U.S. flag

An official website of the United States government, Department of Justice.

NCJRS Virtual Library

The Virtual Library houses over 235,000 criminal justice resources, including all known OJP works.
Click here to search the NCJRS Virtual Library

A Meta-Analysis of Longitudinal Partial Correlations Between School Violence and Mental Health, School Performance, and Criminal or Delinquent Acts

NCJ Number
304817
Journal
Psychological Bulletin Volume: 147 Issue: 2 Dated: 2021 Pages: 115-133
Author(s)
J. R. Polanin; et al
Date Published
2021
Length
19 pages
Annotation

This article reports a meta-analysis of longitudinal partial correlations between school violence and mental health, school performance, and criminal or delinquent acts.

Abstract

The daily challenges resulting from all types of school violence—such as physical aggression, bullying, peer victimization, and general threats—have the potential to affect, longitudinally, students’ mental health, school performance, and involvement in criminal or delinquent acts. Across primary and secondary studies, however, variation in how and how much school violence relates to these outcomes, has persisted. The purpose of this systematic review and meta-analysis, therefore, was to clarify this uncertainty by synthesizing the longitudinal relations. We conducted exhaustive searching procedures, implemented rigorous screening and coding processes, and estimated an underused effect size, the partial correlation from multiple regression models, before estimating a random-effects meta-analysis using robust variance estimation. We meta-analyzed 114 independent studies, totaling 765 effect sizes across 95,618 individual participants. The results of the overall analyses found a statistically significant longitudinal relation between school violence, in any role, and the aggregated outcome variables (rp = .06). Given that this effect size inherently controls for multiple potential confounding covariates, we consider the relation’s magnitude clinically meaningful. We end by discussing ways practitioners and researchers may use these analyses when implementing prevention programming and how the field of meta-analysis should more frequently utilize the partial correlation. (Publisher Abstract Provided)