Evaluation of the Family Unification Program

Last registered on September 16, 2021

View Trial History

Pre-Trial

Trial Information

General Information

Title

Evaluation of the Family Unification Program

RCT ID

AEARCTR-0004670

Initial registration date

January 21, 2020

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published

January 21, 2020, 1:48 PM EST

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Last updated

September 16, 2021, 11:49 PM EDT

Last updated is the most recent time when changes to the trial's registration were published.

Locations

Country

United States of America

Region

Bucks County, Pennsylvania

Country

United States of America

Region

Phoenix, Arizona

Country

United States of America

Region

Orange County, California

Country

United States of America

Region

Seattle-King County, Washington

Primary Investigator

Name

Michael Pergamit

Affiliation

Urban Institute

Contact Primary Investigator

Other Primary Investigator(s)

PI Name

Mark Courtney

PI Affiliation

University of Chicago

Contact Investigator

Additional Trial Information

Status

On going

Start date

2019-04-03

End date

2023-09-29

Keywords

Other

Additional Keywords

child welfare; housing

JEL code(s)

I38

Secondary IDs

HHSP233201500064I_HHSP23337014T

Prior work

This trial does not extend or rely on any prior RCTs.

Abstract

The Family Unification Program (FUP), which is funded by the U.S. Department of Housing and Urban Development (HUD), provides child welfare–involved families with permanent Housing Choice Vouchers. The program provides vouchers to families for whom the lack of adequate housing is a primary factor in (a) the imminent placement of the family’s child, or children, in out-of-home care or (b) the delay in the discharge of the child, or children, to the family from out-of-home-care. The program aims to prevent children’s placement in out-of-home care, promote family reunification for children placed in out-of-home care, and decrease new reports of abuse and neglect.

The goal of the evaluation is to determine whether FUP increased family preservation and reunification for families receiving vouchers and associated services, document how the program is implemented across the sites, and identify how the variations in implementation across sites might explain impact differences across sites. This evaluation has two main components, an implementation study and an impact study. The impact study examines the effect that the program had on outcomes, using quantitative data collection and analysis. The impact study will employ a randomized controlled trial (RCT) in 5-10 sites across the United States. The evaluation will randomly assign families to be referred either to receive a FUP voucher or to receive services as usual. The study will follow all randomized families in the sites in the evaluation. The implementation study examines how the program is executed, the context of FUP within each site's child welfare and housing policies and practices, and the successes and challenges faced by the child welfare and housing agencies.

External Link(s)

Registration Citation

Citation

Courtney, Mark and Michael Pergamit. 2021. "Evaluation of the Family Unification Program ." AEA RCT Registry. September 16. https://doi.org/10.1257/rct.4670-2.0

Sponsors & Partners

Interventions

Intervention(s)

The primary intervention provided by the Family Unification Program is a voucher providing a permanent housing subsidy. The family generally pays 30 percent of their income toward rent and utilities, and the voucher makes up the difference—up to a locally defined rent cap. The voucher is permanent as long as the family continues to meet the eligibility requirements and complies with other requirements (e.g., recertification of income eligibility).

Intervention Start Date

2019-04-03

Intervention End Date

2023-09-29

Primary Outcomes

Primary Outcomes (end points)

1. Child removals into out-of-home care
2. Child reunifications with family from out-of-home care
• Time to reunification

Primary Outcomes (explanation)

Child removals will be measured as the probability that a child who is in the home at the time of randomization has been removed at all at 1 year and 2 years post randomization. Child reunification will be measured as the probability that a child who is out of the home at the time of randomization is reunited with their parents within 1 year and within 2 years of randomization. We will measure time to reunification as the number of days from randomization to reunification. Primary outcomes will be measured from child welfare administrative data.

Secondary Outcomes

Secondary Outcomes (end points)

- Child welfare case closures
- Time to child welfare case closure
- Reduced emergency homeless shelter stays
- Re-entry into child welfare system
- New child maltreatment reports

Secondary Outcomes (explanation)

We will measure case closures as whether the child’s case was closed at 1 year and 2 years post randomization for the full sample as well as whether the child’s case had closed within 1 year and 2 years of randomization. We will measure time to case closure as the number of days from randomization to case closure. While we believe that removal and investigation data will be available in all sites, case-related data, such as dates cases are opened and closed, varies widely in definition, quality and availability. For case-related outcomes, we may have to analyze the results for a subset of sites that have high-quality case-related data. We will measure homeless shelter stays as the number of stays in the two years after randomization.

Experimental Design

The evaluation will employ a household level randomized controlled trial (RCT) in 5-10 sites. The evaluation will select sites where the target population for the FUP program includes many more families eligible for the program than can be accommodated by the program’s limited number of vouchers. Within each site, the evaluation will allocate the limited program vouchers by randomly assigning families to be referred either to receive a FUP voucher or services as usual.

Experimental Design Details

The evaluation will employ a randomized controlled trial (RCT) in 5-10 sites. The evaluation will select sites where the target population for the FUP program includes many more families eligible for the program than can be accommodated by the program’s limited number of vouchers. Within each site, the evaluation will allocate the limited program vouchers randomly assigning families to be referred either to receive a FUP voucher or services as usual.

The evaluation has selected sites based on having a sufficient number of vouchers for families, having excess demand for the vouchers, and being willing and able to participate in a randomized controlled trial. Other considerations included whether the site was not too geographically dispersed and whether referrals could be controlled effectively when a single PCWA was the partner of multiple PHAs. Chicago and Cook County, Illinois, were also excluded because the one previous RCT had been implemented there.

Randomization Method

The specific referral and randomization strategy across the sites is for caseworkers to identify cases eligible either through a screening tool or caseworker judgment. Eligible cases are then referred to a central person (either at Urban or at the public child welfare agency), who randomizes the family into either treatment or control. All sites use a customized online randomization tool, designed by the Urban Institute. The randomization tool collects the family’s identification number, whether they are preservation or reunification, and their referral form and determines whether a family is to be referred to FUP or services as usual. Three out of four sites are using a one-to-one randomization ratio, with one family randomized to treatment for every control family. One site is using a two-to-one ratio, with two families randomized to treatment for every control family.

Randomization Unit

Randomization takes place at the household level.

Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters

N/A

Sample size: planned number of observations

We are still in the process of recruiting new sites to participate in the evaluation. With seven total sites (three additional sites), our goal is 1034 households: 536 treatment and 498 control (see Table 1B).

Sample size (or number of clusters) by treatment arms

498 control households
536 treatment households

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

To determine the appropriate sample size, we estimated Minimum Detectable Effect (MDE) sizes for both Intent-to-Treat (ITT) and Treatment-on-the-Treated (TOT) estimates. The effect size scales the effect of the intervention on an outcome by the control group’s standard deviation on that outcome. Tables 4A and 4B in the analysis plan present Minimum Detectable Effect sizes based on our current expected sample size with four sites as well as estimates based on additional sites. To be able to detect meaningful effect sizes, that is, effect sizes of 0.20 or greater in the TOT, we need a sample of 1,034 families (536 treatment and 498 control). This would require additional sites contributing a total of 150 vouchers. This sample size estimates pool outcomes for families who enter the program intact, whose goal is to prevent removals, and families who enter the program with one or more children in foster care, whose goal is to be reunified. We expect to estimate impacts for each of these two groups separately. Assuming about half of the families will be preservation cases and half reunification cases, the selected sample size would allow us to estimate an ITT effect size in the treated group between 0.22 and 0.28 and a TOT effect size between 0.27 and 0.35. If the current four sites do not provide roughly equal size groups of preservation and reunification cases, we will consider whether to add a criterion to our future site selection that would help even this out.

Supporting Documents and Materials

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

Request Information

IRB