558 - Understanding news articles about teens, social media, and depression

Poster Number: 558
Publication Number: 558.245

Grainne McDonagh, University of Wisconsin School of Medicine and Public Health, Madison, WI, United States; Maggie Bushman, University of Wisconsin School of Medicine and Public Health, Madison, WI, United States; Megan A. Moreno, University of Wisconsin, Madison, WI, United States

Background: The news has been inundated with conversations about a possible relationship between social media and depression among teens over the last decade. The public’s relationship with news media has been complicated; however, the COVID-19 pandemic has increased consumption and trust in the news. Little is known about how journalists report on the relationship between social media and depression among teens.

Objective: The purpose of this study was to understand how journalists represent studies on social media and depression in the news.

Design/Methods: Articles published from 2010 to 2020 were identified from ProQuest: US Newsstream using the search terms depression, social media, and teens. Inclusion criteria limited articles to those that specifically referenced a research article on the topic of teen social media use and depression. Each article was assessed two ways. First, articles were evaluated for three characteristics: 1) the news source, 2) the article author, and 3) quoted sources. Sources were categorized by gender and profession. Second, articles were analyzed using the validated text analysis software, Linguistic Inquiry and Word Count (LIWC). Variables such as anxiety (worried, fearful), female/male references (girl/boy, her/his, mom/dad), and tentative (maybe, perhaps) were included. T-tests were used to compare LIWC output scores by the 3 article characteristics.

Results: Our search yielded 565 articles, and a total of 59 articles met inclusion criteria. Most articles came from regional newspapers (55.9%) and had female authors (56.9%). A total of 149 individuals were quoted as sources: 4.7% researchers, 42.3% healthcare investigators, 11.4% educators, 26.2% stakeholders, and 16.8% family members. There were no significant differences between national and regional newspapers for any LIWC variables. Comparing male and female author articles, anxiety-related words were used more often by female authors than male authors (1.07% vs 0.63%, p=0.032). When comparing quoted sources who were scientists vs. non-scientists, scientist articles used significantly more male references (0.72% vs 0.25%, p=0.004). Further, non-scientist articles used significantly more tentative-related words (2.70% vs 2.27%, p=0.055).Conclusion(s): We found significant differences in news articles based on author gender, and whether scientists were quoted. This information could aid pediatricians to better understand the general public’s perception on the relationship between social media and depression among teens and inform dissemination strategies for research on this topic.