The great between-subjects assumption

Abstract

There seems to be an assumption that some judgment and decision-making (JDM) findings will only be supported in between-subjects designs, even though between-subjects results can be misleading. This paper tested this assumption by taking two famous between-subjects JDM findings (Shafir, 1993; Tversky & Kahneman, 1986) and replicating them within subjects. Replicating within subjects allowed us to conduct a manipulation check, and the qualitative data gathered surprisingly suggest the original experiments lack construct validity