Blended CPD for teaching of reading in isiZulu Home Language: Reflections and lessons from a quasi-experimental study

• In Event: Digital empowerment and transformation

Proposal

Persistent low literacy among South African children underscores the need for effective teacher development interventions. The Early Grade Reading Instruction (EGRI) programme addresses this by implementing a blended Continuous Professional Development (CPD) model for Foundation Phase teachers in isiZulu Home Language instruction. EGRI integrates online learning modules with Professional Learning Communities (PLCs).
This paper explores how blended CPD strengthens teacher knowledge and practice in teaching early grade reading and literacy development.
We present findings from a mixed-methods study employing a convergent parallel design using quantitative and qualitative data. Primary schools participating in the EGRI pilot and a group of non-participating schools, were sampled for data collection. Baseline data were gathered using the (isiZulu) Early Grade Reading Assessment. Other data were collected via a pre-test, surveys, classroom and PLC observations, interviews, focus group discussions, and artefact analysis. Quantitative and qualitative data were analysed independently and later integrated.
Quantitative data were analysed using statistical hypothesis testing frameworks, incorporating randomization-based inference. Permutation tests and bootstrap resampling techniques were employed to account for the clustered sampling design and to ensure robust estimation of between-group differences and treatment effects, mitigating potential biases from reduced sample sizes and non-normal data distributions. Qualitative data were analysed thematically providing contextual insights. Endline data will be collected from October 2024 to April 2025. A pre-test of teacher knowledge of literacy instruction (n 786) showed strong content knowledge in phonics (mean 81%, SD 19.4) and reading strategies (mean 72.4%, SD 22.9), and weaker knowledge in vocabulary and comprehension (mean 65%, SD 25; mean 64.4%, SD 20.8).
A sub-sample of 381 surveyed teachers reveal high self-reported self-efficacy in instructional practices, particularly in addressing reading difficulties, but gaps in self-efficacy in differentiated instruction and learner engagement. However, classroom observations (n 127) for intervention and comparison schools show instructional strategies were less highly rated than some other aspects of teaching.
Assessments of 5192 Grade 2 learners (comparison n 1893; intervention n 3299) indicated significant challenges in foundational reading skills, with over 75% of learners reading fewer than 20 correct words per minute (level expected at the end of Grade 2). Baseline data confirmed equivalency of learners assessed between intervention and comparison groups, validating the design for subsequent impact analysis at endline.
Preliminary implementation data indicated that 29.4% of teachers who attended orientation sessions (n 236/803) registered for the online course, 25.9% of registered teachers (n 315/1217) had completed all available modules at the time of the baseline study, and 68.8% of surveyed teachers (n 262/381) reported participating in PLCs.
Baseline results highlight gaps in teacher self-efficacy in overcoming challenges posed by difficult home environment, instructional practices, as well as learner abilities in letter recognition and decoding.
Endline data collection and analysis (October 2024 to March/April 2025) will elucidate impact on teacher content knowledge, self-efficacy, and the contribution of PLCs to instructional practice, and learner reading outcomes. Lessons from the study will inform scaling of the intervention.

Authors

• Sahar Mohy-Ud-Din, VVOB - Education For Development

• Joshua Engelbrecht, JET Education Services

• Eleanor Hazell, JET Education Services

• Sharon Schroen, VVOB - education for development