Analysing Impact of COVID-19 on US Child Welfare

NLP & computational social science project analysing child welfare casenotes to study COVID-19 impacts on US child welfare systems.

2022 – 2023 Topic Modelling · Sentiment Analysis · NLP · Computational Social Science

Course project for Computational Social Science course by Dr. Ashton Anderson (Computer Science, University of Toronto), in collaboration with Erina Moon.

Analysed over 12,566 casenotes from a US Child Welfare (CW) Agency to investigate how COVID-19 affected child welfare operations and caseworker practices.

Methods

  • Conducted topic modelling, sentiment analysis, and open coding on CW Agency casenotes
  • Applied NLP techniques to uncover patterns in narrative casework data during the pandemic

Outcome: Poster Examining the Impact of COVID-19 on US Child Welfare Systems presented at Upper Bound Conference 2023, Alberta Machine Intelligence Institute (AMII), supported by an AMII Talent Bursary.

Links: Documentation · Poster