Using Self-Generated Priors in Division Leads to Biased Consumer Judgments
Consumers often have to make divisions to evaluate attributes. In six experiments (total N = 3296, four preregistered), this research shows that consumers may rely on their prior for the attribute to generate an estimate rather than divide following a normative procedure. That is, consumers will recruit typical values of the attribute from memory and use them when dividing. This process influences consumers’ judgments. When the actual attribute value is above the attribute prior, it tends to be underestimated, and when it is below the attribute prior, it tends to be overestimated. When the attribute is a product attribute (such the miles-per-gallon a car can drive), this decreases its attractiveness. When the attribute is a company-wide attribute (such as median wage), this intuitive approach results in corresponding changes in company evaluations, willingness-to-pay, and product choice. The authors discuss theoretical implications for consumers’ numerical cognition, anchoring and adjustment, and perceptions of income inequality. Practical implications for the communication and disclosure of product attribute information and income inequality for managers and policymakers are also discussed.