Sample size calculation in medical education research

Abstract

An important aspect in the research methodology, is the calculation of the number of participants that must be included in a study, since the sample size allows the researchers to know how many individuals it is necessary to study in order to estimate a parameter with the desired degree of confidence, or the number needed in order to detect a certain difference between the study groups, assuming that exist actually.
The calculation of the sample size is a mathematical function that expresses the relationship between the variables, amount of participants and statistical power.
A sample from a study should be representative of the population of interest. The main goal of selecting a sample is to make statistical inferences about the population from which comes from. The selection must be probabilistic.
Statistical factors that determine the sample size are: assumptions, error alpha, beta error, statistical power, and variability, losses in the study and size effect.

We review the formulas used for calculating the sample size in the most common situations in research, as well as the revision of formulas for a faster calculation.
It´s included examples of research in medical education.
Also reviewed are important issues such as: sample size for pilot studies, strategies to reduce the required number of subjects, and software for the sample size calculation.