Information processing in choice‐based conjoint experiments : A process‐tracing study

Purpose – This paper investigates how respondents to conjoint experiments process information and choose among product profiles, and how this varies with their knowledge about the product. Models for estimating conjoint attribute weights are almost exclusively based on principles of compensatory decision making. The paper aims to explore to what extent and in what way these basic principles of conjoint modelling are violated. Design/methodology/approach – Data were obtained from a verbal protocol study where 18 undergraduate students each performed a total of 28 stated choice tasks while “thinking aloud”. Findings – Results show that cognitive operations consistent with compensatory decision rules constitute a majority of the total number of operations performed across tasks and respondents. However, few respondents exhibited a consistent use of compensatory‐type processes throughout their choice sets. Results suggest that individual preferences interact with characteristics of the choice sets to instigate changes in information processing. It also appears that complete strategies are seldom used. Finally, respondents' knowledge about the product influences the cognitive operations that respondents use in solving conjoint tasks. Research limitations/implications – Results are based on responses from 18 undergraduate students, which makes generalizations hard. Practical implications – One implication of this work is that one should apply a more flexible model framework to allow detecting the existence of non‐compensatory strategies. Originality/value – This paper is one of few which aim to implement findings in behavvioral decision research within the context of conjoint analysis.