Increasing Access to Minnesota Child Care Stabilization Base Grants

How to cite

Duru, Maya; Turner, Mary Clair; Permut, Stephanie; Girard, Amanda; Norton, Jessica; Merrick, Weston; Fine, Nolan. 2023. "Increasing Access to Minnesota Child Care Stabilization Base Grants." AEA RCT Registry. AEARCTR-0011435. https://doi.org/10.1257/rct.11435-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Maya Duru* , Mary Clair Turner , Stephanie Permut , Amanda Girard , Jessica Norton , Weston Merrick , Nolan Fine

* Primary investigator

Registry ID
AEARCTR-0011435
DOI
https://doi.org/10.1257/rct.11435-1.0
First registered
Last updated
Status
On going
Trial dates
2023-02-10 – 2024-09-01
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/11435

Abstract

The pandemic highlighted the instability of the child care market and put additional financial burdens on child care providers. The American Rescue Plan Act of 2021 (ARP) allocated approximately $24 billion for child care stabilization grants that Health and Human Services (HHS) Administration for Children and Families (ACF), working with states, territories, and tribes, provides as subgrants to child care providers. Minnesota’s Department of Human Services (DHS) is administering several grant programs to help stabilize the child care industry in Minnesota, including the Child Care Stabilization Base Grant (CCSBG), which is available to all eligible child care providers for monthly grant awards. As part of a portfolio to learn what works to support equitable delivery of ARP, the Office of Evaluation Sciences (OES) at the U.S. General Services Administration collaborated with DHS on a randomized evaluation that seeks to understand the extent to which additional outreach strategies and modalities are effective at increasing take-up of stabilization grants among child care providers.

Intervention

OES randomized child care providers to one of three groups: (1) business-as-usual outreach, (2) a communications-bundle of behaviorally-informed emails and text messages; or (3) the same communications-bundle with an additional proactive phone call.

Experimental design

The study providers will consist of providers who apply five or fewer times in the six months prior to the beginning of the study. Providers who have a common program address (e.g., two programs with the same address) or who share the same contact information (e.g., same email address) will be clustered, and considered to be the same provider.
OES randomized child care providers to one of three groups: (1) business-as-usual outreach, (2) a communications-bundle of behaviorally-informed emails and text messages; or (3) the same communications-bundle with an additional proactive phone call.
Among the sample of provider clusters the probabilities of assignment to the business-as-usual group was 50% as was assignment to the text group (comprising either the text and email communication bundle or the text, email, and phone call bundle). Assignment to the text and email group did not differ based on provider characteristics. Some providers in the text and email group also were randomly assigned to the text, email, and phone call group. This probability differed based on the frequency of their applications to CCSBG prior to the start of the evaluation. The overall probability of assignment to the phone call group over the 5 month implementation period was:
43% of provider clusters that had applied zero times (N=674 provider clusters; 684 License IDs);
14% of provider clusters that had applied one to four times (N=132 provider clusters; 139 License IDs); and
17% of provider clusters that had applied five times (N=88 provider clusters; 89 License IDs).
Additionally, we randomly assigned the order in which phone calls were made. Among provider clusters assigned to the phone call group, each provider was randomly assigned to a phone call batch of 50 phone numbers (i.e., Batch 1 - 18).

Design details

Randomization method
Randomization was done in an office using the statistical software Stata.
Randomization unit
We cluster providers according to whether they share the same contact information, or physical address for a center. Outcomes will be collected at the license ID level, so that there is one observation for each license ID.
Clusters
3,265 provider groups
Observations
3,374 license IDs and 5 months of outcome data, for a total of16,870 expected observations.
Arms
1,636 provider clusters (1,698 license IDs) assigned to business-as-usual 1,629 provider clusters(1,676 license IDs) assigned to the text and email group, including 894 provider clusters (912 license IDs) assigned to the phone call group
Primary outcomes
A monthly dichotomous indicator equal to one if the provider was awarded funding for CCSBG for that license ID; and, zero otherwise.
Secondary outcomes
Whether the provider applied in a given month; The monthly amount of funding received (imputed to 0 if the provider did not apply that month); Whether the provider is eligible for funding in a given month; Whether a site's license ID is not active in a given month (3 months after the intervention for the LATE analysis); 0 otherwise; and A dichotomous indicator for whether the center was active (i.e., a center opened with an active license) in a given month (including 3 months after the end of the intervention).

Cited by

No citation data available for this registration yet.

Keywords