An empirical model of screening rule choice

We model consideration set formation involving non-compensatory screening rules. Rather than choosing from among alternatives or sets, consumers in our model choose rules. The best rules strike a balance between maximizing utility and minimizing costs of thinking. By adding psychological insights into an econometric framework, we obtain a parsimonious, interpretable model that delivers a wealth of insights from minimal data requirements. The model requires only stated (or revealed) consideration sets, and alternative attributes, and produces interpretable estimates of partworths, thresholds, common rules used, and thinking costs. In addition, the model can be used for managerial counterfactuals. We estimate the model using a dataset of consideration sets from the automobile industry, and explore substantive findings about common screening criteria in automobile consideration, brand competition and the dollar value of costs of thinking, especially in relation to demographics