DATA-DRIVEN DIFFERENTIATION

Abstract

"The heterogeneous classrooms of today require teachers to differentiate effectively. Effective differentiation however is a very time-consuming process. Teachers are faced with the challenge of first identifying the students in need of differentiated content, be it in the form of more support and easier exercises for struggling students or more challenges for high-performing students. Once these needs are identified, the teacher still needs to come up with the differentiated material that best suits the needs of each student. The identifying of differentiation needs and the delivery of differentiated content should preferably happen as the need arises, not as a delayed reaction based on observations from an exam for example. This trifecta of identifying needs, providing suitable content, and doing it all at the right time is what makes differentiation so difficult. In this article, we present a study where a digital learning platform called Eduten was used to provide automated suggestions for differentiation to teachers. The participants (N=757) were divided into two groups based on whether the teacher followed the suggestions or not. According to results, the differentiated students increased their accuracy significantly, while in the other group the accuracy remained the same. The number of completed exercises also increased more in the differentiated group, suggesting a raise in motivation. Based on the results, automated suggestions for differentiation can be highly useful but only, if the teacher follows them."