National Crime Victimization Survey: MSA Public-Use Data, 2000-2015 (ICPSR 38321)

Version Date: Mar 21, 2022

Cite this study |

Principal Investigator(s):
United States. Bureau of Justice Statistics

Series:

https://doi.org/10.3886/ICPSR38321.v1

Version V1

Slide tabs to view more

Alternate Title

NCVS MSA PUF, 2000-2015

Summary

Through a cooperative agreement, RTI International worked with the Bureau of Justice Statistics (BJS) to create public-use files of victimization data for the 52 largest metropolitan statistical areas (MSAs) covering the 2000-2015 survey years. The National Crime Victimization Survey (NCVS) is one of two national indicators of crime in the U.S. Historically, NCVS estimates of crime were not available at the state or local level because, prior to 2016, the NCVS sample was designed to exclusively produce national estimates. It is important to be able to understand victimization and victimization risk at the local level to inform and improve crime prevention efforts, investigation and victim response practices, and the location and mix of victim services.

To protect respondent confidentiality, with a few exceptions, subnational identifiers are traditionally not included on NCVS public-use files. Instead, information required to conduct analyses of crime at subnational levels must be accessed through a Federal Statistical Research Data Center (FSRDC) by obtaining Special Sworn Status from the U.S. Census Bureau. To provide a greater number of analysts with access to NCVS subnational data, in 2007 the Bureau of Justice Statistics (BJS) released a public-use file containing person- and incident-level data from 1979-2004 for the "core" counties (i.e., self-representing PSUs) within the 40 largest metropolitan statistical areas (MSAs). To accommodate interest from analysts and other interested parties in updating the file with more recent data, BJS has created public-use files for the 52 largest MSAs covering the 2000-2015 survey years.

The 52 MSAs included on these files are those with a 2015 population of 1 million or more persons and an average annual NCVS sample size of at least 250 persons during the period of 2006-2015. While some of the MSAs from these files were also included on the 1979-2004 version, the definitions used to define MSAs are not the same. The 1979-2004 files were based on the "core" counties that were common to the MSA definitions determined by the Office of Management and Budget (OMB) for the 1970-, 1980-, and 1990-based NCVS sample designs. For the current files (i.e., 2000-2015), MSA definitions are based on the most recent delineation files available from OMB at the time of data collection for each survey year included on the files.

Citation

United States. Bureau of Justice Statistics. National Crime Victimization Survey: MSA Public-Use Data, 2000-2015. Inter-university Consortium for Political and Social Research [distributor], 2022-03-21. https://doi.org/10.3886/ICPSR38321.v1

Export Citation:

  • RIS (generic format for RefWorks, EndNote, etc.)
  • EndNote

Funding

United States Department of Justice. Office of Justice Programs. Bureau of Justice Statistics

Subject Terms

Geographic Coverage

Smallest Geographic Unit

Metropolitan statistical area (MSA)

Distributor(s)

Inter-university Consortium for Political and Social Research
Hide

Time Period(s)

2000 -- 2015

Date of Collection

2000 -- 2015
Hide

Sample

Sampling Procedure

Each month the U.S. Census Bureau selects respondents for the NCVS using a "rotating panel" sample design. Households are randomly selected and all age-eligible individuals become part of the panel. Once in the sample, respondents are interviewed every six months for a total of seven interviews over a three-year period. Typically, the first interview is face-to-face and the remainder are by telephone. After the seventh interview the household leaves the panel and a new household is rotated into the sample.

Sample Design

The NCVS sample consists of housing units drawn from the decennial census using a stratified multi-stage cluster design. To take advantage of the availability of data from the most recent census, sample selection from the most recent census is phased in at approximately mid-decade. Prior to that, sample selection is based on the census before the most recent one. For example, prior to 1995, the sample was drawn from the 1980 decennial Census. From January 1995 until December 1997, the sample drawn from the 1990 Census was phased in. From January 1998 until approximately 2005 the complete NCVS sample was drawn from the 1990 Census. In 2005 samples from the 2000 Census began to be phased in. In 2015, samples from the 2010 Census began to be phased in.

The first step of the sample design involves dividing the United States into first-stage sample units or primary sampling units (PSUs) consisting of large metropolitan areas, counties, or groups of bordering counties. The next step is dividing the PSUs into sampling strata. Large PSUs were included in the sample automatically and each is assigned its own stratum. These PSUs are considered to be self-representing (SR) since all of them were selected. The remaining PSUs, called non-self-representing (NSR) because only a subset of them was selected, were combined into strata by grouping PSUs with similar geographic and demographic characteristics, as determined by the decennial Census used to design the sample. One PSU is selected from each strata, using probability proportional to size, to represent the others included in the strata.

The sample of housing units within a PSU was selected in two stages. These stages were designed to ensure a self-weighting probability sample of housing units and group-quarters dwellings within each of the selected areas. The first stage consisted of selecting a sample of Enumeration Districts (ED's) from designated PSUs. ED's were systematically selected proportionate to their size.

In the second stage, each selected ED was divided into segments (clusters of about four housing units each), and a sample of segments was selected. The segments were formed from the list of addresses compiled during the relevant censuses. However, procedures allow for the inclusion of housing constructed after each decennial Census enumeration. A sample of permits issued for the construction of new residential housing is drawn, and for jurisdictions that do not issue building permits, small land-area segments are sampled. These supplementary procedures, though yielding a relatively small portion of the total sample, enable persons living in housing units built after each decennial Census to be properly represented in the survey. In addition, units in group quarters known as "special places" were also selected in special place segments. These units, such as boarding houses and dormitories, constitute a small portion of the total sample.

Time Method

Longitudinal: Panel

Universe

The target population is individuals 12 years of age and older living in households and group quarters within the 52 MSAs with a 2015 population of 1 million or more persons and an average annual NCVS sample size of at least 250 persons during the period of 2006-2015. Excluded are persons not living within these 52 MSAs as well as those who are crews of vessels, in institutions (e.g., prisons and nursing homes) or members of the armed forces living in military barracks.

Unit(s) of Observation

person, household, incident

Data Type(s)

Mode of Data Collection

Response Rates

The unweighted median response rate among MSAs was 90% for households, and it was 86% for persons in participating households. Detailed unweighted and weighted response rates by MSA are included as Appendices D and E in the codebook.

Hide

Original Release Date

2022-03-21

Version History

2022-03-21 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.
Hide

Weight

Along with the household, person, and victimization analysis weights (i.e., WGTHH, WGTPER, and WGTVIC, respectively), 30 replicate weights are provided on each corresponding file. See the codebook for more information.

Hide

Notes

  • The public-use data files in this collection are available for access by the general public. Access does not require affiliation with an ICPSR member institution.