Understanding Adoption Subsidies: An Analysis of AFCARS Data. Variables Used in Analyses


Variables were selected from the AFCARS files to address the goals and hypotheses described above. This section describes the creation and analysis of key variables.

Variables describing time to adoption should be interpreted with caution. These variables represent time to adoption for those children who have exited to adoption during the reporting year, but do not represent the experience of all children who will eventually be adopted. In particular, if the number of adoptions fluctuates from one year to the next (as seems to be the case in some states), these rates will be unstable. These estimates are used as comparisons among states rather than actual estimates of time to adoption, which would ideally be based on analysis of entry cohorts.

Child's age.
Children less than 18 years of age at time of adoption are included in the adoption population. Age at adoption was used for most of these analyses. Children less than 18 years of age as of the end of the reporting period were included in analyses using foster care data. We stratified most of the analysis results by three age groups (0 to 5, 6 to 12, and 13 to 17). These age groups were defined based on developmental stages and their similarities with respect to subsidy needs.
Multiple race designations could apply to each child, beginning with the FY 2000 data. The ethnicity variable was dichotomous: Hispanic and non-Hispanic. After initial analyses shown in Table 3-1, researchers consolidated race and ethnicity categories for simplicity and consistency with other analyses conducted by ACYF. The consolidated categories include white, non-Hispanic; African American, non-Hispanic; Hispanic; and other race/ethnicity. The other category includes non-Hispanic American Indians, Asians, Native Hawaiians, and children with more than one race designation. Researchers used race/ethnicity to describe the children adopted from out-of-home care and to stratify analysis of time from TPR to adoption and subsidy amounts received.
Special needs criteria.
Special needs status with respect to adoption subsidies defines characteristics that would make adoption difficult if a subsidy were not available. Categories broadly defined by the federal government, include race, age, sibling group, medical condition, and other. AFCARS allows only one factor to be assigned to an individual child for reporting purposes. Each state has the latitude to set their own criteria for classifying children as special needs children and to specify a priority for classifying children if they meet multiple criteria. Some caseworkers may report the special needs criteria that is the easiest to document. For these reasons and because states may specify more specific criteria for each special needs category using the other category, special needs data are not entirely comparable across states. Further specification of medical conditions include mental retardation, visually or hearing impaired, physically disabled, emotionally disturbed, and other diagnosed condition. The proportion of adopted children meeting special needs criteria is presented overall and by state and the relationship between time from TPR to adoption and special needs status was examined.
Adoptive family characteristics.
The family structure of the adoptive family (i.e., married couple, unmarried couple, single female, and single male) and the preadoptive parent-child relationship (i.e., foster parent, stepparent, other relative, and nonrelative) is presented. The adoptive mother and father's age and the proportion of children who are adopted by parents of a different race and ethnicity are displayed. These analyses are presented stratified by age.
Time from TPR to adoption.
If available, both the mother's and father's TPR data were reported in the AFCARS adoption file as well as the date the adoption was legalized. For these analyses, we designated the most recent of the two TPR dates to calculate the time from TPR to adoption. These analyses were stratified by child's age at TPR, race/ethnicity, and special needs status.
Time from most recent entry to adoption.
Because the adoption file does not include dates of entry into out-of-home care, we used data from the foster care file to calculate the most recent time in continuous care prior to adoption. This measure was used to describe the population, stratified by age and state.
Proportion of children receiving adoption subsidies.
The proportion of children adopted from public child welfare agencies who receive a monthly subsidy of any kind (i.e., federal or state) is presented at the national level and by state. Tables show the proportion of children receiving federal plus state subsidies, state-only subsidies, and no subsidies. The proportion of children who receive monthly payments and those with deferred agreements are presented. Although AFCARS does not include a field to explicitly indicate a deferred subsidy agreement is in place, we considered cases that were reported to be receiving a subsidy and the amount of subsidy was either $0 or $1 to have deferred payments for these analyses. Because only a small number of cases met this definition, it is likely that these analyses undercount the number of actual deferred agreements since some states may not distinguish between cases with deferred agreements and those without subsidies in AFCARS. In addition, the proportion of adoption subsidies with federal matching funds were compared across states. To examine trends in the federal plus state subsidy rate, data for the three most recent years were analyzed for each state and the percent change from year to year and from 1999 to 2001 is presented.
Proportion of children receiving federal foster care payments.
The proportion of children in nonrelative foster care placement (either current or most recent placement) receiving federal matching funds is calculated from the foster care data and compared to the equivalent adoption variable by state. Children in relative care were excluded from these analyses due to state variability in practices regarding foster care payments to relative caregivers.
Adoption subsidy amounts.
The median subsidy amount is presented for the past 3 years and by state. The state-level data were stratified by age, since foster care payments tend to increase with children's age in most states (likely due to their greater material and service needs) and we hypothesize a correlation between foster care payments and adoption subsidies. To examine trends in the average federal matching monthly subsidy amount, data for the three most recent years were analyzed for each state and the percent change from year to year and from 1999 to 2001 was calculated. Subsidy amounts greater than $10,000 per month were considered invalid data due to the likelihood of errors in the states' data for this field and were treated as missing values.
Foster care payment amounts.
Because the adoption data file does not include the amount the child received in foster care payments, we used the foster care data file to compare foster care payments to adoption subsidy amounts. For this analysis, we compared the subsidy amounts of all adopted children who were adopted by their foster family or by a nonrelative with the monthly payments for children in nonrelative foster care or pre-adoptive homes. Limiting the analysis to children who had been, or were currently placed, in nonrelative foster care excludes those children who were in a group care facility and might be receiving unusually high stipends and eliminates the likelihood that state variability in practices regarding relative caregivers will bias the results. Thus the analysis is reduced to the two groups of children who are most similar. These analyses are also presented for each state. Foster care payments greater than $10,000 were considered invalid data and were treated as missing values due to the likelihood that they were reported in error.
Foster care adoption rate.
Adoption rates are defined as the percentage of eligible children in out-of-home care who were adopted, derived from the foster care data file. Eligible children are defined as those who had a goal of adoption and/or had parental rights terminated, excluding those aged 16 and older with a goal of emancipation.
Deferred subsidy payments.
The proportion of adopted children with deferred subsidy payments are presented for each state. Deferred payments were defined as payments of $0 or $1 for cases where a subsidy was indicated.

View full report


"report.pdf" (pdf, 1.41Mb)

Note: Documents in PDF format require the Adobe Acrobat Reader®. If you experience problems with PDF documents, please download the latest version of the Reader®