Communicating Program Eligibility: A Supplemental Security Income (SSI) Field Experiment
Last registered on February 12, 2018

Pre-Trial

Trial Information
General Information
Title
Communicating Program Eligibility: A Supplemental Security Income (SSI) Field Experiment
RCT ID
AEARCTR-0002423
Initial registration date
September 14, 2017
Last updated
February 12, 2018 1:33 PM EST
Location(s)

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

Request Information
Primary Investigator
Affiliation
Reed College
Other Primary Investigator(s)
PI Affiliation
Office of Evaluation Sciences
PI Affiliation
Social Security Administration
Additional Trial Information
Status
In development
Start date
2017-09-15
End date
2018-06-15
Secondary IDs
Abstract
Survey data from the Health and Retirement Study (HRS) suggest that less than 60 percent of individuals age 65 and over who are eligible for SSI receive SSI and administrative data suggest that take-up may be substantially lower than this. The economic literature has identified at least three main barriers to SSI take-up among individuals age 65 and over. First, individuals may not be aware that they are eligible for SSI (SSA 1976, Warlick 1982), which may be a particularly important barrier for individuals for whom being age 65 and over partly determines eligibility. Second, the expected magnitude of benefits affects take-up (McGarry 2000, McGarry and Schoeni 2015), with individuals with lower expectations about benefits less likely to participate in SSI. Third, potential SSI participants may view the application process as confusing and burdensome (Warlick 1982, McGarry 1996, McGarry and Schoeni 2015).

We have designed four letters to test these hypotheses using a randomized controlled field experiment with nearly 4 million individuals. In this study, individuals age 65-80 and who are likely eligible for SSI will be assigned to receive one of these four letter types through US mail or to a control condition (i.e. business as usual). We will track SSI application filing, SSI application allowed (i.e. approved), and SSI payments using administrative data from the Social Security Administration.
External Link(s)
Registration Citation
Citation
Hemmeter, Jeffrey, Elana Safran and Nicholas Wilson. 2018. "Communicating Program Eligibility: A Supplemental Security Income (SSI) Field Experiment." AEA RCT Registry. February 12. https://www.socialscienceregistry.org/trials/2423/history/25733
Experimental Details
Interventions
Intervention(s)
Four behaviorally-informed letters will be tested against a control condition (i.e. no letter, which is the current standard procedure):

(1) the Basic Letter,
(2) the Maximum Payment Letter,
(3) the Simplifying Application Process Letter, and
(4) a Combined Letter that combines the maximum payment and the simplifying application letters.

All letters include the basic information listed on Letter (1), allowing us to measure the incremental effect of the information on a more detailed letter.
Intervention Start Date
2017-09-15
Intervention End Date
2017-12-31
Primary Outcomes
Primary Outcomes (end points)
The primary outcomes of interest are: (i) SSI application filed, and (ii) SSI application allowed. We will also examine SSI payments.
Primary Outcomes (explanation)
Secondary Outcomes
Secondary Outcomes (end points)
Secondary Outcomes (explanation)
Experimental Design
Experimental Design
The control condition is “no letter”, which is the current standard program procedure. The treatment conditions are:

(1) the Basic Letter,
(2) the Maximum Payment Letter,
(3) the Simplifying Application Process Letter, and
(4) a Combined Letter that combines the maximum payment and the simplifying application letters.

Each treatment condition has a sample size of 100,000 letters, yielding a total letter sample size of 400,000 letters.
Experimental Design Details
Not available
Randomization Method
Randomization done using Stata.
Randomization Unit
The unit of randomization is the individual.
Was the treatment clustered?
No
Experiment Characteristics
Sample size: planned number of clusters
The treatment is not clustered.
Sample size: planned number of observations
Approximately 4 million individuals.
Sample size (or number of clusters) by treatment arms
400,000 individuals will be divided equally into 4 treatment arms. Remaining individuals will be assigned to control arm.
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
Power calculations (conservatively) using 1 million individuals total of whom 400,000 total are assigned to the four treatment study arms yield MDEs as follows. 0.6 percentage points for take-up in any single treatment arm compared to any other single treatment arm. 0.4 percentage points for take-up in any single treatment arm compared to control arm. 0.3 percentage points for take-up in all letter arms (pooled) compared to control arm.
IRB
INSTITUTIONAL REVIEW BOARDS (IRBs)
IRB Name
IRB Approval Date
IRB Approval Number