Curricular policy as a collective effects problem: A distributional approach

• Policies enrolling more 8th grade students in Algebra have iatrogenic effects.
• This is one example of a collective effect, when population and individual effects do not align.
• Collective effects are poorly understood but vitally important for future policy research.
• Distributional approaches hold promise for better understanding population level effects.

Current educational policies in the United States attempt to boost student achievement and promote equality by intensifying the curriculum and exposing students to more advanced coursework. This paper investigates the relationship between one such effort – California’s push to enroll all 8th grade students in Algebra – and the distribution of student achievement. We suggest that this effort is an instance of a “collective effects” problem, where the population-level effects of a policy are different from its effects at the individual level. In such contexts, we argue that it is important to consider broader population effects as well as the difference between “treated” and “untreated” individuals. To do so, we present differences in inverse propensity score weighted distributions investigating how this curricular policy changed the distribution of student achievement. We find that California’s attempt to intensify the curriculum did not raise test scores at the bottom of the distribution, but did lower scores at the top of the distribution. These results highlight the efficacy of inverse propensity score weighting approaches for examining distributional differences, and provide a cautionary tale for curricular intensification efforts and other policies with collective effects.