StatGun logoStatGun Statistics Consulting

data analysis and statistics professionals

t test
an inferential test that compares a group of scores to a population mean, or two groups against each other.

There are several types of t-test, and they perform different tasks. The most commonly used is the two-sample t-test.

A one-sample t-test is a way of seeing if a particular group of scores is unusual (ie, differs from the expected average).

A two sample t-test is a way of seeing if two groups differ in some way.

A matched-sample t-test compares two groups, where each person in one group has a "match" in the other group. Matched individuals have similar characteristics on key variables. For example, they may be the same age, height, and weight in a medical study. Or in an educational study, subject pairs might be matched on education level and intelligence.

   
About Us | Site Map | Privacy Policy | Contact Us | ©2007 Statgun Statistics