Using EGRA data for differentiated instruction: Learning profiles and instructional needs in Uganda

Presentation delivered at CIES2017 (Atlanta). A challenge of large-scale education research projects in international development is determining the most appropriate way to effectively report findings for a wide variety of audiences (e.g., researchers, ministry officials, donors, and other relevant stakeholders). It is important to consider technical rigor and accessibility, while ultimately providing results that can be used to inform policy and instruction. Using Early Grade Reading Assessment (EGRA) data from Uganda in two languages, we are replicating an approach for categorizing students into learning profiles, which are directly tied to their particular instructional needs. We have conducted this analysis previously with data from Indonesia. We had conceptualized on the framework to be used across different context This study reports on the efficacy of using the same method with this sample with Luganda speakers. We divided students into five learning profiles based on their reading ability (Next Grade Ready, Fluent, Instructional, Beginner, and Nonreader) and then examined the relationship among these profiles and their reading skills on a variety of EGRA subtasks to determine the instructional need required to promote students from one profile to the next. Our learner profile method has been used with data in two countries and two languages. EGRA has been used in over 70 countries and in more than 100 languages. This study is an initial attempt to explore the value in this method.