Content uploaded by Dmitri Williams
Author content
All content in this area was uploaded by Dmitri Williams on Jul 17, 2015
Content may be subject to copyright.
Article
Stand by Your Man:
An Examination
of Gender Disparity
in League of Legends
Rabindra A. Ratan
1
, Nicholas Taylor
2
, Jameson Hogan
2
,
Tracy Kennedy
3
, and Dmitri Williams
4
Abstract
Although video gaming is becoming a more widespread activity beyond its histori-
cally core demographic of young males, participation in competitive gaming remains
largely male dominated. Addressing this issue, this research examines the experi-
ence of female players in one of the world’s most popular games, League of Legends.
Two studies—one qualitative (with 15 participants) and the other quantitative (with
16,821 participants)—confirm that although female players accrue skill at the same
rate as males, there remains a dearth of female players in this community. Moreover,
those females who play with a male partner are less confident in their skills and often
focus on supporting their partner’s advancement, not their own. This work suggests
that one way to address the gender gap in gaming is to better understand and
improve the social dynamics within popular games.
Keywords
female gamers, gender in games, League of Legends, mixed methods
1
Michigan State University, East Lansing, MI, USA
2
North Carolina State University, Raleigh, NC, USA
3
Brock University, St. Catharines, ON, USA
4
University of Southern California, Los Angeles, CA, USA
Corresponding Author:
Rabindra A. Ratan, Michigan State University, 404 Wilson Rd, Room 428, East Lansing, MI 48824, USA.
Email: rar@msu.edu
Games and Culture
1-25
ªThe Author(s) 2015
Reprints and permission:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/1555412014567228
gac.sagepub.com
by guest on January 20, 2015gac.sagepub.comDownloaded from
Early and sustained access to digital games and their cultures has been linked to
interest in Science, Technology, Engineering, and Math-based fields (DiSalvo,
2012; Steinkuehler & Duncan, 2008). Similarly, video games have been found to
influence a variety of educational outcomes, such as knowledge acquisition and
motivational improvements (Connolly, Boyle, MacArthur, Hainey, & Boyle,
2012), as well as general cognitive abilities, such as spatial navigation and orienta-
tion skills (Feng, Spence, & Pratt, 2007). Despite recent advances in extending digi-
tal gaming to populations beyond its long-standing ‘‘core’’ demographic, research
points to an ongoing gender disparity in the play of competitive video games
(N. Taylor, 2012; N. Taylor, Jenson, & de Castell, 2009; Witkowski, 2013), partic-
ularly in action-heavy genres such as real-time strategy (e.g., Starcraft), first-person
shooters (e.g., Call of Duty), fighting games (e.g., Mortal Kombat), and more
recently, multiplayer online battle arenas (MOBAs; e.g., Defense of the Ancients and
League of Legends).
Given the educational and vocational benefits associated with gaming, particu-
larly in strategy-driven and/or action-intensive competitive gaming genres, the mar-
ginalization of females from competitive gaming communities is problematic. An
important goal for research in this area, therefore, is to understand the social, psy-
chological, and cultural barriers facing female players, so as to make competitive
gameplay a more inclusive space.
This article reports on a mixed-methods approach to documenting and account-
ing for the gender gap in League of Legends (hereafter League), one of the most
widely played online PC games on the market (Gaudiosi, 2012). We ask why don’t
more females play League? In order to address this question, we broadly examine
the gender-based interactions between players in League. By looking at the playing
fieldmoreclosely,weseektounderstand how gender disparities inside (and out-
side) League shape female experiences of the game and what the implications of
these often tumultuous experiences are to female League players.
This article describes two separate studies that examine male/female gender
dynamics in the game’s community.
1
Design and data collection of these studies
were conducted independently of each other; however, we discovered, through dis-
cussion of our work, that there was significant overlap in research regarding condi-
tions of female involvement in League, despite differences in the larger study
objectives and methodologies. This enabled us to collaboratively pursue and develop
analyses of female players’ experiences, as recorded in two separate data sets, and
then to integrate these analyses into a cohesive and novel examination of gender dis-
parity in this competitive gaming community.
This integration of two studies from different methodological backgrounds pro-
vides an account that is both generalizable and rich in detail. Study 1 is an explora-
tory qualitative study involving 15 participants; it employs ethnographic techniques
to study the co-situated play of participants (most came in pairs, though one group of
three participated) in an effort to explore the lived experience of League play. Study
2 is a quantitative study that uses survey data about player experiences matched with
2Games and Culture
by guest on January 20, 2015gac.sagepub.comDownloaded from
behavior-log data provided by the game operator (Riot Games), which allows for a
broad comparative gender analysis based on 16,821 participants. Working from
these distinct but complementary perspectives, we find that social and psychological
factors, such as pressure on females to fulfill particular in-game roles, negative inter-
actions between players, and the perception of female players as unwelcome and/or
unskilled participants in the community, significantly contribute to the dearth of
female League players. In particular, we draw from Stereotype Threat Theory
(Steele & Aronson, 1995) to consider how female players are negatively affected
by persistent stereotypes that portray the typical gamer as straight, white, and
male, and position females as ‘‘outside’’ of gamer culture (Bryce & Rutter,
2003; N. Taylor et al., 2009). We conclude with the implications of this ‘‘unequal
playing field’’ for the inequitable distribution of rewards and benefits associated
with competitive gameplay.
Background: Gender and Digital Play
The Entertainment Software Association’s (ESA, 2013, p. 5) 2013 report, Essential
Facts About the Computer and Games Industry, states that 45%of the ‘‘game play-
ers’’ are female and, furthermore, that ‘‘women 18 or older represent a significantly
greater portion of the game-playing population (31%) than boys age 17 or younger
(19%).’’. We are not told, however, what females and males are playing; some
research suggests that females outnumber males as players of ‘‘casual’’ games, while
the reverse seems to still be the case for genres such as sports games and first-person
shooters (Trepte, Reinecke, & Behr, 2009). Nor are we given insight into when, how
often, and for how long males and females of different ages play. Similar statistics
have been used by the ESA for years,
2
complicating notions that this represents a
closing of the persistent ‘‘gender gap’’ in gaming (Jenson & de Castell, 2010).
Furthermore, these facts appear incongruous when compared to the ongoing stories
about a digital gaming culture that is beset by misogyny, homophobia, and racism.
These stories include, among others, the backlash against the producer of a series of
web videos critiquing stereotypes of females in games (Lewis, 2012; Moore, 2012),
the sexual harassment of a female e-sports athlete by her male colleague on a live
webcast (Schreier, 2012), and the outrage over a female game developer’s comments
about her dislike of combat-driven gameplay (Griffiths, 2012). Taken together, these
incidents portray mainstream gaming culture as an unpleasant or openly hostile
space for females (Consalvo, 2012).
Game studies scholars have, for some time now, documented these and other
forms of toxicity that female game players confront, particularly when they partic-
ipate in social gaming contexts, either online (Gray, 2012) or offline, particularly in
public gaming contexts (N. Taylor et al., 2009). In her ethnography of Everquest 2
players, for instance, T. L. Taylor (2006) reported on the ways female gamers have
to negotiate gender stereotypes that, on one hand, portray female game characters in
hypersexualized ways, and on the other hand, position females as ‘‘naturally’’
Ratan et al. 3
by guest on January 20, 2015gac.sagepub.comDownloaded from
inferior users of computer technology. Kelly Bergstrom (2012) outlines some possi-
ble reasons as to the large gender disparity in the space-themed massively multi-
player online (MMO) EVE Online, citing the connection between competencies
required by the game and those associated with male-dominated domains of math
and engineering. One mixed-methods study of an all-female Xbox Live gamer com-
munity reported on the kinds of harassment females confront when playing console
games online (as well as the kinds of support they offer each other; T. Kennedy,
2007); ethnographic research with an Xbox Live clan comprised of Hispanic and
African American females extended this exploration, revealing the ‘‘intersecting
oppressions’’ these gamers encounter from other players (male and female) as both
female and non-White (Gray, 2012). With regard to offline contexts, recent mixed-
methods research reports that digital play at LAN parties and Internet cafe´s remains
overwhelmingly male dominated, despite reported increases in the number of
females playing online (N. Taylor, Jenson, de Castell, & Dilouya, 2014). This work
builds on and updates earlier observations of the gender divides that characterize
public gaming contexts, such as arcades (Alloway & Gilbert, 1998), competitive
gaming tournaments (Bryce & Rutter, 2003; N. Taylor et al., 2009), and Internet
cafe´s (Wakeford, 1999). This research demonstrates that female gamers routinely
encounter sexism, misogyny, alienation, and harassment, both online and offline.
As Helen Kennedy (2002) alluded to in her critical analysis of Lara Croft, the her-
oine of Tomb Raider, these broader social and cultural conditions arguably have as
much, if not more effect on the gender gap in gaming than the well-documented
ways that female characters are portrayed in games (Williams, Martins, Consalvo,
& Ivory, 2009).
Approach
This research utilizes both quantitative and qualitative approaches to document and
account for the gender gap in League play, in order to balance the benefits and short-
comings of both approaches. Namely, what quantitative methods offer in terms of
deductive, objective, and generalizable contributions, they often lack in terms of the
inductive, subjective, and contextual understanding provided by qualitative methods
and vice versa (Lingard, Albert, & Levinson, 2008). Specifically, in Study 1 we ana-
lyzed in-depth interviews with players, observed and recorded their actual play, and
invited them to discuss key moments from the video replay of their match; in Study
2, we analyzed matched server-based and survey data from a large sample of online
game players.
The analysis presented reflects an exploration of specifically how these two
approaches provide complementary perspectives on the gender gap in League play.
As we discuss subsequently, qualitative and exploratory observations and interac-
tions with participants in Study 1 provided the impetus for the hypotheses tested
on Study 2 data; additionally, relationships identified in the Study 2 data were furn-
ished with explanations provided through Study 1’s interviews, in ways similar to
4Games and Culture
by guest on January 20, 2015gac.sagepub.comDownloaded from
the mixed-methods approach to other online gaming research (Williams, Kennedy,
& Moore, 2011). In short, the qualitative study finds phenomena and the quantitative
study examines their generalizability.
League
League is a fantasy-themed MOBA game in which players are placed into ad hoc
teams which battle for dominance on a limited map, with the ultimate goal of over-
running their opponents. The most commonly played mode features the two teams’
bases located at either end of the map, separated by forests and streams populated by
computer-controlled monsters. Three lanes connect the two bases, creating lanes that
must be contested and eventually overrun. League is one of the most played PC
video games in the world, with over 30 million players every month (Riot Games,
2012). League players select from a wide assortment of avatars, called Champions,
each with distinctive appearance, abilities, and narrative backstories. Champion
selection is done in conversation with teammates; players often attempt to claim
in-game roles they prefer, and the Champion they choose will generally be suited
to that role. In-game roles include the ‘‘Top’’ and ‘‘Mid,’’ both named for lanes they
guard (i.e., top or middle) and typically requiring a Champion that can take a great
deal of damage. The ‘‘Jungler’’ roams between lanes to harry opponents and assist
teammates when needed. The ‘‘Attack Damage Carry’’ (ADC) deals heavy damage
against the opposing team’s defenses, while ‘‘Support’’ players assist in that effort
by boosting the ADC’s strength or inhibiting enemies. Often, multiple players will
want a particular role, and negotiation tends to be strategic and sometimes
contentious.
Study 1: A Qualitative Case Study of League
Study 1 undertook an exploratory investigation of mentorship and the development
of expertise among League players. The larger investigation was guided by the
deliberately open-ended research question of how League players improve in gen-
eral. For the purposes of this article, we focus specifically on the experiences of our
female participants.
Method
Data collection for Study 1 was undertaken in the summer of 2013 and used semi-
scripted interviews with participants alongside observations of participants’ play in
order to explore the resources and practices through which League players become
more proficient at the game. The study took in 15 participants (2 female and 13 male)
from various skill levels, ranging from players who had been playing for fewer than 2
months to those who had been playing for 4 years. The shortage of female players was
an artifact of actual community statistics—female players are exceedingly rare (the
Ratan et al. 5
by guest on January 20, 2015gac.sagepub.comDownloaded from
presumably representative sample in Study 2 is only 4%female). The majority of par-
ticipants played in co-situated pairs, though one group contained three participants.
The analysis under discussion focused on the two participants from this study who
were female, with special attention paid to their interactions with male players.
Although this is a relatively small number of participants, the goal of this study was
not to generate generalizable claims (which is the focus of Study 2), but instead to dis-
cover what phenomena occur, and the two female participants provided sufficiently
rich data to this end.
Participants were found primarily via social media, with recruitment materials
posted to appropriate group sites such as Facebook and Reddit, as well as the
campus-wide electronic bulletin board system. Matches were played in an on-
campus studio; players logged into their League accounts to play one standard match
in the ‘‘Summoners Rift’’ map, while video and audio of the match were recorded.
Researchers watched the match via League’s ‘‘spectate’’ function, while taking field
notes on the match and participants. Once the match ended, players and researchers
watched the recorded match together, giving players the opportunity to reflect on
their own in-game actions, the composition of the teams in terms of Champion and
role selection, the actions of teammates or opponents, and anything else they felt was
relevant to bring to our attention. Before leaving, we sought participants’ vocal per-
mission for us to spectate future play and assessed their willingness to return and
play for us again.
Results
The original purpose of this research was to explore the learning and teaching prac-
tices of League players at different stages in their involvement. It became apparent to
us over the course of the first few research sessions, however, that issues of gender
with regard to League figured largely in participants’ accounts of their involvement
in this game and its community. Both male and female participants repeatedly
directed us, in interviews and while watching replays, to the dearth of female players
in their peer groups and in the community more generally, and the often toxic, mis-
ogynistic nature of exchanges between players during matches. To be clear, these
observations are not offered as evidence of the larger state of affairs with regard
to gender disparities in League play. Instead, they are exploratory and are followed
up on more systematically in Study 2.
Annie
‘‘Annie,
3
’’ a university student of Chinese descent, came in for two separate play
sessions. In the first session, she accompanied a male friend of her boyfriend (both
of whom she said she plays with), and in the second session she was accompanied by
a classmate of hers, another female recently introduced to the game (Poppy). In her
intake interview, Annie depicted herself as being at somewhat of a crossroads in her
6Games and Culture
by guest on January 20, 2015gac.sagepub.comDownloaded from
League play: introduced to the game through watching her boyfriend play almost a
year and a half before, the first role she learned to play was Support, as a means of
(literally) supporting her boyfriend’s preferred role as ADC. More recently, how-
ever, and during the time of our first encounter with her, Annie had been working
on improving her play in other roles, specifically Mid and Jungler, a shift that coin-
cided with more time spent playing without her boyfriend. In the following interview
excerpt, in which we ask about other female players she knows or plays with, she
expresses frustration with what she seems to regard as a certain stereotyping of
female players into Support roles:
Interviewer: How many other women do you know who play?
Annie: Umm, I can count them on one finger. So it’s like [counting on
hand] four or five?
Interviewer: Ok. Ok. And what do they usually play as?
Annie: Uhhh ... I’m not too sure. Support. Sadly.
Interviewer: Ok. Why sadly?
Annie: So most girls are like, usually Support, that’s how they get car-
ried in ranked
4
games. So, I’m trying to break out of that.
As this exchange indicates, Annie sees other female players she knows (as well as
herself) relegated to Support roles, something many of our participants identified as
being the least popular position to play. This relegation is something Annie is push-
ing back against through her own attempts at acquiring the competencies necessary
for the roles of Mid and Jungler, along with their associated play styles and
Champions.
Annie’s Gameplay
Interestingly, in her first play session (with her boyfriend’s friend, ‘‘Kennen,’’ a
self-identified Caucasian male of similar age), Annie took on a Support role, pos-
sibly because it is the role with which she is most familiar, given that she was play-
ing in an unfamiliar setting and under conditions of observation. Compared to
other participants, Annie seemed to be a competent and experienced player as
demonstrated by her item purchases, leveling, protection of other Champions, map
awareness, and low kill score (all markers of Support competency); in addition, she
demonstrated an understanding of intricate game mechanics and strategies not
shared by all participants.
Ratan et al. 7
by guest on January 20, 2015gac.sagepub.comDownloaded from
In contrast to these numerous demonstrations of competency—particularly in her
role as Support—Annie’s communicative actions during the match suggested a reti-
cence to interact extensively with other players. Although she did a fair degree of
‘‘pinging’’ (using the game’s repertoire of predefined statements to alert teammates;
in her case, primarily to indicate ‘‘Enemies are missing’’ and ‘‘Be careful’’), she
mostly refrained from using the text-based in-game communication system to
‘‘speak’’ with her teammates (or opponents). However, Annie proved to be one of
our more vocal participants, with many of her verbal utterances during the match
being directed at her teammates, who of course, cannot hear her (with the exception
of Kennen who is beside her). These include comments urging her teammates to pur-
sue a particular course of action (Watch out, Udyr) or criticizing them after a bad
play (What is Jayce doing?
5
). When we asked why she does not actually type these
comments into the game’s chat channel instead of directing them verbally at an unre-
sponsive screen, she responded that she did not feel confident enough in her judg-
ment or abilities to warrant the attention (and possibly, negative responses) that
such exchanges might solicit from teammates.
Poppy
In contrast to Annie, ‘‘Poppy’’—the other female (and also of Chinese descent)
player in our study—was a novice at the time we met her. As with Annie’s early
experiences with the game, Poppy plays with her boyfriend and primarily in the role
of Support. During her intake interview, she described getting involved in the game
at the insistence of her brother and her boyfriend. By her own account, her involve-
ment with League and with the game’s community is largely mediated by her boy-
friend; she watches his matches, plays almost exclusively with him, and when they
are both spectating other matches, he decides what to watch.
She was accompanied by Annie, who came in for a second session. Their match
also involved Poppy’s boyfriend, playing remotely (he was not part of the study).
Throughout the match, Poppy followed Annie’s lead in everything from item pur-
chases to skill upgrades, and even in movement around the map. As Support to
Annie’s ADC, Poppy frequently indicated that she was confused by some of the pri-
mary tenets of the role; Annie had to direct her in the placing of ‘‘wards’’ to provide
vision of enemy movements, as well as guide her through the process of ‘‘leashing,’’
in which several team members assist the Jungler in obtaining an early kill. Poppy
also followed Annie’s lead in item purchases, skill upgrades, and movement around
the map.
Discussion
Several aspects of these players’ experiences, as described in their interviews and
observed in their lab-based play, resonate with qualitative research previously car-
ried out with heterosexual couples playing massively multiplayer online role-
8Games and Culture
by guest on January 20, 2015gac.sagepub.comDownloaded from
playing games (MMORPGs) together (Carr & Oliver, 2009; Nardi, 2010; Williams
et al., 2006). Like the female participants in these MMORPG studies, both Annie
and Poppy were introduced to the game by male partners. Annie, despite her increas-
ing self-taught competency in other roles, learned the game primarily through play-
ing the Support role for her boyfriend. Poppy was, at the time of her involvement in
the study, also playing as Support for her boyfriend; but unlike Annie her involve-
ment was almost exclusively contingent on her boyfriend’s play. This is consistent
with the MMORPG research, where female participants in those studies started off
as healers or as party support, essentially parallel to the Support function in League
(Carr & Oliver, 2009; Nardi, 2010; Williams et al., 2006).
Although Annie and Poppy are at different phases in their League career, they
share key similarities. In addition to both playing in Support roles with male part-
ners, both females largely refrain from using in-game chat functions, aside from
Annie’s use of pings, which are preprogrammed and reveal nothing of the player
issuing them. They are hesitant to communicate extensively with online teammates,
whereas our male participants consistently used the chat function for a variety of
purposes, from strategizing to trash talking. For Annie in particular, this suggests
that game-based competence does not necessarily lead to a willingness to actively
communicate in these team-based competitive situations. With the exception of
pings and short (often one word or acronym
6
) utterances, these females were effec-
tively silent—behavior at odds with the game’s reputation for boisterous and often
inflammatory communication between players (Carlson, 2013).
In summary, the qualitative insights about these two female players are as fol-
lows: Both respondents were introduced to the game by their male partners; both
began playing as Support, directly assisting their male partners’ (and in Poppy’s
case, brother’s) efforts to play the more central and dominant role of ADC; both
were observed to participate in text chat far less than study male participants; and
both were unwilling to openly criticize other players (particularly in the case of
Annie). These observations inform the hypotheses articulated subsequently with
regard to Study 2, in an effort to see whether and to what extent the experiences
of these female Study 1 participants are indicative of the experiences of female
League players more generally.
Study 2: A Quantitative Complement
Study 2 complements and extends the qualitative approach of Study 1 by providing a
deductive approach to understanding the gender gap in competitive gaming,
informed both by Study 1 and by the consistent prior research on gaming’s gender
gap. The findings in Study 1 illustrate the prevalence among League players of cer-
tain stereotypes related to females and gaming (e.g., females as less competent
gamers and females typically play the Support role), but also suggest ways in which
players may individually resist these stereotypes (e.g., by playing less with their boy-
friends). It is important to understand these dynamics because such stereotypes are
Ratan et al. 9
by guest on January 20, 2015gac.sagepub.comDownloaded from
easily reinforced through psychological factors (e.g., reminders of the stereotype;
Steele & Aronson, 1995). Conversely, this also suggests the potential to systemically
counteract such stereotypes by avoiding such reinforcement, or even to use positive
stereotypes to encourage behavior that is beneficial to all players.
We start by drawing parallels between the accounts of female League players’
experiences presented in Study 1 to generalizable claims that can be tested with
quantitative data. Although females are demonstrably absent from the League com-
munity (4%of our sample), the responses from Study 1 also suggest that males gen-
erally tend to play the game more intensively,
7
which would be consistent with the
stereotype of the average gamer as an adolescent or young adult male (Griffiths,
Davies, & Chappell, 2003; Kahn, Ratan, & Williams, 2014). Thus, if this stereotype
is supported in League, we would expect that male players spend more time playing
the game. With this in mind, we formulated and tested a series of hypotheses in order
to explore and unpack these expectations.
Hypothesis 1: Male players have a higher average number of matches played than
females.
Game skill (the ability to perform well in the game) is an important factor to con-
sider, as players with more skill should tend to perform better, and may take on the
roles that require more experience. If males play more matches than females
(Hypothesis 1), then it would follow that males should have a higher average skill
level than females. Further, if males are, as the stereotype suggests, more adept at
playing video games in general, then for males and females who have played the
same number of matches, males should still have a higher average skill level. This
expectation was expanded for exploration in our second hypothesis.
Hypothesis 2: (a) Male players’ average level of skill is higher than that of female
players, (b) even for females and males who have played the same number of
matches.
As suggested by our observations of Annie and Poppy, female players may
express less confidence in their abilities—even when (as with Annie) they are as
good as, if not better than, their male counterparts with similar experience and time
investment in the game. If females are not inherently less skilled than men, then how
can we account for this apparent lack of confidence?
One potential explanation comes from Stereotype Threat Theory, which posits
that when people are reminded of a negative stereotype about a demographic to
which they belong, they are more likely to conform to that stereotype (Steele &
Aronson, 1995). For example, African Americans were found to perform worse
on an intellectual test when asked to report their race before the test (compared to
not reporting it), which served to remind them of the negative stereotype about Afri-
can American intellectual performance. Stereotype Threat Theory has been exam-
ined with respect to other social categories (e.g., gender; Spencer, Steele, &
10 Games and Culture
by guest on January 20, 2015gac.sagepub.comDownloaded from
Quinn, 1999) as well as on other types of performance, such as math (Nguyen &
Ryan, 2008), sports (Stone, Lynch, Sjomeling, & Darley, 1999), and automobile
control (Yeung & von Hippel, 2008).
In our review of the literature, we could find only two studies that examined how
stereotypes about gaming ability may impact play behaviors. Richard and Hoadley
(2013) found that among a sample of over 100 gamers, females had lower gaming
self-concept than men, which they explain as a result of the threat induced by the
stereotype of females being inherently less skilled at gaming than males. In an
experimental study, Vermeulen, Nu
´n
˜ez Castellar, and Van Looy (2014) found that
females who thought they were playing a game against a male (as opposed to a
female) felt more stress and perceived their skills as lower. Both studies suggest that
the stereotype that females are not meant to be gamers may hinder female players’
ability and willingness to achieve and compete within such games. In a combat-
driven game such as League, killing opposing players is both a marker of compe-
tency and a crucial element of winning. Thus, we would expect that female League
players are less confident in their ability to kill opponents than males, an expectation
we specifically examine in our third hypothesis.
Hypothesis 3: Females are less confident in their ability to kill opponents than
males.
If female players are treated poorly by the overwhelmingly male player commu-
nity and not expected to compete well, why do they play such games at all? One
common stereotype, which ignores the potential that some females actually enjoy
playing such games, is that of the ‘‘girlfriend gamer,’’ the female who enters the
game space only as a tagalong with a romantic partner (Carr & Oliver, 2009). Both
female participants in Study 1 became involved in the game at the behest of their
romantic partners; while this is certainly not the case for every female player, we
would expect to see a larger proportion of females who are brought into the game
by male partners than who come to it on their own, or via other social relationships.
We would also expect, and hypothesize subsequently, that the proportion of female
players who play with a romantic partner is larger than that the proportion of male
players.
Hypothesis 4: (a) There is a larger proportion of females who play with a roman-
tic partner than those who do not and (b) females are more likely to play with a
romantic partner than males.
Due to the ways they are typically introduced to the game (i.e., via male partners)
and the persistent stereotypes about females’ gaming abilities and predispositions
(Carr, 2005; Jenson & de Castell, 2011; T. L. Taylor, 2003; Thornham, 2008),
females are relegated to game roles that are perceived as less aggressive, and more
collaborative than other in-game roles. In other words, females are discouraged from
playing roles that enhance the skills related to the more competitive elements of the
Ratan et al. 11
by guest on January 20, 2015gac.sagepub.comDownloaded from
game (e.g., killing opposing players) and instead encouraged to facilitate other
(male) players’ pursuit of more aggressive roles. In Study 1, both Annie and Poppy’s
experiences illustrated this phenomenon: Poppy was recently brought into the game
by her boyfriend and told that she would make a good Support player, and while
Annie expressed a desire to learn more central roles in future matches, such as Mid
and Jungler, she told us that she had spent most of her time in the game playing Sup-
port for the other members of her teams (and in particular, during play with her boy-
friend). If the experiences of these females are reflective of a general phenomenon,
then we are led to hypothesize that female players tend to play Support roles more
than males.
Hypothesis 5: (a) Females play Support more than males, and specifically, (b) the
more frequently a female plays with a romantic partner (as a proportion of her
total play), the more likely she is to play a Support role.
If one of the reasons that females choose Support roles is the behest of their male
romantic partners, then it is possible that females who do not play with romantic
partners are less likely to play Support. For instance, Annie’s expressed desire to
change roles in the game was coupled with a sense that playing with her boyfriend
hindered her ability to choose more competitive roles; this led to a desire to play less
frequently with her boyfriend and more frequently on her own. So, by reducing the
proportion of playtime she shared with him, she anticipated being able to increase
her competitive abilities.
Further, if this hypothesis is supported, and given that Support roles (compared to
more competitive roles) are expected to contribute less to the competitive play expe-
rience, then we would expect that females who play with a romantic partner in the
majority of their playtime have weaker competitive abilities than those who do not.
Conversely, for male players, we might expect to see those who play with their
female partners in the majority of their playtime exhibit a higher degree of compet-
itive abilities than those who do not since they would be likely to benefit from the
dedicated support of their female partners. To examine these expectations, we for-
mulated our final hypotheses:
Hypothesis 6: (a) Females who play with a romantic partner more frequently (as a
proportion of total play) have weaker competitive abilities, while (b) males who
play with a romantic partner more frequently (as a proportion of total play) have
stronger competitive abilities.
Method
Although the qualitative study was able to hone in on specific, lived experiences
of both female and male League players, the quantitatively focused Study 2 was
able to make use of a unique data set to explore these hypotheses in the larger
gaming community. The developer of League, Riot Games, provided data from
12 Games and Culture
by guest on January 20, 2015gac.sagepub.comDownloaded from
the game’s back-end databases, and distributed a survey to a large sample of
players. Such large-scale cooperation between an academic research team and
a major game company is rare (but see Williams, Consalvo, Caplan, & Yee,
2009). The collected data were then used to perform statistical analyses in pur-
suit of our hypotheses.
Procedure and Data
Results from a survey conducted in November 2010 were linked with unobtrusively
collected game-based behavioral data, all of which was privacy-protected and anon-
ymous. The survey link was sent via e-mail to a randomly selected set of 113,579
players from the North American League server. Within 1 week, 22,521 completed
the survey, with 18,627 deemed as valid (i.e., not a duplicate and sufficient time—12
min—spent answering questions) for a final response rate of 16.4%. All respondents
were compensated with an in-game ‘‘boost’’ that doubled the rate at which they
accrued points in their next four victories, a form of compensation that appeals to
both male and female players of all experience and skill levels. The survey included
approximately 150 questions, all of which were in multiple-choice or Likert-type
scale format, and many of which required simple ratings of single words (e.g., adjec-
tives describing the self). The topics of most of these questions are not relevant to
this research (e.g., personality traits and play motivations) but are being explored
in other studies. Similarly, the log data set contained approximately 100 variables,
most of which are also not related to this research (e.g., choices of specific characters
and types of damage dealt). The specific measures of interest for this study are
described subsequently.
Participant Sample
Of the valid respondents, the average age was 21.90 (SD ¼5.22), with 93.6%
reporting being male, 4.1%reporting being female, and 2.3%not reporting on
this variable. Because we are primarily interested in the comparison of males
and females, we removed those who did not report their gender from the sam-
ple. Further, in order to increase the likelihood that our sample was represen-
tative of the population of League players at the time of data collection, we
removed respondents who had played fewer than 15 matches, a cutoff point
arrived at first by examining the frequency distribution of matches played (see
Figure 1), which indicated that a large segment of the sample had fewer than 15
matches. In support of this finding, experienced players we consulted con-
firmed that players with fewer than 15 matches were, given the steep learning
curve and time requirements of the game, unlikely to be representative of more
invested players. After these adjustments, our final sample included 16,821
players.
Ratan et al. 13
by guest on January 20, 2015gac.sagepub.comDownloaded from
Measures
In addition to self-reported gender (what is your gender), the present analysis con-
sidered two survey-based measures, namely, confidence at killing opponents and
frequency of playing with a romantic partner. For the measure of confidence at kill-
ing opponents, which we developed in consideration of the notion of self-efficacy
(Bandura, 1977), participants were asked ‘‘How confident are you in your ability
at killing the opponents champions,’’ to which they responded on a 5-point scale
ranging from not confident at all to very confident. For the measure of playing with
a romantic partner, participants were asked ‘‘How often do you play with a romantic
partner (e.g., spouse, fiance´, and boyfriend/girlfriend)?’’ to which they responded on
a 5-point scale ranging from never to always.
The present analysis of the behavior log data focused on aggregated information
over the lifetime of each player’s account. The ‘‘number of matches’’ variable was a
count of the number of matches played over the account’s lifetime. The ‘‘skill’’ vari-
able was based on the game company’s proprietary algorithm for rating the skill
level of each player (a type of Elo rating
8
), which takes a variety of gameplay ele-
ments into account (e.g., wins, kills, etc.). Likelihood of playing Support was
derived from a count of the number of kills and assists, indicative of contributing
to an opponent player’s death but not dealing the final blow (thus, bolstering another
player’s kill count). Specifically, we calculated an assists-to-kills ratio (number of
Figure 1. Frequency distribution of number of matches played across all respondents.
14 Games and Culture
by guest on January 20, 2015gac.sagepub.comDownloaded from
assists divided by number of kills). This reflects the extent of playing a Support role
because the Support player is expected to deal some damage to opponents, but
refrain from dealing the final killing blow (and are often scolded quite severely
by teammates when they do so). Finally, as a measure of competitive abilities, we
calculated a kills-to-deaths ratio because an adept competitive player is likely to kill
more opponents, and to and die fewer times, in a given match. The descriptive
statistics for all variables can be found in Table 1.
Results
Our analyses relied mostly on a series of t-tests (to compare means between groups),
analyses of covariance (ANCOVAs; to compare means between groups while also
considering other continuous factors that may influence these means), w
2
tests (to
compare the number of individuals within different categories), and regressions
(to examine relationships between multiple continuous variables). For Hypothesis
1 (male players have a higher average number of matches played than female),
we found that males had played an average of 347.44 matches (SD ¼237.83), while
females had played an average of 275.66 matches (SD ¼212.23), and this difference
was statistically significant, t(744.10) ¼8.55, with a small effect size (Cohen’s D¼
0.30). This supports our hypothesis: male participants played more matches than
female participants.
Similarly, we found that the average skill rating was slightly higher for males
(M¼1276.15, SD ¼167.00) than for females (M¼1217.94, SD ¼171.23), and
this difference was statistically significant, t(16819) ¼8.85, with a small effect
size (Cohen’s D¼0.35), supporting Hypothesis 2a (male players’ average level
of skill is higher than that of female players). To examine whether this difference
held even for males and females who have played the same number of matches
(Hypothesis 2b), we conducted an ANCOVA with gender as a fixed factor, number
of matches played as a continuous factor (to control for number of matches
played), and skill rating as the dependent variable. The results (Table 2) suggest
that for females and males who have played the same number of matches, gender
significantly predicts player skill level (as previously defined), with skill rated
Table 1. Descriptive Statistics for All Variables.
NMin Max MSD
Opponent kill confidence 16,491 1.00 5.00 3.89 1.06
Play w Partner 16,087 1.00 5.00 1.45 1.02
Number of matches 16,821 16 1749 344.57 237.26
Skill 16,821 508 2091 1,273.82 167.55
Assists-to-kills ratio 16,821 0.22 9.88 1.41 0.53
Kills-to-deaths ratio 16,821 0.06 4.86 1.02 0.36
Ratan et al. 15
by guest on January 20, 2015gac.sagepub.comDownloaded from
higher for males (estimated M¼1,276.05, SE ¼1.31), 95%CI [1,273.48,
1,278.63], than for females (estimated M¼1,220.30, SE ¼6.45), 95%CI
[1,207.66, 1,232.94]. However, the effect size was negligible (Z
2
¼.00), which
suggests that this difference was not meaningful, and thus that females accrue
skillatthesamerateasmales.
For Hypothesis 3 (females are less confident in their ability to kill opponents
than males), we conducted an ANCOVA with gender as a fixed factor, skill rating
as a continuous factor (in order to compare the effects of gender across players
with the same level of skill; i.e., as a control variable), and the measure of confi-
dence at killing opponents as the dependent variable. The results (Table 3) suggest
that females are indeed less confident about killing opponents (estimated M¼
3.45, SE ¼.04), 95%CI [3.37, 3.53], than males (estimated M¼3.91, SE ¼
.01), 95%CI [3.89, 3.92]. This effect size was very small (Z
2
¼.01), so the hypoth-
esis was only marginally supported. In other words, females’ confidence in their
ability to kill opponents was not meaningfully lower than that of male players with
similar levels of skill.
For Hypothesis 4a (there is a larger proportion of females who play with a roman-
tic partner than females who do not), we found that 73%of the female players
reported playing with a romantic partner at least rarely, while 27%reported never
doing so. This difference was significant, w
2
(1, N¼553) ¼68.76, p¼.00, and had
a medium effect size (j¼.33), supporting the hypothesis: more female participants
played with a romantic partner than those who did not.
Table 2. ANCOVA analysis on skill.
Sum of Squares df Mean Square FZ
2
Number of matches 1,107,267.47 1 1,107,267.47 39.71*** .00
Gender 2,001,266.35 1 2,001,266.35 71.78*** .00
Error 468,901,841.97 16,818 27,880.95
Total 27,766,384,390.45 16,821
Note. ANCOVA ¼analysis of covariance.
***p< .001.
Table 3. ANCOVA Analysis on Opponent Kill Confidence.
Sum of Squares df Mean Square FZ
2
Skill 10.78 1 10.78 9.72** .00
Gender 131.79 1 131.79 118.89*** .01
Error 18,277.48 16,488 1.11
Total 267,635.00 16,491
Note. ANCOVA ¼analysis of covariance.
**p< .01. ***p< .001.
16 Games and Culture
by guest on January 20, 2015gac.sagepub.comDownloaded from
For Hypothesis 4b (females are more likely to be playing with a romantic partner
than males), we found that females rated the frequency of their play with a romantic
partner as significantly higher (M¼3.30, SD ¼1.61) than did males (M¼1.38,
SD ¼.90), t(669) ¼30.33, with a large effect size (Cohen’s D¼2.07). Thus, this
hypothesis was strongly supported: female participants played with romantic part-
ners more regularly than male participants did.
We tested Hypothesis 5a (females play Support more than males) in an ANCOVA
with the assists-per-kills metric as the dependent variable in each, gender as the fixed
factor, and skill rating as a continuous factor. The results (Table 4) suggest that
females play more Support roles (estimated M¼1.93, SE ¼.02), 95%CI [1.89,
1.96], than do males (estimated M¼1.39, SE ¼.00), 95%CI [1.38, 1.40]. This dif-
ference is significant and has a small effect size (Z
2
¼.04), supporting the hypoth-
esis: female participants played Support more than males.
Because only female players were of interest for Hypothesis 5b (the more fre-
quently a female plays with a romantic partner, the more likely she is to play Sup-
port), we restricted the sample to only females (N¼673) and ran a linear regression
analysis with the measure of playing with a romantic partner as the independent vari-
able of interest, the measure of skill as a control variable, and the assists-per-kills
metric as the dependent variable. The results (Table 5) suggest that for female play-
ers, playing with a romantic partner significantly predicts playing Support, with a
small effect size (Z
2
¼.03). This supports the hypothesis: female participants who
Table 4. ANCOVA Analysis on Assists-per-Kills Ratio.
Sum of Squares df Mean Square FZ
2
Skill 235.51 1 235.51 912.62** .05
Gender 185.52 1 185.52 718.90*** .04
Error 4,340.12 16,818 .26
Total 38,248.80 16,821
Note. ANCOVA ¼analysis of covariance.
**p< .01, ***p< .001
Table 5. Regression on Assists-to-Kills Ratio for Female Players Only.
Unstandardized Coefficients Standardized Coefficients
bSEbtZ
2
Constant 0.20 0.28 0.71
Play w Partner 0.09 0.02 0.16 4.14*** .03
Skill 0.00 0.00 0.21 5.44*** .04
Note. N ¼673.
***p< .001; F(2, 650) ¼21.206, p< .001, R
2
¼.061, adjusted R
2
¼.058.
Ratan et al. 17
by guest on January 20, 2015gac.sagepub.comDownloaded from
played with a romantic partner played Support more than female participants who
did not play with romantic partners.
Because of their parallel structure, we tested Hypothesis 6a (females who play
with regularly with a romantic partner have weaker competitive abilities) and
Hypothesis 6b (males who play with regularly with a romantic partner have stronger
competitive abilities) in the same ANCOVA, with gender, a dichotomized measure
of playing with a romantic partner (never ¼0 and rarely through always ¼1), the
interaction between the two as the fixed terms, skill rating as a continuous factor,
and the competitive abilities measure as the dependent variable. As the results
(Table 6) suggest, the interaction term was significant, but the effect size negligible
(Z
2
¼.00), and so neither hypothesis was supported. In other words, whether or not
they played with a romantic partner, there was no meaningful difference in compet-
itive abilities for males or females.
Discussion
The findings from Study 2 suggest that League is largely male dominated, with
males playing more matches than females and having higher skill level on average;
this is of course consistent with the gamer-as-adolescent-male stereotype. However,
the difference in actual skill between males and females was found to be negligible
for those who have played the same number of matches. This is inconsistent with the
stereotype, which suggests that males accrue game skill more readily than females.
Thus, this suggests that females are not at an inherent disadvantaged with respect to
developing game skills, a finding consistent with previous research examining the
relationship between gender and gaming expertise, both from qualitative (Jenson
& de Castell, 2008) and quantitative (Lucas & Sherry, 2004) studies. Nonetheless,
females were found to demonstrate less confidence in their ability to kill opponents
than males, albeit this is a very small effect. Together, these findings suggests that
the stereotype of females as being less adept at gaming than males, affects their per-
ception of their own abilities, even when those abilities are no different from males.
This finding is consistent with previous research on stereotype threat in gaming
(Richard & Hoadley, 2013; Vermeulen, Nu
´n
˜ez Castellar & Van Looy, 2014) and
Table 6. ANCOVA Analysis on Assists-per-Kills Ratio.
Sum of Squares df Mean Square FZ
2
Gender 15.80 1 15.80 127.06*** .01
Ever play w Partner 1.78 1 1.78 14.27*** .00
Gender Partner play 2.85 1 2.85 22.88*** .00
Error 2,010.08 16,161 0.12
Total 18,805.72 16,165
Note. ANCOVA ¼analysis of covariance.
**p< .01. ***p< .001.
18 Games and Culture
by guest on January 20, 2015gac.sagepub.comDownloaded from
supports the claim that gender disparity in gaming may result more from cultural
perceptions of gender and gaming, rather than from actual differences in ability or
performance.
The study also found that the portion of females who play the game with a roman-
tic partner is larger than those who do not, as well as larger than the portion of males
who play with romantic partners. These findings are consistent with the girlfriend-
gamer stereotype, which helps explain the further findings that females are more
likely to play Support than males, especially those who play with romantic partners,
suggesting that female players are relegated to play Support, often in support of the
male romantic partners who brought them into the game. In other words, female
players of League find themselves expected, if not required, to stand by their man.
However, while the findings of Study 1 led us to expect that playing with a
romantic partner would have an influence on the competitive abilities of females,
this was not borne out by the results of Study 2. This suggests that female players
may not necessarily be prevented from improving their competitive performance
just because they are fulfilling an expected role within the League community.
Future research should examine this issue more directly, comparing situations in
which playing with a romantic partner can be helpful or detrimental to the perfor-
mance of female players.
Overall Discussion
This research draws from two complementary studies, one qualitative and one quan-
titative, to examine the possible social and psychological factors that may discou-
rage females from playing League, and that prevent those who do from playing as
intensively as their male counterparts. The insights gained from the qualitative
observations were mostly consistent with quantitative data. As explored in Study
1 and supported by Study 2, many female players may be compelled, pressured,
or otherwise directed toward playing the Support role that, though requiring no less
competence than other in in-game roles (and arguably more), is nonetheless seen by
many players as subordinate to, and less desirable than, the role of ADC.
Both studies also suggest that females are not as confident in their gameplay abil-
ity as males. This could be construed as a cause of the gender disparity in the game—
because females are not confident in their abilities, they do not play the game or they
choose to play Support—but we find it unlikely that this is the whole story. Two
other factors, which likely contribute to the systematic gender gap in competitive
games as a whole, are the social climate that is hostile to females, and the stereotype
that females do not belong in, or are not skilled at, the game.
These factors likely hinder the female players from gaining confidence in the
game, much as reminders of a negative stereotype induce people associated with that
stereotype to conform to the negative expectation (Steele & Aronson, 1995). In other
words, many female League players may face a vicious cycle by believing that they
are suitable only for Support roles or, more problematically, that they do not belong
Ratan et al. 19
by guest on January 20, 2015gac.sagepub.comDownloaded from
in the game, female players may refrain from intensive play and/or experimentation
with other game roles, which in turn perpetuates these very stereotypes. Such a cycle
may indeed be demonstrated by the lived experiences of Annie: although she is try-
ing to expand her competencies beyond her normal Support role, she is constantly
compelled to adhere to the gender-normative behaviors we have identified here
by two major pressures, that is, the stereotype of the ‘‘Female Support player’’
(which she herself identifies) and social interactions with male players (e.g., her boy-
friend, other male friends, and online players perceived to be male). Although this
research did not test such theoretical explanations directly, future work in this area
could do so by specifically examining whether reminders of the stereotype that
females are unskilled players has a negative influence on their performance.
As previous research has demonstrated, there is no quick and easy fix to bridge
the gender gap in gaming (Jenson & de Castell, 2010; Lucas & Sherry, 2004). How-
ever, beyond the theoretical implications, our findings suggest that game companies
have the potential to reduce this gap by addressing the social dynamics that propa-
gate these stereotypes, as well as facilitate hostility toward female players. Making
the game more hospitable—both for female and for male players—is at least par-
tially a matter of addressing and curbing the toxicity and negativity that charac-
terizes the majority of players’ in-game verbal communication. Riot Games has
actively taken steps in this direction (McWhertor, 2012), having introduced both
‘‘Honor’’ and ‘‘Reform card’’ systems through which players can either commend
each other for positive and supportive behavior or report each other for abusive
behavior, respectively. Further, Riot Games has instituted a ‘‘Tribunal System’’ in
which community members (i.e., players) review cases of repeated abuse reports and
decide whether to pardon or punish the offending player.
Although encouraging females to join and participate in competitive gaming
spaces would appear to be a worthy goal, we should be cautious of how female
gamers are channeled into these arenas. When we play with others—whether with
a friend, family member, or romantic partner—we must consider whether and how
the existing relationship shapes game play, and one’s experiences within it. As we
have suggested here, playing with a male romantic partner may simultaneously
afford female players entry into the gaming community, while at the same time lim-
iting their participation to remain along stereotypically gendered lines. Moreover,
prior research has found that playing with a romantic partner in general can have
negative impacts on the male partner (Williams et al., 2009). One compelling avenue
for future research is therefore to examine other forms of preexisting player relation-
ships (nonheterosexual romantic couples, nonromantic dyads, etc.) to see whether
and how gender dynamics play out in these different configurations.
Conclusion
These two studies on the perceptions and experiences of female League players
examined the barriers they experience to more equitable participation in, enjoyment
20 Games and Culture
by guest on January 20, 2015gac.sagepub.comDownloaded from
of, and benefits derived from, the play of competitive games. Evidence continues to
mount that digital gaming can be greatly rewarding for many players, not just in
terms of a pleasurable and sociable leisure activity, but as a pathway to positive
interpersonal, vocational, and educational outcomes. As this study documents with
regard to gendered participation in one of the most popular PC games on the market,
these rewards are not distributed equitably. More research is required to not only
document but also possibly help intervene into this current state of play, if female
players are ever to stand apart from their male gatekeepers.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship,
and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of
this article.
Notes
1. Although our focus is on the ways heteronormative relationships play out in the game, we
acknowledge that similar explorations could (and should) be undertaken with regard to the
experiences of Gay, Lesbian, Bisexual, Transgender, and Questioning players (see, e.g.,
Shaw, 2009).
2. In 2007, ‘‘women age 18 or older represent a significantly greater portion of the game-
playing population (31%) than boys age 17 or younger (20%)’’ (ESA, 2007, p. 5).
3. To protect anonymity, we have replaced participants’ names with the names of popular
League of Legends Champions.
4. ‘‘Ranked’’ games refer to games that count toward players’ officially designated status
going from Bronze (lowest) to Platinum (highest). Many participants in Study 1 expressed
frustration with ranked League of Legends play, claiming that these matches included
higher incidences of grievances between players.
5. Both Udyr and Jayce are the names of champions chosen by her teammates.
6. Like many online games, League of Legends has spawned and appropriated a variety of these
short discursive elements (e.g.,‘‘GLHF’’ for ‘‘Good Luck, Have Fun’’ or ‘‘Mid’’ to indicate a
players’ desire and intention to fill that role on the team). As with pings, these statements
carry little potential of revealing any personal information about the issuing player.
7. As seen in the overwhelmingly male player base, as well as the tendency of male players to
play, watch, and in general engage with the game with more frequency.
8. http://leagueoflegends.wikia.com/wiki/Elo_rating_system
References
Alloway, N., & Gilbert, P. (1998). Video game culture: Playing with masculinity, violence
and pleasure. In S. Howard (Ed.), Wired up: Young people and electronic media (pp.
95–114). London, England: UCL Press.
Ratan et al. 21
by guest on January 20, 2015gac.sagepub.comDownloaded from
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change.
Psychological Review,84, 191. doi:10.1037/0033-295X.84.2.191
Bergstrom, K. (2012, May). Virtual inequality: A woman’s place in cyberspace. Proceedings
of the International Conference on the Foundations of Digital Games, ACM, New York,
267–269. doi:10.1145/2282338.2282394
Bryce, J., & Rutter, J. (2003). The gendering of computer gaming: Experience and space. In S.
Fleming & I. Jones (Eds.), Leisure cultures: Investigations in sport, media and technology
(pp. 3–22). Eastbourne, England: Leisure Studies Association.
Carlson, P. (2013, September 12). New League of Legends video uses stats to show how
‘‘rage doesn’t win games.’’ PC Gamer. Retrieved from http://www.pcgamer.com/2013/
09/12/new-league-of-legends-video-uses-stats-to-show-how-rage-doesnt-win-games/
Carr, D. (2005). Contexts, gaming pleasures, and gendered preferences. Simulation Gaming,
36, 464–482. doi:10.1177/1046878105282160
Carr, D., & Oliver, M. (2009). Tanks, chauffeurs and backseat drivers: Competence in
MMORPGs. Eludamos,3, 43–53. Retrieved from http://www.eludamos.org/index.php/
eludamos/article/viewArticle/56/107
Connolly, T. M., Boyle, E. A., MacArthur, E., Hainey, T., & Boyle, J. M. (2012). A systematic
literature review of empirical evidence on computer games and serious games. Computers
& Education,59, 661–686. doi:10.1016/j.compedu.2012.03.004
Consalvo, M. (2012). Confronting toxic gamer culture: A challenge for feminist game studies
scholars. Ada: A Journal of Gender, New Media, and Technology, No.1. doi:10.7264/
N33X84KH
DiSalvo, B. (2012). Glitch game testers: the design and study of a learning environment for
computational production with young African American males. Unpublished doctoral dis-
sertation, Georgia Institute of Technology, Atlanta, GA.
Entertainment Software Association. (2007). Essential facts about the computer and games
industry. Retrieved from http://www.theesa.com/facts/pdfs/esa_ef_2007.pdf
Entertainment Software Association. (2013). Essential facts about the computer and games
industry. Retrieved from http://www.theesa.com/facts/pdfs/ESA_EF_2013.pdf
Feng, J., Spence, I., & Pratt, J. (2007). Playing an Action Video Game Reduces Gender
Differences in Spatial Cognition. Psychological Science,18, 850–855. doi:10.1111/j.
1467-9280.2007.01990.x
Gaudiosi, J. (2012, July 11). Riot Games’ League of Legends officially becomes most played
PC game in the world. Forbes. Retrieved from http://forbes.com/technology
Gray, K. L. (2012). Intersecting oppressions and online communities. Information,
Communication & Society,15, 411–428. doi:0.1080/1369118X.2011.642401
Griffiths, D. N. (2012, February 20). The changing face of games: Feeling angry and ignored,
fans look for someone to blame. Forbes. Retrieved from http://www.forbes.com/sites/
danielnyegriffiths/2012/02/21/bioware-hepler-harassment/
Griffiths, M. D., Davies, M. N. O., & Chappell, D. (2003). Breaking the stereotype: The case
of online gaming. Cyber Psychology & Behavior,6, 81–91.
Jenson, J., & de Castell, S. (2008). Theorizing gender and digital gameplay: Oversights, acci-
dents and surprises. Eludamos. Journal for Computer Game Culture,2, 15–25.
22 Games and Culture
by guest on January 20, 2015gac.sagepub.comDownloaded from
Jenson, J., & de Castell, S. (2010). Gender, simulation, and gaming: Research review and
redirections. Simulation & Gaming,41, 51–71. doi:10.1177/1046878109353473
Jenson, J., & de Castell, S. (2011). Girls @ play. Feminist Media Studies,11, 167–179. doi:10.
1080/14680777.2010.521625
Kahn, A., Ratan, R., & Williams, D. (2014). Why we distort in self-report: The effects cog-
nitive dissonance and balance theory on self-report errors. Journal of Computer-Mediated
Communication,19, 1010–1023. doi:10.1111/jcc4.12056
Kennedy, H. (2002). Lara Croft: Feminist icon or cyberbimbo? On the limits of textual anal-
ysis. Game Studies,2. Retrieved from http://www.gamestudies.org/0202/kennedy/
Kennedy, T. (2007, August). Women’s online gaming communities: Don’t hate the game, hate
the players. Paper presented at the annual meeting of the American Sociology Association,
New York, NY.
Lewis, H. (2012, December 7). Anita Sarkeesian and the gamification of misogyny. New
Statesman. Retrieved from http://www.newstatesman.com/helen-lewis/2012/12/anita-sar-
keesian-and-gamification-misogyny
Lingard, L., Albert, M., & Levinson, W. (2008). Grounded theory, mixed methods, and action
research. BMJ: British Medical Journal,337, 459–461. doi:http://dx.doi.org.prox.lib.ncsu.
edu/10.1136/bmj.39602.690162.47
Lucas, K., & Sherry, J. L. (2004). Sex differences in video game play: A communication-based
explanation. Communication Research,31, 499–523. doi:10.1177/0093650204267930
McWhertor, M. (2012, October 13). The League of Legends team of scientists trying to cure
‘toxic behavior’ online. Polygon. Retrieved from http://www.polygon.com/2012/10/17/
3515178/the-league-of-legends-team-of-scientists-trying-to-cure-toxic
Moore, O. (2012, July 11). Woman’s call to end video game misogyny sparks vicious online
attacks. Globe and Mail. Retrieved from http://www.theglobeandmail.com/news/world/
womans-call-to-end-video-game-misogyny-sparks-vicious-online-attacks/article4405585/
Nardi, B. (2010). My life as a Night Elf priest: An anthropological account of World of
Warcraft. Ann Arbor, MI: University of Michigan Press.
Nguyen, H.-H. D., & Ryan, A. M. (2008). Does stereotype threat affect test performance of
minorities and women? A meta-analysis of experimental evidence. Journal of Applied
Psychology,93, 1314–1334. doi:10.1037/a0012702
Richard, G., & Hoadley, C. M. (2013). Investigating a supportive online gaming community
as a means of reducing stereotype threat vulnerability across gender. Proceedings of
Games, Learning & Society 9.0, ETC Press, Pittsburgh, PA, 261–266.
Riot Games. (2012) Our games. Retrieved from http://www.riotgames.com/our-games
Schreier, J. (2012, February 29). This is whatagamers’sexualharassmentlookslike.
Kotaku. Retrieved from http://kotaku.com/5889415/this-is-what-a-gamers-sexual-har-
assment-looks-like
Shaw, A. (2009). Putting the gay in games: Cultural production and GLBT content in video
games. Games and Culture,4, 228–253. doi:10.1177/1555412009339729
Spencer, S. J., Steele, C. M., & Quinn, D. M. (1999). Stereotype threat and women’s math
performance. Journal of Experimental Social Psychology,35, 4–28. doi:10.1006/jesp.
1998.1373
Ratan et al. 23
by guest on January 20, 2015gac.sagepub.comDownloaded from
Steele, C. M., & Aronson, J. (1995). Stereotype threat and the intellectual test performance of
African Americans. Journal of personality and social psychology,69, 797.
Steinkuehler, C., & Duncan, S. (2008). Scientific habits of mind in virtual worlds.
Journal of Science Education and Technology,17, 530–543. doi:10.1007/s10956-
008-9120-8
Stone, J., Lynch, C. I., Sjomeling, M., & Darley, J. M. (1999). Stereotype threat effects on
Black and White athletic performance. Journal of Personality and Social Psychology,
77, 1213–1227. doi:10.1037/0022-3514.77.6.1213
Taylor, N. (2012). ‘‘A silent team is a dead team’’: Communicative norms in team-based Halo
3play. In G. Voorhees (Ed.), Guns, grenades and grunts: First person shooter games (pp.
251–275). New York, NY: Continuum.
Taylor, N., Jenson, J., & de Castell, S. (2009). Cheerleaders, booth babes, Halo hoes: pro-
gaming, gender and jobs for the boys. Digital Creativity,20, 239–252. doi:10.1080/
14626260903290323
Taylor, N., Jenson, J., de Castell, S., & Dilouya, B. (2014). Public displays of play: Studying
online games in physical settings. Journal of Computer-Mediated Communication,19,
763–779. doi:10.1111/jcc4.12054
Taylor, T. L. (2003). Multiple pleasures: Women and online gaming. Convergence,9, 21–46.
Taylor, T. L. (2006). Play between worlds: Exploring online game culture. Cambridge, MA:
MIT Press.
Thornham, H. (2008). It’s a boy thing: Gaming, gender, and geeks. Feminist Media Studies,8,
127–142. doi:10.1080/14680770801980505
Trepte, S., Reinecke, L., & Behr, K. M. (2009). Creating virtual alter egos or superheroines?
Gamers’ strategies of avatar creation in terms of gender and sex. International Journal of
Gaming and Computer-Mediated Simulations (IJGCMS),1, 52–76. doi:10.4018/jgcms.
2009040104
Vermeulen, L., Nu
´n
˜ez Castellar,E., & Van Looy, J. (2014). Challenging the other: Exploring the
role of opponent gender in digital game competition for female players. Cyberpsychology,
Behavior, and Social Networking,17, 303–309. doi:10.1089/cyber.2013.0331
Wakeford, N. (1999). Gender and the landscapes of computing in an Internet cafe´. In M.
Crang, P. Crang & J. May (Eds.), Virtual geographies: Bodies, space and relations (pp.
178–202). New York, NY: Routledge.
Williams, D., Consalvo, M., Caplan, S., & Yee, N. (2009). Looking for gender (LFG): Gender
roles and behaviors among online gamers. Journal of Communication,59, 700–725.
Williams, D., Ducheneaut, N., Xiong, L., Zhang, Y., Yee, N., & Nickell, E. (2006). From tree
house to barracks: The social life of guilds in World of Warcraft. Games & Culture,1,
338–361. doi:10.1177/1555412006292616
Williams, D., Kennedy, T., & Moore, R. (2011). Behind the avatar: The patterns, practices,
and functions of role playing in MMOs. Games and Culture,6, 171–200. doi:10.1177/
1555412010364983
Williams, D., Martins, N., Consalvo, M., & Ivory, J. (2009). The virtual census: representa-
tions of gender, race and age in video games. New Media & Society,11, 815–834. doi:10.
1177/1461444809105354
24 Games and Culture
by guest on January 20, 2015gac.sagepub.comDownloaded from
Witkowski, E. (2013, August) Following Ms_Fabulous: Women, live-streaming, and do-it-
yourself visibility in e-sports. Paper presented at the meeting of the Digital Games
Research Association (DiGRA), Atlanta, GA.
Yeung, N. C. J., & von Hippel, C. (2008). Stereotype threat increases the likelihood that
female drivers in a simulator run over jaywalkers. Accident Analysis and Prevention,
40, 667–674. doi:10.1016/j.aap.2007.09.003
Author Biographies
Rabindra A. Ratan is an assistantprofessor and AT&T Scholar at Michigan State University’s
Department of Telecommunication, Information Studies and Media. His research focuses on
the psychological experience of media use, with an emphasis on video games and other
interactive environments (e.g., virtual worlds and the road) that include mediated self-
representations (e.g., avatars and automobiles). He is particularly interested in how different
facets of mediated self-representations (e.g., gender and social identity) influence the psycho-
logical experience of media use and how different facets of this psychological experience
(e.g., avatar-body schema integration and identification) affect a variety of outcomes, includ-
ing cognitive performance, learning, health-related behaviors (e.g., food choice and driving
aggression), and prejudicial/prosocial attitudes.
Nicholas Taylor is an assistant professor of Digital Media in the Department of Communi-
cation at North Carolina State University. His work applies critical, feminist and sociotechni-
cal perspectives to experimental and mixed-methods research with digital gaming
communities. In particular, he is interested in the intersections of subjectivity, communicative
practice, technologies, and games, as enacted through both game production and play across a
variety of contexts. He is also the codirector of Circuit Studio, a collaborative research studio
and makerspace at NC State.
Jameson Hogan is a PhD student in the Communication, Rhetoric, & Digital Media program
at North Carolina State University. His scholarly pursuits tend to fall in the areas of digital
humanities, game studies, narratology, and critical making.
Tracy Kennedy is a Digital Culture Instructor at Brock University in St. Catharines, Ontario.
She is also a Gameplay Analyst in the Video Game Industry.
Dmitri Williams is an associate professor at the Annenberg School for Communication at the
University of Southern California. His research focuses on online community behavior, tech-
nology, and methodological development.
Ratan et al. 25
by guest on January 20, 2015gac.sagepub.comDownloaded from
