This study investigated the relationships of teacher-directed approaches with science achievement in Australian schools. The data for this study were drawn from the Program for International Student Assessment (PISA) 2015 database and analysed using multilevel modelling (MLM). MLMs were estimated to test the contribution of each item to students' science achievement scores and to estimate the mediation effect of teacher explanations on these relationships. Only explicit, teacher-directed practices demonstrated a significant, positive association with science achievement. The positive, significant nature of the item 'the teacher explains scientific ideas' (B = 29.61, p < 0.001) suggested that this practice should take place in all science lessons. In the mediation model, the explicit, teacher-directed approaches in the inquiry scale revealed a significant "indirect" effect on science achievement, through the process of the teacher explaining scientific ideas. This indicated that effective explanations also underpin other instructional approaches such as contextualised science learning. These findings, accompanied by an analysis of the teacher-directed items and their relationships to science outcomes, give teachers and policymakers clear guidance regarding the effective use of instructional explanations in the science classroom.