Description
Cross the quality of a prior experience with its relatedness to the current one to separate how past experiences raise satisfaction (endowment) or lower it (contrast).
A past experience affects satisfaction with the present in two opposing ways: a good memory adds to well-being (endowment), but a related good experience also raises the standard against which the present is judged (contrast). Tversky and Griffin separated them with a 2 x 2 design: prior experience good or poor, and related or unrelated to the current one, with the current experience held constant. Endowment is the rating difference after good versus poor unrelated priors; contrast is how much a related prior shifts ratings relative to an unrelated one. Vignettes on a 1 to 10 scale gave endowment around 2.2 and contrast around 0.9; a real-money game gave 2.3 and 2.1, with negative contrast larger than positive. The same work shows choices and satisfaction judgements can diverge.
Attributes
- Discipline
- Experimentation & Causal Inference
- Loss type
- Unmigrated
- Stage
- Design, Evaluation
- Data
- Quantitative
- Effort
- Days
Steps
- Fix the current experience to be evaluated so it is identical for everyone.
- Create good and poor prior experiences, each in a related version (same domain or style) and an unrelated version.
- Randomly assign participants to the four cells and add a filler task between the prior and current experiences.
- Collect satisfaction ratings on the current experience (for example on a 1 to 10 scale).
- Compute endowment E = S(unrelated good) minus S(unrelated poor) and the two contrast terms from related versus unrelated cells.
- Report the asymmetry between positive and negative contrast, and check whether choices diverge from ratings.
Sources
- Preference, Belief, and Similarity (Tversky, 2004)
Also known as
- Endowment and contrast test
- Prior experience decomposition
Categories
- Analytics & Behaviour
- Experimentation & Causal Inference