Over 40 years of research has established the robustness of the similarity effect (SE; Tversky, ), which states that the introduction of new options into choice sets predominantly reduces the choice share of similar options. The present work examines whether the SE systematically reverses when real brands are excluded from assortment subsets, as is the case with stock‐outs in real purchase decisions. To this end, within‐subject decisions are examined under certain out‐of‐stock (OOS) conditions in an enhanced experimental design that resembles real shopping environments. Specifically, unforced choices of experienced consumers, inclusive of real payments, are observed for products in online transactions. The results of two studies corroborate the existence of a reserved SE. Specifically, the OOS‐induced switching patterns systematically refute the assumptions of classic economic theory, since consumers disproportionately switch to alternatives which are similar to the unavailable item in contrast to dissimilar substitutes. Finally, managerial implications and potential directions for follow‐up research in the general domain of marketing are deduced.
Recent findings in research on context effects indicate that, as opposed to choices under hypothetical settings, compromise effects are significantly attenuated in binding settings that include real payments for products. The present investigation examines to what extent this finding applies to yet another context effect, namely the similarity effect (SE). The SE stipulates that, compared with initial shares of two options A and B in a binary choice set, the introduction of an option C that is dissimilar to B but similar to A will hurt the latter more than the nonadjacent option B. The conducted online survey examines the SE in unforced choices of experienced consumers between real brands across three consumer good categories. The results confirm that the SE is evidently reduced under real payment conditions, hence confirming the hypothesis that the applied choice setting moderates the efficacy of the composition of choice sets.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.