|
|---|
|
|
||||
|---|---|---|---|---|
|
|
|
|||
| ACF Home | Services | Working with ACF | Policy/Planning | About ACF | ACF News | HHS Home | ||||
Questions?
|
Privacy
|
Site Index
|
Contact Us
|
Download Reader™
|
Print
 �
|
|---|
This chapter describes additional research activities related to child maltreatment including those using data from the National Child Abuse and Neglect Data System (NCANDS). The U.S. Department of Health and Human Services (HHS), other Federal and State agencies, and other organizations have sponsored these studies. Ideas and suggestions for future research also are included.
Child Welfare Outcomes
The Child Welfare Outcomes 2003: Annual Report is the sixth annual report in the series, which is published by the Children's Bureau. The report contains information by State on key child maltreatment indicators, including the two measures of reducing recurrence of child abuse and neglect, and reducing the incidence of child abuse and neglect in foster care. These measures, as reported in the 2003 annual report, are as follows:
These key measures, as well as other contextual data on child victims, were based on data submitted to NCANDS. The report also contains data about foster care, adoption, and information derived from the Adoption and Foster Care Analysis and Reporting System (AFCARS) as well as from the Child and Family Services Reviews. This report is available on the Children's Bureau Web site at http://www.acf.hhs.gov/programs/cb/stats_research/index.htm#cw.
For further information about Child Welfare Outcomes 2003: Annual Report, contact:
Sharon Newburg-Rinn, Ph.D.
Social Science Research Analyst
Children's Bureau/ACYF/ACF
U.S. Department of Health and Human Services
Portals Building
1250 Maryland Avenue, Room 8378
Washington, DC 20024
202-205-0749
snewburg-rinn@acf.hhs.gov
America's Children: Key National Indicators of Well-Being
This newly restructured annual report is a compendium of indicators drawn from the most reliable official statistics that illustrates both the promises and the difficulties young people confront. The report is a product of collaboration among 22 Federal agencies that form the Interagency Forum on Child and Family Statistics. The 2007 report presents two new sections— Physical Environment and Safety and Health Care—and nine new indicators. One such indicator
is on child maltreatment and used data from NCANDS.
The Interagency Forum on Child and Family Statistics publishes a more detailed report during odd-numbered years and alternates with a condensed version, America's Children in Brief: Key National Indicators of Well-Being, which highlights selected indicators during even-numbered years. The report can be found online at http://childstats.gov.
For further information about America's Children in Brief: Key National Indicators of Well-Being, contact:
Shara Godiwalla
Forum Director
Centers for Disease Control and Prevention
National Center for Health Statistics
3311 Toledo Rd., Room 6114
Hyattsville, Maryland 20782
(301) 458-4256
buh7@cdc.gov
Statistical Abstract of the United States
The Statistical Abstract, prepared by the U.S. Census Bureau, contains a collection of statistics on social and economic conditions in the United States. Selected international data also are included. For many years, two tables using NCANDS data have been published. One table reports the characteristics of child victims by maltreatment, sex, age, and race or ethnicity. The second table reports the number of investigations, the number of children who were subjects of investigations, and the number of victims by State.
The 2007 edition of the Statistical Abstract was published and is available on CD-ROM. An online version is available at http://www.census.gov/compendia/statab/.
For further information about the Statistical Abstract, contact:
Richard P. Kersey
Statistical Abstract
U.S. Census Bureau
301-763-4428
richard.patrick.kersey@census.gov
Outcomes for Children With Allegations of Neglect Who Receive Alternative Response and Traditional Investigations: Findings From NCANDS
This analysis focused on the question of whether children in alternative response systems are kept as safe as children who receive traditional investigations. This study used 2004 and 2005 NCANDS data to examine patterns of reentry into the child protective services (CPS) system among children who receive assessments and those who receive traditional investigations following allegations of neglect. Neglect was chosen as the focus of this study because it is often associated with conditions related to poverty and caregiver behavior and is, therefore, likely to be more chronic in nature. The relatively high rates of repeat victimization seen in children with allegations of neglect justify neglect as an important focus for intervention and prevention of subsequent maltreatment.
Two forms of event history analyses were conducted—trajectory pattern analysis and proportional hazards analysis. The trajectory analysis demonstrated little difference in the rereporting rate of children who received an assessment following allegations of neglect, compared with children who received traditional investigations.
The proportional hazards analysis revealed that children in the assessment track had a somewhat decreased risk of rereporting when compared with children receiving investigations. The risk of rereporting was generally greater for assessed children who were placed in foster care than for investigated children who were placed in foster care.
For additional information on this analysis, contact:
Mary Jo Ortiz
Walter R. McDonald & Associates, Inc.
2720 Gateway Oaks Drive, Suite 250
Sacramento, CA 95833
916–239–4020
mjortiz@wrma.com
National Survey of Child and Adolescent Well-Being
The National Survey of Child and Adolescent Well-Being (NSCAW) is a nationally representative, longitudinal survey that focuses on the well-being of more than 6,200 children—in two samples—who have encountered the child welfare system.
In NSCAW I, direct interviews and assessments were conducted with the children, their caregivers, caseworkers, and teachers, at baseline and again at 18 months and 36 months after a CPS investigation. A 12-month postbaseline followup with caregivers and caseworkers focused on services received during the year after the investigation. More than 80 percent of the children and families interviewed at baseline participated in the 36-month followup interviews. A fourth followup of children was conducted during 2005–2007.
A second longitudinal sample is beginning in 2008. The NSCAW II design and protocol are very similar to the prior study. Data will be collected from 5,700 children, current caregivers, caseworkers, and teachers sampled from the NSCAW I-selected counties using similar measures. Drawing a new sample of children from the same locations will allow researchers to better gauge the effect of changes in policies, practices, and external constraints like budget resources. NSCAW II data will also include administrative data like that provided by the States for NCANDS and AFCARS, to obtain more complete data about rereports, service receipt, and placement history.
The NSCAW data sets represent an important resource for researchers interested in child maltreatment, child welfare, child development, and services to high-risk children and families. Information is available on children's health; development; social, emotional, and cognitive functioning; and both children's and caregivers' service needs and service utilization. Contextual information is provided about the children's household characteristics, as well as the child welfare service system.
NSCAW I data collection is scheduled to be completed in December 2007, and the NSCAW II baseline will begin in March 2008. An 18-month followup is scheduled to begin in September 2009. Study reports and research briefs and more information about NSCAW methods and measures are available at http://www.acf.hhs.gov/programs/opre/abuse_neglect/nscaw/index.html.
The data from NSCAW are available to researchers through licensing agreements with the National Data Archive on Child Abuse and Neglect (NDACAN) at Cornell University. For more information on accessing the NSCAW data sets, please see http://www.ndacan.cornell.edu.
For additional information about the National Survey of Child and Adolescent Well-Being, contact:
Mary Bruce Webb, Ph.D.
Office of Planning, Research and Evaluation/ACF
U.S. Department of Health and Human Services
370 L'Enfant Promenade SW
Washington, DC 20447
202–205–8628
mbwebb@acf.hhs.gov
Longitudinal Analysis of Repeated Child Abuse Reporting and Victimization: Multistate Analysis of Associated Factors
Most children who are subjects of maltreatment reports to CPS are involved just once with CPS; however, some children experience repeated investigations and victimizations. The present study examines individual, maltreatment, and service-related factors associated with maltreatment rereporting and substantiated rereporting in a multi-State context. Case-level NCANDS data with a sample of 505,621 children were analyzed.
Within 24 months, 22 percent of children were rereported, and 7 percent were rereported with substantiation. Children who were younger, white or of mixed race, who had disabilities, and whose caregivers abused alcohol, were more likely to be rereported and substantiated than specific reference groups. Service provision, including foster care placement, was associated with increased likelihood of these subsequent events. These findings suggest that reentry into CPS is a complex interaction of risks to children and systemic factors tied to the intervention they receive.
For additional information about the Longitudinal Analysis of Repeated Child Abuse Reporting and Victimization: Multistate Analysis of Associated Factors contact:
John D. Fluke, Ph.D.
Director
Research Center on Child Protection
American Humane Association
63 Inverness Drive East
Englewood, CO 80112
(303) 925–9416
johnf@americanhumane.org
Fourth National Incidence Study (NIS-4)
HHS is conducting the Fourth National Incidence Study of Child Abuse and Neglect (NIS–4). NIS–4 will measure the incidence and prevalence of child maltreatment by a wide array of demographic characteristics. Like its predecessors, NIS–4 is a Congressionally mandated study. It was mandated by the Keeping Children and Families Safe Act of 2003 (P.L. 108–36). NIS–4 aims to estimate the current national incidence, severity, and demographic distribution of child maltreatment based on standardized research definitions and to assess changes since the previous NIS data were collected. DHHS is conducting NIS–4 through a contract with Westat, a national social science research firm that also conducted all three previous NIS cycles. Assisting Westat with the study planning and CPS recruitment and data analysis is Walter R. McDonald & Associates, Inc. (WRMA).
Earlier research has shown that many more children are abused and neglected in a community than are observed at any single agency. To develop a comprehensive picture of the extent of child abuse and neglect, NIS–4 is pulling together data from a number of agency sources in each study county. The NIS estimates begin with data from the local CPS agency concerning the reports they received and accepted for investigation during the study reference period. Building on this foundation, the NIS estimates also incorporate data on abused and neglected children who were seen by professionals in a number of other community agencies, including the county public health, public housing, juvenile probation departments, and the sheriff or State police. Data were also gathered from scientifically selected samples of other agencies, including voluntary social service and mental health agencies, municipal police departments, schools, hospitals, daycare centers, and shelters for runaway youth and battered women. Designated professionals in the selected community agencies served as study "sentinels" by staying on the lookout for children who were abused or neglected during the study period and by providing descriptive information on the cases they encountered.
Data collection occurred in two phases in a nationally representative sample of 122 counties. These counties were selected using scientific sampling procedures that ensured the necessary mix of geographic regions, urban and rural areas, and other major community characteristics.
More information about the study and its progress is available at http://www.nis4.org. For additional information about the Fourth National Incidence Study, contact:
Maria Woolverton
Office of Planning, Research and Evaluation
Administration for Children and Families
U.S. Department of Health and Human Services
370 L'Enfant Promenade SW
Washington, DC 20447
202–205–4039
maria.woolverton@acf.hhs.gov
The National Resource Center for Child Welfare Data and Technology
The National Resource Center for Child Welfare Data and Technology (NRC-CWDT) is a service of the Children's Bureau that provides a broad range of technical assistance to State and Tribal child welfare agencies and the courts about data and systems issues to improve outcomes for children and families.
The Center helps States, Tribes, and courts improve the quality of data collected, build the capacity to use the information for decisionmaking in daily practice, and develop or improve case management and data collection systems, including Statewide Automated Child Welfare Information Systems (SACWIS). The NRC-CWDT provides technical assistance to States to help improve the quality of data reported to the Federal government in the Adoption and Foster Care Analysis and Reporting System (AFCARS) and the NCANDS. The Center also provides technical assistance during the Child and Family Services Reviews process and on other Federal, State, and local legislative requirements, policies, and initiatives. NRC-CWDT coordinates the data conference sponsored by the Children's Bureau, usually held on an annual basis. Additional dissemination of information and promising practices can be found at www.nrccwdt.org.
For further information about the NRC-CWDT, contact:
Lynda Arnold
Director
NRC-CWDT
2345 Crystal Drive
Suite 250
Arlington, VA 22202
877–672–4829
larnold@cwla.org
The Family Violence and Trauma Research Project
The Family Violence and Trauma Project (FVTP) provides support to the Army Community and Family Support Center Headquarters, Department of the Army Family, Advocacy Committee; the Family Advocacy Research Subcommitttee; Family Advocacy Program Managers; Chiefs of Social Work Services; and Army social workers in the form of briefings, papers, staff studies and a quarterly newsletter on the scientific and medical aspects of child and spouse abuse.
FVTP completed the only two empirical research studies of Family Violence and Deployment. These studies documented the importance of predeployment family violence as a risk factor for postdeployment family violence and the possible role of the "honeymoon phase" of troops returning home.
The FVTP writes and publishes Joining Forces Joining Families, a quarterly newsletter that brings important research to the field. This newsletter is widely distributed to U.S. military organizations and other government and academic institutions with an interest in family advocacy research. The newsletter and additional information on the Center for the Study of Traumatic Stress can be found at the Center's Web site http://www.centerforthestudyoftraumaticstress.org/research.family.shtml.
For further information about the Family Violence and Trauma Research Project, contact:
James E. McCarroll, Ph.D.
Department of Psychiatry
Center for the Study of Traumatic Stress
University School of Medicine
4301 Jones Bridge Road
Bethesda, MD 20814
301–319–8003
Community-Based Grants for the Prevention of Child Abuse and Neglect (CBCAP)
The CBCAP program provides funding to States for the purposes of:
It is estimated that CBCAP programs serve nearly 200,000 families and more than half a million children every year. This number does not include the hundreds of thousands of State residents who benefit from the child abuse prevention public awareness and education activities carried out by lead agencies every year.
CBCAP Conceptual Framework, 2006
This figure displays the conceptual framework of the CBCAP program. The first column discusses the term inputs and outcomes. The next two columns discuss activities and outputs broken down into two categories—direct and indirect. The next column discusses short-term and intermediate outcomes, and the final column discusses the long-term outcomes. Long-term outcomes are broken out by safety, permanency, well-being, and continues improvement categories.
For further information regarding the CBCAP program, contact:
Melissa Brodowski, M.S.W./M.P.H.
Office on Child Abuse and Neglect
Children's Bureau, ACYF, ACF, HHS
1250 Maryland Ave., SW
8th Floor #8127
Washington, DC 20024
202–205–2629
melissa.brodowski@acf.hhs.gov
Researchers interested in using the NCANDS data can apply to the National Data Archive on Child Abuse and Neglect for access to various data files. A description of the National Data Archive is provided below, as well as some suggestions of topics of potential interest for future research. Although far from comprehensive, these topics are of interest to the field.
National Data Archive on Child Abuse and Neglect
Housed in the Family Life Development Center at Cornell University, the National Data Archive on Child Abuse and Neglect (NDACAN) has been established by the Children's Bureau to encourage scholars to use existing child maltreatment data sources in their research. NDACAN acquires data sets from various national data collection efforts and from individual researchers, prepares the data and documentation for secondary analysis, and disseminates the data sets to researchers who have been licensed to use the data.
The Archive seeks to operate as more than a repository of data by providing resources and technical assistance that contribute to the field. In addition to assisting individual researchers as they work with the data, NDACAN also provides many opportunities for scholarly exchange. For example, NDACAN maintains an active electronic mailing list for discussing a range of research issues. NDACAN also is well known for its annual Summer Research Institute. The Institute brings together a group of researchers who are working on projects using Archive data. During the week, participants consult with experts and attend colloquia designed to further progress on their projects. Group computing sessions provide ample opportunity for participants to conduct their analyses and to work together to resolve questions. Through these and other activities, NDACAN serves as a valuable resource to the research community. Information regarding the Archive, its services, and data holdings can be found on the Internet at http://www.ndacan.cornell.edu.
The Archive serves as the official repository of NCANDS data, providing access to both the State-level and case-level data components. Key NCANDS indicators have been available at the State-level in the Summary Data Component (SDC) and the Combined Aggregate File (CAF) data files, which are available to researchers for every year since 1990.
Beginning with the 2004 data year, the Archive—in collaboration with the Children's Bureau, the NCANDS Technical Team, and NCANDS State representatives—adopted a new data release plan for the case-level Child File that strikes a balance between protecting the confidentiality of the data and preserving the utility of the Child File as a research data set. By increasing the contractual responsibilities of researchers wishing to use the Child File data and instituting additional oversight of licensees, the Archive was able to implement a targeted set of confidentiality modifications to the data, focused primarily on elements involving race and Hispanic ethnicity, geography, and dates. This new approach enables the Archive to release comprehensive case-level NCANDS data sets. Data from all States that submitted data using the Child File are available through the Archive starting with data submission year 2004.
For more information about access to NDACAN, researchers may contact:
John Eckenrode, Ph.D.
Director
National Data Archive on Child Abuse and Neglect
Family Life Development Center—Beebe Hall
Cornell University
Ithaca, NY 14853
607–255–7799
jje1@cornell.edu
Each year, states submit their NCANDS Child File data to the Children's Bureau. The resulting file is a rectangular, or flat, data file consisting of rows and columns. There are 146 columns—one for each data element. There are as many rows as there are report-child pairs, because each different child on a CPS report generates a new record in the file, like this:
| Record Number | Report Identifier | Child Identifier | Element 1 | Element 2 | Element 3... |
|---|---|---|---|---|---|
| 1 | Report 1 | Child 1 | value | value | value |
| 2 | Report 1 | Child 2 | value | value | value |
| 3 | Report 2 | Child 1 | value | value | value |
| 4 | Report 3 | Child 1 | value | value | value |
| 5 | Report 3 | Child 2 | value | value | value |
The seeming simplicity of this data file structure conceals quite a bit of complexity. The complexity, and the resulting difficulty in analyzing data elements, stems from the fact that not all data elements contain information about the same entity. The entity defined by a row is a Report-Child. Many of the data elements contain information about this entity. The element "CdEmotnl," for example captures whether this child was recorded as being clinically emotionally disturbed at the time of this report. However, many of the data elements are not about this child on this report, but are about some other entity. The elements "ReportID," "Report Date," "Report Source," and so forth, are not about the report-child, but about the report itself. The information would simply be repeated for each child on the report. To count the number of reports, one must account for multiple report IDs. Similarly, the child's sex and race are permanent characteristics about the child, and are repeated if the child appears on other reports.
Other entities, like maltreatments and perpetrators, are handled differently. Since multiple maltreatments or multiple perpetrators cannot generate new records, the information about these entities must be spread out into other data elements within the same record. For example, a record allows information, such as maltreatment type and maltreatment disposition, to be stored for up to four maltreatments on the same record. For each of these maltreatments, there can be up to three perpetrators. A column is created in the dataset for each of these possibilities, whether data is stored there or not.
To do seemingly simple analysis tasks, such as counting the number of distinct perpetrators in a State, or computing the average number of maltreatments suffered by children, requires counting across these multiple elements, while accounting for empty cells or repeating data. It can be difficult to obtain the appropriate counts. In Child Maltreatment 2005, for example, counts of perpetrators are overestimated because a perpetrator is counted once for each child he or she victimizes.
To facilitate analysis of the Child File, staff at the National Data Archive on Child Abuse and Neglect at Cornell University undertook to restructure the data in such a way that each entity would reside in a separate table, with defined relationships between the tables. Such a structure is known as a relational database. In database terminology, the process of abstracting entities, characteristics of entities, and the relationships among the entities is called "normalization."
The process of normalization requires asking, for each data element, what entity it is describing. In doing so, the Archive identified eight distinct entities. A separate table was created for each entity, and each element was re-assigned to a specific table. The entities and the relationships between them are illustrated below. An arrow should be read as "may appear one or more times in." For example, one Child appears many times in the Report-Child table, and one Perpetrator may appear one or more times in the Report-Perp table.
The entities Child, Report, Perpetrator, and Maltreatment are "atomic". There is one record in the table for each distinct entity (within a State). For example, there is one record in the Child table for each distinct child identifier within each state. The structure of the Child table looks like this:
| State | ChildID | Child Sex | Child Race 1 | Child Race 2... |
|---|---|---|---|---|
| AK | 1 | value | value | value |
| AK | 2 | value | value | value |
| AK | 3 | value | value | value |
| CA | 1 | value | value | value |
| CA | 2 | value | value | value |
Other tables contain information about composite entities. Such entities require more than one identifier to define a record. For example, the Child-Perp table, which stores data elements that describe the relationship between each distinct child-perpetrator pair, looks like this:
| State | ChildID | PerpID | Relationship | Parent Role | Caretaker |
|---|---|---|---|---|---|
| AK | 1 | 1 | value | value | value |
| AK | 2 | 1 | value | value | value |
| AK | 2 | 2 | value | value | value |
| CA | 1 | 1 | value | value | value |
| CA | 2 | 1 | value | value | value |
This restructured dataset makes it much easier to obtain distinct counts of the atomic entities. To obtain a distinct, unduplicated, count of children who were subjects of a CPS investigation, for example, the researcher need only ask how many records there are in the Child table. Similarly, the Perpetrator table contains a separate record for each distinct perpetrator contained in the NCANDS data. Not only does the relational database allow for simple queries of the data, it also provides sophisticated users with the ability to develop complex queries, framing questions that could not realistically be obtained from the flat file structure of the NCANDS Child File. For example, it is possible to find how many distinct children have been physically abused by more than one type of perpetrator or to count how many reports during the year in which a perpetrator has been substantiated for maltreatment.
Problems arise in creating this version of the NCANDS data. In the relational version, a value for a data element appears only once, while in the flat file, the same element may appear multiple times. As a result, different values for the same element may occur. For example, a child's date of birth (DOB) appears in the original dataset once for each report that the child appears on. It is possible that there would be more than one version of the date, or records where no DOB appears. In moving the data from the flat file to the Child table in the relational database, only one DOB is acceptable. The one that is selected may not be the correct one, or it might end up with a missing value where the value is not missing on some other record.
Another problem is that counts produced from the relational database may vary from the published counts in the Child Maltreatment report series, causing possible confusion and making comparisons with historical published values less meaningful.
Due to such considerations, the Archive's intention is to refrain from making this version of the database available to the research community at this time. Before releasing the data or publishing information that appears to conflict with published reports, there is more work to be done. Archive staff need to collaborate with the NCANDS Technical Team and NCANDS State Representatives to compare the national DBMS with State SACWIS systems, and to explore implications for future NCANDS development. Creation of a relational database that accurately reflects the data that have been submitted by the States is the ultimate goal, with the hope that a more complete and nuanced national portrait of child maltreatment in the United States can be achieved.
For more information about the NCANDS relational database, researchers may contact:
Elliott G. Smith, PhD
Associate Director
National Data Archive on Child Abuse and Neglect
Cornell University
Ithaca, NY 14853
607–255–8104
egs1@cornell.edu
Return to Table of Contents
