Appendix K

Home > Statistics & Research > Child Welfare Outcomes 1999: Appendix K

This Appendix provides additional clarification of selected data items in this report. The following items are discussed: Child Maltreatment Victims; Child Fatalities; Race/Ethnicity; Foster Care Counts, Median Length of Stay of Children in Foster Care; Children Waiting to be Adopted; Children Adopted; Recurrence of Maltreatment; Group Homes and Institutions; and Child Welfare and the Juvenile Justice Population.

Child Maltreatment Victims
In the NCANDS, a victim is defined as a child who has been found by the child protective services agency to have been the subject of a substantiated or indicated report of maltreatment. If a child is the subject of more than one substantiated or indicated report, he or she may be counted more than once as a victim.

A “substantiated report ” is one in which the allegation of maltreatment or risk of maltreatment was supported or founded according to State law or State policy. An “indicated report ” is one in which there was reason to suspect maltreatment, but maltreatment was unable to be founded under State law or State policy.

The majority of States report to NCANDS using only the category of “substantiated. ” Those States that use both categories, “substantiated ” and “indicated, ” may have higher rates of child maltreatment victims than those States that use only one category. The counts of victims are based on both substantiated and indicated findings in the following States: Alabama, Alaska, Arizona, Florida, Georgia, Idaho, Kentucky, Montana, South Dakota, and Washington.
(See Child Maltreatment 1999 on the Internet at http://www.acf.hhs.gov/programs/cb for additional information.)

Child Fatalities
The numbers of child fatalities reported by State child welfare agencies are based on information that is available to the State child protective services agency. These numbers are influenced by the specific State and local policies and protocols for investigating child deaths, including suspicious deaths, such as deaths due to drowning or car accidents, which could be either accidental or due to neglect.

Investigations of child deaths are conducted by local law enforcement agencies, coroner’s offices, fatality review boards, and courts. The State statistics on child fatalities due to maltreatment may undercount the actual number of such deaths if the results of these investigations are not provided to the State agency or if suspicious child deaths are not fully investigated.

Race/Ethnicity
Because the NCANDS Summary Data Component collects aggregate data, it has not always been possible to eliminate the duplicated counting of child victims as being of Hispanic ethnicity and of a certain race. Therefore, the total shown for the race/ethnicity of child victims in Context Data: Section B is greater than 100 percent for some States, indicating that Hispanic children are counted both by Hispanic ethnicity and by race. Case-level data from the AFCARS are used in reporting the race and ethnicity of children in foster care and who are adopted. A child’s race/ethnicity is counted as Hispanic if his/her Hispanic Origin is indicated to be “Yes, ” regardless of race, including those children whose race has been identified in AFCARS as “Unable to Determine ” or for whom the race datum is missing. A child is identified by his/her reported race if “Hispanic Origin ” has been coded as “No ” or “Unable to Determine, ” or the Hispanic Origin datum is missing. A child’s race/ethnicity is identified as “Unknown, ” in the following instances: (1) if a child ’s race is indicated as “Unable to Determine ” and Hispanic Origin is “No ” or missing, (2) if a child ’s race is missing and Hispanic Origin is “No ” or “Unable to Determine ”, or (3) if data on both race and Hispanic Origin are missing. The “Not Applicable ” race category includes children for whom more than one race has been designated. The accommodation of reporting more than one race was not in place for fiscal year 1999. In legacy child welfare information systems, data may not be collected for each child on both Hispanic ethnicity and race and therefore the number of children reported as being of Hispanic ethnicity may be an undercount.

Foster Care Counts
In several States there are many more children in foster care on the last day of one fiscal year than on the first day of the next fiscal year. The difference may result from the State’s extraction process. AFCARS uses a computer-generated transaction date (Foster Care #57) for Date of Discharge (Foster Care #56) that occurs within the six-month report period. As a result, even if a child was discharged in Fiscal Year (FY) 1999, the State will report that child’s record with the discharge date in the first 2000 AFCARS report, because the transaction date indicates that the discharge date was not entered into the system until after the FY 1999 period ended.

Unless the State’s AFCARS extraction process uses transaction dates, a child discharged from foster care late in one six-month period may never be reported to AFCARS as discharged from care. When the State’s AFCARS extraction process uses the computer-generated transaction date of all records with a discharge date, then the State’s next AFCARS report will correctly report children discharged in earlier periods, even though these discharges were not entered into the system and reported before. It is important to note that there will always be a slight undercount of the number of children in care on the first day of the report year, because of the way the annual file is constructed from the six-month AFCARS submissions.

Median Length of Stay of Children in Foster Care
The median is the score of the middle case, when the scores for all cases are arranged in order. For example, if data on 125 children are ordered from the shortest length of stay to the longest length of stay, the length of stay of the 63rd child (the child in the middle of the distribution) is the median length of stay.

Children Waiting to be Adopted
The term “waiting children ” is defined for analytical purposes as those children who have a case plan of adoption regardless of whether parents’ rights have been terminated, and those children under 16 years of age whose parents’ rights have been terminated, even if their case goal is “independent living ” rather than adoption.

Children Adopted
Adoption is defined as the establishment of a legal relationship of parent and child between persons who are not so related by birth, with the same mutual rights and obligations that exist between children and their birth parents. The completion of the legal process marks the achievement of adoption. Because adoptions finalized in any fiscal year can be reported to AFCARS without penalties in any subsequent year, the data in this report may reflect some degree of undercount of the total number of children who were adopted during FYs 1998 and 1999.

The data submitted to AFCARS on adoption of children may include children who were adopted without the involvement of the State child welfare system. However, only children who have been designated to have had “state agency involvement ” are reported as adopted children in the Context Data, Section E. Adoptions taking place with “state agency involvement ” include adoptions of (1) children who were placed for adoption by the child welfare agency; (2) special needs children for whom non-recurring expenses were reimbursed and who were adopted in the State, whether or not they were in the public foster care system prior to the adoption; and (3) children for whom an adoption assistance payment or service is being provided based on arrangements made by or through the State agency.

For some States, there are inconsistencies between the number of adoptions reported in Context Data, Section E and the numbers reported for outcome measure 5.1 that assess time to adoption. Most of these inconsistencies are due to fewer children being reported as exiting care to adoption on the AFCARS Foster Care Database than are reported as having finalized adoptions on the AFCARS Adoption Database. One possible reason for this disparity is that adoption is not being used consistently as a “discharge reason ” on State reports to the AFCARS Foster Care Database. Some States in their comment letters suggested that this might be due to staff concerns about confidentiality when a child is discharged to adoption, or to the State’s use of a “discharge code ” that is not AFCARS compliant when an adoptive placement is made and the child is discharged from foster care. Another possible reason for the discrepancy is that this outcome refers only to those children who have been in foster care and have been adopted, while the context data refer to all children who were classified as having been adopted with “State agency involvement. ” However, an analysis conducted on these data indicated that private adoptions occurring with “state agency involvement ” account for only a small percentage of the discrepancy.

The numbers of adoptions reported in Context Data: Section E may differ from those associated with the Adoption Incentive Program because adoptions reported for that program are identified through a different AFCARS data element and must qualify in other ways to be counted toward the award of incentive funds.

Group Homes and Institutions
For measure 7.1, which assesses placements of young children in group homes or institutions, the category of “group home ” is defined as a licensed or approved home providing 24-hour care for children in a small group setting that generally has 7 to 12 children. (Some States may include settings with fewer than seven children as group homes.) “Institution ” is defined as a child care facility operated by a public or private agency and providing 24-hour care and/or treatment for children who require being separated from their own homes and a group living experience. These facilities may include child-care institutions, residential treatment facilities, maternity homes, long-term hospital care, and juvenile justice facilities. In some instances, children may live in a group home in a campus setting, classified as an institution. The category of “other settings ” includes pre-adoptive homes, relative foster family homes, non-relative foster family homes, supervised independent living, and trial home visits.

Several States indicated in their comments that they use short-term emergency shelter placements for young children when there is concern for their safety. Because these shelters meet the AFCARS definitions for group home or institution based on the number of children that can be cared for in the shelter, the States code their emergency shelter placements as either group homes or institutions in their AFCARS files.

Child Welfare and Juvenile Justice Population
Some States include children who are in the State’s juvenile justice system in their child welfare foster care populations. Some of these children were already in foster care when they became involved with the juvenile justice system. Other children became involved with the juvenile justice system and then were placed in an eligible title IVE reimbursable foster care setting. Both of these groups of children are considered children in foster care for whom the State has placement, care, or supervision responsibility, and therefore are reported to AFCARS. The proportion of juvenile justice system children in individual State foster care populations varies extensively.