Exploring the Association Between Suicide Prevention Public Service Announcements and User Comments on YouTube: A Computational Text Analysis Approach

Abstract

In the United States, suicide rates have increased by 30% over the past few decades. Public service announcements (PSAs) are effective health promotion vehicles and social media can help spread PSAs to hard-to-engage individuals who may benefit from intervention efforts, yet the most meaningful characteristics of PSAs for influencing health promotion attitudes and behaviors are inconclusive. This study applied content and quantitative text analyses to suicide prevention PSAs and comments on YouTube to assess the relationships between message frame, message format, and the level of sentiment and help-seeking language within them. Seventy-two PSAs were analyzed for gain/loss-framing and narrative/argument-format, and 4,335 related comments were analyzed for positive/negative sentiment and frequency of help-seeking language use. Results indicate that a higher ratio of positive comments was more likely to be found on gain-framed and narrative-formatted PSAs, and a higher ratio of comments with help-seeking language was more likely to be found on narrative-formatted PSAs. Implications and future research are discussed.