Urban Institute estimates based on the 2009 Current Population Survey. This includes children living with only one biological parent as well as children 14 and under living with no parents.
 Children up to age 21 may also be eligible for child support but are excluded here because we will use a Medicaid eligibility simulation designed for those18 and under.
 Urban Institute estimates based on the 2009 Current Population Survey
 Aron, Laudon. 2002. “Health Care Coverage Among Child-Support Eligible Children.” http://aspe.hhs.gov/HSP/CSE-health-ben02/index.htm
 CHIP coverage has been expanded in many states since 1999 (G. Kenney and J. Yee. 2007. “SCHIP At a Crossroads: Experiences To Date and Challenges Ahead.” Health Affairs 26(2): 356-369.
 G. Kenney, A. Cook and L.Dubay “Progress Enrolling Children in Medicaid/CHIP: Who is Left and What are the Prospects for Covering More Children?” The Urban Institute, December 2009.
 We assign Medicaid/CHIP to the top of the hierarchy to estimate of the total number of children with Medicaid/CHIP coverage, regardless of whether they were also reported to have another type of coverage. This likely allows for more precise adjustments to coverage estimates since the exclusion of children reported to have both Medicaid/CHIP and other coverage from Medicaid/CHIP estimates could potentially overstate the extent of public coverage underreporting on the CPS. In addition, by assigning Medicaid/CHIP to the top of the hierarchy, we are able to assess the full extent to which children are reported to be enrolled in these programs, regardless of their other coverage experiences.
 There are fewer than 500,000 children 15 and over who are classified as having other federal coverage or direct purchase coverage who also reported coverage from outside the household. This represents approximately 1.8 percent of CS eligible children.
 Because many states use the same names for their Medicaid and CHIP programs and because many families are confused about the specific type of public coverage their child has, it is not possible to reliably distinguish between Medicaid and CHIP coverage on the CPS.(C.T. Nelson, and R.J. Mills. “The Characteristics of Persons Reporting State Children’s Health Insurance Program Coverage in the March 2001 Current Population Survey.” U.S. Bureau of the Census, August 2002.)
 The CPS health insurance questions do not specifically identify the type of coverage that is available from outside of the household.
 The undercount adjustment partially adjusts the CPS to administrative estimates of Medicaid and CHIP enrollment. For more information, see Dubay, L., J. Holahan, and A. Cook. “The Uninsured and the Affordability of Health Insurance Coverage.” Health Affairs 26(1): w22-w30. 2007. This adjustment has the effect of reducing the number of uninsured children by 0.8 million (from 8.1 million to 7.3 million) and increasing the number of children with Medicaid/CHIP coverage by 2.5 million (from 23.4 million to 25.9 million.).
 A long standing debate exists regarding whether insurance estimates from the CPS represent people who responded by providing their coverage at the time of the survey or responded about their health insurance coverage over the course of the year (as intended) but with recall error because of the long reference period. The Census Bureau has commented on this issue and stated that CPS estimates are more closely in line with point in time estimates of the uninsured. (DeNavas-Walt, C, BD Proctor, and JC Smith, U.S. Census Bureau, Current Population Reports, P60-236. “Income, Poverty, and Health Insurance Coverage in the United States: 2008.” Washington, DC: Government Printing Office, 2009.)
 CPS estimates of children with both Medicaid/CHIP and private coverage far exceed estimates from the National Health Interview Survey, a point-in-time survey. In 2008, only 1 percent of children report both Medicaid/CHIP and private coverage on the NHIS, while 5 percent do so on the CPS.
 The definition of child support eligibility excludes children who reside with a single parent who has been widowed and children residing with adoptive parent(s). Because children may be adopted by a single parent, we cannot identify which children living with one adoptive parent are child-support eligible.
 For those children living with no parents, we are unable to exclude cases where both parents are deceased.
 U.S. Census Bureau. “Poverty Thresholds for 2008 by Size of Family and Number of Related Children Under 18 years.” http://www.census.gov/hhes/www/poverty/threshld/thresh08.html
 Health insurance units include the member of a nuclear family, including the family head, spouse, and own children under 19 years of age, or own full-time student children 19-22 years of age.
 Census family groupings differ from HIUs because they can include family members who are not part of the nuclear family unit (e.g. grandparents).
 The model takes into account childcare expense, work expense and earnings disregards in determining eligibility, but does not take into account child support disregards.
 To account for the possibility that some foreign born individuals are unauthorized immigrants and therefore not eligible for public health insurance coverage, the model takes into account immigrant status. Immigrant status is predicted based on a simulation model of immigrant status derived from the March 2004 CPS. March 2004 CPS estimates of immigrant status were developed by Passell, and estimates derived from the two sample estimation technique are consistent with those produced using the March 2008 CPS (J. Passel, and D. Cohen. “A Portrait of Unauthorized Immigrants in the United States.” Washington, DC: Pew Hispanic Center, April 2009).
 Child support eligible children in two-parent families live with one biological parent and one step-parent.
 Child support eligible children with family income less than 200 percent of the FPL who are ineligible for Medicaid/CHIP do not qualify for these programs either because they are legal non-citizens who do not meet their state’s citizenship status requirements for eligibility or because they do not meet immigration status requirements. In contrast, the majority of ineligible children with income over 200 percent of the FPL appear to be ineligible on the basis of income.
 Medicaid/CHIP eligible children above 300 percent of the FPL are in states with high CHIP eligibility thresholds or other special programs for children.
 Adjusting for the March CPS overestimate of CS eligible children results in 2.6 million uninsured children
 An estimated 13 percent of child support eligible children appear to be uninsured based on coverage estimates that have not been adjusted for the underreporting of public coverage (See Appendix Table 2).
 Adjusting for the March CPS overestimate of CS eligible children results in an estimate of 2.0 million
 Adjusting for the March CPS overestimate of CS eligible children results in estimates of 36 percent for the proportion of uninsured children that are CS eligible and 28 percent of uninsured children that are both Medicaid/CHIP and CS eligible children
 Adjusting for the March CPS overestimate of CS eligible children results in an estimate 41 percent of Medicaid/CHIP eligible uninsured children that are CS eligible
 According to Urban Institute tabulations, there are 40,123 unweighted households on the 2008 CPS-CSS and 97,502 unweighted households in the 2008 ASEC. Approximately one-quarter of households interviewed in March are in the outgoing rotation group, and so are not in sample for the April CPS and are not included in the CPS-CSS. By restricting the universe to those households in sample for both the March and April interviews, the CPS-CSS appears to also exclude the additional "CHIP" and "Hispanic" samples that substantially increase the ASEC sample size (http://www.census.gov/apsd/techdoc/cps/cpsmar09.pdf).
 The March 2009 file was available in September 2009, while the March/April 2008 file was not available until April 2010.
 Some of 2.5 million excluded children are excluded because the custodial parent indicates that there is a parent living elsewhere, but they have never received or attempted to receive child support. Such children may still be eligible however and our eligibility definition would include these children.
 These coverage estimates do not include the adjustment to reflect the underreporting of public coverage and exclude children living with neither of their parents to be more consistent with the CPS-CSS.