Individual randomization generates exogenous variation in the share of friends who receive a product while holding total friendship network size constant.

Download image · 1200 × 630 PNG

Research design

Allocate cups by public lottery

Assign participant IDs; draw 25 numbers per school from a bag

Selected participants
Remain after the meeting for cups and direct usage training, with guardians.
Unselected participants
Join the control group; receive cups at study end.

Illustrative summary of the source’s public individual lottery across four schools in Chitwan, Nepal. The study enrolled 198 girls. The diagram describes allocation, not adoption outcomes.

See source →

How

When to use

Organisations need to allocate scarce health products transparently among eligible study participants. Standard cluster allocation can obscure peer effects and create perceived unfairness among participants. Public individual lotteries generate exogenous variation in product ownership while maintaining participant trust.

Allocate new technology items individually through a public random drawing at an initial participant gathering.

  1. Assign unique identification numbers to all enrolled participants at the conclusion of an introductory enrollment session.
  2. Place all assigned identification numbers into an opaque bag for a public drawing in front of attendees.
  3. Draw twenty-five identification numbers from the bag at each target location to select treatment recipients.
  4. Assign the selected participants and their co-attending guardians to the treatment group while assigning unselected participants to control.
  5. Retain treatment recipients after the general meeting to receive the distributed technology and direct usage training.
  6. Inform control group participants that they will receive the technology product at the end of the study period.

Why it works

Public drawing prevents selection bias and ensures transparent assignment in front of participants.

Example

In four schools across Chitwan, Nepal, study staff assigned identification numbers to 198 adolescent girls and drew 25 numbers per school from a bag, assigning half the total sample to receive menstrual cups and holding the remaining half as controls.

About the study

Participants

198 adolescent female students in 7th and 8th grades across four schools in Chitwan District, Nepal.

Study Design

Individual public bag lottery drawn at baseline, selecting 25 girls per school for immediate menstrual cup receipt and assigning remaining girls to delayed control.

Results

50% of participants were randomized into treatment; 1 additional treatment friend increased cup usage by 18.6 percentage points at two months.

Source

See source →