Data science is just about everywhere these days: the online marketplace, video streaming services, hospitals, insurance companies, and even baseball teams. But does it impact what social workers do? Increasingly, yes. While relatively new to social work practice, data is used in predictive analytics, an emerging tool for child welfare.
What is predictive analytics? Also known as predictive risk modeling, predictive analytics is a strategic tool that social workers employ to determine whether or not referrals require investigation. Social workers field calls about neglect and abuse from concerned parties on child protection hotlines and other reporting avenues. Typically, a social worker will cross-reference these tips with existing information databases before deciding whether to move forward with an investigation. Predictive analytics helps those in social worker jobs make quicker and more accurate investigation decisions based on data, resulting in better child welfare outcomes.1
According to the U.S. Department of Health and Human Services (HHS), predictive analytics “use data, statistics, and algorithms to answer the question ‘Given past behavior, what is likely to happen in the future?’”2 According to a report by the department, child welfare organizations typically use predictive analytics to address these four main areas of concern:3
- Assessing increased risk after and during preventive services
- Predicting the likelihood of repeated events
- Examining cross-system interactions
- Scrutinizing agency operations to better understand an organization’s own activities, like employee turnover and casework trends
As data collection and interpretation improves, predictive analytics will become increasingly valuable for child welfare agencies. They will help caseworkers uncover insights from large quantities of data and more quickly determine risk across varied circumstances. Regardless, HHS cautions predictive analytics “are intended to supplement good casework practice” and are “an additional tool to add incrementally to the child welfare toolkit.”3
As predictive analytics is so powerful, it must be harnessed ethically and carefully. Social workers should use it to support and enhance their own work, not replace. Agency administrators should create and maintain ethical frameworks to govern the use and application of predictive analytics to support child welfare decisions.4
Like all tools, predictive analytics has some limitations. Critics worry data derived from predictive analytics could become subject to bias. If unchecked, it “could be used disproportionally in a manner that could heighten racial bias and stigmatize, marginalize, and put undue attention on certain parents, particularly those of color, that might lead to the unjust wrenching of children from their families.”1 Social workers may also be susceptible to confirmation bias, letting data derived from predictive analytics dictate child welfare decisions without additional verification from casework.5
To counter these risks, HHS recommends significant caseworker training on predictive analytics. This training enables social workers to interpret risk scores effectively, avoid bias, and better understand predictive modeling practices. To further streamline the implementation of predictive analytics, HHS also suggests inter-agency collaboration through sharing of legal templates, common data definitions, and other resources.3
If you’re interested in the field of social work, or wish to advance your current knowledge in the field, Walden University’s Master of Social Work (MSW) program provides the opportunity to earn your MSW online, with the flexibility you need. Additionally, Walden’s online MSW program is accredited by the Council on Social Work Education (CSWE), which helps assure that the degree you earn meets your professional and educational goals.
Walden University is an accredited institution offering a Master of Social Work (MSW) degree program online. Expand your career options and earn your master’s degree using a convenient, flexible learning platform that fits your busy life.
1Source: www.socialworktoday.com/archive/MA18p10.shtml
2Source: aspe.hhs.gov/predictive-analytics-child-welfare-decision-tool
3Source: aspe.hhs.gov/system/files/pdf/257841/PACWAnAssessmentCurrentEffortsChallengesOpportunities.pdf
4Source: caseyfamilypro-wpengine.netdna-ssl.com/media/Considerations-for-Applying-Predictive-Analytics-in-Child-Welfare.pdf
5Source: www.nccdglobal.org/sites/default/files/inline-files/Principles%20for%20Predictive%20Analytics%20in%20Child%20Welfare-1.pdf
Walden University is accredited by The Higher Learning Commission, www.hlcommission.org.