This article focuses on the relation between student population characteristics and average test scores per school in the final grade of primary education from a dynamic perspective. Aggregated data of over 5,000 Dutch primary schools covering a 6-year period were used to study the relation between changes in school populations and shifts in mean test scores. Path analysis findings indicate that changes in student populations bring about instant changes in school averages. However, the impact of these changes is strongly mitigated by the effects of school results in the past. This reveals long-lasting effects of student population characteristics via past performance and explains the relative stability in raw output measures over time to a considerable extent.