Hypothesis Tests in Practice

Abstract

A common task in clinical research studies is to test whether a characteristic of interest differs between two different groups in a given population. For example, research studies may wish to compare weight among children who are treated versus not treated with stimulant medications, cognitive function among adults with and without a specific genetic variant, or cardiac ejection fraction among persons with and without thyroid disease. Potential differences in these characteristics observed among the people in a particular study may or may not reflect true differences in the underlying population. Two-sample hypothesis tests are statistical procedures that make an inference about the difference in a characteristic between two groups in an underlying population.