Description
Audits survey items and research artefacts for compound, stereotype-matching scenarios that respondents rate as more probable than their simpler components, inflating confidence in specific user stories.
The conjunction fallacy occurs when people judge a specific conjunction of events (A and B) as more probable than one of its constituents (A alone), because the detailed scenario matches a stereotype; the classic example is the Linda problem. In UX research it appears in two places. Survey items that bundle several conditions into a vivid scenario, such as 'a busy parent who shops on mobile at night and abandons at payment', attract higher agreement than their components, overstating how common such users are. Teams commit the same error when detailed personas or journey stories feel more plausible than broader evidence supports. The check decomposes compound items into components, compares ratings, and flags inversions where the conjunction outscores a constituent. It enforces a basic probability constraint on attitudinal and planning data.
Attributes
- Discipline
- Survey & Psychometrics
- Loss type
- Latent
- Stage
- Design
- Data
- Analytical / expert
- Effort
- Hours
Steps
- Audit scenario-based survey questions for conjunction fallacy.
- Compare rates for specific vs. general versions.
- Use probability inflation ratio as distortion measure.
Sources
- Heuristics and Biases — Kahneman (ed.)
- Tversky & Kahneman, Extensional versus intuitive reasoning: The conjunction fallacy in probability judgment (Psychological Review, 1983)
Also known as
- Conjunction fallacy audit
- Linda problem check for surveys
Categories
- Analytics & Behaviour
- Survey & Psychometrics