Multiplicity Problem

Typically, a research project involves testing multiple research hypotheses. These research hypotheses could be evaluated using, for example, comparisons of means, bivariate correlations, or regressions. In fact, most studies consist of a mixture of several different types of test statistics. An important consideration when conducting multiple tests of significance is how to deal with the increased likelihood (relative to conducting a single test of significance) of falsely declaring one (or more) hypothesis(es) statistically significant (i.e., a Type I error). This is termed the multiplicity problem. This multiplicity problem is especially relevant to the topic of research design because the issues associated with the multiplicity problem relate directly to designing studies (i.e., number and nature of variables to include) and deriving a data analysis strategy (e.g., ...