Abstract
To control for response-shift bias in pretest/posttest designs, Howard and his collaborators proposed the use of retrospective pretests, the so-called thentests. If a response-shift bias is present, thentest/posttest difference scores will yield better estimates of a treatment effect than will the usual pretest/posttest difference scores. In this article a method is presented that is aimed at detecting response-shift bias in retrospective pretest/posttest designs. This method, essentially a hierarchical model fitting procedure, is easy to apply, prevents capitalization on chance, and makes optimal use of information supplied by the control group.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have