Census Bureau Issues Working Paper Entitled 'Coverage of Children in American Community Survey Based on California Birth Records'

WASHINGTON, Sept. 28 -- The U.S. Census Bureau issued the following working paper (No. CES-23-46) entitled "Coverage of Children in the American Community Survey Based on California Birth Records."

The paper was written by Gloria G. Aldana.

Here are excerpts:

* * *

Introduction

The Census Bureau recognizes the well-documented undercount of children in Census Bureau surveys (U.S. Census Bureau 2021a). Undercounting of children can occur when an entire housing unit or household was missed in the survey or census, part of the household was not reported, or only the child was not reported. The undercount of children in the Decennial Census can reduce funding for programs such as Head Start and Children's Health Insurance Programs (CHIP) (Reamer 2019; Murphy and Danielson 2018). To reduce the undercount of children, the Census Bureau has focused research on evaluating the coverage of children in surveys and understanding the causes of undercounting (U.S. Census Bureau 2021a). Unreported children tend to have characteristics that are distinct from those of reported children. Child reporting rates can vary by age, with children under 5 years old having the lowest coverage rates (U.S. Census Bureau 2016). Additionally, coverage rates for children vary by race, with Non-Hispanic White children having the highest coverage rates (O'Hare et al. 2019). More complex household structures are associated with lower coverage rates as well (O'Hare et al. 2019; Fernandez et al. 2018; Jensen et al. 2018).

The Census Bureau's American Community Survey (ACS) collects demographic and economic information on individuals and households, such as age, income, and household structure. The ACS provides survey-based estimates of children drawn from a sample of the U.S. population. However, survey responses may not match administrative records, such as birth records. Birth records are assumed to provide a virtually complete account of all births, along with child-parent relationships and demographic characteristics. California has both the largest share of the population of U.S. children and a known high undercount rate for young children (U.S. Census Bureau 2022; O'Hare et al. 2019; O'Hare 2014b). This paper uses California as a case study to examine differences between reported versus unreported children in a Census Bureau survey based on state birth records.

* * *

Conclusion

This paper takes advantage of administrative records from California, which has both a large population of children and a high undercount for young children, to characterize undercounted children. Patterns of child reporting in this analysis generally follow those of prior research, offering further support for the characteristics known to be associated with the undercounting of children, such as younger child age, race and Hispanic ethnicity, English proficiency, and household complexity. Moreover, this analysis provides support for using state birth records to supplement demographic analyses of undercounting.

Given the noticeable variations in child reporting by poverty status and mother's race and ethnicity, future research should take advantage of other administrative records housed at the Census Bureau, such as 1040 and W2 data from the Internal Revenue Service (IRS). Use of IRS income data could further refine the analysis by using IRS reported income rather than survey-reported income and by examining child reporting rates for children who were not claimed as dependents by their parents. Additionally, this analysis can be expanded by linking birth records to the Decennial Census, which is intended to have a complete count of all children in the United States.

* * *

The paper is posted at: https://www2.census.gov/library/working-papers/2023/adrm/ces/CES-WP-23-46.pdf