We have documented differences in the profiles of the support systems created to care for children who end up living in homes where they are cared for by someone other than one of their parents. Although some of these differences are related to the age of the child and the type (formal versus informal) of care provided, the clearest contrasts observed so far have involved the type of place in which the child lives. The primary cities in each state show higher levels of both formal and informal kinship care than other places in these states. We can only hypothesize about the factors underlying this fundamental difference. The largest cities include substantial concentrations of persons in poverty, disproportionate numbers of minorities, and, when considering formal kinship care, huge child welfare agencies and court systems straining in their capacity to handle growing and complex caseloads. Our largest cities have shown many symptoms of social dislocation; problems such as unemployment, drug use, crime and violence, teenaged parenthood, etc. occur in much greater magnitude, if not more frequently, in large urban places.
The national analysis of children in kinship care (Section I) presented evidence of clear racial differences in the likelihood that children will live with relatives other than a parent. (See Table 3.4). Overall, the combined CPS panels for 1989-91 showed 6.2 percent of African American children living in kinship situations, as opposed to 2.4 percent of Hispanic children, 1.2 percent of white children, and 2.1 percent of the children of Asian, Native American, and other backgrounds. However, the data as reported in Section I, do not provide a cross-classification of race and region. Using information from published 1994 CPS results, we have computed the following breakdown for the likelihood of living in all kinship care (formal and informal) for American children ages 0-14 by age, race/ethnicity, metro/non-metro status, and age category.(15)
Within each age-racial/ethnic group, the nonmetropolitan percentages for kinship care either equal the metropolitan percentages or exceed them by up to one-third. Observed age differences are also relatively small, and seem to have an effect only among all Hispanics and metropolitan African Americans. The racial effect that clearly persists, both across age groups and across metropolitan/nonmetropolitan places, with African American levels averaging about five times higher than white levels and about 2.5 times higher than Hispanic levels.
Census tabulations of children living in kinship care unfortunately do not include any racial/ethnic categorization, so the influence of this factor cannot be directly introduced for consideration in comparing formal and informal kinship care for four states. However, ethnic classifications are part of the descriptive information in the records of each child tracked by the Multistate Foster Care Data Archive, and we can identify the race/ethnicity of each child in a formal kinship foster care placement. So, examination of the extremely important racial/ethnic distribution can currently be made only for the formal component of the kinship care population. It is unfortunate that the informal/formal comparison cannot be analyzed fully along ethnic lines. In the remainder of this analysis, we will simplify comparisons by discussing race only in terms of African American and "all other races."
Table 3.5 presents several types of information that help us examine levels of kinship foster care participation by ethnicity/race and by region of the state. The observed levels of kinship foster care for race-region subgroups, measured by prevalence rates per 1,000 children, are presented in the bold box at the center of Table 3.5, and are portrayed graphically in Figure 3.5. It is once again clear that kinship foster care is most common among African American children and in the largest cities. The highest race-region specific rate observed is in Los Angeles, where 28.5 of each 1,000 African American children are in kinship foster placements. The formal kinship prevalence rate among African American children in New York City is the next largest at 21.2 per thousand. In each state, the lowest rates are those for "others" (not African American) in the "balance" of the state. Observed rates for this group (the largest in population size) are as small as 0.2 children per 1,000 in "Upstate" New York. This provides a striking contrast -- the kinship foster care prevalence level for non-African Americans in Upstate New York is less than one one-hundredth of the level observed for African American children in New York City.
In California and Illinois (and probably Missouri), the race effect is the more dominant of these two factors. Within each regional category, an African American child is eight to ten times more likely to be in a kinship foster placement than any other child. In Chicago, where almost one-half the total child population is African American, African American children comprise over 90 percent of the kinship foster care population. As a result, almost two out of every three kinship foster cases in Illinois are African American children from Chicago. It is significant that in both California and Illinois, African American children living in areas outside of the primary city (i.e. in the "balance of the state") are still more likely to live in kinship foster care settings than are the "other" (non African American) children living within the primary city.
New York presents a different pattern. Although both African American ethnicity and a New York City location are each related to higher levels of kinship foster care,the "city" effect appears more dominant. In New York, the city rates for both racial groups are relatively high, while the upstate rates for both racial groups are low. Non African American ("other") children living in New York City are four times more likely to be placed in kinship foster care (8.4 per thousand) than African American children from "upstate" (2.2 per thousand) -- a relation unique among the four states examined.
Because the actual number of children in kinship foster care can be viewed as the net result of a process that applies these race-region specific rates to the race-region specific populations, we must consider the influence of population composition on the final net result. These "marginals," fully independent of any kinship foster care patterns, provide the context within which the differential tendencies suggested by the prevalence rates can operate. The percentage of children in each state that live in the defined "primary urban place" varies from a low in Missouri where St. Louis contains only 7.6 percent of all children in the state, to a high in New York City where almost 40 percent of the state's children live. Racial composition also varies widely between and within states. About one of every eight children in Los Angeles is African American, compared to about one in three for New York City, one in two in Chicago, and almost two in three in St. Louis. African American children are far less likely to live in the remaining portions of any of these states, with the "balance of state" percentages varying from a low of 6.9 percent in California to a high of 9.6 percent in Missouri.
(12) Much more information about kinship dynamics is available from Archive data. Kinship arrangements are likely to be established fairly early in a child's foster care experience. Apart from the earliest short-term temporary custody placements, most children in kinship foster care placements tend to be "pure" kinship cases, and most children in non-relative placements tend to be "pure" non-relative cases. There is not a high level of movement in between the two statuses. Also, kinship cases tend to have a much longer duration than other foster care. See Goerge, Wulczyn, Harden (1993, 1994) for more detailed Archive reporting.
(13) This "other unrelated children" group, which we use here to describe all unrelated children who do not live in foster care, can also be further subdivided. For all children 0-17, it is possible to differentiate "non-relative family foster care," "child welfare placements in congregate care facilities," "other children in institutional settings," "other children in group quarters," and "other unrelated children living in households."
(14) The "primary urban areas" have been defined as follows for the purpose of this work: Los Angeles County in California, the city of Chicago in Illinois, Saint Louis City in Missouri and New York City in New York. These delineations are somewhat arbitrary and they could be quite arguable. However, the concentration of kinship foster care within these areas is so dramatic that issues of precision are rendered moot.
(15) These numbers are based on computations from document P20-484,Table 3. It should be noted that the "Metropolitan" category is much more broadly defined than our construct of "primary urban place", and includes much smaller cities and suburban counties. Also, children 15-17 are not included in this tabulation.