Description
Measures how much a pre-selected option drives choice, and why, by comparing outcomes under different defaults or active choice.
Defaults are among the most powerful levers in choice architecture: Madrian and Shea found 401(k) participation rose from about 49% to 86% under automatic enrolment, and organ donor consent differs dramatically between opt-in and opt-out countries. Default effect analysis treats every pre-selected setting as an intervention and measures its effect, using randomised comparison of defaults or active choice where possible, or regression discontinuity and before-after designs where not. It also decomposes the mechanism: inertia and effort, implied endorsement by the designer, and reference-point shifts. Rigour comes from randomisation, measuring long-run outcomes rather than just initial selection, and checking whether people stuck with a default that harms them, which raises an ethical as well as a statistical question.
Attributes
- Discipline
- Behavioral Science & Economics
- Loss type
- Siloed
- Stage
- Design, Evaluation, Post-launch
- Data
- Quantitative
- Effort
- Weeks
Steps
- Inventory all pre-selected options, pre-filled fields and pre-ticked choices in the flow, and record each one's current uptake rate from logs.
- Select defaults with material consequences for users (money, privacy, safety, workload) and state for each the hypothesised mechanism: effort, endorsement or reference point.
- Design arms: current default, alternative default and enforced active choice; calculate sample size for the expected difference (default effects are often 20 to 40 percentage points, so a few hundred users per arm frequently suffices).
- Randomise users to arms and measure initial selection, later switching, and downstream outcomes over a pre-registered window of at least 30 to 90 days.
- Probe mechanism with a short follow-up survey or 6 to 10 interviews asking whether users noticed the default and whether they read it as a recommendation.
- Evaluate welfare: check whether outcomes for defaulted users are good for them, choose the default that best serves the majority, and preserve easy opt-out.
Sources
- Paradox of Choice (Schwartz)
- Think Bigger (Iyengar)
- Happiness by Design (Dolan)
- Methods Atlas (Gopinath)
- The Inquiry Series, Ch 22 — Choice Architecture, Default Effects & Option Set Design
Also known as
- Default option testing
- Opt-in versus opt-out analysis
- Default nudge evaluation
Categories
- Analytics & Behaviour
- Behavioral Science & Economics