Core Dataset Project: Child Welfare Service Histories. Previous Child Welfare Population Research


In addition to this problem of linking across the two domains, there is also the fundamental problem of how little population-level child welfare statistics have been done. Most research and analysis is not longitudinal and only counts children or cases during a period of time. Neither does most research unduplicate to the child or family level. Without a method for unduplicating them, multiple reports for the same child being counted within a year may result. Also, on the service side, much of the reporting is for a particular point-in-time, which mixes children who have been in the system from 1 day with those who have been in the system for up to 18 years. This point-in-time research is extremely important for administrators and service providers as it tells them the characteristics of children that must be served at a point in time, but it does little to show the outcomes for children who have either been in the system for varying amounts of time or who enter because of different reasons.

There are numerous reasons for the absence of longitudinal, child-level, population-wide work in addition to the historical separation between the two systems and a lack of an ID link between the two. The simple volume of data which one must process and analyze to understand the experiences of the populations of children in both domains is so large that it has been nearly impossible and exorbitantly expensive to compute without the recent decrease in the cost of computing and increase in computing power. Also, states have been unwilling to share this data without extensive confidentiality agreements that often take months and sometimes years to develop. These factors, in addition to poor documentation, often provided obstacles that were too difficult to overcome. Finally, the complexity of analyzing sequences has only recently been addressed with methods of pattern matching. This study utilizes these methods to analyze the results.

The information from this type of analysis is important for a number of reasons that respond to today’s child welfare system challenges including:

  • Outcomes of service - linking future events (e.g. reunification or adoption) to the initial events that led to intake.
  • Managed care - understanding service histories are critical to having good information to both plan and implement managed care around foster care or other services that children and their parents may receive while in the child welfare system.
  • Family Preservation - Understanding the front end of the child welfare system is necessary so that family preservation services can be designed more effectively through more effective triage and resources allocation.
  • Children who are not reported as abused or neglected - The percent of children who are in the child welfare system for a reason other than abuse or neglect varies over time. Depending on the availability of resources and the "policy" environment at a point in time in a state, these children may be more or less welcome in the foster care system.

View full report


"chapin1.pdf" (pdf, 71.08Kb)

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®