Social websites like Facebook enable users to upload self-created digital images; it is therefore of interest to see how gender is performed in this domain. A panel used a literature review of pictorial features associated with gender traits, and a sample of Facebook pictures to assess gender stereotypes present in Facebook images. Traits emerging in greater prominence in pictures of males included active, dominant, and independent. Those prominent with female users included attractive and dependent. These findings generally conform to gender stereotypes found in prior research and extend the research regarding stereotypical gender traits displayed in professional media depictions to self-selected social media displays. They also extend the research on gender differences in impression management generally, in both interpersonal communication and social media, to include gender-specific traits that are part of young mens and women’s impression management. Keywords: Facebook; Gender Display; Impression Management; Role Theory; Social Media
Jessica Rose (B.A., Villanova University, 2011) is a marketing and communications professional in the Greater Philadelphia Area. Susan Mackey-Kallis (Ph.D., Penn State University, 1986) is an associate professor in the Department of Communication at Villanova University. Len Shyles (Ph.D., Ohio State University, 1981) is an associate professor in the Department of Communication at Villanova University. Kelly Barry (B.A., Villanova University, 2011) is a marketing and communications professional in the Greater New York area. Danielle Biagini (B.A., Villanova University, 2011) is a marketing and communications professional in the Greater San Diego area. Colleen Hart (B.A., Villanova University, 2011) is a student at the University of North Carolina School of Law. Lauren Jack (B.A., Villanova University, 2011) is a marketing and communications professional in the Greater New York area. The authors would like to thank Dr. Jesse Frey of the Mathematics Department of Villanova University for his help in creating the tables presented in this article. Correspondence: Susan Mackey-Kallis, Department of Communication, 800 E. Lancaster Avenue, Villanova, PA 19085; E-mail: [email protected] ISSN 0146-3373 print/1746-4102 online # 2012 Eastern Communication Association DOI: 10.1080/01463373.2012.725005
The centrality of gender embodiment has animated recent debates in media studies about the relationship among gender representations in media, gendered bodies in virtual space, and gender as performance. With the emergence of social media websites, such as Twitter, Facebook, and MySpace, users have an online platform that allows them to communicate widely, to virtually manage others’ impressions of them, and to even express gendered identities in cyberspace. With over 500 million active users as of 2011 (http://www.facebook.com/press/info.php?statistics), Facebook dominates the social media market. Offering a highly interactive platform, Facebook users can leave comments on their friends’ walls, provide status updates and photos, and can even access one another wirelessly through Facebook Mobile. As of 2010, users spent over 700 billion minutes per month on Facebook (http://www.facebook.com/ press/info.php?statistics), often updating their Facebook profiles to add relationships to their friends lists.
One well-known feature of Facebook is the user’s profile picture, displayed in the upper left-hand corner of each user’s homepage. Intended to be the first thing seen, it is arguably one of the most important features of the user’s Facebook page. The profile picture offers friends, acquaintances and even potential employers a firstimpression of the user’s appearance and, perhaps, their character. Therefore, the content of users’ Facebook profile pictures is an important object of study for researchers interested in how people practice impression management. According to role theory, people follow unwritten social and cultural rules and norms as they behave ‘‘in ways that are different and predictable depending on their respective social identities and the situations they find themselves in’’ (Biddle, 1986, p. 68).
Gender identity and gender roles are a significant part of everyday life and, according to Goffman, are actually constituted through social interaction (1976). Gender shapes how people make sense of themselves and their social relationships. However, as Wood notes, ‘‘What gender means depends heavily on cultural values and practices; a culture’s definitions of masculinity and femininity shape expectations about how individual men and women should communicate; and how individuals communicate establishes gender that, in turn, influences cultural views’’ (Wood, 2009, p. 20). Gender display, as a continuous communication loop, is defined by society and expressed by individuals as they interact while shaping evolving societal expectations regarding gender.
In contemporary media and culture, women’s and men’s social desirability and gender have often been defined in terms of their bodies. For women, this has often involved comparing themselves to and even replicating the ‘‘thin ideal’’ (presented in modern mass media despite the looming specter of anorexia and bulimia), altering their bodies to heighten perceived sexuality or youthfulness (through cosmetic surgery, exercise or eating), or conforming to traditional definitions of femininity including qualities such as submissiveness or sentimentality (through dress, cosmetics, style, etc.). For men, gender-based definitions of success frequently revolve around presenting or developing their bodies as strong, youthful, active, and physically dominant. The evolution of social media (such as Facebook) and online digital gaming environments (such as massive, multiple online games [MMOs]) now offer venues where individuals can consciously self-select and present virtual versions of themselves that can either conform with, challenge, or defy societal expectations and media presentations.
Digital formats, on the one hand, represent exciting possibilities for individuals who can explore the freedom of presenting a physical self that might differ from the one they present or perform in everyday life or from socially-defined expectations. As Rettberg argues, ‘‘our fascination with creating digital self-portraits is indicative of our collective coming of age where we as a culture are discovering that we have voices online and can express ourselves rather than simply accepting the mass media’s views of the world’’ (2009, p. 453). On the other hand, digital formats may simply offer a chance to replicate cultural and mass media normative versions of the individual, specifically as they relate to gender. Since how people present and perform their bodies in virtual spaces offer specific impressions, many of which relate to gender, one objective of the current study is to examine how gender portrayals manifest themselves in self-selected social media displays.
Specifically, can social media website content help us understand more about gender roles and the way people present themselves in the virtual social world? Do the ways they manage their images reinforce existing gendered stereotypes? Because social networks such as Facebook are relatively recent phenomena, the content of self-presentation profile pictures has not been analyzed in great depth. Extant literature supports the idea of expected gender roles unique to males and females (Goffman, 1976; Lauzen, Dozier, & Horan, 2008; Wanta & Legett, 1989; Williams & Best, 1990) and the idea that society advocates these roles through various media (Bell & Milic, 2002; Hancock & Toma, 2009), with both males and females engaging in impression management in order to control their public image (Dominick, 1999; Jones, 1997; Leary, 1996).
Witmer and Katzman (1997) argued that females may display more emotional graphics than males while communicating on the Internet. Extant literature, however, only reveals one study that has examined gender differences in self-selected portraits in self-posed photographs (Mills, 1984) and one study that has examined gender differences in Facebook profile pictures (Strano, 2008). Strano’s study, focusing exclusively on gender difference in impression management, found that women engage in management more than men (Strano, 2008). In a related area of literature, self-presentation in computer-mediated communication (CMC), some studies suggest a great deal of ‘‘gender-swapping’’ on the Internet (Bruckman, 1993; Roberts & Parks, 1999; Witmer & Katzman, 1997) with some estimates as high as 60% (Roberts & Parks, 1999), and with males ‘‘gender-swapping’’ more than females (Bruckman, 1993; Suler, 1999).
The fantastical and fantasy-based nature of many gaming environments, however, and the anonymous nature of most CMC in general might suggest that the freedom to reinvent oneself not only in terms of gender but also race, ethnicity, and other variables is much broader than in Facebook profile pictures. Facebook friends, who may know the person in real life, recognize a photograph as a self-selected presentation but, researchers argue, most likely do not assume that the profile picture reflects extensive alteration or photo retouching. The current study, therefore, makes a unique contribution by investigating whether self-selected Facebook profile pictures exhibit stereotypical gender roles consistent with traits emerging from existing research.
Literature Review Gender Roles Some researchers suggest that gender differences result from a variety of factors including socialization and biology; as such, gender roles are often manifested through communication and culture (Goffman, 1976; Lauzen et al., 2008; Wanta & Legett, 1989; Williams & Best, 1990; Wood, 2009). West and Zimmerman (1987) claim gendering is a routine interaction of everyday life. ‘‘Both gender role and gender display focus on behavioral aspects of being a man and a woman’’ (p. 127). They use Goffman’s (1976) account of ‘‘gender display’’ to suggest gender is also constituted through interaction. Says Goffman, ‘‘If gender [can] be defined as the culturally established correlates of sex (whether in consequence of biology or learning), then gender display refers to conventionalized portrayals of these correlates’’ (1976, p. 69). Goffman cites sports as a framework to explain masculine tendencies, asserting that the male gender is categorically viewed as aggressive, strong, and competitive. In agreement, West and Zimmerman (1987) assert that ‘‘Doing gender is unavoidable . . . because of the social consequences of sex-category membership: [this includes] the allocation of power and resources not only in the domestic, economic, and political domains but also in the broad arena of interpersonal relations’’ (p. 145).
As men and women tend to assume ‘‘proper’’ societal gender roles, associated behaviors are viewed as cultural markers that indicate norms of social interaction. Williams and Best (1990) searched for gender stereotypes among respondents from 25 nations worldwide. Participants were presented with a list of 300 character traits and instructed to indicate whether the trait was ‘‘more frequently associated with men than with women,’’ ‘‘more frequently associated with women than with men,’’ or ‘‘not differentially associated with the two sexes.’’ Table 1 presents results for the traits most commonly associated with men and women and indicates the traits isolated for this analysis. Gender Roles in the Media Society often promotes gender role markers as social norms through photographs and other visual displays used in advertising. Wanta and Legett (1989) studied the media images of male and female athletes of the 1987 Wimbledon Tennis Tournament, concluding that men and women were depicted differently in terms of emotion, dominance, and power.
Goffman (1976) accounts for these traits in his research of magazine and newspaper photography, finding women to be pictured in more submissive positions while men are depicted in more elevated positions. Based on Goffman’s inquiries, Wanta and Legett (1989) hypothesized that female tennis players would be shown more often in positions implying helplessness than male tennis players. Goffman’s (1976) studies of power within photographs asserted that the more dominant a person’s face was (i.e., the more full-front, direct-to-camera orientation of the face, and the greater the percentage of photo space taken up by the face), the more power was held and=or portrayed by the person pictured. Wanta and Legett used these ideas to predict that the photographs of female tennis players would focus more on the players’ bodies, while male tennis players would have more concentrated images of their faces. However, the majority of Wanta and Legett’s (1989) hypotheses remained unsubstantiated; their gender stereotypes were not confirmed. In fact, opposite portrayals often emerged.
They concluded that the photographer was trying to break gender stereotypes. In contrast to the work of Wanta and Legett, the research conducted in the current study does not rely on images shot by professional photographers. Rather, the present analysis is based on self-selected and, almost exclusively, self-created Facebook profile pictures. Gender roles, present in everyday interaction, are also enacted on television. Lauzen and colleagues (2008) examined gender roles enacted by men and women on television.
Using a stratified random sample of 124 prime-time television series airing on six broadcast networks during the 2005–06 seasons, they looked at the rates at which men and women fell into categorically different social roles. Taking a category scheme developed by earlier research, Lauzen and colleagues defined social roles as the things ‘‘people do in daily life’’ (see Eagly & Steffan, 1984, p. 735). These roles vary from childcare and household chores to workplace activities. Through a content analysis, they found male characters on prime-time television were more likely to inhabit work roles, including blue collar, white collar, and extracurricular activities, while women were portrayed in more interpersonal roles involving romance, friendship, and family.
Similarly, in their content analysis of 827 Australian magazine advertisements from 1997–98 to determine the presence of stereotypical gender roles, Bell and Milic (2002) concluded that ‘‘Males were more frequently shown in ‘narrative’ ways (as actors) than females, and this is true of both groups and individuals. Women were more likely than men to ‘behave’ (or to express emotion)’’ (p. 215). Their findings suggest stereotypical gender traits of men and women consistent with those of Williams and Best (1990) and consistent with Goffman’s (1976) analysis of advertising which found that women were ‘‘more likely to be portrayed performing submissive or appeasing gestures such as head or body canting, bending one knee inward (‘bashful knee-bend’), smiling, clowning, and acting less seriously’’ and were ‘‘often portrayed as being under the physical care and protection of a man’’ (as cited in Bell and Milic, 2002, p. 205).
Ragan (1982) analyzed gender differences in 1,296 portrait photos from high school and university yearbooks, concluding there are gender differences; females smiled more than males, smiled more expansively than males, tilted their heads at greater angles than males, faced the camera less directly than males, and wore glasses less frequently than males. While this research identifies gender differences, it was limited by an influential factor: Photographers posed the subjects (Ragan, 1982). In hopes of accounting for this limitation, Mills (1984) conducted a study in which 34 men and 34 women were asked to present themselves as typical college students in pictures. Mills’ findings reinforced the suggestion that females smile more, and smile more expansively, than males. Gender stereotypes also abound in video games.
Female characters are represented as highly sexualized while male characters possess exaggerated strength, are hypermasculine, aggressive, and, with the exception of showing hostility, lack emotion. They are also less likely to display helping or nurturing qualities (Robinson, Callister, Clark, & Phillips, 2008). One study found that the central role for male characters was ‘‘competitor’’ while females’ central roles were ‘‘victim,’’ ‘‘damsel in distress,’’ or ‘‘evil obstacle’’ for the hero to overcome (Heintz-Knowles et al., 2001). The findings of these video game content analyses have remained fairly consistent over time and have also been shown to be perceived by audiences (Robinson et al., 2008).
Based on the work from several decades of research on gender roles from the fields of advertising, television, photography, digital gaming studies, and cultural studies, it is apparent that masculinity often implies strength, ambition, and independence, whereas femininity implies physical attractiveness, reverence, and sentimentality (Wood, 2009). The rapid growth of digital media invites researchers interested in the cultural impact of gender to investigate this alternative outlet for self-presentation. This study provides a natural extension of such work in investigation of gender differences in self-selected Facebook profile pictures. Self-Presentation Goffman (1959) argued that individuals were concerned with self-presentation during all social encounters. This is because, among other reasons, impressions impact the opinions of others regardless of an individual’s intentions. Burr (2002) claims J. Rose et al.
The other people making up our audience can, by their own conduct, either legitimate or reject our claim to be a certain kind of person, and Goffman (1959) argues that this is done by carefully monitoring the match or mismatch between what we ‘give’ (the things we say or do to create an impression) and what we ‘give off’ (the body language, our general demeanor—the communicative aspects of our conduct that are harder for us to control and manipulate). The creation and maintenance of impressions is therefore a two-way street (2002, p. 73).
Hence, for Goffman (1959), the ‘‘presentation of self in everyday life’’ and the roles maintained are pertinent to everyday interaction. People constantly play characters to avoid embarrassment and to ‘‘fit-in’’ with social norms. Gender role, then, focuses on the collectivity of logical, behavioral, cognitive, and emotional responses to social situations (Burr, 2002). Most self-presentation studies have examined the concept only in face-to-face communication (Goffman, 1959; Leary 1996). Recent studies (Oh, 2004; Cho, 2006) about self-presentation in personal websites analyze only the styles and not gender display specifically. Similar to face-to-face contexts, individuals do make choices about gender-related impressions over the Internet. A number of studies, for example, have demonstrated extensive ‘‘gender-swapping’’ in ‘‘avatar’’ creation for online gaming and in text-based CMC (Bruckman, 1993; Roberts & Parks, 1999; Suler, 1999).
In these virtual environments, physical identity markers are not apparent and, as a result, the self is more fluid and changeable (Gergen, 1991) and offers increased opportunities for strategic self-presentation (Walther, 1993; Walther, Anderson, & Park, 1994). Gender is often one of the variables that communicators can consciously shape in these mediated environments (Roberts & Parks, 1999; Bruckman, 1993; Wilbur, 1996). Some researchers have gone so far as to argue that the computer-mediated environment is a gender-bending world (Witmer & Katzman, 1997). Modern gaming environments, in particular, allow gamers to design or choose ‘‘avatars,’’ their virtual self in the gaming world, that possess a variety of differing characteristics such as height, weight, age, gender, dress, and profession. In these environments, the avatar becomes ‘‘inextricably linked to their performance of self and engagement in [a virtual] community’’ (Taylor, 1999, p. 438).
Despite the above cited studies of ‘‘gender-swapping’’ and self-presentation in CMC, no researchers have examined the extent to which social media users ascribe to gender stereotypes in their presentation of self on the Internet. Samp, Wittenberg, and Gillett (2003) examined the extent to which ‘‘gender schematic’’ individuals (individuals with either strong masculine or feminine gender orientations versus androgynous orientations) and individuals who were high (versus low) self-monitoring engaged in gender-swapping on the Internet. The researchers in this study used self-report data from Internet users about their online gender-swapping behavior generally.
The researchers in the current study, by contrast, provide a content analysis of actual Facebook profile pictures in terms of the presence or absence of gender stereotypes and do not focus on gender swapping. Hancock and Toma’s 2009 study of profile pictures on online dating websites created and posted with the intention of creating relationships comes closest to the focus of the current study. In line with Goffman’s (1959) suggestion that self-presentation is the process of packaging and editing the self in order to create a certain impression for an audience, Hancock and Toma (2009) examined the impact of gender on self-presentation and social desirability. They found that both women and men ‘‘edit’’ their profiles to create a better self-presentation through self-enhancement (Hancock & Toma, 2009). Some participate in ‘‘selective self-presentation,’’ an even more controlled act of impression management in which images are changed or distorted, often leading to further inaccuracy portrayed by the profile (Hancock & Toma, 2009).
Having the ability to ‘‘select’’ or specifically change or display particular points of interest, the users can greatly affect the impression made of them (Hancock & Toma, 2009). Both men and women on the online dating social network use ‘‘selective self-presentation’’ to their advantage to give the impression of being more desirable to their audiences. Hancock and Toma (2009) suggest men and women can control their self-presentation through social networks (i.e., online dating sites). Such sites comprise one segment of social networks; Facebook is another. Realizing stereotypical gender roles are present in society, Dominick (1999) studied how men and women presented themselves on personal homepages. Dominick (1999) coded 500 randomly sampled personal homepages based on demographic and personal information, creative expressions, and photographs. He used Jones’ (1997) five strategies of image construction: ingratiation (statements of modesty, familiarity, and humor); competence (statements of abilities and achievements); intimidation (statements of anger and unpleasantness); exemplification (acts of moral superiority); and supplication (images of helplessness, while acting self-deprecating; Dominick, 1999).
He concluded that females released more information than males while both males and females were equally likely to have photographs on their pages. Women’s photographs tended to be more sentimental in nature, while men’s more often were ‘‘joke images’’ and images that made them seem more competent and capable. He concluded that ‘‘A personal web page can be viewed as a carefully constructed selfpresentation’’ (Dominick, 1999, p. 647). Dominick asserted that the concept of impression construction exposes the different strategies men and women use to present themselves through images and information to gain a higher level of likeability, respect, and power in society (1999). Jones (1997) noted that individuals strive to be liked and accepted, resulting in social rewards such as friendship, social support, companionship, romance, and social status.
Because smiling is associated with being liked and competent, Jones (1997) correlates gestures with the ingratiation and competence strategies of image construction. As Facebook was not founded until 2004, Dominick’s (1999) study is expanded upon in this study (http://www.facebook.com/press/info.php?factsheet). Buffardi and Campbell (2008) studied whether photographs from a variety of social networks are self-promoting. They state, ‘‘Self-promoting connoted persuading others about one’s own positive traits’’ (p. 1307) and define physical attractiveness as the degree to which an individual appears self-promoting and vain in a photograph. While not examining gender differences in self-promotion per se, the researchers did examine ‘‘how sexy and modest . . . the individual in the main photo appeared to be’’ (Buffardi & Campbell, 2008, p. 1307).