553 - Smiling through the pain: A mixed-methods study of adolescent emoji use

Poster Number: 553
Publication Number: 553.245

Matt Minich, University of Wisconsin-Madison, School of Journalism and Mass Communication, Madison, WI, United States; Bradley Kerr, University of Wisconsin School of Medicine and Public Health, Madison, WI, United States; Megan A. Moreno, University of Wisconsin, Madison, WI, United States

Background: Adolescents often use text messaging to communicate with peers and adults, including healthcare providers. Many messages contain emojis, which can convey emotional context. Adults sometimes struggle to understand the emotional significance of these emojis.

Objective: We sought to better understand the associations between emojis and moods among adolescents.

Design/Methods: This study used a mixed-methods approach. First, adolescents ages 12-17 were recruited via Qualtrics panels for an ecological momentary assessment (EMA) study. A total of 30 short text message surveys were delivered as EMAs over 6-day periods. Participants were asked to describe their current mood with a number (1-7) and an emoji. Face emojis from these responses were coded by features (e.g. mouth up/down), which were entered into a multiple regression model as dichotomous predictors of numerical mood ratings. Dominance analysis identified the features most important in predicting ratings. We used those features to construct categories and tested whether ratings differed across categories using repeated-measures ANOVA. We calculated an average rating for each emoji and tested mean differences with a one-way ANOVA. Second, adolescents ages 13-18 were recruited for virtual focus groups and asked to describe their use of emojis. Thematic analysis was conducted using the constant comparative method.

Results: The 258 EMA participants completed 6,351 EMAs; participants were an average of 14.9 (SD=1.4) years old and 68% were female. Participants submitted 101 different “face” emojis, comprising 94% of the sample (N = 5999). Dominance analysis yielded 5 ordered categories of emojis. A compilation of all EMA mood ratings differed significantly across these categories, F(4,5737) = 442.1, p < 0.001. Averaged mood ratings also differed across categories, F(4,96) = 14.67 , p < 0.001. Focus group participants included 31 adolescents with mean age 6.2 (SD = 1.5 ), 48% were female. We observed three relevant themes. First, participants described that they were less likely to use emojis when experiencing a negative mood versus a positive mood. Second, participants described using mouth up emojis in sarcastic or ironic ways. Third, participants described emojis as having context-dependent meanings.Conclusion(s): Within the highly structured setting of EMAs, trends suggested certain features of face emoji were associated with mood patterns. Focus group findings described a more complicated reality in which the significance of emojis varies depending on the intended audience and conversational context.
Matt Minich CVCURRICULUM VITAE.docx.pdf
Focus group themesExcerpts of focus group transcripts that exemplify three relevant themes identified through constant comparitive analysis.