Using Machine Learning to Predict Teacher Research Reading

• In Event: Predicting Teacher Learning and Mindsets

Abstract

Autonomous teachers’ reading of materials related to their job contributes significantly to their professional development. The present study sampled 10,469 language teachers using the teacher questionnaire of PISA 2018 to investigate the factors that could affect their professional reading. The performances of two machine learning models, logistic regression and Support Vector Machines (SVM), in correctly classifying low and high readers based on 19 predictors were very similar (with accuracies around 65%). The length of the reading text that the teachers assign their students, instruction of reading comprehension strategies, and teachers’ own reading habits of general books and news were found to be the most important predictors of their professional reading time.

Authors

• Mehrdad Yousefpoori-Naeim, University of Alberta
  Presenting Author

• Surina He, University of Alberta
  Non-Presenting Author

• Ying Cui, University of Alberta
  Non-Presenting Author

• Maria Cutumisu, McGill University
  Non-Presenting Author