Association for Public Policy Analysis & Management

Education policy in the United States is designed and implemented with the goal of improving student achievement with little concern about potential unintended consequences on children’s mental health. While student achievement and mental health are typically positively correlated, various education policies aimed at boosting student achievement could theoretically worsen child mental health by labeling children as relatively low performing (e.g., tracking), reducing time for recreation (e.g., extended school days or year), or lowering student-relative-to-peer ability (e.g., desegregation). Any such negative mental health effects could offset positive achievement effects and negatively impact adult outcomes. Identifying whether prominent education reform strategies affect mental health is necessary to holistically evaluate such reforms, and help policy-makers carefully consider the design of future education policy.

In this paper, I examine the effects on child mental health of one of the most prominent and widespread education policies of the past three decades – the expansion of statewide standardized testing and test-based accountability. During this period in which nearly annual testing has become ubiquitous across the country, rates of childhood anxiety and mental health issues increased (Boyle et al., 2011). Many have suggested a causal link between these two phenomena: Psychology research documents harmful test-anxiety can occur due to standardized test-taking, particularly high-stakes testing (e.g., Segool et al., 2013). Children, parents, school staff, and the media have argued that expanded testing has increased anxiety levels, hurt mental health, and reduced student enjoyment of school (e.g., Washington Post, 2011; USA Today, 2015). Despite these claims, while numerous studies have examined the effects of accountability on student achievement and school and district behavior, few have examined effects on student mental health.

To examine the effect of statewide testing on children’s mental health, I create a dataset documenting which states required statewide testing in which grades from 1985 to 2012. I use mental health measures for 9,260 children in public elementary and secondary school during this period from the Children of the National Longitudinal Survey of Youth–1979 (CNLSY79). I conduct a difference-in-differences style analysis, exploiting plausibly exogenous variation in student exposure to state-mandated testing across states, grades, and student cohorts. I find that cumulative exposure to testing, defined as the fraction of grades through the present in which a child was exposed to testing, increases anxiety and reduces children’s satisfaction with school. However, I find no detectable effects of cumulative testing exposure on broader measures of overall mental health among children or adolescents, such as measures of behavioral issues, self-esteem or depression. The results confirm anecdotal evidence that testing has increased anxiety among children, but at the same time alleviates concerns that these increases led to meaningful deteriorations in children’s mental health.