The purpose of this symposium is to present three levels of findings from the Early Grade Reading Study (EGRS) mixed method impact evaluation. The study, led by the South African Department of Basic Education in partnership with the University of the Witwatersrand and the Human Sciences Research Council, is designed to advance knowledge of sustainable system-wide instructional reform in low and lower-middle income systems and provide an evidence-based warrant for scaling up interventions. The EGRS is anchored on a randomised experiment complemented with 60 structured classroom observations and in-depth case studies designed to investigate the cost, efficacy and mechanisms of three intervention models intended to improve reading outcomes in the Setswana language (Home Language) in field trials in 230 primary schools in the North West province of South Africa.
The study investigates the cost-effectiveness of two versions of the ‘education triple-cocktail’ model, otherwise referred to as structured pedagogic programmes (Snilstviet et al, 2016) or a combined approach (McEwan, 2015). The education triple cocktail consists of curriculum aligned systematic and structured lesson plans, provision of quality reading materials including graded or levelled readers, and capacity building. In the ‘lite’ (less expensive) model, teachers receive centralised training on the pedagogic programme with lesson plans and materials. In the ‘full’ model, teachers are assigned instructional coaches who provide one-on-one support in their classrooms approximately twice a month and ongoing support via Whatsapp in addition to the lesson plans and reading materials. The third model consists of the training and deployment of local community reading coaches to work with parents on how to improve and support reading in the home.
This symposium reports on findings of three datasets. The first paper presents the findings of the baseline, midline and endline results of EGRS student testing. Having tracked 3700 students over two years, at three data collection points, this paper represents a rigorous analysis of the comparative growth of reading skills and competencies in schools in the four groups (3 treatments [models] and 1 control). The second paper presents the findings from observations of 60 literacy lessons in the two ‘triple cocktail’ intervention groups and control group. The classroom observation paper provides valuable findings on how the interventions impact teachers’ instructional practices and offers some important insights about changed practice. The third paper, which reports on the case studies, analyses of classroom practices, informal conversations, and semi-and unstructured interviews with teachers and school leaders. The case studies provide insight into contextual realities, teachers’ language of practice, and mechanisms of change.
Papers
Improving Early Reading in Setswana: Preliminary findings of the randomized control trial in South Africa - Stephen Taylor & Nompumelelo Mohohlwane, South African Department of Basic Education, Department of Basic Education and Cas Prinsloo, Human Sciences Research Council
Structured Observation of Classroom Practice in the Early Grade Reading Study in South Africa - Janeli Kotze, Department of Basic Education; Jacobus Cilliers, Georgetown University
Contextual constraints and mechanisms from the case studies of the Early Grade Reading Study in South Africa - Brahm Fleisch and Kerryn Dixon, University of the Witwatersrand
Discussant: Crain Soudien
