Methodology & Data Sources

Data Sources

All data comes directly from U.S. federal agencies and is in the public domain:

  • ACF AFCARS (FY2022) - The Adoption and Foster Care Analysis and Reporting System is administered by the Administration for Children and Families (ACF) within the U.S. Department of Health and Human Services. AFCARS is the federally mandated reporting system through which all 50 states, the District of Columbia, and Puerto Rico submit case-level data on every child in foster care and every child who has been adopted with public agency involvement. Data includes foster care entries, exits, placement types, time in care, permanency outcomes, and demographic characteristics of children and foster families.
  • HHS NCANDS (2022) - The National Child Abuse and Neglect Data System is a voluntary state-level reporting system published annually by HHS. NCANDS collects data on screened-in referrals, substantiated maltreatment victims by type (physical abuse, neglect, sexual abuse, emotional maltreatment), and child fatalities due to maltreatment. While participation is voluntary, all 50 states and D.C. submit data in most years.
  • U.S. Census Bureau ACS - American Community Survey five-year estimates for child population (under 18) by state, used to calculate per-capita foster care rates that allow meaningful comparisons between states of different sizes.

Processing Pipeline

  1. AFCARS FY2022 annual data files are downloaded from the ACF data portal, containing case-level records for every child in state-supervised foster care during the fiscal year.
  2. NCANDS 2022 child file is downloaded from HHS, providing state-level aggregate counts of maltreatment reports, investigations, and substantiated victims.
  3. Census ACS five-year population estimates for children under 18 are downloaded by state to serve as denominators for per-capita rate calculations.
  4. Per-capita rates are computed by dividing foster care counts and maltreatment counts by each state's child population, enabling fair cross-state comparison regardless of state size.
  5. Eight performance metrics are aggregated per state: reunification rate, adoption rate, aging-out rate (percentage of exits aging out at 18+ without permanency), median length of stay, re-entry rate (percentage of children who re-enter foster care within 12 months of exit), placement stability (percentage of children with two or fewer placements), maltreatment recurrence (percentage with a repeat maltreatment report within 6 months), and timely permanency (percentage achieving permanency within 12 months of entry).
  6. A through F letter grades are assigned based on the reunification rate, the primary permanency goal established by the federal Adoption and Safe Families Act (ASFA) - using a transparent grade scale documented in our grading guide.
  7. State rankings are computed for each of the eight metrics, and composite profiles are assembled showing each state's performance across all dimensions.
  8. All data is compiled into a structured dataset that powers state profiles, metric comparison pages, and national overview summaries.

Data Vintage and Update Frequency

The current dataset reflects FY2022 (October 1, 2021 through September 30, 2022) for AFCARS, and calendar year 2022 for NCANDS. ACF publishes AFCARS reports annually, typically 18–24 months after the end of the fiscal year due to the time required for states to submit data, for ACF to validate submissions, and for the national dataset to be compiled. NCANDS data follows a similar publication timeline. PlainFoster updates its database when new federal data releases become available from either source.

Data Neutrality and Accuracy Commitment

PlainFoster presents government data factually without editorializing. Foster care outcomes are shaped by complex structural factors, funding levels, caseworker caseload sizes, rural service access gaps, judicial practices, and state policy decisions. Grades are educational tools designed to help users understand system-level performance patterns, not judgments on any community, demographic group, or individual family situation. All statistics are reproduced exactly as reported by federal agencies. When state-submitted data has known quality issues flagged by ACF, PlainFoster notes this where applicable rather than excluding the state from analysis.

Limitations

  • AFCARS covers only children in state-supervised foster care; privately arranged placements, kinship care arrangements not supervised by the state, and tribal foster care systems operating under separate federal authority are excluded.
  • NCANDS participation is voluntary, while most states participate, some have incomplete maltreatment data in certain years, and reporting definitions for neglect and emotional maltreatment vary by state statute.
  • Reunification rates vary by state policy, court practices, and legal timelines, not solely by outcome quality. Some states prioritize reunification more aggressively while others may lean toward adoption or guardianship as permanency goals.
  • Reunification, adoption, and aging-out rates shown on state pages do not sum to 100% of exits -- AFCARS tracks additional exit categories not broken out on this site, including guardianship, transfer to another agency, and other case closures.
  • Data reflects FY2022 conditions. Current state performance may differ due to policy changes, funding shifts, or other developments since the reporting period.
  • PlainFoster is not affiliated with ACF, HHS, the Census Bureau, or any government agency.

Reuse and Citation

PlainFoster's state profile pages are freely citable and linkable, please link back to the specific state page you drew a figure from rather than a screenshot, since AFCARS figures are periodically revised. The underlying AFCARS and NCANDS datasets are U.S. government works and are in the public domain; PlainFoster's own narrative framing, grading methodology, and page design are not licensed for wholesale republication or scraping.

Contact

Questions about our methodology? Contact us.