Browsing by Author "Piellusch, Emily K."

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Detecting substance-related problems in narrative investigation summaries of child abuse and neglect using text mining and machine learning
    (Elsevier, 2019-12) Perron, Brian E.; Victor, Bryan G.; Bushman, Gregory; Moore, Andrew; Ryan, Joseph P.; Lu, Alex Jiahong; Piellusch, Emily K.; School of Social Work
    Background State child welfare agencies collect, store, and manage vast amounts of data. However, they often do not have the right data, or the data is problematic or difficult to inform strategies to improve services and system processes. Considerable resources are required to read and code these text data. Data science and text mining offer potentially efficient and cost-effective strategies for maximizing the value of these data. Objective The current study tests the feasibility of using text mining for extracting information from unstructured text to better understand substance-related problems among families investigated for abuse or neglect. Method A state child welfare agency provided written summaries from investigations of child abuse and neglect. Expert human reviewers coded 2956 investigation summaries based on whether the caseworker observed a substance-related problem. These coded documents were used to develop, train, and validate computer models that could perform the coding on an automated basis. Results A set of computer models achieved greater than 90% accuracy when judged against expert human reviewers. Fleiss kappa estimates among computer models and expert human reviewers exceeded .80, indicating that expert human reviewer ratings are exchangeable with the computer models. Conclusion These results provide compelling evidence that text mining procedures can be a cost-effective and efficient solution for extracting meaningful insights from unstructured text data. Additional research is necessary to understand how to extract the actionable insights from these under-utilized stores of data in child welfare.
  • Thumbnail Image
    Item
    Prevalence and context of firearms-related problems in child protective service investigations
    (Elsevier, 2020-09) Sokol, Rebeccah L.; Victor, Bryan G.; Piellusch, Emily K.; Nielsen, Sophia B.; Ryan, Joseph P.; Perron, Brian E.; School of Social Work
    Background: Despite the significance of firearm safety, we need additional data to understand the prevalence and context surrounding firearm-related problems within the child welfare system. Objective: Estimate proportion of cases reporting a firearm-related problem during case initiation and the contexts in which these problems exist. Sample and setting: 75,809 caseworker-written investigation summaries that represented all substantiated referrals of maltreatment in Michigan from 2015 to 2017. Methods: We developed an expert dictionary of firearm-related terms to search investigation summaries. We retrieved summaries that contained any of the terms to confirm whether a firearm was present (construct accurate) and whether it posed a threat to the child. Finally, we coded summaries that contained firearm-related problems to identify contexts in which problems exist. Results: Of the 75,809 substantiated cases, the dictionary flagged 2397 cases that used a firearm term (3.2 %), with a construct accuracy rate of 96 %. Among construct accurate cases, 79 % contained a firearm-related problem. The most common intent for a firearm-related problem was violence against a person (45 %). The co-occurrence of domestic violence and/or substance use with a firearm-related problem was high (41 % and 48 %, respectively). 49 % of summaries that contained a firearm-related problem did not provide information regarding storage. Conclusion: When caseworkers document a firearm within investigative summaries, a firearm-related risk to the child likely exists. Improved documentation of firearms and storage practices among investigated families may better identify families needing firearm-related services.