Abstract: Do Perceived Message Effectiveness Ratings Change Over Time?

◆ Youjin Jang, University of North Carolina at Chapel Hill
◆ Nisha Gottfredson O’Shea, RTI International
◆ Marissa G. Hall, University of North Carolina at Chapel Hill
◆ Noel T. Brewer, University of North Carolina at Chapel Hill
◆ Seth Noar, University of North Carolina at Chapel Hill

Introduction. Perceived message effectiveness – both effects and message perceptions – are used for both message pre-testing and as an early indicator of campaign receptivity. Researchers use these two types of PME interchangeably, assuming that they operate similarly. Studies have yet to examine, however, whether PME changes over time in response to multiple message exposures and whether this differs for effect vs. message perceptions.

Objectives. We sought to understand 1) how initial and repeated exposure to the Food and Drug Administration’s Real Cost vaping prevention ads affected adolescents’ PME ratings, and 2) whether the impact differed for effect versus message perceptions.

Methods. Data came from a 3-arm randomized clinical trial with parallel assignment. Participants were n=1,514 US adolescents aged 13 to 17 years who were susceptible to vaping or current e-cigarette users, recruited from online panels. Participants were randomized to 1 of 2 Real Cost vaping prevention intervention arms (health harms or addiction-themed advertisements) or to a control arm (neutral videos about vaping). Participants watched 3 randomly ordered 30-second video advertisements online in their assigned condition at each of 3 weekly study visits (visits 1, 2, and 3). Participants completed 4 weekly online surveys at visits 1-4 over a 3-week period. Analyses combined the intervention arms and compared with control.

Results. Adolescents were primarily in high school (71%), white (71%) or Black (25%), male (75%), and straight (95%). Sixty-one percent had vaped in the past 30 days and the remaining 39% were susceptible to vaping. After first message exposure, effects and message perceptions of Real Cost ads were both greater than control ads (p<.05), with message perception scores being higher than effect perception scores in both trial arms. After repeated exposures, effects perceptions grew in both trial arms, with greater increases in the control trial arm (p<.05 for interaction). However, message perceptions did not change over time. Adding demographic characteristics to the repeated message exposures analyses revealed no interactions for effect or message perceptions.

Conclusions. First message exposure resulted in similar effects on effects and message perceptions, suggesting that both measures may be appropriate for message pre-testing. However, only the effects perception measure changed over time in response to multiple message exposures. This indicates that when applying PME in real-world campaign receptivity studies, researchers can use effect perceptions (but not message perceptions) as an early indicator of message impact.