Copyright © 2019 (Valerie Barker and Nathian Shae Rodriguez). Licensed under the Creative Commons
Attribution Non-commercial No Derivatives (by-nc-nd). Available at http://ijoc.org.
NATHIAN SHAE RODRIGUEZ
San Diego State University, USA
Prior studies have described selfies as narcissistic vehicles of self-presentation; by
examined how selfies signify forms of personal and social identity. Identity motivations
for selfies, social capital affinity on social media, and racial identity were predictors of
selfie intensity. Confirming other research, women were most likely to share selfies, but
also reported differences to men in selfie identity motivations and contexts. Among
LGBTQ participants, selfies for empowerment correlated with online activism.
Keywords: selfies, personal, social identity
some of the most frequent selfie sharers. Of interest are identity motivations for selfies, the extent to
which selfies relate to forms of social identity (with close friendship groups, gender, race, sexual
orientation), how selfies represent performance of identity—doing/being who we are (Stets & Burke,
2000)—and how much selfies provide affirmation from valued others. Whether technological affordances
and social media facilitate (or inhibit) these processes are also examined. Among young people, selfie-
taking is often regarded as normative behavior (Pew Research Center, 2014; “Investigating the style of
self-portraits,” 2019; YouGov, 2017); however, an air of negativity surrounds selfie-taking, especially for
women (Burns, 2015). Pamela Rutledge, director of the Media Psychology Research Center, stated,
“Selfies frequently trigger perceptions of self-indulgence or attention-seeking social dependence that
raises the damned-if-you-do and damned-if-you-don’t specter of either narcissism or very low self-
esteem” (quoted in Barakat, 2014, para. 10). Prior research (e.g., Barry, Doucette, Loflin, Rivera-Hudson,
& Herrington, 2017) does provide evidence for narcissistic motivations for selfies, but with Mendelson and
Papacharissi (2010), it might be more insightful to see this “as a step toward self-reflection and self-
actualization, rather than instances of uncontrollable self-absorption” (p. 30). Thus, the current study
focuses on the positive potential of selfies regarding personal and social identity.
Some researchers (Hernández, 2009; Noland, 2006; Yefimova, Neils, Newell, & Gómez, 2015) have
build relationships within marginalized groups and allowing participants to speak for themselves. Tiidenberg
Valerie Barker: email@example.com
Nathian Shae Rodriguez: firstname.lastname@example.org
Date submitted: 2018–05–22
International Journal of Communication 13(2019)
and Gómez Cruz (2015) argue that selfies represent a practice of freedom. They are “narrative acts and
signifiers of community belonging” (Tiidenberg & Tekobbe, 2014, p. 22). Posting selfies may teach
community members new ways of seeing themselves or act as pleas for support (Tiidenberg & Tekobbe,
2014). In short, selfies can be forms of identity work in which the in-group of the individual is either implicit
or clearly enacted. Identity motives for selfies may be particularly important among victimized groups, for
example, women, racial minorities, and/or LGBTQ people (e.g., Wargo, 2017). For examples, see Figure 1.
Therefore, using the theoretical lens of social identity theory, the current study examines the role of selfies
in identity construction and maintenance among young adults and specifically among traditionally
This Is Who I Am 1145
Social identity theory (Tajfel & Turner, 1986) posits that social identity stems from “that part of
groups) together with the value and emotional significance of that membership” (Tajfel, 1978, p. 63). People
construct group norms during interactions with valued in-group members and internalize and enact such
norms as part of their social identity (Turner, 1981). Hogg, Abrams, Otten, and Hinkle (2004) contend that
people possess many social identities and personal identities as there are groups they belong to or personal
relationships they have. Context affects the salience and form of social identity. Thus, identities vary in
subjective importance and value and situational accessibility (Kiang, Yip, & Fuligni, 2008). It is also
important to note that these social groups (or categories) are not monolithic, but rather overlap with one
another—a concept called intersectionality (Crenshaw, 1989).
Via social media, personal and social identity cues are manifested. Stets (1995) describes person
personal identity is “self-defined and evaluated in terms of idiosyncratic personal attributes and close
personal relationships with specific other people” (Hogg, Abrams, & Brewer, 2017, p. 2). Selfie-takers may
post shots relating to their personal identity, but also may include group markers (e.g., clothing, tattoos).
Social media function as public displays of connection (Donath & boyd, 2004), very often involving pictures.
Research indicates that, among young people, the camera is a frequently used mobile affordance for
celebration of friends (Senft & Baym, 2015; YouGov, 2017). Instagram and Snapchat, both popular with
18- to 24-year-olds (Pew Research Center, 2018), provide opportunities to connect through pictures,
Thus, we proposed the first hypothesis:
Selfies and Forms of Social Identity
In addition to how friendship group identity, gender, sexual orientation, and race identity relate to
Typically, people know the members of their social media networks (boyd, 2007), but many of these
connections are not close, unlike family or friends. Also, some people are unknown offline. Prior research,
outlined below, highlights the importance of social media connections regarding involvement with online
content and in-group membership, specifically social capital affinity, friendship groups, gender, sexual
orientation, race/ethnicity, and activism.
Social Capital Affinity
Affinity is defined as sympathy marked by community of interest and likeness based on casual
people online who are weak ties (Granovetter, 1983). Even though such people may be known only casually
or not at all offline, their opinions may be of interest, and their presence may enhance the online experience
by providing a sense of camaraderie. Walther and colleagues (2011) suggest that “social identification and
peer group influence in computer-mediated communication should be a useful element in explaining a
variety of influence effects in the new technological landscape” (p. 25). Related to this, Lange (2009) argues
that affinity spaces exist online as social settings where people congregate to reinforce group connection or
like-mindedness. Thus, it is expected that people are motivated by social capital affinity in posting selfies
for intergroup communication (Katz & Crocker, 2015).
Friendship Group Identity
Group identity is particularly meaningful to young people (Williams & Thurlow, 2005). In terms of
associated with individualization from family, with an additional challenge in maintaining existing peer group
identity (Manago, 2015) and simultaneously transitioning to new friendship groups. In doing so, young
people invest a lot of time communicating on social media. In focus group discussions with young people,
Walsh, White, and McDonald-Young (2009) found that such contact facilitated connectedness and enhanced
feelings of belonging. They concluded that, because young people are developing their identity outside the
immediate family, they rely on friends and peers to provide a sense of community. It is plausible to assume
a relationship between friendship group identity and selfie-taking/posting.
Gender identity is a person’s experience of his/her gender (Human Rights Campaign, 2019; Morrow
social identity (Moghadam, 1992). Gender attributes are archetypally assigned to males and females,
including expectations about masculinity and femininity, biological sex, and gender expression (Eller, 2015).
Some people do not identify with specific, or all, of the aspects of gender assigned to their biological sex;
they may identify as transgender, genderqueer, or nonbinary (Zastrow, 2013). It is possible, however, that
gender identity might relate to selfie-taking because women are more likely to post selfies (e.g., Dhir,
Pallesen, Torsheim, & Andreassen, 2016; “Investigating the style of self-portraits,” 2019 Sorokowska et al.,
The dominant media discourse about selfies is that this is a “girl thing” (Burns, 2015; Mascheroni,
Dictionary (Selfie, n.d.) states that selfies are
stupid photos that 14-year-old girls take of themselves. They take these photos to let
taking of a selfie when posted on a website of some sort. Some examples of these are:
#nomakeup #twerk and other stupid words that girls think make themselves sound cool.
This Is Who I Am 1147
But research indicates that selfies are much more than that for women and girls. Katz and Crocker
(2015) report that many of their interviewees felt that selfies told a story about their lives and elicited
conversations within chosen in-groups: “Often selfies referenced shared experiences, favorite movies,
people the users knew, or other ingroup referential images” (p. 1868). Also, Williams and Marquez (2015)
show that women and men produce and consume selfies differently: Selfie poses conform to gender
stereotypes, but also varying norms determine what is acceptable in selfies for men versus women.
Sexual Orientation Identity
Sexual orientation identity involves self-identification as a person enacting a specific sexual
identification (Ashmore, Deaux, & McLaughlin-Volpe, 2004). A person who identifies as heterosexual likely
experiences this form of social identity differently from other sexual orientations. Members of traditionally
stigmatized sexual orientations face a challenge in developing positive, stable, and secure social identity
within a cultural framework replete with negative stereotyping and treatment (Crocker, Major, & Steele,
1998). But selfies may be used to gain affirmation from others who identify similarly; for example, Katz and
Crocker (2015) mention that a participant reported “purposely using selfies to play with both male and
female presentations of self and in general to identify as queer gendered” (p. 1866). Similarly, Duguay
(2016) believes that selfies “reflect and propagate counter-discourses of sexuality and gender to oneself,
peers, and publics” (p. 2).
Racial identity is often a frame wherein individuals categorize others based on skin color (O’Hearn,
consider like or different from themselves (Chávez & Guido-DiBrito, 1999). By contrast, race identity may be
treated as a social construction associated with sense of group belonging: the perception that individuals share
a common origin with in-group members (Helms, 1993). Race identity is constructed in concert with in-group
others and, in comparison to, out-groups, a point of connection that helps group members interpret the world
and feel pride in themselves. Selfies may help in this process. Williams and Marquez (2015) found that Black
and Latino respondents reported producing selfies more frequently than White respondents. In addition, Black
and Latino women were fairly accepting of men in general taking selfies. By contrast, White women reported
definitive rules about White men’s selfies, favoring natural poses.
Accordingly, the second hypothesis predicted:
Aside from levels of identity, potential intercorrelations among demographic attributes, identity
Selfies, Empowerment, and Activism
Much selfie research targets self-presentation and impression management. However, Lindgren
used to take control in various ways” (p. 119). As mentioned, extant research on selfies hints at showing
solidarity, activism, and/or empowerment with an in-group. Other research about selfies and self-photography
documents presence in ways that reinvent what it is to be, as in gender, disability, or societal status.
Karadimitriou and Veneti (2016) argue that selfies present opportunities for new forms of interaction
that they provide a sense of intimacy, and, ultimately, may gain mainstream media attention. By attracting
wider public interest, selfies can garner support for a cause. Eler (2017) cites the Standing Rock protests (see
Figure 1, upper right photo), where Energy Transfer was to build a massive oil pipeline. Standing Rock selfies
evidenced what was taking place there. When news circulated that law enforcement was using Facebook check-
ins to track who was at the protest camp, more than a million people “checked in” at Standing Rock in support.
Consequently, the possible intercorrelation of selfie-taking, political/activist events, sense of
Selfies and Social Media
Hess (2015) sees selfies as an assemblage of dimensions. The selfie is a version of self. The place
and the pose also speak about performance of self. Finally, the networked audience (on social media) that is
invited to like or share the selfie also supports the motivation behind the selfie. Selfies posted on Snapchat
versus Instagram versus Facebook convey something different and may be perceived differently by different
Katz and Crocker (2015) found that, on Snapchat, the rapid exchange of selfies formed a dialogue.
common, or other in-group images. For Duguay (2016), Instagram’s affordances and content generation tools
were said to encourage users to focus on aesthetic appearance, whereas Vine’s limited editing tools and support
of creative sharing allowed users to highlight personal experiences. These findings generate the final research
This Is Who I Am 1149
Survey participants (N = 472) were recruited online via Sona Systems, which provides subject pool
was 53% White (n = 260). The remaining participants identified as Hispanic/Latino/Latinx (25%) = 117,
African American (6%) = 26, Asian American (20%) = 95, and Pacific Islander (8%) = 39. Some participants
reported more than one ethnicity. Most of the participants were female (73%; n = 343; male = 125,
transgender = 1, other = 2). Fifty-one participants identified as LGBTQ (10.8%). Participants’ average age
was 20.69 years (range = 18–63; SD = 3.90).
Measurements included demographic information (age, gender, race, sexual orientation) and
selfie editing, use of filters); level of friendship group identification, a diagrammatic gender, race, sexual
orientation identification measure, and a measure of affinity with others on social media (social capital
affinity); measures of identity motivations and perceptions’ of other people’s motivations for selfies; selfie
contexts and liking for other people’s selfie contexts; and online activism. The participants’ scores were the
overall means of the scale items. Questions were closed-ended, and participants typically responded on a
5-point range (e.g., 1 = very strongly disagree, 5 = very strongly agree). The background and explanation
of the scales appear below.
To measure social media and phone apps usage, two groups of items asked participants for their
Pew Research Center, 2018) and an opportunity to determine whether specific platforms/apps correlated
with selfie motivations/contexts.
Frequency of Social Media Platform Use
Using measures adapted from Gil de Zúñiga, Garcia-Perdomo, and McGregor (2015), participants
Instagram, followed by Snapchat, YouTube, and Facebook. See Table 1.
Frequency of Phone Apps Use
Seven items adapted from Auter (2007) measured frequency of phone app use (7-point scale):
frequently used apps were texting, music, and camera use.
How often do
you use the
(range = 1–
(range = 1–
Three way Selfies
This scale captured the effort invested in constructing selfies. Hess (2015) observed that, in
disseminating them via social networks” (p. 1639). Selfies are re-presented and enhanced; thus, how often
people take selfies is only a part of selfie intensity. Two additional items (5-point scale) asked how often
participants edit and use filters on selfies. These items posted an alpha of .84.
Identity Motives for Selfies
Prior qualitative and quantitative research (e.g., Diefenbach & Christoforakos, 2017; Krämer et al.,
2014) has investigated selfies as symbols of self disclosure, self-presentation, self-promotion, signs of
belonging, and sense of community. However, Hess (2015) found that “the same networked audiences that
celebrate sharing selfies also lambaste those who somehow do not fit the technological and cultural decorum
found online” (p. 1643). Similarly, Diefenbach and Christoforakos (2017) report that people are often critical
of other people’s selfies compared with their own. These findings reveal a disconnect between what people
report about their own selfies and what they believe about other people’s selfies. Therefore, participants
were asked about their own selfie motivations (and contexts; see below) and those of others.
Reflecting extant research, nine items (5-point scale) measured identity motivations for selfies. These
This Is Who I Am 1151
Analysis section) indicated that these items formed a unidimensional scale posting Cronbach alphas of .89 for
participants’ motivations and .92 for other people’s motivations. Analysis of separate items revealed that, for
participants, “to illustrate something about me” (M = 3.51, SD = 1.08) posted the highest mean and “to
manage others’ opinion of me” (M = 2.70, SD = 1.19) posted the lowest mean. For participants’ assessments
of other people’s motivations, the highest mean was “to receive positive feedback” (M = 4.21, SD = 0.90) and
the lowest mean was “to identify with others like them” (M = 3.68, SD = 1.02).
Where people photograph themselves affects how others perceive them (Lobinger & Brantner,
during special moments and events (such as travels)” (Lobinger & Brantner, 2015, p. 1854). Thus,
participants were asked to evaluate how often (5-point scale) they take selfies in eight contexts: family
occasions, holidays, vacation spots, bathroom, graduation, political event, activist event, or other (write in).
Vacation spots posted the highest mean (M = 3.72, SD = 1.23); political events posted the lowest mean (M
= 1.65, SD = 0.10). In addition, participants reported how much they liked seeing other people’s selfies in
these settings (7-point scale). Again, vacation spots posted the highest mean (M = 5.48, SD = 1.29); other
people’s bathroom shots posted the lowest mean (M = 3.25, SD = 1.58). Overall, participants did not
particularly like to see other people’s selfies (range = 1–5; M = 2.43, SD = 0.90).
Social Identity Measures
Friendship group identification. Research shows that for young people, members of their immediate
peer group are important and influential (e.g., Larson, Whitton, Hauser, & Allen, 2007). Close friends, of
comparable age and who exhibit similar ideas, allegiances, clothing styles, and other group markers, help
young people to individuate from family and acquire a sense of strong in-group connection. According to
social identity theory, social identity involves the value and significance attached to group membership—
collective self-esteem. Thus, there is an element of social comparison in assessing social identity. Friendship
group identification was assessed using six items adapted from the 16-item Luhtanen and Crocker (1992)
Collective Self-Esteem Scale to provide a measure of self-evaluation of identity with participants’ closest
group of friends. Luhtanen and Crocker conducted three initial studies to validate the scale, where it was
found to positively correlate with collectivism scales. Because of space considerations, eight negatively
worded items were omitted. Of the remaining eight items, two were excluded because they were a differently
worded repetition of others in the scale. The remaining six items were representative of the original
dimensions of the scale and have been effectively employed in prior research. The items were (1) “The
group I belong to is an important reflection of who I am,” (2) “I’m glad to be a member of my group,” (3)
“Others respect my group,” (4) “I feel good about the group I belong to,” (5) “I participate in the activities
of my group,” and (6) “Others consider my group good.” In the present study, the scale showed high internal
reliability (.91) and the overall mean was high (M = 4.14, SD = 0.68, range = 1–5). The means did not
differ across gender, White/non-White, or sexual orientation (each median = 4.00). Also, the scale was
positively related to the other identity measures (see Table 2).
Selfie intensity: never/rarely/sometimes/often/very often (α = .84; M = 3.00, SD = 1.10)
How often do you take selfies?
Edit your selfies?
Use filters on your selfies?
Identity motivations for selfies: I take selfies to . . . (strongly disagree–strongly agree)
Participant (α = .89; M = 3.14, SD = 0.88)
Perceptions of others’ motivations: I think other people take selfies to . . . (α = .92; M = 3.95, SD = 0.75)
Illustrate something about me
Feel better about myself
Identify with others like me
Keep others up to date with my life
Receive positive feedback
Manage others’ opinions of me
Other (please write in)
Online activism: How many times have you performed the following activities? (never/once/twice/three
Sent an e-mail to a political official when requested by a group you support
Sent an e-mail to a political official under my own volition
Signed an online petition
Donated money to a political organization
Followed or liked a political figure on social media
Tweeted or commented about a political official using social media
Asked a question of a political figure using social media
Friendship group identification: Please say how much you agree with following statements about your
closest group of friends: (strongly disagree–strongly agree; α = .91; M = 4.13, SD = 0.68)
The group I belong to is an important reflection of who I am
I’m glad to be a member of my group
Others respect my group
I feel good about the group I belong to
I participate in the activities of my group
Others consider my group good
Social capital affinity–social media: Thinking about the social media platform you use most, say how
much you agree with the following statements: (strongly disagree–strongly agree; α = .88; M = 3.58,
The opinions of those visiting this platform interest me
Interacting with people visiting this platform makes me feel like part of a community
When visiting this platform, hearing what others say enhances the experience
Communication with the people visiting this platform raises points of interest for me
Being with people visiting this platform makes me want to follow up on things
This Is Who I Am 1153
Intercorrelations for social identifications
3. Sexual orientation
4. Friendship group
5. Social capital affinity–social media
range = 1–7; M = 5.58, SD = 1.58; sexual orientation identification: range = 1–7; M = 5.67, SD = 1.77.
Friendship group identification: range = 1–5; M = 4.13, SD = 0.67; social capital affinity–social media:
range = 1–5; M = 3.58, SD = 0.75.
*p < .05. **p < .01.
Gender, sexual orientation, race identification. “When individuals categorize themselves as group
members, the ingroup becomes included in the self and individuals recognize the characteristics of the
ingroup as representing part of themselves” (Tropp & Wright, 2001, pp. 586–587). The current study
assessed gender identification as distinct from actual gender. Participants were asked how much they
identify with their gender without reporting at that point how they categorize themselves in terms of gender.
The inclusion of in-group in self-measure separately assessed identification with gender, sexual orientation,
and race (as ascribed groupings). It consisted of seven pairs of circles varying in degree of overlap.
Participants selected the pair of circles that best represented their level of identification with a specific in-
group. Its visual representation captured in-group identification as the interrelationship of self and group.
This measure provided data on how salient an individual’s identity was relative to gender, sexual orientation,
and race; however, it did not measure intersections of all three. Identities, specifically social identities, do
overlap with one another (Crenshaw, 1989). Some evidence for this was present in the current data in that
each identity measure was intercorrelated with the others (see Table 2).
Social capital affinity (SCA)–social media. Social media feature heavily in the lives of young adults.
This scale measured sense of affinity with like-minded individuals who may be weak ties (Granovetter, 1983)
on social media. Five items measured affinity with others on the participants’ favorite platform. For example,
they were asked how much they agreed that “interacting with people on this platform makes me feel like
part of a community.” The scale posted a Cronbach alpha of .88.
Online activism. Participants were asked to evaluate the frequency with which they performed
seven activities (1 = never, 2 = once, 3 = twice, and 4 = three or more times), for example, followed or
liked a political figure on social media and tweeted or commented about a political official using social media.
The scale posted a Cronbach alpha of .78 and correlated with reported selfies taken at political (r = .33, p
< .001) and activist events (r = .35, p < .001). Measures are summarized in Table 2.
Scale items were coded positively: A high score indicated higher race identification, identity
for participants’ perceptions about other people’s motivations for selfies, we employed factor analysis using
principal axis factoring and Oblimin rotation. Regarding participants’ responses about themselves, one major
factor accounted for 52.8% of the variance; however, a lesser factor emerged accounting for 6.8% of the
variance in motivations. Inspection of the rotation revealed that there were strong cross-loadings for all of
the variables. Thus, a unidimensional scale was constructed for participants’ identity motivations. For the
participants’ perceptions of other people’s motivations for selfies, only one factor resulted. This accounted
for 58.8% of the variance in selfies. A paired sample t test revealed a significant difference in scores for
participants’ reported selfie-taking motivations (M = 3.10, SD = 0.88) compared with their perceptions
about other people’s motivations for selfie-taking (M = 3.95, SD = 0.75), t(457) = -19.00, p = .0001.
Subsequently, bivariate tests using Pearson’s correlation coefficients were performed, followed by analysis
of variance (ANOVA) testing to assess differences among groups of participants. Regression analysis
determined the relative value of selfie intensity correlates..