Uses machine learning on digital tutoring-platform data from Copenhagen Municipality schools (the MatematikFessor platform) to measure student math knowledge and study algorithmic fairness, how classroom and non-classroom instruction affects progression, and the sustained effects of individualized tutoring.

Related papers: Fairness Over Time: A Nationwide Study of Evolving Bias in Dropout Prediction·Trading off performance and human oversight in algorithmic policy: evidence from Danish college admissions

← All research