I said that in any experiment where we have pretest and posttest scores we could either look at the difference scores (compared across groups) or use the pretest as a covariate. These two analyses...

I said that in any experiment where we have pretest and posttest scores we could either look at the difference scores (compared across groups) or use the pretest as a covariate. These two analyses will be equivalent only when the slope relating posttest to pretest scores is 1.00. How likely do you think it is that such a condition would be met (or at least approximated)? What does b 5 1.00 actually imply?