Investigating the comprehension iceberg: Developing empirical benchmarks for early-grade reading in agglutinating African languages

Abstract

Background: Reading development in agglutinating African languages is a relatively under-researched area. While numerous studies highlight the low comprehension levels among learners reading in African languages in South Africa, little has been done to probe beneath this ‘comprehension iceberg’ in terms of decoding skills.

Aim: As a tentative step towards benchmarking in African languages, we analyse the sub-components of reading across three languages (Northern Sotho, Xitsonga and isiZulu), to better understand the nature of alphabetic knowledge, word reading and fluency in these languages, how these relate to one another, and how accuracy and speed relate to comprehension.

Setting: Data was obtained from 785 Grade 3 learners across three African languages in three provinces in South Africa.

Methods: The early grade reading assessment (EGRA) framework was adapted to the written features of the three languages to assess letter-sounds, single-word reading, non-word reading, oral reading fluency (ORF) and oral comprehension.

Results: We present results on fluency, accuracy and comprehension and their interrelationships in these morphologically rich languages. While differences emerged between the conjunctive and disjunctive orthographies, strong relations occurred across the languages between letter-sound knowledge and word reading, word reading and oral reading fluency, and ORF and reading comprehension. Results suggest minimum thresholds of accuracy and ORF in each language, below which it is virtually impossible to read for meaning.

Conclusion: There is a strong need for language-specific norms and benchmarks for African languages. Preliminary minimum decoding thresholds for comprehension found in these three languages serve as a move in that direction.