Matching 5.0 Alpha (Limited release): Enhanced private pay estimation

A major enhancement with Matching 5.0 Alpha is the improved estimation of the number and distribution (by age, income, and household employment status) of children served by private pay child care in different income and household employment segments based on survey data from the U.S. Census Bureau’s Survey for Income and Program Participation (SIPP). SIPP is a nationally representative longitudinal survey that provides comprehensive information on the dynamics of income, employment, household composition, and government program participation. SIPP is also a leading source of data on economic well-being, family dynamics, education, wealth, health insurance, child care, and food security. The survey interviews individuals for several years and provides monthly data about changes in household and family composition and economic circumstances over time. With Matching 5.0 Alpha, CUSP estimates of children who are served by private pay child care within each income-employment segment will be informed by actual SIPP survey data, collected longitudinally over several years and using weighting provided by the Census Bureau to ensure that this data is correctly generalized from the SIPP sample to the population. This integration will therefore generate estimates that are more verifiable and reliable than prior CUSP versions.

  • Integration with the Census Bureau’s 2021 SIPP data (which includes survey panels from 2018, 2020, and 2021) to determine the distribution, by income and employment segment, of the percentage share and number of children served by private pay child care.
  • Separate modeling of private pay child care for children 0 through 5 years and school age children 6 through 12 years to account for differences in child care needs and utilization among children below school age and children of school age.
  • Estimation takes into account cross-sectional and longitudinal variation in the number of children served by private pay child care by pooling survey data across U.S. regions and over multiple years.
  • Consistent use of sampling weights provided by the U.S. Census Bureau ensures that the SIPP sample data is appropriately generalized to the full population and across all geographic units of analysis.
  • The use of mutually exclusive, collectively exhaustive (MECE) age-household income-household employment segments in distributing children served by private pay ensures that the SIPP data is comprehensively and correctly integrated with the CUSP model.


  • Enhanced accuracy of matching between children served by providers and funding sources, including private pay child care, based on child and household characteristics.
  • Integration with empirical survey data from the Census Bureau makes estimates of children served by private pay (which represents the largest share of children served by child care, notably for children under 5 years) more verifiable and reliable than prior CUSP versions.
  • Eliminates need for simplifying analytical assumptions about utilization of private pay child care in different income and employment segments, thus making CUSP outputs more robust and valid.