My thesis focuses on PSAs and their level of influence on college students. Specifically, if they influence the student's drinking/driving habits and if this influence is more or less effective for students who belong to a Greek organization.
An associate brought up a great point that I may also want to explore, possibly in another project unless I can incorporate it into this one. He said that determining the most effective medium would be valuable information as well.
There are a few studies that show unrealistic, doomsday-style announcements are subject to mocking instead of positively influencing actions. I'm sure you've seen them. Here's just one: http://www.sciencedaily.com/releases/2011/08/110822121721.htm
But the below study is the first time I've seen that allowing *any* kind of discussion or commentary on the PSA detracts from its effectiveness.
Text below is from http://www.asc.upenn.edu/News/PressDetail.aspx?id=137
************
Shi, Messaris, and Cappella worked with a group of nearly 600 adult regular smokers. Participants completed an online survey where they were told they were testing a new website where people can share health related video clips and PSAs. Everyone was shown and asked to respond to an equal number of anti-smoking PSAs. Each PSA had a carefully balanced mix of comments that were:
- Positive and civil
- Positive and uncivil
- Negative and civil
- Negative and uncivil
- Mixed positive and negative comments
- No comments, just PSAs
The results showed the PSAs with no commentary were rated the most effective overall. Any form of commentary – positive, negative and mixed -- made the PSA less persuasive.
I work in the field of normative influence or social norms. Our media posters / billboards / screen savers / buttons / radio PSA that have content describing injunctive norms (norms of approval) for sober driving and descriptive norms (norms of actual sober driving) as well as methods of avoiding intoxicated driving are most effective with college students. See Perkins and Linkenbach article.