Repeated follow-up as a method for reducing non-trading behaviour in discrete choice experiments.

Eliciting individuals' preferences using discrete choice experiments is becoming increasingly popular. An emerging issue is that some people do not trade, that is, their choices are consistent with them having a dominant preference. However, it may be the case that these individuals have not been presented with the 'right' trade-offs. This experiment presents individuals with trade-offs that are systematically varied in response to their previous answer. It is thus assessed whether individuals have truly dominant preferences or whether they will trade given the 'right' choices. The preferences studied are time preferences over future health states. After being presented with an initial discrete choice, 203 university students were asked a series of follow-up questions using a web-based questionnaire. Very little evidence of any dominant preferences was found in this sample of respondents. Only one subject did not trade duration and timing following repeated follow-up questions. This finding suggests that non-trading behaviour could be virtually eliminated by asking the 'right' questions.