By Allen W. Harden of the Chapin Hall Center for Children at the University of Chicago and Rebecca L. Clark and Karen Maguire of The Urban Institute for the Office of the Assistant Secretary for Planning and Evaluation, U.S. Department of Health and Human Services, June 20, 1997.
This report presents the results of work pursued by analysts at two separate research institutions in a collaboration designed to describe the population ofAmerican children living in kinship care arrangements.
The Task Order was to examine existing national data sources in order to describe the characteristics of children in kinship living arrangements, and to identify recent trends in the pattern of kinship caregiving. Particular importance was attached to developing information that could support comparison between formal kinship care arrangements (i.e. care provided by relatives as foster care under auspices of the state) and informal kinship arrangements (all other caregiving provided by relatives in the absence of a parent).
Kinship foster care has attracted much attention in recent years within the context of the child welfare system. The extensive placement of children with relatives has created a new, rapidly growing, and poorly understood segment of the child welfare caseload that has great impact on the size and nature of the foster care population in many states. Children in formal kinship placements can be viewed as a subgroup of a broader category of family-based alternatives to parental care -- the population of all children living in kinship care settings across the country. Most American children who live in kinship care arrangements are not foster children. We cannot yet determine whether most current kinship foster care placements are "formalizations" of kinship arrangements that would likely exist without agency intervention, or whether these are mostly new arrangements created as a result of recent child welfare practices. But it is clear that children in informal kinship settings are potentially of crucial importance for the child welfare system -- as a reference group, as a potential "feeder" population, and as an alternate model of caregiving.
By virtue of the similarity between formal and informal kinship arrangements, any policy actions directed towards one of these groups is likely to affect the other in a parallel or reactive manner, whether or not this is intended by those who frame these actions. Even though our understanding of the recent interdependence between these two kinship subgroups is weak, the importance of anticipating their future interrelationship becomes increasingly apparent -- especially as our questions move from the strict realm of child welfare policy into the broader arena of family supports and welfare reform.
This report presents the results of four separate, and relatively independent, research tasks, each approaching these questions with a different set of information tools. Taken as a whole, they provide us with a greatly improved picture of kinship care in the United States, and provide an enriched context for discussing these issues. The first task was produced by the Urban Institute, the remaining three by the Chapin Hall Center for Children at the University of Chicago. A brief description of each task and a summary of substantive findings from each follows.
I. National Patterns and Trends in Kinship Care:
Section I describes the population of children in kinship care settings in the United States, the characteristics of these children and their caretakers, and trends that have been observed since 1983. These descriptions are based on information drawn from 12 years of the Current Population Survey (CPS), a large and ongoing national sample of the full United States population. At the national level, the CPS provides the richest and most reliable information available about children's living arrangements and households -- including identification of kinship care relationships. Information is collected about the children, their relative caretakers, and the families which they share. The following are among the key findings reported in this section.
About Children in Kinship Care:
- In 1994, approximately 2.15 million children, or just over 3 percent of all children in the United States, were estimated to live in the care of relatives without a parent present.
- Nationally, the prevalence of kinship care probably increased between 1983 and 1993, and it certainly did not decrease. There is no evidence of any increase in kinship care among the white (non-Hispanic) children in recent years; all observed growth in kinship care has been among white Hispanic and non-white sectors of the population.
- Non-Hispanic white children are substantially less likely to live without their parents in the care of relatives than are the children of any other racial/ethnic group. African American children are most likely to live in kinship care settings, at levels four to five times as great as those for white non-Hispanic children. The gap between African Americans and the other ethnic groups widened throughout the 12 year period examined.
- Kinship care has been more prevalent in the South, for children living outside of Metropolitan areas, and for older children, although the size of the differences due to each of these factors has diminished gradually over the 12 years studied.
About Kinship Caregivers:
- Roughly two-thirds of kinship caregivers are the child's grandparent . About half of the kinship caregivers are currently married, while over 85 percent of the single kinship caregivers are female.
- The kinship caregiver population is much older than the parent caregiver population. Although over 95 percent of the parents who live with their own children are below the age of 50, over one-half of all kinship caregivers are 50 years of age or greater.
- Compared to parents who live with their own children, kinship caregivers tend more often to be currently unmarried, to be less-educated, to be unemployed or out of the labor force, to live in poverty, and to receive benefits through government social welfare programs.
The portrait of kinship care that emerges from the CPS is of a population of children that live in arrangements with strained resources of many types. This population is disproportionally composed of minority children being cared for by relatives that, as a group, show fewer advantages than own-parent caregivers.
II. Living Arrangement Patterns by State: 1990
Section II describes the living arrangement patterns for all children state by state. This analysis is based on data made available from the 1990 Census of Population. The census does not provide as much substantive detail as the Current Population Survey, but the estimates it provides are reliable for much smaller geographic areas.
- The national pattern of child living arrangements in 1990 showed most American children living with at least one of their own parents. Over 70 percent lived with two parents, 20 percent with their mother only, and 4 percent with their father only. Just over 2 percent of all children lived with relatives (parent absent), and just over 2 percent in the care of unrelated persons.
- Although this fundamental pattern persists across all states, substantial variation in the distributions is seen between states. The percentage of children living with two parents varies from under 62 percent to over 83 percent. For kinship care, state percentages varied from under 1 percent of all children to well over 3 percent. Much of this variation follows regional lines, with the southern states consistently showing the highest levels of kinship care arrangements.
- In every state, older children (6-17) are more likely to live in kinship care settings than are younger children (0-5).
- In general, kinship care levels across states tend to be positively associated with levels of mother-only care, and weakly or negatively associated with father-only and unrelated care. The levels of kinship care and mother-only care also each vary directly with the total percentage of children not living with two parents, while father-only and unrelated arrangements do not.
A tentative argument is developed that higher levels of mother-only care and relative care appear to be direct products of higher levels of social disruption and family disorganization, because they consistently vary strongly and inversely with the proportion of children living within a traditional two-parent family structure.
III. Formal and Informal Kinship Care Patterns: Four States
Section III introduces data developed directly from administrative foster care records in four states: California, Illinois, New York, and Missouri. Kinship foster care counts obtained from these child welfare records are used to split the census-based counts of children living with relatives into the separate categories of formal and informal kinship care. This information becomes available in the form of aggregate counts for the four states and certain sub-state places.
Findings include the following:
- Informal kinship care is far more common than formal kinship care. In the four states combined, only 15.5 percent of all kinship children were in a formal foster care placement.
- Levels of informal kinship care are rather similar across each of these four states, while the levels of formal kinship foster care vary dramatically.
- Younger children in kinship care are more likely to be in foster care than are older kinship care children. Formal kinship levels were 58 percent higher for 0-5 year olds than for 6-17 year olds, while informal kinship levels were over twice as high for 6-17 year olds as for 0-5 year olds.
Within each state, the analysis compares the "primary urban place" (i.e. Los Angeles County; Chicago City; St. Louis City; and New York City) to the "balance," or remainder, of the state.
- In two states, New York and Missouri, formal kinship foster care appears almost exclusively in the primary urban place, and is virtually absent across the balance of the state. In California and Illinois, formal kinship is still concentrated in the primary urban place and a few other counties.
- Informal kinship care is also consistently higher in the primary urban places than in the balance of each state, although it is distributed far more evenly than formal kinship care.
- In larger cities, where formal kinship care is most common, there appears to be an inverse relationship between the levels of formal and informal kinship care. This might suggest that the children in the two types of kinship care are drawn from the same pool of children, and that the observed differences in formal versus informal care levels between cities are mostly due to different agency practices involving the use of formal kinship care.
Looking only at formal kinship foster care:
- In each of the four states, African American children are more likely to experience kinship foster care than are children from other racial or ethnic groups. Overall, African American children are about eight times as likely as all others to be in formal kinship placements. The racial effect holds across regions and across age groups.
- This racial effect and the "primary urban place" effect become compounded because of the high representation of African American children in the primary urban places in each state. The interaction can be huge: for example, African American children in New York City are one hundred times more likely to be in a kinship foster care placement than are non-African American children in the remainder of New York State.
- In California and Illinois, the race appears to be a stronger predictor of kinship foster care levels than primary urban place. In New York, the "urban place" factor appears to be a stronger predictor of kinship foster care than race.
IV. Formal and Informal Kinship Care Dynamics in Illinois To gain at least one "window" for comparing characteristics of children in formal and informal kinship care settings, information was accessed from the Illinois Child Multiservice Database that is being developed at Chapin Hall. Individual-level records were examined for all recent (1990-95) child AFDC grant recipients and all foster children in the state. The population of AFDC children living in kinship care arrangements is treated as a biased sample of all Illinois children in informal kinship care -- sort of a "semi-formal" kinship group.
Looking at characteristics of these groups:
- Compared to the AFDC/Relative group, the formal kinship care group is younger, overrepresents African Americans, and is disproportionally comprised of children from Cook County (Chicago). No gender differences are apparent. Both of these groups are younger and more likely to live in Cook County than the remainder of Illinois's informal kinship care population.
- Compared to AFDC/Parent cases, the AFDC/Relative cases are more likely to have two or more adults present and the caretaker is more likely to be currently married. But, the relative caretakers are significantly older, and four out of five are the child's grandparent.
- The Illinois formal kinship care group more than tripled (from 8,000 to 27,000) between 1990 and 1995, while the AFDC/Relative group remained constant at 16,000 children.
- Within each racial category, the prevalence of AFDC/Relative cases is similar for children from Cook County and children from the remainder of Illinois, while the prevalence of formal foster care is more than twice as high in Cook County than for the balance of the state. For both types of care, the prevalence of kinship care for African American kinship exceeds that of "all others" combined by ten times or more.
It was possible to track movements of individual children between these statuses across the 5-year time period (via annual snapshots).
- Most children "stay" in their current status from year to year. Over 70 percent of AFDC/Relative children and 80 percent of formal kinship children can be expected to remain in their current status after a given one-year period.
- Viewed as a transition from their current status, AFDC/Relative children are about twice as likely to move into formal kinship care as are AFDC/Parent children, although the likelihood of such a change was small (less than 2 percent per year) for both groups.
- Viewed as sources of transition into formal kinship care, a new entrant to kinship foster care is ten times more likely to have moved from an AFDC/Parent setting than from an AFDC/Relative setting. The apparent anomaly between this and the previous finding is explained by the fact that the AFDC/Parent population is more than twenty-five times as large as the AFDC/Relative population.
- Even though less than 1 percent of AFDC/Parent children are expected to move into kinship foster care in the course of one year, over one-half of all new children in kinship foster care moved into this status from AFDC/Parent settings.
- Children who move between the different AFDC and kinship settings tend to be younger, while children who "stay put" or who leave the system entirely tend to be older.
V. Summary, Observations, and Potential Next Steps A final section summarizes these findings, describes some of the data limitations that acted as obstacles in the production of this report., discusses some conceptual issues in the study of kinship care, and proposes certain paths for future data gathering and analysis. Some of the issues discussed include:
- The difficulty of clearly defining family relationships, as opposed to just the relation of members to the household head, in much data collected through surveys. Presence or absence of a child's parent is often not identifiable for complex households.
- Kinship care arrangements are relatively uncommon, so only censuses, very large population surveys, or specially targeted surveys can enumerate a sufficient number of kinship care cases to support a meaningful comparative analysis.
- Having access to individual-level data is extremely important in order to allow observed relationships to be controlled for such key variables as race/ethnicity and poverty status.
Some possible next steps include:
- Maintaining a baseline of information on kinship care by continuing to monitor the annual CPS results and by supporting more detail in the analyses created from them.
- Extending the aggregate reporting from census data to provide more detailed information on the living arrangements of children, particularly to classify reported data by race/ethnicity.
- Extending the work in formal kinship care to more than four states, possibly by accessing the new AFCARS data being reported directly to HHS by the states.
- Continuing new efforts to create linked and integrated information resources describing the full range of children's contacts with social services and other public systems. This information is potentially rich for describing process, child needs, and outcome indicators.
The discussion concludes by arguing that kinship care arrangements should be studied within a framework that emphasizes their role in ongoing child and family processes. It is the context in which the need for kinship care occurs, and not the fact that relatives are providing care, that carries the information that has the most ongoing relevance to social policy formulation.
A much more refined body of information would be needed to support an effort to examine these processes, observe causes, track movements, classify kinship care cases, compare subgroups, and evaluate trends and changes. Information of this quality could only be gathered through a survey that is longitudinal and comprehensive in scope.