Classificatory stream analysis in the prediction of expected reading readiness: understanding student performance

Abstract

This research describes the application of a neural networks approach in the prediction of readiness for reading upon entry to primary education. Machine-learning techniques used offer an iterative methodology that is capable of discovering complex relationships and interactions in the inputs and outcomes. The approach maximized classification accuracy, and was able to model various outcome patterns from the over 700 students studied. Results based on hypotheses of student characteristics using these predictive modeling achieved a total accuracy of 98% in the identification of "students-below-readiness-threshold". The presentation explains the processes and the stream analysis technique utilized, and explores various alternative models.