crime in schools,
reactions to crime,
Smallest Geographic Unit:
Date of Collection:
Unit of Observation:
All public elementary and secondary schools and school
districts in the United States excluding schools in the outlying
United States territories, ungraded schools, and those with a high
grade of kindergarten or lower.
Data Collection Notes:
Users interested in obtaining a restricted-use data file containing more detailed information can file an application for the data with the Institute of Education Sciences/National Center for Education Statistics (IES/NCES), Restricted-data Licenses.
The SSOCS is the primary source of school-level data on crime and safety for the United States Department of Education. It provides nationwide estimates of crime, discipline, disorder, programs, and policies in public schools. Data on crime, violence, and disorder in the nation's schools are collected to provide policymakers, parents, and educators with the information necessary to identify emerging problems and to gauge the safety of American schools.
SSOCS:2000 was conducted as a mail survey with telephone follow-up. Six months before the onset of data collection, NCES began working with the school districts of sample schools that required prior approval to participate in the survey. In late February 2004, advance letters were sent to school administrators of sample schools that included the date of the first questionnaire mailing and a toll-free number to call with any questions. Approximately one week later, SSOCS questionnaires were mailed to administrators with a cover letter describing the importance of the survey and a brochure providing additional information about it. On the same day the questionnaires were mailed to schools, letters were sent to sample district superintendents and the Chief State School Officer of each state to inform them that schools within their districts and states, respectively, had been selected to participate in SSOCS:2004. The letters included information about the survey and were accompanied by a copy of the questionnaire and brochure that were sent to schools. The letter was not designed to ask for permission from these officials to participate in the survey but rather was designed by NCES as a vehicle to enhance participation. Starting approximately one week after the first questionnaire mailing, follow-up telephone prompts were used to verify that the questionnaire was received and to encourage survey response. As an alternative to replying by mail, data were also accepted by fax submission and over the telephone. Data collection ended on June 4, 2004. Returned questionnaires were examined for quality and completeness using both manual and computerized edits. Out of 227 items on the questionnaire, 101 were identified as key (critical) items. If the survey had more than 40 percent of all items missing or more than 20 percent of critical items missing, the respondent was recontacted to resolve issues related to the missing data. In cases where the recontacts failed to produce a satisfactory resolution, imputation was used to resolve data quality issues for questionnaires that had at least 60 percent of all items and 80 percent of critical items completed. Schools whose questionnaires did not meet the 60/80 criterion and for which recontact was not successful were reclassified as nonrespondents.
The sampling frame for SSOCS:2000 was constructed from the public school universe file created for the 2000 Schools and Staffing Survey (SASS). However, only the approximately 81,000 regular schools (excluding schools in the outlying United States territories, ungraded schools, and those with a high grade of kindergarten or lower) in the 1997-1998 NCES Common Core of Data (CCD) Public School Universe File within the SASS frame were eligible for the study. The SASS frame was derived primarily from the 1997-1998 CCD, which includes charter schools. These CCD schools, including charter schools, were included in the SSOCS:2000 study. (The SASS frame also includes a supplement made up of additional charter schools as well as a small number of Bureau of Indian Affairs and Department of Defense schools not represented in the 1997-1998 CCD file. Schools from this supplement were not included in the SSOCS:2000 study.)
Since the schools were selected with unequal probabilities, sampling weights are required for analysis to inflate the survey responses to population levels. The use of sampling weights is essential not only for calculating aggregate statistics (e.g., the total number of crimes of a particular type in United States public schools would be greatly understated without the use of weights) but also for other statistics such as percentages. For example, if a particular school policy was less common at elementary schools than at other schools, then the fact that elementary schools were chosen with a lower probability than other schools could lead to overstating the percentage of schools with that policy if weights are not used. The FWT (final sample weight) provided with the data file takes into account three components. The essential component of the sampling weight is the "base weight," which is defined to be the reciprocal of the probability of selecting a school for the sample. In addition, the weight includes a correction for unit nonresponse (i.e., sampled schools that did not respond to the survey) and a poststratification adjustment based on the Public School Universe File of 1998-1999 Common Core of Data. To compensate for unit nonresponse in SSOCS:2000, the base weights were adjusted by applying appropriate nonresponse adjustment factors within groups of schools (cells) expected to be homogeneous in terms of response propensity. The nonresponse adjustment cells were formed by applying a CHAID (Chi-square Automatic Interaction Detector) analysis. CHAID is a classification technique that divides a population or sample based on available auxiliary information into homogeneous subgroups with respect to a target characteristic. The target characteristic used in SSOCS:2000 was response status, and the characteristics that were used as auxiliary information were instructional level, enrollment size of school, region, type of locale, minority status, free lunch category, pupil to teacher ratio, district enrollment size, the ratio of guidance counselors to teaching staff in district, and the ratio of graduates to dropouts in district. Based on the CHAID analysis, a total of 49 adjustment cells were formed and the adjustment factors were computed as the ratio of the number of schools sampled to the number of schools responded in each cell. The base weights were multiplied by the computed adjustment factors to derive the nonresponse adjusted weights. A final poststratification (raking ratio) adjustment was applied to the nonresponse adjusted weights. The purpose of the poststratification was to ensure that the estimated counts of schools by various characteristics are in agreement with the known counts obtained from the most recent (1998-1999) CCD file. The counts of schools in the CCD file by instructional level and enrollment size, and by instructional level and type of locale were used in the poststratification adjustment. By forcing the weighted sample counts to agree with known current population counts, poststratification may help reduce any bias associated with the noncoverage of new schools in the SSOCS:2000 sample.
Mode of Data Collection:
mail survey with telephone followup
Unweighted response rate: 68 percent Weighted response rate: 70 percent
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:
Performed consistency checks.
Performed recodes and/or calculated derived variables.
Checked for undocumented or out-of-range codes.