The purpose of this research was to build and test a conceptual model of the complex interrelationships between students' learning in science (learning approaches and self-regulation), their reading comprehension, question-asking in class and science achievement. These variables were measured by means of a test and a series of questionnaires administered to 604 ninth-grade students, and the data collected were analysed using a correlational, cross-sectional design. Results of a path analysis indicated that (a) students' self-regulated and intentional knowledge-constructing activity (self-regulated strategy use, deep approach and knowledge-building) were what chiefly accounted for their question-asking in class; (b) question-asking (high and low levels) was related directly to reading comprehension and indirectly, through its contribution to the this, to academic achievement; (c) reading comprehension was directly and negatively associated with surface approach and indirectly and positively related to deep approach and knowledge-building; and (d) some of these variables, particularly reading comprehension, accounted for academic achievement in science. This model explained nearly 30% of the variance in academic achievement and provided a substantial and distinctive insight into the web of interrelationships among these variables. Implications for future research and science teaching and learning are discussed (e.g. the importance of supporting students' efforts to learn science in a meaningful, active and self-regulated way and of improving their reading comprehension).