Logo image
Applying machine learning models in multi-institutional studies can generate bias
Conference proceeding

Applying machine learning models in multi-institutional studies can generate bias

Rebeckah K. Fussell, Meagan Sundstrom, Sabrina McDowell and N. G. Holmes
2024 PHYSICS EDUCATION RESEARCH CONFERENCE, PERC, pp 144-149
01 Jan 2024
url
https://doi.org/10.1119/perc.2024.pr.FussellView
Published, Version of Record (VoR) Open

Abstract

Education & Educational Research Education, Scientific Disciplines Social Sciences
There is increasing interest in deploying machine learning models at scale for multi-institutional studies in physics education research. Here we investigate the efficacy of applying machine learning models to institutions outside of their training set, using natural language processing to code open-ended survey responses. We find that, in general, changing institutional contexts can affect machine learning estimates of code frequencies: either previously documented sources of uncertainty increase in magnitude, new unknown sources of uncertainty emerge, or both. We also find an example where uncertainties do not change between the institution used in the training data and an institution not in the training data. Results suggest that attention to uncertainty is critical, especially when making measurements of student writing across multi-institutional data sets.

Metrics

22 Record Views

Details

UN Sustainable Development Goals (SDGs)

This publication has contributed to the advancement of the following goals:

#4 Quality Education

InCites Highlights

Data related to this publication, from InCites Benchmarking & Analytics tool:

Web of Science research areas
Education & Educational Research
Education, Scientific Disciplines
Logo image