Date of Award
Program or Major
Doctor of Philosophy
This dissertation examines a set of behavioral and cognitive issues in the stated preference non-market valuation framework and economic pedagogy. Drawing from the fields of behavioral and environmental economics, this research examines two empirical deviations between theory and behavior, namely, anchoring effects and attribute non-attendance. This work contributes to the non-market valuation literature by attempting to correct for and explain these behavioral anomalies, both ex-ante and ex-post. In the presence of these anomalies, corrections are necessary if the tenets of welfare economics are to remain a viable analytical foundation for valuing environmental goods.
In the first chapter, I examine the issue of price anchoring in a discrete choice experiment. Price anchoring occurs in a DCE when respondents anchor their preferences for a good on the price presented in the choice exercise, causing bias in welfare estimates when the ‘anchor’ deviates from respondents’ true underlying preferences. The cheap talk approach of Cummings and Taylor (1999) is adapted as an ex-ante mechanism to mitigate anchoring bias. This approach is applied to a choice experiment used to examine consumers’ willingness-to-pay for local and organic agriculture in Northern New England. Results suggest that anchoring effects are present, though only for certain vegetable/attribute combinations. In particular, price anchoring increases WTP estimates between 26%-187%. Further, exposure to the anchoring-specific cheap-talk treatment can be associated with a reduction of anchoring bias between 30% and 82%.
In the second chapter, we attempt to explain the presence of attribute non-attendance (ANA) in the discrete choice experiment through previous purchasing experience. Attribute non-attendance arises in the DCE framework if decision-makers do not attend to all of the attributes presented in the choice exercise and is inferred via the method developed in Hess and Hensher (2010). Results show that the inferred ANA method of Hess and Hensher (2010) uncovers non-attendance in both the local and organic attribute in this choice experiment for tomatoes. In particular, we find that inferred measures of ANA fall when estimating separately by those respondents with and without previous market experience. With respect to welfare estimates, we find that after controlling for previous purchasing experience, WTP estimates increase from $0.91 to $1.12 per pound on the local attribute and increase from $0.98 per pound to $1.18 per pound on the organic attribute. A latent class model is used to corroborate these findings, with the intent of uncovering groups of respondents within the data. This analysis uncovered two groups of respondents who vary in their underlying preference structure, one of which groups exhibited zero preference for the local and organic attributes, mimicking the behavior of those respondents who are thought to be “not-attending”, while the other exhibits strong preferences for each of these attributes. Given this finding, there was no statistical difference between the two groups in terms of previous market experience for either attribute. Overall, this analysis contributes to the literature explaining ANA in the choice experiment framework and provides methodological recommendations for the design of choice experiments.
Finally, in the third chapter, our attention turns to economics education. With the intent of developing students who are self-regulated learners, this study attempts to (1) measure students’ able to accurately self-evaluate the quality of their own work, (2) determine if self-evaluation skills of students improve over the course of a semester, (3) investigate and explain differences in self-evaluation skills across students, and (4) understand whether increased self-evaluation skills improve learning outcomes. The data is generated from a cohort of undergraduate students in an Environmental Economics course at a US university. A repeated self-evaluation exercise was required on a series of writing assignments to test whether self-evaluation skills improve after repeated experience with the self-evaluation process. Results show an increase in self-evaluation abilities by about 32% across the sample, where females and those with past-evaluation experience explaining much of this improvement. Further, we find both over-confidence from those of lesser abilities and under-confidence from those of greater abilities, consistent with that found in Dunning and Kruger (1999). These effects largely disappear by the final self-evaluation procedure, suggesting that both low and high ability students can improve self-evaluation skills by simply engaging in repeated self-evaluation procedures.
Lemos, Scott, "Essays on Behavioral and Cognitive Anomalies: Implications for Stated Preference Valuation and Economic Pedagogy" (2019). Doctoral Dissertations. 2479.