The Effects of the Sexualization of Female Video Game Characters on Gender Stereotyping and Female Self-Concept

Abstract

The present study utilized an experimental design to investigate the short term effects of exposure to sexualized female video
game characters on gender stereotyping and female self-concept in emerging adults. Bussey and Bandura’s (1999) social cognitive theory of gender development and differentiation was used to explicate this relationship. Undergraduate
students (N = 328) at a large U.S. Southwestern university participated in the study. Students were randomly assigned to play a “sexualized”
heroine, a “non-sexualized” heroine, or no video game; then completed an online questionnaire. Female self-efficacy was negatively
affected by game play with the sexualized female character. Results cautiously suggest that playing a sexualized video game
heroine unfavorably influenced people’s beliefs about women in the real world.

Full text

ORIGINAL ARTICLE
The Effects of the Sexualization of Female Video
Game Characters on Gender Stereotyping
and Female Self-Concept
Elizabeth Behm-Morawitz &Dana Mastro
Published online: 1 August 2009
#Springer Science + Business Media, LLC 2009
Abstract The present study utilized an experimental
design to investigate the short term effects of exposure to
sexualized female video game characters on gender stereo-
typing and female self-concept in emerging adults. Bussey
and Bandura’s(1999) social cognitive theory of gender
development and differentiation was used to explicate this
relationship. Undergraduate students (N=328) at a large U.S.
Southwestern university participated in the study. Students
were randomly assigned to play a “sexualized”heroine, a
“non-sexualized”heroine, or no video game; then completed
an online questionnaire. Female self-efficacy was negatively
affected by game play with the sexualized female character.
Results cautiously suggest that playing a sexualized video
game heroine unfavorably influenced people’s beliefs about
women in the real world.
Keywords Video games .Gender roles .
Gender stereotyping .Media effects .
Social cognitive theory
Introduction
The present study experimentally investigates the short
term effects of exposure to sexualized female video game
characters on gender stereotyping and female self-concept
among U.S. college students. Social cognitive theory of
gender development and differentiation (Bussey and
Bandura 1999) is used to explicate the process through
which mass media images influence gender-related beliefs
and self-concept. Although this study is based in the U.S.,
the findings are of consequence to scholars, parents,
gamers, and the gaming industry not only in the U.S. but
also abroad. The gender images under investigation here are
mass produced and distributed in many countries across the
world. Accordingly, the implications of exposure to the
images and themes characterized in video games take on
global importance. As such, the goal of this study is to shed
light on the impact of exposure to these characterizations on
young adult users.
Over the past three decades, the video game market has
developed into a $10 billion a year industry in the U.S.
(CNNmoney.com 2006). As the popularity of video games
has increased, the profile of the gamer has shifted,
reflecting the wider variety of consumers that play video
games today. The image that comes to mind when picturing
a video gamer should no longer be an adolescent, or even
teenage, boy, as women and adults are playing games in
greater numbers. An industry survey indicates that 40% of
all game players in the U.S. are female (ESA 2009), and
80% of girls (grades 4–12) report playing games in their
homes (Walsh et al. 2005). Further, the average age of video
game players is 35 years old (ESA 2009), demonstrating that
gaming is no longer just a childhood pastime.
Thus, the effects of video game play on both men and
women should not be trivialized. The growing popularity of
E. Behm-Morawitz (*)
Department of Communication, University of Missouri-Columbia,
115 Switzler Hall,
Columbia, MO 65211, USA
e-mail: behmmorawitze@missouri.edu
D. Mastro
Department of Communication, University of Arizona,
211 Communication Building,
Tucson, AZ 85721, USA
Sex Roles (2009) 61:808–823
DOI 10.1007/s11199-009-9683-8

this medium has provoked concern among parents and
advocacy groups regarding the potentially harmful effects
of playing video games. Although this has led to significant
research on the influence of violent video games on
subsequent aggression (e.g., Gentile and Anderson 2003;
Sherry 2001), the effects of exposure to many other content
features have largely been ignored. This is particularly
disconcerting when considering the stereotypical manner in
which females are commonly portrayed in these games.
Media effects theories and empirical research support the
idea that individuals can and do learn about gender roles as
well as gender-related attitudes and beliefs from mass
media offerings including: television and advertising (e.g.,
Davidson et al. 1979; Herrett-Skjellum andAllen 1996;
McGhee and Freuh 1980; Morgan 1987; Signoriellli 1989)
as well as magazines (e.g., Carpenter 1998; Hatoum and
Belle 2004; Morrison et al. 2004). Moreover, exposure to
idealized images of the female body in the media has been
shown to negatively affect girls’and women’s general
feelings of self worth. Specifically, exposure to idealized
images of the female body affects self-esteem and self-
efficacy (Bessenoff 2006; Clay et al. 2005; Hawkins, et al.
2004). Taken together, this research indicates that media
use has a measurable influence on gender roles and gender-
based cognitions. Although little to no empirical inves-
tigations have examined the effects of gender stereotypes in
video games on video game players, the findings based on
traditional media point to the gender-related outcomes that
can be anticipated with new entertainment technology. As
such, the present study examines the effects of exposure to
sexualized portrayals of women in video games on
individuals’gender attitudes and beliefs and on female
self-esteem and self-efficacy.
Portrayals of Females in Video Games
In terms of the characterization of women in video games,
content analytic research indicates that females are vastly
underrepresented in popular video games and are often
hypersexualized when depicted (Beasley and Standley
2002; Dietz 1998; Glaubke et al. 2001; Ivory 2006; Miller
and Summers 2007). Not unlike other media, video games
offer a narrow range of roles to female characters. Overall,
research suggests that when female characters appear in
video games they most often serve as victims or prizes
(Provenzo 1991) and occupy stereotypical gender roles
such as brazenly sexualized beings and objects of sexual
desire (Miller and Summers 2007).
Glaubke, et al.’s(2001) work examining the content
features of video games demonstrated that “female sexuality
[is] often accentuated with highly revealing clothing”(p.14);
underscoring the prominence and perpetuation of the
sexualized image of females in video games. Further,
Beasley and Standley (2002) found that 70% of female
characters in Mature-rated video games and 46% of
female characters in Teen-rated video games were depicted
with abundant cleavage, 86% of female characters were
portrayed wearing clothing with low/revealing necklines, and
48% of female characters were dressed in outfits with no
sleeves. This is in contrast to only 22% of male characters
represented in clothing with no sleeves and 14% of male
characters wearing clothing with low/revealing necklines.
Moreover, females were twice as likely as males to be shown
wearing revealing clothing.
In addition, the vast majority of female characters have
been found to be non-playable, meaning that they cannot be
played by the gamer (Miller and Summers 2007)—thus
underscoring their secondary and exiguous status. When
playable female characters do appear in video games, they
are typically overtly sexualized and portrayed wearing
promiscuous dress and engaging in seductive acts (Dietz
1998). Notably, however, these characters are often high
status, powerful characters, such as the heroic figures more
commonly associated with men. Although this type of
female character fits the normative characteristics of an
action hero (i.e. male action hero) by demonstrating
strength, speed, intellect, and independence (Richard and
Zaremba 2005), her sexuality is her defining feature,
relegating her status to that of an object to be gazed upon
(Mikula 2003). In other words, her role as an action hero is
tied to her sexuality and body.
Thepresentstudyisconcernedwithmeasuringthe
influence of exposure to such videogame content.
Specifically, this study investigates the effect of playing a
highly sexualized video game heroine compared with a less
sexualized (i.e. falling on the low end of the sexualization
continuum) heroine or no exposure at all. For ease of
discussion, the characters used in this study will be referred
to as either sexualized or non-sexualized. Although, strictly
speaking, the “sexualized”character is conceptualized as
being highly sexualized and the “non-sexualized”character is
conceptualized as being low in terms of her sexualization.
Generally, sexualization is defined by the degree to which the
female body is exposed and ‘idealized’with larger breasts and
a smaller waist.
Considering the scarcity of females in video games, the
mere presence of a prominent female video game character
may trigger attention and serve as a motivator for
individuals to adopt congruent gender-related beliefs
(Bandura 1986). In particular, gamers may adopt beliefs
and standards that are in line with these sexualized
portrayals, resulting in the desire to be like the characters
(among women) and to judge self and others based upon
the characters (among both men and women). Although it
could be the case that exposure to images of strong,
powerful female heroines in video games may empower
Sex Roles (2009) 61:808–823 809

girls and women through these characters’embodiment of
female success, strength, and intelligence, the overwhelming
presence of female sexualization is likely to diminish positive
effects that may emerge. The blending of stereotypical and
counter-stereotypical attributes into female heroines, like Lara
Croft, complicates the task of understanding the influence of
such portrayals on media consumers.
On the one hand, such female characters are strong,
bold, intelligent, and independent, but on the other hand
they are ‘made-up’(with makeup and clothing), sexualized,
and objectified (Inness 1998). These latter characteristics
are what maintain females’vulnerable and non-threatening
status (Inness 1998). Thus, the powerful role of the female
heroine is diminished by the emphasis on her physical
feminine appearance.
In particular, it is the sexualization of female characters
in video games that seems likely to negatively influence
video game players’perceptions of self and women in
society. Based on the assumptions of social cognitive
theory of gender development and differentiation (Bussey
and Bandura 1999), exposure to sexually objectified
women and girls in video games would be expected to
influence social perceptions about gender and women’s
gendered self-concept.
Theoretical Framework
Social cognitive theory of gender development and differen-
tiation (Bussey and Bandura 1999) offers a framework for
understanding how exposure to mediated models (e.g.,
video game characters) may impart gender lessons to
consumers, influencing their attitudes and beliefs about
gender and their own gender-related self-concept. According
to Bussey and Bandura (1999) media messages serve as one
source for “the development of gender-linked knowledge
and competencies,”(p. 686) influencing perceptions of
appropriate gender-based conduct, normative gender roles,
self-evaluative gender-specific standards, and even self-
efficacy beliefs. Accordingly, video game portrayals of the
female body, for example, may be used to help form an
individual’s social and moral standards about gender-
appropriate dress, ideal female body-type, and even evalua-
tions of female (self)worth. It is the symbolizing capability
that allows one to observe and cognitively organize mediated
symbols so they having meaning for that individual. Further,
Bandura (1986) argues that media messages are a powerful
source of information in our culture, greatly expanding
consumer’s exposure to issues, images, and phenomenon
they would otherwise never encounter in their daily lives.
The pervasiveness and unique role of media in both
reflecting and creating culture (Bandura 1986)suggeststhat
media may be an important source of learning about gender
norms and values.
Research demonstrates that there is a significant rela-
tionship between media exposure to modeled gender
stereotypes and individuals’stereotypical gender role
beliefs and expectations (e.g., Herrett-Skjellum and Allen
1996; Signorielli 1993). “Modeling is one of the most
pervasive and powerful means of transmitting values,
attitudes, and patterns of thought and behavior”(Bandura
1986, p. 686). As evidenced by content analytic work (see
above discussion), video games often present exaggerated
gender stereotypes (e.g., hypersexualized female body),
communicate unrealistic standards for women’s bodies, and
encourage the treatment of women as sexual objects. From a
social cognitive perspective, then, it would be expected that
exposure to such sexualized, stereotypical portrayals of female
characters has the potential to diminish self-esteem and self-
efficacy in female players. Indeed, traditional media research
findings indicate that exposure to sexualized images of
women can negatively influence female self-concept (Lavine
et al. 1999). More specifically, viewing idealized images of
female bodies has been repeatedly linked to lower feelings
of worth in women and girls. This may, at least in part, be
explained by research which suggests that acquisition of
gender stereotypes is related to lower feelings of efficacy,
and this decreased self-efficacy is a function of holding
stereotypic beliefs—not a person’s actual capabilities
(Bussey and Bandura 1999). This is particularly true of
females, (rather than males), due to the differential value
ascribed to male versus female gender roles. Typically,
gender stereotypes of men (e.g., representations of men as
strong) are more positive and have more desirable out-
comes (e.g., high status and dominance) than gender
stereotypes of women (e.g., representations of women as
sexual objects) (Berscheid 1993). Thus, exposure to
sexualized images of women in video games may have
damaging effects on women’s self-efficacy (i.e. belief in their
abilitytoaccomplishthingsinlife).
Additionally, findings from Funk and Buchman (1996)
lend support to the assertion that video game play may be
negatively associated with self-esteem in women. Their
investigation into the influence of video game play on
adolescent self-concept found gaming to be negatively
associated with self-esteem for girls. For girls, then, as
video game playing increased their self-esteem decreased.
However, Funk and Buchman only examined the influence
of time spent playing video games, in general, and did not
specifically test the effects of gender portrayals in video
games. Further, they did not identify the causal mechanisms
that may be at work. The present study will help elucidate
the processes underlying this relationship and lend insights
into why female self-esteem may be negatively associated
with video game play by examining the effects of playing
specific types of female characters. It stands to reason that
the type of character played may have an influence on the
810 Sex Roles (2009) 61:808–823

self-esteem and self-efficacy in female gamers. More
specifically, from the perspective of social cognitive theory
of gender development and differentiation, it is expected
that playing the sexualized female character will result in
lower self-esteem and self-efficacy than playing the non-
sexualized female character or no video game at all. Thus,
the following hypotheses are predicted.
H1a: Among female players, video game condition
will predict player self-esteem, such that playing a
sexualized female character will result in lower self-
esteem, in comparison to playing a non-sexualized
female character or no video game.
H1b: Among female players, video game condition
will predict player self-efficacy, such that playing a
sexualized female character will result in lower self-
efficacy, in comparison to playing a non-sexualized
female character or no video game.
The effects of exposure to such images are likely not
isolated to women’s conceptualizations of self. Of equal
consequence are the implications of gaming on men’s
and women’s attitudes and beliefs about women in
general. Given the types of characterizations of women
found in video games, it is not unreasonable to suggest
that video gaming may increase male and female players’
stereotyping toward women and girls, and even serve to
encourage and/or justify negative and unfair treatment of
women.
Gender stereotypes are generally defined as “beliefs
about what it means to be female or male …[including]
information about physical appearance, attitudes and
interests, psychological traits, social relations, and occupa-
tions”(Golombok and Fivush 1994, p. 17). Deaux and
Lewis (1983) have organized these components of gender
stereotypes into four dimensions: traits, physical character-
istics, role behaviors, and occupations. First, when it comes
to making gender-related judgments based on traits, people
often base these evaluations on attributes such as intelli-
gence, aggression, and emotionality (Glick and Fiske 1999;
Spence and Hahn 1997). In this case, women are
stereotypically judged to be less intelligent, more emotional,
and less aggressive, for example, than men. Second, in
terms of physical appearances women are expected to be
youthful, beautiful, soft, voluptuous but thin, and
feminine in appearance. No such expectations hold for
men. Third, different role behaviors are stereotypically
expected of men and women. For instance, traditionally,
women are expected to take on nurturing roles. Last,
there are stereotypic beliefs about the different occupa-
tions that men and women should hold. Women are more
likely to be expected to occupy jobs that involve care-
taking, lower status jobs (e.g., secretarial positions), and
jobs that do not involve manual labor.
In the context of the current study, the insights offered
from Deaux and Lewis’s dimensions of gender stereo-
types alongside research on gender roles suggest that
beliefs about: (a) women’s body and appearance (Cash et al.
1997; Deaux 1985); (b) roles in society (Spence and
Helmreich 1978); (c) cognitive capabilities (Deaux 1985;
Glick and Fiske 1999;Spenceetal.1974); and (d) physical
capabilities (Glick and Fiske 1999;Spenceetal.1974), each
should be taken into consideration when examining
gender-based attitudes and stereotypes. Research demon-
strates that these dimensions are relevant and long-
standing domains of gender perceptions and identity.
Moreover, recent research on gender-related constructs
and measures further supports the idea that gender
attitudes and beliefs are multidimensional (Ashmore et al.
1995;Kingetal.1997).
Not surprisingly, the media have been found to serve as
one source of learning gender-related attitudes and beliefs,
including those linked to gender stereotyping. Content
analytic research has consistently demonstrated that the
media often adhere to traditional gender stereotypes in the
presentation of men and women (Dietz 1998; Glaubke et al.
2001; Mastro and Behm-Morawitz 2005: Signorielli 1989;
Signorielli and Bacue 1999; Signorielli and Kahlenberg
2001). Eagly (1987) argues that such culturally transmitted
gender stereotypes influence individuals’gender role beliefs
and behaviors. In other words, one’s gender expectations and
behaviors are, at least in part, determined by society’s
endorsement of gender stereotypes. Indeed, Michael
(1982) conducted a longitudinal survey examining the
influence of television consumption on adolescents’
gender stereotyping. His findings indicated that television
exposure was positively related to traditional gender
stereotyping in adolescents. Zuckerman, et al.’s(1980)
survey research yielded similar results for girls, revealing
an association between television consumption and gender
stereotyping.
Additionally, Ward and Friedman (2006) experimentally
examined the effects of exposure to stereotypical media
depictions of gender and found that adolescents who
viewed portrayals of women as sex objects (for a period
of 12 min) were more likely to report greater adherence to
gender stereotypes and to condone stereotype-consistent
treatment of women. Thus, the media’s propensity to
emphasize women’s sexuality and to treat women as sex
objects may have real world consequences when it comes
to individuals’judgments about women. This suggests that
exposure to the sexualized images of women in video
games may result in gender stereotyping of women by both
men and women. More specifically, we argue that exposure
to sexualized female video game characters may promote
more traditional, less egalitarian beliefs about women in the
real world.
Sex Roles (2009) 61:808–823 811

However, not all portrayals of women in the media are
stereotypical in nature. Although not as frequently, media
periodically portray women in roles that do not conform to
traditional gender role stereotypes. Moreover, some re-
search suggests that exposure to these counter-stereotypical
portrayals may decrease negative gender attitudes and beliefs
in consumers (Aubrey and Harrison 2004; Eisenstock 1984).
In other words, exposure to positive, non-traditional gender
portrayals may be linked to decreases in stereotypical
gender role attitudes and beliefs. Thus, there is potential
for strong, non-sexualized/less sexualized (i.e. counter-
stereotypical) female video game characters to have a
positive effect on individuals’attitudes toward women.
When this body of research is considered alongside the
tenets of social cognitive theory, it would be expected that
video game play would influence a variety of gender-based
attitudes and beliefs about women in society; ranging from
perceptions regarding appropriate and viable careers to views
on women’s cognitive and physical abilities to attitudes
about femininity, sexuality, and attractiveness. Specifically,
individuals’responses to each dimension of gender role
attitudes and beliefs would likely be more traditional, and less
egalitarian, after being exposed to sexualized media images.
Past research (e.g., Ward and Friedman 2006)supportsthis
contention. Thus, the following hypotheses are offered.
H2a: Video game condition will predict player beliefs
about gender career and domestic roles, such that
playing a sexualized female character will result in
greater belief in traditional gender roles, in comparison
with playing a non-sexualized female character or no
video game.
H2b: Video game condition will predict player beliefs
about gender appearance, such that playing a sexualized
female character will result in greater support for
traditional gender appearances, in comparison with
playing a non-sexualized female character or no video
game.
H2c: Video game condition will predict player beliefs
about gender cognitive capabilities, such that playing
a sexualized female character will result in greater
support for traditional gender cognitive capabilities, in
comparison with playing a non-sexualized female
character or no video game.
H2d: Video game condition will predict player beliefs
about gender physical capabilities, such that playing a
sexualized female character will result in greater
support for traditional gender physical capabilities,
in comparison with playing a non-sexualized female
character or no video game.
In addition, it is possible that the gender of the consumer
may moderate the relationship between exposure and
gender attitudes and beliefs. However, previous research
does not provide sufficient insight to allow for the
formulation of specific predictions regarding the nature of
these associations. To address the possible role of gender in
this relationship, the following research question was
developed.
RQ1: Will the gender of the participant interact with
condition to determine gender stereotyping?
The social cognitive framework additionally suggests
that certain media characteristics may enhance learning;
thus the features of the medium and media product, as well
as the experience of the individual with the media product,
should be considered when making predictions. Bandura
(2002) identifies attributes of media stimuli that are likely
to increase the potential for observational learning. He
argues that specific features of media content as well as the
varying attributes inherent to different media play a role in
social learning. In terms of garnering attention (the first
process necessary for observational learning), the realistic
and often stunning graphics of video games would be
expected to play an influential role. The more distinctive/
notable the content, the greater the chance it will gain and
maintain attention. Indeed, the cornerstone of the gaming
industry’s success has been its focus on innovative
technology and cutting edge imagery. Social cognitive
theory also suggests that repetition of messages positively
influences the adoption of attitudes, values and behaviors
depicted in the media by increasing retention in media
consumers (Bandura 2002). Accordingly, the repetitive
action sequences in video games alongside the tendency
for gamers to repeatedly play the games would likely
encourage retention of the messages provided by the
gaming experience.
Moreover, unlike traditional media, video games allow
individuals to actually play the character depicted on the
screen, possibly increasing the level of identification felt
with this mediated model. Individuals take on the persona
of the video game character, control the actions of the
character, and ultimately influence that character’s story or
outcome. This “active participation”(Gentile and Anderson
2003) increases the likelihood that one will learn from the
video game due to greater identification and immersion.
Active participation with video games is likely to heighten
one’s attention to it and one’s feelings of identification with
the mediated models. This makes for a fertile environment
for learning to occur, given the assumptions of social
cognitive theory.
In applying the social cognitive model to gaming, what
becomes clear is that the features of video games and game
play may be likely to enhance the potential for learning
over less interactive media (Schneider et al. 2004). By
incorporating presence into the social cognitive theory
framework, our understanding of how video games operate
812 Sex Roles (2009) 61:808–823

uniquely in this learning process can be better recognized.
Presence, also termed ‘telepresence’(McMahan 2003;
Tamborini and Skalski 2006) refers to the feeling of being
‘present’in a mediated environment (McMahan 2003;
Schloerb 1995; Sheridan 1992; Steuer 1992; Witmer and
Singer 1998). When one experiences presence they over-
look the superficiality or virtuality of the media environ-
ment and become immersed in it. Occurring on a continuum,
presence is the degree to which a video game player feels
present in the game. During game play, a person experiencing
a high level of presence would feel more immersed in the
game and involved with the game characters than engaged in
the goings on in their actual, physical space (e.g., their living
room).
Findings from Tamborini et al.’s(2004) study on video
games and violence, underscore the important role of
immersion in predicting outcomes associated with video
game play. Their results reveal increases in aggression
following video game play to be positively related to the
players’experience of immersion during game play. This
indicates that learning outcomes from video game play may
be enhanced by the experience of presence. Generally, it is
understood that the more present an individual is with the
video game, the more likely they are to identify with the
characters and to be influenced by their gaming experience
(Kiousis 2002; Pinhanez et al. 2000). Given the important
role of identification within the social cognitive model, this
association is of considerable consequence. Thus, the
following research question was developed to explore the
influence of feelings of presence during the video game
experience on the outcomes under study.
RQ2: Will presence interact with condition in predicting
female self-esteem, female self-efficacy, and gender
stereotyping?
In order to test these relationships the following pilot
study and experimental study were conducted.
Method
Pilot Study
A pilot study was conducted to determine the video game
stimuli. In particular, the pilot study helped to ensure that:
(1) the two video games effectively manipulated the
sexualization of the female character (i.e. one “sexualized”
heroine and one “non-sexualized”heroine); and (2) the
games were comparable to one another except in their
physical portrayal of the female heroine. Three top-rated
video games were played and evaluated by 56 undergraduate
students. Women (n=34) made up 61% of the sample. The
majority, approximately 71%, of the participants were White
(n=40). The average age of the participants was 20 years
old. Each participant was randomly assigned to play a video
game for 15 min and fill out a questionnaire about the game
and its character after game play was concluded. Participants
played solo on PC computers, using headphones to hear the
audio.
After playing the game, participants completed an online
questionnaire, which asked questions about the difficulty of
game play; the level of violence in game play; how
engaging the video game was; and how enjoyable the
game was to play. For example, participants were asked to
respond to the following questions on a seven-point likert
scale: “How violent was your video game play?;”“How
difficult was the game to play?;”and “How involving was
the video game?”Additionally, the questionnaire contained
11 items from the Presence Scale.
Second, participants responded to questions about the
characteristics of the character they played. They were
asked to identify the gender, approximate age, and race
of the character they played and to assess their character
based on adjectives and phrases set on a seven-point
semantic differential scale. Participants were asked to
judge their character based on the following adjectives:
attractive/unattractive; strong/weak; aggressive/submissive;
violent/passive; scantily clad/fully clothed; revealingly
dressed/conservatively dressed; skillful/incompetent; good/
evil (moral character); sexualized/not sexualized. The
responses to these semantic differential items were used to
assess the perception of the portrayal of the female character in
each game.
One video game, Tomb Raider: Legend (2006), emerged
as the best video game to use for both the sexualized and
non-sexualized female character conditions. Notably, this
game is a top seller, selling over 2.6 million copies within a
few weeks of its April 11, 2006 release in the U.S. alone
(www.tombraidercentral.net). Lara Croft, the featured fe-
male character in this game, is depicted in a highly
sexualized manner throughout most of the game, however
some portions of the game feature Lara in less sexualized,
more conservative attire. Please see http://tombraidergirl.
com/pics/outfits/tr7/eveningripped.jpg for image of the
“sexualized”Lara (outfitted in “ripped evening gown”),
and http://tombraidergirl.com/pics/outfits/tr7/winter.jpg for
image of the “non-sexualized”Lara portrayal (outfitted in
“winter”clothing). The character, Lara, differs only in attire
(e.g., sexualization, body exposure) throughout the game,
and her skills, actions, and weaponry remain consistent.
Thus, internal validity is strengthened as the consistency in
game play with the two different Lara portrayals ensures
that the only content feature varying across conditions of
game play is the degree of sexualization of the character. In
other words, confidence is increased that the experiment is
actually testing the difference between playing a “sexualized”
Sex Roles (2009) 61:808–823 813

and “non-sexualized”female character, and any emerging
differences between gaming conditions are due to the
character portrayal and not differences in type of game play,
quality of game, etc.
Results from the pilot study revealed that the two
gaming conditions were seen as equivalent on all aspects
of game play and characteristics, other than the sexualized
appearance of the two different Lara Croft portrayals. No
significant differences between the two portrayals were found
in terms of: aggression, t(33)=−1.41, p=.17; strength,
t(33)= −2.36, p=.82; attractiveness, t(33)=−0.56, p=.58;
skill, t(33)=−1.62, p=.12; violence, t(33)=.03, p=.98;
moral character, t(33)= 0.09, p=.93; game violence, t(33)=
−1.09, p=.28; game difficulty, t(33)= −1.44, p=.16; game
enjoyment, t(33)= 1.22, p=.23; or quality of game graphics,
t(33)= 0.22, p=.83. The level of presence experienced while
playing the two different portrayals also did not differ
significantly, t(33)=0.88, p=.39. Additionally, there was
general agreement in terms of the characters’age, race, and
sex. All of the participants identified their character as being
female and of adult age, and approximately 94% identified
their character as White.
Participants rated the characters as significantly different
in terms of their clothing, t(33)= 6.84, p<.001, r
2
=.59,
indicating that the “sexualized Lara”(M=1.18, SD= .39)
was in fact scantily clad, whereas the “non-sexualized
Lara”(M=4.70, SD=2.03) was deemed to be more fully
clothed. Similarly, participants found the “sexualized Lara”
to be dressed in a much more revealing manner compared
to the “non-sexualized Lara,”t(33)=−4.77, p<.001, r
2
=.41.
The “sexualized Lara”(M=1.24, SD = .56) was perceived as
dressing in a revealing manner, whereas the “non-sexualized
Lara”(M=3.22, SD= 1.63) was perceived as more con-
servatively dressed. Additionally, the “sexualized Lara”
(M=3.70, SD= .75) was rated as significantly more
sexualized compared to the “non-sexualized Lara”(M=
6.65, SD=.49), t(33)=13.53, p<.001, r
2
=.85. These
findings are consistent with how sexualization of women
in video games has been conceptualized by content
analytic research, and the physical appearance of the
“sexualized Lara”falls along the sexualized end of the
continuum. In particular, the “sexualized Lara”is depicted
wearing sleeveless clothing with a very low neckline and
short hemline. Additionally, her body is characterized by a
thin frame with a large chest. In contrast, the “non-
sexualized Lara”(i.e. the less sexualized Lara) is fully
clothed from neck to toe, and her proportions appear to be
more realistic in this attire.
Thus, the two character portrayals and the game play
associated with these portrayals were rated as equivalent
other than the sexualized appearance of the character. Based
on the high internal validity afforded by the use of just one
video game and one character for both the non-sexualized
and sexualized character conditions, Tomb Raider: Legend
was chosen for the main experiment. The game chapter
entitled “Japan”was used for the sexualized condition, and
the game chapter called “Kazakhstan”was used for the
non-sexualized game condition.
Main Experiment
Participants
A total of 328 undergraduate students from a large, U.S.
Southwestern university took part in this experimental
study on a voluntary and anonymous basis. Sixty-three
percent (n=206) of participants were women and 37% (n=
122) were men. No video gaming experience was required
for participation in the experiment, however the majority of
students had some experience playing video games. Eighty-
two percent (N=270) (female, n= 167; male, n= 103) of
participants reported having played a video game prior to
participating in this experiment. Of these participants: 66%
(N=177) (female, n=130; male, n= 47) reported playing for
0–1 h per week; 11% (N=30) (female, n=10; male, n= 20)
reported playing 2–3 h per week; 4% (N= 11) (female, n=3;
male, n=8) reported playing 4–5 h per week; 9% (N= 25)
(female, n=10; male, n=15) reported playing 6–7 h per
week; 3% (N=8) (female, n=4; male, n=4) reported
playing 8–9 h per week; and 7% (N=19) (female, n=9;
male, n=10) reported playing for more than 10 h per week.
An undergraduate sample is particularly appropriate to this
project given that college-age individuals make up a large part
of the primary target market for video games. An industry
survey indicated that the average age of the video game player
in the U.S. is 35 years old (ESA 2009), suggesting that video
gaming is becoming more popular among teens and adults
and is no longer just a childhood activity. Additionally,
research on emerging adults (age 18–24) indicates that this
developmental period of life is marked by increased identity
exploration (Arnett 2007). College students, in particular, are
often away from home for the first time, experiencing
heightened independence and new experiences (Arnett and
Tan ner 2005). During this time of life they are exploring who
they are and what role they have in the world. Part of this
exploration is in forming a gendered adult identity.
Procedure
Participants were randomly assigned to one of three
conditions: the sexualized video game condition, the non-
sexualized video game condition, or the control (no video
game) condition. Random assignment was accomplished by
using an online random assignment generator tool (http://
www.randomizer.org/). Up to ten students signed up per
session, and the random numbers generator was used to
814 Sex Roles (2009) 61:808–823

ensure random assignment of each student to a condition.
Gender of participant was evenly distributed across each
condition. Each participant assigned to a video game
condition was briefed on how to play the game on a PC
computer. The researcher provided instruction sheets for
each person assigned to an experimental condition, and the
researcher spent a few minutes familiarizing the participants
with how to play the game. The instructions provided a
summary of the game, specific game play tips to advance in
the game, and the commands/actions available during game
play. The participants were told that they did not have to
use the detailed game instructions if they preferred to
explore the game and play without using the hints given on
the instruction sheet. They were informed that their game
play was not being monitored or recorded and that their
skill level was not being assessed in any way. Also, they
were asked not to play with or ask questions of the other
participants in the room. Headphones were provided to
each participant to hear the auditory aspects of the game
and to provide a more private playing experience; limiting
distractions from others in the room.
After the instruction period, the participants were told to
play the game on their own for 30 min. This timeframe was
selected based on previous research designs which have
typically utilized game play periods from 10–30 min (e.g.,
Carnagey and Anderson 2005; Cicchirillo and Chory-Assad
2005; Eastin 2006). Additionally, traditional media effects
research (e.g., Ward and Friedman 2006) has demonstrated
effects on self-esteem, self-efficacy, and gender beliefs after
exposure to idealized images of the female body for as little
as 10–15 min. Thus, it was determined that playing a video
game for 30 min should be sufficient time for effects to be
observed, if they are present.
Upon completion of the game playing period, participants
filled out an online questionnaire asking questions about their
video gaming habits; recognition of the game, previous
exposure to the character; presence experienced during
game play; gender role beliefs; self-esteem; and self-
efficacy. The participants randomly assigned to the
control condition did not play a video game. They simply
completed a comparable online questionnaire, excluding the
questions about presence.
Independent Variables
Independent variables included portrayal of female video
game character, gender of participant, and presence.
Exploratory factor analysis was conducted in the construc-
tion of the scale measuring the construct presence. Items
were assessed using principal-axis factoring and with a
varimax rotation, and only items that achieved a factor
loading of 0.50 or higher were retained. Cronbach’s alpha is
reported below for this measure.
Portrayal of Female Video Game Character The portrayal
of the female video game character was manipulated, such
that participants in experimental conditions were exposed to a
sexualized female heroine or a non-sexualized female heroine.
The sexualized character was promiscuously dressed; had
large breasts and a small waist; and exposed skin in the
stomach, chest, and thigh regions. The non-sexualized
character was more modestly dressed; had less emphasis on
her figure; and was fully clothed, not revealing much skin in
the stomach, chest, or thigh region. The two versions of Lara
Croft differed in physical appearance but not in aptitude.
Gender The participant’s gender was assessed based on
self-report of being male or female.
Presence Presence (α=.81) was measured using items from
the Presence Questionnaire (PQ) (Witmer and Singer 1998),
Kim and Biocca’s(1997) eight-item Presence Scale (PS),
and items related to character identification. The items
taken from the PQ and PS were adapted to represent the
video game medium. For instance, Kim and Biocca’s scale
refers to presence experienced with a television program, so
any reference to television were replaced with “video
game”or “video game environment.”Both of these scales
contain items measuring immersion and involvement, two
key components of presence. The character identification
items asked participants about how well they related to the
character, how similar they felt the character was to them,
and how well they could empathize with the character. All
items were measured on a seven-point scale, ranging from
“not at all”(1) to “very much/very well”(7), where low
numbers indicate low presence and high numbers indicate
high presence. A median split was performed to categorize
participants’experience of presence as high or low. See
Appendix A for presence items.
Dependent Variables
Dependent variables include: self-esteem, self-efficacy, and
gender attitudes and beliefs. Exploratory factor analysis was
again conducted in the construction of the scales measuring
the construct gender attitudes and beliefs (Cronbach’s alpha
reported below).
Self-Esteem Self-esteem (α=.85, M=5.74, SD = .87) was
measured using Rosenberg’s Self-Esteem Scale. Partici-
pants responded to 10 items and rated their current feelings
about self on a scale from 1 (strongly disagree) to 7
(strongly agree), where lower numbers indicate low self-
esteem and higher numbers indicate high self-esteem.
Negatively stated items (e.g., “At times I think I am no
good at all.”) were reverse coded.
Sex Roles (2009) 61:808–823 815

Self-Efficacy Self-efficacy (α=.79, M=5.60, SD=.62) was
measured using the 10-item General Self-Efficacy Scale
(Sherer et al. 1982). The scale has demonstrated high
construct validity, as evidenced by the scale’s correlation
with, but distinction from, related constructs, such as self-
esteem (Sherer et al. 1982). Participants responded to
statements on a seven-point scale, where lower numbers
indicate low self-efficacy and higher numbers indicate
high self-efficacy. Negatively worded items were reverse
coded.
Gender Attitudes and Beliefs Measures Four dimensions of
gender attitudes and beliefs were measured using the
Attitudes toward Women Scale (AWS) (Spence and
Helmreich 1972) and additional statements about gender
roles and characteristics. It has been noted that the AWS is
primarily a measure of individuals’attitudes toward
women’s rights, rather than gender characteristics, or traits,
and roles (Spence and Hahn 1997). To create a more robust
measure of gender attitudes and beliefs, that is consistent
with current theorizing on this subject, statements about
gender appearance and capabilities were added.
The four gender role dimensions included: career and
domestic labor; appearance; cognitive capabilities; and
physical capabilities. These domains were selected based
on their relevance to the present study’s context, design,
and past gender attitudes and gender roles research. Items
were derived from existing gender-related scales (e.g.,
Personal Attributes Questionnaire, Sex-Role Egalitarianism
Scale, and Gender Attitude Inventory). See Appendix B for
all gender attitudes and beliefs items. The items were
measured on a seven-point scale, ranging from “strongly
disagree”(1) to “strongly agree”(7), with low numbers
indicating egalitarian, or positive, attitudes toward women
and high numbers indicating negative attitudes toward
women. Negatively stated items were reverse coded. Factor
analysis revealed that four subscales emerged, as expected.
Career and domestic labor (α=.86). Participants
responded to statements about men’s and women’s respon-
sibilities related to household chores and childrearing, as
well as the appropriateness of their roles in professional and
manual labor jobs.
Appearance (α=.82). Participants responded to items
about men’s and women’s attire and appearance in public.
These statements reflect gender expectations about how
women should maintain their bodies and appearance for
others.
Cognitive capabilities (α=.79). Participants responded
to items regarding men’s and women’s rationality and
problem-solving skills.
Physical capabilities (α=.80). Participants responded to
statements about men’s and women’s physical strength and
their ability to handle physical challenges.
Covariates
General Gaming Experience Participants’general experi-
ence playing video games was measured by asking about
the number of hours they play video games per week.
Specifically, they were asked “If you have played a
video game prior to this experiment, on average, how
many hours do you spend playing video games per week?”
They selected one of the following options: 0–1h;2–3h;
4–5h;6–7h;8–9h;morethan10h.
Specific Gaming Experience Previous play of the video
game used in the study was measured with one item simply
asking whether or not the participant had played this game
(Tomb Raider: Legend) in the past.
Satisfaction with Performance Participants in the gaming
conditions also reported on the level of satisfaction with
their performance in the game. They were asked to respond
to the following two items: “I am unhappy with my
performance in the video game”(reverse coded); and “I
think I performed well in the video game.”The items were
measured on a seven-point scale, ranging from “strongly
disagree”(1) to “strongly agree”(7), with low numbers
indicating low satisfaction with personal performance in the
game. The correlation of item one (M= 3.32, SD = 1.83, N=
220) and item two (M=4.16, SD= 1.59) was significant,
r(218)= .36, p<.001.An average score was constructed
from the two items to determine level of player’s
satisfaction with their performance in the game. This
variable served as a covariate, and was included to account
for differences in participants’feelings about their experience
with the game during the experiment.
Recognition of Character Prior exposure to the character,
Lara Croft, in movies or other media appearances may
influence the effects of short term game play with the
character. Thus, in addition to measuring prior experience
with the game, general recognition of the character also was
assessed. Participants who played the video game were
asked to answer “yes”or “no”to the question: Did you
recognize the character you played during the experiment
from previous exposure (e.g., advertisements, video games,
movies)?
Results
Analysis of variance and multivariate analysis of variance
tests were used to test the hypothesized relationships. For
Hypotheses 1a and 1b, only females were included in
analysis. For Hypothesis 2 as well as Research Questions 1
816 Sex Roles (2009) 61:808–823

and 2, both males and females were included in analyses.
Any interaction effects that emerged were decomposed by
conducting independent sample t-tests. Initially, ANCOVA
was used to include the covariates in analysis, however the
covariates did not significantly affect the hypothesized
relationships so they were dropped from analysis to avoid
diminishing statistical power.
Overall Gender Differences
Multivariate analysis of variance was used to test for overall
sex differences for participant gender role beliefs, self-
efficacy, and self-esteem. First, sex differences for each of
the gender attitudes and beliefs subscales were tested. Indeed,
men were found to hold less egalitarian beliefs in comparison
to women on all four subscales. Sex differences emerged
for physical capabilities gender beliefs, F(1, 326)=52.76,
p<.001, η
2
=.14, such that male participants (M=4.39, SD =
1.41) reported less favorable attitudes toward women than
female participants (M=3.30, SD=1.25). In other words
male participants reported less confidence in female physical
capabilities, in comparison to male physical capabilities, than
did female participants. A main effect for sex emerged on the
cognitive capabilities subscale, F(1, 326)= 111.48, p<.001,
η
2
=.25. Men and women reported differences in attitudes
toward women in terms of cognitive capabilities, such
that male participants (M=3.50, SD= 1.15) reported less
favorable attitudes toward women than female participants
(M=2.25, SD=.92). In looking at the means, it is evident that
neither male nor female participants strongly agreed with the
idea that men are more mentally capable than women,
however male participants did adhere more strongly to this
belief than did female participants. A main effect for sex also
emerged on the female appearance component, F(1, 326)=
34.62, p<.001, η
2
=.10. Men and women reported differ-
ences in attitudes toward women with regard to female
appearance, such that male participants (M=4.19, SD=1.11)
reported less favorable attitudes toward women than female
participants (M=3.47, SD = 1.02). Overall, men held more
traditional ideas related to how females should dress and
appear in public. Last, a main effect for sex emerged on the
career/domestic work component, F(1, 326)=7.30, p=.01,
η
2
=.02. Specifically, men and women reported differences in
attitudes toward women in terms of career and domestic
work, such that male participants (M=3.98, SD =.96)
reported less favorable attitudes toward women than female
participants (M=3.67, SD= 1.0). In other words, men were
more likely to be supportive of statements that were
reflective of traditional gender roles in terms of career
choices and domestic work.
Additionally, there were significant gender differences
found between men and women on self-efficacy, F(1, 326)=
5.08, p=.03, and self-esteem, F(1, 326)= 5.91, p=.02, such
that women reported lower self-esteem (M=5.65, SD=0.88)
and self-efficacy (M=5.55, SD=0.59) overall in comparison
to men (self esteem, M=5.89, SD=0.84; self-efficacy, M=
5.70, SD=0.65).
Hypothesis 1a
Hypothesis 1a postulated that playing the sexualized female
character would result in lower self-esteem for females. No
support was found for this prediction, F(2, 204) = .98,
p=.38. Thus, there were no significant differences in terms
of playing the non-sexualized video game character (M=
5.53, SD=.58) in comparison with playing the sexualized
character (M=5.52, SD= 1.00) or no video game (M=5.68,
SD=.80).
Hypothesis 1b
Hypothesis 1b stated that playing the sexualized female
character would result in lower self-efficacy in women.
This hypothesis was supported. A main effect for condition
was revealed, F(2, 204)=3.55, p=.03, η
2
=.03. Specifi-
cally, Tukey pairwise comparisons (p<.05) indicated that
playing the sexualized character (M= 5.40, SD = .66) versus
playing no video game (M= 5.67, SD = .52) resulted in
lower self-efficacy for female participants. In other words,
playing a sexualized female video game character nega-
tively affected feelings of self-efficacy in women, compared
to playing no video game character.
Hypotheses 2a, 2b, 2c, and 2d
Hypothesis 2(a–d) proposed that condition would predict
attitudes and beliefs about gender among both male and
female participants. Some support was found for this
relationship. Video game condition was found to have a
significant main effect on subsequent beliefs about gender-
related capabilities, F(2, 325)= 3.38, p= .04, η
2
=.02.
Specifically, participants who played the sexualized character
(M=2.93, SD=1.17) reported less favorable attitudes toward
women’s cognitive capabilities than did participants who did
not play a video game (M=2.52, SD = 1.01). There were no
significant differences involving the non-sexualized character
condition (M=2.71, SD=1.30).
No significant differences were found for gender
attitudes and beliefs about career/domestic roles, F(2,
327)=0.21, p=.81, appearance, F(2, 327) = 0.29, p= .75, or
physical capability, F(2, 327)= 2.64, p= .07.
Research Question 1
Research Question 1 asked whether sex of the participant
would interact with condition in predicting participants’
Sex Roles (2009) 61:808–823 817

gender attitudes and beliefs. A significant two-way interac-
tion between sex and condition was found for the physical
capabilities component of gender attitudes and beliefs, F(2,
322)=2.98, p=.05, η
2
=.02 (see Table 1). T-tests were run to
decompose this relationship. These tests revealed that men
and women who played the sexualized character reported
differences in attitudes about women’s physical capabilities,
t(105)=−2.00, p=.05, r
2
=.04, such that male participants
(M=4.29, SD= 1.59) reported less favorable attitudes
toward women than female participants (M=3.71, SD=
1.40) in the sexualized game condition. Men and women
who played the non-sexualized character reported differ-
ences in attitudes about women’s physical capabilities,
t(105)=−5.56, p<.001, r
2
=.23, such that male participants
(M=4.39, SD= 1.34) reported less favorable attitudes
toward women than female participants (M=3.06, SD=
1.17) in the non-sexualized game condition. This is not
surprising given that a main effect for sex was found for all
four gender role components.
Interestingly, significant differences emerged for
females when looking at differences across the sexual-
ized and non-sexualized conditions. Specifically, women
reported differences in attitudes about women’s physical
capabilities based on condition, F(2, 205)= 5.08, p=.01,
η
2
=.05, such that female participants in the sexualized
condition (M=3.71, SD= 1.40) reported less favorable
attitudes toward women than female participants in the
non-sexualized condition (M=3.06, SD = 1.17) and the no
video game condition (M=3.17, SD=1.13). This suggests
that playing the sexualized character resulted in less
favorable attitudes toward women with regard to women’s
physical capabilities. This was not the case for male
participants. Men did not report differences in attitudes
toward women’s physical capabilities based on condition,
F(2, 121)=0.31, p=.73.
No significant interaction effects for sex by condition
were found for gender attitudes and beliefs about cognitive
capabilities, F(2, 322)=1.01, p=.37, career/domestic work,
F(2, 322)=0.85, p=.42, or appearance, F(2, 322) = 0.79,
p=.50.
Research Question 2
Research Question 2 investigated the influence of
presence on the relationship between video game play
and women’s self-esteem and self-efficacy as well as all
players’gender stereotyping. It was postulated that
playing the sexualized female character might result in
lower self-esteem for women, particularly those women
who experience a high level of presence during game
play. No support was found for this prediction, F(1,
125)=0.00, p=.96. Secondly, it was proposed that playing
the sexualized female character may result in lower self-
efficacy in women, particularly among those high in
presence. This predicted interaction did not emerge, F(1,
125)=.96, p=.33. Last, research question 2 asked whether
presence would moderate the relationship proposed in
hypothesis two, such that a high level of presence
experienced during game play would result in greater
negative attitudes toward women. No support was found
for this prediction. Level of presence did not interact with
condition to determine participants’attitudes toward
women in terms of cognitive capabilities, F(1, 212)= .01,
p=.94, physical capabilities F(1, 212)=0.10, p=.76, female
appearance, F(1, 212)=2.12, p=.15, or career and domestic
work, F(1, 212)=1.52, p=.22.
Condition Cognitive Capabilities Physical Capabilities Appearance Career/Home
M(SD)M(SD)M(SD)M(SD)
Sexualized
Women 2.46(1.01) 3.71(1.40)
abcd
3.51(1.04) 3.52(.98)
Men 3.55(1.08) 4.29(1.59)
a
4.07(1.23) 4.01(.95)
All 2.93(1.17)
e
3.96(1.50) 3.75(1.15) 3.73(1.00)
Non-Sexualized
Women 2.13(.83) 3.06(1.17)
b
3.42(1.15) 3.69(1.12)
Men 3.58(1.30) 4.39(1.34)
bc
4.32(1.11) 4.01(1.01)
All 2.71(1.30) 3.59(1.39) 3.78(1.21) 3.82(1.08)
No Video Game
Women 2.19(.80) 3.17(1.13)
d
3.47(.89) 3.77(.89)
Men 3.32(1.03) 4.55(1.24) 4.17(.94) 3.88(.93)
All 2.52(1.01)
e
3.57(1.32) 3.67(.95) 3.80(.90)
Tab l e 1 Means and standard
deviations for gender attitudes
and beliefs.
abcde
Means with the same
superscript differ significantly at
p<.05 and reflect results
reported for Hypothesis 2(a–d)
and Research Question 1. The
items were measured on a
seven-point scale, with low
numbers indicating egalitarian
attitudes and higher numbers
indicating negative attitudes
toward women
818 Sex Roles (2009) 61:808–823

Discussion
Findings from this study inform media effects researchers;
parents who are concerned about monitoring their child-
ren’s video game play; legislators involved in policy-
making related to video game content and ratings; and
video game programmers. Although the results were
somewhat limited, they shed light on the influence of
exposure to sexualized female video game characters on
individuals’gender attitudes and beliefs and self-concept.
Specifically, these data cautiously indicate that gender
portrayals in video games can, in fact, affect people’s
beliefs about women in the real world, and women’s
self-efficacy.
Hypotheses 1a and 1b
A significant relationship emerged between condition and
female self-efficacy. For women, playing the sexualized
heroine resulted in lower self-efficacy in comparison to
playing the non-sexualized heroine or no video game at all.
This suggests that exposure to sexualized images of women
in video games (among female players) may reduce
confidence in their ability to succeed in the real world.
This is particularly notable in that it supports the idea that
the sexualization of strong female media models may
negate the potentially positive effects of exposure to a
female character who exhibits many counter-stereotypical
characteristics, such as physical strength, independence,
and power. Given that the two character portrayals were
identical in every way except for the degree of sexualiza-
tion of their appearance, the findings from this study
demonstrate that sexualized video game images of females
can, just by the nature of their sexualization, negatively
affect women’s feelings of self-efficacy. One can speculate
that this may have occurred because the sexualization of the
character confines her and limits her power by making her
sexualized body her most prominent feature. Further, this
finding is consistent with those in the domain of traditional
media effects, revealing that exposure to idealized images
of females in the media impacts on feelings of self-worth.
Bussey and Bandura (1999) contend that “perceived
efficacy is…the foundation of human agency”(p. 691).
This study and others indicate that stereotypical media
portrayals of women can negatively influence women’s
belief in their own ability to succeed and do well in the
world.
Counter to previous research findings, however, in the
present study video game play was not associated with
diminished self-esteem in female players. Unlike the survey
study conducted by Funk and Buchman (1996) which
found video game play, in general, to be associated with
lower self-esteem in girls, this experimental study suggests
that playing video games may not always result in lower
self-esteem in female players. Although, it was reasoned
that playing a non-sexualized female video game player
may result in greater self-esteem in female players, in
comparison to playing a sexualized female character, the
present findings did not bear this out. As the means in
Table 1reveal, there is very little difference across the
conditions. In other words, this study does not provide any
support for the idea that, for women, playing a non-
sexualized character, or even a same-sex character of either
a sexualized or non-sexualized nature, positively (or
negatively) affects level of self-esteem.
Consideration of the differences between self-efficacy
and self-esteem may be helpful in uncovering why playing
the sexualized character resulted in lower self-efficacy but
not lower self-esteem for women. Pajares and Schunk
(2001) point out that these two aspects of self are unique
and distinct from one another. Self-efficacy relates to how
well an individual feels they can do something, whereas,
self-esteem is concerned with one’s belief in their own
worth, value, and their liking of themselves. Thus, self-
efficacy reflects one’s confidence in their ability to
accomplish something, whereas self-esteem reflects one’s
value in themselves (Pajares and Schunk 2001). It is,
therefore, possible to feel generally positive about oneself
while feeling somewhat less confident in one’s abilities to
accomplish tasks or goals. In the present case, for women,
playing the sexualized video game character resulted in
lower confidence in their abilities to accomplish things but
did not significantly reduce feelings of self-worth and
liking among female players.
Although it is possible that the distinct aspects of self-
efficacy and self-esteem may account for the results found
here, one additional explanation should not be dismissed.
Based on traditional media effects research, the influence of
exposure to idealized images of the female body on
people’s self-concept and body image is notable but small.
Accordingly, it is possible that this study did not have
sufficient power to observe these effects. Indeed, with the
sample of 206 females in the present study, power analysis
reveals that the ability to reject the null (if an effect is, in
fact, present) is only 60%. It is possible that non-significant
findings are attributable to lack of power. Power was
calculated assuming a small effect size and an alpha of .05.
Hypothesis 2 and Research Question 1
Overall, findings from this study suggest that the level of
sexualization of female video game characters has limited
effects on players’gender-stereotyping. It was predicted
that playing the sexualized female video game character
would result in less favorable attitudes toward women. This
was found to be true in two instances. First, playing the
Sex Roles (2009) 61:808–823 819

“sexualized”character resulted in less favorable attitudes
toward women’s cognitive capabilities. This was the case
for both male and female participants. Second, playing the
“sexualized”character for female participants resulted in
less favorable judgments about female physical capabilities.
Specifically, women who played the “sexualized”character
were more likely to agree with statements suggesting that
women were less physically capable in comparison to men,
whereas the women who played the “non-sexualized”
character were less likely to agree with such assertions.
This indicates that even though the sexualized heroine was
very physically able, playing this character led to greater
gender-stereotyping on the dimension of physical ability.
It is unclear, however, why similar results did not emerge
on the two other gender role dimensions (i.e. appearance
and career/domestic work). Looking at the descriptive
statistics (see Table 1) across conditions for female
participants, the means for three of the four gender role
dimensions are highest in the sexualized character condi-
tion. This trend indicates that playing the sexualized
character, for women, is associated with less favorable
attitudes toward women, but only significantly so for the
capabilities dimensions. There is no clear trend that
emerges from examining the means for the male partic-
ipants. The failure to achieve significance may be a
consequence of inadequate statistical power, or it may be
that no such effects exist. Another explanation is that these
findings are analogous to research studies examining the
interrelationships between the dimensions of gender atti-
tudes. It has been suggested that when making gender-
based judgments about others and gendered behavioral
choices for the self, individuals may not apply the
components of gender stereotypes in a consistent manner
(Signorella 1999; Windle 1987). Additional research will be
required to draw more definitive conclusions regarding
these associations.
Research Question 1 probed the relationship between
gender, condition, and gender stereotyping. It emerged that
the gaming condition had no significant effect on male
players’gender stereotyping. One explanation for the lack
of significant results for tests of the interaction between
gender and videogame condition may come from the
features of the game itself. It is possible that the
stereotypical appearance of the sexualized character may
not have had a negative influence on men’s attitudes toward
women (in comparison to playing the non-sexualized
character or no video game) due to other traits of the
character, which are arguably positive/counter-stereotypi-
cal. This suggests that powerful female video game
characters may have a positive influence on male players
in that (regardless of sexualization) they are much stronger
and more powerful representations of women than is typical
of many other popular media products. Exposure to such
powerful images may decrease the tendency to gender-
stereotype based on appearance.
Research Question 2
Research question 2 asked whether level of presence would
influence the associations under investigation in the present
study. Despite expectations, the level of presence a game
player experienced had no significant effect on these
relationships. This is somewhat surprising given that
previous literature suggests that heightened presence is
associated with greater learning effects. In this case,
perhaps the mere exposure to the image of the female
character was enough to trigger the subsequent gender-
related responses from participants. Future research should
continue to examine the relationship between presence and
gender-related outcomes, considering long term exposure
and exposure among individuals who play games with
similar female images.
Limitations and Future Research
One limitation of this study is the use of only one video
game in the experiment. Although this helped to ensure
high internal validity, results from this study may not be
representative of effects of other video games on the
market. Future studies should examine different genres of
games, different female and male portrayals, and utilize
multiple games in analyses. This will provide a more
complete picture of the effects of playing gendered
portrayals in video games on people’s gender-related
attitudes and perceptions of self. Additionally, it is possible
that the degree of sexualization of the non-sexualized
character was too great to demonstrate the unique effects
of playing a truly non-sexualized character versus a
sexualized character. In other words, perhaps the difference
in appearance between the two character portrayals was not
great enough to demonstrate effects that may have emerged
if the characters had been more extreme in their differences
in appearance. Future studies should examine different
video game heroines to test this idea.
Additionally, the particular game playing experience in the
present study may have presented a limitation. Individuals
who play games often/regularly (i.e. gamers) have longer term
exposure to the images studied in this experiment, and the
game play experience likely differs from the game play that is
experienced in a laboratory. In other words, the design of this
study is limited by its experimental nature in that the
participants and game setting may not be representative of
the gaming experience in the real world. Future studies should
consider triangulation, utilizing multiple methods to compare
results between a laboratory experiment, survey design, and a
natural setting experiment.
820 Sex Roles (2009) 61:808–823

Although this study provides some interesting insights
into the relationship between the sexualization of video
game heroines on gender-related outcomes, the role of
presence in this relationship needs clarification. The results
of this experiment do not provide support for the idea that
presence is positively related to learning outcomes. It is
unclear why presence did not moderate the significant
relationship (among women) found between video game
condition and outcomes such as self-efficacy and percep-
tions regarding women’s physical capabilities. The effect on
self-efficacy and beliefs about female physical capabilities
emerged for all female players regardless of their level of
presence during game play. Future studies should test
different types of game play (e.g., shooter games, sports,
role playing, strategy, etc.) and manipulate the time spent
playing the game to see if either game type of length of
exposure influences the role of presence in learning from
video games.
Lastly, future research should take care to better measure
players’pre-existing relationships with the characters under
study. The present study measured and controlled for
participants’recognition of the character, Lara Croft. In
other words, participants’familiarity with the character
before participating in the study was accounted for.
However, this study did not measure their pre-existing
level of identification with or feelings about the character.
Future studies should consider this. Perhaps, how much a
participant likes and/or identifies with the character (due to
prior exposure) would impact the results of exposure to that
character.
Despite the limitations of this study, the findings provide
initial insights into gender and identity related outcomes
that may be associated with video game use. This area of
research is still largely under-examined, and this study
provides just a glimpse into the potential negative and
positive effects of playing female video game characters,
for both men and women. Continued research in this area
will help to reveal learning trends based on character
portrayal and offer much needed explication in terms of the
theoretical underpinnings for such effects.
Appendix A
Presence
All items were measured on a seven-point scale, ranging
from “not at all”(1) to “very much/very well”(7)
How well were you able to control your actions in the
video game?
How much did the visual aspects of the game involve
you?
How much did the noises/music in the game involve you?
How well were you able to look around or search the
game environment?
How well were you able to interact with objects and/or
people in the game?
How involved were you in the video game experience?
How well did you adjust to the video game experience?
While playing the video game, I felt I was in the world
of the video game.
While playing the video game, I NEVER forgot that I
was in the middle of an experiment.
When I stopped playing the video game, I felt like I
came back to the “real world”after an experience.
While playing the video game, I was unaware of the
noises in the room in which I am sitting.
While playing the video game, I was unaware of the
movements of others in the room in which I am sitting.
I was not at all involved in the video game while I was
playing it.
I felt like my character was a real person.
During my game play, I felt like I really was my
character.
I felt upset when my character did not do well in the
game (e.g., was injured, killed, got lost).
I felt happy when my character did well in the game.
I could relate to my character.
Appendix B
Gender Attitudes and Beliefs
The items were measured on a seven-point scale, ranging
from “strongly disagree”(1) to “strongly agree”(7).
Appearance
Women should dress in a traditionally feminine manner.
Compared to men, it is more important that women look
their best when appearing in public.
It looks worse for a woman to be drunk than a man.
Women should not look masculine in appearance.
Women should dress in a way that pleases and attracts
men.
A woman who dresses in a sexy or provocative manner
is more powerful than a woman who does not.
Career/Domestic Work
Women should assume the same positions in business
and all the professions along with men.
Jobs that require manual labor (e.g., construction, heavy
lifting, etc.) should be done by men rather than women.
A woman’s children should come before her career.
A women’s marriage should come before her career.
Men and women should share housework chores
equally.
Women rather than men should do the cooking at home.
Sex Roles (2009) 61:808–823 821

Women rather than men should be in charge of child-
rearing.
Cognitive Capabilities
Men are more rational than women.
Men are better at problem-solving than women.
Men are better at handling mental challenges than
women
Physical Capabilities
Women are as strong as men.
Men are better at handling physical challenges than
women.
Women are as athletic as men.
References
Arnett, J. J. (2007). Suffering, selfish, slackers? Myths and reality
about emerging adults. Journal of Youth Adolescence, 36,23–29.
Arnett, J. J., & Tanner, J. L. (2005). Emerging adults in America:
Coming of age in the 21st century. Washington, DC: American
Psychological Association.
Ashmore, R. D., Del Boca, E. K., & Bilder, S. M. (1995).
Construction and validation of the Gender Attitude Inventory, a
structured inventory to assess multiple dimensions of gender
attitudes. Sex Roles, 32, 753–786.
Aubrey, J., & Harrison, K. (2004). The gender-role content of
children’s favorite television programs and its links to their
gender-related perceptions. Media Psychology, 6, 111–146.
Bandura, A. (1986). Social foundations of thought and action: A
social cognitive theory. Englewood Cliffs: Prentice Hall.
Bandura, A. (2002). Social cognitive theory of mass communication.
In J. Bryant & D. Zillmann (Eds.), Media effects: Advances in
theory and research (pp. 121–153). Mahwah: Lawrence Erlbaum
Associates.
Beasley, B., & Standley, T. C. (2002). Shirts vs. skins: Clothing as an
indicator of gender role stereotyping in video games. Mass
Communication & Society, 5, 279–293.
Berscheid, E. (1993). Forward. In A. E. Beall & R. J. Sternberg (Eds.),
The psychology of gender (pp. vii–xvii). New York: Guilford
Press.
Bessenoff, G. R. (2006). Can the media affect us? Social comparison,
self-discrepancy, and the thin ideal. Psychology of Women
Quarterly, 30, 239–251.
Bussey, K., & Bandura, A. (1999). Social cognitive theory of gender
development and differentiation. Psychological Review, 106,
676–713.
Carnagey, N. L., & Anderson, C. A. (2005). The effects of reward and
punishment in violent video games on aggressive affect,
cognition, and behavior. Psychological Science, 16, 882–889.
Carpenter, L. M. (1998). From girls into women: Scripts for sexuality
and romance in Seventeen magazine, 1974–1994. Journal of Sex
Research, 35, 158–169.
Cash, T. F., Ancis, J. R., & Strachan, M. D. (1997). Gender attitudes,
feminist identity, and body images among college women. Sex
Roles, 36, 430–447.
Cicchirillo, V., & Chory-Assad, R. M. (2005). Effects of affective
orientation and video game play on aggressive thoughts and
behaviors. Journal of Broadcasting and Electronic Media, 49,
435–449.
Clay, D., Vignoles, V. L., & Dittmar, H. (2005). Body image and self-
esteem among adolescent girls: Testing the influence of sociocul-
tural factors. Journal of Research on Adolescence, 15,451–477.
CNNmoney.com. (2006, January 14). Video game set sales record in
2005: Game Boy, PSP lift the industry, despite languid holiday
season. Retrieved February 6, 2006, from http://money.cnn.com/
2006/01/13/technology/personaltech/gamesales/.
Davidson, E., Yasuna, A., & Tower, A. (1979). The effects of
television cartoons on sex-role stereotyping in young girls. Child
Development, 50, 597–600.
Deaux, K. (1985). Sex and gender. Annual Review of Psychology, 36,
49–81.
Deaux, K., & Lewis, L. L. (1983). Components of gender stereotypes.
Psychological Documents, 13, 25.
Dietz, T. L. (1998). An examination of violence and gender role
portrayals in video games: Implications for gender socialization
and aggressive behavior. Sex Roles, 38, 425–442.
Eagly, A. (1987). Sex differences in social behavior: A social-role
interpretation. Hillsdale: Erlbaum.
Eastin, M. (2006). Video game violence and the female game player:
Self and opponent gender effects on presence and aggressive
thoughts. Human Communication Research, 32, 352–372.
Eisenstock, B. (1984). Sex-role differences in children’s identification
with counterstereotypical televised portrayals. Sex Roles, 10,
417–430.
Entertainment Software Association (ESA) (2009). Top 10 Industry
Facts. Retrieved on February 12, 2008, from http://www.theesa.
com/facts/top_10_facts.php.
Funk, J. B., & Buchman, D. D. (1996). Playing violent video and
computer games and adolescent self-concept. Journal of Commu-
nication, 46,19–32.
Gentile, D. A., & Anderson, C. A. (2003). Violent video games: The
newest media violence hazard. In D. A. Gentile (Ed.), Media
violence and children: A complete guide for parents and
professionals (pp. 131–152). Westport: Praeger.
Glaubke, C. R., Miller, P., Parker, M.A. & Espejo, E. (2001). Fair
play? Violence, race, and gender in video games. Children NOW.
Glick, P., & Fiske, S. T. (1999). Sexism and other “isms”: Interdepen-
dence, status and the ambivalent content of stereotypes. In W. B.
Swann, J. H. Langlois & L. A. Gilbert (Eds.), Sexism and
stereotypes in modern society: The gender science of Janet T.
Spence (pp. 193–221). Washington, DC: American Psychological
Association.
Golombok, S., & Fivush, R. (1994). Gender development. New York:
Cambridge University Press.
Hatoum, I. J., & Belle, D. (2004). Mags and abs: Media consumption
and bodily concerns in men. Sex Roles, 51, 397–408.
Hawkins, N., Richards, P. S., Granley, H. M., & Stein, D. M. (2004).
The impact of exposure to the thin-ideal media image on women.
Eating Disorders, 12,35–50.
Herrett-Skjellum, J., & Allen, M. (1996). Television programming and
sex stereotyping: A meta-analysis. Communication Yearbook, 19,
157–185.
Inness, S. (1998). Tough girls: Women warriors and wonder women in
popular culture. Philadelphia: University of Pennsylvania Press.
Ivory, J. D. (2006). Still a man’s game: Gender representation in
online reviews of video games. Mass Communication & Society,
9, 103–114.
Kim, T., & Biocca, F. (1997). Telepresence in television. Journal of
Computer-Mediated Communication, 3. Retrieved on July 24,
2009 from http://jcmc.indiana.edu/vol3/issue2/kim.html
King, L. A., King, D. W., Gudanowski, D. M., & Taft, C. T. (1997).
Latent structure of the sex-role egalitarianism scale: Confirmatory
factor analyses. Sex Roles, 36,221–234.
Kiousis, S. (2002). Interactivity: A concept explication. New Media &
Society, 4, 355–383.
Lavine, H., Sweeney, D., & Wagner, S. (1999). Depicting women as sex
objects in television advertising: Effects on body dissatisfaction.
Personality & Social Psychology Bulletin, 25,1049–1058.
822 Sex Roles (2009) 61:808–823

Mastro, D., & Behm-Morawitz, E. (2005). Latino representation on
primetime television: A content analysis. Journalism & Mass
Communication Quarterly, 82,110–130.
McGhee, P., & Frueh, T. (1980). Television viewing and the learning
of sex-role stereotypes. Sex Roles, 6, 179–188.
McMahan, A. (2003). Immersion, engagement and presence: A method
for analyzing 3-D video games. In M. Wolf & B. Perron (Eds.), The
video game theory reader (pp. 67–86). New York: Routledge.
Michael, M. (1982). Television and adolescents’sex-role stereotypes:
A longitudinal study. Journal of Personality and Social Psychol-
ogy, 43, 947–955.
Mikula, M. (2003). Gender and videogames: The political valency of
Lara Croft. Continuum: Journal of Media & Cultural Studies, 17,
80–87.
Miller, M. K., & Summers, A. (2007). Gender differences in video
game characters’roles, appearances, and attire as portrayed in
video game magazines. Sex Roles, 57, 733–742.
Morgan, M. (1987). Television, sex-role attitudes, and sex-role
behaviors. Journal of Early Adolescence, 7, 269–282.
Morrison, T. G., Kalin, R., & Morrison, M. A. (2004). Body-image
evaluation and image investment among adolescents: A test of
sociocultural and social comparison theories. Adolescence, 39,
571–592.
Pajares, F., & Schunk, D. H. (2001). Self-beliefs and school success:
Self-efficacy, self-concept, and school achievement. In R. Riding
& S. Rayner (Eds.), Perception (pp. 239–266). London: Ablex.
Pinhanez, C., Davis, J., Intille, S., Johnson, M., Wilson, A., Bobick,
A., et al. (2000). Physically interactive story environments. IBM
Systems Journal, 39, 438–455.
Provenzo, E. F. (1991). Video kids: Making sense of Nintendo.
Cambridge: Harvard University Press.
Richard, B., & Zaremba, J. (2005). Gaming with grrls: Looking for
sheroes in computer games. In J. Raessens & J. Goldstein (Eds.),
Handbook for computer game studies (pp. 283–300). Cambridge,
MA: MIT Press.
Schloerb, D. W. (1995). A quantitative measure of telepresence.
Presence: Teleoperators and Virtual Environments, 4,64–80.
Schneider, E. F., Lang, A., Shin, M., & Bradley, S. D. (2004). Death
with a story: How story impacts emotional, motivational, and
physiological responses to first-person shooter games. Human
Communication Research, 30, 361–375.
Sherer, M., Maddux, J. E., Mercadante, B., Prentice-Dunn, S., Jacobs,
B., & Rogers, R. W. (1982). The self-efficacy scale: Construction
and validation. Psychological Reports, 51, 663–671.
Sheridan, T. B. (1992). Musings on telepresence and virtual presence.
Presence: Teleoperators and Virtual Environments, 1, 120–126.
Sherry, J. A. (2001). The effects of violent video games on aggression:
A meta-analysis. Human Communication Research, 27, 409–431.
Signorella, M. L. (1999). Multidimensionality of gender schemas:
Implications for the development of gender-related characteristics.
In W. B. Swann, J. H. Langlois & L. A. Gilbert (Eds.), Sexism and
stereotypes in modern society: The gender science of Janet T.
Spence (pp. 107–126). Washington, DC: American Psychological
Association.
Signorielli, N. (1989). Television and conceptions about sex roles:
Maintaining conventionality and the status quo. Sex Roles, 21,
337–356.
Signorielli, N. (1993). Television and adolescents’perceptions about
work. Youth & Society, 24(3), 314–341.
Signorielli, N., & Bacue, A. (1999). Recognition and respect: A
content analysis of prime-time television characters across three
decades. Sex Roles, 40, 527–544.
Signorielli, N., & Kahlenberg, S. (2001). Television’s world of work
in the nineties. Journal of Broadcasting & Electronic Media, 45,
4–22.
Spence, J. T., & Hahn, E. D. (1997). The attitudes toward women scale
and attitude change in college students. Sex Roles, 21,17–34.
Spence, J. T., & Helmreich, J. (1972). The attitudes toward women
scale: an objective instrument to measure attitudes toward the
rights and roles of women in contemporary society. Journal
Supplement Abstract Service Catalog of Selected Documents in
Psychology, 2, 66.
Spence, J. T., & Helmreich, J. (1978). Masculinity and femininity:
Psychological dimensions, correlates, and antecedents. Austin:
University of Texas Press.
Spence, J. T., Helmreich, J., & Stapp, J. (1974). The Personal
Attributes Questionnaire: A measure of sex role stereotypes and
masculinity-femininity. Journal of Consulting and Clinical
Psychology, 44, 766–770.
Steuer, J. (1992). Defining virtual reality: Dimensions determining
telepresence. Journal of Communication, 42,73–93.
Tamborini, R., Eastin, M. S., Skalski, P., Lachlan, K., Fediuk, T. A., &
Brady, R. (2004). Violent virtual video games and hostile thought.
Journal of Broadcasting and Electronic Media, 48,335–357.
Tamborini, R., & Skalski, P. (2006). The role of presence in electronic
games. In P. Vorderer & J. Bryant (Eds.), Playing video games:
Motives, responses, and consequences (pp. 225–240). Mahwah,
NJ: Lawrence Earlbaum.
Walsh, D., Gentile, G., Walsh, E., Bennett, N., Robideau, B., Walsh,
M., et al. (2005). Tenth annual media wise video game report
card. National Institute on Media and the Family. Retrieved
August 15, 2006, from http://www.mediafamily.org/research/
report_vgrc_2005.shtml.
Ward, L., & Friedman, K. (2006). Using TV as a guide: Associations
between television viewing and adolescents’sexual attitudes and
behavior. Journal of Research on Adolescence, 16, 133–156.
Windle, M. S. (1987). Measurement issues in sex roles and sex typing.
In D. B. Carter (Ed.), Current conceptions of sex roles and sex
typing (pp. 33–45). New York: Praeger.
Witmer, B. G., & Singer, M. J. (1998). Measuring presence in virtual
environments: A presence questionnaire. Presence, 7, 225–240.
Zuckerman, D., Singer, C., & Singer, J. (1980). Children’s television
viewing, racial, and sex role attitudes. Journal of Applied Social
Psychology, 10, 281–294.
Sex Roles (2009) 61:808–823 823