Originally posted by: optimistic
Yeah, that's what I'm going for 'variables' .......wait..that could work. Hmm, anyways I thought there was a word that existed that had had the combine meaning of "variables + unknown."
Maybe you're right. The word is probably already staring me at the face.
Well, confounding variables are those variables that are unknown and immeasurable that can influence the results of an experiment. In a closed laboratory setting, you would restrict and limit as many variables as possible that could possibly affect the results, limiting the number of confounding variables.
When doing a correlational analysis, you might compare TV watching to intelligence. You find that there is a direct correlation showing that the more TV you watch, the less intelligent you are. However, there are lots of sources of confounding. For instance, you watch more TV because you don't have access to books. Because you don't have access to books, you read less and are less intelligent. You get what I'm saying I think.