Hypothesis 1 predicted a positive relationship between identity motivations for selfies and selfie
statistically significant (see Table 3).
Table 3. Correlations: Identity Motivations for Selfies and Selfie Intensity.
Feel better about myself
Receive positive feedback
Keep others up to date with my life
Identify with others like me
Illustrate something about me
Manage others’ opinion of me
Note. All relationships significant at p < .01.
identity motivations. The dependent variable was selfie intensity and the independent variables were
identity motivations for selfies (including other people’s motivations). Four identity motivations
remained in the model: to feel better about myself (β = .23), keep others up to date (β = .16), receive
positive feedback (β = .17), and to illustrate something about me (β = .13). These four predictors
explained 34% of the variance in selfie intensity, F(4, 459) = 45.77, p < .0001. However, when the
This Is Who I Am 1155
identity motivations scale (average composite score for all the identity motivations) was added to the
model, the other variables were excluded (β = .53), F(1, 462) = 181.78, p < .0001. Hypothesis 1 was
Hypothesis 2 predicted that forms of social identity would be positively related to selfie
intercorrelations did not indicate multicollinearity (see Table 2).
The bivariate tests revealed positive relationships between selfie intensity and closest friendship
.01). Again, regression analysis with stepwise entry was conducted with the dependent variable as selfie
intensity and the independent variables as race identity, closest friendship identity, and SCA-social
media. Two forms of identity remained in the model: SCA-social media (β = .28) and race identity (β =
.11). These variables explained 9% of the variance in selfie intensity, F(2, 461) = 25.13, p < .0001.
Hypothesis 2 was partially confirmed.
Research Question 1 inquired about the relationship among gender, race, and sexual
A one-way ANOVA test revealed that women (M = 3.20, SD = 1.04) posted the highest mean
= 2.12), F(3, 466) = 24.83, p < .0001.
To assess gender differences in identity motivations and contexts, we employed two
test, the dependent variables were the identity motivations scale, other people’s identity motivations,
and the motivation items composing the identity motivations scale, overall F(9, 445) = 32.18, p < .0001.
In the second test, the dependent variables were the selfie contexts (see Table 4). The results indicated
that, for motivations, there were statistically significant differences on the means (higher for women)
on all items except for other people’s motivations and “to manage other people’s opinions of me.” For
selfie contexts, all of the means were statistically different (women higher) except for political and
activist events, overall F(7, 457) = 10.30, p < .0001.
International Journal of Communication 13(2019)
Men M (SE)
Women M (SE)
Overall other people’s motivations
4.00 (0.041) ns
Something about me
To feel better
To identify with others like me
Keep people up to date
Manage other people’s opinions
2.78 (0.066) ns
1.66 (0.054) ns
1.94 (0.063) ns
Race and Sexual Orientation
Identity motivations. Bivariate tests indicated that Black participants were more likely to report the
motivation to take selfies to “identify with others like me.” A one-way ANOVA confirmed the bivariate
relationship, F(1, 468) = 6.90, p < .001 (Black: M = 3.38, SD = 1.10; other race: M = 2.79, SD = 1.12).
Selfie contexts. Non-White participants were more likely than Whites to report taking selfies at
family occasions, F(1, 468) = 7.0, p < .01 (non-White: M = 3.10, SD = 0.08; White: M = 2.81, SD = 0.08);
graduation, F(1, 468) = 4.38, p < .05 (non-White: M = 3.39, SD = 0.09; White: M = 3.13, SD = 0.09);
and activist events, F(1, 468) = 6.72, p < .01 (non-White: M = 2.01, SD = 0.08; White: M = 1.74, SD =
0.07). Black participants were more likely to report taking selfies at activist events, F(1, 470) = 4.00, p <
.05 (other race: M = 1.84, SD = 1.14; Black: M = 2.31, SD = 1.41). For Black participants, Twitter use and
using activist events as selfie contexts were correlated (Twitter: r = .12, p < .05; activist event: r = .09, p
LGBTQ participants were more likely to take selfies at political events, F(1, 471) = 4.91, p < .05
Research Question 2 asked about relationships among selfie-taking, gender, race, sexual
This Is Who I Am 1157
LGBTQ participants were more likely to take selfies to feel empowered, F(1, 467) = 4.63, p < .05
(heterosexual: M = 3.07, SD = 1.20; LGBTQ: M = 3.46, SD = 1.18). Similarly, LGBTQ participants were
more likely to say that they take part in online activism, F(1, 460) = 18.70, p < .01 (heterosexual: M =
1.65, SD = 0.58; LGBTQ: M = 2.04, SD = 0.77). Overall, online activism was related to feeling empowered
via selfies (r = .13, p < .01). Therefore, we used analysis of covariance to investigate a possible interaction
regarding sexual orientation in the relationship between selfie empowerment and online activism. The results
confirmed that LGBTQ participants who feel empowered by selfies are also more likely to report online
activism, F(1, 460) = 20.12, p < .0001 (heterosexual: M = 1.67, SD = 0.58; LGBTQ: M = 2.00, SD = 0.77).
See Figure 2.
------- LGBTQ, R
Linear = .07
Linear = .008
Research Question 3 investigated the relationship between identity motivations for selfies and
social media platforms. Table 5 illustrates the bivariate relationships between identity motives for selfies
and selected social media platforms. Instagram posted the strongest correlations with identity motivations
for selfies and selfie intensity.
Table 5. Correlations: Social Media Platform Use and Identity Motivations.
Identify with others
To feel empowered
To manage impressions
*p < .05. **p < .01.
Based on social identity theory, this study investigated correlates of selfie-taking among young
orientation, race identity, and social capital affinity–social media).
Participants reported that they take selfies to say something about who they are, connect with
themselves. The highest single-item selfie motivation correlates of selfie intensity were to feel better and to
feel empowered. However, it was clear that all of the assessed motivations contributed to form a holistic
contribution to selfie-taking. Selfies can be used as tools to communicate multiple identity dimensions. The
group is one foundation for identity (who one is) and what one does (role vis-à-vis significant others) is
another, both of which complement the self as an individual with idiosyncratic personal attributes (Stets &
Race identity contributed to selfie intensity in this sample. Race identity speaks to the inclusion of
language, clothes, hair, music, tattoos. People can see and compare others who are like them and those
who are not via selfies. However, race identity was not correlated with any racial group in our study; thus,
it may simply be that those who are high identifiers with their racial group feel the need to celebrate it via
selfies. Interestingly, the item “to identify with people like me” was statistically related to identifying as
Black. This is consistent with findings from Williams and Marquez (2015) who reported that African
Americans regard selfies as a positive way to present to others of their race. Also, being Black, Twitter use,
and use of activist events as selfie contexts were intercorrelated. Other research (e.g., Florini, 2014)
highlights African American use of Twitter for “signifyin’” Blackness, which involves “millions of Black users
This Is Who I Am 1159
on Twitter networking, connecting, and engaging with others who have similar concerns, experiences, tastes
and cultural practices” (Florini, 2014, p. 225).
Although closest friendship group was a correlate of selfie intensity, the strongest predictors were
therefore, this may provide evidence of intersecting identities among women. Women reported stronger
feelings of close friendship identity and affinity with those in their social media networks, some of whom are
close and others who are not well known although like-minded and similar. Also, women, when compared
with men, reported differing estimates of motivations for selfies and contexts for taking them. In fact, men’s
estimates for motivations and for contexts were low. This may indicate that men have motives and contexts
for selfies not included in the survey. Conversely, it may be that young men just do not choose to take
selfies. Selfies may be more of part of being a woman in a cultural sense, especially as women show a
greater presence on social media and report relational reasons for doing so (Krasnova, Veltri, Eling, &
Women and LGBTQ participants were more likely to report taking selfies to feel empowered. LGBTQ
activism. This, then, is something apart from the narcissism that is often associated with selfies. Instead, it
appears that selfies can be used to make a personal and/or political statement. Indeed, the findings suggest a
complexity behind this apparently simple act. This is not to deny evidence of ambivalence about posting selfies,
as demonstrated by the disconnect between estimates about participants’ own motives for selfies and others’
motivations. That aside, the data imply that social media users value the opportunity to tell others something
about themselves and to engage with others in the process by posting selfies. Arguably, this is productive on
a personal level in terms of enacting identity, but also on a wider network level in connecting with others
whether the common denominator is gender, race, and/or sexual orientation, or all of these. Considering the
#MeToo movement and the push to prohibit transgender people from serving in the military and even to
eradicate the definition altogether (Green, Benner, & Pear, 2018), online empowerment, affirmation, and
activism via selfies require further research.
Although not directly tested, the results suggest that issues of intersectionality are at play
personal and social identities and use selfies to reify group belonging. Identities interact to inform the
experiences of each individual. Women of color, transwomen, and LGBTQ members tend to be
disproportionately oppressed in society. In addition to this structural intersectionality, issues of political
intersectionality—conflicting political agendas—also invoke additional negotiation. For example, White
feminism has long excluded women from racial and ethnic minority groups and women who are economically
disadvantaged (Watters, 2017). How does the intersectionality of these identities influence selfie-taking,
posting, and social identity salience? Further research on intersectionality, identity, and selfies is warranted.
Social capital affinity on social media was another significant predictor of selfie intensity. Social
that can send and may be intended to send different messages to different individuals, communities, and
audiences. Duguay (2016) also conceptualizes selfies as part of a conversation, stating, “Messages
communicated through selfies can feature in conversations reinforcing dominant discourses within existing
publics or form counterpublics, gathering people around alternative and opposing discourses” (p. 2). The
selfie, then, acts a tool of connection and community-building via social media.
Duguay (2016) describes social media as “non-human actors” (p. 2) in communication interactions.
is invited to like, comment, and/or share the selfie. As is typical of this age cohort, most participants reported
taking selfies at least sometimes. Instagram and Snapchat were the most popular platforms. Instagram posted
the strongest relationship with selfie intensity; however, the data provide evidence that specific social media
platforms are outlets for specific types of identity motivations for selfies. For example, Instagram potentially
allows selfie producers to receive positive feedback, Twitter is a platform on which one may feel empowered
by posting a selfie, and Snapchat offers selfie-takers opportunities to say something about themselves. Other
research might investigate platform attributes that encourage certain types of selfies, conversations about
selfies, and solidarity with selfie producers.
As a cross-sectional study using a convenience sample, no claims are made about causal
(Pew Research Center, 2018). The findings form a basis for understanding trends about identity and selfie-
taking among young adults. Regarding the subsamples, however, there was a clear gender bias (women),
the racial composition of the sample was not reflective of minorities in the wider population, and, in fact,
the LGBTQ presence was greater than that estimated in the U.S. population. Gallup (2017) estimates that
approximately 4% identify as LGBTQ; here, about 11% identified as LGBTQ. Regardless, the statistical
outcomes reflect existing research in this domain. Overall, the scale reliabilities were good, but future
research should expand the measurement of selfie motivations, contexts, and types of activism.
Experimental designs could manipulate contexts, poses, and software involving selfies. In addition,
qualitative research such as focus groups would add depth to these findings.
Maddox (2018) suggests that selfies negotiate between lived experience and mass media and allow
takers craft their identities online and communicate them to others in their chosen audiences/communities.
This study sheds light on the ways in which online practices such as selfies enhance people’s lives. As well,
the findings indicate that potentially marginalized groups can use selfies to say something about who they
are and, in doing so, feel a sense of affirmation, connection, and empowerment. This goes against the
perception that taking selfies is a frivolous, narcissistic practice with little meaning or value.
Affinity. (n.d.). In Merriam-Webster’s online dictionary. Retrieved from https://www.merriam-
Ashmore, R. D., Deaux, K., & McLaughlin-Volpe, T. (2004). An organizing framework for collective
This Is Who I Am 1161
Auter, P. J. (2007). Portable social groups: Willingness to communicate, interpersonal communication
gratifications, and cell phone use among young adults. International Journal of Mobile
Barakat, C. (2014, April 16). Science links selfies to narcissism, addiction and low self-esteem. Retrieved
Barry, C. T., Doucette, H., Loflin, D. C., Rivera-Hudson, N., & Herrington, L. L. (2017). “Let me take a