By adding new information obtained directly from the foster care case records held in the Archive, the census-based living arrangement categories described previously in the analysis of living arrangement patterns in fifty states can be extended in several ways. Most important for this work, the counts of children in kinship foster care for April 1990 can be subtracted from the count of children living with relatives that was enumerated in the 1990 census to derive a new count of informal kinship cases. Thus, the "related child" group described in the previous section can now be divided into two subgroups: a formal kinship group -- those children observed in kinship foster care through the Archive data; and the residual informal kinship group -- those children living in relative settings who are not observed in kinship foster care. In a similar fashion, we have broken the "unrelated child" category into two parts -- non-relative foster care and other unrelated children.(13) These operations can be performed for any geographic or substantive subgroup for which the census and Archive both track data. For this analysis, we have tabulated informal versus formal kinship for four states, the counties within those states, and for the 0-5 and 6-17 year age groups within each of these geographic areas.
Table 3.1 presents these modified living arrangement data for each the four study states. These numbers are fundamentally the same as the state numbers in Table 2.2a, except that the detail within the relative and non-relative categories is expanded here using the new information extracted from the foster care data systems in each state. Across these four states, which together include over 16 million children (or over one-quarter of the U.S. child population), we observe that almost 400,000 children lived in kinship care settings in 1990, with the preponderance (331,521) in informal kinship care. A substantial, but much smaller, number (61,023) lived in formal kinship foster care placements. The same basic relationship describes the children living in unrelated care situations, where formal non-relative foster care represents only a modest share of the total children who were housed and cared for by unrelated individuals. Thus, while child caretaking through non-parental living arrangements is a relatively uncommon phenomenon (over 95 percent of all children in these states live with one or more parent), the number of children living in each of the four "atypical" care arrangements described here is still substantial.
Living arrangement patterns can be compared across these four states with percentage distributions. Panel B of Table 3.1 shows that the allocation of children across the non-traditional living arrangements varies in the four states being examined. Missouri had very low levels of formal kinship care in 1990 ( 0.5 children per thousand), while New York had levels over ten times as high ( 5.0 children per thousand). In each of the four states, informal kinship care was a more probable care arrangement than formal kinship care. Missouri, Illinois, and New York showed similar levels of informal kinship care (between 17 and 19 children per thousand), while California had a slightly higher level higher level (23 per thousand). The basic state patterns observed for kinship care are once again reflected in unrelated care arrangements, with New York having the higher prevalence of formal child welfare (foster care) arrangements with non-relatives and California having the highest level of non-relative placements arranged outside of the formal foster care system.