Longitudinal data from adolescents and young adults show no evidence that social media use predicts later depressive symptoms, according to research published in Clinical Psychological Science. However, the findings do show that relatively higher depressive symptoms predicted later social media use among adolescent girls.
This research stands in contrast with recent claims that adolescents’ use of social media may lead to depression. Those claims are based primarily on studies that examined associations between average social media use and average well-being measured at a single point in time.
“You have to follow the same people over time in order to draw the conclusion that social media use predicts greater depressive symptoms,” says lead author Taylor Heffer of Brock University. “By using two large longitudinal samples, we were able to empirically test that assumption.”
Beginning in 2017, Heffer and coauthors surveyed 6th, 7th, and 8th graders in Ontario, Canada once a year for 2 years. The researchers also conducted annual surveys of undergraduate participants, beginning in their first year of university over a span of 6 years.
To measure depressive symptoms, the researchers used the Center for Epidemiological Studies Depression Scale for the young adults and an age-appropriate version of the same scale for the adolescents. All participants answered two questions about their average daily hours spent on social media – one measuring weekday use and the other measuring weekend use. The participants also answered questions about other screen time, such as watching TV, and non-screen activities including doing homework and exercising.
Heffer and colleagues analyzed the data separately for each age group and gender.
The results showed that social media use did not predict later depressive symptoms among adolescents or college undergraduates. Rather, greater depressive symptoms predicted more social media use over time, but only among adolescent girls.
“This finding contrasts with the idea that people who use a lot of social media become more depressed over time. Instead, adolescent girls who are feeling down may turn to social media to try and make themselves feel better,” says Heffer.
Overall, the research suggests that the fear surrounding social media use and its impact on mental health may be premature. “Policymakers also have recently been debating ways to tackle the effects of social media use on mental health,” Heffer says. Evaluating whether fears about the impacts of social media have merit requires prospective longitudinal studies that allow researchers to examine whether it is social media use that predicts depressive symptoms (rather than the other way around), while controlling for other potential influences. As Heffer and colleagues note, individual differences in personality, motivation, and current well-being are likely to play a critical role in the relationship between media use and future well-being.
“There may be different groups of people who use social media for different reasons,” Heffer explains. “For example, there may be a group of people who use social media to make social comparisons or turn to it when they are feeling down, while another group of people may use it for more positive reasons, such as keeping in contact with friends.”
Examining the role that these differences play will help clarify the ways in which social media interacts with mental health, with implications for parents, policymakers, and healthcare professionals alike.
Coauthors on the research include Marie Good of Redeemer University College and Owen Daly, Elliott MacDonell, and Teena Willoughby, all of Brock University.
Willoughby received funding for this study from Social Sciences and Humanities Research Council of Canada and Canadian Institutes of Health Research.
Heffer, T., Good, M., Daly, O., MacDonell, E., & Willoughby, T. (2019). The longitudinal association between social-media use and depressive symptoms among adolescents and young adults: An empirical reply to Twenge et al. (2018). Clinical Psychological Science. doi.org/10.1177/2167702618812727