Research Critique

Computers in Human Behavior 27 (2011) 1828–1833

Contents lists available at ScienceDirect

Computers in Human Behavior

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / c o m p h u m b e h

A picture is worth a thousand words: A content analysis of Facebook profile photographs

Noelle J. Hum a, Perrin E. Chamberlin a, Brittany L. Hambright a, Anne C. Portwood a, Amanda C. Schat a, Jennifer L. Bevan b,⇑ a Department of Communication Studies, Chapman University, 1 University Drive, Orange, CA 92866, United States b Department of Communication Studies/M.S. Program in Health and Strategic Communication, Chapman University, 1 University Drive, Orange, CA 92866, United States

a r t i c l e i n f o

Article history: Available online 4 May 2011

Keywords: Facebook Profile photographs Online identity construction Gender

0747-5632/$ - see front matter � 2011 Elsevier Ltd. A doi:10.1016/j.chb.2011.04.003

⇑ Corresponding author. Address: Department of Program in Health and Strategic Communication, Chap Drive, Orange, CA 92866, United States. Tel.: +1 714 53

E-mail address: [email protected] (J.L. Bevan).

a b s t r a c t

This research examines identity construction and gender roles in social networking sites by studying and comparing the profile photographs of male and female Facebook users. Specifically, the number of photos in the profile album and the content of the main profile picture are studied by coding specific pictures, and determining if the content and amount of profile pictures differ significantly by gender. Participants include male and female Facebook users between the ages of 18 and 23 who are currently enrolled in a college or university. Profile pictures tended to be inactive, posed, appropriate, and only including the subject. The content and amount of Facebook profile photographs also did not significantly vary by gen- der. Implications of these findings, as well as suggestions for future research, are discussed.

� 2011 Elsevier Ltd. All rights reserved.

1. Introduction

‘‘OKAY. seriously, bye Facebook. SEE YOU GUYS JULY 16th.’’

‘‘I tried deactivating my facebook and it worked for 48 h. epic fail.’’

These ‘‘status updates,’’ short posts by Facebook users to update their online friends on what their current state of affairs or emo- tions are, are examples of how pervasive social networking sites (SNSs) such as Facebook can be. At the end of 2009, 75% of online adults ages 18–24 had a profile on an SNS (Lenhart, 2009). Couple that growing number with the 500 million active Facebook users who spend over 700 billion minutes per month on the site, and it becomes quite obvious just how encompassing and time-consum- ing SNSs like Facebook are today (Facebook, 2010). It is no wonder that, for the first time, Facebook was the most-visited website in 2010 (Experian Hitline, 2010).

Facebook was launched in 2004 as a ‘‘social utility that helps peo- ple communicate more efficiently with their friends, family and coworkers’’ (Facebook Factsheet, 2010, para. 1). By building a profile, each Facebook user is able to post notes, photos, links, and videos to be shared with ‘‘friends’’; that is, other members who are connected to an individual’s online social network, and thus granted access to view the individual’s profile. The ‘‘Home Page’’ allows each Facebook user to be constantly updated on the most recent postings and

ll rights reserved.

Communication Studies/M.S. man University, 1 University 2 7768; fax: +1 714 6287237.

interactions of and among friends. Facebook users can also enable ‘‘Facebook Chat’’ to instant message online friends in real time (Face- book Factsheet; Junco & Cole-Avent, 2008).

Taken together, these elements of Facebook combine to allow its users to construct an image or identity to communicate to the great- er online community. Perhaps one of the most telling pieces of self-disclosure or image construction is the profile photo, the single default photo by which Facebook users choose to identify them- selves within the entire network (Watson, Smith, & Driver, 2006). Hancock and Toma (2009) noted that, ‘‘With the emergence of profile-based social networking sites like MySpace and Face- book. . .online self-presentations are no longer limited to text-based descriptions. The profile photograph is now a central component of online self-presentation, and one that is critical for relational success’’ (p. 368). Indeed, 98.7% of college students in one study re- ported that they posted a photo of themselves on Facebook (Young & Quan-Haase, 2009). Despite the prevalence and importance of this image, little known research has analyzed the content of Facebook profile photographs. As such, the goal of our study is to explore what Facebook users are posting in their profile photos, as well as how this content differs by gender. Doing so will expand on the existing knowledge surrounding social media and the limits of it with respect to how and why different subpopulations of young adults use and perceive the multitude of benefits and aspects of SNSs.

1.1. Facebook and identity construction

It is not the mere use of Facebook that has warranted social media research, but the motives and behaviors behind the online

N.J. Hum et al. / Computers in Human Behavior 27 (2011) 1828–1833 1829

activity. Recent studies have examined these motivations, citing various reasons such as to gain social capital by initiating and maintaining friendships (Lenhart, 2009; Valenzuela, Park, & Kee, 2009) and to create and enhance a self-image (Utz, 2010; Zhang, 2010). These studies are a part of the increasing literature on the effects of computer mediated communication (CMC) to support the idea that the use of SNSs enhances, not detracts from, the growth of interpersonal relationships and social skills (Valenzuela et al.).

As mentioned, there has recently been a substantial increase in scholarly literature investigating the various motives for using SNSs as a way to identify the self, and to connect the self with oth- ers. Social capital is a concept that is separated into three domains: interpersonal, intrapersonal, and behavioral (Scheufele & Shah, 2000). The interpersonal domain encompasses the interactions one has with his or her social network; the intrapersonal domain describes the inner emotions and levels of self-satisfaction within the individual; and the behavioral domain involves the individual’s participation in civic and political activities (Scheufele & Shah). The use of Facebook was found to be positively associated with higher levels of self-satisfaction and social trust (Valenzuela et al., 2009); in other words, individuals who maintained an online identity felt more connected with their peers and possessed an overall higher level of happiness and social contentment. This finding was also supported by Ellison, Steinfield, and Lampe (2007), who concluded that the perception of being connected to others is partly due to the convenience and free-of-cost services that Facebook provides, such as daily reminders of friends’ birthdays.

Additionally, establishing social connections is positively linked with establishing a social identity, and Facebook users successfully do this by indicating membership of certain subgroups (race, gen- der, sexuality, etc.) and subcultures (music, movies, etc.) (Pempek, Yermolayeva, & Calvert, 2009). Respondents to the Pempek et al. study indicated that one of the most important uses of Facebook was to not only learn information about others, but to reconnect with real-life friends. One of the largest appeals of Facebook there- in lies in the ability to expand and strengthen one’s social network.

To highlight another dimension of social capital, Facebook Groups (an application feature that allows a user to join a subnet- work of people with a common interest) have the power to in- crease civic and political participation through a reciprocal relationship (Valenzuela et al., 2009). The fact that young voters use Facebook to communicate their political opinions and that Facebook (Groups) uses the community of registered voters to dis- seminate political information illustrates the effectiveness of Face- book to unite and empower communities with similar interests and views (Valenzuela et al.).

If one of the benefits of Facebook is to bring individuals in a community together, then it also makes sense that Facebook pro- vides a means for self-expression in order to form these social, geo- graphical, and political connections. The idea of self-construction in a nonymous (i.e., the opposite of ‘‘anonymous’’) setting such as Facebook is critical to our understanding of how and why indi- viduals communicate on and through SNSs. Zhao, Grasmuch, and Martin (2008, p. 211) identified ‘‘hoped-for, possible identities’’ as social networking identities that were not necessarily consistent with real-life personalities. These online identities were often shaped through positive word affirmations that described an indi- vidual as socially desirable and outgoing (Zhao et al.). Furthermore, these identities were carefully constructed to reflect social and cul- tural norms; researchers noted that the creation of seemingly sep- arate online identities is not an innate behavior, but rather a response to the social conditions and environment that the indi- vidual is placed in (Zhao et al.).

Researchers have also found that Facebook users generally con- struct their self-identity through indirect, mimetic ways that show

viewers glimpses of their personalities instead of explicit, direct cues (Zhao et al., 2008). For example, a participant’s Facebook sta- tus publicly claimed her love and devotion to her husband. This behavior not only reinforces what Zhao et al. describes as affirma- tions of accepted societal norms (in this case, heterosexuality), but also illustrates how users create identities through implicit com- munication, leaving ‘‘clues’’ for viewers to pick up and interpret. Utz (2010) expanded on this theory when she studied online impression formation through inferences made when looking at an individual’s Facebook friends. When looking at communal traits, which enhance one’s social desirability, respondents judged the le- vel of these traits based on the impressions received from the indi- viduals’ friends; that is, the profiles of the individual’s online followers (Utz). This type of online behavior is often equated to nonverbal behavior in traditional face-to-face communication; context cues derived from implicit sources are perceived as more indicative of one’s identity, thus working to assist greatly with impression formation (Tidwell & Walther, 2006).

Perceptions of social desirability are not the only inference Facebook users attempt to determine from others’ profiles. Walther, Van Der Heide, Kim, Westerman, and Tong (2008) found that the level of attractiveness of one’s friends had a significant im- pact on the perception of attractiveness of the user whose profile was being examined. Additionally, Walther et al.’s results were consistent with Utz’s (2010) finding that friends who were per- ceived as outgoing and social helped to increase the perception of social attractiveness of the Facebook user. These perceptions of sociable friends were formed after analyzing photos and wall posts (Walther et al.). A Facebook profile then, is the product of not only self-generated information, but of a combination of that and the inferences made from indirect sources of online communication.

This idea of self-construction in an online environment war- rants further investigation into how subgroups of young adults/ college students utilize SNSs to create an image of themselves. Sev- enty-four percent of students reported that their Facebook profiles were accurate representations of themselves (Stern & Taylor, 2007), suggesting that identity construction is a key consideration when determining which information to post or not. According to Bugeja (2006), Christine Rosen, a fellow at the Ethics and Public Policy Center in Washington DC, commented that Facebook and similar sites are not really about the fostered connections in the online community, but rather serve as vehicles by which individu- als can participate in ‘‘egocasting, [which is] the thoroughly per- sonalized and extremely narrow pursuit of one’s personal taste.’’ Rosen further argues that Facebook users ‘‘have a tendency to de- scribe themselves like products’’ (Bugeja, 2006, p. 2). Indeed, a study conducted by Pempek et al. (2009) found that self-presenta- tion was one of the most popular reasons college students were at- tracted to, and continued to use, Facebook. Pempek et al. argued that perhaps one of the biggest reasons for being attracted to care- fully selected self-presentation is to ‘‘resolve key developmental is- sues’’ such as identity and intimacy development (p. 236).

The findings by Zhao et al. (2008) that established the idea of showing, not telling, when constructing online identities support our argument that the study of content of implicit communication cues of Facebook profiles, i.e., profile photos, would add to the growing knowledge of identity construction in social media con- texts. Mesch and Beker (2010) concluded that the norms of offline self-disclosure did not necessarily coincide with the separately developed norms of online self-disclosure. Online self-disclosure was more open and expressive, perhaps due to the perception of anonymity in online communities (as opposed to face-to-face interactions) or the more intense need for uncertainty reduction (Mesch & Beker). Foon Hew’s (2011) review of students’ uses of Facebook similarly indicated that greater self-disclosure occurred on Facebook than offline. Whatever the real reason may be,

1830 N.J. Hum et al. / Computers in Human Behavior 27 (2011) 1828–1833

researchers understand that there are separate rules that establish and govern the norms of self-disclosure in online identities versus face-to-face communication settings (Mesch & Beker). With the above research on Facebook and identity construction in mind, we thus offer our first research question, which explores exactly how many Facebook profile pictures college students utilize and what the content of these pictures consists of:

RQ1: What is the content and amount of college students’ Face- book profile photographs?

1.2. Gender and Facebook behavior

The rules that govern online disclosure norms are culturally and socially created, but do all groups, specifically males and females, respond in the same way? According to social role theory (Eagly, 1987; Eagly, Wood, & Diekman, 2000), socially constructed gender roles are considered when accomplishing primary goals and creat- ing positive impressions or identities. Support for gendered social roles – which characterize females as more communal caregivers and males as more agentic providers – in online environments was recently provided by Guadagno, Muscanell, Okdie, Burk, and Ward (2011), who determined that Second Life behaviors differed according to gender. On Second Life, which is a virtual, interactive environment where users create visual avatars, female users were found to be more likely to meet new people, shop, and change their avatars’ appearance than males.

Further, some of the only known research that examined the content of Facebook profile photographs considered this content according to user gender. Specifically, the notion of face-ism, which is a cross-cultural form of gender stereotyping that assumes the fo- cus of males is on his face and head, whereas females’ bodies are more central, was examined (Reichart Smith & Cooley, 2008). Cross-culturally, male Facebook users’ profile photographs were found to display greater scores on the face-ism index (i.e., the rel- ative prominence of one’s face in an image) than female users’ pho- tos (Cooley & Reichart Smith, 2010; Reichart Smith & Cooley). The authors conclude that gender stereotypes remain present even when users themselves can select the images that represent them in their Facebook profile photographs. As such, in line with social role theory and previous face-ism findings, we seek to continue this line of research to determine if other aspects of Facebook pro- file photograph content also varies by the user.

We also hope to add to the minimal, but growing, research on gender and the use of SNSs. For example, one study found that, upon analyzing groups of Facebook users who posted images of alcohol and marijuana-related photos, females posted photos of alcohol use more often and males were more likely to post photos of marijuana use (Morgan, Snelson, & Elison-Bowers, 2010). This finding was interpreted to possibly mean that for young adult males, marijuana use was considered more socially acceptable, or even socially desirable, whereas young adult females posted pho- tos of alcohol use to foster connections with others or to maintain relationship management (Morgan et al.). These findings were con- sistent with Bond (2009), who found that females were more likely than males to ‘‘include images pertaining to friends, family, signif- icant others, holidays, school, and alcohol’’ (p. 5). Male participants, in comparison to the females, uploaded more sports-related photos (Bond). Interestingly, Morgan et al. also noted that photo disclo- sure by women could be generally motivated by the desire to store data and memories.

The act of online photo disclosure was also studied by Mesch and Beker (2010), who found that while women are more likely than men to post photos (and allow these photos to be publicly ac- cessed), men are more likely to post videos of themselves on Face- book. Generally speaking, however, both men and women

disclosed close to the same amount of personal information online and were equally as likely to post an image of themselves (Young & Quan-Haase, 2009). This is a phenomenon that is not seen in tradi- tional face-to-face interaction, thus supporting the idea that online self-disclosure follows a separate, and sometimes different, set of norms and rules.

Other research indicates that women are generally more active than men in online communities where information is shared and topics are discussed (Sussman & Tyson, 2000). The behavior of men and women on the Internet was reflective of social theories sur- rounding gender differences. In political and news areas, men posted six times more often than women about their opinions and stances on specific situations; women were more likely to ini- tiate discussions (Sussman & Tyson). Additionally, Bond (2009) found that females self-disclose more than males on SNSs. These findings are consistent with social role theory’s idea that females behave in communal ways (Eagly, 1987).

Online behavior and gender is especially interesting when delv- ing into the online dating community. While the general purpose of Facebook and SNSs is not to provide an online dating commu- nity, many SNS users have created and/or maintained romantic relationships through the convenience and affordability of SNSs. Hancock and Toma (2009) highlight the use of selective self- perception, that is, the transferring of text-based elements to online profile photos. The researchers argue that there are two motivations for selective self-perception: (1) the simple desire to enhance the self; and (2) the simultaneous need to present an accurate view of the self. Selective self-perception, at least with respect to posting profile photos, creates a tension between enhancement and accuracy that was seen more in women’s pro- files than men. Researchers infer that this may be due to the beauty regimens that can alter a woman’s outward appearance to such an extent that it creates an inaccurate physical representation of the woman (Hancock & Toma).

While there seems to be several clues to gain insight into how men and women differ in their use of SNSs, specifically Facebook, there are limitations to the existing literature that justify addi- tional research into this fairly new type of computer mediated communication. Further, does the preliminary support for Eagly’s (1987) social roles theory in online contexts also extend to Face- book profile photograph content? If existing literature tells us that we can infer such a significant portion of a person’s identity through his or her social network profile, then is it safe to say that this is true across all subgroups? And if not, where do these differ- ences lie? These questions surrounding self-construction through online identities according to gender encompass our second re- search question:

RQ2: Does the content and amount of college students’ Face- book profile photographs vary by gender?

2. Method

2.1. Sample and general procedures

All potential participants were current college students selected at random from the researchers’ list of active Facebook friends by utilizing a table of random numbers. If a prospective participant was chosen and was not a current college student, the researchers proceeded to the next eligible person on their list of Facebook friends, and continued randomly selecting individuals from there. The researchers’ own Facebook friends were exclusively employed for this study for three reasons: (1) due to Facebook policies and university IRB regulations, personal email (rather than Facebook messaging) was required for contacting participants and acquiring their consent; this information is typically only available to a

N.J. Hum et al. / Computers in Human Behavior 27 (2011) 1828–1833 1831

Facebook user’s active friends; (2) only active Facebook friends can access the profile photograph album, which we needed to view when coding the total number of profile photographs category; and (3) we reasoned that coding our own Facebook friends’ profile photographs would increase their privacy and confidentiality, thus reducing the likelihood that participants would change their pho- tographs to ones that were more ‘‘appropriate’’ during the coding time range.

Once each researcher had randomly selected approximately 40–50 prospective participants, each candidate was emailed on his or her personal email account, requesting the consent to partic- ipate in the study. Pertinent study details were provided to all potential participants, as well as the specific range of dates that profile picture coding would be conducted. This process continued until each researcher had received 35 consenting participants; five of these were randomly selected from each researcher’s sample to be used for coder training and achieving reliability (final N = 150). The final participants were 40% male and 60% female.

The unit of analysis was each participant’s current Facebook profile photograph. The six categories to be coded for each profile picture unit of analysis were: sex, quantity of profile photos in the participant’s profile picture album, level of physical activity, can- didness, level of appropriateness, and the number of subjects. Level of physical activity and number of subjects were of interest be- cause they reflect elements of social role theory (Eagly, 1987; Eagly et al., 2000). Candidness and level of appropriateness were selected to reflect the recent increased importance of personal online infor- mation in forming professional impressions (e.g., Microsoft, 2010; Watson et al., 2006). Finally, number of profile pictures could tap the extent to which college student Facebook users may be inter- ested in various, multiple online identities.

The researchers participated in 3 h total of training in order to achieve coding reliability. Twenty profile pictures (13.33% of the fi- nal sample), which were not part of the final sample, were used for training and calculating reliability. Differences in coding were re- solved through discussion and the codebook was revised and clar- ified as needed. Once reliability was achieved for each of the six coding categories (as assessed by averaging Cohen’s kappa values across the five coders), the researchers then coded their 30 Face- book friends’ profile photos during a week-long time period that was disclosed to participants when they provided consent.

2.2. Measures

The sex category included two options: male or female (average j = 1.00). The quantity of profile photos included the following re- sponses: 1–10, 11–20, 21+, or other (i.e., not able to view profile al- bum) (average j = 1.00). Under the level of physical activity category, the four options were: completely physically active, moderately active, inactive, and other (i.e., the Facebook account holder is not in the profile photograph) (average j = 1.00). The can- didness category included four options: candid, between candid and posed, posed, and other (average j = .85). The level of appro- priateness category included the following options: inappropriate, moderate, appropriate and other (average j = .86). Finally, the sub- ject category provided the options of alone, couple (i.e., two indi- viduals, but does not have to be just romantic in nature), group (i.e., three or more individuals), or other (average j = .94).

3. Results

3.1. Number of photos

The vast majority of participants had 21 or more profile pictures in their profile picture album (86%, n = 129). Following that, 11

(7.3%) participants had 11–20 profile pictures in their album. Only 8 (5.3%) had 1–10 pictures and 1 (0.7%) fit in the other category (i.e., this person’s number of pictures was unknown). There was no significant difference between the number of profile pictures in males’ Facebook profile picture albums in comparison to fe- males’ albums, v2 (3) = 4.66, p = .20.

3.2. Physical activity

Most of the profile pictures were completely inactive (82%, n = 123). Next, 20 (13.3%) pictures studied had moderate physical activity and only 2.7% of the pictures had people who were com- pletely physically active. Three profile pictures (2%) did not have any subjects in them. There was no significant difference between males’ and females’ level of physical activity in the profile pictures, v2 (3) = 2.25, p = .52.

3.3. Candidness

The majority of the profile pictures (76%, n = 114) were posed pictures. Twenty-two (14.7%) of the participants’ pictures were be- tween candid and posed, and only 11 (7.3%) of participants had pic- tures that were totally candid. There was no significant gender difference in candidness of profile pictures, v2 (3) = 5.98, p = .11.

3.4. Appropriateness

Most of the pictures studied (86.7%, n = 130) were found to be appropriate for all audiences. Only 18 (12%) of the pictures were moderately appropriate, and 2 (1.3%) of pictures fell into the ‘other’ category. No profile pictures were found to be completely inappro- priate. There were no significant appropriateness differences by gender, v2 (2) = .76, p = .68.

3.5. Number of subjects

The majority of the profile pictures had the subject as the only person in the picture (42%, n = 63). Next, we found that 48 (32%) of pictures had the subject and one other person in it. Only 36 (24%) of pictures included three or more people. Three (2%) pictures con- tained no people because they were an advertisement or a logo and were thus coded as other. The number of subjects in the profile pic- tures did not significantly vary by gender, v2 (3) = 4.38, p = .22.

In sum, in answering RQ1, our college student participants had multiple Facebook profile pictures and their current photographs tended to be inactive, posed, appropriate for all audiences, and in- clude only the subject. Further, for RQ2, no significant difference between male and female profile pictures for any of the six content and amount categories were observed.

4. Discussion

The purpose of this research was to analyze college students’ Facebook profile pictures and determine if the quantity and con- tent of the photos differed according to gender. The study looked specifically at the number of photos in participants’ profile picture albums, as well as the level of physical activity, candidness, appro- priateness, and the number of total individuals in the photos. Par- ticipants’ identity as constructed by their profile photographs was one where they were inactive, posed, appropriate, and alone (RQ1). Further, participants typically had over 20 profile pictures in their albums. The content of these profile pictures also did not vary according to gender (RQ2). A variety of conclusions can be derived from these results, as well as a plethora of potential new research areas.

1832 N.J. Hum et al. / Computers in Human Behavior 27 (2011) 1828–1833

As discussed by Zhao et al. (2008), a SNS user may leave a trail of personality traits for other users to find that offer clues as to who the profile user is as a whole. The profile picture is often the first item viewed on a SNS; indeed, on Facebook, the profile photo- graph is included in the results of a search for the name of a user, meaning that anyone with a Facebook account – not just a user’s friends – can view this image. Thus, as the Facebook profile photo- graph offers an important first attempt to construct one’s online identity, it is logical to assume that a significant amount of delib- erate and conscious thought is put into offering this first important clue.

The results of our first research question found that our college student participants may be aware of the importance of construct- ing such an identity, as the majority of their profile photographs were inactive, posed, appropriate, and contained only the subject. In addition, most of our college student participants’ profile picture albums contained 20 or more images, suggesting that they are choosing to provide multiple identity clues to other Facebook users via their profile pictures. As the time in college is considered to be one of transition (e.g., Arnett, 1994), our participants may also be ‘‘trying on’’ different identities by frequently changing their Face- book profile pictures.

These findings refute critics who charge that students could post photos of themselves that are inappropriate, thus hurting their future employment prospects (Foon Hew, 2011). However, our results – particularly that the majority of college students’ pro- file photographs are appropriate in nature – are consistent with Watson et al.’s (2006) research, which observed little to no Face- book profile photographs that contained alcohol use, drug use, sex- ually suggestive content, or partial nudity. Together, these studies support Watson et al.’s conclusion that ‘‘the incidence of such neg- ative depictions are far less frequent than the media seems to indi- cate’’ (p. 22). Studying the content of Facebook profile photographs for different age groups, particularly high school students who may not yet be as concerned with presenting themselves professionally, could yield different results than ours, and should be considered in the future.

Indeed, college students may be aware of the fact that potential employers may be seeking out their Facebook pages. In response, college students could subsequently be crafting an online identity that is professional and appropriate in preparation for entering the job market. With the recent spate of incidences where Facebook content – typically of a photographic nature – has led to university discipline or dismissal from a job or expulsion from school (e.g., Beals, 2010; Buffington, 2009; Haning, 2011; Watson et al., 2006) and a report (Microsoft, 2010) that 70% of employers have not hired job candidates based on information that they found online, this can be a very useful strategy for managing one’s online iden- tity. Microsoft’s study also found that 85% of employers were pos- itively influenced by a positive online reputation, which involves a ‘‘publicly held social evaluation of a person based on his or her behavior, what he or she posts, and what others (such as individu- als, groups, and Web services) share about the person on the Inter- net’’ (p. 3). This is further evidence of the importance of identity management in the context of Facebook, particularly in the case of the profile photograph, which can typically be viewed by anyone with a Facebook account.

For RQ2, be it the quantity of photos or the material con- tained in them, the Facebook profile photos examined in this study were not found to differ by sex. When considering that the total impact that sex or gender has on individuals’ commu- nication is minimal (e.g., Canary & Hause, 1993; Dindia & Wood, 1998; Hall, 1984), this finding is not surprising. In social interac- tions, knowing the sex of the subject ‘‘gives us little predictive power’’ in determining the behavior that will ensue (Aries, 1998, p. 77). However, it is not consistent with the tenets of

Eagly’s (1987) social role theory nor with the observations of face-ism in profile photographs (Cooley & Reichart Smith, 2010; Reichart Smith & Cooley, 2008), perhaps because the role of college student was more central to our participants’ choice of Facebook profile photograph than gender.

As we discussed in the literature review, a reason for the sim- ilarity across genders in online behaviors in this study may lie within the creation of a separate set of social norms in the online environment. These SNSs can be viewed as mini-cultures with a distinct set of social norms that can perhaps neutralize any differ- ences in gender roles. Dindia and Wood (1998, p, 35) suggest that it is unwise to look at the biological differences of males and fe- males but rather that ‘‘the social structures and practices’’ that tend to shape one’s behaviors should be at the forefront, and our findings for RQ2 provide evidence for this assertion. Our re- sults thus suggest that men and women may aim for similar per- ceptions when constructing their online identity via their profile pictures.

4.1. Limitations, future research suggestions, and conclusion

As with all research, certain limitations are present. Due to Facebook and IRB regulations, informed consent had to be obtained for each participant, meaning that the researchers of this study could only recruit their Facebook friends. As such, our sample was neither random nor representative of the Facebook user pop- ulation. However, our findings are consistent with related research that observed randomly selected college student profile photo- graphs (Watson et al., 2006). Relatedly, those who provided con- sent were made aware of the dates of data collection and may have consequently changed their profile photos to better represent themselves. Further, a willing participant may be an individual with different traits or online behaviors compared with someone unwilling to consent. These issues should be considered in future research on online identity construction.

This study’s findings offer a variety of further research ideas. For example, future research may choose to incorporate a wider array of SNSs, such as Twitter, MySpace, or LinkedIn, or perhaps consider a broader age range. Certain SNSs may be more relevant to one gender or age group over another, in much the same way that SNSs geared toward making business contacts over new relationships will draw in different types of users. Examining a broader range of SNSs and age groups in regards to gender roles in the online community will extend and expand upon our findings. In addition, including a self-report component, where Facebook profile photo- graphs are objectively coded and users are also surveyed regarding why they chose the content that they did, would be a useful ap- proach to future research.

Overall, the implication that Facebook and other social websites may host a separate list of social rules and norms indicates that on- line behavior of an individual may differ from face-to-face interac- tions. This study was able to shed light on the use of social networking sites and the role (or lack thereof) that gender plays within them. Facebook advertisers may find this information valu- able in order to better target different types of individuals, as Face- book’s personalized advertisements may be ultimately ineffective if the information being self-disclosed on Facebook has a separate social regulation than that of everyday in-person encounters be- tween sexes. Common users of Facebook and other SNSs may find this information of interest when they make new friends or estab- lish new relationships using the help of online networking. It can be implied that self-regulated construction of online identities may be skewed from reality in order to keep in line with alterna- tive social constructs of the SNS in question. As such, our study of- fers new knowledge about interpersonal interactions and nonverbal self-disclosure on the Internet.

N.J. Hum et al. / Computers in Human Behavior 27 (2011) 1828–1833 1833

References

Aries, E. (1998). Gender differences in interaction: A reexamination. In D. Canary & K. Dindia (Eds.), Sex differences and similarities in communication (pp. 65–81). Mahwah, NJ: Lawrence Erlbaum Associates Inc., Publishers.

Arnett, J. J. (1994). Are college students adults? Their conceptions of the transition to adulthood. Journal of Adult Development, 1, 213–224.

Beals, S. R. (2010, July 28). Superintendent’s Facebook posts could be trouble. Hartford Courant. <http://articles.courant.com/2010-07-28/news/hc-windsor- locks-superintendent-0728-20100727_1_superintendent-s-job-new-school- superintendent-social-media>.

Bond, B. J. (2009). He posted, she posted: Gender differences in self-disclosure on social networking sites. Rocky Mountain Communication Review, 6(2), 29–37.

Buffington, M. (2009, November 10). Former teacher sues BCS. Barrow Journal. <http://www.barrowjournal.com/archives/1960-Former-teacher-sues-BCS. html>.

Bugeja, M. J. (2006). Facing the Facebook: Unless we reassess our high-tech priorities, issues of student insensitivity, indiscretion, and fabrication will consume us. Chronicle Careers, 52(21), C1.

Canary, D. J., & Hause, K. S. (1993). Is there any reason to research sex differences in communication? Communication Quarterly, 41, 129–144.

Cooley, S. C., & Reichart Smith, L. M. (2010, June). International Facebook faces: An analysis of self-inflicted face-ism in online profile pictures. Paper presented at the annual convention of the International Communication Association, Singapore.

Dindia, K., & Wood, J. (1998). What’s the difference? A dialogue about differences and similarities between women and men. In D. Canary & K. Dindia (Eds.), Sex differences and similarities in communication (pp. 19–39). Mahwah, NJ: Lawrence Erlbaum Associates Inc., Publishers.

Eagly, A. (1987). Sex differences in social behavior: A social role interpretation. Hillsdale, NJ: Erlbaum.

Eagly, A. H., Wood, W., & Diekman, A. B. (2000). Social role theory of sex differences and similarities: A current appraisal. In T. Eckes & H. M. Trautner (Eds.), The developmental social psychology of gender (pp. 123–174). Mahwah, NJ: Erlbaum.

Ellison, N. B., Steinfield, C., & Lampe, C. (2007). The benefits of Facebook ‘‘friends:’’ Social capital and college students’ use of online social network sites. Journal of Computer-Mediated Communication, 12, 1143–1168.

Experian Hitline (2010). Facebook was the top search term in 2010 for second straight year. <http://www.hitwise.com/us/press-center/press-releases/facebook-was- the-top-search-term-in-2010-for-sec/>.

Facebook Factsheet (2010). <http://www.facebook.com/press/info.php?factsheet>. Foon Hew, K. (2011). Students’ and teachers’ use of Facebook. Computers in Human

Behavior, 27, 662–676. Guadagno, R. E., Muscanell, N. L., Okdie, B. M., Burk, N. M., & Ward, T. B. (2011). Even

in virtual environments women shop and men build: A social role perspective on Second Life. Computers in Human Behavior, 27, 304–308.

Hall, J. A. (1984). Nonverbal sex differences: Communication accuracy and expressive style. Baltimore, MD: Johns Hopkins University Press.

Hancock, J. T., & Toma, C. L. (2009). Putting your best face forward: The accuracy of online dating photographs. Journal of Communication, 59, 367–386.

Haning, E. (2011, January 4). Facebook unhealthy for doctor-patient relationship. WTOP.com. <http://www.wtop.com/?nid=25&sid=2221190>.

Junco, R., & Cole-Avent, G. A. (2008). An introduction to technologies commonly used by college students. New Directions for Student Services, 124, 3–17.

Lenhart, A. (2009). PEW Internet Project Data Memo. PEW Internet & American Life Project, 2, 17.

Mesch, G. S., & Beker, G. (2010). Are norms of disclosure of online and offline personal information associated with the disclosure of personal information online? Human Communication Research, 36, 570–592.

Microsoft (2010). Online reputation in a connected world. <http://www.microsoft. com/privacy/dpd/research.aspx>.

Morgan, E. M., Snelson, C., & Elison-Bowers, P. (2010). Image and video disclosure of substance use on social media websites. Computers in Human Behavior, 26, 1405–1411.

Pempek, T. A., Yermolayeva, Y. A., & Calvert, S. L. (2009). College students’ social networking experiences. Journal of Applied Developmental Psychology, 30, 227–238.

Reichart Smith, L. M., & Cooley, S. C. (2008, May). Face-ism on Facebook: An analysis of self-inflicted face-ism in online profile pictures. Paper presented at annual convention of the International Communication Association, Chicago, IL.

Scheufele, D. A., & Shah, D. V. (2000). Personality strength and social capital: The role of dispositional and informational variables in the production of civic participation. Communication Research, 27, 107–131.

Stern, L. A., & Taylor, K. (2007). Social networking on Facebook. Journal of the Communication, Speech, and Theatre Association of North Dakota, 20, 9–20.

Sussman, N. M., & Tyson, D. H. (2000). Sex and power: Gender differences in computer-mediated interactions. Computers in Human Behavior, 16, 381–394.

Tidwell, L. C., & Walther, J. B. (2006). Computer-mediated communication effects on disclosure, impressions, and interpersonal evaluations: Getting to know one another a bit at a time. Human Communication Research, 28, 317–348.

Utz, S. (2010). Show me your friends and I will tell you what type of person you are: How one’s profile, number of friends, and type of friends influence impression formation on social network sites. Journal of Computer-Mediated Communication, 15, 314–335.

Valenzuela, S., Park, N., & Kee, K. F. (2009). Is there social capital in a social network site? Facebook use and college students’ life satisfaction, trust, and participation. Journal of Computer-Mediated Communication, 14, 875–901.

Walther, J. B., Van Der Heide, B., Kim, S., Westerman, D., & Tong, S. T. (2008). The role of friends’ appearance and behavior on evaluations of individuals on Facebook: Are we known by the company we keep? Human Communication Research, 34, 28–49.

Watson, S. W., Smith, Z., & Driver, J. (2006). Alcohol, sex, and illegal activities: An analysis of selected Facebook central photos in fifty states. (ERIC Document Reproductive Service No. ED493049).

Young, A. L., & Quan-Haase, A. (2009). Information revelation and Internet privacy concerns on social network sites: A case study of Facebook. In Proceedings of the 4th international conference on communities and technologies (pp. 265–274). New York: ACM.

Zhang, J. (2010). Self-enhancement on a self-categorization leash: Evidence for a dual-process model of first- and third-persona perceptions. Human Communi- cation Research, 36, 190–215.

Zhao, S., Grasmuch, S., & Martin, J. (2008). Identity construction on Facebook: Digital empowerment in anchored relationships. Computers in Human Behavior, 24, 1816–1836.