What Type Of Test Do Most Behavioral And Social Science Research Articles Use?
There is a larger variety of research methods in behavioral science than in the other sciences. Among the most frequently published are true experiments, quasi-experiments (including longitudinal research), surveys, ethnography, content analysis, and in-depth interviews (including focus groups). Each methodology tends toward different statistical tests. Experiments and quasi-experiments tend to use a variety of analysis of variance procedures, including multivariate, time series, and contrast analyses. Surveys tend toward regression techniques, ranging in complexity from simple zero order correlation to multivariate hierarchical non-linear analysis. Content analyses typically report frequency statistics with confidence intervals around the parameter estimates, dependent on sampling technique. Ethnography and in-depth interviews do not typically use statistical tests, though they may report descriptive statistics such as frequencies, cross-tabs, etc. Essentially, the tool (method or sta