Deconstructive replication: How to tell if a thought experiment-population pairing is fit for purpose

Abstract

Thought experiments/scenarios are often rolled out to help answer research questions The idea is that they tell us what people really think, value, etc. about the research question But how do we know whether the judgments elicited are really about the research question? Scenarios might work well for some people but not others. What we need is a way to assess a scenario-population pairing’s fitness for purpose in relation to a particular research question