ATTRIBUTE NON-ATTENDANCE IN FOOD CHOICE EXPERIMENTS UNDER VARYING INFORMATION LOAD

It is increasingly recognized that respondents use simple heuristics such as attribute non-attendance to make decisions in the discrete choice experiments. This paper use the latent class model to investigate different choice strategies and explore robust welfare estimates under varying attribute information load. We find that respondents are more likely to rely on simple heuristics to make choices when information load increases. Reinforcing previous findings, we also observe that willingness-to-pay estimates decrease, with and without accounting for attribute non-attendance.