Within- vs. between-subjects

Within-subjects

Also known as repeated-measures
The same subject, but repeated measurements

Between-subjects

Comparing results of subject_i with subject_k
Avoids carry-over effects (where the subject learns from one trial and this causes a difference in subsequent trials)

Mixed design

Transcript

Next, I need to think about: "Am I doing comparisons within or between?". Within - that means that I'm concerned with what's called repeatability measures - it's the same subject but repeated measurements. So, I have a person and a task, how long does it take to do that the first time, the second time, the third time, etc. Or I may say, "Ah! it's about a comparison between subjects?" How long does it take person A to do it versus how long does it take for person B to do it? But now, I have a problem so-called carryover problems - the method that they applied the first time may lead to learning - so when they apply the second method, it takes them less time because they actually learned things from the first time. So, we need to be very, very careful. So, then, of course, we make sure that we have another group that does them in the opposite order to compensate for that. So we need to think about avoiding the carryover effect, and of course, we can go very sorts of mixed designs.