Date of Graduation

8-2022

Document Type

Thesis

Degree Name

Master of Science in Agricultural Economics (MS)

Degree Level

Graduate

Department

Agricultural Economics and Agribusiness

Advisor/Mentor

Rodolfo M. Nayga, Jr.

Committee Member

Di Fang

Second Committee Member

Alvaro Durand-Morat

Keywords

Choice Experiments, Choice modelling, Hypothetical Bias, Information Provision Modalities, Plant-based meat, Uncertainty

Abstract

Stated discrete choice experiments are extensively used in applied economics to study preferences and valuation for new products, as well as costs and benefits of new policies and programs. Moreover, information provision experiments widely use the method to examine information effects on different outcomes. This thesis explores two methodological issues in discrete choice experiments: (i) information provision modalities and (ii) hypothetical bias. The first study examines the effect of information modality by testing the effect of using combined text script and audio clip (treatment) versus only text script (control) to convey information in discrete choice experiments. Specifically, the study elicits willingness to accept agricultural field jobs of low-skilled nonmigrant workers in the US amid the COVID-19. Using an online discrete choice experiment, subjects were randomly assigned to one of the information modalities. The findings indicate that respondents treated with the combined text script and audio clip were willing to accept more for three out of seven attributes. The treatment effect was detected for two more attributes when estimates were conditional on attribute attendance. Moreover, the treatment lowered the prevalence of attribute non-attendance. The second paper assesses the effectiveness of budget reminder to mitigate the hypothetical bias relative to cheap talk and cheap talk with budget reminder. Moreover, it explores the impact of accounting for uncertainty directly in the choice tasks on respondents’ choices. We conducted a laboratory experiment and randomly assigned subjects to a control group and five treatments to elicit their willingness to pay for animal-based and plant-based burgers. The results suggest that the choice task uncertainty adjustment reduced the likelihood of choosing the no-buy option. Furthermore, respondents exhibited hypothetical bias and overstated their willingness to pay by a factor of 1.29 and 1.40 for the animal-based and the plant-based burgers, respectively. Budget reminder reduced the hypothetical bias for the animal-based burger, while cheap talk and cheap talk combined with budget reminder eliminated it for both products. This thesis revisits the potential of budget reminder and cheap talk to mitigate hypothetical bias and magnifies the importance of information provision modalities in discrete choice experiments.

Share

COinS