Reactive Effects of a Survey On the Television Viewing Behavior of a Telemetric Television Audience Panel

Abstract

Data from telemetric television audience panels are invaluable for commercial and scientific television reseanch. For the study of most research questions, however, it is necessary to request additional information from the panel participants. Such interventions can only be permitted if they do not threaten the validity of the telemetric data. This study investigates whether a questionnaire intervention disrupts the viewing behavior of a television audience panel. The results indicate that, on the day they completed a questionnaire, experimental subjects watched a small amount more television than controls who received no questionnaire. It is argued that such a small effect on one day should not be accorded practical relevance. In addition, the article introduces a combined time-series and control-group design with which intervention-effect hypotheses can be investigated. It is suggested and illustrated that this design has a number of important advantages over traditional approaches.