Testing the Causal Links Between School Climate, School Violence, and School Academic Performance

This post is from Rami Benbenishty (Bar Ilan University) and co-authors

Many studies show that positive school climate and low levels of student bullying and violence are associated with improved academic performance. Currently, scientists and policy makers interpret this robust finding as evidence that climate improvement and violence reduction cause academic improvement. Nevertheless, this causal link has rarely been tested. In fact, no study has explored the possibility that improvements in academic performance would lead to climate improvement and violence reduction. The present study is one of the few to examine the causal link between school climate, school violence, and the school’s academic performance over time. We hypothesized that reductions in school violence and climate improvement would lead to schools’ overall improved academic performance.

In order to test these causal links, we employed a longitudinal design and the statistical method of cross-lagged panel autoregressive modeling designed to estimate causal links in nonexperimental studies. We analyzed the California Healthy Kids Survey (CHKS) at three points in time (2007-2009; 2009-2011; 2011-2013). The CHKS is a bi-annual statewide survey conducted in approximately 85% of all public school districts in California for the California Department of Education (CDE) by WestEd. We merged these data on each school with the CDE’s statewide Academic Performance Index (API) per each school. The total sample was 1,862 middle schools and 1,310 high schools at each point in time. 

Contrary to what we hypothesized, findings offer strong evidence that a school’s overall improvement in academic performance is a central causal factor in reducing violence and enhancing a school’s climate. The findings indicate that the improvement of academic performance has an impact on improved school climate and reduced violence. This pattern was true at each wave and for both high schools and middle schools. These findings go in the reverse causal direction of that which most researchers, policy makers, and educators assume.

These findings have implications for school reform and the need to include social-emotional, school climate and violence reduction programs into academic school reform as a combined strategy. These results would suggest that stand-alone climate or bullying reduction that are not tied with or part of academic improvement plans may not have any impact on academic improvement. A combined approach can improve all three.

This study should be replicated in additional contexts to examine its generalizability. It is important to conduct studies that combine quantitative and qualitative methods to explore how this causal process happens on the ground. We think that when schools make strong efforts to improve academics, they should address issues of climate and victimization as part of those academic reform efforts. Improved academic performance of students may also have positive impact on how the teaching staff perceive and relate to students.

The new Every Student Succeeds Act creates unique opportunities for states to develop their own accountability systems surrounding school reform. Based on the study findings, we recommend that California school accountability systems and school academic reform efforts should examine together climate, violence, and academics, and disseminate this information to each and every school, to help school improvement efforts. Over time, such an integrated accountability system will provide invaluable insights as to how academic performance could be improved, while ensuring that our students feel safe and supported, and school promote well being and pro-social behaviors.

The full study is in Rami Benbenishty, Ron Avi Astor, Ilan Roziner and Stephani L. Wrabel, “Testing the Causal Links Between School Climate, School Violence, and School Academic Performance: A Cross-Lagged Panel Autoregressive Model,” EDUCATIONAL RESEARCHER April 2016 vol. 45 no. 3 197-206.


  • 1 of 234

PACE thanks these funders and sponsors for their financial support