I'm interested in understanding what does it mean to build a statistically optimal design. Treats et al., 2020 (Generating Optimal Designs for Discrete Choice Experiments in R: The idefix Package) writes that this design "aims to select those choice sets that force the respondent to make trade-offs, hereby maximizing the information gained from each observed choice". What does it mean exactly to "force respondents to make trade-offs"?
Thank you in advance,
Alice
Forced trade off means that two profiles on the choice set do not share the same attribute levels forcing the respondent to consider which attribute level is more valuable. When profiles share the same attribute levels then the respondent can see they are identical and not need to consider them in their decision other than if you have a demand response such as would you consider either option against a none option. Efficient designs generally maximise such differences so that respondents are forced to trade off across multiple levels. For example am I willing to forgo level 1 in lieu of level 2 on attribute 1 in order to have level 1 on attribute 2.
- Badges
- Science topic
- Similar topics
- Geoscience
- Asia
- India