A Model of Grocery Shopping Decision Process Based on Verbal Protocol Data

Behavioral decision research has offered two principal classes of models as candidates for representing human decision processes: compensatory (or linear additive) and noncompensatory (or hierarchical). Experimental research has found evidence for both types of models, and it has been argued that decision processes may depend largely on the context of the decision task being performed. This study takes a fresh look at these competing models by examining behavioral data from a natural decision task: American grocery shopping. The methodology is based upon participant observation and the analysis of verbal protocols recorded in supermarkets. In contrast to some experimental results, very few of the cases examined in this study show clear evidence for compensatory behavior. Shoppers appear to evaluate grocery items in ways that allow them to avoid trade-offs on attributes of quality, quantity, and price.