(PS7-A21) Effects of Instagram and Twitter Use on Depressed and Anxious Mood: A Multimethod Approach

Social media (SM) use among young adults has increased in the last decade, and during this period, rates of depression and anxiety have also increased. Many studies link SM use to anxiety and depression (NIMH, 2017; Twenge et al., 2019). and some have argued that increased rates of emotional disorders among young adults are at least partially due to SM use (Twenge at al., 2018). However, it is difficult to draw conclusions because most studies use cross-sectional data and do not investigate specific SM platforms. The current study investigated the effects of Instagram and Twitter use on depressed and anxious mood using a multimethod approach by collecting cross-sectional, experimental, and longitudinal data.  

Participants were 591 college students. For cross-sectional data, Instagram and Twitter use data were taken directly from participants phones and participants completed self-report measures of depressed and anxious mood. To experimentally investigate the effects of SM on state mood, participants were randomly assigned to one of four groups - Instagram for 30 or 5 minutes or Twitter for 30 or 5 minutes – and completed self-report measures of mood pre- and post-SM use. For longitudinal data, participants’ current depressed and anxious mood and recent SM use were sampled 6 times per day over 7 days through ecological momentary assessment (EMA).

In our cross-sectional data, Twitter use was negatively associated with anxious mood (r = -.094, p =.029) but not depressed mood (p > .05). Instagram was not associated with depressed or anxious mood (ps > .05).

A 4 (condition: Instagram 30 minutes, Instagram 5 minutes, Twitter 30 minutes, Twitter 5 minutes) x 2 (time: pre, post) x 2 (mood: depressed, anxious) repeated measures ANOVA did not reveal a significant condition by time by mood interaction. There was no interaction with mood type and no significant main effect of condition. However, there was a significant main effect of time with participants experiencing lower depressed and anxious mood from pre- to post-SM use regardless of condition (p < .001).

Our EMA data contained 7,629 observations across 207 participants. We evaluated the effect of SM use on mood with multilevel modeling. The fixed effects of SM use on depressed, ß = .11, p < .001, R²_marginal = .011, and anxious mood, ß = .11, p < .001, R²_marginal = .008, were significant but of small effect size. We also investigated the effect of mood on subsequent SM use. The fixed effect of depressed mood on SM use was significant, ß = .012, p < .001, R²_marginal = .011, but the effect of anxious mood was not, ß = .002, p = .07.

Aside from a cross-sectional (negative) relationship between Twitter use and anxious mood, our cross-sectional and experimental results did not reveal a relationship between SM use and depressed and anxious mood. We did find a weak, but statistically significant, relationship between SM use and increased depressed and anxious mood. We found a reciprocal relationship indicating that depressed mood may motivate increased SM use. Our results indicate that general SM use across two platforms does not play a large role in depressed or anxious mood. As such, it may be important to focus on specifically problematic SM use, rather than general use, when evaluating the relationship between SM and mental health.