification. Psy- The Body Esteem Scale: Multidimensional
chology of Women Quarterly, 33, 120–133. structure and sex differences in a col-
doi:10.1111/j.1471-6402.2008.01479.x lege population. Journal of Personality As-
Cochrane, C. C., Frazier, P. A., & Olson, sessment, 48, 173–178. doi:10.1207/s1532
A. M. (1997). Predictors of responses to 7752jpa4802 12
unwanted sexual attention. Psychology of Fredrickson, B. L., & Roberts, T. (1997).
Women Quarterly, 21, 207–226. doi:10. Objectification theory: Toward understand-
1111/j.1471-6402.1997.tb00109.x ing women’s lived experiences and mental
Crowne, D. P., & Marlowe, D. (1960). A health risks. Psychology of Women Quar-
new scale of social desirability independent terly, 21, 173–206. doi:10.1111/j.1471-
of psychopathology. Journal of Consulting 6402.1997.tb00108.x
 256 INTERNATIONAL JOURNAL OF SEXUAL HEALTH

Fuller-Tyszkiewicz, M., Reynard, K., Sk- Principles, advances, and applications.

outeris, H., & McCabe, M. (2012). An The Counseling Psychologist, 34, 684–718.
examination of the contextual determi- doi:10.1177/0011000006286347
nants of self-objectification. Psychology of Kurtz, J. E., Tarquini, S. J., & Iobst, E. A. (2008).
Women Quarterly, 36, 76–87. doi:10.1177/ Socially desirable responding in personality
0361684311426721 assessment: Still more substance than style.
Gelfand, M. J., Fitzgerald, L. F., & Drasgow, F. Personality and Individual Differences, 45,
(1995). The structure of sexual harassment: 22–27. doi:10.1016/j.paid.2008.02.012
A confirmatory analysis across cultures and Landis, J. R., & Koch, G. G. (1977). The
settings. Journal of Vocational Behavior, 47, measurement of observer agreement for
164–177. doi:10.1006/jvbe.1995.1033 categorical data. Biometrics, 33, 159–174.
Gordon, M. K. (2008). Media contribu- doi:10.2307/2529310
tions to African American girls’ focus on Liss, M., Erchull, M. J., & Ramsey, L. R.
beauty and appearance: Exploring the con- (2011). Empowering or oppressing? De-
Downloaded by [McMaster University] at 08:48 06 March 2015

sequences of sexual objectification. Psychol- velopment and exploration of the Enjoy-

ogy of Women Quarterly, 32, 245–256. ment of Sexualization Scale. Personality
doi:10.1111/j.1471-6402.2008.00433.x and Social Psychology Bulletin, 37, 55–68.
Greenleaf, C., & McGreer, R. (2006). Disor- doi:10.1177/0146167210386119
dered eating attitudes and self-objectification McKinley, N. M. (1998). Gender differences
among physically active and sedentary fe- in undergraduates’ body esteem: The me-
male college students. Journal of Psychology: diating effect of objectified body con-
Interdisciplinary and Applied, 140, 187–198. sciousness and actual/ideal weight discrep-
doi:10.3200/JRLP.140.3.187-198. ancy. Sex Roles, 39, 113–123. doi:10.
Harper, B., & Tiggemann, M. (2008). The effect 1023/A:1018834001203
of thin ideal media images on women’s self- McKinley, N. M., & Hyde, J. S. (1996). The
objectification, mood, and body image. Sex Objectified Body Consciousness Scale:
Roles, 58, 649–657. doi:10.1007/s11199- Development and validation. Psychol-
007-9379-x ogy of Women Quarterly, 20, 181–215.
Hart, E. A., Leary, M. R., & Rejeski, W. J. (1989). doi:10.1111/j.1471-6402.1996.tb00467.x
The measurement of social physique anxiety. Mercurio, A. E., & Landry, L. J. (2008). Self-
Journal of Sport & Exercise Psychology, 11, objectification and well-being: The impact
94–104. of self-objectification on women’s overall
Hurt, M. M., Nelson, J. A., Turner, D. L., Haines, sense of self-worth and life satisfaction. Sex
M. E., Ramsey, L. R., Erchull, M. J., & Liss, Roles, 58, 458–466. doi:10.1007/s11199-
M. (2007). Feminism: What is it good for? 007-9357-3
Feminine norms and objectification as the Moffitt, L. B., & Szymanski, D. M. (2011).
link between feminist identity and clinically Experiencing sexually objectifying environ-
relevant outcomes. Sex Roles, 57, 355–362. ments: A qualitative study. The Counsel-
doi:10.1007/s11199-007-9272-7 ing Psychologist, 39, 67–106. doi:0.1177/
Jang, M. H., & Lee, G. (2013). Body image 0011000010364551
dissatisfaction as a mediator of the associa- Myers, D. G. (2008). Exploring psychology in
tion between BMI, self-esteem and mental modules (7th ed.). New York, NY: Worth.
health in early adolescents: A multiple-group National Heart, Lung, and Blood Institute.
path analysis across gender. Journal of the (2013). Assessing your weight and health risk.
Korean Academy of Nursing, 43, 165–175. Retrieved from http://www.nhlbi.nih.gov/
doi:10.4040/jkan.2013.43.2.165 health/public/heart/obesity/lose wt/risk.htm
Kahn, J. H. (2006). Factor analysis in counseling Nordholm, L. A. (1974). A note on the reli-
psychology research, training, and practice: ability and validity of the Marlowe-Crowne
 S. J. GREGUS ET AL. 257

Scale of Social Desirability. The Journal of So- 7-item Social Physique Anxiety Scale using
cial Psychology, 93, 139–140. doi:10.1080/ a test–retest method. Measurement in Physi-
00224545.1974.9923140 cal Education and Exercise Science, 8, 57–62.
O’Connor, B. P. (2000). SPSS and SAS programs doi:10.1207/s15327841mpee0802 1
for determining the number of components Shloim, N., Hetherington, M. M, Rudolf, M.,
using parallel analysis and Velicer’s MAP test. & Feltbower, R. G. (2013). Relationship
Behavior Research Methods, Instrumentation, between body mass index and women’s
and Computers, 32, 396–402. body image, self-esteem and eating be-
Pearl, R. L., & Puhl, R. M. (2013). Mea- haviours in pregnancy: A cross-sectional
suring internalized weight attitudes across study. Journal of Health Psychology. doi:
body weight categories: Validation of 10.1177/1359105313502568
the Modified Weight Bias Internaliza- Silberstein, L. R., Striegel-Moore, R. H., Timko,
tion Scale. Body Image, 13, 113–117. C., & Rodin, J. (1988). Behavioral and
doi:10.1016/j.bodyim.2013.09.005 psychological implications of body dissatis-
Downloaded by [McMaster University] at 08:48 06 March 2015

Ponterotto, J. G., &.Ruckdeschel, D. E. (2007). faction: Do men and women differ? Sex
An overview of coefficient alpha and a relia- Roles, 19, 219–232. doi:10.1007/BF00290
bility matrix for estimating adequacy of inter- 156
nal consistency coefficients with psycholog- Szymanski, D. M., & Henning, S. L. (2007).
ical research measures. Perceptual & Motor The role of self-objectification in women’s
Skills, 105, 997–1014. depression: A test of objectification theory.
Preacher, K. J., & MacCallum, R. C. (2003). Sex Roles, 56, 45–53. doi:10.1007/s11199-
Repairing Tom Swift’s electric factor analy- 006-9147-3
sis machine. Understanding Statistics, 2, 13– Tabachnick, B. G., & Fidell, L. S. (2007). Using
43. multivariate statistics. Boston, MA: Allyn &
Rosenberg, M. (1965). Society and the ado- Bacon.
lescent self-image. Princeton, NJ: Princeton Tiggemann, M., & Boundy, M. (2008). Effect
University Press. of environment and appearance compliment
Rudiger, J. A., & Winstead, B. A. (2013). on college women’s self-objectification,
Body talk and body-related co-rumination: mood, body shame, and cognitive per-
Associations with body image, eating formance. Psychology of Women Quar-
attitudes, and psychological adjustment. terly, 32, 399–405. doi:10.1111/j.1471-
Body Image, 4, 462–471. doi:10.1016/j. 6402.2008.00453.x
bodyim.2013.07.010 Wilson, R. E., Latner, J. D., & Hayashi,
Schoenefeld, S. J., & Webb, J. B. (2013). Self- K. (2013). More than just body weight:
compassion and intuitive eating in college The role of body image in psychological
women: Examining the contributions of dis- and physical functioning. Body Image, 4,
tress tolerance and body image acceptance 644–647. doi:10.1016/j.bodyim.2013.04.
and action. Eating Behavior, 4, 493–496. 007
doi:10.1016/j.eatbeh.2013.09.001 Yoder, J. D. (2007). Women and gender: Making
Scott, L. A., Burke, K. L., Joyner, A. B., & Brand, a difference (3rd ed.). Cornwall-on-Hudson,
J. S. (2004). Examining the stability of the NY: Sloan.