When a child is in need of temporary care due to abuse or neglect, the decision regarding where the child should live is often an imperfect one, based more on availability of a home rather than suitability. Child protection agencies striving to improve the placement process must look first to relatives using innovative and proven family search and engagement (FSE) strategies. When relatives are not an option, a child is often placed with the first foster parent who says “yes.” While this might provide a bed for the night, and perhaps even the foreseeable future, this kind of hasty, haphazard decision-making process does not support the likelihood that the child will be able to stay in the same foster home until returning to his or her own home, or achieving another form of permanency.

All children need stability and security, especially children in foster care. Child protection agencies can do more to improve placement stability by using tools to identify the right foster home so that the first placement is the only placement.

How have jurisdictions implemented placement matching tools?

Some child protection agencies have been harnessing the power of data and predictive analytics through placement matching tools that help identify the suitability of a foster or adoptive family for a child in need of a temporary or permanent home. Using an algorithm that is data-driven and based on empirical research, a placement matching tool may be used to supplement and enhance a caseworker’s clinical decision-making process, not substitute it. Using information about a child’s characteristics as well as a family’s, including location, demographics, and other considerations, placement matching tools rate the compatibility of available families to help a caseworker make more informed decisions about which families to contact regarding potential placement.

Placement matching tools are often compared to dating sites that use compatibility algorithms. In fact, one such tool involved the work of the former head of research and development at e-Harmony, Dr. Gian Gonzaga, and Heather Setrakian, research scientist and co-founder of eHarmony Labs. Together, they supported Thea Ramirez’s vision in the creation of Family-Match, a data-driven adoption matching tool. Launched in 2014 by Adoption-Share, a nonprofit organization, Family-Match¹ is based on an extensive review of the empirical literature regarding the predictors of disruption in foster care and adoption placements. Family-Match is being piloted in Florida and Virginia for the purpose of matching adoptive families to children in need of adoptive homes and, understanding the need for better child-family matches earlier in a child’s experience of out-of-home care, Family-Match is also being integrated into Tennessee’s statewide automated child welfare information system (SACWIS) so that it can support improved matching for foster care placements as well.

Another placement matching tool available to the field is Every Child A Priority (ECAP).² Originally created by a multistate private foster care agency, this research-backed matching tool was developed using the expertise of seasoned social workers with decades of experience in determining placements for children in foster care. ECAP, which is now distributed and maintained by Foster Care Technologies, can be interfaced with any agency’s case management system. Since 2010, ECAP has helped guide over 24,000 placements and is being used by foster care networks in seven states.

In addition to these two off-the-shelf placement matching tools, New Jersey has developed its own tool that is embedded in its SACWIS system, and New York state and Washington, D.C., are developing their own placement matching tools as well. Whether using an in-house or off-the-shelf model, these predictive tools suggest which families a worker might want to consider talking to first, in order to support a decision about who might be the best match. The tool narrows the pool of potential foster or adoptive parents, which is then further assessed by caseworkers tasked with determining the best match for the child.

Are placement matching tools effective?

Our understanding of the power of placement matching tools is still in its infancy. However, an evaluation³ of ECAP by the University of Kansas School of Social Welfare indicates that this tool has had significant impact on placement stability and timely permanence, which have also led to cost savings for private providers and the public child protection agency. These cost savings are based on fewer days spent in care, as well as the impact on staff workload every time a child changes placement. Although foster parent retention was not included in the evaluation, preliminary data indicate that placing children who are better suited to their resource home has improved foster parents’ experience with providing out-of-home care, increasing the likelihood that they will continue to foster.

Although Family-Match is new in its implementation, it has already demonstrated an impact on time to adoption; the tool has supported placements of children into pre-adoptive homes in as little as six weeks, even children who have been legally free for adoption for years. Family-Match is also committed to tracking outcomes by following up with families and caseworkers at one, three, six, and 12 months post-placement. At each of these intervals, families are sent a 10-minute online questionnaire, and caseworkers are sent a brief survey that takes about 2.5 minutes to complete. The questionnaires are designed to measure not only whether the child is still in the original resource home but, more importantly, the quality of the match. The outcome data allows the matching model to continually improve over time.

As a result of ECAP

• Placement stability improved by 22.5%
• Median time to permanency decreased by 53 days
• Contracting foster care agency saved $731,732 in one year
• The state of Kansas saved $3,543,436 in one year

What are some implementation considerations?

There are a number of considerations that need to be addressed when contemplating the addition of a matching tool to support placement decision-making for youth in out-of-home care:

• High quality data is essential: As with all computer algorithms, the quality of the match is only as good as the quality of the data available to the system. Therefore, it is imperative that the tool have as much information as possible about the child’s characteristics and needs, as well as the family’s characteristics and skills. For a placement matching tool to be useful to caseworkers and beneficial to children, caseworkers must understand the importance of entering quality data into the system. This process can be aided by use of standardized screening and assessment tools. In Tennessee, for example, the Child and Adolescent Needs and Strengths (CANS) 2.0 is used to assess a child’s strengths and needs, and this information is entered into the state’s SACWIS, which then automatically populates the Family-Match tool to streamline the process for staff. In Florida and Virginia, families are responsible for creating their own profiles as another strategy for ensuring the quality of the data and reducing the impact on staff workload. ECAP can also incorporate data from other assessments, such as the CANS or the Child and Adolescent Functional Assessment Scale (CAFAS), which can then be considered alongside ECAP’s own research-backed assessment.
• A sufficient supply of resource families is necessary: High-quality matches can only be made if there is a sufficient array of foster homes available, so that the tool can identify those families that can best meet the needs of the child. Investing in a placement matching tool should only be one part of a comprehensive resource home recruitment and retention plan, one that includes prioritizing kinship caregivers and recruiting and retaining a diverse array of foster parents.
• The tool improves with use: The more a placement matching tool is used, the more data the algorithm has to work with, which improves the accuracy of the tool in recommending quality matches. As a placement matching tool such as ECAP is used more widely, the quality of the tool should improve over time. Quality control data such as the information gathered through Family-Match’s follow-up questionnaires also serve to improve the tool.
• The top match isn’t necessarily the best match: Placement matching tools, by themselves, are not “the” answer. A tool can provide guidance, but agencies should not always assume that the top match is automatically the best match, nor should they disregard the intuition and experience of their placement staff. For example, a family might be a good match for one child, but not for the child’s sibling. There is no one single “best” placement for a child, but rather, there are potentially a number of homes that can provide equally good care for any given child. It is important that caseworkers consider a few of the top matches for a child or sibling group, engage those families in exploration of the “goodness of fit,” and then apply their clinical skills to making the best decision possible for that particular child or children.

1 All information related to Family-Match is from personal communication with Thea Ramirez, Adoption-Share Founder + Chief Sharer, April 9, 2018, and with Thea Ramirez, Dr. Gian Gonzaga, Adoption-Share Data Scientist + Compatibility Guru, and Heather Setrakian, Adoption-Share Marriage and Family Researcher, April 13, 2018.
2 Unless otherwise noted, all information related to ECAP is from personal communication with Eve Anderson, Foster Care Technologies Chief Marketing Officer, and Paul Epp, Foster Care Technologies Chief Operating Officer, May 23, 2018.
3 Foster Care Technologies. (2018). A new standard in placement stability: How placement matching can lead to better outcomes in foster care. Retrieved from https://www.fostercaretech.com/