home glossary blog contact












a b c d e f g h i j k l m
n o p q r s t u v w x y z

T-test

This is the name of an inferential and parametric test of difference. 

 

The related T-test assesses the difference between related samples (e.g. Repeated Measures Design or matched pairs design) and the unrelated T-test is used for unrelated samples (e.g. independent measures design).

 

The T-test should only really be used if the distribution is normal and the level of measurement is at the interval or ratio level.