This study investigates how different types of prior knowledge influence student achievement and how different assessment measures influence the observed effect of prior knowledge. We introduce a model of prior knowledge that distinguishes between different types of prior knowledge and uses different assessment measures to assess different types of knowledge. The sample consists of 202 mathematics students who completed the prior knowledge test during the first lesson. The student achievement was measured by the final grade on the course. The results indicate that the type of prior knowledge makes a difference: The measures assessing procedural knowledge predicted the final grades best whereas measures assessing declarative knowledge did not predict final grades. Additionally, previous study success was the best predictor of student achievement. These results are discussed in relation to assessment measures and their implications for practice.