Association for Public Policy Analysis & Management

This paper exploits a period of industrial action in the UK in 2009/10 where one third of primary schools stated that they would not teach towards or administer the national exams of age 11 students. Using administrative data on the entire English state school population over seven cohorts, we track the progress of students that did and didn’t experience “teaching to the test” at this age. Using a classic difference-in-difference design, we find that this industrial action had small but lasting positive impacts on teacher assessments of student performance and externally graded high stakes examinations. However, we also provide evidence of the boycotting schools having higher growth in the years prior to the boycott. The paper then shows that propensity score matching can be used to generate a sample of comparable schools with common trends. Using this reduced sample we find no long run positive impact from the industrial action and small negative impact on end of high school math and science test scores.