EVALUATING THE NATURE OF CHANGES IN SCHOOL VIOLENCE DURING “THE GREAT AMERICAN CRIME DECLINE”

Open Access
- Author:
- Sanders, Amber Nicole
- Graduate Program:
- Criminology
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- May 18, 2018
- Committee Members:
- Eric P Baumer, Dissertation Advisor/Co-Advisor
Eric P Baumer, Committee Chair/Co-Chair
Jeremy Staff, Committee Member
David M Ramey, Committee Member
Paul Morgan, Outside Member - Keywords:
- crime trends
crime decline
school violence - Abstract:
- Since the 1970s, a wealth of research looking at the cross-sectional patterns and predictors of school violence has accumulated, but there has been little empirical research devoted to exploring trends of violence within the school setting. In contrast, the current research evaluates the nature of changes in school violence between 1990 and 2002, a period that corresponds to what some have called the “Great American Crime Decline” (Zimring 2007). The current research uses data from the National Education Longitudinal Study of 1988 (NELS) and Education Longitudinal Study of 2002 (ELS) to evaluate the nature of changes in school violence from the early 1990s to the early 2000s. The study seeks to answer two questions. First, what is the overall pattern of change in school violence since the early 1990s? This question considers the overall pattern of change for school violence and whether the change varied by students’ socio-demographic characteristics (e.g. sex, race, and class). Second, what factors are associated with observed changes in school violence over that period? This question considers how various social and self-controls have changed over time and how such controls may relate to changes in school violence from 1990 to 2002. This research finds evidence of statistically and substantively significant temporal variation in the prevalence of fighting at school over the course of the 1990s, akin to aggregate-level declines in violence during this time period. Multivariate logistic regression models indicate a 20.5% reduction in the prevalence of fighting at school from 1990 to 2002. Further, analyses show that the reduction in fighting at school was shared across socio-demographic groups (i.e. sex, race/ethnicity, and socioeconomic status), indicating a robust cohort effect; however, the magnitude of change did vary across subgroups. Specifically, the decline in fighting was largest among males, which is constant with recent evidence from the crime trends literature (Lauritsen, Heimer, and Lynch 2009; Lauritsen, Rezey, and Heimer 2014). Conversely, this research does not find evidence that the decline in fighting prevalence was greater among racial minority youth or those of lower SES backgrounds, contrary to expectations based on evidence from the crime trends literature. Next, this research finds increases in student educational expectations and intimacy of parental communication significantly accounted for portions of the observed temporal variation in fighting prevalence. Such findings are consistent with theoretical expectations that strengthening social bonds to various institutions can help account for declines in violence (LaFree 1998; Roth 2010). Results also indicate that commitment to school and student effort decreased from one cohort to the next, contrary to expectations (Eisner 2014; LaFree 1998; Roth 2010). Despite these two indicators decreasing, both are significantly related to fighting, and therefore appear to suppress the decline in fighting at school.