Child welfare outcomes vary widely in North Carolina. Data can help explain why.

Published on
November 30, 2020

Taking a child from a home is usually a last resort. Social workers are required by federal and state laws and policies to follow specific protocols when deciding whether and when to intervene.

All 100 counties in North Carolina are supposed to operate by the same rules and policies. But those state-level rules are sometimes ambiguous. In practice, local policies lack uniformity, leading to widely divergent outcomes for children and families.

For example, studies by state program evaluators show a wide discrepancy in the intake procedures counties use, resulting in an uneven application of child welfare services statewide.

The lack of consistency in screening abuse reports is concerning enough for state legislators to consider changes to North Carolina’s system. In one in four counties in North Carolina, workers incorrectly screen maltreatment or abuse reports, according to a recent legislative study. 

That study also found that not every county screens complaints about child maltreatment the same way, leaving open the possibility that similar cases in different counties will have different responses and outcomes.

This lack of consistency is possible because of the way North Carolina administers its child welfare system. Most states have a centralized administration of child welfare services, with an overarching state department that monitors regional hubs.

While North Carolina’s Department of Health and Human Services oversees child welfare offices in each of the state’s 100 counties, the state relies on counties to decide how to allocate scarce resources of time and money to protect children.

Our research and experience in North Carolina shows the state’s aversion to centralized oversight and accountability is leading to erratic outcomes, with extremes of severity or laxity. 

This also pointed us toward a project that could identify policy and legislative solutions. Additional research showed very few states have the combination of policies that result in this pattern in North Carolina.

Carolina Public Press has written about issues in child protective services agencies for years. In Cherokee County, agency workers were taking children from parents without judicial authority — which has since resulted in felony indictments for two former workers and the former director of social services. These workers and the office were ordered by a judge to not destroy any documents — they did anyway. The situation is also the subject of a federal civil suit.

Editor Frank Taylor has for years followed irregular practices at a variety of child welfare offices. The situations seemed to defy any pattern — some counties were simply sloppy, others appeared overly aggressive in seizing children, some refused to intervene at all despite evidence of obvious abuse and neglect — and some played nepotism games.

Carolina Public Press hopes to examine publicly available data, under the auspices of the 2020 Data Fellowship, to explore inequities in the state’s child welfare system, including those that result in the termination of parental rights.

We will analyze how children in other states that have converted from a decentralized model to one with more state oversight have fared — specifically looking at Georgia, which recently made this change and is comparable to North Carolina in size, demographic makeup and regional identity.

We know some of the policies are borne out of what local officials believe is a necessity. Living arrangements can vary widely from community to community. We hope to identify the outlier counties using data analysis and focus our efforts toward interviewing the officials there and examining outcomes for children.

Undoubtedly the response to the COVID-19 pandemic has changed how some child welfare offices operate. We intend through our reporting to learn how the pandemic changed social workers’ daily jobs, how families are coping with a change to those services, and whether children are suffering more or less. Early anecdotal evidence indicates that because children are not in physical schools, teachers no longer spot abuse that may be visible due to a change in a child’s demeanor, a fading bruise or broken bones.

This will not be an easy task. Through the fellowship, I’ll be seeking to learn more about statistical methods for data analysis using a programming language called R. A deep analysis of publicly available data may yield results that will help illuminate problems in North Carolina that can be made better with a little sunlight. It may reveal which counties have deeply irregular practices when compared to other counties.