Classifying the quality of robotics-enhanced lesson plans using motivation variables, word count, and sentiment analysis of reflections

• Classified lesson plan quality using motivation, word count and sentiment analysis variables. Strongest classifiers were of field experience teaching coding task; task. According to ordinal logistic regression, performance goal orientation was a significant predictor front-end quality. Classifiers can be used flag preservice teachers for provision additional support before plans are created. While planning, it is critical uncover signs that needed. The conceptual framework this study grounded in two fundamental perspectives as learners designers guided our research inquires motivational factors reflection processes. In study, discriminant model incorporating motivation variables, count, reflections classify Next, vector machines identified misclassified plans. Finally, we regression estimate Betas each predictor. strongest scores counts. fairly accurate, yielding 12.82%, 7.69%, 23.08% misclassification rate. Ordinal indicated