Fritz Sang

Developmental Trends in Television Coviewing of Parent-Child Dyads

In this study, long-term trends in joint parent-child television viewing are taken as prototype for developmental changes of the parent-child relationship during adolescence. To describe and compare trends of television coviewing in different configurations of family members, trajectories of daily television viewing, tmeasured with Nielsen-type people meters over a time period of three years, were analyzed using time series methods. The findings show the transferability of hypotheses about general developments in relations between the generations on coviewing behavior of adolescents aged 14–16 with their parents to predict differential declines in the four parent-child dyads.

Individuation and television coviewing in the family: Developmental trends in the viewing behavior of adolescents

This study tests the validity of two competing models of family coviewing. Trajectories of daily television viewing measured with people meters over a 3‐year period were analyzed using time‐series methods. The findings show that current models of coviewing are in need of refinement. The proposed expansion by a longitudinal perspective allows for differential predictions about age‐related trends in adolescent viewing behavior in relation to parental coviewing patterns.

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

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.

Time-Series Analysis of Longitudinal People-Meter Data

Realizing that worldwide longitudinal television research has exclusively been based on short sequences of cross-sectional data, it is argued that future developmental television research should use people-meter data in combination with time-series methods. To support this claim, several issues are investigated. It is demonstrated how telemetric TV data, which are collected continuously in all industrialized countries, can be used not only to validate findings based on cross-sectional data but also to expand our knowledge about developments in TV use by individuals and groups. This is an implicit plea to use these already-existing data pools for social and media scientific purposes. The examples presented are based on individual and aggregated trajectories of daily television viewing time covering a period of three years.