Article Text
Statistics from Altmetric.com
The investigator must juggle sample size and effect size to produce a study with a priori credibility and ex post facto utility
Sample size and effect size are interrelated parameters that have been given insufficient consideration in analyses of two major outcome measures in the field of sexually transmitted infections (STI) prevention—STI incidence and condom use. Their inter-relation highlights two critical features of interventions—statistical significance and epidemiological importance.
Though there are a myriad variations on the theme of sample size, the calculation usually depends on four parameters. The investigator must designate an acceptable level for type I error (the probability of falsely rejecting a null hypothesis); type II error (the probability of failing to reject a false null hypothesis); variance (the amount of dispersion in the result that would be acceptable); and the effect size (the size of the detectable difference that is deemed important). Cynics will be quick to point out that sample size depends on only one parameter—the amount of money available for the study—but even if true, researchers must still deal with the consequences of the terms so dictated. Because type I and type II error are often set by convention, the investigator must juggle sample size and effect size to produce a study with a priori credibility and ex post facto utility.
SOME STATISTICAL CONSIDERATIONS
An important difference among studies is whether or not they are measuring continuous or dichotomous outcomes. In the former case, effect size can be assessed by transformation into units of standard error, and a test of statistical significance is straightforward. Yet, continuous outcomes are rare in STI research. Incidence of STIs is inherently a dichotomous outcome and condom use is typically dichotomised, based on the assumption that consistent (that is, 100%) use is protective against infection whereas lower rates of use are not protective. …