Comparing Choice Question Formats for Evaluating Natural Resource Tradeoffs

Choice questions are increasingly being used to scale competing natural resource programs. Respondents choose between two alternatives with varying levels of program characteristics and costs. Complexity in the choice task can increase the randomness (variance) in the choices and the estimation of preferences, and the magnitude of randomness is examined using scope tests and scale parameters. We provide an empirical comparison of response variance from three formats. A simple resource-to-resource format appears superior to simple referendum and composite formats in terms of coherence. The application stems from a study addressing PCB-caused natural resource losses in Green Bay, Wisconsin.