ABSTRACT
Content analysts have a variety of sampling techniques at their disposal for selecting content. The appropriate one depends on the theoretical issues and practical problems inherent in the research project. If the number of recording units involved is small, a census of all content should be conducted. If the number of units is large, a probability sample is likely to be more appropriate because it allows inference to the population from the sample. Available probability sampling methods include: simple random, systematic, stratified, cluster, and multistage sampling. The appropriate probability sample also depends on the nature of the research project. However, probability samples are necessary if one hopes to use statistical inference. Efficient sampling of mass media to infer to content for a given time period often involves stratified sampling because mass media content varies systematically with time periods. Guidelines are provided in the chapter for sampling content from legacy media. Content analysts need to be aware that sampling may involve probability samples based on time, content, or both.
