Essays on decision making processes, information acquisition, and preferences in stated choice experiments: Applications to economic valuation of consumer and environmental goods

PhD defence

Kennet Christian Uggeldahl

Abstract

Economic valuation methods are tools used to value goods and services in situations where markets fail to provide an adequate measure of value. This thesis is based on four papers using the stated choice experiment (SCE) method to value environmental and consumer goods. In a SCE, respondents are presented with a scenario, and asked to choose their most prefer alternative from a choice set.

In paper I, the benefits of riparian buffer strips, an agri-environmental measure, are valued using a SCE. The findings show the value of co-benefits, such as aesthetic and recreational benefits, are significant, and depend on the characteristics of the buffer strips (such as the width).

The three other papers focus more explicitly on the decision process leading up to the choice. In conventional SCEs, like Paper 1, the underlying assumptions treat humans as perfectly informed utility maximizers. A growing body of research, however, has shown that including the decision process is important in deriving valid value estimates. Papers II and III explore and incorporate the decision process using eye-tracking in SCEs concerning food items. Eye-tracking provides a measure of respondents' information acquisition. Paper II shows that respondents that more frequently switch their attention (i.e. where they look) between the presented alternatives are more uncertain in their choices. In paper III, a novel eye-tracking setup in SCE is used, which precisely measures the information that people ignore. In both papers, eye-tracking information improves the performance of the choice model. Paper IV focusses on improving the content validity of an online SCE survey, by presenting the information in the survey as a narrated video compared to text and illustrations. The findings in paper IV are mixed, with choices from the narrated video treatment being more accurate, however, this might be due to certain types of respondents choosing not to complete the survey in the video treatment.

Overall, findings will provide stepping stones for future analyses of the decision process in SCEs, which, in turn, will pave the way for more realistic behavioral models and reliable value estimates from SCEs.

Supervisors

Associate Professor, Søren Bøye Olsen
Associate Professor, Thomas Hedemark Lundhede

Assessment Committee

• Professor Jette Bredahl Jacobsen (chair), Department of Food and Resource Economics, University of Copenhagen
• Research Associate Jürgen Meyerhoff, Technische Universität Berlin
• Professor Fredrik Carlsson, University of Gothenburg