t-test

#1
I'm confused about when to use a t test for two independent samples and when to use a repeated measures t test. When there is a pacebo group in an experiment, would I do a t test for 2 indep. samples? When is it related? Also how would I know if I should run a 2 tailed or a 1 tailed test..?
 
Last edited:

JohnM

TS Contributor
#2
When the items / objects / particiapnts in the two groups have nothing to do with each other, then they are independent, and you use the independent samples t-test.

When they are related (i.e., two measurements on the same items, objects, people, or if the two samples are related in some way, such as convenience, or a sibling, spouse, etc.) then you use the dependent samples test.

A one-tail test is used when you are interested in the direction (large than, smaller than) of the difference between groups. If all you want to know is whether they are different, and you don't particularly care about the direction of the difference, then use a two-tailed test.