x

Please fill out this short user survey of only 3 questions in order to help us improve the site. We appreciate your feedback!
Algorithmic Nudges
Last registered on October 27, 2020

Pre-Trial

Trial Information
General Information
Title
Algorithmic Nudges
RCT ID
AEARCTR-0006660
Initial registration date
October 23, 2020
Last updated
October 27, 2020 7:20 AM EDT
Location(s)

This section is unavailable to the public. Use the button below to request access to this information.

Request Information
Primary Investigator
Affiliation
University of Chicago
Other Primary Investigator(s)
Additional Trial Information
Status
In development
Start date
2020-10-01
End date
2020-12-31
Secondary IDs
Abstract
Nudges are a combination of words and contexts chosen by the sender to elicit a (desired) response from the receiver without a material change in incentives. We aim to develop and test a framework that algorithmically constructs individual nudges and test the value of this approach in the context of the SNAP program.
External Link(s)
Registration Citation
Citation
Misra, Sanjog. 2020. "Algorithmic Nudges." AEA RCT Registry. October 27. https://doi.org/10.1257/rct.6660-1.0.
Experimental Details
Interventions
Intervention(s)
The intervention involves the construction and sending of algorithmically targeted messages (via sms and email) to those SNAP participants who consented to receiving a reminder. The messages remind the participants that they need to fill out the SAR 7 update form to continue receiving SNAP benefits.
Intervention Start Date
2020-11-01
Intervention End Date
2020-11-07
Primary Outcomes
Primary Outcomes (end points)
SNAP SAR 7 Update filings.
Primary Outcomes (explanation)
Secondary Outcomes
Secondary Outcomes (end points)
Duration between reminder and action.
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
The experiment has four treatments arms:
(1) Control/Status Quo nudge decided by and used by Code for America
(2) Random (Random allocation of nudges derived from optimal design theory)
(3) Optimal (Model based optimal message - estimated using pilot data)
(4) Algorithmic (Model based targeted messages - estimated using pilot data)

Assignment to arms is random with 10%,10%, 30% and 50% to each of the arms listed above.
Experimental Design Details
Not available
Randomization Method
Done by computer.
Randomization Unit
Individual
Was the treatment clustered?
No
Experiment Characteristics
Sample size: planned number of clusters
9966 individuals planned but will depend on how many become eligible.
Sample size: planned number of observations
9966 individuals planned but will depend on how many are actually eligible.
Sample size (or number of clusters) by treatment arms
Control: 996
Optimal: 997
Randomized: 2992
Algorithmic: 4981
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Supporting Documents and Materials

There are documents in this trial unavailable to the public. Use the button below to request access to this information.

Request Information
IRB
INSTITUTIONAL REVIEW BOARDS (IRBs)
IRB Name
University of Chicago IRB
IRB Approval Date
2020-09-29
IRB Approval Number
IRB20-0834