Publication Date: 2003-12-00
Author: Larsen, Eric
Source: ERIC Clearinghouse on Urban Education
Violence in U.S. Public Schools: A Summary of Findings. ERIC Digest.
A recent report from the National Center for Education Statistics, "Violence in U.S. Public Schools," deals with a subject that has received increasing attention in recent years as a result of several highly publicized cases of violence in public schools. The report provides an initial analysis of data gathered in the 2000 School Survey on Crime and Safety (SSOCS), the latest in a series of government-sponsored studies of school safety that goes back to the 1970s. A follow-up SSOCS will be conducted in 2004. Building on the experience of previous studies, a wide-ranging survey was designed that asked principals about incidents of crime and violence, disciplinary problems and actions, use of security measures, policies and practices relating to safety, and violence prevention programs and activities in their schools. In addition, principals were asked to provide information about particular school characteristics.
The survey was sent to the principals of 3,314 schools, which make up a nationally representative cross-sectional sample of regular elementary, middle, secondary, and combined public schools in all 50 states and the District of Columbia. In all, there were 2,270 responses (a response rate of 68.5 percent), which provided a basis for estimating national totals. The salient points for teachers and administrators in this analysis of survey findings have to do with the relationships that exist between school and student characteristics and the incidence of violence and serious violence in schools; this digest reviews the key findings with respect to those relationships.
SCHOOL AND STUDENT CHARACTERISTICS
The present analysis is concerned with a section of the survey that lists specific kinds of crime and violence and asks principals how many incidents of each occurred at their school in the 1999-2000 school year. The report focuses on two categories, crimes that constitute violence and a subset of these that constitute serious violence: rape, sexual battery other than rape, physical attack or fight with or without a weapon (in serious violent incidents, with a weapon), threat of physical attack with or without a weapon (in serious violent incidents, with a weapon), and robbery with or without a weapon.
To document the relationship between school characteristics and school violence, the report considers: school demographic characteristics, school administrative practices, and general school disorder. For school demographic characteristics, the following are examined: school level (elementary, middle, secondary, combined); enrollment; urbanicity (city, urban fringe, town, rural), and neighborhood crime. In terms of administrative practices, the report looks at student-teacher ratios, whether and how often students changed classrooms between class periods, and whether the school had law enforcement personnel. The school disorder characteristics considered for the report include reports of serious discipline problems, transfers, schoolwide disruptions, and unexcused absences.
In identifying characteristics that are related to school violence, the report also looks at the student population of schools, specifically at the percent of: minority students, students eligible for free or reduced-price lunch, students scoring below the 15th percentile on standardized tests, students likely to attend college, students for whom academic achievement is important, and males.
Findings on the 2000 School Survey on Crime and Safety indicated that a significant number of the characteristics studied in the analysis, including size of school enrollment, absenteeism, reported crime level of the school neighborhood, and percentage of males in the student population, appeared to be positively related to the incidence of school violence. It was not clear, however, which of these characteristics, examined by itself without the influence of others, had a positive relationship to the prevalence of school violence. In order to control for the influence of school and student characteristics on one another, the report analyzes all of the factors simultaneously in relation to violence and serious violence.
SCHOOL AND STUDENT CHARACTERISTICS AND THE PREVALENCE OF VIOLENCE
When each group of school and student characteristics are examined together, it is found that just six out of the seventeen factors accounts for the prevalence of school violence.
Among the school demographic characteristics, two remain: level and location.
* Secondary schools are more likely to report at least one incident of violence than elementary and middle schools, with 92 percent of secondary schools, 87 percent of middle schools, and 61 percent of elementary schools reporting a violent incident. For serious violence, secondary and middle schools (29 percent) report the same incidence of serious violence, while 14 percent of elementary schools report at least one serious incident.
* Similarly, compared to city schools, those on the urban fringe are less likely to report a violent incident, while there are no differences in reports of violent incidents between city schools and those located in towns or rural areas.
After controlling for other characteristics, the only school management practice that remains associated with reports of violence is the number of classroom changes students make during a typical school day; as the number increases, the probability of experiencing a violent incident also increases.
Two characteristics associated with school disorder--the number of schoolwide disruptions and the number of serious discipline problems--remain positively related to the likelihood of a violent incident.
* Schools that reported at least one widely-impacting disruption, such as a bomb threat, were more likely to also report an incident of violence or serious violence than schools that had no schoolwide disruptions.
* Schools in which three or more serious disciplinary problems occurred were more likely to experience violence or serious violence than schools that had two or fewer incidents of disciplinary problems to report.
Finally, all the student population characteristics, including percentage of minority students, fall away except the importance of academic achievement, which is negatively related to the prevalence of violent incidents. As the percentage of students who principals believe consider academic achievement to be very important increases, the likelihood of schools reporting a violent incident decreases.
PREVALENCE OF SERIOUS VIOLENCE
As with reports of school violence, many characteristics of schools and students seem to relate to serious school violence. In the same statistical analysis performed for incidence of school violence, all factors associated with serious violence were analyzed simultaneously. In this analysis, only five characteristics continue to show a positive relationship to serious violence.
* School enrollment size is positively related to the prevalence of serious violence.
* The only student characteristic that remains associated with serious violence is the percentage of males in the student population; as this increases, reports of serious incidents of violence also increase.
* In terms of school disorder, the number of schoolwide disruptions and the number of serious discipline problems reported both relate positively to the likelihood of one or more incidents of serious violence.
* In connection with urbanicity, rural schools are less likely than city schools to report an incident of serious violence, while no differences are found between schools in cities, the urban fringe, or towns.
PATTERNS OF SCHOOL VIOLENCE
Reported incidents of school violence are distributed among schools in similar patterns to crime and violence in the general society. Findings from the 2000 SSOCS indicate that a disproportionately small number of schools reported a large amount of violence and serious violence. Findings show that 18 percent of schools accounted for 75 percent of the reported incidents of violence, while 6.6 percent accounted for 50 percent. With respect to serious violence, 1.9 percent of schools reported 50 percent of the incidents.
To determine the characteristics of schools with high levels of violence, schools were divided into three groups: those with no violence, those with low to moderate levels (1-59 incidents), and those with high levels (60 or more incidents). These cutoffs were chosen so that schools with high levels of violence were defined as those that accounted for 50 percent of the reported incidents. When these high-level schools were compared to those with no violence or low to moderate levels, the following factors were found to be significant: school level, enrollment size, urbanicity, crime where students live, number of classroom changes, number of serious discipline problems, and number of schoolwide disruptions.
To determine the characteristics of schools with high levels of serious violence, schools were again divided into three groups: those with no serious violence, those with low to moderate levels of serious violence (1-8 incidents), and those with high levels of serious violence (9 or more incidents). A somewhat different list of characteristics was found to account for high levels of serious violence: enrollment size, students below the 15th percentile on standardized tests, student-teacher ratio, number of serious discipline problems, number of students transferring from the school, and number of schoolwide disruptions.
During the 1999-2000 school year, 71 percent of U.S. public schools reported experiencing at least one violent incident, and the total number of incidents reported was 1,466,395. With respect to serious violence, 20 percent reported at least one incident, and a total of 60,719 incidents were reported. Correspondingly, 36 percent of schools reported at least one violent incident to the police and a total of 256,876 incidents, whereas 15 percent of schools reported at least one incident of serious violence to police and a total of 34,281 incidents.
It is important to note that this survey finds, as previous research has concluded, that academic engagement and school discipline are significantly related to the prevalence of school violence. To understand the nature of school violence, teachers and administrators may analyze the characteristics of schools with low levels of violence, high levels of violence, and high levels of serious violence. For example, schools with high levels of serious violence also have large enrollments, high student-teacher ratios, and high percentages of students performing well below academic standards.
Comparing and analyzing these outcomes may provide information that focuses administrative practices and school policies on the factors most clearly associated with school violence.
National Center for Education Statistics. (2003). Violence in U.S. public schools: 2000 School Survey on Crime and Safety (NCES-2004-314). Statistical Analysis Report. Washington, DC: Author.