The main objective of this research was to explore how teachers teach reading comprehension to Grade 3 Tshivenda-speaking learners. The research was prompted by the low performance of reading amongst these Grade 3 learners in this area. This study was a qualitative study. The paradigmatic position of the study was the interpretive paradigm. Data was collected through individual teachers’ interviews, focus group interviews, classroom observations and content analysis. Interviews and focus group discussions were taped, transcribed, analysed, and data was categorised into themes. Three schools, each with two Grade 3 classes, were selected. Learners were involved because the main aim was to observe teachers teaching reading comprehension to them. This study did not attempt to measure learner performance, but investigated the teaching of reading comprehension to Grade 3 Tshivenda-speaking learners. The aim of this was to investigate how teachers teach reading comprehension in their classrooms. The study developed a conceptual framework from the literature review, consisting of five phases. These phases were identified and developed as it became clear that, in order to teach reading comprehension effectively, various parties involved in schools need to be actively involved in the different phases when teaching reading comprehension education of learners. In addition, the learning and developmental theories consulted in this study were related to the conceptual framework. The research implies that reading comprehension can be taught and learnt, which further supports the development of the conceptual framework for this study. The findings of the study revealed that the participating teachers had limited understanding of reading comprehension strategies. They regarded teaching comprehension as a challenge as they do not know what comprehension strategies are and how to apply them. Additionally, they are stressed, confused and frustrated, because learners cannot read and understand the text. Furthermore, responses showed that teachers are uncertain of how to teach reading comprehension. Teachers spent little time on reading comprehension with learners, with no uniform approach amongst them on how to teach reading comprehension. Guided practice and time to practise comprehension strategies were absent in classrooms. The teachers also showed a lack of theoretical knowledge and practical experience about teaching comprehension strategies, which seems to result in teachers’ developing a negative attitude towards their learners (who struggle to read). Interviews and classroom observations revealed that, although the participating teachers said they understood what reading comprehension was, there was no correlation between what they said and what they did in practice in their classrooms. The results measured against Zimmerman’s (1998) applied social model of self-regulated learning, show that teachers lack the theoretical knowledge of teaching reading comprehension. In addition, participants were not satisfied with the intervention strategies and policies provided by the Department of Education. They felt neglected because guidelines were only in English and not in African languages, like Tshivenda. As such, no guidelines to teachers written in Tshivenda with Tshivenda examples exist. Workshops had not been helpful to Tshivenda teachers either. The study also revealed a lack of learning support materials, no variety of reading materials for learners and, in some schools, no readers at all. Lastly, the study shows that these teachers do not have access to research literature on the teaching of reading comprehension and rely on their own experience.