'Disappearing Day Care:' Analysis of Missouri child care deserts and pandemic relief grants

This repository contains data and findings for an ongoing collaboration between MuckRock and The Missouri Indepedent on Missouri's "child care deserts," areas where there are more than three children ages 5 and under for every licensed child care slot or, in some cases, no licensed slots at all.

The findings in the article Missouri child care deserts now include nearly half of kids 5 and under are summarized in the file analysis/findings.qmd.

Data

Supply of Missouri child care programs

• We determind the supply of child care programs, data/raw/ChildCareAware_nov_2019_mar_2023, by the licensed capacity of programs by ZIP code, through data provided by the advocacy group Child Care Aware. That group has pulled data twice a year from Missouri’s active facility list since 2019.

Demand for Missouri child care

• We determind the demand for child care by the number of children ages 5 and under by ZIP Code Tabulation Area (ZCTA), using American Community Survey 5-Year Data (2017-2021). This methodology is similiar to the one used by Child Care Aware and in child care research more broadly.
• We pulled ACS population estimates using the tidycensus package in etl/2_get-zcta-2017-2021-acs and stored in processed/pop_zcta_5acs.

Pandemic funding and grant applications

• Through open-records requests, MuckRock and The Missouri Indepedent recieved data on every application through seven grants in two federal COVID-19 relief programs — the Biden administration’s American Rescue Plan (ARPA) and the Coronavirus Response and Relief Supplemental Appropriations Act of 2021 (CRRSA).
• The raw data for each application can be found in data/raw with "ARPA" and "CRRSA" and the grant type in the file name, as in ARPA Paycheck Protection Round 3 with Paid.
• Data/processed/dese_funds_update is an updated list as of April 14, 2023, of the several types of funding for child care through Missouri Department of Elementary and Secondary Education and how much has been spent of each funding source and grant.

Methodology

• We crosswalked the supply of Missouri child care programs by ZIP code to supply by ZCTA using the popular UDS crosswalk in etl/3_zip-zcta-crosswalk-ChildCareAware. This is necessary to both map the data and to join the data to ACS population estimates, which are in ZCTA format, not ZIP.
• We joined ZCTA-level supply to demand in etl/4_append_all_pop_to_ChildCareAware.
• We cleaned and crosswalked the grant applications with the address of each facility to ZCTAs in etl/5_clean-relief-data. Several addresses had incorrect ZIP codes. We manually corrected these by searching the address on Google Maps, after which each application's unique ID was marked in the code and with a comment that the ZIP code that was manually corrected.
• The file analysis/findings.qmd brings the above sources of data together to map ZCTA areas that qualify as child care deserts and assess whether funding from the grants we tracked went to those areas.

Questions / Feedback

Contact Dillon Bergin at dillon@muckrock.com