crime in schools,
Smallest Geographic Unit:
Date of Collection:
- 1997--2006 (Part 1: 1997-1998 school year through the 2005-2006 school year)
- 2004--2007 (Part 2: Fall 2004, Spring 2005, Fall 2005, Spring 2006, Fall 2006, Spring 2007)
Unit of Observation:
Part 1: school by year,
Part 2: student by wave
Part 1: All middle schools and high schools in Broward County, Florida, Hillsborough County, Florida and Pinellas County, Florida from the 1997-1998 school year through the 2005-2006 school year.
Part 2: All middle school students in a sample of 14 middle schools and all high school students in a sample of 16 high schools across the Broward County, Florida, Hillsborough County, Florida and Pinellas County, Florida school districts between 2004 and 2007.
administrative records data,
Data Collection Notes:
With the exception of the student process survey data which are included in Part 2, data collected for the process analysis are not available as part of this data collection.
Users should be aware that, in order to replicate Exhibit 3.5 and Exhibit 3.6 in "Impact Evaluation of Youth Crime Watch Programs" (NCJ 226356) with the Student-Level Data (Part 2), they need to filter the dataset using the NOT_NEW_YCW "RESPONDANT WAS NOT ACTUALLY NEW WHEN TAKING THE BASELINE" and TYPE "SCHOOL TYPE--MIDDLE OR HIGH" variables. Even when accounting for the NOT_NEW_YCW and TYPE variables, users should also be aware that there still might be a small count case discrepancy between some figures referenced in the original final report and the case counts in the dataset that is being distributed as part of this data collection.
The purpose of this study was to assess both the school-level effects and the participant-level effects
of Youth Crime Watch (YCW) programs.
The school-level impact study general research question was: Do schools with YCW programs have lower levels of crime and drug use than schools without YCW programs? Recognizing that there are multiple ways of measuring crime, the specific research questions were: Do schools with YCW programs have fewer reported crimes, fewer major disciplinary actions taken against students, and a student population that feels safer, indicates it has been victimized less frequently, and believes that fewer crimes are committed than in comparable schools without YCW programs?
The participant-level impact study had two general research questions: (1) Does participation in YCW improve students' leadership skills, self-esteem, problem solving skills, and ability to work effectively with others?, and (2) Does participation in YCW cause students to be more willing to report different types of misconduct and crime?
Abt Associates conducted a four-year impact evaluation of Youth Crime Watch (YCW) programs in three Florida school districts (Broward, Hillsborough, and Pinellas Counties). School-based YCW programs implement one or more of a variety of crime prevention activities, including youth patrol, in which YCW participants patrol their school campus and report misconduct and crime. The evaluation collected both School-Level Data (Part 1) and Student-Level Data (Part 2).
The School-Level Data (Part 1) contain 9 years of data on 172 schools in the Broward, Hillsborough, and Pinellas school districts, beginning in the 1997-1998 school year and continuing through the 2005-2006 school year. A total of 103 middle schools and 69 high schools were included, yielding a total of 1,548 observations. These data provide panel data on reported incidents of crime and violence, major disciplinary actions, and school climate data across schools and over time. The research team conducted the school-level analysis by employing a quasi-experimental design which analyzed data from all the middle schools and high schools in the Broward, Hillsborough, and Pinellas school districts (by school and by school year) using a time series approach (i.e., before and after the introduction of YCW) and across groups of schools with a cross-sectional approach (i.e., schools with and without YCW). Abt Associates collected data on YCW programs from the 172 secondary schools via telephone surveys administered to YCW advisors by project staff. The Florida Department of Education provided data regarding reported incidents and disciplinary actions. The Broward, Hillsborough, and Pinellas School Districts provided school climate survey data via student responses to surveys on victimization and perceived safety in schools. The research team aggregated these data by school and school year. The years of school climate data vary by district.
The Student-Level Data (Part 2) were collected between 2004 and 2007 and are comprised of two major components: (1) self-reported youth attitude and school activities survey data that were administered to a sample of students in middle schools in the Broward, Hillsborough, and Pinellas School Districts as part of a participant impact analysis, and (2) self-reported youth attitude and school activities survey data that were administered to a sample of YCW continuing middle school students and YCW high school students in the same three school districts as part of a process analysis.
Concerning the participant impact analysis, surveys were administered to students in 14 middle schools across the Broward, Hillsborough, and Pinellas School Districts to compare students who had just signed up for YCW for the first time ("new YCW" students) with a comparison group of students not in YCW ("non-YCW" students) and assess the effect of participation on a number of different individual development measures and on willingness to report different types of misconduct and crime. Middle schools were used instead of high schools because the researchers believed, based on interviews with YCW adult leaders and Youth Crime Watch of America (YCWA), that it might be more likely to detect measurable impact at middle schools. Abt Associates developed four survey instruments for the participant impact analysis: a baseline and follow-up survey for both the new YCW group and the non-YCW comparison group. Project staff enrolled three cohorts of students in the study, one each at the beginning of the 2004-2005, 2005-2006, and 2006-2007 school years. The researchers surveyed students in the first 2 cohorts 3 times -- at baseline, 6 months after baseline, and 18 months after baseline. They surveyed students in the third cohort two times -- at baseline and six months after baseline.
A survey was also administered to YCW participants in high schools and YCW continuing middle school students to inform the process analysis. Student process surveys were initially conducted in late April and early May 2005. The research team conducted a second round of process surveys in late April and early May 2006.
For Part 2, a total of 3,386 completed surveys were collected including 1,319 "new YCW" student surveys, 1,581 "non-YCW" student surveys, and 486 "Pro" or "Process" student surveys.
This evaluation sampled schools and students in the Broward, Hillsborough, and Pinellas school districts. The three school districts coincide with the Florida counties with the same names. Broward County is centered around Fort Lauderdale, Hillsborough around Tampa, and Pinellas around St. Petersburg. All three school districts are among the 25 largest in the Nation. The research team chose a design that included a small number of school districts -- as opposed to a design with sites from across the nation -- both for cost efficiency reasons and to minimize cross area effects or cross district-level effects. Broward, Hillsborough, and Pinellas were selected because data provided by Youth Crime Watch of America (YCWA) at the beginning of the study indicated that these three districts had more YCW programs than any other school districts in the nation, with approximately 25 percent of middle and high schools in the three districts having submitted a site survey to YCWA. All three school districts are in Florida and are therefore subject to the same state-level requirements for reporting of incidents, disciplinary actions, student achievement data, and school demographic information.
The School-Level Data (Part 1) sample included all middle schools and high schools in the Broward, Hillsborough, and Pinellas school districts (172 schools), with the exception of combined elementary and middle schools, charter schools, and other special-purpose schools. Data were collected annually for a total of 9 years, yielding 1,548 cases.
Regarding the participant impact analysis which was comprised of baseline and follow-up surveys for both the new YCW group and the non-YCW comparison groups in the Student-Level Data (Part 2), the researchers selected both a school sample and a student sample. Concerning the school sample, the student-level survey was conducted in 14 middle schools across the Broward, Hillsborough, and Pinellas school districts. These schools were not randomly selected from among all schools with YCW programs in the three districts, nor did the research team attempt to influence a school's decision whether or not to run a YCW program. Rather, the researchers intentionally tried to recruit those middle schools considered to have active and well-respected YCW programs. Because no complete data existed on which schools had YCW programs or their characteristics, the research team relied on recommendations from staff at school district security offices and commanders of the law enforcement units that provide School Resource Officers for the relevant schools in the three school districts. YCW advisors at recommended schools in turn recommended other schools. A total of 28 schools were recommended, some more highly than others. After an initial screening interview with the YCW advisor, Abt Associates sent recruiting materials to principals at 16 schools, from which 11 schools were recruited for data collection in the 2004-2005 school year. Of the 11 schools, 2 dropped out of the study at the beginning of the 2005-2006 school year and were replaced by 3 other schools. In the end, 9 schools participated in all 3 years of data collection; 3 participated for 2 years, and 2 participated for 1 year. Of the 14 schools, 5 were from Broward, 8 were from Hillsborough, and 1 was from Pinellas.
Concerning the student sample, research staff attempted to enroll into the study all new YCW students at the start of the 2004-2005, 2005-2006, and 2006-2007 school years who were in either the 6th or 7th grade. The research team did not include 8th graders, because they would be at different schools at the time of the 18 month follow-up survey, making them difficult to re-survey. For the non-YCW comparison group, Abt Associates enrolled in the study all students in either one or two classrooms in each school. (The research team was able to identify YCW members in selected classrooms because they included a question about YCW membership in the surveys.) Classrooms were selected in
consultation with the YCW advisor and the school administration. The final sample included 1,319 New YCW surveys and 1,581 Non-YCW surveys.
For the process analysis component of the Student-Level Data (Part 2), a subset of 16 high schools in the 3 districts was selected. The research team attempted to sample all YCW high school students within these 16 high schools in late April and early May 2005 and in late April and early May 2006, yielding 407 responses. Additionally, continuing middle school students were sampled during the same time period as part of the process analysis, yielding 79 responses.
Mode of Data Collection:
paper and pencil interview (PAPI),
Questionnaires administered to YCW advisors by project staff via telephone surveys of all secondary schools collected YCW program data including implementation status and program characteristics (Part 1).
The Florida Department of Education provided data on measure of crime and the level of school safety (including reported incidents and in-school and out-of-school suspensions) as well as other school characteristics (school size, per pupil expenditures, standardized test results, the percentage of students eligible for free or reduced lunch, and dropout and graduation rates) (Part 1).
The Broward, Hillsborough, and Pinellas County School Districts collected data on measures of crime and the level of school safety via student surveys on school climate (Part 1).
Self-reported surveys administered to students at their schools (Part 2).
Description of Variables:
The 138 variables in the School-Level Data (Part 1) include Youth Crime Watch (YCW) program data, measures of crime and the level of school safety in a school, and other school characteristics. YCW program data variables include implementation status (whether a school had a YCW program) and program characteristics (details of the YCW program operations). Measures of crime and the level of school safety include reported incidents, in-school and out-of-school suspensions, and student surveys on school climate. Other school characteristics include school size, per pupil expenditures, standardized test results, the percentage of students eligible for free or reduced lunch, and dropout and graduation rates.
The 99 variables in the Student-Level Data (Part 2) include two groups of questions for assessing participant impact: (1) how the respondents felt about themselves, in terms of self esteem, leadership skills, ability to work with others, and problem-solving ability, and (2) whether the respondent would report certain types of problems or crimes that they observed at the school. Part 2 also includes administrative variables including group, cohort, wave, and school identification number and demographic/background information (grade, race/ethnicity, gender). Other variables in Part 2 pertain to the respondent's involvement in school-based extracurricular activities, involvement in volunteer and community activities, attitudes toward school, attitudes about home environment, future education plans, attitudes toward the YCW advisor, attitudes about effects of YCW on school as a whole and on respondent personally, duration and extent of participation in YCW, and reasons for joining YCW and reasons for remaining in YCW.
Presence of Common Scales:
Part 1: Several Likert-type scales were used. Part 2: Several Likert-type scales were used. The principal investigator also selected items from the California Healthy Kids Survey and the Review of Personal Effectiveness and Locus of Control (ROPELOC).
Extent of Processing: ICPSR data undergo a confidentiality review and are altered when necessary to limit the risk of
disclosure. ICPSR also routinely creates ready-to-go data files along with setups in the major
statistical software formats as well as standard codebooks to accompany the data. In addition to
these procedures, ICPSR performed the following processing steps for this data collection:
Created variable labels and/or value labels.
Standardized missing values.
Checked for undocumented or out-of-range codes.