Application of machine learning approaches to analyse student success for contact learning and emergency remote teaching and learning during the COVID-19 era in speech-language pathology and audiology.

The data came from a university warehouse and existing student records, so the patterns may not apply to other speech-language pathology and audiology programmes. The study did not test a teaching or clinical intervention, so it cannot show that remote or hybrid learning caused any change in performance. Course codes were introduced, discontinued or missing across years, which limits year-to-year comparisons. The paper uses age and quintile as key attributes, but later says bias-introducing features such as age would not be included, creating uncertainty about the intended model. The accuracy figures are from model testing on this data set, not from clinical decision-making with patients.

Do not read the risk groups or model accuracy as evidence that a student will fail or that remote teaching harmed performance. The paper is a pattern analysis of existing student records, not a tested intervention.