Statistical Power

Researchers in criminology and criminal justice have placed much more emphasis on the statistical significance of a study rather than on the statistical power of a study. The lack of attention given to issues of statistical power by criminologists likely reflects a lack of familiarity with the techniques used for assessing the statistical power of a study. Consequently, the purpose of this chapter is to present the key components in an assessment of statistical power, so that criminologists will have a basic understanding of how they can estimate the statistical power of a research design or to estimate the size of sample necessary to achieve a given level of statistical power. Our discussion of statistical power presents the basic conceptual and statistical background on statistical power, with an emphasis on the three key components of statistical power. We illustrate the computation of statistical power estimates, as well as estimates of sample size, for some of the most common (and basic) types of statistical tests researchers will confront. We conclude by highlighting some of the more recent developments in assessing statistical power for more complex multivariate models and future directions for estimating statistical power in criminology and criminal justice.