Tell them what they don't know
Information
Information interventions provide a fact the recipient didn't have or hadn't focused on — a financial aid eligibility, a school quality rating, a deadline. The mechanism assumes that the gap between behavior and intent is partly informational.
23
Experiments
9
Policy areas
1976–2024
Year span
14 / 23
Positive
When it works
When the gap actually is informational — the recipient would have acted differently if they'd known. Information about a specific local opportunity that the recipient can immediately act on produces the largest, most consistent effects.
Watch out for
Information alone rarely works for behaviors that are dominated by other factors (time, money, executive function). Vaccination information campaigns aimed at already-informed skeptics are the canonical null finding. Information needs to meet the recipient at a point where they have the capacity to use it.
Information across policy areas
Public Health· 7 experiments
- Positive
SMS Reminders for Childhood Vaccination — Meta-Analysis
Systematic review (international) · Global (multiple countries) · 2024
Effect: RR = 1.11 (95% CI: 1.05–1.17); effect consistent across income settings
- Null
COVID-19 Vaccination SMS Reminders
Rhode Island Department of Health · Rhode Island, USA · 2021
Effect: Largest effect: +0.2 pp (2.0% control vs. 2.2% best treatment arm) — statistically significant but practically small
- Positive
Community-Level Mask Promotion — COVID-19
Yale / Stanford research partnership with Bangladeshi NGOs · Bangladesh · 2020
Effect: Mask-wearing increased from 13% to 42% in treatment villages; symptomatic COVID-19 reduced by ~11%; symptomatic seroprevalence reduced by ~9% (larger effects for surgical vs. cloth masks)
- Null
Clean Cookstove Adoption in Rural India
University of Chicago / Harvard / IPA · India (Orissa state) · 2016
Effect: Stove adoption high initially but usage fell to 50% after 1 year and 25% after 4 years; no significant improvement in health outcomes detected
- Positive
Vaccination Reminders and Incentives — India
J-PAL / Seva Mandir · Rajasthan, India · 2010
Effect: Control: 6% fully immunized; reliable schedule only: 17%; schedule plus lentil incentive: 38%
- Mixed
Oregon Medicaid Lottery
Oregon Health Authority / MIT / Harvard · Oregon, USA · 2008
Effect: ER use: +40%; doctor visits: +35%; financial hardship: −25% (catastrophic medical bills); depression: −30%; blood sugar, blood pressure, cholesterol: no significant improvements at 2 years
- Mixed
Mexico Seguro Popular — Health Insurance Coverage Expansion
Harvard School of Public Health / Mexican Ministry of Health · Mexico · 2002
Effect: Catastrophic health expenditure: −23% reduction in treatment municipalities at 10 months; coverage rates rose substantially; outpatient care utilization increased; no significant effect found on mortality or child health in the short follow-up window — consistent with Mexico Oregon Medicaid parallels
Public Safety· 6 experiments
- Null
Police Body-Worn Cameras — Washington DC Randomized Trial
The Lab @ DC / Metropolitan Police Department · Washington, DC, USA · 2016
Effect: Use of force: no statistically significant difference between camera and no-camera officers (effect size near zero). Civilian complaints: no statistically significant difference. Arrests: no significant difference. The null findings were precisely estimated, ruling out effects larger than approximately 2 per 1,000 officer hours.
- Mixed
Ban the Box Employment Policy
New York City Commission on Human Rights · New York City, NY, United States · 2015
Effect: Employment of people with records: modest increase; callback rates for Black applicants without records: declined 7% as employers statistically discriminated
- Positive
Court Date SMS Reminders to Reduce Failure to Appear
City of New York / ideas42 · New York City, NY, United States · 2015
Effect: FTA reduced by 21%; equivalent to 30,000 fewer arrest warrants annually if scaled citywide
- Mixed
Pretrial Risk Assessment Tool (Arnold Foundation)
Laura and John Arnold Foundation · United States (multiple jurisdictions) · 2014
Effect: Mixed: some jurisdictions reduced detention without increasing FTA; others saw no change or increased detention of low-risk defendants
- Null
D.A.R.E. — Drug Abuse Resistance Education
Los Angeles Police Department / Purdue University / multiple evaluators · Multiple US cities · 1983
Effect: Virtually zero effect on drug use at all follow-up intervals. Ennett et al. (1994): effect size d=0.06 on drug use (near zero and below threshold of practical significance). Lynam et al. (1999): DARE and control students showed no significant differences in drug use, attitudes, or drug-related behaviors at 10-year follow-up. GAO (2003): none of 6 long-term evaluations found significant reductions in drug use.
- Negative
Scared Straight — Juvenile Awareness Project
Rahway State Prison / Rutgers University · New Jersey, USA · 1976
Effect: Scared Straight programs produced a 13% increase in re-arrest rates compared to controls (pooled RCT estimate); individual studies ranged from −3% to +28% on re-arrest; no study found significant positive effect
Voter Engagement· 3 experiments
- Positive
Voter Registration SMS — Local Authorities
UK Electoral Commission / Local authorities · United Kingdom · 2022
Effect: +1.5 to +3 pp registration rate depending on message frame
- Positive
Voter Registration Postcards to Unregistered Eligible Citizens
State election agencies (two states) · United States · 2019
Effect: +2.7 pp registration (treatment 9.6% vs. control 6.9%); statistically significant turnout increase
- Positive
Text Message Get-Out-the-Vote Campaigns
Stanford / Yale / IPA (Malhotra, Michelson, Rogers, Valenzuela and others) · California and multiple US states · 2011
Effect: Average effect: 0.3–0.9 percentage points across studies; effects are reliable but small per message; at scale (statewide campaigns reaching millions), effects translate to tens of thousands of additional votes
Financial Services· 2 experiments
- Positive
Buy-Now-Pay-Later SMS Late Payment Reduction
AFM (Dutch Authority for the Financial Markets) + Riverty · Netherlands · 2024
Effect: −20% of customers charged late fees (1 in 5 avoided a fee)
- Positive
M-Pesa Mobile Money and Poverty Reduction
MIT / IGC / Safaricom · Kenya · 2016
Effect: M-Pesa access increased per capita consumption 2%; 2% of households lifted from extreme poverty; female-headed households: +22% consumption