Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)



First Advisor

Robert C. Hornik


One construct that is useful when discussing media effects is the notion of encoded exposure, described here as a retrievable memory trace in an individual. Based on past research, encoded exposure to electronic media content should be associated with a variety of predictors on multiple levels of conceptualization, including variables related to the environmental prevalence of media content in question, individual media use, interest, processing ability and tendency, conversation with others, and formal content features. Past work also suggests that an explicitly multilevel model of encoded exposure including such predictors should be more useful than single-level prediction efforts alone. This volume describes and validates a recognition-based measure of encoded exposure developed as part of an evaluation of a national health communication effort, namely an antidrug mass media campaign sponsored by the U.S. Office of National Drug Control Policy. In order to test a multilevel model of encoded exposure, this study assesses three types of data. The National Survey of Parents and Youth, administered by Westat and the University of Pennsylvania’s Annenberg School for Communication to a nationally representative sample of U.S. adolescents, contributed both the encoded exposure measure and a variety of other individual measures. Television gross rating point estimates from the campaign provided environmental prevalence indicators. Lastly, electronic copies of campaign television advertisements also were a rich source of data concerning formal content features. Not all hypothesized predictors garnered significant coefficients in the final analyses. As hypothesized, nonetheless, a multilevel model of encoded exposure (including significant individual-level predictors and significant content-level predictors) found strong support among a sample of U.S. adolescents with regards to television content from the campaign. In short, encoded exposure appears to be both related to individual-level variables, such as media use and conversation with others, and a function of content-level variables, such an environmental prevalence and formal features related to the depiction of time and space.