Sample size planning for pilot studies

Abstract

Pilot studies are often the first step of experimental research. It is usually on a smaller scale and the results can inform intervention development, study feasibility and how the study implementation will play out, if such a larger main study is undertaken. This paper illustrates the relationship between pilot study sample size and the performance study design of main studies. We present two simple sample size calculation methods to ensure adequate study planning for main studies. We use numerical examples and simulations to demonstrate the use and performance of proposed methods. Practical heuristic guidelines are provided based on the results.