| The growing level of inactivity and overweight among youth requires health communication strategies that can have a widespread impact with reasonable costs. Interventions delivered via print and interactive media have shown promise for increasing physical activity among adults, but little is known about their effectiveness among youth. This study employed a randomized trial design to compare the effectiveness of identical physical activity messages delivered via Web and print media among a sample of middle school girls. Both Web and print groups had significant changes in the primary outcome variables of physical activity self-efficacy and intentions, as well as in physical activity behaviors among those were active three days per week or less at baseline. The print group demonstrated significantly greater changes in intentions and behavior compared to the Web group, and was more likely to recall the materials and recommend them to a friend.;The relative effectiveness of the two media was examined using the Elaboration Likelihood Model (ELM) of persuasion developed by Petty & Cacciopo (1986). A measure of elaboration likelihood was adapted and tested for reliability and validity among this study sample. The scale proved to be reliable and valid for this population, and was significantly correlated with communication outcomes for both the Web and print groups. The findings suggest that differences in information processing between Web and print groups may account for some of the variation in communication outcomes.;It has been suggested that perceived interactivity is a key feature of Web-based media that may enhance learning and persuasion. However, existing measures of this construct have not been validated among youth and do not allow for comparisons with non-Web based media. A measure of perceived interactivity suitable for Web and print media was developed and tested for reliability and validity and found not to perform well among a sample of adolescent girls. These findings are likely due to challenges relating to the study population, message content, and the multi-dimensional nature of this construct. |
