Description: Recorded lecture on the basic concepts of power and sample size for experimental studies.
Links:
Who is this resource for: Clinical investigators, statisticians
Who created this resource: This is part of a lecture series ‘Foundations of Biostatistics and Epidemiology’ which was developed by the Oklahoma Shared Clinical and Translational Resources (OSCTR), part of the NIH-funded Oklahoma Clinical and Translational Sciences Institute (OCTSI) at the University of Oklahoma Health Sciences Center. This particular lecture was given by Julie Stoner, PhD, Professor and Chair, Department of Biostatistics and Epidemiology.
What does this resource offer:
This lecture concisely presents general power/sample size concepts, without giving details about the calculations for specific statistical tests. Dr. Stoner writes, “In this module I will briefly review information related to hypothesis testing, which will form the framework for our discussion of sample size requirements. I will also present a summary of the primary factors that influence sample size estimates, namely, the significance level, power, effect size and variability. Keep in mind that additional study design factors, such as [study] withdrawal and longitudinal repeated measures, also impact the sample size.”