Paired or un-paired?

[MultiD Support]

When shall I choose paired nd when shall I choose unpaired t-test?

[MultiD Support]

Hello,
You should use paired test when a measurement in one group is correlated to a measurement in the other group. Typical case is if you measure twice on each subject; once before treatment and once after treatment. If there is no correlation between the subjects you should use un-paired tests.

A paired test is more powerful than an unpaired test. Hence, if possible paired experiments should be designed.

Mikael