ABSTRACT
In chapter 3 (this volume), one of the questions posed in the model of content analysis asked, “How much data would be needed to test the hypotheses or answer the research questions?” To begin to answer this question, a distinction must be made between all relevant content and a subset of this content. In an ideal world, sampling would not be an issue for social scientists. Researchers would include all relevant content in their studies. For example, a study concerning gender representation on television would examine every program on every channel during all pertinent time periods.
