Ethical considerations in research when building predictive risk modelling in child and family welfare

Authors

DOI:

https://doi.org/10.31265/jcsw.v19i1.619

Keywords:

ethics, child and family welfare, Child Protection, predictive risk modelling, machine learning, decision-making, notifications

Abstract

This article presents and discusses ethical issues and implications in research when building a predictive risk model for potential use in Danish child and family welfare. The idea is to build a predictive risk model in order to study whether such a model can be valuable to child and family welfare services in the assessment of risk – aimed specifically at the decision-making process regarding notifications.

Based on a framework developed especially for this field, we present and discuss ethical considerations, reflections and actions in relation to four main ethical principles: non-maleficence, autonomy, justice and explicability. We hope that our reflections on these ethical challenges can inspire research – and potentially also the field of practice  when taking a deep dive into the difficult field of digitalization in social work.

Author Biographies

Anne Marie Villumsen

Senior Researcher
VIVE – The Dannish Centre for Social Sciences Research
Denmark
E-mail: amav@vive.dk

Michael Rosholm

Professor
Department of Economics and Business Economics, Aarhus University
Denmark
E-mail: rom@econ.au.dk

Simon Tranberg Bodilsen

Postdoc
Department of Economics and Business Economics, Aarhus University
Denmark
E-mail: sibo@econ.au.dk

Sanne Dalgaard Toft

Project Manager
Department of Economics and Business Economics, Aarhus University
Denmark
E-mail: sanne@econ.au.dk

Line Svolgaard Berg

Associate Professor
VIA University College
Denmark
E-mail: libe@via.dk

Liesanth Yde Nirmalarajan

Research Assistant
Department of Sociology and Social Work, Aalborg University
Denmark
E-mail: lin@socsci.aau.dk

References

Ada Lovelace Institute. (2022). Looking before we leap: Ethical review processes for AI and data science research. https://www.adalovelaceinstitute.org/report/lookingbefore-we-leap/Ethics and accountability in practice

Cheng, H. F., Stapleton, L., Kawakami, A., Sivaraman, V., Cheng, Y., Qing, D., ... & Zhu, H. (2022, April). How child welfare workers reduce racial disparities in algorithmic decisions [Conference presentation]. 2022 CHI Conference on Human Factors in Computing Systems, New Orleans. https://doi.org/10.1145/3491102.3501831

Chouldechova, A., Benavides-Prado, D., Fialko, O. & Vaithianathan, R. (2018). A case study of algo-rithm-assisted decision making in child maltreatment hotline screening decisions [Conference presentation]. 1st Conference on Fairness, Accountability and Transparency https://proceedings.mlr.press/v81/chouldechova18a.html.

Coulthard, B., Mallett, J., Taylor, B.J., 2020. Better decisions for children with ‘big data’: can algo-rithms promote fairness, transparency and parental engagement? Societies, 10(4), 97. https://doi:10.3390/soc10040097.

Cuccaro-Alamin, S., Foust, R., Vaithianathan, R. & Putnam-Hornstein, E. (2017). Risk assessment and decision making in child protective services: Predictive risk modeling in context. Children and Youth Services Review, 79, 291-298. https://doi.org/10.1016/j.childyouth.2017.06.027

Devlieghere J., PGillingham, P. & Roose, R. (2022): Dataism versus relationshipism: a social work perspective, Nordic Social Work Research, 12(3), 328-338. https://doi.org/10.1080/2156857X.2022.2052942

Gilbert, N. (ed). 2007. Combatting Child Abuse: International Perspectives and Trends. Oxford uni-versity press.

Gilbert, P., Parton, N. & Skivenes, M. (eds.). (2011). Child Protection Systems: International Trends and Orientations. Oxford University Press. https://doi.org/10.1093/acprof:oso/9780199793358.001.0001

Gillingham, P (2016): Predictive Risk Modelling to Prevent Child Maltreatment and Other Adverse Outcomes for Service Users: Inside the ‘Black Box’ of Machine Learning. British Journal of So-cial Work (2016) 46(4), 1044-1058. https://doi.org/10.1093/bjsw/bcv031

Goldhaber-Fiebert, J. D. & Prince, L. (2019). Impact evaluation of a predictive risk modeling tool for Allegheny county’s child welfare office. Allegheny County.

High-Level Expert Group. (2019). Ethics Guidelines for trustworthy AI. European Commission. https://digital-strategy.ec.europa.eu/en/library/ethics-guidelines-trustworthy-ai

High-Level Expert Group on Artificial Intelligence. (2018). A definition of AI: main capabilities and scientific disciplines. European Commission. https://digital-strategy.ec.europa.eu/en/library/definition-artificial-intelligence-main-capabilities-and-scientific-disciplines

Kawakami, A., Sivaraman, V., Cheng, H. F., Stapleton, L., Cheng, Y., Qing, D., ... & Holstein, K. (2022, April). Improving human-AI partnerships in child welfare: understanding worker practic-es, challenges, and desires for algorithmic decision support. [Conference presentation]. 2022 CHI Conference on Human Factors in Computing Systems, New Orleans. https://doi.org/10.1145/3491102.3517439

Keddell, E. (2023). The Devil in the Detail: Algorithmic Risk Prediction Tools and Their Implications for Ethics, Justice and Decision-making. In B. Taylor, J. D. Fluke, J. C. Graham, E. Keddell, C. Kil-lick, A. Shlonsky & A. Whittaker (eds.) The Sage Handbook of Decision Making, Assesment and Risk in Social Work (pp. 405-420). Sage Publications. https://doi.org/10.4135/9781529614657.n51

Kriz, K. & Skivenes, M. (2013). Systemic Differences in Views on Risk: A Comparative Case Vignette Study of Risk Assessment in England, Norway and the United States (California). Child and Youth Services Review, 35(11), 1862–1870. https://doi.org/10.1016/j.childyouth.2013.09.001

Leslie, D., Holmes, L., Hitrova, C. & Ott, E. (2020). Ethics review of machine learning in Children’s social care. The Alan Touring Institute. Oxford University. https://whatworks-csc.org.uk/research-report/ethics-review-of-machine-learning-in-childrens-social-care/

Lehtiniemi, T. (2024). Contextual social valences for artificial intelligence: anticipation that mat-ters in social work. Information, Communication & Society, 27(6), 1110-1125. https://doi.org/10.1080/1369118X.2023.2234987

Lundberg, S.M. & Lee. S. I. (2017) A unified approach to interpreting model predictions. In I. Guy-on, U. Von Luxburg, S. Bengio, H. Wallach, R. Fergus, S. Vishwanathan & R. Garnett (eds.), Ad-vances in Neural Information Processing Systems 30 (NIPS 2017). https://papers.nips.cc/paper_files/paper/2017/hash/8a20a8621978632d76c43dfd28b67767-Abstract.html

Mittelstadt, B. D., Allo, P., Taddeo, M., Wachter, S. & Floridi, L. (2016). The ethics of algorithms: Mapping the debate. Big Data & Society, 3(2), 1-21. https://doi.org/10.1177/2053951716679679

Morley, J., Floridi, L., Kinsey, L. & Elhalal, A. (2020). From what to how: an initial review of publicly available AI ethics tools, methods and research to translate principles into practices. Science and engineering ethics, 26(4), 2141-2168. https://doi.org/10.1007/s11948-019-00165-5

Munn, L. (2023). The uselessness of AI ethics. AI and Ethics, 3(3), 869-877. https://doi.org/10.1007/s43681-022-00209-w

Pösö, S., Skivenes, M. & Hestbæk, A.-D. (2013). Child Protection Systems within the Danish, Finnish and Norwegian Welfare States – Time for a Child Centric Approach?. European Journal of Social Work 17(4), 475–490. https://doi.org/10.1080/13691457.2013.829802

Rosholm, M., Bodilsen, S. T., Michel, B. & Nielsen, S. A. (2024). Predictive risk modeling for child maltreatment detection and enhanced decision-making: Evidence from Danish administrative data. PLOS ONE, 19(7), e0305974. https://doi.org/10.1371/journal.pone.0305974

Søbjerg, L. M., L. Nirmalajaran & A. M. Villumsen. (2020). Perceptions of Risk and Decisions of Referring Children at Risk. Child Care in Practice 26(2), 130-145. https://doi.org/10.1080/13575279.2019.1685460

Søbjerg, L.M., Taylor, B.J., Przeperski, J., Horvat, S., Nouman, H. & Harvey, D. (2020). Using risk-factor statistics in decision making: prospects and challenges. European Journal of Social Work, 24(5), 788-801. https://doi.10.1080/13691457.2020.1772728

Taylor, B. J. (2020). Teaching and learning decision making in child welfare and protection social work. In J. Fluke, M. López López, R. Benbenishty, E. J. Knorth, & D. J. Baumann (Eds.), Decision making and judgement in child welfare and protection: Theory, research and practice (pp. 281–298). Oxford University Press. https://doi.org/10.1093/oso/9780190059538.003.0013

Vaithianathan, R., Maloney, T., Putnam-Hornstein, E. & Jiang, N. (2013). Children in the public benefit system at risk of maltreatment: Identification via predictive modeling. American Journal of Preventive Medicine, 45(3), 354-359. https://doi.org/10.1016/j.amepre.2013.04.022

Villumsen, A. M. (2017). Hvorfor det ikke er så lige til med udsathed hos børn og unge. In. D. Graversen (ed.), Pædagogik: introduktion til pædagogens grundfaglighed (1st ed.). Hans Reitzels Forlag.

Villumsen, A. M., & Søbjerg, L.M. (2020).: Informal Pathways Informal pathways as a response to limitations in formal categorization of referrals in child and family welfare. Nordic Social Work Research, 13(2), 176-187. https://doi.org/10.1080/2156857X.2020.1795705

Downloads

Published

2024-10-03

How to Cite

Villumsen, A. M., Rosholm, M., Bodilsen, S. T., Toft, S. D., Berg, L. S., & Nirmalarajan, L. Y. (2024). Ethical considerations in research when building predictive risk modelling in child and family welfare. Journal of Comparative Social Work, 19(1), 102–126. https://doi.org/10.31265/jcsw.v19i1.619