Research

This chapter presents recommendations for designing experimental studies to test variations in message elements on persuasive outcome variables, for analyzing and interpreting data, and for reporting results. The section on research design covers the importance of using theory to guide message design research, including a control condition in experiments, keeping attribute versus variable status of message elements in mind, recruiting an adequate number of research participants, testing multiple instantiations of messages, and being aware of the implications for internal validity and ecological validity in how test messages are presented to research participants; the section also addresses issues relevant to study selection for meta-analyses. The section on data analysis and interpretation covers random versus fixed effects and the importance of not confusing individual-level and message-level correlations, not confusing effect sizes and effects, and not making claims about changes in dependent variables based on posttest data alone. Finally, the section on reporting results discusses the importance of following established reporting standards for original studies and meta-analyses and presents a list of 10 reporting standards for message testing studies. Each section includes a list of recommended readings.