38603-User guide

National Crime VictimizationSurvey, [United States], 2022ICPSR 38603United States. Bureau of Justice StatisticsUser GuideInter-university Consortium forPolitical and Social ResearchP.O. Box 1248Ann Arbor, Michigan 48106www.icpsr.umich.eduTerms of UseThe terms of use for this study can be found at:http://www.icpsr.umich.edu/web/ICPSR/studies/38603/termsInformation about Copyrighted ContentSome instruments administered for studies archived with ICPSR may contain in wholeor substantially in part contents from copyrighted instruments. Reproductions of theinstruments are provided as documentation for the analysis of the data associated withthis collection. Restrictions on "fair use" apply to all copyrighted content. More informationabout the reproduction of copyrighted works by educators and librarians is available fromthe United States Copyright Office.NOTICEWARNING CONCERNING COPYRIGHT RESTRICTIONSThe copyright law of the United States (Title 17, United States Code) governs the makingof photocopies or other reproductions of copyrighted material. Under certain conditionsspecified in the law, libraries and archives are authorized to furnish a photocopy or otherreproduction. One of these specified conditions is that the photocopy or reproduction isnot to be "used for any purpose other than private study, scholarship, or research." If auser makes a request for, or later uses, a photocopy or reproduction for purposes inexcess of "fair use," that user may be liable for copyright infringement.-1-Study DescriptionCitationTitle StatementTitle: National Crime Victimization Survey, [United States], 2022Alternative Title: NCVS 2022Identification No.: 38603Responsibility StatementAuthoring Entity: United States. Bureau of Justice StatisticsProduction StatementProducer: Inter-university Consortium for Political and Social ResearchPlace of Production: Ann Arbor, MI: Inter-university Consortium for Political and Social ResearchFunding Agency: United States. Bureau of Justice StatisticsDistribution StatementDistributor: Inter-university Consortium for Political and Social ResearchSeries StatementSeries Name: National Crime Victimization Survey SeriesStudy ScopeSubject InformationKeyword(s): assault, auto theft, burglary, crime, crime costs, crime rates, crime reporting, crime statistics,offenders, offenses, property crimes, rape, reactions to crime, robbery, sexual offenses, victimization, victimsTopic Classification(s): Social Institutions and Behavior, Crime and the Criminal Justice SystemAbstractThe National Crime Victimization Survey (NCVS), previously the National Crime Survey (NCS), has beencollecting data on personal and household victimization through an ongoing survey of a nationally-representativesample of residential addresses since 1972. The survey is administered by the U.S. Census Bureau (under theU.S. Department of Commerce) on behalf of the Bureau of Justice Statistics (under the U.S. Department ofJustice).The NCS and NCVS were both designed with four primary objectives:

1. To develop detailed information about the victims and consequences of crime,
2. To estimate the numbers and types of crimes not reported to the police,
3. To provide uniform measures of selected types of crimes, and
4. To permit comparisons over time and types of areas.

Beginning in 1992, the NCVS categorizes crimes as "personal" or "property" covering the personal crimes of rapeand sexual assault, robbery, aggravated and simple assault and purse-snatching/pocket-picking; and the property-2-crimes of burglary, theft, and motor vehicle theft. Beyond a simple count of victimizations, the NCVS gathersdetails on the incident itself insofar as the victim can report them. Such details include the month, time, andlocation of the crime, the relationship between victim and offender, characteristics of the offender, self-protectiveactions taken by the victim during the incident and results of those actions, consequences of the victimization,type of property lost, whether the crime was reported to police and reasons for reporting or not reporting, andoffender use of weapons, drugs, and alcohol. Basic demographic information, such as age, race, gender, andincome, is also collected to enable analysis of crime by various sub-populations for select years. The Glossary(see Other Materials below) describes terms relevant to the NCVS.Analysts who are interested in the history of the 1992 redesign of the NCVS or are interested in conductinganalyses using both NCS and NCVS data should refer to the NCVS Resource Guide(https://www.icpsr.umich.edu/web/pages/NACJD/NCVS/index.html) on the NACJD Web site.Summary Data DescriptionTime Period: January 1, 2022 - December 31, 2022Date(s) of Collection: January 1, 2022 - December 31, 2022Country: United StatesGeographic Coverage: United StatesGeographic Unit: RegionUnit of Analysis: Household, Individual, Crime IncidentUniverse: The NCVS target population is individuals 12 years of age or older living in households and groupquarters within the United States and the District of Columbia. Excluded are persons who are crews of vessels, ininstitutions (e.g., prisons and nursing homes) or members of the armed forces living in military barracks.Interviews are translated for non-English speaking respondents where possible.Kind of Data: Survey dataComparisons Across YearsEvery 10 years, the U.S. Census Bureau (Census) conducts the official population count of the United States. In2006 and 2016, redesigns of the NCVS sample were necessary to account for shifts in the population identifiedthrough the Decennial Census. Examination of data in 2006 and 2016 suggested that these sample redesignshad a significant impact on survey estimates, meaning that estimates for 2006 and 2016 are not comparable toother years. Care should be taken in interpreting trend results including data from 2006. See CriminalVictimization, 2006 Technical Notes for more information. For 2016, BJS worked with the U.S. Census Bureau todevise a methodology for creating a revised 2016 NCVS data file to allow for comparisons between 2016 dataand prior (and future) data years. See Criminal Victimization, 2016: Revised for more information.2020 Weighting AdjustmentThe 2020 NCVS applied temporary changes to the weighting methods to adjust for the changes in data collectionprocedures due to the coronavirus pandemic. For additional information, please see the documentation for thedata from the 2020 NCVS, ICPSR Study 38090.NCVS Methodological Changes in 2016From 2015 to 2016, the National Crime Victimization Survey (NCVS) sample size increased by 41% (from 95,760to 134,690 households interviewed) to facilitate the ability to produce aggregate state- and local-level victimizationestimates for the largest 22 states. At the same time, the sample was adjusted to reflect the U.S. populationcounts in the 2010 decennial census (rather than being based on the 2000 decennial census, as was the case forthe 2006 through the 2015 NCVS).In Criminal Victimization, 2016 (NCJ 251150, BJS web, December 2017), BJS concluded that the 2016 NCVSvictimization rates could not be compared to 2015 or any prior year. BJS observed that "a comparison of the 2015-3-and 2016 victimization estimates showed that the violent and property crimes had increased" but that "it was notpossible to isolate ... the degree to which the change in rates from 2015 to 2016 resulted from the sampleredesign rather than real changes in U.S. victimization levels." Subsequently, we worked to develop a viablesolution that would permit comparisons between the 2016 NCVS and other years. See Criminal Victimization,2016: Revised for more information.Data AnomaliesNCVS Methodological Changes in 2006 and 2007A number of methodological changes were implemented in the NCVS in 2006 that impacted the victimization rateestimates for that year. These effects were reversed in 2007, suggesting that the 2006 findings represent atemporary anomaly in the data. The changes and their impact upon the survey's estimates are described inCriminal Victimization 2006 Technical Notes, available online at http://www.bjs.gov/content/pub/pdf/cv06.pdf. Adetailed discussion of the analyses of the 2007 data by BJS and the U.S. Census Bureau are available inCriminal Victimization, 2007 at http://www.bjs.gov/content/pub/pdf/cv07.pdf.The methodological changes implemented in 2006 that have been linked to the data anomaly for that yearinclude:• The introduction of a new sample based on the 2000 decennial census;• A transition from paper-and-pencil interviewing (PAPI) to computer-assisted interviewing (CAPI); and• The use of first-time interviews in the new sample areas in the production of survey estimates.In 2007, three additional changes were made largely for budgetary reasons to the NCVS program:• The sample was reduced by 14% in July 2007;• First-time interviews from all sample areas were used in the production of 2007 estimates; and• Computer-assisted interviews from centralized telephone centers were discontinued in July 2007.Analyses of the 2007 estimates indicate that the program changes made in 2007 had relatively small impact onNCVS estimates. As discussed in Criminal Victimization, 2006, the substantial increases in victimization ratesfrom 2005 to 2006 do not appear to be due to actual changes in crime during that period. Based on researchcompleted to date, there is a high degree of confidence that survey estimates for 2007 are consistent with andcomparable to those for 2005 and with 2008 through 2015. While the estimates for 2007 are close to the levels for2005, they are substantially below the levels for 2006.NCVS Sample ReinstatementTo restore the quality of the NCVS data, BJS and the U.S. Census Bureau (Census) implemented a samplereinstatement. The sample reinstatement project was designed to restore sample cases that were cut in the mid-2000s due to budget constraints as a means of improving the precision of the national victimization estimates.The sample size was returned to levels last seen in the late 1990s to increase the precision of the estimates forcrime. Beginning in October 2010, Census began restoring the sample to existing areas using existing fieldrepresentatives. The sample size increased approximately 24%, from about 8,500 households per month to10,500. The reinstatement was fully implemented by June 2011.FORM AND CONTENT OF NCVS DATA FILESThe form and content of NCVS data files have changed over the years. A brief discussion of this is warranted. Inparticular, the concepts of collection-year versus data-year format and bounded versus unbounded interviewsaffect the data.Collection year versus data yearHistorically the NCVS data files were made available data-year format. Under the data-year format, crimeincidents were analyzed based on the year that the incident actually occurred. Thus, given that the NCVSemploys a six month retrospective reference period, 18 months of interviews were necessary to estimate a year ofcrime incidents. For example, a respondent interviewed in March, 2022 may indicate that a crime occurred inDecember, 2021. Under the data-year format that crime would be counted in 2021.-4-Under the collection-year format, crime incidents are reported based on when the interview was conducted, notwhen the crime occurred. In the previous example, the crime would be counted in 2022. Whereas the data-yearformat requires 18 months of interviews to estimate crime incidents, the collection-year format requires only 12months. Therefore, estimates of crime can be reported in a more timely fashion under the collection-year format.NCVS data files made available under the data-year format included all of the data that comprise the collection-year format. However, the data-year format inherently resulted in delays in analyzing data and publishing results.All current BJS publications use the collection-year format. Consequently, BJS has decided to release acollection-year version of the NCVS.The NCVS data files have several variables that contain year information. Two of them are particularly important:V4015 (YEAR INCIDENT OCCURRED) and YEARQ (YEAR AND QUARTER OF INTERVIEW). As their labelsindicate they record different information. Under the collection-year format V4015 will contain more than one year,and YEARQ will contain only one year (recorded as yyyy.q to indicate both year and quarter of interview). Thecollection-year data files are prepared by using YEARQ to select records.NCVS data collections with titles such as NATIONAL CRIME VICTIMIZATION SURVEY, YYYY [RECORD-TYPEFILES] are data-year format. Data collections with titles such as NATIONAL CRIME VICTIMIZATION SURVEY,YYYY [COLLECTION YEAR RECORD-TYPE FILES] are collection-year format.Starting with 2009 data, the NCVS is released in collection-year format only. The 2009 and forward datacollections are titled: NATIONAL CRIME VICTIMIZATION SURVEY, YYYY. Refer to the 2009 codebook for moreinformation on collection year versus data year.Bounded versus Unbounded InterviewsNCVS households are interviewed a total of seven times at six-month intervals over a three-year period. From1992-2005 the first interview was excluded from the traditional release of NCVS data (except for replacementhouseholds). The first interview is the bounding interview and is used to control for respondent telescoping. It isused to place reported crime incidents in time to avoid reporting them again in subsequent interviews. Thetraditional release of NCVS data files is bounded data-year files. This changed in 2006. The years 2006 and 2007were a phase-in period when households new to the sample began having their first interviews included in theNCVS data files. Starting in 2008, all new households have their first interview included in the data; the data areunbounded. Data collections titled: NATIONAL CRIME VICTIMIZATION SURVEY, YYYY [RECORD-TYPEFILES] are bounded for years 1992-2005; for 2006 and 2007 they are a mix of bounded and unboundedinterviews; and for 2008 and onward they are unbounded.In anticipation of the NCVS data becoming unbounded, BJS made several years of NCVS data (1999-2006)available in unbounded format. This allowed users interested in methodological issues to work with both forms ofthe data. These data collections are titled: NATIONAL CRIME VICTIMIZATION SURVEY: UNBOUNDED DATA,YYYY.For years where there are both bounded and unbounded formats of NCVS data available, the unbounded datafile will contain more records. However, other than the additional records, the only difference between theunbounded data and the bounded data is the values of the weight variables. The household, person, and incidentweights have all been adjusted to reflect the larger number of records in the unbounded file.Weighted estimates of housing units and population are comparable in the unbounded and bounded data,although not exactly the same because of rounding as the weights were recalculated. Crime victimizationsestimated using the unbounded data will be higher than those estimated using the regular NCVS because morereported incidents are included with the data.ADDITIONAL DEMOGRAPHIC QUESTIONS ADDED IN 2016In July 2016, BJS added additional socio-demographic questions to the crime screener instrument (NCVS-1),specifically, respondent's veteran status, citizenship, disability status, sexual orientation, and gender identity. BJSalso expanded the household income response categories. However, because these variables were onlycollected for half the year, they are not available on the 2016 Public Use Files (PUF) with the exception ofdisability status. These variables are included on the Restricted Use Files through the U.S. Census Bureau'sFederal Statistical Research Data Centers (FSRDC). More information on the FSRDC is available at-5-(https://www.census.gov/fsrdc). Beginning in 2017, these variables are included in the Public Use Files. Currently,for these socio- demographic variables added to the NCVS in July 2016, previous responses stay with a personon the PUF even if they are not asked again at that next time-in-sample (TIS) interview. Therefore, once a personis asked these socio-demographic questions, those responses would be carried forward in subsequent interviews,and would appear on subsequent PUFs.Starting in July 2019, the sexual orientation and gender identity (SOGI) questions were administered only to crimevictims age 16 or older. In addition, these questions will only be asked once of all victims during the time theyremain in sample, rather than asking at their first, fifth, and seventh interview. From July 2016 to June 2019, thesequestions were administered to all respondents age 16 or older at their first, third, fifth, or seventh interview, or ifthey had not been asked the questions before. Because responses to the SOGI questions, like all other socio-demographic questions, are carried forward in subsequent interviews, the full effect of the July 2019 changes tothe administration of these questions won't be apparent on the PUF until all the June 2019 cases are no longer inthe NCVS sample, which happened in July 2022. Even after the July 2019 SOGI changes were implemented,some respondents with no crime incidents on the PUF will still have been asked the SOGI questions. This is afunction of the data collection and editing process, and can happen for various reasons.Starting in January 2022, BJS reverted back to administering the SOGI questions to all respondents (not proxyrespondents) age 16 or older, regardless of whether they completed a Crime Incident Report. As a result of thischange, populations estimates can be generated and used to calculate crime rates by SOGI status. Thesequestions are asked at the first, third, fifth, and seventh interviews or if the respondent had not been asked thequestions before.Annual Updates to NCVS PUFData users should always refer to the most recent versions of released NCVS data for updates or changes toPublic Use Files.Record-Type FilesData from the National Crime Victimization Survey are now available from ICPSR in four record-type files:address, household, person, and incident. Previously NCVS data were available in a hierarchical formatcontaining four record types. The use of four record-type files - each a rectangular file - simplifies the use of thedata while maintaining all of the information and functionality present in the hierarchical format.The address and household record-type files contain information about the household as reported by therespondent and characteristics of the surrounding area as computed by the U.S. Census Bureau. The personrecord-type file contains information about each household member age 12 years and older as reported by thatperson or proxy, with one record for each qualifying individual. The incident record-type file contains informationdrawn from the incident report completed for each household incident or person incident mentioned during theinterview.ID variables have been added to the record-type files to facilitate merging the separate files. Each of the fourrecord-type files contains the variable YEARQ, which indicates the year and quarter an interview was conductedin, or was to be conducted in. This variable is necessary because NCVS ID variables are unique only within eachsix-month interview period. The address record-type file is merged to the household record-type file using YEARQand IDHH (the household ID). The household record-type file is merged to the person record-type file also usingYEARQ and IDHH (the household ID). The person record-type file is merged to the incident record-type file usingYEARQ, IDHH (the household ID), and IDPER (the person ID). An example of matching the record-type files isincluded with this documentation in Other Materials below. The example demonstrates how to create the incident-level extract file included with this data collection.In addition to the single-year record-type files contained in this data collection, a separate data collection containsmulti-year concatenated incident-level extract files.Methodology and ProcessingData Collection MethodologyTime Method: Rotating panel survey-6-Data Collector: United States Department of Commerce. U.S. Census BureauFrequency of Data Collection: Semi-annuallySampling 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. Typicallythe first interview is face-to-face and the remainder are by telephone. After the seventh interview the householdleaves the panel and a new household is rotated in to the sample.Sample DesignThe NCVS sample consists of approximately 200,000 sample housing units selected with a stratified, multi stagecluster design. The NCVS sample is drawn from the decennial census.To take advantage of the availability of data from the most recent census, sample selection from the most recentcensus is phased in at approximately mid-decade. Prior to that, sample selection is based on the census beforethe most recent one. For example, prior to 1995, the sample was drawn from the 1980 decennial Census. FromJanuary, 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, samplesfrom the 2000 Census began to be phased-in. In 2015, samples from the 2010 Census began to be phased in.The sample of housing units within a PSU was selected in two stages. The first step of the sample designinvolves dividing the United States into 1,987 first-stage sample units or PSUs consisting of large metropolitanareas, counties, or groups of bordering counties. These areas are grouped together within each state using datafrom the 2010 Census on characteristics such as total land area, current and projected population counts, largemetropolitan areas, and natural barriers, such as rivers and mountains. With the exception of some PSUs inAlaska, the required population size for a PSU was at least 7,500 persons. The preferred maximum land areawas 3,000 square miles, but some PSUs exceeded this threshold because counties had a large geographic areaand small population. These limits were chosen to provide sufficient workload for at least one interviewer per PSUwhile ensuring travel time for individual interviewers was not overly onerous.The next step is dividing the PSUs into sampling strata. Some PSUs are so large and therefore important to theestimates that they are included in every sample. They are known as self-representing (SR) PSUs because theyrepresent themselves and no other PSUs. For NCVS, PSUs are identified as SR if the counties are within largeCore Based Statistical Areas (CBSAs). All SR PSUs make up their own sampling strata.The remaining PSUs are non-self-representing (NSR) and are grouped within state with similar NSR PSUs toform strata. The data used for grouping the NSR PSUs consist of decennial census demographic data, AmericanCommunity Survey data, and administrative crime data. NSR PSUs are grouped to be as similar or homogeneousas possible. Just as the SR PSUs must be large enough to support a full workload, so must each NSR strata beof that size. The most efficient stratification scheme is determined by minimizing the between-PSU variance andthe within-PSU variance.The first stage selection identifies a sample of PSUs, and this occurs once every ten years. The SR PSUs areautomatically selected for sample or selected with certainty. The NSR PSUs are sampled with probabilityproportional to the population size using a linear programming algorithm. One PSU is selected from each NSRstratum. There was no effort to maximize the amount of overlap between 2010 and 2000 design PSUs.The 2010 design NCVS sample includes 339 SR PSUs and 203 NSR PSUs. These 542 PSUs include 933counties and county equivalents, 505 of which are continuing areas because they were also in sample in the 2000design NCVS. The remaining 428 counties and county equivalents are new to NCVS beginning with the 2016data collection year. Because PSUs were redefined for the 2010 design, some sampled PSUs contain both newand continuing counties. Therefore, the concept of new or continuing first-stage sample areas must be applied atthe county level and not at the PSU level.For second stage selection, a systematic random sample of housing units and GQs is selected within the firststage sample PSUs. The sample size is the number of housing units and GQs expected to result in enoughcompleted person interviews to meet NCVS reliability goals.-7-Rotating Panel DesignBecause of the continuing nature of the National Crime Victimization Survey, a rotation scheme was devised toavoid interviewing the same household indefinitely. The sample of housing units is divided into six rotationgroups, and each group is interviewed every six months for a period of three and a half years.Within each of the six rotation groups, six panels are designated. A different panel is interviewed each monthduring the six-month period. Until 2006, the first interview for a housing unit in a group was used only forbounding and was not included in the regular public use data. However, if a new household moved into a sampleunit, the members of that household were interviewed and their unbounded interviews were included in the data.Thus, the number of incident reports associated with a household may be affected by its tenure in the sample.New households will have unbounded information while older units may display "time in sample" effects. Time insample effects include telescoping and fatigue. Telescoping is a tendency to report incidents that occurredoutside the period of interest, and fatigue is a tendency to report fewer incidents in later waves of a panel survey.See Census Sample Information and NCVS Rotation Chart in Other Materials below for an illustration of theNCVS rotation scheme.Mode of Data Collection: Telephone interviews, computer-aided telephone interviews, face-to-face interviewsType of Research Instrument:Three basic forms are used to collect the required data for the National Crime Victimization Survey. The controlcard contains the address of each sample unit and the basic household data, such as the names of all personsliving there and their age, race, sex, marital status, education, etc. The basic screen questionnaire contains a setof questions used to elicit reports of crimes against the household and its members. The crime incident reportcontains a set of questions used to gather details (e.g., location of incident, relationship of victim to offender,extent of injuries and/or economic loss) on crimes reported in the Basic Screen Questionnaire. These forms aredescribed in detail in the Glossary (see Other Materials below, under "Control Card", "Basic ScreenQuestionnaire", and "Incident Report"; also see Questionnaire Details in Other Materials below for information onthe NCVS questionnaire and interviewer training).A number of changes have been made over time in the NCS and NCVS instruments and the procedures foradministering them. For example, in some cases response categories were changed, question wording wasaltered, and questions were added or deleted. Major changes relating to the NCVS redesign are described in theNCVS Redesign document contained in the NCVS Resource Guide(https://www.icpsr.umich.edu/web/pages/NACJD/NCVS/index.html).Due to the changes that the National Crime Victimization Survey instrument has undergone over the years, astandard format for all files is not possible. Users are strongly urged to carefully review each variable used in theiranalyses.The variables that are relevant to an individual year of NCVS are contained in the current year's documentation.Variables no longer in use are excluded from the current individual year data files and documentation.As new questions are added to the NCVS, ICPSR attempts to add the variables to the data files so that theyparallel the flow of the questionnaire. Inevitably, this means that new variables will be inserted between existingones. The original NCVS variables were named V1000's for address variables, V2000's for household variables,V3000's for person variables, and V4000's for incident variables. Suffixes are added to new variables that areinserted between existing variables. The presence of a suffix does not mean that the variable is a continuation ofthe previous one. Rather, it is most likely a completely new variable.Actions to Minimize Nonresponse:Prior to an interview visit, each sample unit is sent a letter from the U.S. Census Bureau informing the householdabout the National Crime Victimization Survey. During each enumeration period, interviews are conducted with asmany eligible household members age 12 and over as possible. Questions pertaining to the entire household areasked only once of the Household Respondent (see Glossary in Other Materials below), a single householdmember who could reasonably be expected to know the answers pertaining to the household. Individual ScreenQuestions on the Basic Questionnaires are asked of all household members age 12 or older.-8-Currently, 12 and 13 year olds are interviewed directly unless a knowledgeable household member insists on aproxy. (Prior to July, 1986, information about each household member 14 years and older was obtained by self-response; however, information about members 12 and 13 years old was obtained by a proxy from either thehousehold respondent or some other knowledgeable household member.)The average face-to-face household interview lasts 25 minutes. Subsequent interviews are conducted primarilyby telephone. However, in some cases subsequent interviews are conducted face-to-face for respondents withoutphones, who are difficult to reach by phone, or prefer that mode over the telephone.A sample unit for which an interview could not be obtained is classified as one of three non-interview types. TypeA non-interviews consist of households occupied by persons eligible for interviews but from whom no interviewswere obtained because, for example, no one was found at home in spite of repeated visits or the householdrefused to give any information. Because Type A non-interviews are considered avoidable, every effort is made toconvert them to interviews.Type B non-interviews are for units which are unoccupied or which are occupied solely by persons who have ausual residence elsewhere (URE).Type C non-interviews are units ineligible for the sample because the unit is demolished, condemned, or noaddress was listed. Occasionally, the field representative is unable to obtain an interview for a particularhousehold member in an otherwise interviewed household. This person is classified as a type Z person (seeGlossary in Other Materials below under "Non-Interview" for more detail).Weighting:See Weights Details in Other Materials below.Address Versus Household:Once an address is selected, the "household" (persons 12 and over living at that address) is interviewed. If thathousehold moves out and a different household moves in, the survey is conducted with this new household. Thatis, the survey maintains contact with the ADDRESS originally sampled, not necessarily the INDIVIDUALSoriginally sampled. The variable V1006 will change to indicate that a different household is being interviewed fromthe previous period.Original Variables and Allocated Variables:During the data collection process, responses may be keyed in incorrectly. For most variables, any invalid entry isrecoded as missing or "Residue". For some variables, however, the interviewer's original keystroke entry ismaintained in the dataset. These variables are called "Original" variables, each of which has a corresponding"Allocated" variable. Allocated variables result from a U.S. Census Bureau recoding scheme which forces anyinvalid entry into a valid code. Original variables may contain the original invalid entries. Variable labels in thePUF note whether the variable is "Original" or "Allocated."Multiple Response and Category Variables:Variable labels for multiple response questions are preceded by (LI) which indicates the "Lead in" question, (C)which indicates any number of category questions associated with the lead-in question, and (RESIDUE) whichindicates the residue variable. Responses to the "Lead in" question may fit into any number of categoriesassociated with it. The "Lead in" code indicates whether at least one positive response was obtained for any ofthe categories. Variables for each individual category follow the "Lead in" variable. If respondents answer "Yes" tothe main question, but none of the categories is applicable, the Lead-in code is equal to 1, the category codes areequal to 0 and the residue code is equal to 8.Data Appraisal InformationResponse Rate: The 2022 NCVS data file includes 143,794 household interviews. Overall, 64% of eligiblehouseholds completed an interview. Within participating households 226,962 persons completed an interview in2022, representing an 82% response rate among eligible persons from responding households. See the Sourceand Accuracy Statement for the 2022 National Crime Victimization Survey in the Other Materials section of thisdocument for more information on response rates.-9-Estimates of Sampling Error: See Variance Details in Other Materials below.Data AccessDataset AvailabilityLocation: Inter-university Consortium for Political and Social ResearchData Use Statement:Citation Requirement: Publications based on ICPSR data collections should acknowledge those sources bymeans of bibliographic citations. To ensure that such source attributions are captured for social sciencebibliographic utilities, citations must appear in footnotes or in the reference section of publications.Deposit Requirement: To provide funding agencies with essential information about use of archival resources andto facilitate the exchange of information about ICPSR participants' research activities, users of ICPSR data arerequested to send to ICPSR bibliographic citations for each completed manuscript or thesis abstract. Visit theICPSR Web site for more information on submitting citations (http://www.icpsr.umich.edu).Data Files DescriptionFile-by-File DescriptionAddress Record-Type FileFile Name: Household Record-Type FileContents of Files: The address record-type files contain information about the household as reported by therespondent and characteristics of the surrounding area as computed by the U.S. Census Bureau.Missing Data: There are four missing data codes used in this collection:

1. RESIDUE - This type of missing data is represented by 8's or a combination of 9's and 8's. Residue can

indicate a keying error which resulted in an out-of-range code, an incorrect or unusable answer by therespondent, or the absence of an entry for a question that should have been asked.There are two types of residue on multiple response variables on NCVS data files. The first type of residue occurswhen the entire item was not ascertained. In this instance, there will be a code 8 on the lead-in variable, an 8 ineach of the multiple response category variables, and an 8 on the residue variable for that multiple responsequestion.The second type of residue occurs when there is a "yes" for the lead-in question but the specific categories of themultiple response items are not ascertained. In this situation there will be a code 1 on the lead-in variable and an8 on the residue variable. The multiple response category variables will all have 0 entered.

2. OUT OF UNIVERSE/BLANK - This type of missing data is represented by 9's. This code indicates that the

case is outside of the applicable range or universe of questions to be answered. For instance for V4127,"Received Medical Care for Injury", only victims who report being injured are asked whether they receivedmedical care. All other victims are skipped past this question and are coded "9" on this variable.In BJS publications, Residue categories, such as "Don't know" are often included in tabulations and are thus notclassed as missing data. Data users may choose to define these Residue codes as missing data if it is useful fortheir analytic purposes.See the Notes for a tabular summary of these two missing data codes and types.

3. ITEM INVALID UNTIL YYYY Q# - This type of missing data is represented by -1. Variables with this code

were created during the data year, and the code of -1 means that the missing data is due to the item notbeing in use the full data year. The code distinguishes it from normal out of universe missing data.-10-

4. ITEM INVALID AFTER YYYY Q# - This type of missing data is represented by -2. Variables with this code

were dropped from use during the data year. The code distinguishes it from normal out of universe missingdata.Codes -1 and -2 are user specified missing values.MISSING DATA CODE TYPE OF MISSING DATA99999 Out of Universe8 or 98 or 998 Not Ascertained Residue:(A) Single ResponseIncorrect Response or Keying Error(B) Multiple Response

1. Entire item Not Ascertained

a) Lead-in: coded 8b) Each multiple response category variables: coded 8c) Residue variable: coded 8

2. "Yes" on Lead-In but Categories on Multiple Responses are Not

Ascertaineda) Lead-in: coded 1b) Multiple response category variables: coded 0c) Residue variable: coded 8

Household Record-Type FileFile Name: Household Record-Type FileContents of Files: The household record-type file contains information about the household as reported by therespondent and characteristics of the surrounding area as computed by the U.S. Census Bureau.Unique Identifier: IDHHThe variable IDHH - NCVS ID FOR HOUSEHOLDS - uniquely identifies a household within the data file.However, given that a year of NCVS data includes 18 months of interviews, and interviews are conducted everysix months, the same IDHH may appear in the data file up to three times. It may also appear in the data file oneor two times depending on attrition. The variable YEARQ - YEAR AND QUARTER OF INTERVIEW (YYYY.Q) - inconjunction with IDHH uniquely identifies a household within a six month interview period.

Person Record-Type FileFile Name: Person Record-Type FileContents of Files: The person record-type file contains information about each household member age 12 yearsand older as reported by that person or proxy, with one record for each qualifying individual.Unique Identifier: IDPER-11-The variable IDPER - NCVS ID FOR PERSONS - uniquely identifies a person within the data file. However, giventhat a year of NCVS data includes 18 months of interviews, and interviews are conducted every six months, thesame IDPER may appear in the data file up to three times. It may also appear in the data file one or two timesdepending on attrition. The variable YEARQ - YEAR AND QUARTER OF INTERVIEW (YYYY.Q) - in conjunctionwith IDPER uniquely identifies a person within a six month interview period.

Incident Record-Type FileFile Name: Incident Record-Type FileContents of Files: The incident record-type file contains information drawn from the incident report completed foreach household incident or person incident mentioned during the interview.Incidents vs. Victimizations:For personal crimes, the NCVS makes a basic distinction between "incidents" and "victimizations." The number ofvictimizations reflects how many criminal acts were experienced by NCVS survey respondents; the number ofincidents reflects the number of criminal acts committed against NCVS respondents and others present duringsuch incidents, as reported by survey respondents. If there was more than one victim, the incident weight isadjusted to compensate for the possibility that the incident could be reported several times by multiple victims andthus be over-counted. For example, if a survey respondent and two other people (non-respondents) were robbedduring the same incident, the NCVS counts this as one-third of a personal incident and as one victimization. If allvictims were respondents, this criminal act would be counted as one incident and three victimizations. Therefore,estimates of the number of personal incidents are usually lower than estimates of personal victimizations.Estimates of incidents and victimizations can be made by using weights. For personal crimes the variable"INCIDENT WEIGHT ADJUSTED FOR SERIES CRIMES" [SERIES_IWEIGHT] was computed by dividing theperson's weight by the total number of persons victimized during the incident and is used to estimate incidenttotals. To calculate a victimization count, users should use the variable "ADJUSTED VICTIMIZATION WEIGHT"(WGTVICCY or WGTVICDY for collection year or data year respectively). To calculate a victimization count tomatch recent BJS reports, users should use variable "SERIES_WEIGHT" or the code to calculate series weightsfound in the Crime Estimate Details section in Other Materials below.

Household Crimes:By definition household crimes are committed against the household as a whole. Therefore, a household crimecan only have one "victim" -- the household. Accordingly, the victimization weight equals the incident weight forhousehold crimes.

Changes in the Definition of Theft:Due to a change in the NCVS definition of Theft, most property crimes are now classified as Household Theft (theexceptions are purse-snatching and pocket-picking). Therefore, the Household Weight should be used for allproperty theft without contact and the Person Weight should be used for all purse-snatching and pocket-pickingcrimes. The decision of whether to use the person weight or household weight has been incorporated into theadjusted victimization weights for collection year and data year (WGTVICCY and WGTVICDY). See Glossary inOther Materials below for details on the Theft redefinition.

Type of Crime Code (TOC):During 1992 Quarters 1 and 2 the type of crime (TOC) "Unwanted Sexual Contact Without Force" (code "16"V4528 and V4529) did not exist. For this TOC and related crimes, therefore, comparisons of the 1992 data withfuture years should be avoided.

2022 Collection Year Incident-Level Extract FileFile Name: 2022 Collection Year Incident-Level Extract File

Contents of Files:The incident-level extract file is not the same as the incident record-type file. The incident-level extract file is a file-12-created by prepending household and person variables to incident records using the YEARQ (YEAR ANDQUARTER OF INTERVIEW), IDHH (NCVS ID FOR HOUSEHOLDS), and IDPER (NCVS ID FOR PERSONS)variables as match keys. For data-year formats, this file has been "bounded" to contain incidents occurring withinthe specific calendar year, regardless of when the interview was conducted. Under the collection-year format, theincident-level extract file is not "bounded" based on when the incidents occurred; rather, it contains incidentsreported in 2022, regardless of when they occurred.Example code for creating a collection year incident-level extract file is included in Matching Files to Create theCollection Year Incident- Level Extract File in Other Materials below.

Other Materials

Crime Estimate DetailsThe method of producing estimates of victimization is different depending on whether the DATA-YEAR orCOLLECTION-YEAR format is used.To produce rates of victimization under the COLLECTION-YEAR format, one only needs to identify the year thatthe interview was conducted.Producing estimates of victimization under the DATA-YEAR format is more complicated. Under the DATA-YEARformat, adjustments must be made to account for the rotational period of the sample design and the reported dateof the incident. The table below shows the relationship between collection month and the related referencemonths.-13-

NCVS REFERENCE PERIODS FOR ANNUAL ESTIMATES

REFERENCE MONTH WITHIN BOUNDING PERIODCollectionMonthJAN FEB MAR APR MAY JUN JUL AUG SEP OCT NOV DECJANFEB XMAR X XAPR X X XMAY X X X XJUN X X X X XJUL X X X X X XAUG X X X X X XSEP X X X X X XOCT X X X X X XNOV X X X X X XDEC X X X X X XJAN X X X X X XFEB X X X X XMAR X X X XAPR X X XMAY X XJUN XThe above table illustrates the collection month from which information about a given calendar or reference yearmay be drawn due to the six-month bounding period. For this reason, a respondent may report a victimizationincident which occurred up to six months before the time of the interview. When producing DATA-YEARvictimization estimates, one must identify incidents occurring in the reference year. To identify all the incidentswhich occurred in a reference year, all the incidents reported as having occurred before the beginning of that yearmust be excluded. Also, all the incidents reported in the following year as having occurred before the end of thereference year must be identified.Under the COLLECTION-YEAR format, when the interview is conducted determines when the victimization iscounted.COMMENT The following is an example calculation of crime rates forCOMMENT households and persons. The example uses SPSS as theCOMMENT statistics package, and the calculations are on aCOMMENT collection-year basis.COMMENT For the example we assume that the household, person, and incidentCOMMENT record-type files have been downloaded and renamed: hh_rec2.sav,COMMENT per_rec3.sav, and incid_rec4.sav respectively.

COMMENT CALCULATE HOUSEHOLD VICTIMIZATIONS AND THE HOUSEHOLD COUNTCOMMENT ON A COLLECTION YEAR BASIS.-14-COMMENT Household victimization count.get file = "incid_rec4.sav".COMMENT Select only property crimes (TYPE OF CRIME > 23).select if ( V4529 > 23 ).COMMENT To be consistent with BJS publications prior to 2012, exclude series crimes.COMMENT Exclude series crimes (use for data prior to 2012 only)COMMENT V4019 - ENOUGH DETAIL TO DISTINGUISH INCIDENTS.COMMENT select if ( V4019 ne 2 or missing(V4019) ).COMMENT To be consistent with BJS publications from 2012 on (CriminalCOMMENT Victimizations, 2011), include series crimes and use the syntax below.COMMENT Include up to 10 occurrences in a series crime.do if (v4019=2).recode v4016 (11 thru 996=10). end if.execute.if ( v4019 = 2 ) WGTVICCY = WGTVICCY * v4016.COMMENT To be consistent with BJS publications for all years, exclude crimesCOMMENT committed outside the U.S.select if ( V4022 ne 1 or missing(V4022) ).weight by WGTVICCY.frequencies variables = V4529.COMMENT The total (13,373,327) is the number of victimizationsCOMMENT against households on a collection year basis.COMMENT Household count.get file = "hh_rec2.sav".weight by WGTHHCY.COMMENT Use any variable for a frequency (V2006 - HOUSEHOLD NUMBER).frequencies variables = V2006.COMMENT The total (131,259,549) is the number of households on aCOMMENT collection-year basis.COMMENT Thus the victimization rate for households (per thousandCOMMENT households) is:COMMENT 13,373,327COMMENT ------------------------ = 101.9COMMENT (131,259,549/1,000)COMMENT .

COMMENT CALCULATE PERSON VICTIMIZATIONS AND THE PERSON COUNT ON ACOMMENT COLLECTION-YEAR BASIS.COMMENT Person victimization count.get file = "incid_rec4.sav".-15-COMMENT Select only crimes against persons (TYPE OF CRIME <= 23).select if ( V4529 <= 23 ).COMMENT To be consistent with BJS publications prior to 2012, exclude series crimesCOMMENT and crimes committed outside the U.S.COMMENT Exclude series crimes (use for data prior to 2012 only)COMMENT V4019 - ENOUGH DETAIL TO DISTINGUISH INCIDENTS.select if ( V4019 ne 2 or missing(V4019) ).COMMENT To be consistent with BJS publications from 2012 on (CriminalCOMMENT Victimizations, 2011), include series crimes and use the syntax below.COMMENT Include up to 10 occurrences in a series crime.do if (v4019=2).recode v4016 (11 thru 996=10). end if.execute.if ( v4019 = 2 ) WGTVICCY = WGTVICCY * v4016.COMMENT To be consistent with BJS publications for all years, exclude crimesCOMMENT committed outside the U.S.select if ( V4022 ne 1 or missing(V4022) ).weight by WGTVICCY.freq vars = V4529.COMMENT The total (6,772,351) is the number of victimizationsCOMMENT against persons.COMMENT Person count.get file = "per_rec3.sav".weight by WGTPERCY.COMMENT Use any variable for frequency (V3018 - SEX ALLOCATED). frequencies variables = V3018.COMMENT The total (282,304,639) is the number of persons age 12 andCOMMENT older on a collection-year basis.COMMENT Thus the victimization rate for persons (per thousandCOMMENT persons) is:COMMENT 6,772,351COMMENT -------------------- = 24.0COMMENT ( 282,304,639/1,000 )COMMENT .

INCIDENT-LEVEL EXTRACT FILEIn the examples above the victimization counts were calculated using the incident record-type file. The incident-level extract file (not the same as the incident record-type file) can also be used to calculate victimizations.

Data-Year Files-16-

CONSTRUCTING ADDITIONAL NCVS DATA FILES - DATA-YEAR FORMATThis data collection includes 2022 NCVS data files constructed on a collection-year basis. Data-year files forthe previous year - 2021 - can be constructed by using the current files, the previous year's files, and thestatistical software syntax that follows.A common use of the NCVS is to estimate victimization rates. These rates can be estimated on both acollection-year and a data-year basis. For either basis this involves determining estimates of the number ofhouseholds, the number of persons, and the number of household and personal victimizations.

DATA-YEAR HOUSEHOLD FILECreating a 2021 data-year household file involves concatenating the 2021 and 2022 collection-year householdfiles, subsetting the resulting file to include only records for interviews conducted from January 2021 through June2022, and calculating the household weight to reflect the households' time in sample.In constructing the 2021 data-year household file records from interviews conducted in January through June2022 need to be included along with records from interviews conducted in 2021. This is because of the six-monthretrospective interview period; respondents interviewed in June of 2022 may report victimizations that occurred inDecember 2021.The household weight needs to be constructed such that it represents the households' opportunity to report acrime incident. For example, a household interviewed in June 2022 has an opportunity to report crimes occurringin one month of 2021 - December. Whereas a household interviewed in December of 2021 has an opportunity toreport crimes occurring in six months of 2021 - June through November. Variable V2116 (HOUSEHOLDWEIGHT) is adjusted to reflect the varying opportunities to report crime incidents based on the households' timein the NCVS sample.The SPSS code below is an example of how to prepare a data-year household file.--- BEGIN SPSS SYNTAX FOR CREATING A 2021 DATA-YEAR HOUSEHOLD FILE.Concatenate the 2021 and 2022 collection-year householdrecord-type files.add files/file = "household2021CY.sav"/file = "household2022CY.sav".The data year consists of ALL the 2021 collection-year recordsand the FIRST SIX MONTHS' of the 2022 collection-year records.select if ( YEARQ >= 2021.1 and YEARQ <= 2022.2 ).Adjust the six month household weight (V2116) to account for time insample.YEARQ - YEAR AND QUARTER OF INTERVIEW.V2130 - MONTH ALLOCATED FROM PANEL/ROTATION NUMBER.do if (YEARQ=2021.1 and v2130=01). compute WGTHHDY=0.else if (YEARQ=2021.1 and v2130=02). compute WGTHHDY=(V21161/12).else if (YEARQ=2021.1 and v2130=03). compute WGTHHDY=(V21162/12).else if (YEARQ=2021.2 and v2130=04). compute WGTHHDY=(V21163/12).else if (YEARQ=2021.2 and v2130=05). compute WGTHHDY=(V21164/12).else if (YEARQ=2021.2 and v2130=06). compute WGTHHDY=(V21165/12).-17-else if (YEARQ=2021.3 or YEARQ=2021.4 and (v2130=07 or v2130=08 or v2130=09or v2130=10 or v2130=11 or v2130=12)).compute WGTHHDY=(V21166/12).else if (YEARQ=2022.1 and v2130=01). compute WGTHHDY=(V21166/12).else if (YEARQ=2022.1 and v2130=02). compute WGTHHDY=(V21165/12).else if (YEARQ=2022.1 and v2130=03). compute WGTHHDY=(V21164/12).else if (YEARQ=2022.2 and v2130=04). compute WGTHHDY=(V21163/12).else if (YEARQ=2022.2 and v2130=05). compute WGTHHDY=(V21162/12).else if (YEARQ=2022.2 and v2130=06). compute WGTHHDY=(V21161/12).end if.Save the 2021 data-year household file.The file is currently unweighted.save outfile = "household2021_DY.sav".Weight the data by the adjusted household weight. weight by WGTHHDY.Run a frequency on any variable.freq vars = V2001.NOTE - A warning message about a zero value for a weight isnormal. It is not an error.RESULT:The 2021 household count on a data year basis is: 129,669,277.--- END SYNTAX.

DATA-YEAR PERSON FILECreating a 2021 data-year person file involves the same steps as described above for creating a household file.The 2021 and 2022 collection-year person files are concatenated, subsetted to include only records from January2021 through June 2022, and the person weight calculated to reflect the households' time in sample.--- BEGIN SPSS SYNTAX FOR CREATING A 2021 DATA-YEAR PERSON FILE.Concatenate the 2021 and 2022 collection-year personrecord-type files.add files/file = "person2021CY.sav"/file = "person2022CY.sav".The data year consists of ALL the 2021 collection-year recordsand the FIRST SIX MONTHS' of the 2022 collection-year records.select if ( YEARQ >= 2021.1 and YEARQ <= 2022.2 ).Adjust the six month person weight (V3080) to account for time in sample.YEARQ - YEAR AND QUARTER OF INTERVIEW.V3025 - MONTH INTERVIEW COMPLETED.do if (YEARQ=2021.1 and V3025=01). compute WGTPERDY=0.else if (YEARQ=2021.1 and V3025=02). compute WGTPERDY=(V30801/12).else if (YEARQ=2021.1 and V3025=03). compute WGTPERDY=(V30802/12).-18-else if (YEARQ=2021.2 and V3025=04). compute WGTPERDY=(V30803/12).else if (YEARQ=2021.2 and V3025=05). compute WGTPERDY=(V30804/12).else if (YEARQ=2021.2 and V3025=06). compute WGTPERDY=(V30805/12).else if (YEARQ=2021.3 or YEARQ=2021.4 and (V3025=07 or V3025=08 or V3025=09or V3025=10 or V3025=11 or V3025=12)).compute WGTPERDY=(V30806/12).else if (YEARQ=2022.1 and V3025=01). compute WGTPERDY=(V30806/12).else if (YEARQ=2022.1 and V3025=02). compute WGTPERDY=(V30805/12).else if (YEARQ=2022.1 and V3025=03). compute WGTPERDY=(V30804/12).else if (YEARQ=2022.2 and V3025=04). compute WGTPERDY=(V30803/12).else if (YEARQ=2022.2 and V3025=05). compute WGTPERDY=(V30802/12).else if (YEARQ=2022.2 and V3025=06). compute WGTPERDY=(V30801/12).end if.Save the 2021 data-year person file.This file is unweighted.save outfile = "person2021_DY.sav".Weight the data by the adjusted person weight. weight by WGTPERDY.Run a frequency on any variable. freq vars = V3001.NOTE - A warning message about a zero value for a weight is normal.It is not an error.RESULT:The 2021 person count on a data year basis is: 279,499,176.--- END SYNTAX.

DATA-YEAR INCIDENT FILECreating a 2021 data-year incident file is different from creating the household and person files.As with the previous examples, the 2021 and 2022 incident files are concatenated together. Unlike the previousexamples, the file is not subsetted based on when the interviews were done. Rather, it is subsetted based on theyear that the incidents occurred. To create a 2021 data-year incident file only incidents occurring in 2021 areselected.The collection-year victimization weight does not need to be adjusted. However, to reinforce that this is now adata-year file a new weight is created and specifically indicated data year.Victimization rates published by BJS often exclude series crimes and crimes committed outside of the U.S. Thisneeds to be taken into consideration to tie to BJS publications.--- BEGIN SPSS SYNTAX FOR CREATING A 2021 DATA-YEAR INCIDENT FILE.Concatenate the 2021 and 2022 collection-year incidentrecord-type files.add files/file = "incident2021CY.sav"/file = "incident2022CY.sav".Select only incidents OCCURRING in 2021.-19-V4015 - YEAR INCIDENT OCCURRED.select if ( V4015 = 2021 ).In the subsetted file the data-year victimization weightis the same as the collection-year victimization weight.numeric WGTVICDY (F12.5).variable labels WGTVICDY "ADJUSTED VICTIMIZATION WEIGHT - DATA YEAR". compute WGTVICDY =WGTVICCY.Save the 2021 data-year incident file.This file is unweighted.save outfile = "incident2021_DY.sav".To be consistent with BJS publications prior to 2012,exclude series crimes (use only with data prior to 2012).V4019 - ENOUGH DETAIL TO DISTINGUISH INCIDENTS.select if ( V4019 ne 2 or missing(V4019) ).Exclude crimes outside the U.S. - this is a BJS convention.Do this in order to tie to BJS publications for all years.V4022 - IN WHAT CITY, TOWN, VILLAGE.select if ( V4022 ne 1 or missing(V4022) ).To be consistent with BJS publications from 2012 on (CriminalVictimizations, 2011), include series crimes and use the syntax below.Include up to 10 occurrences in a series crime.do if (v4019=2).recode v4016 (11 thru 996=10). end if.execute.if ( v4019 = 2 ) WGTVICDY = WGTVICDY v4016.Weight the data by the adjusted victimization weight. weight by WGTVICDY.Run a frequency on the TYPE OF CRIME (V4529) variable.freq vars = V4529.RESULT - The total victimization count on a data-year basis is:15,607,478.*--- END SYNTAX.

DATA-YEAR INCIDENT-LEVEL EXTRACT FILEData analysts are often interested in examining crime incidents (victimization) while taking into considerationcharacteristics of the person and the household. One way to do this is to prepend the household and personvariables that are related to the crime incident records. This requires merging three files to create a Data-YearIncident-Level Extract File.The following example builds on the examples above. The files that are used are already data-year files.-20-Consequently the resulting extract file will be a data-year file.*--- BEGIN SPSS SYNTAX FOR CREATING A 2021 DATA-YEAR INCIDENT-LEVEL EXTRACT FILE.Two merges are done. The first merges the household variables ontothe person record using the YEARQ (YEAR AND QUARTER OF INTERVIEW)and IDHH (NCVS ID FOR HOUSEHOLDS) variables.In the match files statement the "table" indicates that thehousehold file will be used as a one-to-many match. For example, agiven household record may match to several person records, and thesame household information will be copied onto all of the matchingperson records. It is possible that a household will have nomatching person records. In these instances using a tablespecification prevents creating a new record with only householdvariables and no person variables. Thus the table specificationcopies household variables onto all of the matching person records,but it does not create a new record when there is not a match. Thisissue is more prominent when merging household and person variablesonto crime incident records.match files/table = "household2021_DY.sav"/file = "person2021_DY.sav"

/by YEARQ IDHH.Now merge the "in-process" file (denoted by the asterisk) createdabove with the incident record file. The table in the match filesstatement is the in-process file of person records with householdvariables prepended to them.match files/table = */file = "incident2021_DY.sav"

/by YEARQ IDHH IDPER.Save the 2021 data-year incident-level extract file. This filecontains incident records where the incident occurred in

2021. Household and person variables are prepended to the incident

records.save outfile = "incident2021_DYextract.sav".Since the resulting file is incident level, it may be used toestimate victimizations just as with the previously createddata-year incident file.To be consistent with BJS publications prior to 2012,exclude series crimes (use only with data prior to 2012).select if ( V4019 ne 2 or missing(V4019) ).Exclude crimes outside the U.S. - this is a BJS convention.-21-Do this in order to tie to BJS publications for all years.select if ( V4022 ne 1 or missing(V4022) ).To be consistent with BJS publications from 2012 on (CriminalVictimizations, 2011), include series crimes and use the syntax below.Include up to 10 occurrences in a series crime.do if (v4019=2).recode v4016 (11 thru 996=10). end if.execute.if ( v4019 = 2 ) WGTVICDY = WGTVICDY * v4016.Weight the data by the adjusted victimization weight. weight by WGTVICDY.Run a frequency on the TYPE OF CRIME (V4529) variable.freq vars = V4529.RESULT - The total victimization count on a data-year basis is:15,607,478.*--- END SYNTAX.

Matching Files to Create the Collection Year Incident-Level

Extract FileCOMMENT The following is an SPSS example of matching the household,COMMENT person, and incident record-type files. It shows how theCOMMENT collection year incident-level extract file contained withCOMMENT this dataset was constructed.COMMENT For the example we assume that the household, person, andCOMMENT incident record-type files have been downloaded and renamed:COMMENT hh_rec2.sav, per_rec3.sav, and incid_rec4.sav respectively.COMMENT Match the household records onto the corresponding personCOMMENT records.match files/table = "hh_rec2.sav"/file = "per_rec3.sav"/by YEARQ IDHH .COMMENT Match the resulting file onto the corresponding incidentCOMMENT records.match files/table = */file = "incid_rec4.sav"/by YEARQ IDHH IDPER .execute.COMMENT In the above match files statements the "table" indicatesCOMMENT that the specified file should be used as a one-to-manyCOMMENT match. For example, a given household record may match toCOMMENT several person records, and the resulting household-personCOMMENT records may match to several incident records. However, theCOMMENT resulting household-person records may not match to anyCOMMENT incident records - most persons do not have crimeCOMMENT incidents. In those instances the "table" specification alsoCOMMENT prevents those records from being added to the resultingCOMMENT file. Hence, the resulting file will be comprised ofCOMMENT incident records with the household and person information-22-COMMENT prepended to them.COMMENT Recode the variable victimization weight for data yearCOMMENT (WGTVICDY) to zero to prevent inadvertently using it.recode WGTVICDY ( else = 0 ).COMMENT Save the resulting incident-level extract file.save outfile = "incident.sav"

/keep V2001 YEARQ IDHH IDPERV2002 to TYPEINTV_TIS7 V3001 to V3098 V4001 to SERIES_IWEIGHT.COMMENT The above save command finalizes the incident-level extractCOMMENT file and places the variables in a specified order.

Weights DetailsBecause the data collected by the National Crime Victimization Survey represent the total U.S. population 12years and older, records can be weighted to produce population estimates from the sample cases. Weights arecarried in survey records and are numbers which one adds or accumulates to obtain universe estimates ofparticular events. The final weight is a multiplier that indicates how many times a particular sample record is to becounted. Traditionally, the NCVS has provided three classes of weights -- household, personal, and incident --and the appropriate weight to select for an analysis depends on the type of crime, victim, or unit of analysisinvolved. There are multiple weights within each of these three classes of weights. These weights are defined as:• HOUSEHOLD WEIGHT: This weight is attached to the household record-type file and is the weight of the"Principal Person" in the household. In husband-wife households, this is the weight for the wife, excluding thewithin-household non-interview adjustment (see Components Of Weights below). For other households, thehousehold weight is that of the individual identified as owning, buying, or renting the dwelling (the "ReferencePerson"), excluding the within-household non-interview adjustment. This weight, after proper adjustment, is usedto generate the population of households which is also used to form the denominator in the calculation of propertycrime rates. Household weights are V2116 and WGTHHCY.• PERSON WEIGHT: Attached to the person record-type file, this variable provides an estimate of thepopulation represented by each person in the sample. Person weights are most frequently used to computeestimates of crime victimizations of persons in the total population. This weight, after proper adjustment, is usedto generate the population of persons which is also used to form the denominator in the calculation of personalcrime rates. Person weights are V3080 and WGTPERCY.• INCIDENT WEIGHT: This weight is attached to the incident record-type file. For personal crimes, it isderived by dividing the person weight of a victim by the total number of persons victimized during an incident asreported by the respondent. For property crimes, the incident weight and the household weight are the same,because the victim of a property crime is considered to be the household as a whole. The incident weight is mostfrequently used to compute estimates of the number of crimes committed against a particular class of victim (e.g.,by victim race or Hispanic origin, age, or sex).

VICTIMIZATION, HOUSEHOLD, AND PERSON WEIGHTSThe sample weights V2116 (household weight) and V3080 (person weight) included with the data have beenconstructed by the U.S. Census Bureau to be representative of the national household and population age 12 orolder at the time of interview. Since the sample is interviewed every six months, the sample weights must beadjusted for use in creating annual estimates to reflect the average population size during the entire referenceyear. The resulting adjusted weights are WGTHHCY (adjusted household weight) and WGTPERCY (adjustedperson weight). In addition, the value of the victimization weight depends on whether the crime is against ahousehold or a person.Six weights are included with the data that have all of the relevant adjustments incorporated to create nationalestimates of victimization, household counts, and person counts. Refer to the NCVS Technical Documentation(www.bjs.gov) for a detailed description of how each weight is created. Table 1 shows the distribution of these sixweights.-23-Table 1: 2022 National Crime Victimization Survey Weight Distributions

Weight Min 1stQuartile

Mean Median 3rdQuartileMax SumWGTPERCY 98 682 1,244 1,065 1,599 12,109 282,304,639

Person Victimization(SERIES_WEIGHT)309 1,337 3,113 2,194 3,468 60,908 6,786,936

Person Incident(SERIES_IWEIGHT)65 1,240 2,932 2,004 3,356 43,069 6,392,133WGTHHCY 83 533 913 796 1,215 5,326 131,259,549

Household Victimization(SERIES_WEIGHT)197 980 1,823 1,466 2,378 47,798 13,426,686

Household Incident(SERIES_IWEIGHT)197 980 1,823 1,466 2,378 47,798 13,426,686Source: U.S. Census Bureau data from the 2022 National Crime Victimization Survey public use file.Victimization and incident weights can be larger than sample unit weights because they are for different timeperiods. Each household and person is counted once per year, while victimizations are measured separatelyacross the two six month periods. Therefore, the unbounded victimization weights are twice as large as thehousehold and person weights for each individual sample unit. Incident weights are equal to victimization weightsunless the crime incident involved more than one person – a concept that only applies to personal crimes. It doesnot apply to household crimes because the victim of a household crime is counted as one victim: the affectedhousehold. The weighted sums represent the total person/household populations and total number ofvictimizations and incidents.The variables WGTVICCY is the adjusted victimization weight for collection year. This weight is included on theincident record-type file and used to calculate an estimate of victimizations. In a calculation of victimization ratethis weight is used to determine the numerator. WGTVICCY is not adjusted for series crimes.The variable WGTHHCY is the adjusted household weight for collection year. This weight is included on thehousehold record-type file and used to calculate an estimate of households covered by the NCVS. In a calculationof HOUSEHOLD victimization rate - which requires selecting only crimes against households (property crimes) -this weight is used to determine the denominator.The variable WGTPERCY is the adjusted person weight for the collection year. This weight is included on theperson record-type file and used to calculate an estimate of persons covered by the NCVS. In a calculation ofPERSON victimization rate - which requires selecting only crimes against persons (violent crimes and personallarceny) - this weight is used to determine the denominator.Adjusted household and person weights, included on the household and person record-type files respectively, willbe zero in certain circumstances. Noninterviewed households and persons will have zero weight for bothcollection year and data year. In addition, the adjusted household and person weights for the COLLECTIONYEAR will be zero for the last six months of the 18 month period that makes up an individual NCVS data year.This is because the collection year only includes the 12 months of interviews that occurred during the calendaryear.The adjusted victimization weight, included on the incident record-type file, for COLLECTION YEAR (WGTVICCY)will be zero for the last six months of the 18 month period that makes up an individual NCVS data year. Again,this is because the collection year only includes the 12 months of interviews conducted during the calendar year.In the incident record-type file, WGTVICCY will be greater than zero for the 12 months of the calendar year andzero for the remaining six months. Under the collection-year format, victimizations are counted based on whenthe interview occurred not on when the incident occurred, and they are counted for the 12 calendar months ofinterviews.In the incident record-type file, the adjusted victimization weight for DATA YEAR will be zero in somecircumstances. Under the data-year format a victimization is counted based on the year the crime incident-24-actually occurred, not when the interview was conducted. Consequently, if an interview was conducted inFebruary 2022 and a crime was reported as having occurred in October 2021, then the value of WGTVICDY for2022 will be zero. However, the weight will be greater than zero for the 2021 data year.The adjusted household, person, and victimization weights were added to the NCVS data so that analysts wouldnot need to compute them directly. Those interested in the calculations necessary to compute the adjustedweights should refer to the NCVS Resource Guide. The resource guide also contains victimization, household,and person counts for collection-year and data-year formats.An example of calculating crime rates using weighted counts of victimizations, households, and persons isincluded in the Crime Estimate Details section of this documentation.

SERIES INCIDENTS WEIGHTINGBJS uses victimization weight in reports to account for the number of persons present during an incident and forrepeat victims of series incidents. Incidents are distinguished from victimizations in that one criminal incident mayhave multiple victims or victimizations. SERIES_WEIGHT is the victimization weight adjusted for series crimes.SERIES_IWEIGHT is the incident weight adjusted for series crimes. The weight counts seriesincidents/victimizations as the actual number of incidents/victimizations reported by the victim, up to a maximumof 10 incidents/victimizations. Series victimizations/incidents are similar in type but occur with such frequency thata victim is unable to recall each individual event or describe each event in detail (for example, intimate partnerviolence or domestic violence may be a series victimization). Survey procedures allow NCVS interviewers toidentify and classify these similar victimizations/incidents as series and to collect detailed information on only themost recent incident in the series. In 2022, about 1.5% of all victimizations and 3.3% of all violent victimizationswere series incidents. Weighting series incidents/victimizations as the number of incidents/victimizations up to amaximum of 10 produces more reliable estimates of crime levels, while the cap at 10 minimizes the effect ofextreme outliers on the rates (see code for weight in the Crime Estimate Details section of this documentation).Additional information on the series enumeration is detailed in the report Methods for Counting High FrequencyRepeat Victimizations in the National Crime Victimization Survey, NCJ 237308, BJS website, April 2012,http://www.bjs.gov/content/pub/pdf/mchfrv.pdf.USE OF WEIGHTED V. UNWEIGHTED DATABJS analyses of NCVS data typically utilize weighted data. However, analysts have used both weighted andunweighted data to examine victimization-related research questions. Generally, using weighted data increasesvariance, but decreases sampling bias and bias that is introduced through unequal probabilities of selection andobservation. Weighted data are more typically used to compute representative population estimates ofvictimization dynamics and to compare victimization patterns over time.Unweighted data have been used to develop and explore predictive models of victimization. However, the analystmust assume that the model is insensitive to small temporal variations in the composition of the NCVS sampleand reporting rates and that variables in the model are unrelated to time, if multiple years of data are utilized (seeWeighting And Deriving Longitudinally Representative Data below).

COMPONENTS OF WEIGHTSThe final weight (household, V2116; person, V3080; or incident, SERIES_IWEIGHT) adjusts for unequalprobabilities of selection and observation, and for known age, sex, and race ratios based on the AdjustedDecennial Census Population Totals. Specifically, the final person and incident weights are the product of thevalues of the following six components weights. Household weights exclude the within-household non-interviewadjustment.

1. BASE WEIGHT: The inverse of the national sampling rate for the stratum of that unit (person or

household).

2. WEIGHTING CONTROL FACTOR: Adjusts for any sub-sampling due to unexpected events in the field,

such as new construction, area segments larger than anticipated, and other deviations from the overallstratum sampling rate.

3. HOUSEHOLD NON-INTERVIEW ADJUSTMENT: Adjusts for nonresponse at the household-level by

increasing the weights of interviewed households most similar to households not interviewed in terms of-25-race, MSA status of residence, and urban/suburban/rural status of residence. This inflates the weightvalue assigned to interviewed households so that they represent themselves and non-interviewedhouseholds. The non-interviewed cases are assigned a weight of zero, thereby excluding them frompopulation estimates.

4. WITHIN-HOUSEHOLD NON-INTERVIEW ADJUSTMENT: Adjusts for non-response at the person-level

by increasing the weight of interviewed persons most similar to persons not interviewed in terms ofregion, age, race, sex, and household composition. The adjustment inflates the weight value assigned tocompleted interviews, so that they represent themselves and sampled individuals who were notinterviewed. The non-interviewed cases are assigned a weight of zero.

5. FIRST STAGE RATIO ESTIMATES FACTOR: Adjusts for differences between characteristics of the

sample non-self-representing (NSR) primary sampling units (PSUs) and independent measures of thepopulation NSR PSUs. (For self-representing PSUs this factor is set to 1). This factor adjusts for PSUdifferences on region, MSA status, urban/suburban/rural status, and racial composition.

6. SECOND STAGE RATIO ESTIMATE FACTOR: A post-stratification factor defined for each person to

adjust for the difference between weighted counts of persons (using the above five weight components)and independent estimates of the number of persons, within certain age by race by sex categories. Theseindependent estimates are based on the Census population controls adjusted for the undercount.The weights discussed above (household, V2116; person, V3080; and incident, SERIES_IWEIGHT) are not thesame as the adjusted household, person, and victimization weights. The adjusted weights, both collection-yearand data-year versions, were calculated using V2116 and V3080 as necessary. It is the adjusted weights forhouseholds, persons, and victimizations (WGTHHCY, WGTHHDY, WGTPERCY, WGTPERDY, WGTVICCY, andWGTVICDY) that should be used to directly calculate counts of households, persons, and victimizations on eithera collection-year or data-year basis.

WEIGHTING AND DERIVING LONGITUDINALLY REPRESENTATIVE DATANot only are weights useful for calculating cross-sectional estimates, but they play a crucial role in ensuring thatany analysis of NCVS data over time is accurate. When the analyst plans to compare NCVS data across time, it isimportant to bear in mind the following cautions regarding the use of weights and unweighted data:The NCVS sample size has fluctuated over the years, particularly during the phase-in period of the NCVSquestionnaire. (During the completion of the NCVS phase-in during 1993, the number of sample cases alsochanged monthly from January to June.) Consequently, attempts to compare unweighted cases across yearsmay produce artifactual differences that are due to variations in sample size and not to any underlying temporalchanges in victimization patterns. Weights compensate for these variations in sample size across years. Anyattempt to compare victimization counts and patterns across time should utilize weighted data to eliminate samplesize variation as a confounding factor in the analysis.Although the NCVS has a rotating panel design, it is important to recognize that NCVS weights are cross-sectional, not longitudinal. That is, they are intended to provide estimates that are representative of the populationof individuals and households at the time an interview was taken. They do not adjust for changes in thecomposition of any panel of NCVS respondents over time or for panel attrition and should not be used for thispurpose in any analysis that involves controlling for these changes to a panel.

Variance Details

DIRECT VARIANCE CALCULATIONAlthough generalized variance functions are useful for calculating population estimates of NCVS data, they arenot well adapted to more complex statistical modeling and testing. A number of methods for calculating variancesfor complex sample designs directly from the data have been developed over the last decade, and these arebetter suited for statistical modeling.However, data files must provide either clustering information or "pseudo-strata" that will allow the dataset to besubsetted for multiple "replicate" calculations. NCVS public use files have been updated to provide variablescontaining pseudo-strata and half sample coding - V2117 and V2118. With this enhancement, software can nowbe used with the public use data files to calculate direct variances. Additional information can be found in theUser's Guide to National Crime Victimization Survey (NCVS) Direct Variance Estimation.

GENERALIZED VARIANCES-26-The 2022 data files contain 8 sets of replicate weights:

1. HHREPWGT1-HHREPWGT160 (6-month household replicate weights)
2. HHREPWGTCY1-HHREPWGTCY160 (household replicate weights adjusted for 12-month collection

period)

3. PERREPWGT1-PERREPWGT160 (6-month person replicate weights)
4. PERREPWGTCY1-PERREPWGTCY160 (person replicate weights adjusted for 12-month collection

period)

5. SERIESINCREPWGT1-SERIESINCREPWGT160 (series incident replicate weights)
6. SERIESVICREPWGT1-SERIESVICREPWGT160 (series victimization replicate weights)
7. INCREPWGT1-INCREPWGT160 (non-series incident replicate weights)
8. VICREPWGT1-VICREPWGT160 (non-series victimization replicate weights)

Generalized variance functions (GVF) provide a simple way to estimate the variance of any NCVS estimatebased on the value of the estimate. The Census Bureau creates GVF parameters by fitting all NCVS estimates toa model that predicts the relative variance of the crime rate as a function of the crime rate.The methodology used to form the replicate weights is not traditional orthogonal replicates, but instead uses Fay’smethodology. This requires an additional adjustment option in statistical software packages in order to accuratelycalculate standard errors.Example syntax code in SUDAAN is as follows -

REPWGT HHREPWGTCY1-HHREPWGTCY160/ADJFAY=4;Example syntax code in SAS is as follows -VARMETHOD=BRR(Fay);

WEIGHT WGTHHCY;repweights HHREPWGTCY1-HHREPWGTCY160;Before the 1992 NCVS redesign implementation, a set of standard errors was estimated using the TaylorExpansion Method and was generalized to be applicable to all survey estimates. Generalized curves wereproduced from subsets of computed standard errors, and formulae developed for these curves, which would allowanalysts to calculate the standard error of any survey estimate. These formulae and the constant applicable foreach year and general domain of crime have been provided for each year of published NCS data.In the 1992-1997 NCVS data, a Jackknife Repeated Replication procedure was used to estimate standard errors,and a new set of formulae and constants was provided to calculate standard errors. Study of NCVS data haddetermined that the design effect was not constant across all survey estimates and that the new proceduresprovided a more accurate method for computing standard errors. Beginning in 1998, a Balanced RepeatedReplication procedure was used to estimate standard errors.During GVF calculations, estimates are removed from processing if they have absolute studentized residualsabove 3.0 and absolute measure of influence (DFFITS) values above 0.22. A complete list of estimates that wereremoved from the data during iterative GVF modeling is included in the Source and Accuracy Statement for the2022 National Crime Victimization Survey.Four formulae are available to calculate the appropriate standard error, depending on the type of estimateinvolved. Each formula requires the analyst to use a set of constants that varies between year and domain ofcrime. Table 2 presents the current and two previous years' NCVS GVF parameters for overall and domainestimates of person and property crimes.-27-Comparisons of direct and GVF variance estimates of household and person victimizations are also included inthe Source and Accuracy Statement for the 2022 National Crime Victimization Survey. The GVF parametersperform reasonably well for small estimates, but tend to overestimate the variance for larger estimates. This ismost notable in the person domain estimates. When making comparisons to estimates from previous years, usethe year-to-year correlation coefficients in Appendix A of the Source and Accuracy Statement for the 2022National Crime Victimization Survey. Refer to the NCVS Technical Documentation for more on how to use GVFparameters and correlation coefficients to estimate variances. The 2013 NCVS Technical Documentation alsoincludes GVFs for years 2008 to 2013. Additional information can be found in the NCVS TechnicalDocumentation on the BJS website, www.bjs.gov.

Questionnaire Details

QUESTIONNAIRE PARTSRather than asking a simple yes/no question ("Have you been victimized by crime?"), the NCVS questionnaire ismade up of two parts: the "Screening Section" and the "Incident Report." The Screening Section describescriminal incidents using everyday language in order to ascertain whether the respondent has been victimizedduring the six months prior to interview. Any positive responses are followed by an Incident Report, a series ofdetailed questions on the incident itself. The screening section is a series of questions and cues describingvarious behaviors and features of a criminal incident. For example, to elicit reports of aggravated assault onescreen question asks: "Has anyone attacked or threatened you in any of these ways: with any weapon, forinstance, a gun or knife; with anything like a baseball bat, frying pan, scissors, or stick, by something thrown, suchas a rock or bottle..." Because the NCVS measures both property and personal crime, a separate ScreeningSection is administered for each. The Household Respondent (see Glossary) is asked to report on crimes againstthe household as a whole (e.g. burglary, motor vehicle theft) as well as personal crimes against him/herself. Othermembers of the household are asked only about personal crimes, although if they report crimes against thehousehold not already reported by the Household Respondent an Incident Report is filled out.If any Household or Individual screen question elicits a positive response, an Incident Report is filled out to gatherdetails of the victimization. The incident report is designed to obtain detailed data on the characteristics andcircumstances of the victimization. Items such as time and place of occurrence, injuries suffered, medicalTable 2: National Crime Victimization Survey Generalized Variance Function Parameters

Year ParameterSet #

Description a b c

2022 1 Overall Person CrimeEstimates-0.00017101 5,590 2.541

2 Person Crime DomainEstimates-0.00038364 3,991 6.208

3 Overall Property CrimeEstimates-0.00012374 2,536 1.196

4 Property Crime DomainEstimates-0.00012797 2,767 1.225

2021 1 Overall Person CrimeEstimates-0.00030659 2,185 4.992

2 Person Crime DomainEstimates-0.00049937 1,913 8.229

3 Overall Property CrimeEstimates-0.00009503 2,761 0.838

4 Property Crime DomainEstimates-0.00012910 2,601 1.239

2020 1 Overall Person CrimeEstimates-0.00026442 6,177 4.039

2 Person Crime DomainEstimates-0.00068320 2,338 11.25

3 Overall Property CrimeEstimates-0.00013449 2,998 1.256

4 Property Crime DomainEstimates-0.00015281 2,890 1.473Source: U.S. Census Bureau internal data from 2020-2022 National Crime Victimization Survey interviews.-28-expenses incurred, number, age, race, and sex of offenders, relationship of offenders to victim (stranger, casualacquaintance, relative, etc.), and other detailed data relevant to a complete description of the incident areincluded in the report. This information is used during subsequent computer processing to determine whether theincident fits into one of the crimes within the scope of the survey and to place the incident in the appropriatecategory (see Glossary for NCVS Crime Classification Schemes and Crimes).

FIELD REPRESENTATIVESField representatives' (FRs) instructions are explicit and detailed concerning the administration of thequestionnaires, adherence to question wording, and the use of probes. The questions and all short cuesaccompanying the questions are read in full. Field representatives are instructed to read these questions andcues only, and offer response categories only as a last resort. Field representatives are explicitly instructed toavoid reading the answer categories for the question, "How were you threatened?" if at all possible. Residue isused in the codebook documentation to designate those cases where the respondent could not or did not providean answer, or for some reason the response given did not fall into the predetermined categories.Field representatives receive extensive training prior to any field work. The quality of interviewing is maintained byperiodic direct observation of all field representatives, office edits of completed work, and verification of their workby re-interviewing a sample of completed sample units. FRs also measure errors in coverage of the populationresulting from missing units, incorrect address listing, etc. The Bureau of the Census circulates a series ofperiodic updates dealing with the field procedures followed by the National Crime Victimization Survey fieldrepresentatives.

Census Sample Information and NCVS Rotation Chart

Census Sample Information and NCVS Rotation ChartThe NCVS is a panel, or longitudinal, survey in which persons in each housing unit or GQ are interviewed everysix months, for a total of seven interviews. The NCVS interviews are conducted continuously throughout the yearin a design that divides the NCVS sample into seven rotating groups. Within each of the seven rotating groups,six panels are designated, each of which is interviewed in a given month and every six months thereafter, eitherin-person or by telephone, for a total of seven interviews. New sample units rotate into the sample on an ongoingbasis to replace outgoing households that have been in sample for the 3-year period.The rotation chart in Figure 2 illustrates the sampling and interviewing schedule for eighteen months of NCVSinterviewing. The years along the top indicate when the sample was selected, and the years and months alongthe left side are when the sample cases are interviewed. The two-digit numbers represent the panel and rotationnumbers assigned to each sample unit at the time of sampling. In July through December 2019, for example,2015 rotation 5 cases were removed from the sample and 2018 rotation 5 cases were added to the sample.Sample cases in panel 3 within 2018 rotation 5 (the two-digit number 35 in the 2018 column) began interviewingin September 2019, are interviewed again in March 2020, and will interview every six months until their seventhand final interviews September 2022 (not pictured).-29-Figure 2. NCVS Rotation Chart for January 2019 through June 2020

Interview Year

and Month

Sampling Year2015 2016 2017 2018 20192019 JANFEBMARAPRMAYJUN15 16-25 26-35 36-45 46-55 56-65 66-17 11-27 21-37 31-47 41-57 51-67 61-12 13-22 23-32 33-42 43-52 53-62 63-142434445464JULAUGSEPOCTNOVDEC16-26-36-46-56-66-17 11-27 21-37 31-47 41-57 51-67 61-12 13-22 23-32 33-42 43-52 53-62 63-14 1524 2534 3544 4554 5564 652020 JANFEBMARAPRMAYJUN17 11-27 21-37 31-47 41-57 51-67 61-12 13-22 23-32 33-42 43-52 53-62 63-14 1524 2534 3544 4554 5564 65162636465666Source: Bureau of Justice Statistics, National Crime Victimization Survey, 2016

Measuring Rape and Sexual AssaultIn 2016, The Government Accountability Office (GAO) issued findings and recommendations from study 16-546,“Sexual Violence Data: Actions Needed to Improve Clarity and Address Differences Across Federal DataCollection Efforts.” A key recommendation was that in order to enhance the clarity and transparency of sexualviolence data, federal agencies should ensure that details about the survey items used to generate theseestimates are made publicly available. This section describes the variables that feed into the NCVS measures ofrape and sexual assault obtained in variable v4529 (Type of Crime (TOC) Code, NEW).When classifying a crime as a rape or sexual assault, the Census Bureau assign the codes in a specific order ofimportance. Codes 1-4 (rape or sexual assault categories) are assigned first, and then codes 5-14 are assigned(in no particular order) before codes 15-16 (rape or sexual assault categories) are assigned. Finally, code 17 isassigned before codes 18-19 (rape or sexual assault categories) are assigned. Because of this specificity anddetail in assigning codes, a user may not get the same TOC as what the Census Bureau came up with.

Variable Name

and ValueDescription

TOC 1 – Completed RapeIFV4045 =1 Victim present? = Yes ANDV4094 = 1 How attacked? = Rape ORV4112 = 1 Injuries suffered? = Rape injuries THENV4529 = 1 Completed rape

TOC 2 – Attempted RapeIF V4045 =1 Victim present? = Yes ANDV4095 = 1 How attacked? = Attempted rape OR-30-V4113 = 1 Injuries suffered? = Attempted rape injuries ORV4078 = 1 How threatened? = Verbal threat of rape THENV4529 = 2 Attempted rape

TOC 3 – Sexual Attack with Serious AssaultIF V4045 = 1 Victim present? = Yes ANDV4096 =1 How attacked? = Sexual assault other than rape orattempted rape ORV4114 = 1 Injuries suffered? = Sexual assault other than rape orattempted rape ORV4082 = 1 How threatened? = Unwanted sexual contact with force(grabbing, fondling, etc.) ORV4083 = 1 How threatened? = Unwanted sexual contact withoutforce (grabbing, fondling, etc.) ORV4050 = 1 What was the weapon? = Any weapon type ORV4050 = 3 What was the weapon? = Indicates weapon but noweapon type was provided ORV4050 = 7 What was the weapon? = Indicates weapon but no otherspecify was provided ORV4119 = 1 Injuries suffered? = Knocked unconscious ORV4120 = 1 Injuries suffered? = Bruises, black eye, cuts, scratches,swelling, chipped teeth ANDV4138 >= 2 and <=200 Hospital? = Hospital stay between 2-200 days THENV4529 = 3 Sexual attack with serious assault

TOC 4 – Sexual Attack with Minor AssaultV4045 = 1 Victim present? = Yes ANDV4096 = 1 How attacked? = Sexual assault other than rape orattempted rape ORV4114 = 1 Injuries suffered? = Sexual assault other than rape orattempted rape ORV4082 = 1 How threatened? = Unwanted sexual contact with force(grabbing, fondling, etc.) ORV4083 = 1 How threatened? = Unwanted sexual contact withoutforce (grabbing, fondling, etc.) ANDV4119 = 1 Injuries suffered? = Knocked unconscious ORV4120 = 1 Injuries suffered? = Bruises, black eye, cuts, scratches,swelling, chipped teeth ORV4110 = 1 Injuries suffered? = Indicated injury but did not specifyinjury type ORV4082 = 1 How threatened? = Unwanted sexual contact with force(grabbing, fondling, etc.) ORV4083 = 1 How threatened? = Unwanted sexual contact withoutforce (grabbing, fondling, etc.) ANDV4089 = 1 How threatened? = Followed or surrounded ORV4090 = 1 How threatened? = Tried to hit, slap, knock down, grab,hold, trip, jump, push, etc. THEN-31-V4529 = 4 Sexual attack with minor assault

TOC 15 – Sexual Assault without InjuryV4045 = 1 Victim present? = Yes ANDV4096 = 1 How attacked? = Sexual assault other than rape orattempted rape ORV4082 = 1 How threatened? = Unwanted sexual contact with force(grabbing, fondling, etc.) THENV4529 = 15 Sexual assault without injuryTOC 16 – Unwanted Sexual Contact without Force (classified as sexual assault)V4045 = 1 Victim present? = Yes ANDV4083 = 1 How threatened? = Unwanted sexual contact withoutforce THENV4529 = 16 Unwanted sexual contact without force

TOC 18 – Verbal Threat of Rape (classified as rape)V4045 = 1 Victim present? = Yes ANDV4078 = 1 How threatened? = Verbal threat of rape THENV4529 = 18 Verbal threat of rapeTOC 19 – Verbal Threat of Sexual Assault (classified as sexual assault)V4045 = 1 Victim present? = Yes ANDV4081 = 1 How threatened? = Verbal threat of sexual assault otherthan rape THENV4529 = 19 Verbal threat of sexual assaultNCVS measurement of rape or sexual assaultThe NCVS uses a two-stage measurement approach in the screening and classification of criminal victimization,including rape or sexual assault. In the first stage of screening, survey respondents are administered a series ofshort-cue screening questions designed to help respondents think about different experiences they may have hadduring the reference period. (See NCVS-1 at https://bjs.ojp.gov/content/pub/pdf/ncvs22_bsq.pdf) This designimproves respondents’ recall of events, particularly for incidents that may not immediately come to mind ascrimes, such as those committed by family members and acquaintances. Respondents who answer affirmativelyto any of the short-cue screening items are subsequently administered a crime incident report (CIR) designed toclassify incidents into specific crime types. (See NCVS-2 at https://bjs.ojp.gov/content/pub/pdf/ncvs22_cir.pdf)First stage of measurement. Two short-cue screening questions are specifically designed to target sexualviolence:

1. Other than any incidents already mentioned, has anyone attacked or threatened you in any of these

ways—a. with any weapon, such as a gun or knifeb. with anything like a baseball bat, frying pan, scissors, or stickc. by something thrown, such as a rock or bottled. by grabbing, punching, or chokinge. any rape, attempted rape, or other types of sexual attackf. any face-to-face threatsg. any attack or threat or use of force by anyone at all? Please mention it even if you are not-32-certain it was a crime.

2. Incidents involving forced or unwanted sexual acts are often difficult to talk about. Other than any

incidents already mentioned, have you been forced or coerced to engage in unwanted sexual activityby—a. someone you did not knowb. a casual acquaintancec. someone you know well?Respondents may screen into a CIR if they respond affirmatively to another short-cue screening question. Forinstance, a separate screening question cues respondents to think of attacks or threats that took place in specificlocations, such as at home, work, or school. Respondents who recall a sexual victimization that occurred athome, work, or school and answer affirmatively would be administered a CIR even if they did not respondaffirmatively to the screening question targeting sexual violence.Second stage of measurement. The CIR is used to collect information on the attributes of each incident. Thekey attributes of sexual violence that are used to classify a victimization as a rape or sexual assault are the typeof attack and physical injury suffered. Victims are asked if “the offender hit you, knock[ed] you down, or actuallyattack[ed] you in any way;” if “the offender TR[IED] to attack you;” or if “the offender THREATEN[ED] you withharm in any way?” The survey participant is classified as a victim of rape or sexual assault if he or she respondsaffirmatively to one of these three questions and then responds that the completed, attempted, or threatenedattack was—• rape• attempted rape• sexual assault other than rape or attempted rape• verbal threat of rape• verbal threat of sexual assault other than rape• unwanted sexual contact with force (e.g., grabbing or fondling)• unwanted sexual contact without force (e.g., grabbing or fondling).If the victim selects one of these response options to describe the attack, he or she is also classified as a victim ofrape or sexual assault if the injuries suffered as a result of the incident are described as—• rape• attempted rape• sexual assault other than rape or attempted rapeCoercion. The CIR does not ask respondents if psychological coercion was used, nor make any reference to thevictim being unable to provide consent (e.g., in incidents involving drugs or alcohol use). One screening questiontargeted to rape and sexual assault asks respondents if force or coercion was used to initiate unwanted sexualactivity. The final classification of incidents by the CIR results in the following definitions of rape and sexualassault used in the NCVS:Rape. Coerced or forced sexual intercourse. Forced sexual intercourse means vaginal, anal, or oral penetrationby the offender(s). This category could include incidents where the penetration was from a foreign object such asa bottle. It includes attempted rape, threatened rape, male and female victims, and both heterosexual and same-sex incidents.Sexual assault. A wide range of victimizations, separate from rape, attempted rape, or threatened rape. Thesecrimes include attacks or threatened attacks involving unwanted sexual contact between the victim and offender.Sexual assaults may or may not involve force and include such things as grabbing or fondling.-33-GlossaryThis glossary describes definitions and procedures currently employed in the National Crime Victimization Survey(NCVS). These definitions and procedures reflect a major redesign of the survey implemented in two stages:"Near-Term" changes and "Long-Term" changes. Generally, the "Near-Term" changes refer to improvementsmade to the Crime Incident Report and implemented in 1986. The "Long-Term" changes refer to improvementsmade to the Screening Section that were fully implemented in 1993 and where "Long-Term" changes created abreak in series by affecting the likelihood that a victimization would be reported. Data collected before this breakare referred to as NCS, and data gathered afterwards as NCVS. Specific changes are noted (by "NCS NOTE"and "NCVS NOTE") where relevant and indicate when the changes were implemented in the field. See NCVSRedesign document in the NCVS Resource Guide for a description of why and how these changes were made.A

ADDRESS SEGMENTSee: SegmentAGEAge is determined by asking the respondent for month, day, and year of birth. From this birth date, the fieldrepresentative determines the respondent's age as of the last day of the month previous to the interview month.The respondent is asked to verify the calculated age. This age is used to determine a respondent's eligibility forthe survey according to the following rules:

1. under 12: not interviewed (though the number of such children living in the household is asked of the

household respondent)

2. 12 and older: universe for interviews
3. 16 and older: additional questions are asked

NCS NOTE: Prior to July, 1986, 12- and 13-year-olds were interviewed by proxy respondent (a respondent whoanswers questions for another person). As of July 1986, respondents age 12 and older are interviewed directly,unless they are ill or otherwise unavailable, or a knowledgeable adult objects to the interviewing of 12and 13-year-olds in the household.See also: Interview, Type ofASSAULTThe NCVS defines assault as an unlawful physical attack, whether aggravated or simple, upon a person. Itincludes attempted assaults with or without a weapon, but excludes rape, attempted rape, and sexual assaults(these are categorized separately). Assaults also exclude attacks involving theft or attempted theft (these areclassified as robbery). Severity of assaults are classified into two major sub-categories:

1. Simple Assault: an attack without a weapon resulting either in minor injury (e.g., bruises, black eyes,

cuts, scratches, swelling) or an undetermined injury requiring less than two days of hospitalization. Italso includes attempted assault without a weapon and verbal threats of assault.

2. Aggravated Assault: an attack or attempted attack with a weapon, regardless of whether or not an injury

occurred, and attack without a weapon when serious injury results. Serious injury includes brokenbones, loss of teeth, internal injuries, loss of consciousness, and any injury requiring two or more days ofhospitalization.See also: Rape, Sexual Assault, Crimes, Robbery

ARMED FORCESSee: Military StatusATTACK-34-Some form of actual physical contact between the victim and the offender. If something was thrown at the victimbut did not hit him/her or if s/he was shot at but not struck by the bullet, s/he is not considered to have beenattacked, but threatened.See also: ThreatATTEMPTDefinitions of attempted crimes vary by type of crime. See entries for individual crimes for definition.

AUTO THEFTSee: Motor Vehicle TheftB

BASIC SCREEN QUESTIONNAIREA set of questions used to elicit reports of crimes against the household and its members (for a list of the types ofcrimes measured by the NCVS, see CRIMES). The screening is also used to obtain characteristics of thehousehold and individual household members 12 years of age and older. The screen questionnaire is composedof several types of items:

1. Cover page for household identification items (e.g.: address, segment number) and general information

(e.g.: number of members in the household, type of living quarters)

2. Personal characteristics items for obtaining general information on persons (e.g.: age, marital status,

sex, race)

3. Household screen questions, asked only of the Household Respondent, to elicit information on whether

any property crimes were committed during the reference period

4. Individual screen questions, asked of all household members 12 and older including the Household

Respondent, to elicit information on whether any crimes against individual household members werecommitted during the reference period

5. Questions dealing with household and individual vulnerability to crime.

The household and individual screen questions are designed to describe all crimes within the scope of the NCVSusing everyday language (not legal or technical definitions). Multiple examples of features likely to accompany acriminal incident are provided as cues to jar respondents' memories and to provide a more complete picture of thekinds of incidents included in the survey. If a respondent answers yes to any of the screen questions or cues, acomplete Incident Report is filled out and the crime is classified according to the NCVS Crime Classificationscheme.See also: Household Screen Questions, Individual Screen Questions, Questionnaire Forms, Interviewing

Sequence, NCVS Crime ClassificationNCVS NOTE: Beginning in January 1993, significant changes to the household and individual screen sectionswere implemented. See the NCVS Redesign document in the NCVS Resource Guide for a discussion of thesechanges.

BOUNDING PROCEDUREA technique used in the NCVS to establish a reference point during the first interview to avoid recording duplicatereports of incidents in subsequent interviews. The first interview conducted at the household is considered abounding interview. Data collected from this interview do not become part of the final dataset but are comparedwith incidents reported in the second interview. If the interviewer determines that a newly-reported incident is thesame incident that has already been reported in the bounding interview, the incident is eliminated. Incidentsreported in this second interview then become the new reference point for comparison with incidents reported inthe third interview, and so on.NCVS NOTE: Reporting procedures have been slightly modified. Previously, interviewers collected information on-35-incidents only within the 6-month reference period which ends on the last day of the month preceding theinterview if respondents reported very recent incidents -- those that occurred after the reference period ended butbefore the interview actually took place -- details on the incident were not recorded at that time but during thesubsequent interview. Deferring the Incident Report may have resulted in less accurate recall during thesubsequent interview. To address this concern, NCVS procedures were modified to allow an Incident Report tobe filled out even if the crime occurred after the 6-month reference period. Data on these incidents are held andtabulated along with data from the subsequent interview.NOTE: In the NCS National Samples of 1972 and the first half of 1973, the first-time interviews were unbounded.Bounded interviews started in January 1973 and July 1973 for the first rotation groups; thus, by July 1973,bounded interviews were being conducted in all returning rotation groups. Because of the large number ofunbounded interviews, the 1972 National Sample files have not been released.Before 2007, the initial interview for a person or sample unit was excluded from the estimates. This interview wasunbounded, meaning it did not have a prior interview to frame the reference period. In order to minimize over-reporting, the first interview was not used in estimates and was only used as a bounding interview for the secondinterview. For budgetary reasons, since January 2007, the bounding interview has been included in thecomputation of annual estimates. A bounding weighting adjustment is applied to compensate for the use of thefirst interview and the potential inclusion of victimizations that are out of the reference period. New constructioncases and replacement households entering the sample during the second through seventh interview are notbounded, and do not receive the bounding weighting adjustment. See also: Sample Rotation, Reference PeriodBURGLARYOne of the six major types of crimes covered by the NCVS, burglary is unlawful or forcible entry or attemptedentry of a residence. This crime usually, but not always, involves theft. The illegal entry may be by force, such asbreaking a window or slashing a screen, or may be without force by entering through an unlocked door or anopen window. As long as the person entering has no legal right to be present in the structure, a burglary hasoccurred. Furthermore, the structure need not be the house itself for a burglary to take place; illegal entry of agarage, shed, or any other structure on the premises also constitutes household burglary. If breaking and enteringoccurs in a hotel or vacation residence, it is still classified as a burglary for the household whose member ormembers were staying there at the time the entry occurred.See also: Crimes, Forcible Entry, Unlawful Entry Without Force, Attempted Forcible EntryC

CENTRAL CITYThe incorporated city (or cities) which is central to an urbanized area and/or a Metropolitan Statistical Area(MSA). It is bounded by city limits. A central city or the combined central cities must have a population of 50,000or more.See also: Urbanized Area, Metropolitan Statistical Area, Place Description Code

CHECK DIGITIn the NCVS National Sample, the check digit is computed on the nine digits of the sample number PSU, andsegment number. The check digit is part of the control number which uniquely identifies a sample unit.See also: Control Number Since July 1986, the five elements of the control number have been scrambled topreserve confidentiality.

CHILDREN UNDER 12, CRIMES AGAINSTBecause of the complexity of the NCVS questionnaire and the standardized interviewing procedures, crimesagainst children under 12 are not measured by the NCVS. In keeping with NCVS crime definitions, however,there is one exception: if items belonging to a child were stolen and there was no contact between the offenderand a household member, the incident is treated as a property crime, provided the incident occurred at or nearthe parent's place of residence.-36-

CLASS OF WORKERThe employment status of a respondent. Categories include:

1. Private Employee: persons who work for a private employer for wages, salary, commission, tips, piece-

rates, or pay in kind. This applies regardless of the occupation at which the employee worked, whethergeneral manager, file clerk, or porter. Also includes persons working for pay for settlement houses,churches, unions, and other private non-profit organizations.

2. Government Employee: persons who work for any branch of federal, state or local governments. This

includes persons who were elected to paid federal offices and members of the Armed Forces. It alsoincludes employees of international organizations and employees of foreign governments.

3. Self-Employed: persons who work for profit or fees in own business, farm, shop, office, etc. This does

not apply to managers, superintendents, or other executives hired to manage a business.

4. Working Without Pay: persons working without pay on a farm or business operated by a related member

of the household. Room and board and a cash allowance are not counted as pay for these familyworkers; however, if the worker receives money which is definitely considered to be wages for workperformed, he or she should be considered a private employee.See also: Industry Code, Occupation CodeCLOSEOUTThis refers to the time period in which either interviewers or the regional offices try to obtain interviews from non-interviewed persons prior to transmitting the questionnaire data for keying.CLUSTERA group of four housing units. Clustering is done so that selected sample units are in relatively close geographicproximity in order for interviews to be carried out in an efficient manner. Because of this clustering, the varianceestimates are slightly larger than they would be if the sample design used a simple random sample selectionscheme within the sample PSUs.See: Segment

COLLECTION QUARTERA calendar quarter of the year (three months) during which interviews are conducted in the NCVS. For example,the first collection quarter consists of interviews conducted in January, February, and March. The six-monthreference period for the January interviews is July 1 through December 31. The reference period for Februaryinterviews is August 1 through January 31, and so on. Because the reference period is six months long andbecause interviews are bounded (see "Bounding" above), 18 months of data collection (or 6 Collection Quarters)are necessary to yield one full year's worth of (bounded) data under the DATA-YEAR format.See also: Reference Period, Bounding Procedure (NCVS Note)

COMPLETE SAMPLE FILES NCVSData files containing data for both interviewed and non-interviewed households, victims and non-victims. The filesare hierarchical in structure, consisting of four record types:

1. Address ID records
2. Household records
3. Person records
4. Incident records (including series incidents)

The National Complete Sample files are arranged by the quarter in which the interviews were conducted(collection quarters).

COMPUTER-ASSISTED TELEPHONE INTERVIEWING (CATI)-37-Computer-Assisted Telephone Interviewing (CATI) involves bringing interviewers into a centralized telephonefacility to conduct the survey over the phone by reading questions off of a computer screen and enteringresponses directly into the computer. CATI stores information provided by the respondent as the surveyprogresses, automatically presenting the interviewer with the appropriate next question in the survey sequence.CATI software is designed with automated internal consistency checks to alert the interviewer of responses thatmay need to be clarified with the respondent. The CATI technique is contrasted with the field technique, in whichinterviewers around the country conduct telephone or face-to-face interviews and record responses in pencil on aprinted questionnaire. CATI is not utilized for the following circumstances:

1. All NCVS interviews for the first (of seven) scheduled interviews for households in sample will be

conducted in person so as establish and maintain rapport with respondents. No household will receiveits initial NCVS contact from a CATI facility.

2. Respondents without access to a telephone or who refuse during the interview to receive subsequent

interviews by telephone will continue to be interviewed in person.

3. To minimize disruption of field interviewer staffing and workloads and to avoid elimination of some

single-interviewer Primary Sampling Units (PSUs), CATI is fully implemented only in PSUs that arecovered by more than one field representative. On single field representative PSUs, only half are eligiblefor CATI, and in these PSUs only half the segments are assigned to CATI interviewing.NCS/NCVS NOTE: Phase-in of CATI began in 1987. From 1987 to 1992, 5 to 10 percent of the NCS sampleused CATI. From 1992 on, CATI was used for 30 percent of the NCVS sample. Since 2007, CATI has beendiscontinued as a mode.CONFIDENTIALITYThe law (U.S. Code 34, Sections 10131, 10231, and 10134) provides that all information given to the U.S. U.S.Census Bureau for this survey must be held in confidence and may be used only for statistical purposes. TheBureau may not release any statistics which might disclose the identity of any individual or household. Thisprotection explains why public data often do not provide the level of detail many researchers might wish. Currentrestrictions prohibit identification of less than 0.5% of the sample, when releasing individual or microdata.

CONTROL CARDThe control card (sometimes referred to as the "cover sheet") is one of four basic forms used to collect the NCVSdata. It is the first form the interviewer completes during the initial interview and is updated on each subsequentinterview as long as the household is in the sample. It contains the address of each sample unit and the basichousehold data, such as the names of all persons living there and their age, race, sex, marital status, education,etc. In addition, such items as family income, tenure of the unit, and pertinent information about non-interviewedunits are included on the Control Card. The Control Card also serves as a record of visits, telephone calls,reasons for interview and non-interview, etc.See also: Interviewing Sequence, Questionnaire Forms

CONTROL NUMBERA control number is assigned to each sample unit. In the National Sample, the control number is composed of fiveelements: the sample, PSU, Segment number, Check Digit, and Serial Number. Since July 1986, these fiveelements have been scrambled together to preserve confidentiality.See also: PSU, Segment Number, Check Digit, Serial Number, Identification Code, Sample Designation

CRIME INCIDENT REPORTA set of questions used to gather details (e.g.: location of incident, relationship of victim to offender, extent ofinjuries and/or economic loss) on crimes reported in the Basic Screen Questionnaire. This information is used tocategorize the incident (see NCVS Crime Classification). Details from the Incident Report also serve as adatabase for analysis of crime (e.g.: susceptibility to crime). One Crime Incident Report is completed for eachincident reported in answer to the screen questions (with the exception of "Series Crimes;" see the section - Typeof Research Instrument for details on series crimes incident reports). The following conditions must be met beforea series incident is reported: (1) incidents must be of the same type or very similar in detail; (2) there must be at-38-least six incidents in the series (this threshold was changed from three incidents to six incidents in January 1993);and (3) the respondent must not be able to recall dates and other details of the individual incidents well enough tohave reported them separately. Interviewers are instructed to try through probing to get individual reportswhenever possible and only accept series reports as a last resort.See also: Questionnaire Forms, Series Incident Report, Interviewing SequenceNCS NOTE: The Crime Incident Report was significantly modified in 1986 to provide more information oncharacteristics of the crime incident. See the NCVS Redesign document in the NCVS Resource Guide.CRIMESThe National Crime Victimization Survey focuses on measuring the extent of victimization by the major crimescommonly referred to as rape, robbery, assault, burglary, motor vehicle theft, and theft. The survey does notprovide measures for other offenses such as murder, kidnapping, shoplifting, fraud, gambling, etc.NCVS NOTE: Prior to January, 1993, the NCVS defined 32 distinct types of crime. The NCVS modificationsredefined some crimes (Theft and Series Crimes) and added crimes not previously measured (Sexual Assaultother than Completed and Attempted Rape, and Vandalism). Since 2009, the NCVS no longer collectsinformation about vandalism. See the NCVS Redesign document in the NCVS Resource Guide.

TYPES OF CRIME, LISTED IN ORDER OF SERIOUSNESS

(The Seriousness Hierarchy)

1. Crimes of Violence
1. Completed Rape
2. Attempted Rape
3. Sexual Assault with Serious Assault
4. Sexual Assault with Minor Assault
5. Completed Robbery with Injury from Serious Assault
6. Completed Robbery with Injury from Minor Assault
7. Completed Robbery without Injury
8. Attempted Robbery with Injury from Serious Assault
9. Attempted Robbery with Injury
10. Attempted Robbery without Injury
11. Completed Aggravated Assault with Injury
12. Attempted Aggravated Assault with Weapon
13. Threatened Assault with Weapon
14. Simple Assault Completed with Injury
15. Sexual Assault without Injury
16. Unwanted Sexual Contact without Force

NCS Note: This category did not exist during Quarters 1 and 2, 1992. Comparisons of the 1992 data for this crimecategory with future years should be avoided.

17. Assault without Weapon without Injury
18. Verbal Threat of Rape
19. Verbal Threat of Sexual Assault
20. Verbal Threat of Assault
2. Purse-snatching/pocket-picking
21. Completed Purse-Snatching
22. Attempted Purse-Snatching
23. Completed Pocket-Picking
3. Property Crimes
31. Completed Burglary, Forcible Entry
32. Completed Burglary, Unlawful Entry without Force
33. Attempted Forcible Entry
40. Completed Motor Vehicle Theft

-39-

41. Attempted Motor Vehicle Theft
54. Completed Theft Less than $10
55. Completed Theft $10 to $49
56. Completed Theft $50 to $249
57. Completed Theft $250 to Greater
58. Completed Theft Value NA
59. Attempted Theft

See also: NCVS Crime Classification SchemeD

DATA COLLECTION OR ENUMERATION PERIODThe length of time allotted to interviewers for completing their assigned workload for a survey or census;commonly referred to as the interview period. For the NCVS, households are enumerated during the first twoweeks of every month.See also: Household Status, Enumeration

DISABILITY STATUSThe NCVS defines disability as the product of interactions among individuals' bodies; their physical, emotional,and mental health; and the physical and social environment in which they live, work, or play. A disability existswhere this interaction results in limitations of activities and restriction to full participation at school, work, home, orin the community.EEDSee: Enumeration District (ED)EDUCATIONCurrent education level of household members listed on the control card. Respondents are asked the highestgrade or year of regular school attended, and whether that year was completed. Regular schools include allgraded public, private, parochial schools, colleges, universities, and professional schools which advance a persontoward an elementary or high school diploma, or a college degree. Regular schools do not include vocational,trade, business, correspondence, or other specialized schools, unless credits obtained are accepted in theregular school system. For persons still attending regular school, highest grade attended is the one in which theyare currently enrolled. Education questions, which are on the control card, are asked routinely at the 1st, 3rd, 5th,and 7th visits only.

EDUCATIONAL ATTAINMENTThe highest year completed. This is a recoded variable for use in BJS published reports. It is classified as follows:Never attended or kindergarten, Elementary, High School, College, and N.A. Post-graduate work is coded as"College."ENUMERATIONThe process of obtaining information about selected units of study in a survey or census.

ENUMERATION DISTRICT (ED)A small geographic area with well-defined boundaries set up for the 1990 Census and containing about 250households. A single PSU contains numerous EDs. In the NCVS National Sample, each selected ED issubdivided into clusters of four households, and a sample of these clusters is systematically selected. All four-40-households in the cluster are selected to be in the sample.See also: Primary Sampling Unit (PSU), Cluster

ENUMERATION PERIODSee: Data Collection PeriodESTIMATEA numeric value obtained from a sample used to describe the measurement which would be obtained if acomplete census were carried out. An estimate of a total for a population is obtained by multiplying the value for asampled record by a systematically designed weight which has the effect of inflating the sample count torepresent the total population. Multiple factors are used to compute this weight (see the section on - Weights).Basically, the sample count is inflated by the reciprocal of its probability of selection. Each estimate has its ownunique variance and standard error which express the precision of the estimate.See also: Variance, Weight, Standard ErrorETHNICITYA household respondent's self-reported statement of the national, cultural, or linguistic group with which eachmember of the household identifies. Currently, the NCVS gathers data on Hispanic ethnicity only.See also: Race

ETHNICITY OF OFFENDERThe ethnicity of offender as perceived and reported by the respondent. Categories are whether offender wasHispanic or Latino. When recording the ethnicity of multiple offenders, categories are: mostly Hispanic or Latino,mostly non-Hispanic, equal number of Hispanic and non-Hispanic, and don't know.See also: Race of offender

EXTRA UNITSHousehold units or OTHER units which are discovered at time of interview or during updating and are not alreadyentered on the listing sheet for the segment.

1. Area Segments:

a. In Area segments, EXTRA units must be in the same structure or on the same property as thesample unit being interviewed.b. In Special Places within Area segments, an EXTRA unit must be within the specific address of theoriginal unit.

2. In Address and Special Place segments, EXTRA units must be within the specific address of the sample unit

being interviewed, but need not be within the same structure or on the same property.

3. In Permit segments, EXTRA units must be within the specific address of the sample unit being interviewed

and must be within the same structure. When housing units or OTHER units which qualify as EXTRA unitsare discovered, they are interviewed and allocated to the sample without bias.See also: Segment, Housing Unit, Other Unit, Special Places, Listing, Household Status, UpdatingF

FINAL WEIGHTA final adjustment factor used to inflate each sample observation so that it properly represents the population.The final weight is a factor of six component weights and ratio adjustments (see section Weights). For personalcrimes, the final weight equals the person weight times five adjustment factors; for property crimes, the finalweight equals the household weight times these adjustment factors.-41-See also: Household Weight, Person Weight

FORCIBLE ENTRYA form of burglary in which force is used to gain entrance; e.g., by breaking a window or slashing a screen.Forcible entry may or may not be accompanied by theft or property damage. Evidence of force must be physicaland visible, i.e., able to be seen after the occurrence of the incident. An open or unlocked door is not physical,visible evidence.See also: Burglary, Unlawful Entry Without Force, Attempted Forcible EntryG

GROUP QUARTERSCensus term used to describe living quarters occupied by five or more persons unrelated to the head ofhousehold. Quarters with no designated head but with six or more unrelated persons are also group quarters.Some quarters occupied by fewer than six persons may also be group quarters by definition; e.g., living quartersin dormitories occupied by students. Group quarters were selected from Census listings.See also: Housing Unit, Other Unit, Special PlaceH

HATE CRIMEFrom 2003-2009, hate crime variables V4526AA (source code 910), V4526HA1 (Source Code 915), V4526HA2(SC916), V4526HA3 (SC917), V4526HA4 (SC918), V4526HA5 (SC919), V4526HA6 (SC920), V4526HA7 (SC921), V4526HA8 (SC922), and V4526HA9 (SC908), were suppressed from the public use files. During this time,the hate crime recode (V4526I) was created to designate an incident as a hate crime, a possible hate crime, ornot a hate crime. Beginning in 2010, all hate crime variables were included in the public use files and the hatecrime recode (V4526I) was removed. The statements below can be used to create a recode comparable to thehate crime recode. Please note that in keeping with the definition of hate crime used by the Hate Crime StatisticsAct (HCSA) of 1994 (28 U.S.C '534'), crimes that were committed because of gender bias were not included inthe hate crime recode. In 2009, the HCSA was expanded to include crimes committed because of bias againstthe victim's gender or gender identity.The hate crime recode is set to 1 (incident is a hate crime) when there is an entry of 1 in any of the followingvariables: V4526A, V4526B, V4526C, V4526D (SC896-SC899), V4526F (SC901), V4526G (SC911), V4526H(SC913), and there is also an entry of 1 in V4526HA2 (SC916), V4526HA3 (SC917), or V4526HA4 (SC918).The hate crime recode is set to 2 (incident is a possible hate crime) when there is an entry of 1 in any of thefollowing variables: V4526A, V4526B, V4526C, V4526D (SC896-SC899), V4526F (SC901), V4526G (SC911),V4526H (SC913), and there is also an entry of 1 in V4526HA5 (SC919), V4526HA6 (SC920), or V4526HA7(SC921), but there is not an entry of 1 in any of the following variables: V4526HA2 (SC916), V4526HA3 (SC917),and V4526HA4 (SC918).The hate crime recode is set to 3 (incident is not a hate crime) when there is an entry of 1 in any of the followingvariables: V4526A, V4526B, V4526C, V4526D (SC896-SC899), V4526F (SC901), V4526G (SC911), V4526H(SC913), but there is not an entry of 1 in any of the following variables: V4526HA2 (SC916), V4526HA3 (SC917),V4526HA4 (SC918), V4526HA5 (SC919), V4526HA6 (SC920), and V4526HA7 (SC921). The hate crime recodeis also set to 3 when there is not an entry of 1 in any of the following variables: V4526A, V4526B, V4526C,V4526D (SC896-SC899), V4526F (SC901), V4526G (SC911), and V4526H (SC913), but there is an entry of 1 inV4526E (SC900) or an entry of 1 or 2 in V4526AA (SC910).The hate crime recode is set to 4 (don't know lead-in of V4526AA (SC910)) when there is not an entry of 1 in anyof the following variables: V4526A, V4526B, V4526C, V4526D (SC896-SC899), V4526F (SC901), V4526G(SC911), and V4526H (SC913), and there is not an entry of 1 in V4526E (SC900) but there is an entry of 3 inV4526AA (SC910).-42-The hate crime recode is set to 8 (no valid entry in V4526AA (SC910)) when there is not an entry of 1 in any ofthe following variables: V4526A, V4526B, V4526C, V4526D (SC896-SC899), V4526F (SC901), V4526G(SC911), and V4526H (SC913), and there is not an entry of 1 in V4526E (SC900) but there is an entry of 8 inV4526AA (SC910).The hate crime recode is set to 9 (out of universe) when there is not an entry of 1 in any of the following variables:V4526A, V4526B, V4526C, V4526D (SC896-SC899), V4526F (SC901), V4526G (SC911), and V4526H (SC913),and not an entry of 1 in V4526E (SC900) and not an entry of 1, 2, 3, or 8 in V4526AA (SC910).Following is a list of all the hate crime related variables.

Variable Source Code DescriptionV4526A SC896 HATE CRIME: RACE (START 2003 Q1)V4526B SC897 HATE CRIME: RELIGION (START 2003 Q1)V4526C SC898 HATE CRIME: ETHNICITY (START 2003 Q1)V4526D SC899 HATE CRIME: DISABILITY (START 2003 Q1)V4526E SC900 HATE CRIME: GENDER (START 2003 Q1)V4526F SC901 HATE CRIME: SEXUALITY (START 2003 Q1)V4526HA9 SC908 EVER TELL POLICE YOU THOUGHT INCIDENT WAS A HC (START 2010Q1)V4526AA SC910 SUSPECT INCIDENT JUST DISCUSSED WAS HATE CRIME OR CRIME OF

PREJUDICE OR BIGOTRY (START 2010 Q1)V4526G SC911 HATE CRIME: ASSOCIATE (START 2003 Q1)V4526H SC913 HATE CRIME: PERCEPTION (START 2003 Q1)

V4526HA1 SC915 EVIDENCE OF HC - OFFENDER TARGET YOU FOR

CHARACTERISTICS/REGLIGIOUS BELIEFS (START 2010 Q1)V4526HA2 SC916 EVIDENCE OF HC - OFFENDER USE ABUSIVE LANGUAGE (START 2010Q1)V4526HA3 SC917 EVIDENCE OF HC - OFFENDER USE HATE SYMBOLS (START 2010 Q1)

V4526HA4 SC918 EVIDENCE OF HC - POLICE CONFIRM YOU TARGETED FOR

CHARACTERISTICS/RELIGIOUS BELIEFS (START 2010 Q1)V4526HA5 SC919 EVIDENCE OF HC - KNOW IF OFFENDERS PREVIOUSLY DID SIMILAR

CRIMES (START 2010 Q1)V4526HA6 SC920 EVIDENCE OF HC - DID INCIDENT OCCUR NEAR HOLIDAY, EVENT,

LOCATION ASSOCIATED WITH PARTICULAR GROUP (START 2010 Q1)V4526HA7 SC921 EVIDENCE OF HC - HAVE OTHER HC HAPPEND TO YOU OR IN

NEIGHBORHOOD (START 2010 Q1)V4526HA8 SC922 EVIDENCE OF HC - BELIEVE INCIDENT WAS HC BUT NOT ENOUGH

EVIDENCE (START 2010 Q1)V4526HA8 SC922 EVIDENCE OF HC - BELIEVE INCIDENT WAS HC BUT NOT ENOUGH

EVIDENCE (START 2010 Q1)

HEAD OF HOUSEHOLDPrior to 1979, one person in each household was designated as "head," usually the person reported by themembers of the household to be the head. Beginning in 1979 a non-head of household approach was institutedand interviewers were instead instructed to designate a "reference person" -- the household member who owns, isbuying, or is renting the housing unit. The reference person may be any one household member 18 years of ageor older -- usually the first person listed in the household membership. If all household members are under 18, theinterviewer will choose the person identified as owning or renting the living quarters.-43-See also: Reference Person, Relationship to Reference PersonHOUSEHOLDA group of occupants of a sample unit who meet the criteria for household membership.See also: Sample Unit, Household Member

HOUSEHOLD CRIMESee: Household Victimization; This classification of crime has been revised in the NCVS and is now labeled"property crimes".

HOUSEHOLD INCIDENTSee: Household Victimization; This classification has been revised in the NCVS and is now labeled "propertycrimes."

HOUSEHOLD INCOMEThe sum of income received by all household members (14 years of age or older) living in a sample housing unit.The income may include wages, salaries, net income from business, farm or rent, pension, dividends, interest,social security payments, alimony, public assistance, child support, and any other money received by householdmembers age 14 or older, and other money income (not identified as income are such things as room and board,insurance payments, lump sum inheritances, occasional gifts, money from selling property, withdrawals fromsavings accounts, or tax refunds) for the 12-month period immediately preceding the month of interview. It is notthe calendar year unless the month of the interview was January. Family income questions are asked routinelyduring the first, third, fifth, and seventh interviews only.

HOUSEHOLD MEMBERPersons in a sample unit are members of the household if:

1. their usual place of residence at the time of the interview is the sample unit, and
2. they have no usual place of residence elsewhere. Usual place of residence must be specific living quarters held

by the person, to which s/he is free to return at any time, and at which s/he usually lives and sleeps. Householdmembers not only include members of the family, but may also include: lodgers, servants, other employees wholive in the unit and consider it their usual place of residence. Usual residents also include persons who live in thesample unit but are temporarily absent. Armed Forces members are considered household members if they arestationed nearby and usually sleep in the sample unit. Students attending school away from home are notconsidered household members at their parents' home.See also: Household, Sample Unit

HOUSEHOLD NUMBERIdentifies the sequence of households that occupy a particular address. The household interviewed at the initialvisit receives a household number of "l". If members of that household change, but at least one householdmember remains at the address during subsequent interviews, the household number remains a "1". If a newfamily moves into a sample unit being interviewed for a second time, the household number will be "2" indicatingthat this is a different household from that previously interviewed.See also: Household Status

HOUSEHOLD RESPONDENTA household member chosen by the interviewer who appears to be the most knowledgeable about the household-44-composition and is able to answer the household screen questions dealing with crimes against the household.This person is generally 18 years of age or older; however, if no one in the household is at least 18 years of age,the interviewer will choose someone who appears to be knowledgeable about the entire household. Thehousehold respondent is not necessarily the reference person or the principal person.See also: Reference Person, Principal Person

HOUSEHOLD SCREEN QUESTIONSThese questions ask about any crimes against the household, such as actual or attempted break-ins, theft ofhousehold goods, and motor vehicle thefts. These screen questions are asked of the Household Respondent onlyand are used to determine whether a Crime Incident Report should be completed.See also: Basic Screen Questionnaire, Individual Screen Questions, Crime Incident Report

HOUSEHOLD STATUSConsists of the following categories:

1. Same household as last enumeration, if the same household was interviewed during the previous interview

period.

2. Replacement household since last enumeration if the former occupants have moved and the unit is now

occupied by a new household.

3. Previous non-interview or not in sample before -- for an interviewed unit that was previously non-interviewed for

any reason, for a unit in its first period of enumeration, or for an EXTRA unit being interviewed for the first time.See also: Extra Units, Enumeration

HOUSEHOLD THEFTSee: Theft

HOUSEHOLD VICTIMIZATIONA specific criminal act as it affects the household. This NCS class of crime has been replaced by a revisedclassification -- property victimization in the NCVS. The NCS defined household crimes as burglary, motor vehicletheft, and theft. Each criminal act against a household is considered to involve a single victim: the affectedhousehold (therefore, the terms household victimization and household incident are synonymous). If anyhousehold member is attacked or threatened during the crime, it is considered a personal crime and reports aretaken from each person who was personally victimized. Typically, household victimizations were reported by theHousehold Respondent. However, any interviewed member could report household crimes in response toindividual screen questions.See also: Personal Victimization, Household Screen Questions, Property Victimization NCVSNCVS NOTE: Beginning in January 1993, information on vandalism against a household was also collected.Beginning in 2009 the NCVS no longer collected information on vandalism against the household.

HOUSEHOLD WEIGHTIn the NCVS the household weight equals the person weight of the "Principal Person." The weight for the principalperson in husband-wife households equals the weight for the wife, excluding the within-household non-interviewadjustment. For non-husband-wife households, the principal person weight equals the weight for the referenceperson, excluding the within-household non-interview adjustment.See also: Principal Person, Person Weight, Reference Person, Weight, Incident Weight, Non-interviewAdjustment

HOUSING UNIT-45-A group of rooms or a single room, occupied as separate living quarters. Separate living quarters exist when theoccupants do not live and eat with any other persons in the structure, AND when there is either direct access fromthe outside or through a common hall, OR complete kitchen facilities for this unit only. Vacant rooms or groups ofrooms, which are intended for occupancy as separate living quarters, are also considered housing units. A unithas complete kitchen facilities when it has an installed sink with piped water AND a range or cooking stove AND amechanical refrigerator. All units in the National Sample are defined as either being a housing unit or an OTHERunit.See also: Group Quarters, Living Quarters, Extra Unit, Household, Occupied Housing Unit, Vacant Housing Unit,

Other UnitI

IDENTIFICATION CODEA sample designation and control number assigned to each sample unit.See also: Sample Designation, Control NumberINCIDENTA specific criminal act involving one or more victims and offenders. In the NCVS, victims may be persons 12years of age or older or households. An incident is technically distinct from a victimization; in criminal acts againstpersons, the number of victimizations is determined by the number of victims. Hence, one incident may compriseseveral victimizations.See also: Crimes, Victimization, Household Victimization, Personal Victimization, Personal Incident

INCIDENT REPORTSee: Crime Incident Report

INCIDENT WEIGHTFor personal crimes the incident weight is computed by dividing the person's weight by the total number ofpersons victimized during an incident as reported by the respondent. For example, if a respondent reports havingbeen robbed at the same time as two friends, three people are considered to have been victimized by the robberyincident. The incident weight in this case would be computed by dividing the person weight by three. Forhousehold crimes, the incident weight and the household weight are the same, since there can be only onehousehold that sustains a crime incident.See also: Weight, Household Weight, Person Weight, VictimizationINCOMESee: Family Income

INDIVIDUAL SCREEN QUESTIONSA set of questions asked of each household member 12 years of age and older. These questions are designed toelicit reports of both violent crimes (rape, sexual assault, robbery, simple and aggravated assault) and purse-snatching or pocket-picking committed against individual household members. A Crime Incident Report iscompleted for each crime recorded in the Individual Screen Questions.See also: Basic Screen Questionnaire, Household Screen Questions, Crime Incident Report

INDUSTRY CODENCS NOTE: Prior to March 1986, a three-digit numeric code was assigned to persons using the 1970 or 1980-46-Alphabetic Index of Industries and Occupations, based on the kind of business or industry at the location wherethe person works. For unemployed persons, the industry code was assigned on the basis of the last job heldwithin 5 years. Industry coding was discontinued in March 1986.See also: Occupation Code, Class or Worker, Major ActivityINJURYBodily hurt or damage sustained by a victim as a result of criminal assault. The types of injuries suffered are usedto distinguish between serious and minor assaults. Serious injuries include knife or gunshot wounds, brokenbones, loss of teeth, and loss of consciousness. A completed rape is classified as a serious injury. Minor injuriesinclude bruises, black eyes, cuts, scratches, swelling. Other injuries that can't be identified as serious or minor aredistinguished by the amount of hospitalization required. Injuries suffered from an attack during a crime incidentinclude any and all PHYSICAL (bodily) damage experienced by the victim such as broken bones, bruises, cuts,internal injuries, and so on. Emotional and psychological trauma are not included.

INTERVIEW PERIODSee: Enumeration Period

INTERVIEW PROCEDURESBefore the scheduled field interview, a letter informing each household about the NCVS and the interviewer'simpending visit is sent to each sample unit. The initial contact between the interviewer and the household is apersonal visit, at which time interviews are obtained for as many household members 12 years or older aspossible. For those eligible household members not interviewed during the initial visit to the household,interviewers make telephone callbacks to obtain interviews. During the initial interview, respondents are asked ifthey would be willing and able to conduct subsequent interviews by phone. If they answer yes, all but the fifth aredone by phone. In 1987 computer-assisted telephone interviewing (CATI) was used for some phone interviews.Since 2007 CATI has been discontinued as a mode. Prior to March 1986, three of the seven interviews (thesecond, fourth, and sixth) were done by phone whenever possible. Prior to January 1980, all interviews weredone face-to-face.See also: Computer-Assisted Telephone Interviewing (CATI), Interviewing Sequence, Noninterview Procedures

INTERVIEW, TYPE OFThe NCVS uses a combination of personal and telephone interviews, and uses proxy interviews wherenecessary.

1. Personal Interviews: A face-to-face interview between survey interviewer and respondent usually taking place in

the respondent's home. Each enumeration period, initial interviews in a household are conducted in person.Follow-up interviews to reach persons not available at the time of the initial visit may be made by telephone.

2. Telephone Interviews: A telephone interview between survey field representative and respondent. The

respondent usually conducts the interview from his/her home; the field representative may be calling from a fieldoffice or from a centralized telephone facility. Telephone interviewing is used to:a) conduct interviews with respondents willing and able to be interviewed over the telephone for all but the firstand fifth interview, andb) conduct interviews with household members not available for interview at the time of the initial personalinterview at the household

3. Self Respondent: A person who answers questions for him/herself. Information about each household member is

usually obtained by self-response; that is, each of these persons is interviewed for him/herself.

4. Proxy Respondent: A proxy respondent is a respondent who answers questions for another person.

NCVS NOTE: Prior to July, 1986, information about each household member aged 12 and 13 was obtained byasking the questions for these persons of the household respondent or some other knowledgeable householdmember. Currently, unless a knowledgeable household member objects, 12- and 13-year-olds are intervieweddirectly by the interviewer. If a particular household member 12 or older is physically or mentally unable to answerthe individual questions, or is temporarily absent and is not expected to return before the enumeration closeoutdate, the interviewer is instructed to accept information from another knowledgeable household member.-47-See also: Computer-Assisted Telephone Interviewing (CATI)

INTERVIEWING SEQUENCEThe general interview sequence for the NCVS is to:

1. Fill a Control Card for the Unit.
2. Interview the household respondent. This involves asking all appropriate personal characteristics and Basic

Screen Questions (for the individual AND the household).

3. For each incident against the household or the individual, an Incident Report is completed.
4. Interview all other eligible household members. This involves asking all appropriate personal characteristics and

Basic Screen Questions (from the Individual Screen section only)

5. For each incident reported by individual household members, an incident report is filled out. An entire interview is

completed for a household member before proceeding with the next person.See also: Interview Procedures, Control Card, Basic Screen Questionnaire, Crime Incident ReportL

LAND USEDesignated as urban or rural, based on Census classification criteria. Urban areas are places of 2,500 or moreinhabitants incorporated as cities, boroughs, villages, and towns (except towns in New England, New York, andWisconsin); the densely settled urban fringe, whether incorporated or unincorporated, or urbanized areas; andunincorporated places of 2,500 or more inhabitants. All remaining areas are classified as rural.See also: Rural, UrbanLARCENYSee: TheftLIFESTYLEThe Individual Screen Section begins with a series of questions on lifestyle and frequency of behaviors that haveshown some association with crime victimization. Such behaviors include shopping, spending the evening awayfrom home, and riding public transportation.(See: Individual Screen Section, Items 30, 31 and 32)

LINE NUMBERA two-digit number assigned during the interview to identify each individual in a household. Line numbers serve tolink Control Card data to the Personal Characteristics data, and both of these to the Incident Reports data.LISTINGA procedure used to note the exact location of every place where people live or might live within a given addressor segment. Interviewers record street names, house numbers, apartment numbers, and unit locations. This list isthen used to select sample units randomly. In general, listing is done by observation guided by detailed andextensive procedures, and varies somewhat depending on the type of segment.See also: Updating, Segment, Extra Units

LIVING QUARTERSThis is a general term for structures occupied or intended for human occupancy. Living quarters are classified aseither housing units or OTHER units. Six types of housing units are distinguished for NCVS:-48-

1. House, apartment, flat: includes such housing units as an apartment over a garage or behind a store, janitor's

quarters in an office building, and housing units in converted barns or sheds, as well as ordinary houses andapartments.

2. Housing unit in a non-transient hotel, motel, etc. All separate living quarters in non-transient hotels, motels, etc.,

are housing units by definition.

3. Housing unit permanent in transient hotel, motel, etc. Occupied or intended for occupancy by permanent guests

or resident employees.

4. Housing units in rooming house.
5. Mobile home or trailer: includes those with permanent foundations.
6. Housing unit not specified above. Five types of OTHER units are distinguished for NCVS:

a) Quarters, not housing unit, in rooming or boarding house.b) Unit not permanent in transient hotel, motel, etc. Quarters not occupied or intended for occupancy bypermanent guests or resident employees.c) Vacant tent site or trailer site.d) Student quarters in a college dormitory.e) Not specified above (e.g.: quarters for nurses, quarters in bunkhouses.See also: Housing Unit, Other Units, Special Places, Transient/Nontransient, Place of Residence, TenureM

MARITAL STATUSThe NCVS defines five categories of marital status:

1. Married: includes common-law marriage
2. Widowed
3. Divorced
4. Separated: those married persons who have a legal separation or who have parted because of marital discord,

but have not yet obtained a divorce. Those who have parted temporarily for reasons other than marital discord(such as employment, Armed Forces, etc.) are recorded as married.

5. Never Married: includes all children under 14, and persons who have never been married or whose only

marriages have been annulled.

MEDICAL CAREVictims who were injured during an incident are asked whether they received any medical care for their injuries.NOTE: Prior to 1979, medical care included only treatment rendered by trained professionals. In 1979 this waschanged to include paraprofessionals, non-professionals and self-treatment.

MEDICAL EXPENSES, TOTAL AMOUNT OFThis is the total amount of medical expenses which results directly from the victim's injuries in the incident.Included are such items as doctor and hospital bills, ambulance services, and any special devices or aids thevictim was forced to obtain as a result of the injury. If the victim is still undergoing treatment for injuries related tothe incident, the total amount of medical expenses is an estimate of the total projected medical expense.

METROPOLITAN STATISTICAL AREAThe Office of Management and Budget (OMB) defines this as a population nucleus of 50,000 or more, generallyconsisting of a city and its immediate suburbs, along with adjacent communities having a high degree ofeconomic and social integration with the nucleus. MSAs are designated by counties, the smallest geographicunits for which a wide range of statistical data can be obtained. However, in New England, MSAs are designatedby cities and towns since these subcounty units are of great local significance and considerable data is availablefor them. Currently, an area is defined as an MSA if it meets one of two standards:

1. a city has a population of at least 50,000
2. the Census Bureau defines an urbanized area of at least 50,000 people with a total metropolitan population of at

least 100,000 (or 75,000 in New England).-49-The Census Bureau's definition of urbanized areas, data on commuting to work, and the strength of the economicand social ties between the surrounding counties and the central city determine which counties not containing amain city are included in an MSA. For New England, MSAs are determined by a core area and related cities andtowns, not counties. A metropolitan statistical area may contain more than one city of 50,000 and may cross Statelines.

MILITARY STATUSDetermined for males and females 18-65 years of age. "In Armed Forces" includes those serving on active duty atthe time of the interview in U.S. Army, Navy, Air Force, Marine Corps, Coast Guard, or the reserve branch of anyof the above only if currently on active duty status for several months. Also includes commissioned officers of theU.S. Public Health Service who are attached to any branch of the Armed Forces, and National Guardsmen onactive duty. Cadets in U.S. Military academies also are considered to be with the Armed Forces.

MINOR ASSAULTSee: Robbery, Sexual Assault

MOTOR VEHICLE THEFTStealing or unauthorized taking of a motor vehicle, including attempted thefts. A motor vehicle is defined as anautomobile, truck, motorcycle, or any other motorized vehicle legally allowed on public roads or highways.

MOTOR VEHICLES, TOTAL NUMBER OFAll motor vehicles owned during the reference period are counted, including those sold, given away, junked,stolen, or abandoned. Motor vehicles used SOLELY by a recognizable business are not included, nor arevehicles loaned to the respondent for private use by a business, unless the respondent owns the business. Motorvehicles used by a student, owned by his/her parents, and kept away from home are not included.

MULTI-STAGE SAMPLINGA process used to select sample units. The NCVS sample was selected using two stages. In the first stage,Primary Sampling Units (PSUs) were selected. In the second stage, a sample of addresses was selected withinthe sample PSUs. See "Sample Design" for a full description.See also: Primary Sampling Unit (PSU)N

NCVS CRIME CLASSIFICATION SCHEMECrimes reported in the NCVS are classified using three main factors:

1. Details respondents provide in the Incident Report
2. Type of Crime Categories (see "CRIMES")
3. Seriousness Hierarchy (see "CRIMES")

Details of the incident are reviewed in order to place the incident in a Type of Crime Category. If the incidentmatches a category, that Type of Crime code is assigned. If the incident fits more than one Type of Crimecategory, the Seriousness Hierarchy is used to arrive at a single Type of Crime code for each incident. Anexample of this is a break-in that could be classified as both a burglary and an assault (e.g.: if a residentencountered the burglar and was physically assaulted). Assault is higher than Burglary on the SeriousnessHierarchy, thus the incident would be coded as an Assault. However, all details collected from the IncidentReports are preserved and available for users to create their own classification scheme.NON-INTERVIEW-50-For various reasons, interviewers are unable to obtain interviews for all sample units. Such units are classified asnon-interviews and are handled as follows:

1. TYPE A: when an interview for an eligible household is not obtained for some reason such as refusal. These

consist of households occupied by persons eligible for interview but from whom no interviews are obtainedbecause:a. no one is at home in spite of repeated visitsb. the entire household is temporarily away during all of the interview periodc. the household refuses to give any informationd. the unit cannot be reached due to impassable roadse. interview is not conducted due to a serious illness or death in the familyf. the interviewer is unable to locate the sample unit.Every effort is made to avoid this type of non-interview in order to keep the unit and maintain a representativesample. A Non-Interview adjustment is applied to interviewed household and person weights so that Type A unitsare represented in the sample.

2. TYPE B: when an interview is not obtained because the unit is ineligible for interview at the present time (such as

a vacancy) but could become eligible in the future. These consists of units which are not eligible for interviewbecause:a. unit is a vacant regular housing unitb. unit is vacant and used for storage of household furniturec. unit is occupied by persons usually residing elsewhered. unit unfit for habitation or to be demolishede. unit under construction and not ready for occupancyf. unit temporarily converted to business or storageg. address identifies an unoccupied tent or trailer siteh. permit granted but construction not startedi. household consists entirely of persons who are ineligible for the survey. Since these units might becomeeligible in the future, they are revisited in subsequent periods that the unit is in the sample. If they becomeeligible, the household is interviewed.The sample size is sufficiently large to allow for Type B Non-Interviews.

3. TYPE C: when the unit is physically ineligible. These consist of units not eligible for sample because:

a. no address was listed on the sample line of the listing sheetb. unit demolished by time of enumerationc. house or trailer has been movedd. unit converted to permanent business or storagee. unit has been merged with another unitf. unit condemned and unoccupiedg. outside the segment boundaryh. built after April 1, 1980, 1990, 2000 (based on decennial census)i. permit was abandoned.The sample size is sufficiently large to allow for Type C Non-Interviews.

4. TYPE Z: when a particular member is not interviewed in an otherwise interviewed household, i.e.: a within-

household non-interview. These occur when the interviewer is unable to obtain an interview for a particularhousehold member in an otherwise interviewed household, e.g., the person is never available, refused, or isphysically or mentally unable to answer and no proxy is available. For this type of non-interview person, only afew personal characteristic items are filled in on the Basic Screen Questionnaire. A Non-Interview adjustment isapplied to interviewed person weights to account for within-household non-interviews.See also: Basic Screen Questionnaire

NON-INTERVIEW/NON-RESPONSE ADJUSTMENTSAdjustments made to the weights of interviewed households and persons in order to compensate for those-51-households and persons not interviewed. A household non-interview adjustment accounts for occupied sampleunits qualified to participate in the survey, but from which an interview was not obtained (Type A Non-Interview).A within-household non-interview adjustment accounts for eligible persons who were not interviewed in aninterviewed household (Non Type Z Non-Interview.) See section - Weight for a full description.See also: Household Weight, Person Weight, Final WeightO

OCCUPIED HOUSING UNITA housing unit is considered occupied if it is the usual place of residence of the person or persons living in it atthe time of the interview.See also: Place of Residence, Housing UnitOFFENDERThe person(s) who committed or was/were believed by the respondent to have committed the crime.

OPERATION OF BUSINESSFor the purposes of the NCVS, a distinction is made between two types of businesses: recognizable andunrecognizable. A recognizable business is one that is observable and identifiable from the outside by a sign orsome other indication. Crimes involving any recognizable business, such as property stolen from the business,are not included in the NCVS. Only the loss of personal property of the respondent or of another householdmember, or personal threat or injury to the respondent or other household member(s) is included if a crimeinvolves a recognizable business. An unrecognizable business has no outside identification, such as a sign in theyard or window, which indicates that it is a business. An example of an unrecognizable business is a mail orderbusiness run from the home, if there is no sign outside the house advertising the business. Crimes involvingunrecognizable businesses, including property stolen from the business, are included in the NCVS estimates.

OTHER UNITSRefers to living quarters which do not qualify as housing units. OTHER units are located in special places such asinstitutions, dormitories, and boarding houses where the residents have their own rooms, groups of rooms, orbeds, and also have some common facilities such as a dining room, lobby, or living room or recreational areas.Some of the rooms, groups of rooms, etc., in these places are always classified as OTHER units, even thoughthey have direct access or complete kitchen facilities. Other rooms or groups of rooms are classified as OTHERunits if they do NOT meet the housing unit definition. All units in the NCVS sample are defined as either being ahousing unit or an OTHER unit.See also: Housing Unit, Living Quarters, Special Places, Group Quarters, Extra Units

OUT OF UNIVERSEOut of Universe (or INAP) is used in the codebook documentation to designate those items on the questionnairethat were not appropriate for certain respondents (based on information gathered throughout the interview) andtherefore should not have been asked. For example, hospital tenure questions are not asked of victims who werenot injured.See also: ResiduePPANELA systematically selected group of sample units comprising one-sixth of the entire NCVS sample. Each panel isinterviewed twice per year (at six-month intervals, e.g.: January and July) for three years total during the panel's-52-time in sample.See also: Sample Rotation

PANEL NUMBERIn the National Sample, an identification number (l-6) assigned to each case in the sample identifying the monthsin which the unit is to be interviewed (e.g., Panel 1 housing units are interviewed in January and July ). The panelnumber is the first digit of the segment number.See also: Segment Number

PERSON WEIGHTA variable associated with each person in the sample to provide an estimate of the population represented byeach person. The person weight is calculated as the product of six factors, whose values depend on thecharacteristics of the household or the person:

1. Base Weight
2. Weighting Control Factor
3. Household Non-interview Adjustment
4. Within-Household Non-interview Adjustment
5. First-Stage Factor
6. Second-Stage Factor

For a complete description of these factors, see section - Weight Details.See also: Weight, Household Weight, Incident Weight, Non-interview Adjustments

PERSONAL CRIMESSee: Personal Incident, Personal Victimization.

PERSONAL INCIDENTA specific criminal act involving one or more persons as victims of one or more offenders. The NCVS definescrimes against persons as rape, sexual assault, robbery, simple or aggravated assault, and purse-snatching\pocket-picking. The NCVS makes a general distinction between an Incident and a Victimization. Thenumber of victimizations reflects how many criminal acts were experienced by NCVS survey respondents; thenumber of incidents reflects the number of criminal acts committed against NCVS survey respondents and otherspresent during such incidents as reported by survey respondents. If there were more than one victim, the incidentweight is adjusted to compensate for the possibility that the incident could be reported several times by multiplevictims and thus be over-counted. For example, if a survey respondent and two other people (non-respondents)were robbed during the same incident, the NCVS counts this as one-third of a personal incident but onevictimization. If all victims were respondents, this criminal act would be counted as one incident and threevictimizations. Therefore, estimates of the number of personal incidents are usually lower than estimates ofpersonal victimizations.See also: Personal Victimization, Victimization, Incident Weight

PERSONAL INTERVIEWSSee: Interview, Type of

PERSONAL VICTIMIZATIONA specific criminal act as it affects one victimized person. The NCVS defines crimes against persons as rape,sexual assault, robbery, simple or aggravated assault, and purse-snatching/pocket-picking. Estimates of thenumber of personal victimizations are determined by the number of victims of such acts as reported by the surveyrespondent. Ordinarily, the number of victimizations is somewhat higher than the number of incidents because-53-more than one person may be victimized during certain incidents, and because details of personal crimesoccurring during the course of a commercial crime, as related to the victimized person, are reflected in theseestimates.See also: Personal Incident, Victimization, IncidentPLACEThe term place refers to a concentration of population, regardless of the existence of legally prescribed units,powers, or functions. However, most of the places identified in the census are incorporated as cities, towns,villages, or boroughs. In addition, the larger unincorporated places are identified as places. Places are classifiedin the National Sample files by a two-digit place size code and a one-digit place description code.See also: Place Size Code, Rural, Urban

PLACE DESCRIPTION CODEA code used in NCVS files to identify the type of place in which a sample unit is located.See also: Central City, Standard Metropolitan Statistical Area

PLACE OF OCCURRENCEThe following detailed place of occurrence categories are used to describe more specifically where the incidentoccurred:

1. In respondent's home or lodging

a. In own dwelling, own attached garage, or enclosed porchb. In detached buildings on own property, such as detached garage, storage shed, etc.c. In vacation home/second homed. In hotel or motel room respondent was staying in

2. Near own home

a. Own yard, sidewalk, driveway, carport, unenclosed porch (does not include apartment yards)b. Apartment hall, storage area, laundry room (does not include apartment parking lot or garage)c. On street immediately adjacent to own home

3. At, in, or near a friend's/relative's/neighbor's home

a. At or in home or other building on their propertyb. Yard, sidewalk, driveway, carport (does not include apartment yards)c. Apartment hall, storage area, laundry room (does not include apartment parking lot/garage)d. On street immediately adjacent to their home

4. Commercial places

a. Inside restaurant, bar, nightclubb. Inside other commercial building, such as, store, bank, gas stationc. Inside office, factory, or warehouse

5. Parking Lots/Garages

a. Commercial parking lot/garageb. Noncommercial parking lot/garagec. Apartment/townhouse parking lot/garage

6. School

a. Inside school buildingb. On school property (school parking area, play area, school bus, etc.)

7. Open Areas, on street or public transportation

a. In apartment yard, park, field, playground (other than school)b. On street (other than immediately adjacent to own/friend/relative/neighbor's home)c. On public transportation or in station (bus, train, plane, airport, depot, etc.)

8. Other

Crimes that occur outside the United States are identified on question 4a of the incident report form; however,BJS does not tally these crimes in its annual report.NCVS NOTE: Prior to January 1993, enclosed and unenclosed porch were not explicitly asked.-54-

PLACE OF RESIDENCESpecific living quarters held by respondent, to which s/he is free to return at any time, AND at which s/he usuallylives and sleeps. Not necessarily the same as legal residence or voting residence. Also described as "UsualPlace of Residence".See also: Living Quarters, Occupied Housing Unit, Vacant Housing Unit

PLACE SIZE CODEA code used to identify the population size range for the place in which the sample unit is located.NOTE: Places under 2,500 are considered rural by definition.See also: PlacePOCKET-PICKINGTheft or attempted theft of property or cash directly from the victim by stealth, without force or threat of force.

POLICE INFORMEDA question on the Crime Incident Report asks: "Were the police informed of this incident in any way?" The "police"include all regular police and sheriff's departments at the city, county, state, or federal level of government. Alsoincluded are specialized police forces authorized to make arrests in a special area or jurisdiction, e.g., campuspolice, park police, transit police, harbor police, and airport police. Excluded are security guards, fish and gamewardens, and such officers without authority to make police arrests.

PRESENCE OF WEAPONSee: Weapon

PRIMARY SAMPLING UNIT (PSU)Survey estimates are based on data obtained from a stratified, multi-stage cluster sample. The primary samplingunits (PSUs) composing the first stage of the sampling were counties, groups of counties, or large metropolitanareas. Large PSUs were included in the sample automatically and are considered to be self-representing (SR).The remaining PSUs, called non-self-representing (NSR), were combined into strata by grouping PSUs withsimilar demographic characteristics, as determined by the census. One PSU was selected from each stratum bymaking the probability of selection proportionate to the population of the PSU. Since July 1986, the PSU isscrambled with four other elements into a control number, to preserve confidentiality.See: Control Number

PRINCIPAL PERSONIn husband/wife living arrangements, this person is the wife. In non-husband/wife living arrangements, this personis the reference person. The principal person's demographic characteristics are used in forming the householdweight.See also: Head of Household, Household Respondent

PROPERTY CRIMEProperty crimes include burglary, motor vehicle theft, and theft. The category includes both attempted andcompleted crimes. Property crime corresponds roughly to the NCS classification of Household Crime, except thatcrimes formerly classed as personal larceny without contact are now classed as thefts and thus property crimes.The household weight is the correct weight to use with any property crime.-55-

PROPERTY VICTIMIZATIONA specific criminal act that may be classified as a property crime. Each criminal act against a household isconsidered to involve a single victim: the affected household. If any household member is attacked or threatenedduring the crime, it is considered a personal crime, and reports are taken from each person who was personallyvictimized. Typically, property victimizations are reported by the Household Respondent. However, anyinterviewed member may report property crimes in response to individual screen questions.See also: Personal Victimization, Household Screen Question

PSU NUMBERS"PSU numbers" are three-digit codes which identify each PSU, and are part of the Control Number which uniquelyidentifies each sample unit. For purposes of confidentiality, these numbers have been scrambled on the publicuse files.

PROXY RESPONDENTSee: Interview, Type of.PURSE-SNATCHINGTheft or attempted theft of properties or cash directly from the victim by stealth without force or threat of force.Q

QUESTIONNAIRE FORMSThere are three basic forms used to collect the data required for the NCVS:

1. Control Card
2. Basic Screen Questionnaire (Personal Characteristics, Household and Individual Screen Sections)
3. Crime Incident Report

See section Type of Research Instrument for more detail on these forms and procedures.See also: Basic Screen Questionnaire, Control Card, Crime Incident Report, Interview ProceduresRRACERespondent's self-reported race. The categories are: White, Black or African American, American Indian orAlaska Native, Asian, Native Hawaiian or Other Pacific Islander, and Other.See also: Ethnicity

RACE OF OFFENDERThe race of offender as perceived and reported by the respondent. Categories are White, Black or AfricanAmerican, American Indian or Alaska Native, Asian, Native Hawaiian or Other Pacific Islander, and Don't know.When recording the racial designation(s) of multiple offenders, categories are: Mostly White, Mostly Black orAfrican American, Mostly American Indian or Alaska Native, Mostly Asian, Mostly Native Hawaiian or OtherPacific Islander, Equal number of each race, and Don't Know.NCVS NOTE: Categories of race of offender (single and multiple) were modified in 1986.Categories of race of offender (single and multiple) were expanded in 2012 to add American Indian or AlaskaNative, Asian, and Native Hawaiian or Other Pacific Islander.-56-RAPEOne of the six major types of crime covered by the NCVS, rape is defined as:"Forced sexual intercourse and includes both psychological coercion as well as physical force. Forced sexualintercourse means vaginal, anal, or oral penetration by the offender(s). This category also includes incidentswhere the penetration is from a foreign object such as a bottle."Respondents are asked a series of questions in the Basic Screen Section about attack, rape, attempted rape,sexual attack, and forced or coerced unwanted sex. These questions are asked directly and are accompanied bycues about the offender (casual acquaintance, co-worker, neighbor, friend, relative, etc.). If the respondent repliesin the affirmative to any of these questions or cues, an Incident Report is filled out to gather details. In gatheringthese details, interviewers are instructed to ask "Do you mean forced or coerced sexual intercourse?" todetermine whether the incident should be recorded as a rape. If the respondent requests clarification of the terms"rape" or "sexual intercourse", interviewers may read the definition from the manual above.NCVS NOTE: Prior to January 1993, the Basic Screen Questionnaire did not ask direct questions about rape andrelated sexual attacks. Rapes may have been reported in response to any screen question, especially thosewhich asked about attacks, threats, and attempted attacks. Since the questions did not ask directly about rape, itwas up to the respondent to acknowledge any such incidents as warranting "yes" responses and to specify that arape occurred in response to questions on the incident form. Rape sub-categories included completed andattempted rape only. Beginning in January 1993, the categories of rape and sexual attack were expanded. Seethe NCVS Redesign document in the NCVS Resource Guide for a complete discussion of these changes.See also: Crimes, AssaultRECODEA single code, mechanically derived by combining codes for several related items, used for convenience andefficiency in tabulating data.

RECOGNIZABLE BUSINESSSee: Operation of Business

REFERENCE PERIODThe period of time for which respondents are asked to recall victimizations experienced. The NCVS uses areference period of six months. For crime tabulation purposes, the reference period is exactly six months endingon the last day of the month prior to the interview; for data collection purposes, the length of the reference periodvaries slightly, depending on the exact day of interview. For example, for interviews taking place in July, thereference period for tabulating victimizations is January 1 through June 30. For data collection purposes,respondents are asked to report on incidents beginning January 1 and ending on the day prior to interview. Incases where victimizations are reported as having occurred during the month in which the interview takes place,an Incident Report is filled out but the data on the victimization(s) are held and tabulated with crimes reported inthe following reference period (in this case, July 1 through January 31).See also: Collection Quarter, Bounding Procedure

REFERENCE PERSONThe person identified by the household respondent as owning, buying, or renting the living quarters. This personwill generally be at least 18 years of age or older; however, if no one in the household is 18 years of age, theinterviewer will choose the person who is identified as owning or renting the living quarters. The reference personreplaces the former concept of "head of household" and is not necessarily the same as the household respondentor principal person.See also: Household Respondent, Principal Person, Head of Household, Relationship to Reference Person-57-REINTERVIEWA reinterview is a second interview of a previously interviewed sample unit, with the purpose of evaluating someaspect of the original interview. It re-asks a small subsample of the original respondents all or a portion of thequestions from the original interview and refers to the same time point as the original interview. For NCVS, amember of the supervisory staff or a supervisory field representative conducts the reinterview and compares thereinterview answers to the original answers. The reinterviewer uses this comparison primarily to detect and deterfalsification. Falsification occurs when field representatives intentionally falsify interviews or misclassifynoninterviews. The reinterviewer can also use reinterview data to do the following:

1. Detect system-wide problems. For example, inadequate training, deficient procedures, and poor

questionnaire design

2. Monitor interviewer performance, give feedback, and decide when an interviewer needs additional

training.

RELATIONSHIP TO REFERENCE PERSONOnce interviewers have established the reference person in a household, other household members are identifiedby their precise relationship to the reference person, e.g., wife, husband, son, daughter, step-son, step-daughter,father, mother, grandson, daughter-in-law, aunt, nephew, lodger, servant, hired hand, visitor, and so on.See also: Reference PersonRESIDENCESee: Place of ResidenceRESIDUEUsed in the codebook documentation to designate those cases where the respondent could not or did not providean answer, or for some reason the response given did not fall into the predetermined categories. Attempts weremade during the edits to convert such cases into valid codes based on the presence or absence of certain keyelements.See also: Out of UniverseRESPONDENTA person selected to be part of the NCVS sample who participates in the survey.See also: Interview, Type of

RESPONSE RATEThe rate at which selected sample units participated in the survey. The response rate is generally calculated bydividing the number of sample units participating in the survey by the number of sample units selected.ROBBERYOne of the six major crime categories, robbery is defined as completed or attempted theft, directly from a person,of property or cash by force or threat of force, with or without a weapon. There are two main subcategories ofrobbery: WITH INJURY and WITHOUT INJURY. ROBBERY WITH INJURY is theft or attempted theft from aperson, accompanied by an attack, either with or without a weapon, resulting in injury. ROBBERY WITH INJURYmay be further classified as occurring with serious or minor assault. As used in the NCVS classification hierarchy,SERIOUS ASSAULT is a form of assaultive violence in which the attack is either carried out with a weapon, orresults in serious injury, such as broken bones, internal injuries, loss of consciousness, or an undetermined injuryrequiring 2 or more days of hospitalization. This definition is equivalent to that for aggravated assault, except thatan aggravated assault is classified when other forms of victimization, such as theft of property or sexual attack,are not present during the same incident. An injury is classified as resulting from a MINOR ASSAULT if it is theconsequence of an attack carried out without a weapon and results in minor injury, such as bruises, black eye,-58-cuts, and scratches or an undetermined injury requiring less than 2 days of hospitalization. This definition isequivalent to that for completed simple assault, except that a completed simple assault is classified when other,more serious forms of victimization are not present during the same incident.See also: Theft, AssaultROTATIONSee: Sample Rotation, Rotation Group Number

ROTATION DESIGNSee: Sample Rotation, Rotation Group Number

ROTATION GROUP NUMBERA number used to place the sample unit in the rotation scheme. The rotation group number is denoted by thehundreds digit of the four-digit segment number assigned to each sample unit. Each rotation group is asystematic one-sixth sample of all current NCVS housing units as well as a one-sixth sample of the sampleinterviewed each month.See also: Segment Number, Sample RotationRURALAll remaining areas not falling into one of the following categories:

1. Places of 2,500 or more inhabitants incorporated into cities, boroughs, villages, and towns (except towns

in New England, New York, Wisconsin)

2. Densely settled urban areas
3. Unincorporated places of 2,500 or more inhabitants.

See also: Urban, Place, Land UseSSMSASee: Metropolitan Statistical AreaSAMPLEA selection of units of a population from which estimates about the general population are made. If sample unitsare chosen according to laws of probability and randomization, inferences may be made from the sample to thegeneral population within specified margins of error. For the NCVS a stratified cluster sample was used wherebyeach element of the population has a known chance of being included in the sample, allowing such inference tobe made.See also: Estimate

SAMPLE DESIGNATIONA portion of the identification codes assigned to each sample unit in the NCVS, consisting of the letter "J" and twodigits. The letter distinguishes NCVS surveys from other Census surveys, and the digits identify the sequenceand source of selection of sample addresses. For example, the first sample selected was designated as J01.See also: Identification Code, Control Number

SAMPLE ROTATION-59-The following rotation scheme is used for the NCVS: each sample of approximately 50,000 households is dividedinto six rotation groups. When the rotation plan became fully operative in January 1977, households in a particularrotation group and sample were interviewed once every six months for three years, the first interview being onlyfor the purpose of bounding. Each rotation group is further divided into six panels or months. Thus l/6th of eachrotation group is interviewed each month during a six-month period. One rotation group enters the sample everysix months, while the corresponding rotation group from a previous sample is phased out. This enablespreservation of the overall sample size and survey procedures while averting undue imposition to, and biases in,reporting by respondents through repeated interviews.See also: Bounding Schedule, Panel, Rotation Group Number

SAMPLE UNITA unit selected for inclusion in a sample or a census.See also: Sample, Housing Unit, Other Units, Special Places

SCREEN QUESTION NUMBEROn each Crime Incident Report, the number of the screen question that elicited the report of the incident isrecorded.See also: Individual Screen Question, Household Screen Questions

SCREEN QUESTIONNAIRESee: Basic Screen QuestionnaireSEGMENTA string of clusters of sample units in the same PSU selected to be part of the NCVS sample. Sample units arelocated in one type of segment, depending on the type of Enumeration District in which the segment is located:

1. Address Segments consist of completely recorded specific addresses selected from the Decennial

Census list, in address type Enumeration Districts.

2. Special Place Segments consist of special places (e.g.: dormitories, rooming houses) in address type

Enumeration districts which can be identified from Census listings.

3. Area Segments consist of relatively small designated areas of land with well-defined boundaries, such

as streets, roads, rivers, or railroad tracks in area type Enumeration Districts. The interviewer lists all ofthe living quarters in a designated segment, and a sample of these is selected for interview.

4. New Construction Segments represent units built since the Census.

See also: Enumeration District, Extra Units, Listing, Segment Number, Serial Number, Updating

SEGMENT NUMBERPart of the Control Number assigned to each sample unit. In the NCVS Sample it consists of four digits: thethousands digit identifies the panel; the hundreds digit identifies the rotation group; tens and units digits together(the serial number) identify clusters which are numbered sequentially starting with "01" in each panel and rotationcombination. Since July 1986, the segment number has been scrambled into the control number with othercontrol number elements to preserve confidentiality.See also: Segment, Control Number, Panel Number, Rotation Group Number, Enumeration District, SerialNumberSELF-EMPLOYEDSee: Class of Worker-60-SELF-RESPONDENTSee: Interview, Type ofSELF-WEIGHTINGThe result of a sampling procedure in which all units have an equal probability of being selected in the sample.Because each unit has an equal probability of selection, the inverse of that probability of selection is the weightitself, with no need for weighting adjustment factors due to the design.See also: Weight, Sample

SERIAL NUMBERA unique two-digit number assigned in sequence within each segment to each sample household in the NCVS. Itis used for administrative control, and is part of the Control Number. Since July 1986, the serial number has beenscrambled into the control number to preserve confidentiality.See also: Segment, Control Number

SERIES INCIDENT REPORTSix or more incidents similar in nature for which the respondent is unable to recall dates and other details of theindividual incidents well enough to report them separately. If a reported series of incidents meets this descriptionthe following information is recorded:

1. The incident is marked as a series crime.
2. An estimate of the number of times the incident occurred is asked of the respondent.
3. Details of the most recent incident in the series are gathered in the Crime Incident Report.
4. Information on the nature of the series is gathered (e.g.: how many times the incident occurred, whether

all incidents occurred in the same place, the victim/offender relationship, how similar the incidents wereto each other, and whether the problem persists).NCVS NOTE: Prior to January 1993, series crimes were defined as three or more incidents similar in nature forwhich the respondent was unable to recall dates and other details separately for each incident. Also, detailsdescribed in point #3 were added. See the NCVS Redesign document in the NCVS Resource Guide for adescription of this redefinition.

SERIOUS ASSAULTSee: Robbery, Sexual Assault

SERIOUSNESS HIERARCHYSee: Crimes, NCVS Crime Classification Scheme

SEXUAL ASSAULTA wide range of victimizations, separate from rape or attempted rape. These crimes include attacks or attemptedattacks generally involving (unwanted) sexual contact between victim and offender. Sexual assaults may or maynot involve force, such as grabbing or fondling, for example and may also include verbal threats. Sexual assaultsresulting in injury are further classified by the seriousness of the attack. A sexual assault with SERIOUSASSAULT is a form of assaultive violence in which the attack is either carried out with a weapon, or results inserious injury, such as broken bones, internal injuries, loss of consciousness, or an undetermined injury requiring2 or more days of hospitalization. This definition is equivalent to that for aggravated assault, except that anaggravated assault is classified when other forms of victimization, such as theft of property or sexual attack, arenot present during the same incident. A sexual attack with MINOR ASSAULT is the classification of an attackcarried out without a weapon and which results in minor injury, such as bruises, black eye, cuts, and scratches oran undetermined injury requiring less than 2 days of hospitalization. This definition is equivalent to that forcompleted simple assault, except that a completed simple assault is classified when other, more serious forms of-61-victimization are not present during the same incident.See also: Rape, Assault

SIMPLE ASSAULTSee: Assault

SPECIAL PLACE SEGMENTSee: Segment

SPECIAL PLACEPlaces in which the occupants have special living arrangements.Special places include: correctional institutions; homes for the aged, infirm, or needy; mental institutions; nursing,convalescent, or rest homes; specialized hospitals; Armed Forces installations; boarding and rooming houses;convents and monasteries; dormitories and other residences at schools; missions; halfway houses; hotels andmotels; nurse's homes; private residential clubs and tourist camps; recreational, educational and religious camps;tent and trailer camps; migratory workers' camps; vacation cabins; YMCAs and YWCAs; and so on. Specialplaces may include housing units and OTHER units. For example, a dormitory consisting of rooms for studentsand an apartment for a resident manager would include both OTHER units and a housing unit. Living quarterswithin special places are designated as either Staff Units, Non-staff Units, or Unclassified Units, as determined byoccupancy or intended occupancy. Non-staff Units are located in certain special places and are defined as thepersons for whom the special places are being operated, OR the units in which these persons live, such as roomsor suites or rooms. Non-staff units are never interviewed for the NCVS. Staff Units are located in the same typesof places which contain Non-staff Units and are defined as the houses, apartments, rooms, or suites of roomsoccupied or intended for occupancy by resident employees and their families, if any. Unclassified Units arelocated in certain special places and are the houses, apartments, rooms, etc. occupied or intended to beoccupied by all persons. Staff Units and Unclassified Units are interviewed.See also: Living Quarters, Group Quarters, Other Units, Extra Units

STANDARD ERRORAn expression of sampling error which states the amount of deviation of an estimate because a sample wasdrawn instead of a census taken. Standard error is a measure of how precise the sample estimate is compared tothe true value that would have been obtained if the entire population was measured. A measure of standard erroris calculated in the following way: under any given sample design, there are many possible samples of the samesize that could be drawn. In theory, each of these possible samples could actually be drawn, estimates could becalculated for each, and the average of these estimates could be computed. The difference between the estimatederived from the sample that actually was drawn and the average of the estimate calculated if all possiblesamples were drawn is called sampling deviation. The standard error of a sample estimate is a measure of thevariation among these estimates from all possible samples, and thus is a measure of the precision with which anestimate from a particular sample approximates the average result of all possible samples. An estimate of thestandard error of a sample estimate can be computed from the data obtained in the sample.See also: Variance, Estimate

STANDARD METROPOLITAN STATISTICAL AREA (SMSA)See: Metropolitan Statistical AreaSTRATAGroups of Non-self-representing (NSR) PSUs which have specified characteristics in common. The process ofmaking these groupings is called stratification. NSR PSUs were stratified by combining PSUs with similarpopulation characteristics which include:-62-

1. Crime rate (based on Uniform Crime Report data from the Federal Bureau of Investigation)
2. Percent urban
3. Average value of owned housing units
4. Average rent of rented housing units

T

TARGET POPULATIONThe population that sample estimates are designed to describe. The NCVS estimates crime levels andvictimization rates for the population of households and of individuals 12 years of age and older living inhouseholds and group quarters in the United States and the District of Columbia. Excluded are persons who arecrews of vessels, in institutions (e.g., prisons and nursing homes), or members of the armed forces living inmilitary barracks.

TELEPHONE INTERVIEWSSee: Interviews, Type ofTENUREManner in which living quarters are held.

1. Owned or being bought: if owner or co-owner is a household member of the unit, even if the unit is

mortgaged or not fully paid for. Owner need not be head of household. A cooperative apartment orcondominium unit is owned or being bought only if the owner lives in it.

2. Rented for cash: if any money rent is paid or contracted for by any persons.
3. Without cash rent: Includes units not owned, being bought, or rented for cash; usually provided in

exchange for services or as a gift.See also: Living QuartersTHEFTCompleted or attempted theft of property or cash without personal contact. Incidents involving theft of propertyfrom within the same household would classify as theft if the offender has a legal right to be in the house (such asa maid, delivery person, or guest). If the offender has no legal right to be in the house, the incident would classifyas a burglary. Theft is classified as a property crime, and the household weight should be used for estimates. Thisdefinition applies only from January 1993. Before that date, theft was classified as household (household larceny)or personal (personal larceny without contact, personal larceny with contact, i.e., pocket-picking or purse-snatching). The primary distinguishing attribute of personal and household larceny was location, with thefts awayfrom the home classed as personal and thefts at home defined as household.See also: Crimes, robbery, forcible entry, unlawful, entry, attempted forcible entry, burglary, motor vehicle theftTHREATA threatening situation is one in which there was no physical contact between the offender and victim but thevictim felt that physical harm could have occurred. This includes nonverbal threats, e.g., brandishing a weapon,and verbal threats of physical harm which are made in person. Threats made over the telephone or threateningletters are not included.See also: Attack

TIME LOST FROM WORKNumber of cumulative days all affected household members were absent from work due to a crime. It can be timelost because injuries sustained in the incident kept the victim from returning to work immediately, or time taken offwork by a household member to do things related to the incident such as visiting a doctor, reporting the incidentto the police, seeing an insurance agent, or testifying in court. Since 1979, victims are asked in more detail about-63-their work time losses.TRANSIENT/NONTRANSIENTA transient place is a special place in which more than 25 percent of the rooms, suites, or cabins are occupied orintended for occupancy by transient guests. Transient guests usually stay less than a month and pay at dailyrates. A nontransient place is one in which 75 percent or more of the rooms, suites, or cabins are occupied orintended for occupancy by nontransient (permanent) guest or resident employees. Permanent guests usually staya month or more at reduced monthly or weekly rates.See also: Living QuartersUUASee: Urbanized Area (UA)

UNCLASSIFIED UNITSSee: Special PlaceUNEMPLOYEDIn the NCVS, civilians age 16 and older who were not "at work" during the reference week are consideredunemployed if they were available for work and:

1. had been "looking for work" within the past four weeks
2. were waiting to be called back to a job from which they had been laid-off or
3. were waiting to report to a new job within 30 days.

See also: Major Activity

UNIT SEGMENTSee: Segment

UNLAWFUL ENTRY (WITHOUT FORCE)A form of burglary in which the offender has no legal right to be on the premises, even though no force was usedto gain entrance.See also: Burglary, Forcible Entry, Attempted Forcible Entry

UNRECOGNIZABLE BUSINESSSee: Operation of BusinessUPDATINGAddition to, and correction of, original listing sheets which insures that every unit in the segment has a chance tobe selected for interview. Remodeled, no longer existing, replacement, demolished, converted-to-commercial,merged, newly-constructed, and converted-to-residential structures or units moved into are accounted for inupdating procedures. Permit segments are not updated.See also: Listing, SegmentURBAN-64-Census designation for persons living in places of 2,500 or more inhabitants incorporated as cities, boroughs,villages, and towns (except towns in New England, New York and Wisconsin); the densely settled urban fringe,whether incorporated or unincorporated, or urbanized areas; and unincorporated places of 2,500 or moreinhabitants.See also: Rural, Place, Land Use

URBANIZED AREA (UA)A broad geographical location containing a central city (or twin cities) of 50,000 or more inhabitants and thesurrounding closely settled incorporated and unincorporated areas. It differs from an SMSA chiefly in excludingthe rural portions of counties in an SMSA and those places separated from the densely populated fringe by ruralterritory. The boundaries are not considered permanent. Urbanized areas may not be uniquely identified withNCVS data files.See also: Central City, Standard Metropolitan Statistical Area, Place Description CodeV

VACANT HOUSING UNITA housing unit where no persons are residing during the interview period, or any persons found living thereusually reside elsewhere.See also: Housing Unit, Place of Residence

VALUE OF PROPERTY TAKENExcludes stolen cash and does not include property stolen from non-household members or recognizablebusinesses, nor does it include stolen checks, credit cards, travelers checks, cashier's checks, etc.VANDALISMDeliberate, intentional damage to or destruction of household property (e.g.: breaking windows, slashing tires,painting graffiti on walls).VARIANCEA statistic equal to the standard error squared used to express the precision of an estimate.See also: Standard Error, EstimateVICTIMIn the NCVS, the person or household involved in a criminal act during the specified reference period, for whoman Incident Report was completed.See also: Victimization

VICTIM-OFFENDER RELATIONSHIPThe respondent is asked to specify one of six broad categories indicating how well the victim knew the offender ina crime incident in which the victim was present:

1. Stranger: someone the victim had never seen before
2. By sight only: someone the victim had seen before, but had not said more than "hello" to
3. Casual acquaintance: someone the victim knew well enough to say more than "hello" to
4. Well-known: someone the victim had considered as such. The victim's judgment alone distinguishes

-65-between "casual acquaintances" and "well known". Well-known offenders are classified as relatives ornon-relatives.

5. Relative: by type of relationship. Subcategories are:

a. spouse at times of incidentb. ex-spouse at time of incidentc. parent or step-parentd. own child or step-childe. brother/sisterf. other relative (specified by respondent)1979 NOTE: The category "ex-spouse" was added as a separate category in 1979. Prior to that, interviewerswere instructed to code "ex-spouse" as "other relative".

6. Non-relative (i.e.: friends or close acquaintances). Subcategories are:

a. boyfriend, girlfriend, ex-boyfriend, ex-girlfriendb. friend or ex-friendc. roommate, boarderd. schoolmatee. neighborf. someone at work, customerg. other non-relative (specified by respondent)In crime incidents involving multiple offenders, the respondent may specify whether the offenders were:

1. All strangers to the victim
2. All relatives (by type of relationship)
3. Some relatives (by type of relationship)
4. All known
5. Some known

VICTIMIZATIONA specific criminal act as it affects a single victim. In criminal acts against persons, the number of victimizations isdetermined by the number of victims of such acts. Each criminal act against a household is assumed to involve asingle victim, the affected household.See also: Crimes, Incident, Victim, Victimization Rate, Household Victimization, Personal Victimization

VICTIMIZATION RATEA measure of crime occurrence among population groups at risk. Rates can be calculated for the generalpopulation (e.g.: the number of criminal victimizations per 1,000 persons or households) or for any crime type anddemographic measure (e.g.: the number of burglaries per 1,000 urban households or the number of robberies per1,000 males). As used in BJS publications:

1. For crimes against persons, the victimization rate is computed on the basis of the number of

victimizations per 1,000 resident population age 12 and over

2. For crimes against households, victimization rates are calculated on the basis of the number of incidents

per 1,000 households. See the separate note on computation of rates.See also: Weight, VictimizationWWEAPONItems such as guns (pistols, revolvers, rifles, shotguns, etc.) and knives are always considered weapons. Otherobjects are not considered weapons unless they are used as weapons; if the respondent felt threatened by theobject, then it is considered a weapon. The following items are not considered as weapons for the purposes of theNCVS: mace, tear gas, items of food, small empty cans, animals, parts of the body, BB and tear gas guns (exceptwhen used as clubs). Weapons other than guns or knives, e.g., rocks, clubs, etc., are not individually identifiable-66-in the data files.WEIGHTA factor applied to a sample element that allows computation of the size of the population the element represents.When a sample is interviewed, each incident or victimization represents a probable percentage of all cases if thetotal population were to be interviewed. Accordingly, each case is assigned a numerical coefficient, or weight,expressing its relative importance in a frequency distribution equaling that of the sample to the whole. Thus,multiplying each case by its assigned weight and summing over all cases provides an estimate of the entirepopulation. Because there are different proportions in the samples of households, persons and incidents withrespect to all possible households, persons, and incidents in the surveys, there are different weights for each.See also: Household Weight, Incident Weight, Person Weight, Estimate, Self-WeightingWORKINGWorking for pay or profit; working without pay on family farm or business; or working on jury duty, if being paid forit.See also: Major Activity

DRB Clearance Number CBDRB-FY23-POP001-0082 census.govJune 21, 2023MEMORANDUM FOR Heather N. Brotsos

Chief, Victimization Statistics Unit

Bureau of Justice StatisticsThrough: Meagan M. Meuchel

Survey Director, National Crime Victimization SurveyFrom: Anthony G. Tersine, Jr.

Chief, Demographic Statistical Methods DivisionSubject: Source and Accuracy Statement for the 2022 National Crime

Victimization SurveyThe purpose of this memorandum is to record and communicate the source and accuracy of the2022 National Crime Victimization Survey (NCVS). This document provides quality measures andnonresponse bias analysis in accordance with the U.S. Census Bureau’s Quality Standards.The U.S. Census Bureau reviewed the underlying data product for this report for unauthorizeddisclosure of confidential information and approved the disclosure avoidance practices appliedto this release. CBDRB-FY23-POP001-0082.If you have any questions about this document, please contact Forest Krueger atforest.krueger@census.gov or 301.763.5061, Zackary Scalyer at zackary.scalyer@census.gov or301.763.0445.Attachment: Source and Accuracy Statement for the 2022 National Crime Victimization Survey.cc:

Rachel Morgan (BJS)

Jennifer Truman

Steven Bittner (ADDP)

Edward Madrid

Emily Mohn

Megan Ruhnke

Christopher Seamands

Scott Raudabaugh (DSD)

James Farber (DSMD)

David Hornick

Forest Krueger

Consuela Payton

Zackary Scalyer

DRB Clearance Number CBDRB-FY23-POP001-0082Page intentionally left blank

DRB Clearance Number CBDRB-FY23-POP001-0082

Demographic Statistical Methods Division

Sample, Design, and Estimation Pool

Source and Accuracy Statement for the 2022

National Crime Victimization SurveyVersion 1.0June 21, 2023

Forest Krueger

Zackary Scalyer

James Farber, Assistant Division ChiefDavid V. Hornick, Lead Scientisti

DRB Clearance Number CBDRB-FY23-POP001-0082

Table of ContentsINTRODUCTION ..................................................................................................................... 1SOURCE OF THE DATA ........................................................................................................... 1KEY MEASURES................................................................................................................................ 1SAMPLE DESIGN............................................................................................................................... 1WEIGHTING .................................................................................................................................... 2ACCURACY OF THE ESTIMATES............................................................................................... 3RESPONSE RATES ............................................................................................................................. 4Overall Response Rates ........................................................................................................... 5Response Rates by Domain ..................................................................................................... 7Respondent Distributions ........................................................................................................ 9Modeled Response Estimates ............................................................................................... 11ITEM RESPONSE AND IMPUTATION RATES ........................................................................................... 14COVERAGE .................................................................................................................................... 15SAMPLE ATTRITION......................................................................................................................... 16VARIANCE ESTIMATION ....................................................................................................... 17DIRECT VARIANCE ESTIMATION ......................................................................................................... 17GENERALIZED VARIANCE FUNCTION ................................................................................................... 18SUMMARY .......................................................................................................................... 19REFERENCES ........................................................................................................................ 20ii

DRB Clearance Number CBDRB-FY23-POP001-0082TablesTable 1: 2022 National Crime Victimization Survey Weight Distributions ............................................ 3Table 3: Response Rates and Standard Errors for 2022 National Crime Victimization SurveyHouseholds ........................................................................................................................... 7Table 4: Weighted Response Rates and Standard Errors for 2022 National Crime VictimizationSurvey Persons ..................................................................................................................... 9Table 5: Weighted Respondent Distributions for 2022 National Crime Victimization SurveyHouseholds ......................................................................................................................... 10Table 6: Respondent Distributions for 2022 National Crime Victimization Survey Persons ............... 11Table 7: 2022 National Crime Victimization Survey Response Model Observations and C-Statistics. 12Table 8: 2022 National Crime Victimization Survey Modeled Response Estimates ............................ 13Table 11: 2022 National Crime Victimization Survey Person Coverage Ratios by Race/Ethnicity/Ageand Sex ............................................................................................................................... 16Table 12: 2022 National Crime Victimization Survey Attrition Rates by Time in Sample ................... 17Table 13: Coefficient of Variation (CV) of 2022 National Crime Victimization Survey Key Estimates . 18Table 14: National Crime Victimization Survey Generalized Variance Function Parameters ............. 19Table 15: Correlations for comparing 2022 National Crime Victimization Survey Estimates toPrevious Years .................................................................................................................... 22Table 16: Model Variables for 2022 National Crime Victimization Survey (NCVS) Nonresponse BiasAnalysis ............................................................................................................................... 23FiguresFigure 1: National Crime Victimization Survey Annual Workloads........................................................ 2Figure 2: Annual Weighted National Crime Victimization Survey Response Rates ............................... 6Figure 3: 2022 National Crime Victimization Survey Variance Estimates – Overall Property Crimes . 27Figure 4: 2022 National Crime Victimization Survey Variance Estimates – Domain Property Crimes 28Figure 5: 2022 National Crime Victimization Survey Variance Estimates – Overall Person Crimes .... 29Figure 6: 2022 National Crime Victimization Survey Variance Estimates – Domain Person Crimes ... 301

DRB Clearance Number CBDRB-FY23-POP001-0082IntroductionThe National Crime Victimization Survey (NCVS) is a nationally representative household surveysponsored by the Bureau of Justice Statistics (BJS). It is the nation's leading source ofinformation on criminal victimization, both reported and not reported to police. The U. S.Census Bureau (Census Bureau) serves as the primary data collection organization for the NCVSby conducting interviews and processing sample data on a monthly basis.This source and accuracy statement presents quality measures and nonresponse bias analysisfrom the NCVS interviews collected in 2022. The Census Bureau and BJS follow statistical qualitystandards and data quality guidelines to ensure the utility, objectivity, and integrity of allstatistical information released to sponsors and the public. Any NCVS quality measures that donot meet the Census Bureau quality standards are noted in this report. For detaileddocumentation of the Census Bureau statistical quality standards, refer to the Census Bureauwebsite (www.census.gov). For detailed documentation of the BJS data quality guidelines, referto the BJS website (www.bjs.ojp.gov).For a detailed description of NCVS sample design, data collection, weighting, and estimation,refer to the NCVS Technical Documentation on the BJS website (www.bjs.ojp.gov).

Source of the Data

Key MeasuresThe NCVS provides estimates of the number and rate of victimizations in a year for all crimecategories. These estimates are created for the entire United States as well as for smallerdomains defined by demographic characteristics of the victim or whether the crime wasreported to the police.The key measures for the NCVS are rates and totals for:• crimes of violence,• serious violence (rape/sexual assault, robbery, and aggravated assault),• property crimes, and• crimes not reported to the police.Key domains are:• sex,• race/Hispanic origin, and• age.

Sample DesignThe NCVS is a panel, or longitudinal, survey in which persons at each housing unit or groupquarters address are interviewed every six months, for a total of up to seven interviews. Theannual workloads presented in Figure 1 count every interview attempt at each sampled2

DRB Clearance Number CBDRB-FY23-POP001-0082address, so many addresses are counted twice within a year. The annual workload increasedfrom 256,500 in 2021 to 257,500 in 2022.Figure 1: National Crime Victimization Survey Annual WorkloadsSource: U.S. Census Bureau internal data from the 1996-2022 National Crime Victimization Survey.WeightingThe NCVS uses six sets of weights to create estimates of the population, crime victimizations,and crime incidents1 for households and persons. Refer to the NCVS Technical Documentation(www.bjs.ojp.gov) for a description of how each weight is created.Starting in 2020, the NCVS implemented a new extreme weight adjustment that will apply to allfuture NCVS cycles. The purpose of this adjustment is to reduce the impact on the estimatesfrom respondents who report an unusually high number of victimizations. Extreme cases areidentified as any single interview where the respondent reported at least 11 incidents(including series incidents2 up to ten times) and the sum of final victimization weights for theinterview is in the top three percent among all victims of the same crime type3. Thevictimization weights for extreme cases are capped at the 97th percentile among all victims ofthe same crime type. Final person and household weights are adjusted by the same factor for1 Incidents are distinguished from victimizations in that one criminal incident may have multiple victims orvictimizations.2 Series crimes are high frequency, repeat victimizations. When one type of crime occurred more than five timeswithin one six-month period, the different incidents were similar to each other, and the respondent could notrecall details to distinguish the different incidents from each other, respondents were recorded as havingexperienced series crimes and details were collected about the most recent incident only.3 Crime categories used to identify extreme household interviews are all property, burglary/trespassing, motorvehicle theft, and other theft. Crime categories used to identify extreme person interviews are all violent,rape/sexual assault, robbery, aggravated assault, and simple assault. An interview is flagged as extreme if it meetsextreme criteria within at least one crime type.050,000100,000150,000200,000250,000300,000199619971998199920002001200220032004200520062007200820092010201120122013201420152016201720182019202020212022

Annual HU WorkloadYear3

DRB Clearance Number CBDRB-FY23-POP001-0082extreme cases, and the removed weights are allocated to all other respondents (includingvictims and non-victims). This adjustment is done at the national level after the population ratioadjustment, so population totals for individual states and demographic groups are slightlydifferent from independent population controls used in the ratio adjustment.Table 1 shows the distribution of the six final weights used in the NCVS.Table 1: 2022 National Crime Victimization Survey Weight Distributions

Weight Min 1st Quartile Median Mean 3rd Quartile Max SumPerson 98 682 1,065 1,244 1,599 12,109 282,304,639Person Victimization 309 1,337 2,194 3,113 3,468 60,908 6,786,936Person Incident 65 1,240 2,004 2,932 3,356 43,069 6,392,133Household 83 533 796 913 1,215 5,326 131,259,549Household Victimization 197 980 1,466 1,823 2,378 47,798 13,426,686Household Incident 197 980 1,466 1,823 2,378 47,798 13,426,686Source: Census Bureau data from the 2022 National Crime Victimization Survey public use file.Victimization and incident weights can be larger than sample unit weights because they are fordifferent time periods. Two full six-month samples are used to produce estimates for a wholeyear. When counting crimes, the total crimes in the year are counted, regardless of the six-month period within which they occurred. However, when counting total persons orhouseholds, only the equivalent of one full sample is counted and the weights are adjusted byone half. Therefore, the unbounded single victimization weights are twice as large as thehousehold and person weights for each individual sample unit. Incident weights are equal tovictimization weights unless the crime incident involved more than one person – a concept thatdoes not apply to household crimes. The weighted sums represent the total person/householdpopulations and total number of victimizations and incidents.

Accuracy of the EstimatesA sample survey estimate has two types of error: sampling and nonsampling. The accuracy ofan estimate depends on both types of error. The nature of the sampling error is known giventhe survey design; the full extent of the nonsampling error is unknown.Since the NCVS estimates come from a sample, they may differ from counts or rates based onan enumeration of the entire population using the same questionnaires, instructions, andinterviewers. For a given estimator, the difference between an estimate based on a sample andthe estimate that would result if the sample were to include the entire population is known assampling error. Standard errors, which are presented below, are primarily measures of themagnitude of sampling error. However, they may include some nonsampling error.For a given estimator, the difference between the estimate that would result if the sample wereto include the entire population and the true population value being estimated is known asnonsampling error. There are several sources of nonsampling error that may occur during the4

DRB Clearance Number CBDRB-FY23-POP001-0082development or execution of the survey. It can occur because of circumstances created by theinterviewer, the respondent, the survey instrument, or the way the data are collected andprocessed. For example, nonsampling errors could occur because:• The interviewer records the wrong answer, the respondent provides incorrect information,the respondent estimates the requested information, or an unclear survey question ismisunderstood by the respondent (measurement error).• Some individuals who should have been included in the survey frame were missed(coverage error).• Responses are not collected from all sample units, or the respondent is unwilling toprovide information (nonresponse error).• The values are estimated imprecisely for missing data (imputation error).• Data may be incorrectly keyed, coded, or recoded (processing error).To minimize these errors, the Census Bureau applies quality control procedures during allstages of the production process including the design of the survey, wording of questions,review of the work of interviewers and coders, and statistical review of reports.This section describes the sources of nonsampling error that can be examined andmeasurements of sampling error. In accordance with Census Bureau and Office of Managementand Budget Quality Standards (U.S. Census Bureau, 2023; Office of Management and Budget,2006), this section also includes an assessment of nonresponse bias in the 2022 NCVS. Allsignificance testing uses α=0.05. All counts are rounded to the nearest hundred, and weightedsums are rounded to the nearest thousand. Unweighted and weighted values may not sum tooverall totals due to rounding. Rates and significance testing are based on unrounded numbers.All comparisons in this report are significant unless stated otherwise.

Response RatesThe NCVS has two different levels of unit response: household and within-household (orperson). If no persons within the household complete the person interview, the sample unit isconsidered a household nonresponse. If at least one person completes the person interview,the household is a sufficient partial response (not complete, but enough information to beconsidered a response). If all persons in the household complete person interviews, thehousehold is a complete response. Person interviews are considered within-householdresponse provided the person completes the screener questions and the crime incident reportquestions (if applicable). If two or more screener questions are not answered for any reason,the person is converted to within-household nonresponse in post data collection processing.The following sub-sections present overall response rates and nonresponse bias analysis in theform of response rates by subpopulation domains, respondent and nonrespondentdistributions, and modeled response estimates. Household response rate calculations usehousehold sampling weights, which are the inverse of the probability of sample selection5

DRB Clearance Number CBDRB-FY23-POP001-0082multiplied by subsampling factors. Person response rate calculations use the household weightsadjusted for household nonresponse.The NCVS collects demographic information during the interview, so households that have notpreviously responded to the NCVS have missing demographic data. Since missing data arecorrelated with nonresponse, all cases with missing data were removed from the analysis toallow for evaluation of reported demographics.

Overall Response RatesThe NCVS calculates separate household and within-household response rates as the ratio ofcomplete respondents and sufficient partials over in-scope households or persons. The productof these two response rates is the overall person response rate. Table 2 shows the weightedand unweighted household, within-household, and overall response rates for the 2022 NCVS.Table 2: 2022 National Crime Victimization Survey Response Rates

Response Category Count Weighted Sum Response/Nonresponse Rates

Unweighted WeightedSampled addresses 268,000 151,700,000In-scope 225,000 126,200,000Household response 144,000 82,440,000 63.9% 65.3%Type A nonresponse 81,000 43,790,000 30.2% 28.9%Refusal 66,000 35,090,000 29.3% 27.8%No one home 8,500 5,148,000 3.8% 4.1%Temporarily absent 550 346,000 0.3% 0.3%Language barrier 850 476,700 0.4% 0.4%Other nonresponse reason 5,200 2,723,000 2.3% 2.2%Out-of-scope 43,500 25,440,000Type B noninterview – vacant/URE* 30,000 17,520,000Type C noninterview – out-of-scope 14,000 7,921,000Persons 12+ within responding households 278,000 244,500,000Within-household response 227,000 201,700,000 81.7% 82.5%Within-household nonresponse 51,000 42,850,000 18.3% 17.5%Refusal 29,000 23,800,000 10.4% 9.7%Temporarily absent 850 776,400 0.3% 0.3%Never available 19,500 16,500,000 7.0% 6.7%Other nonresponse reason 2,000 1,784,000 0.7% 0.7%Overall person response 52.2% 53.9%Source: U.S. Census Bureau internal data from the 2022 National Crime Victimization Survey.*URE – Housing units that are temporarily occupied by persons who have a usual residence elsewhere.Type A nonresponse contains sample addresses that are eligible for interview but none of thepersons could be interviewed during the interview period. Type B noninterviews contain unitsthat are vacant or occupied by persons who have a usual residence elsewhere (URE). They areineligible during the current interview period, but they could become eligible at a later time.Type C units are permanently ineligible, and include demolished units, condemned units, unitsthat have been converted to permanent business or storage, unlocatable addresses, etc.6

DRB Clearance Number CBDRB-FY23-POP001-0082Because Type C cases are permanently ineligible for the NCVS, they are not sent out for laterinterviews. The total number of sampled addresses (268,000) in Table 2 includes approximately10,500 previous Type C units that were originally scheduled to be interviewed in 2021 but werenot sent out for another interview in 2022. The sample size presented in the Source of the Datasection (257,500) was the number of addresses sent out for interview in 2022. Figure 2 belowshows the weighted response rates for past NCVS years.Figure 2: Annual Weighted National Crime Victimization Survey Response RatesSource: U.S. Census Bureau internal data from the 1996-2022 National Crime Victimization Survey.Some degree of nonresponse bias and variance are normal features of almost all statisticalsurveys. The NCVS produces crime estimates using the answers from responding householdsand persons. These crime estimates will be biased if answers from respondents differ from thepotential answers of nonrespondents. The magnitude of nonresponse bias is a function of theresponse rate and differences between respondents and nonrespondents.One of the ways in which the NCVS attempts to minimize nonresponse bias is by increasingresponse rates and adjusting weights for potential differences between respondents andnonrespondents. Measures to increase response rates include conducting in-person interviewswith new sample units, mailing advanced letters for all sample units, providing a Spanishlanguage questionnaire for potential respondents who do not speak English, allowinginterpreters for potential respondents who do not speak English or Spanish, traininginterviewers to gain respondent cooperation, allowing proxy respondents in specialcircumstances, and mailing follow-up letters to nonresponding households.0%10%20%30%40%50%60%70%80%90%100%199619971998199920002001200220032004200520062007200820092010201120122013201420152016201720182019202020212022

Weighted Response RateYearHouseholdPersonOverall7

DRB Clearance Number CBDRB-FY23-POP001-0082Despite the measures taken to reduce nonresponse bias, there may still be some amount ofnonresponse bias that cannot be corrected without knowing the victimizations of thenonrespondents. The next three sub-sections examine potential nonresponse bias in the 2022NCVS.

Response Rates by DomainThe response rates represent the percentage of eligible sample cases that responded to theNCVS. It is useful to compare response rates for different subgroups to understand themagnitude of potential biases.Response rates are defined as:𝑅𝑅𝑅𝑅 = ∑ 𝑤𝑤𝑖𝑖𝑅𝑅𝑖𝑖𝐷𝐷𝑖𝑖𝑖𝑖∈𝑠𝑠∑ 𝑤𝑤𝑖𝑖𝐷𝐷𝑖𝑖𝑖𝑖∈𝑠𝑠where:𝑤𝑤𝑖𝑖= the appropriate weight (1 if unweighted) for the response rate calculation𝑅𝑅𝑖𝑖 = the response indicator (1 for respondents, 0 for nonrespondents)𝐷𝐷𝑖𝑖 = the domain indicator (1 if within domain of interest, 0 otherwise)𝑠𝑠 = the set of all eligible households or personsTable 3 provides the unweighted and weighted household counts, weighted response rate, andstandard error for each subgroup. Standard errors are conditional on the sample and representexpected variability in the response process, rather than traditional sampling error. Replicateweights were used to calculate the standard errors to account for the sample design.Table 2: Response Rates and Standard Errors for 2022 National Crime Victimization Survey Households

Characteristic UnweightedHouseholdsWeightedHouseholds

Weighted ResponseRate %StandardError %SigGrp *

Metropolitan Statistical Area (MSA) StatusPrincipal city within MSA 73,000 42,890,000 63.5% 0.47% CNot part of principal city within MSA 122,000 66,260,000 65.6% 0.35% BOutside of a MSA 30,500 17,070,000 68.8% 1.01% A

Urban/Rural StatusUrban 181,000 102,500,000 64.5% 0.68% BRural 43,500 23,690,000 68.9% 0.28% A

Race of Reference Person^White only 143,000 80,060,000 81.9% 0.25% ABlack only 21,500 12,490,000 77.9% 0.50% BOne other race only 11,000 6,550,000 78.9% 0.52% BTwo or more races 2,600 1,440,000 81.7% 0.93% AOverall 225,000 126,200,000 65.3% 0.29%Source: U.S. Census Bureau internal data from the 2022 National Crime Victimization Survey.

• Significance grouping - response rates with the same letter are not significantly different within each characteristic.

^ May not sum to overall totals because households with missing or imputed characteristic data are excluded.Households with missing or imputed data for each characteristic were excluded from theanalysis in Table 3. The sampling frame contains Metropolitan Statistical Area (MSA) status and8

DRB Clearance Number CBDRB-FY23-POP001-0082urban/rural status for all sample units, but race of reference person is collected during theNCVS interview. For nonresponding households, the race of reference person comes from priorinterviews, or from data collected from households that were converted to noninterviews inpost data collection processing due to insufficient data. Therefore, nonresponding householdswith no prior interviews are excluded from the denominator of the response rates for race ofreference person, making these rates all higher than the overall response rate.The significance grouping column (Sig Grp) indicates which demographic groups are significantlydifferent from the others4. All response rates identified with the same letter are notsignificantly different from each other, with A indicating the highest response rates, B indicatingthe next highest rates, and so on. In the race of reference person category, for example, whiteonly and two or more races had a significantly higher response rate the other two categories,but white only and two or more races were not significantly different from each other.The overall weighted response rate for 2022 NCVS households was 65.3 percent with astandard error of 0.29 percent. Housing units in the three MSA Status categories are allsignificantly different, where outside of a MSA had a significantly higher response rate than theother categories, and principal city within MSA had the lowest response rate. Rural householdshad higher response rates than urban households. Households with white only referencepersons and reference persons with two or more races had higher response rates than blackonly and one other race only reference persons households categories.Table 4 below provides within-household response rates, which do not account for householdnonresponse, by subgroup. For all persons in the NCVS, the within-household weightedresponse rate was 82.5 percent with a standard error of 0.24 percent. Persons with missing orimputed data for each characteristic are excluded from the analysis, and nonrespondents areassigned to characteristics reported in a prior interview.Females had a higher response rate than males. White only and black only persons had higherresponse rates than both Hispanic and non-Hispanic other races. The South region had thehighest response rate, and the other regions were not significantly different from each other.The different age groups all had different response rates, with higher response rates for olderrespondent groups. The largest differences in response rates were seen in the age variable,where the rates ranged from 54.3 percent for persons age 12 to 17 to 92.2 percent for personsage 65 and older.4 P-values adjusted for multiple comparisons within each demographic characteristic using the Tukey-Kramercorrection (Hayter, 1984 and Rafter et al., 2002).9

DRB Clearance Number CBDRB-FY23-POP001-0082Table 3: Weighted Response Rates and Standard Errors for 2022 National Crime Victimization Survey Persons

Characteristic UnweightedPersons^WeightedPersons^Weighted

Response Rate%

Standard Error% Sig Grp*SexMale 134,000 118,100,000 81.5% 0.27% BFemale 143,000 126,300,000 83.4% 0.25% AAge12-17 20,000 17,570,000 54.3% 0.76% F18-24 22,500 19,770,000 67.7% 0.57% E25-34 37,500 33,170,000 82.9% 0.33% D35-49 58,500 51,780,000 85.2% 0.26% C50-64 63,000 55,030,000 86.6% 0.24% B65 and older 66,000 56,970,000 92.2% 0.19% A

Race/Hispanic OriginHispanic 42,000 42,050,000 79.0% 0.46% BNon-Hispanic white only 184,000 153,900,000 83.9% 0.27% ANon-Hispanic black only 28,000 26,040,000 83.3% 0.52% ANon-Hispanic other races 22,000 20,550,000 79.5% 0.47% BRegionNortheast 43,000 42,970,000 80.6% 0.41% BMidwest 77,500 51,610,000 79.9% 0.61% BSouth 93,500 91,940,000 85.4% 0.45% AWest 64,000 58,010,000 81.5% 0.39% BOverall 278,000 244,500,000 82.5% 0.24%Source: U.S. Census Bureau internal data from the 2022 National Crime Victimization Survey.^ Subgroups may not sum to overall totals because persons with missing/imputed characteristic data are excluded.

• Significance grouping - response rates with the same letter not significantly different within each characteristic.

Respondent DistributionsRespondent and nonrespondent distributions show the relative percent contribution ofmembers of a domain subset to respondent and nonrespondent populations separately. This isdifferent than the response rates, which are the relative percent of respondents within thedifferent domain subsets. Respondent distributions are defined as:𝑅𝑅𝐷𝐷 = ∑ 𝑤𝑤𝑖𝑖𝑅𝑅𝑖𝑖𝐷𝐷𝑖𝑖𝑖𝑖∈𝑠𝑠∑ 𝑤𝑤𝑖𝑖𝑅𝑅𝑖𝑖𝑖𝑖∈𝑠𝑠This definition assumes the same eligibility criteria, weights, and indicators as the response ratecalculations in the previous section. Nonrespondent distributions use the same formula, butwith the 𝑅𝑅𝑖𝑖 variable indicating nonrespondents instead of respondents. The Rao-Scott Chi-square test (Rao and Scott, 1984) determines if respondent and nonrespondent distributionsdiffer, using replicate weights to account for the sample design.Table 5 shows the percent of total sample distribution, as well as comparisons of respondentand nonrespondent distributions for NCVS households within the different domain subgroups.As in Table 3, households that had missing or imputed values were excluded from this analysis,and race of reference person came from prior interviews for nonrespondents.10

DRB Clearance Number CBDRB-FY23-POP001-0082The Rao-Scott Chi-square test in the two right columns evaluates whether each characteristichas different respondent and nonrespondent distributions. All characteristics show significantdistribution differences, but this does not necessarily indicate that the differences are largeenough to suggest nonresponse bias. Nonresponse bias occurs when differences are correlatedwith the outcome of interest, namely criminal victimization. If the differences betweenrespondents and nonrespondents are related to different rates of criminal victimization, thereis nonresponse bias. There is no way to know the nonrespondents’ victimizations, so theamount of nonresponse bias is unknown.Table 4: Weighted Respondent Distributions for 2022 National Crime Victimization Survey HouseholdsPercent

of TotalSampleHouseholdRespondent%HouseholdNonrespondent%Chi-SquareStatistic{df*}P-value

Metropolitan Statistical Area (MSA) StatusPrincipal city within MSA 34.0% 33.0% 35.8% 28.6{2} <.0001Not part of principal city within MSA 52.5% 52.7% 52.1%

Outside of a MSA 13.5% 14.2% 12.2%

Urban/Rural StatusUrban 81.2% 80.2% 83.2% 42.1{1} <.0001Rural 18.8% 19.8% 16.8%

Race of Reference PersonWhite only 79.6% 80.3% 76.7%113.7{3} <.0001Black only 12.4% 11.9% 14.6%One other race only 6.5% 6.3% 7.3%Two or more races 1.4% 1.4% 1.4%Source: U.S. Census Bureau internal data from the 2022 National Crime Victimization Survey.*df – Degrees of freedom.Table 6 shows the percent of total sample distribution, as well as comparisons of respondentand nonrespondent distributions for NCVS persons within the different domain subgroups.Similar to households, the Rao-Scott Chi-square tests show significant differences for thedistributions of all variables. The distributions for age show substantial differences between therespondent and nonrespondent distributions, which correspond to the magnitude of the Chi-square test statistic.The differences found in response rates and distributions among domain groups betweenrespondents and nonrespondents are accounted for in the nonresponse weighting adjustmentsat both household level and person level. All domain variables examined in the nonresponseanalysis are included to form the nonresponse adjustment cells. Nonresponse adjustmentfactors are then calculated by allocating the weight of nonrespondents to respondents.The nonresponse adjustment is designed to minimize the nonresponse impact in the finalvictimization estimates.11

DRB Clearance Number CBDRB-FY23-POP001-0082Table 5: Respondent Distributions for 2022 National Crime Victimization Survey PersonsCharacteristic

Percent ofTotalSamplePersonRespondent%PersonNonrespondent%Chi-SquareStatistic{df*}P-valueSexMale 48.3% 47.8% 51.0% 96.1{1} <.0001Female 51.7% 52.2% 49.0%Age12-17 7.5% 4.9% 20.3%9988.2{5} <.000118-24 8.4% 6.9% 16.1%25-34 14.2% 14.1% 14.4%35-49 22.1% 22.7% 19.3%50-64 23.5% 24.5% 18.6%65 and older 24.3% 27.0% 11.3%

Race/ Hispanic OriginHispanic 17.3% 16.6% 21.0%178.8{3} <.0001Non-Hispanic white only 63.5% 64.4% 58.8%Non-Hispanic black only 10.7% 10.8% 10.3%Non-Hispanic other races 8.5% 8.2% 10.0%RegionNortheast 17.6% 17.2% 19.5%102.1{3} <.0001Midwest 21.1% 20.4% 24.3%South 37.6% 38.9% 31.3%West 23.7% 23.5% 25.0%Source: U.S. Census Bureau internal data from the 2022 National Crime Victimization Survey.*df – Degrees of freedom.

Modeled Response EstimatesNonresponse bias is a function of the response rate and the difference between therespondents and the nonrespondents on a specific variable of interest. One challenge ofnonresponse bias analysis is that there is no way to know the key items for nonrespondents.The nonresponse bias analysis in this section applied multivariate regression to predictresponses to key items for both respondents and nonrespondents, then compared the model-based estimates for respondents and the full sample (methods described in Andridge and Little,2011).Several sources of data were used, in addition to the 2022 NCVS interview data, for thisanalysis. Census block and tract demographic characteristics were included from the 2020Census and from 2016-2020 American Community Survey (ACS) results. Census data comesfrom internal Census Bureau files and ACS data is available to the public on the PlanningDatabase (U.S. Census Bureau, 2022). County-level crime data were included from the FederalBureau of Investigation’s Uniform Crime Reporting (UCR) Program from 2016, and the State ofIllinois data from 20165. The models were constructed using the NCVS respondent data, thenapplied to respondents and nonrespondents to produce complete response predictions.A total of five responses were predicted using five different models:5 Historically, Illinois has not provided their crime reports to the FBI, so the UCR contains imputed data for Illinois.Therefore, instead of using UCR Illinois data, this analysis used the data directly from the State of Illinois.12

DRB Clearance Number CBDRB-FY23-POP001-0082• Households:o Did the household disclose any property crimes (yes/no)?o Did the household disclose any property crimes NOT reported to the police (yes/no)?• Persons:o Did the person disclose any crimes of violence (yes/no)?o Did the person disclose any crimes of serious violence (yes/no)?o Did the person disclose any personal crimes (including purse snatching and pocketpicking) NOT reported to the police (yes/no)?For each question, a stepwise logistic regression model was generated, excluding observationsmissing any covariate data. Each model produced a predicted propensity of victimization(values ranging from 0 to 1) for all sample units with complete covariate information. Table 7indicates the number of completed NCVS interviews and the number of observations that hadno missing covariates and were ultimately used to develop each model. For a complete list ofthe variables used within each model, see Appendix B.Table 6: 2022 National Crime Victimization Survey Response Model Observations and C-Statistics

Model CompletedInterviewsModelObservationsModelC-Statistic

Property Crimes 144,000 144,000 0.687

Property Crimes Not Reported to Police 144,000 144,000 0.701

Personal Violent Crimes 227,000 217,000 0.764

Personal Serious Violent Crimes 227,000 219,000 0.793

Personal Crimes Not Reported to Police 227,000 217,000 0.786Source: Federal Bureau of Investigation 2016 Uniform Crime Report data; U.S. Census Bureau 2016-2020 AmericanCommunity Survey data, internal data from the 2022 National Crime Victimization Survey, and 2010 Census data.Table 7 also presents the c-statistics for each model, which measure the goodness-of-fit of alogistic regression and range from 0.5 to 1.0. A c-statistic of 0.5 indicates that the model is nobetter than chance at predicting whether a member belongs to a particular group, whereas a c-statistic of 1.0 indicates that the model perfectly predicts the inclusion within a group. Modelsare typically considered reasonable when the c-statistic is higher than 0.7 and strong when thec-statistic exceeds 0.8 (Hosmer & Lemeshow, 2000). Using these criteria, the first model,“property crimes”, is not reliable enough to use for the remaining analysis.Outcomes were predicted for each person (or household) in the sample using the victimizationpropensities. For each model, the victimization propensities for the sample were ordered fromform largest to smallest, and the top propensities above a calculated cutoff were assigned aone (predicted to be a victim of a crime). The cutoff was calculated as the sum of all samplepropensities, rounded to the next whole number. For example, the sum of the samplepropensities in the serious violent crimes model was 871.39, and thus the top 871 propensitieswere assigned a one.13

DRB Clearance Number CBDRB-FY23-POP001-0082The modeled response estimates in Table 8 are calculated as the percent of persons with atleast one predicted victimization in a six-month period for the nonrespondents, respondents,and the full sample. This definition is different from the NCVS victimization rate estimates,which are the annual victimizations per 1,000 persons. The household base weight used forthese estimates is the sampling base weight, while the person base weight is the sampling baseweight after adjusting for household nonresponse. All persons within the same respondinghousehold have the same person base weight. Estimates using nonresponse adjusted weightsare also included to show how the nonresponse adjustment affects the estimates from just therespondents.Table 7: 2022 National Crime Victimization Survey Modeled Response Estimates

Model Weight Full

Sample Nonrespondents Respondents 95% CI* forRespondents

Property Crimes Not

Reported to PoliceBase weight 2.85% 2.57% 2.99% 2.61%-3.38%Nonresponse adjusted weight 2.95% 2.61%-3.29%

Violent Crimes Base weight 0.76% 0.29% 0.85% 0.76%-0.95%Nonresponse adjusted weight 0.83% 0.74%-0.93%

Serious Violence Base weight 0.36% 0.10% 0.41% 0.33%-0.50%Nonresponse adjusted weight 0.39% 0.31%-0.48%

Person Crimes Not

Reported to PoliceBase weight 0.44% 0.20% 0.48% 0.40%-0.57%Nonresponse adjusted weight 0.48% 0.40%-0.56%Source: Federal Bureau of Investigation 2016 Uniform Crime Report data, U.S. Census Bureau 2016-2020 AmericanCommunity Survey data, internal data from the 2022 National Crime Victimization Survey, and 2020 Census data.CI – Confidence Interval.Using replicate weights, 95 percent confidence intervals were included for each of therespondent means in Table 8. For all crime types, the full-sample estimates fell inside thecorresponding 95 percent confidence interval for the respondent mean.6 Therefore, for thesecrime types, this analysis could not detect a difference in the modeled estimates between thefull sample and the respondents alone, indicating no evidence of nonresponse bias.It is important to remember that these estimates are calculated from predictions of personalcrime victimization for both respondents and nonrespondents and may not match actualresponse data. Due to the model design, the predicted values mimic the victimization patternsobserved among respondents, but there is no guarantee that these predictions match whatwould be observed if all persons responded to the NCVS. There is no way to know the trueresponses of nonrespondents, but these modeled predictions provide the best available insightinto nonresponse bias.6 Due to rounding, the full-sample estimate of modeled violent crimes is shown as equivalent to the lower 95%confidence bound for the respondent mean.14

DRB Clearance Number CBDRB-FY23-POP001-0082

Item Response and Imputation RatesRespondents sometimes do not answer specific questions because they do not know theanswer, or they refuse to provide the answer. NCVS editing and coding operations may fill inmissing values or create new missing values if the interviewer notes indicate that the interviewshould have taken a different path through the instrument. The item response rates presentedin Table 9 are based on the original interview data before editing and coding. For exactquestion wording, refer to the NCVS questionnaires available on the BJS website(www.bjs.ojp.gov).Table 9: 2022 National Crime Victimization Survey Weighted Item Response Rates

Key item Response Rate*

Person DemographicsAge 96.6%Race 99.1%Sex 100.0%

Hispanic Origin 99.8%

Screener QuestionsS1-Has anything been stolen? 100.0%S2-Has any vehicle or vehicle part been stolen or used without permission? 100.0%S3-Has anyone attacked or threatened you in any way? 100.0%S4-Have you been forced or coerced to engage in unwanted sexual activity? 99.9%

Incident Report QuestionsI1-Was a household member present when the incident occurred? 98.9%I2-Did the offender hit you, knock you down, or attack you in any way? 99.8%I3-Was something stolen? 99.9%I4-Were police informed? 99.9%Source: U.S. Census Bureau internal data from 2022 National Crime Victimization Survey interviews.*Item response rate calculations use final weights. All values of 100% are due to rounding.The NCVS accounts for unit nonresponse through weighting adjustments by assigning theweights of nonrespondents to respondents with similar characteristics. Most item nonresponseis not corrected, and missing values remain in the data. However, the NCVS processing requiresvalid values for some items, and a subset of items are corrected through hot-deck imputation7.Table 10 presents unweighted rates of item imputation for all such items.The eligible units column indicates all sample cases that required valid values. Some itemsrequire values for only respondents, and some items require values for respondents and Type Anonrespondents. The NCVS household data contain imputed values for vacant and out-of-scope(Type B and Type C) noninterviews, but they are not included in this table.7 Both the original and imputed values are available on the NCVS public use files.15

DRB Clearance Number CBDRB-FY23-POP001-0082Table 10: 2022 National Crime Victimization Survey Unweighted Item Imputation Rates

Item Eligible Units Imputed ImputationRateHousehold income 144,000 35,500 24.7%Household land use 144,000 2,200 1.5%Household farm sales 144,000 550 0.4%Household reference person race* 225,000 47,500 21.1%

Person age 278,000 11,000 4.0%

Person sex 278,000 200 0.1%Person race* 278,000 2,900 1.0%Person ethnicity 278,000 850 0.3%Source: U.S. Census Bureau internal data from the 2022 National Crime Victimization Survey.

• Household reference person race differs from the person race in the definition of eligible units. Household

reference person race is all eligible housing units including nonresponding units, while person race is all eligiblepersons within responding housing units.CoverageCoverage errors occur when the total population that could be selected for a sample differsfrom the survey’s target population. Missed housing units and missed people within samplehouseholds create undercoverage, which adds bias to survey estimates if the missed populationdiffers from those interviewed on the key survey items. Duplicate housing units or persons andout-of-scope units (for example vacant housing units, businesses, and institutional groupquarters) create overcoverage, which can diminish the precision of estimates.Coverage ratios are the uncontrolled estimated totals divided by independent populationcontrols. The occupied household coverage ratio compares the NCVS weighted sums to the ACSestimates of occupied housing units. The person coverage ratios use annual independentpopulation controls available within the Census Bureau.Table 11 presents the 2022 NCVS person coverage ratios by demographic group and type ofweights used. The uncontrolled coverage ratios are calculated using NCVS sampling weightsthat are adjusted for nonresponse but not adjusted to match independent controls. Thecontrolled coverage ratios are calculated using final weights that are adjusted to matchindependent controls and include the new extreme weight adjustment. A coverage ratio of oneindicates that the survey estimate perfectly matches the independent control. A coverage ratioless than one indicates undercoverage, and a coverage ratio greater than one indicatesovercoverage.16

DRB Clearance Number CBDRB-FY23-POP001-0082Table 8: 2022 National Crime Victimization Survey Person Coverage Ratios by Race/Ethnicity/Age and Sex

Demographic Category Uncontrolled Controlled

Male Female Male FemaleOverall 0.85 0.88 1.00 1.00Hispanic 0.80 0.86 1.00 1.00Non-Hispanic white only 0.90 0.92 1.00 1.00Non-Hispanic black only 0.72 0.78 1.00 1.00Non-Hispanic other races 0.83 0.79 1.00 1.00Age 12-17 0.69 0.66 1.00 1.00Age 18-24 0.71 0.66 1.00 1.00Age 25-34 0.77 0.80 1.00 1.00Age 35-49 0.85 0.89 1.00 1.00Age 50-64 0.89 0.94 1.00 1.00Age 65 and older 1.04 1.05 1.00 1.00Source: U.S. Census Bureau internal data from the 2022 National Crime Victimization Survey.The coverage ratios in Table 11 are provided separately for males and females. Theuncontrolled coverage ratio for females age 12 to 17 and 18 to 24 and for males age 12-17 arebelow the Census Bureau standard of 0.7, which indicates potential data quality issues. Thecombined overall person coverage ratios were 0.86 using uncontrolled weights and 1.00 usingcontrolled weights. The overall occupied housing unit coverage ratios were 0.98 usinguncontrolled weights and 1.02 using controlled weights.

Sample AttritionThe NCVS sample housing units and group quarters are assigned for seven different interviewsover a three-year period. The first interview is TIS 1, the second interview is TIS 2, and so onthrough TIS 7. Each occupied sample unit may respond during one TIS but not respond duringanother.Table 12 lists the household and person attrition rates for the 2022 NCVS. The attrition rate isthe difference in response rates between the current and previous TIS. These measures includeall sample units from January through December 2022 and compare their response during 2022to their responses from six months prior (July 2021 through June 2022). During typical years,the NCVS sample contains an approximately equal distribution of sample units on their firstthrough seventh interviews. During 2020, TIS 1 cases were excluded from six months of datacollection (April through September) due to pandemic restrictions. These cases startedinterviewing for the first time in October 2020 and are counted as TIS 3 and TIS 4 in Table 12.17

DRB Clearance Number CBDRB-FY23-POP001-0082Table 92: 2022 National Crime Victimization Survey Attrition Rates by Time in Sample

Time in Sample Unweighted Households Net Attrition Rate

Households Persons1 33,000 - -2 33,000 -2.30% -0.11%3 40,500 0.08% -0.80%4 40,000 -0.25% 0.25%5 24,000 1.89% -0.08%6 23,500 -0.96% -0.83%7 31,500 -0.96% -1.04%Source: U.S. Census Bureau internal data from the 2022 National Crime Victimization Survey.Negative net attrition rates indicate that cases within that time in sample had better overallresponse rates than their previous time in sample. For example, the household attrition rate forTIS 2 is negative, which indicates that the response rate for persons on TIS 2 in 2022 was higherthan the response rate for the same sample cases on TIS 1.

Variance Estimation

Direct Variance EstimationVariance estimation is a way to quantify the sampling error in a survey estimate. The CensusBureau calculates point estimates for all crime categories in the NCVS crime taxonomy for theentire nation as well as for demographic subsets of the population, or domains. Each estimateis the sum of final victimization/incident weights among respondents who reported each typeof incident. All occurrences of series crimes are included in these estimates and are countedbased on the number of times the incident occurred within the series, up to a maximum of 10.The variance of these point estimates is calculated using Fay’s balanced repeated replicationformula (Fay, 1989) with 160 replicate weights.The coefficient of variation (CV), or relative standard error, provides a standardized way tocompare the magnitude of standard errors across different estimates. Table 13 presents theCVs, which are the standard errors divided by the estimates, for key estimates from the 2022NCVS. Census Bureau standards require that the CV for most key estimates be under 30 percentin order for the data to be released without restrictions.Household demographics are based on the demographics of the reference person, who is anyresponsible adult household member who is not likely to permanently leave the household. Thereference person is usually (but not always) the person who completes the householdinterview. Age categories are different for property crimes because age distributions ofhousehold reference persons are different than age distributions of all persons.18

DRB Clearance Number CBDRB-FY23-POP001-0082Table 10: Coefficient of Variation (CV) of 2022 National Crime Victimization Survey Key Estimates

Key Domains

Person Crimes Property Crimes

Violent Crimes Serious Violence Crimes

per 1000 CVCrimesper 1000

CV Crimesper 1000 CVNational total 23.5 3.8% 9.8 5.6% 101.9 2.0%Race = black only 22.2 9.3% 9.4 12.4% 105.7 5.3%Race = not black only 23.7 4.3% 9.9 6.2% 101.3 2.1%

Ethnicity = Hispanic 22.6 8.9% 11.0 11.6% 132.3 4.0%Age = 12-24* 32.4 8.0% 13.9 10.9% 147.4 3.6%Age = 25-49** 29.3 6.3% 11.8 8.0% 114.8 3.6%Age = 50+ 14.1 7.1% 6.1 12.2% 76.1 2.8%Crime reported to police 9.7 5.7% 4.7 7.3% 32.4 3.3%Crime not reported to police 13.2 5.1% 5.0 7.9% 68.0 2.3%

Sex = Male 23.5 5.2% 9.5 8.6% N/A N/A

Sex = Female 23.4 5.3% 10.0 7.7% N/A N/ASource: U.S. Census Bureau data from the 2022 National Crime Victimization Survey public use file.For property crimes, this age category is 12-34For property crimes, this age category is 35-49

Generalized Variance FunctionGeneralized variance functions (GVF) provide a simple way to estimate the variance of anyNCVS estimate based on the value of the estimate. The Census Bureau creates GVF parametersby fitting all NCVS estimates to a model that predicts the relative variance of the crime rate as afunction of the crime rate.During GVF calculations, estimates are removed from processing if they have absolutestudentized residuals above 3.0 and absolute measure of influence (DFFITS) values above 0.22.A complete list of estimates that were removed from the data during iterative GVF modeling isincluded in Appendix C.Table 14 presents the current and two previous years’ NCVS GVF parameters for overall anddomain estimates of person and property crimes.19

DRB Clearance Number CBDRB-FY23-POP001-0082Table 11: National Crime Victimization Survey Generalized Variance Function Parameters

Year Parameter Set # Description a b C2022

1 Overall Person Crime Estimates -0.00017101 5,590 2.541

2 Person Crime Domain Estimates -0.00038364 3,991 6.208

3 Overall Property Crime Estimates -0.00012374 2,536 1.196

4 Property Crime Domain Estimates -0.00012797 2,767 1.2252021

1 Overall Person Crime Estimates -0.00030659 2,185 4.992

2 Person Crime Domain Estimates -0.00049937 1,913 8.229

3 Overall Property Crime Estimates -0.00009503 2,761 0.838

4 Property Crime Domain Estimates -0.00012910 2,601 1.2392020

1 Overall Person Crime Estimates -0.00026442 6,177 4.039

2 Person Crime Domain Estimates -0.00068320 2,338 11.25

3 Overall Property Crime Estimates -0.00013449 2,998 1.256

4 Property Crime Domain Estimates -0.00015281 2,890 1.473Source: U.S. Census Bureau public data from National Crime Victimization Survey 2020-2022.Using these parameters, data users can calculate variance estimates of any NCVS estimate 𝑡𝑡̂ as:𝑣𝑣(𝑡𝑡̂) = 𝑎𝑎𝑡𝑡̂2 + 𝑏𝑏𝑡𝑡̂ + 𝑐𝑐𝑡𝑡̂3/2Comparisons of direct and GVF variance estimates of household and person victimizations arealso included in Appendix C. The GVF parameters perform reasonably well for small estimatesbut tend to overestimate the variance for larger person crime estimates.When making comparisons to estimates from previous years, use the year-to-year correlationcoefficients in Appendix A. Refer to the NCVS Technical Documentation for more on how to useGVF parameters and correlation coefficients to estimate variances.SummaryThe overall response rate of the 2022 NCVS was 53.9 percent, which causes concern ofpotential bias in the NCVS estimates. The nonresponse bias analysis in this report foundsignificant differences in response rates between different demographic subgroups.Respondent and nonrespondent distributions are significantly different for all key domains.However, nonresponse weighting adjustments are expected to correct for these differencesbecause the key domains are used in the adjustment.The NCVS model-based estimates found no evidence of nonresponse bias after nonresponseweighting adjustments are applied.Item response in the 2022 NCVS is generally good; age has the lowest item response rateamong the key questions with 96.6 percent.There is undercoverage in the NCVS sample, but the weighting adjustments correct for theundercoverage of the population.20

DRB Clearance Number CBDRB-FY23-POP001-0082Since the full extent of the nonsampling error is unknown, one should be particularly carefulwhen interpreting results based on small differences between estimates. Caution should alsobe used when interpreting results based on a relatively small number of cases. Crime measures(such as totals and means) likely are not reliable when computed on a subpopulation smallerthan 1,500,000. Note that this subpopulation cutoff specifically applies to violent crimes.ReferencesAndridge, R. R. and Little, R. J. A. (2011) Proxy Pattern-Mixture Analysis for SurveyNonresponse. Journal of Official Statistics, Vol. 27, No. 2, 2011, pp. 153-180.Berzofsky, A., Moore, A., Couzens, G., Langton, L., and Krebs, C. (2020). Potential Survey ErrorDue to a Panel Design: A Review and Evaluation of the National Crime VictimizationSurvey. RTI Research Report ISSN 2378-7902, issued July 2020. Retrieved fromhttps://www.rti.org/rti-press-publication/potential-survey-error-panel-design/fulltext.pdfBureau of Justice Statistics (2017). National Crime Victimization Survey, 2016: TechnicalDocumentation. Retrieved fromhttps://bjs.ojp.gov/sites/g/files/xyckuh236/files/media/document/ncvstd16.pdfFay, R. E. (1989). Theory and Application of Replicate Weighting for Variance Calculations.Proceedings of the Survey Research Methods Section, 212-217, American StatisticalAssociation. Retrieved fromhttps://ww2.amstat.org/sections/srms/Proceedings/papers/1989_033.pdfHayter, A. J. (1984). A Proof of the Conjecture that the Tukey-Kramer Multiple ComparisonsProcedure is Conservative. The Annals of Statistics 12 no. 1, 61-75. Retrieved fromhttps://projecteuclid.org/euclid.aos/1176346392Hosmer, D. W. and Lemeshow, S. (2000). Applied Logistic Regression (2nd Edition). New York,NY: John Wiley & Sons.Office of Management and Budget. (2006). Standards and guidelines for statistical surveys.Retrieved fromhttps://obamawhitehouse.archives.gov/sites/default/files/omb/inforeg/statpolicy/standards_stat_surveys.pdfRafter, J., Abell, M., and Braselton, J. (2002). Multiple Comparison Methods for Means. Societyfor Industrial and Applied Mathematics, 44(2), 259-278. Retrieved fromhttp://www.jstor.org/stable/414835521

DRB Clearance Number CBDRB-FY23-POP001-0082Rao, J. N. K. and Scott, A. J. (1984). On Chi-Squared Tests for Multiway Contingency Tables withCell Proportions Estimated from Survey Data. The Annals of Statistics 12 (1984), no. 1,46-60. Retrieved from https://projecteuclid.org/euclid.aos/1176346391U.S. Census Bureau (2022). 2022 Planning database. Retrieved fromhttps://www.census.gov/topics/research/guidance/planning-databases.htmlU.S. Census Bureau (2023). U.S. Census Bureau statistical quality standards. Retrieved fromhttps://www2.census.gov/about/policies/quality/quality-standards.pdf22

DRB Clearance Number CBDRB-FY23-POP001-0082

Appendix A – Year-to-Year Correlation CoefficientsTable 12: Correlations for comparing 2022 National Crime Victimization Survey Estimates to Previous Years

Crime Category 2021 2020

Total Personal Crimes 0.101 0.113

Crimes of Violence 0.119 0.127Completed 0.087 0.049Attempted 0.143 0.134

Rape/Sexual Assault 0.196 0.044Rape 0.111 0.038Completed 0.178 0.010Attempted 0.044 0.020

Sexual Assault 0.113 0.145Robbery 0.026 -0.067Completed 0.102 -0.087

With Injury 0.044 -0.109

Without Injury 0.088 -0.046Attempted -0.047 -0.037

With Injury 0.059 -0.018

Without Injury 0.050 0.038Assault 0.126 0.166Aggravated -0.117 0.095

Completed with Injury -0.055 -0.043Attempted -0.037 0.006Simple 0.178 0.151

Completed with Injury -0.032 0.077Attempted 0.193 0.138

Purse Snatching/Pocket Picking 0.026 -0.008Completed 0.010 0.005Attempted 0.037 0.022

Pocket Picking -0.023 0.013

Serious Violence 0.057 0.062

Total Property crimes 0.105 0.066Burglary 0.101 0.118Completed 0.053 0.114

Forcible Entry -0.021 0.075

Unlawful Entry without Force -0.030 0.090

Attempted Forcible Entry 0.031 -0.091

Motor Vehicle Theft 0.126 0.064Completed 0.099 0.043Attempted -0.017 0.093Theft -0.009 0.007Completed -0.037 -0.026Less than $50 0.136 -0.070$50 - $249 -0.020 -0.021$250 or more -0.156 -0.121Amount not available -0.012 0.126Attempted -0.020 0.097Source: U.S. Census Bureau public data from National Crime Victimization Survey 2020-2022.23

DRB Clearance Number CBDRB-FY23-POP001-0082

Appendix B – List of Variables in Each Response ModelThe models for nonresponse bias analysis were created using stepwise logistic regression with avariety of variables potentially related to the dependent variable from the 2020 NCVS data, the2020 Census, the 2016-2020 ACS, the 2016 UCR data and the State of Illinois data. The tablebelow lists all covariates in the final models.Let the following indicate the corresponding models:

1. Property Crime
2. Property Crime Not Reported to Police
3. Violent Crime
4. Serious Violent Crime
5. Personal Crime Not Reported to Police

Table 13: Model Variables for 2022 National Crime Victimization Survey (NCVS) Nonresponse Bias Analysis

Covariate Model1 2 3 4 5

2022 NCVS Sampling and Response Data – Household levelHousehold income~ X X X X XHousehold tenure (rent/own/occupy free) X X XHousehold time in sample X XMetropolitan Statistical Area (MSA) principal city status X X X

State X X X X X

Stratification PSU Type (self-representing/non-self-representing) X XYear and quarter of data collection XNumber of persons in the household X X XSquared number of persons in the household X X X

Poverty Code X X

Urban Area Type X X

2022 NCVS Sampling and Response Data – Person levelAge X XAge squared X X XAge group X X X

Education X XPerson time in sample X X XRace/Hispanic origin X XRelationship to reference person X X XNumber of persons in the household squared * race/Hispanicorigin interaction X

2020 Census Measures

Percent of Race/Hispanic origin X X

Percent of Race/Black origin X

Percent of Race/Pacific Islander origin X

Percent of Multi Racial/Non-Hispanic origin X X X X XPercent of all housing units that are classified as the usualplace of residence of the individual or group living in it X24

DRB Clearance Number CBDRB-FY23-POP001-0082

Covariate Model1 2 3 4 5Percent of all housing units that have no regular occupants onCensus Day; housing units with its usual occupants temporarilyaway X

2016-2020 American Community Survey EstimatesTotal population X XPopulation density (based on ACS population estimates andCensus land area) X XThe percentage of all housing units that are in a structure thatcontains two to nine housing units XPercent of age 25+ persons with no high school diploma orequivalent X XThe percentage of occupied housing units that are not owneroccupied X X XPercent of population between 25 and 44 years old X XPercent of all occupied housing units where a Spanish orSpanish Creole language was assigned as the householdlanguageX X

Federal Bureau of Investigation Uniform Crime Report MeasuresTotal crime rate X X X XProperty crime rate X XViolent crime rate X XSource: Federal Bureau of Investigation 2016 Uniform Crime Report data, State of Illinois 2016 crime data, U.S.Census Bureau internal data from 2022 National Crime Victimization Survey, 2020 Census, and 2016-2020American Community Survey (ACS).~ The NCVS household income is used only when reported by respondents. When missing, it is imputed with theACS median household income for the census block group.25

DRB Clearance Number CBDRB-FY23-POP001-0082

Appendix C – GVF Models and EstimatesThe tables below list the property crime and violent crime estimates removed from processingif they have absolute studentized residuals above 3.0 and absolute measure of influence(DFFITS) values above 0.22.Table 17: 2022 National Crime Victimization Survey Outliers Removed from Generalized Variance FunctionsIteration

Removed Domain Crime Type Estimate 𝝈𝝈� Studentized

Residual DFFITS*

Overall Person Crime Estimates

2 P-All Persons Purse snatchingcompleted 31,000 36,000 4.63 4.29

Domain Person Crime Estimates1 P-In MSA, outsideprincipal cityPurse snatchingattempted 1,000 1,000 -3.73 -1.131 P-Household incomeless than $25,000Purse snatchingattempted 1,000 1,000 -3.73 -1.13

1 P-Hispanic Purse snatchingattempted 1,000 1,000 -3.73 -1.132 P-Not reported topolicePurse snatchingcompleted 27,000 36,000 5.14 0.35

2 P-Outside MSA Robbery completed 40,000 35,000 5.54 0.33

2 P-Outside MSA Robbery completed noinjury 35,000 35,000 6.64 0.422 P-In MSA, outsideprincipal cityPurse snatchingcompleted 27,000 36,000 5.14 0.352 P-Household incomeless than $25,000Purse snatchingcompleted 27,000 36,000 5.14 0.352P-Divorced/separated/never marriedPurse snatchingcompleted 31,000 36,000 5.87 0.38

2 P-Non-Hispanic Purse snatchingcompleted 29,000 36,000 5.53 0.362 P-Not black-only Purse snatchingcompleted 31,000 36,000 5.87 0.38

2 P-Age 12-24 Purse snatchingcompleted 27,000 36,000 5.14 0.35

2 P-Female Purse snatchingcompleted 29,000 36,000 5.49 0.364 P-In MSA, outsideprincipal cityPurse snatching/pocketpicking 91,000 43,000 4.81 0.244 P-Household income$100,000 and greaterRobbery completedwith injury 35,000 23,000 3.67 0.234 P-Household income$50,000-$99,999Robbery completed noinjury 73,000 37,000 4.23 0.225 P-Age 50 and over Robbery completedwith injury 87,000 40,000 4.36 0.22

Overall Property Crime Estimates

2 H-All Households Forcible Entry 556,000 68,000 3.28 2.3426

DRB Clearance Number CBDRB-FY23-POP001-0082Iteration

Removed Domain Crime Type Estimate 𝝈𝝈� Studentized

Residual DFFITS*

Domain Property Crime Estimates2 H-Tenure - rent/livefreeForcible entry 356,000 65,000 3.83 0.242 H-1 person household Forcible entry 324,000 64,000 4.36 0.28

2 H-Black-only Forcible entry 155,000 51,000 6.96 0.58

2 H-Age 12-34 Forcible entry 167,000 51,000 6.24 0.52 H-In principal city ofMSAForcible entry 291,000 60,000 4.18 0.272 H-Household income$25,000-$49,999Forcible entry 223,000 58,000 5.78 0.42

3 H-Non-Hispanic Forcible entry 479,000 67,000 3.96 0.24

3 H-Black-only Burglary/Trespassing 380,000 59,000 3.83 0.24

3 H-Black-only Burglary/Trespassingcompleted303,000 57,000 5.08 0.343 H-Reported to police Forcible entry 374,000 61,000 4.56 0.293 H-Household income$25,000-$49,999Burglary/Trespassingcompleted499,000 70,000 4.2 0.254 H-1 person household Burglary/Trespassingcompleted818,000 86,000 4.21 0.25

4 H-Age 35-49 Forcible entry 147,000 33,000 3.66 0.3

4 H-Age 12-34 Burglary/Trespassingcompleted441,000 62,000 4.41 0.274 H-Household income$25,000-$49,999Burglary/Trespassing 548,000 71,000 4.49 0.275 H-1 person household Burglary/Trespassing 972,000 88,000 3.79 0.23

5 H-Age 12-34 Burglary/Trespassing 548,000 66,000 4.24 0.25Source: U.S. Census Bureau internal data from 2022 National Crime Victimization Survey interviews.

• DFFITS - The DFFITS statistic is a scaled measure of the change in the predicted value for the observation when it

is excluded from the regression. A large DFFITS value indicates an influential observation.The figures below present plots of the direct NCVS variance estimates compared to GVFvariance estimates for overall and domain estimates of household and person victimizations.27

DRB Clearance Number CBDRB-FY23-POP001-0082Figure 3: 2022 National Crime Victimization Survey Variance Estimates – Overall Property CrimesSource: U.S. Census Bureau data from the 2022 National Crime Victimization Survey public use file.0501001502002503000 5 10 15

Standard ErrorThousandsEstimateMillions

Direct Estimate Direct Estimate Outliers Generalized Variance Function28

DRB Clearance Number CBDRB-FY23-POP001-0082Figure 4: 2022 National Crime Victimization Survey Variance Estimates – Domain Property CrimesSource: U.S. Census Bureau data from the 2022 National Crime Victimization Survey public use file.0501001502002503000 2 4 6 8 10 12 14

Standard ErrorThousandsEstimateMillions

Direct Estimate Direct Estimate Outliers Generalized Variance Function29

DRB Clearance Number CBDRB-FY23-POP001-0082Figure 5: 2022 National Crime Victimization Survey Variance Estimates – Overall Person CrimesSource: U.S. Census Bureau data from the 2022 National Crime Victimization Survey public use file.0501001502002503000 1 2 3 4 5 6 7 8

Standard ErrorThousandsEstimateMillions

Direct Estimate Direct Estimate Outliers Generalized Variance Function30

DRB Clearance Number CBDRB-FY23-POP001-0082Figure 6: 2022 National Crime Victimization Survey Variance Estimates – Domain Person CrimesSource: U.S. Census Bureau data from the 2022 National Crime Victimization Survey public use file.0501001502002503003500 1 2 3 4 5 6 7

Standard ErrorThousandsEstimateMillions

Direct Estimate Direct Estimate Outliers Generalized Variance Function1

NCVS-500 Control Card (2010 Sample Design)KEY

 Blue Diamond and text = Field Representative Instructions

[ ] Red Brackets and text = Instrument Instructions/Skip Instructions[Green Text] = Automatically filled by instrument as appropriateBold Text = Question text is read aloud to the respondent

Gray Text = Optional for Field Representative to read out loudNOTES“Empty” means never previously answered“Verify” means the Field Representative sees the answer prefilled from the previous enumeration period

Instrument Screen NameSTART_CP

[Appears during ALL Enumeration Periods]

CENSUS CATI/CAPI SYSTEM

National Crime Victimization SurveyCase Status is: New CaseDate is: [Today’s Date Appears]Time is: [Current Time Appears]Incoming/Continuing Case: [Will specify whether this is an Incoming Case or Continuing Case]Interview Number: [Will specify which interview this is from 01 to 07]Confirmed Refusal: [Will Specify Yes or No]

1. Telephone interview
2. Personal interview (See notes)
3. Personal interview (SKIP all notes and go to GEN_INTRO_CP)
4. Nointerview (TYPE A/B/C)
5. Ready to transmit case – no followup (TYPE Z’s)
6. Quit: Do not attempt now

SHOW_CP_ROSTER[Verify 2nd, 3rd, 4th, 5th, 6th, and 7th Enumeration Periods] Status of household composition

LN NAME REL MEMBER SEX AGE STATUS PHONE #

1. Enter 1 to continue

2SHOW_INFO_CP[Appears during 2nd, 3rd, 4th, 5th, 6th, and 7th Enumeration Periods]

 INFORMATION FROM PREVIOUS SCREEN INTERVIEWPress Control-F7 to view case level notesPress Control-T to view interview time preference

1. Enter 1 to Continue

GEN_INTRO_CP

[Appears during ALL Enumeration Periods] Identify yourself and show I.D.Hello. I’m (YOUR NAME)… from the U.S. Census Bureau. Here is my identification. If new household – give introductory letter and allow time to read

[Appears during 1st Enumeration Period] ASK TO SPEAK TO: Eligible respondent (a household member at least 18 years of ageand knowledgeable about the household)[Appears during 2nd, 3rd, 4th, 5th, 6th, and 7th Enumeration Periods] ASK TO SPEAK TO: [Household Respondent]

[Appears during ALL Enumeration Periods] Household address is:[Address Displayed]

1. Respondent available
2. Respondent not available (SKIP to VERIFY [F10])
3. Unavailable through closeout

[Pop Up Screen:  Are you sure this person is unavailable through closeout? If so, youmust select a new household respondent.]CAPI_INTRO_B Is respondent ready to complete the interview?

1. Continue

(SKIP to GETLETTER_CP during 1st Enumeration Period)(SKIP to HHNUM_VR_CP during 2nd, 3rd, 4th, 5th, 6th, and 7th Enumeration Periods)

2. Inconvenient time. Callback needed.

(SKIP to VERIFY [F10])

3. Reluctant respondent – hold for refusal follow-up

(SKIP to VERIFY [F10] during 1st Enumeration Period)(SKIP to HHNUM_VR_CP during 2nd, 3rd, 4th, 5th, 6th, and 7th Enumeration Periods)

4. Other outcome-problem/select new household respondent

(SKIP to HHNUM_VR_CP during 2nd, 3rd, 4th, 5th, 6th, and 7th Enumeration Periods)

5. Wrong address (wrong case selected) (SKIP to VERIFY [F10])

VERIFY[This screen appears only when GEN_INTRO_CP = 2 and when CAPI_INTRO_B = 2, 3, or 4] Review information listed below Do you need to make any changes? Press “Control T” to review/update interview time preference.TELEPHONEOK: YesTelephone1: () – Extn: Type:Telephone2: () – Extn: Type:Telephone3: () – Extn: Type:3HHNUM_VR_CP[Appears during 2nd, 3rd, 4th, 5th, 6th, and 7th Enumeration Periods] Is this a replacement household? If unsure, enter 2 (No)

1. Yes (SKIP to CK_REPLACE_CP)
2. No (SKIP to GETLETTER_CP)

CK_REPLACE_CP[Appears during 2nd, 3rd, 4th, 5th, 6th, and 7th Enumeration Periods] A replacement household means that there are NO members of the householdinterviewed during the previous enumeration period living at this address. Are you sure this is a replacement household?

1. Yes

[Pop Up Screen: This is the last screen before the roster and all incoming data isdeleted and must be re-entered. You are about to start a new case and this actioncannot be undone without the case being restarted. If “Yes”, click SUPPRESS]

2. No

GETLETTER_CP IF NECESSARY: Hello I’m (your name) from the U.S. Census Bureau.I’m here concerning the National Crime Victimization Survey. The Census Bureau isconducting a survey here and throughout the Nation to determine how often peopleare victims of crime.

[Read during 1st Enumeration Period]Did you receive our introductory letter in the mail?[Read during 2nd, 3rd, 4th, 5th, 6th, and 7th Enumeration Periods]We contacted your household for this survey several months ago. Did you receive ourintroductory letter in the mail? If “No” or “Don’t know” give respondent an introductory letter and allow time toread.

1. Yes
2. No
3. Don’t know

VERADD_CPI have your address listed as … Read below

[Address Appears]Is that your exact address?

1. SAME address

(SKIP to MAILINGSAME_CP during 1st Enumeration Period)(SKIP to CHNGPH_CP during 2nd, 3rd, 4th, 5th, 6th, and 7th Enumeration Periods)

2. MOVED (NOT same address)
3. Haven’t moved, but address has changed (SKIP to CHNGPH_CP)
4. Incorrect address previously recorded (SKIP to CHNGPH_CP)

4MOVED_CP[This screen appears only when VERADD_CP = 2 during 2nd, 3rd, 4th, 5th, 6th, and 7th Enumeration Periods]Since your address rather than you personally was chosen for inclusion in the survey, no interview isrequired of you at this time. Thank you for your past cooperation. The help you gave us was animportant contribution to the National Crime Victimization Survey data.

1. Enter 1 to continue

[Pop Up Screen: This case needs to be made a replacement household. Press GOTO tocontinue.]CHNGPH_CP Do not askDo you need to change the current phone number?CURRENT NUMBER: () – EXT:

1. Yes (SKIP to NEWPH_CP)
2. No (SKIP to NEWADDHNO_CP during 1st Enumeration Period)

(SKIP to MAILINGSAME_CP during 2nd, 3rd, 4th, 5th, 6th, and 7th Enumeration Periods)

Update Physical Address Information Screen BeginsNEWPH_CP through NEWADDGQDESCRIPTION_CP appear on the same screen.NEWPH_CPWhat is the area code and telephone number where you would like to be called? Record new number Enter 0 for no telephone numberNEWADDHNO_CP

 OLD ADDRESS:

[Address Appears] If incorrect or missing, enter new house number; otherwise press “Enter” tocontinueNEWADDHNOSUF_CP

 OLD ADDRESS:

[Address Appears] If incorrect or missing, enter new house suffix; otherwise press “Enter” to continueNEWADDSTRNAME_CP

 OLD ADDRESS:

[Address Appears] If incorrect or missing, enter new street name; otherwise press “Enter” to continueNEWADDUNITES_CP

 OLD ADDRESS:

[Address Appears] If incorrect or missing, enter new unit designation; otherwise press “Enter” tocontinueNEWADDNONCITYSTYLEADDRESS_CP

 OLD ADDRESS:

[Address Appears] If incorrect or missing, enter non-city style address; otherwise press “Enter” tocontinue5NEWADDPHYSDESCRIPTION_CP

 OLD ADDRESS:

[Address Appears] If incorrect or missing, enter new physical location description; otherwise press“Enter” to continueNEWADDCITY_CP

 OLD ADDRESS:

[Address Appears] If incorrect or missing, enter new city name; otherwise press “Enter” to continueNEWADDSTATE_CP

 OLD ADDRESS:

[Address Appears] If incorrect or missing, enter new state abbreviation; otherwise press “Enter” tocontinueNEWADDZIP5_CP

 OLD ADDRESS:

[Address Appears] If incorrect or missing, enter 5-digit ZIP code; otherwise press “Enter” to continueNEWADDZIP4_CP

 OLD ADDRESS:

[Address Appears] If incorrect or missing, enter new 4-digit ZIP code; otherwise press “Enter” tocontinueNEWADDBUILDINGNAME_CP

 OLD ADDRESS:

[Address Appears] If incorrect or missing, enter new building name; otherwise press “Enter” tocontinueNEWADDGQDESCRIPTION_CP

 OLD ADDRESS:

[Address Appears] If incorrect or missing, enter new group description; otherwise press “Enter” tocontinue

Update Physical Address Information Screen EndsADRCHECK_CP Enter reason why address information provided by respondent did not matchdisplay address

OLDADDRESS NEWADDRESS

[Address Displayed] [Address Displayed]

 HOUSE NUMBER NON-CITY STYLE ADDRESS

11. House number was incorrect
12. House number was missing/blank

 HOUSE NUMBER SUFFIX

13. House number suffix was incorrect (e.g. A instead of B)
14. House number suffix was missing

 STREET NAME6

15. Street name was correct but misspelled (e.g. Pak instead ok Oak Street)
16. Street name was not correct (e.g. Oak instead of Pickford)

 UNIT DESIGNATION

17. Unit designation was incorrect (e.g. A instead of 1)
18. Unit design was missing
19. Non city style address was incorrect (P.O. Box 12 instead of P.O. Box 121)

 ZIP CODE

20. ZIP code was missing or incorrect

 STATE

21. State was missing or incorrect (ME instead of MD)

 CITY

22. City name was missing or incorrect

 GROUP QUARTERS

23. Group quarters name was missing or incorrect
24. Building name was missing or incorrect
25. 911 Address Conversion

MAILINGSAME_CPIs your mailing address still the same as your physical address?

1. Yes (SKIP to TENURE during 1st , 3rd, 5th, and 7th Enumeration Periods)

(SKIP to STUDENTHOUSING during 2nd, 4th, and 6th Enumeration Periods)

2. No (SKIP to VERIFYMAILING_CP)

VERIFYMAILING_CP Read mailing addressI have your mailing address as…

MAILING ADDRESS:

[Address Displayed]Is that correct?

1. Yes (SKIP to TENURE)
2. No (SKIP to NEWMAILHNO_CP)

Update Mailing Address Information Screen BeginsNEWMAILHNO_CP through NEWMAILGQDESCRIPTION_CP appear on the same screen.NEWMAILHNO_CP

 OLD MAILING ADDRESS:

[Previous Address Displayed] If incorrect or missing, enter new house number; otherwise press “Enter” tocontinueNEWMAILHNOSUF_CP

 OLD ADDRESS:

[Address Appears] If incorrect or missing, enter new house number suffix; otherwise press “Enter” tocontinueNEWMAILSTRNAME_CP

 OLD ADDRESS:

[Address Appears] If incorrect or missing, enter new street name; otherwise press “Enter” to continue7NEWMAILUNITES_CP

 OLD ADDRESS:

[Address Appears] If incorrect or missing, enter new unit designation; otherwise press “Enter” tocontinueNEWMAILNONCITYSTYLEADDRESS_CP

 OLD ADDRESS:

[Address Appears] If incorrect or missing, enter non-city style address; otherwise press “Enter” tocontinueNEWMAILCITY_CP

 OLD ADDRESS:

[Address Appears] If incorrect or missing, enter new city name; otherwise press “Enter” to continueNEWMAILSTATE_CP

 OLD ADDRESS:

[Address Appears] If incorrect or missing, enter new state abbreviation; otherwise press “Enter” tocontinueNEWMAILZIP5_CP

 OLD ADDRESS:

[Address Appears] If incorrect or missing, enter 5-digit ZIP code; otherwise press “Enter” to continueNEWMAILZIP4_CP

 OLD ADDRESS:

[Address Appears] If incorrect or missing, enter new 4-digit ZIP code; otherwise press “Enter” tocontinueNEWMAILGQDESCRIPTION_CP

 OLD ADDRESS:

[Address Appears]If incorrect or missing, enter new group description; otherwise press “Enter” tocontinue

Update Mailing Address Information Screen EndsTENURE[Appears during 1st Enumeration Period or when empty. Verify during 3rd, 5th and 7th Enumeration Periods]Ask or verifyAre your living quarters... Read answer categories.

1. Owned or being bought by you or someone in your household?
2. Rented for cash?
3. Occupied without payment of cash rent?

8STUDENTHOUSING

[Verify each Enumeration Period] If apparent, enter precode without asking.Are your living quarters presently used as student housing by a college or university?

1. Yes
2. No

PUBLICHOUSING[Appears during 1st Enumeration Period or when empty][If box 1 is marked in item TENURE, SKIP to INDIANRESERVATIONHU][If box 2 or box 3 is marked in TENURE, ask below]Is this building owned by a public housing authority?

1. Yes, public housing – (SKIP to PUBLICHOUSINGMGRVERIFY)
2. No, not public housing – (SKIP to INDIANRESERVATIONHU)

PUBLICHOUSINGMGRVERIFY

[Appears when PUBLICHOUSING = 1] If possible, verify PUBLICHOUSING entry of 1 (Yes) with the manager of building. Able to verify

1. Public housing
2. Not public housing

 Unable to verify

3. Telephone
4. Other – Specify

PUBLICHOUSINGMGRVERIFYSPEC

[Appears when PUBLICHOUSINGMGRVERIFY = 4] Specify the reason why you are unable to verify the public housing status for thishousing unit.INDIANRESERVATIONHU[Appears during 1st Enumeration Period, or when empty] If apparent, enter precode without asking.Are your living quarters located on an American Indian Reservation or on AmericanIndian Lands?

1. Yes
2. No

FARMSALES[Appears during 1st Enumeration Period, or when empty, and address is in rural area]During the past 12 months did sales of crops, livestock, and other farm products fromthis place amount to $1,000 or more?

1. Yes
2. No

ACCESS[Appears during 1st Enumeration Period, or when empty] Please mark whether or not the sample household has direct access to their livingquarters.

1. Direct
2. Through another unit - Not a separate HU; combine with unit through which access is

Gained9TYPEOFHOUSINGUNIT[Appears during 1st Enumeration Period, or when empty] Please select one box that describes the type of housing unit.

1. House, apartment, flat
2. HU in nontransient hotel, motel, etc.
3. HU permanent in transient hotel, motel, etc.
4. HU in rooming house
5. Mobile home or trailer with no permanent room added
6. Mobile home or trailer with one or more permanent rooms attached
7. HU not specified above - Describe
8. Quarters not HU in rooming or boarding house
9. Unit not permanent in transient hotel, motel, etc.
10. Unoccupied site for mobile home, trailer, or tent
11. Student quarters in college dormitory
12. Other unit not specified above – Describe

TYPEOFHOUSINGUNITSPEC7

[Appears when TYPEOFHOUSINGUNIT = 7] Please specify the other type of "Housing Unit".TYPEOFHOUSINGUNITSPEC12

[Appears when TYPEOFHOUSINGUNIT = 12] Please specify the other type of "Other Unit".NUMBEROFUNITS[Appears during 1st Enumeration Period, or when empty]

 Observe or askHow many housing units are in this structure?

1. 1
2. 2
3. 3
4. 4
5. 5-9
6. 10+
7. Mobile home/trailer
8. Only OTHER units

DIRECTENTRANCETOUNIT[Appears during 1st Enumeration Period, or when empty]

 Observe or askDoes the unit have an outside entrance, patio doors, or windows, etc., on the groundlevel - or outside stairs leading directly to this unit?

1. Yes
2. No
3. Don't know

GATEDWALLEDCOMMUNITY[Appears during 1st Enumeration Period, or when empty] Ask if unsureIs this unit in a gated or walled community that restricts access by non-residents orrequires entry codes, key cards, or security guard approval to access?

1. Yes
2. No

10RESTRICTEDACCESS[Appears during 1st Enumeration Period, or when empty] Ask if unsureIs this unit in a building that requires a special entry system such as entry codes, keycards, or security guard approval to access?

1. Yes
2. No

Begin Initial Roster LoopAppears when a household roster is being first established.(Repeat for each household member)HHROSTER_FNAME[Appears during 1st Enumeration Period, or when empty]What are the names of all people living or staying here?Start with the person or one of the people who [owns/rents] this home. Enter first name on this screen. To change a non-member already listed on this roster to a member use theup/down arrow to go to MEMBERCHANGES and enter the reason why this person isnow a member. Enter 999 to leave the table.HHROSTER_LNAME[Appears during 1st Enumeration Period, or when empty]What are the names of all people living or staying here?Start with the name of the person or one of the people who [owns/rents] this home. Enter last name on this screen.SEX[Appears during 1st Enumeration Period, or when empty] Ask if necessaryIs [household member name] male or female?

1. Male
2. Female

RELATIONSHIP[Appears during 1st Enumeration Period, or when empty] Review the relationship of other household members to the new reference personand update these relationships if necessary.What is [household member name]’s relationship to [reference person]?

11. Husband
12. Wife
13. Son
14. Daughter
15. Father
16. Mother
17. Brother
18. Sister
19. Other Relative
20. Non Relative

11HHMEMBER[Appears during 1st Enumeration Period, or when empty]Does [household member name] usually live here? If "No", probe for usual residence elsewhere.

1. Yes
2. No

HSEMEMURE[Appears during 1st Enumeration Period, or when empty]Does [household member name] have a usual place of residence elsewhere?

1. Yes
2. No

End Initial Roster Loop

Begin Verify Roster SectionAppears if a roster has been established in a previous enumeration.NAMECHECK[Verify during 2nd, 3rd, 4th, 5th, 6th, and 7th Enumeration Periods]I have [Read names below] listed as living or staying at this address.[The name and information of each household member from the previous enumeration period will appear in a tablebelow.]

LINE NUMBER NAME RELATIONSHIP AGE SEX MARITAL STATUS STATUSAre ALL of these people still living or staying at this address?

1. Yes (SKIP to MEMBERCHANGES if missing demographic data, else SKIP to HHLDCOVERAGE)
2. No (SKIP to REFPERSTILLLIVE)

REFPERSTILLLIVE

[Appears when NAMECHECK = 2]Does [reference person name] still live at this address?

1. Yes
2. No (SKIP to NEWREFPER)

NEWREFPER

[Appears when REFPERSTILLLIVE = 2]What is the name of the person (or one of the persons) who owns or rents that home?Would that be you? Enter line number, or 31 if someone not listed[The name and information of each household member from the previous enumeration period will appear in a tablebelow.]

LINE NUMBER NAME RELATIONSHIP SEX AGE MARITAL STATUS STATUS01.-30. Line number 01 – Line number 30 [Names on line number 1 up to line number 30 aredisplayed.]

31. Someone not listed above

Begin Verify Roster Loop(Repeat for each household member)MEMBERCHANGES[Appears during 2nd, 3rd, 4th, 5th, 6th, and 7th Enumeration Periods when new household member enters or leaves] Enter reason why there is a change in household membership for this person. If no change is needed for this person, press the ENTER key without selecting aprecode.12 Use the arrow keys to move through the table and REVIEW/UPDATE demographics.When done, press Page Down.

WHY ENTERED HOUSEHOLD:

11. Returned from school or college
12. Returned from institution
13. Entered because of marriage/separation/divorce
14. Person entered household for reasons other than above

WHY LEFT HOUSEHOLD:

15. Person died
16. Left for school or college
17. Entered institution
18. Left because of marriage/separation/divorce
19. Person left household for reasons other than above
20. Visitor - residence elsewhere

HHROSTER_FNAME[Appears during 2nd, 3rd, 4th, 5th, 6th, and 7th Enumeration Periods when new household member is added to roster]What is the name of the person/people that is/are new to the household? Enter first name on this screen. To change a non-member already listed on this roster to a member use theup/down arrow to go to MEMBERCHANGES and enter the reason why this person isnow a member. Enter 999 to leave the table.HHROSTER_LNAME[Appears during 2nd, 3rd, 4th, 5th, 6th, and 7th Enumeration Periods when new household member is added to roster]What is the name of the person/people that is/are new to the household? Enter last name on this screen.SEX[Appears during 2nd, 3rd, 4th, 5th, 6th, and 7th Enumeration Periods when new household member is added to roster] Ask if necessaryIs [household member name] male or female?

1. Male
2. Female

RELATIONSHIP[Appears during 2nd, 3rd, 4th, 5th, 6th, and 7th Enumeration Periods when new household member is added to roster] Review the relationship of other household members to the new reference personand update these relationships if necessary.What is [household member name]’s relationship to [reference person]?

11. Husband
12. Wife
13. Son
14. Daughter
15. Father
16. Mother
17. Brother
18. Sister
19. Other relative
20. Nonrelative

13HHMEMBER[Appears during 2nd, 3rd, 4th, 5th, 6th, and 7th Enumeration Periods when new household member is added to roster]Does [household member name] usually live here? If "No", probe for usual residence elsewhere.

1. Yes (SKIP to HHLDCOVERAGE)
2. No (SKIP to HSEMEMURE)

HSEMEMURE[Appears during 2nd, 3rd, 4th, 5th, 6th, and 7th Enumeration Periods when new household member is added to roster]Does [household member name] have a usual place of residence elsewhere?

1. Yes
2. No

End Verify Roster Loop

End Verify Roster SectionHHLDCOVERAGE

[Verify each Enumeration Period]Have I missed anyone else living or staying here/there such as any babies, any lodgers,or anyone who is away at present traveling or in the hospital?[The name of each household member given will appear in a table below.]

LINE NUMBER NAME

1. Yes
2. No

ENTIREHHURE[Appears when all people on the household roster have a usual residence elsewhere] All people on the household roster have a usual residence elsewhere. Enter 1 to exit this case. Then, reenter this case to code it a Noninterview (Type B - Entire Household URE)via START_CP.

 Enter 1 to Continue[Appears when all household members no longer live at residence .]PICK1STHHRESP[Appears during 1st Enumeration Period, or when household respondent is changed] Enter the line number for the person with whom you are speaking. This person should be at least 18 years of age and knowledgeable about thehousehold.[The name of each household member given will appear in a table below.]

LINE NUMBER NAME

Begin Additional Demographic LoopAppears when a household is being first established AND when a household roster has been previouslyestablished.(Repeat for each household member)AGECHECK[Verify during 2nd, 3rd, 4th, 5th, 6th, and 7th Enumeration Periods]I have [you/household member name] listed as [calculated age] years old.Is that correct?

1. Yes, age IS correct
2. No, age is NOT correct

14BRTHDATEMO

[Ask or Verify each Enumeration Period]What is [your/household member name] date of birth? Enter month on this screen.

1. January
2. February
3. March
4. April
5. May
6. June
7. July
8. August
9. September
10. October
11. November
12. December

BRTHDATEDY

[Ask or Verify each Enumeration Period]What is [your/household member name] date of birth? Enter day on this screenBRTHDATEYR

[Ask or Verify each Enumeration Period]What is [your/household member name] date of birth? Enter year on this screen If year is less than 1890, enter 1890VFYAGE

[Ask or Verify each Enumeration Period]That would make [you/household member name] [calculated age] years old.Is that correct?

1. Yes
2. No

ESTAGE[Appears when BRTHDATEYR is Don’t Know][Appears for household respondent]Even though you don't know your exact birthdate, what is your approximate age?[Appears for household member]Even though you don't know [household member name]’s exact birthdate, what isyour best guess as to how old he/she was on his/her last birthday?AGERNG[Appears when BRTHDATEYR is Don’t Know, and ESTAGE is Don’t Know or Refused] OR[Appears when BRTHDATEYR is Refused][Appears for household respondent]Are you between 18 and 65 years old? 66 or older?Are you…[Appears for household member]Is [he/she] a child, a teenager, or an adult?Is [he/she]…15 Read appropriate age categories.

1. 0 - 11 years old?
2. 12 - 13 years old?
3. 14 - 15 years old?
4. 16 - 17 years old?
5. 18 - 24 years old?
6. 25 - 34 years old?
7. 35 - 49 years old?
8. 50 - 65 years old?
9. 66 years old or older?

MARITAL

[Ask or Verify each Enumeration Period] Enter new marital status for [household member name]LAST REPORTED AS: MARITAL If in doubt, ask:[Are/Is] [you/household member name] now married, widowed, divorced, separatedor [have/has] [you/he/she] never been married?

1. Married
2. Widowed
3. Divorced
4. Separated
5. Never married

ARMEDFORCES[Ask or Verify each Enumeration Period for persons 18-65 years old]LAST REPORTED AS: MILITARY[Are/Is] [you/household member name] now in the Armed Forces?

1. Yes
2. No

EDUCATIONATTAIN[Ask or Verify during 1st, 3rd, 5th, and 7th Enumeration Periods for persons 12 years of age and older]

<Flashcard Booklet> (Page 6)LAST REPORTED AS: EDUCATIONATTAINWhat is the highest level of school [you/household member name] completed or thehighest degree [you/he/she] received?

1. 1st grade
2. 2nd grade
3. 3rd grade
4. 4th grade
5. 5th grade
6. 6th grade
7. 7th grade
8. 8th grade
9. 9th grade
10. 10th grade
11. 11th grade
12. 12th grade (No diploma)
13. High school graduate (Diploma or the equivalent)
14. Some college (No Degree)
15. Associate's degree

16

16. Bachelor's degree (e.g. BA, AB, BS)
17. Master's degree (e.g. MA, MS, MEng, MSW, MBA)
18. Professional School degree (e.g. MD, DDS, DVM, LLB, JD)
19. Doctorate degree (PhD, EdD)
20. Never attended, preschool, kindergarten

ATTENDINGSCHOOL[Ask or Verify each Enumeration Period for persons 12 years of age and older]LAST REPORTED AS: ATTENDINGSCHOOL[Are/Is] [you/household member name] currently attending or enrolled either full-time or part-time in a college or university, trade, or vocational school?

1. Regular school
2. College/University
3. Trade school
4. Vocational school
5. None of the above schools

SP_ORIGIN[Appears during 1st Enumeration Period or when new household member is added]

<Flashcard Booklet> (Page 8)[Are/Is] [you/household member name] Spanish, Hispanic, or Latino?

1. Yes
2. No

RACE[Appears during 1st Enumeration Period or when new household member is added]

<Flashcard Booklet> (Page 10) If personal interview show flash card. If telephone interview read answer categories. Do not probe.Please choose one or more races that [you/household member name]consider/considers [yourself/himself/herself] to be.

1. White
2. Black or African American
3. American Indian or Alaska Native
4. Asian
5. Native Hawaiian or Other Pacific Islander
6. Other – Specify

RACE_SPECIFY Specify the other race for this person.

End Additional Demographic LoopOMB No. 1121-0111: Approval Expires 11/30/2024Implementation Date: (1/1/2022)

NATIONAL CRIME VICTIMIZATION SURVEY

NCVS-1 BASIC SCREEN QUESTIONNAIRENCVS-1

____________ Months (1-11) - SKIP to 33e505____________ Years (Round to nearest whole year)5065 years or more - If HHLD Respondent SKIP to 34, else SKIP to 36aLess than 5 years - ASK 33eCHECKITEM A How many years are entered in 33a?Altogether, how many times have you moved in the last5 years, that is, since _________________, 20 ___?Enter number of times.__________ Number of times -508

MOBILITY QUESTIONSDoes anyone in this household operate a business from this address?If this is a PERSONAL visit - Fill by observation.If this is a TELEPHONE contact - Ask.Is there a sign on the premises or some other indication to thegeneral public that a business is operated from this address?

35. BUSINESSSIGN (Asked of Household Respondent Only)

12Yes - ASK 35

No - SKIP to 36a53012Yes (Recognizable business)No (Unrecognizable business)53133a. TIMEATADDRESSHow many months?If = 0 ASK 33b

If = DK or RF SKIP to 33c

Else SKIP to 33d33c. TIMEATADDRESSPROBEHave you lived here...Read categories 1-4:1234More than 5 years - If HHLD Respondent ASK 34, else SKIP to 36a.Less than 5 years, but more than 1 yearLess than 1 year, but more than 6 months6 months or less

Don't Know5SKIPto 33e33d.33b. MONTHSATADDRESS33e. TIMEMOVEDIN5YEARS If HHLD Respondent ASK 34,

Else SKIP to 36a

34. BUSINESS (Asked of Household Respondent Only)

BUSINESS OPERATED FROM SAMPLENOTE: Questions are listed in the order asked. Skips in question numbering are due to questionnaire changes over time.Before we get to the crime questions, I have some questions that arehelpful in studying where and why crimes occur.Ask or verify -How long have you lived at this address?(Enter 0 if less than 1 year.)Page 136a. SQTHEFTPage 2Did any incidents of this type happen to you?How many times?37a. SQBREAKIN (Asked of Household Respondent Only)Did any incidents of this type happen to you?Ask only if necessary37b. SQBREAKINTIMES (Asked of Household Respondent Only)Briefly describe incident(s)1Yes - ASK 37b532

If Household Respondent ASK 37a; else SKIP to 40a533

No - SKIP to 382Briefly describe incident(s)1534Number of times (37b)5352Yes - ASK 36b

No - If Household Respondent ASK to

37a; Else SKIP to 40a36c. SQTHEFTSPEC37c. SQBREAKINSPEC (Asked of Household Respondent Only)What happened?What happened?Number of times (36b)I'm going to read some examples that will give you an idea of thekinds of crimes this study covers.As I go through them, tell me if any of these happened to you in thelast 6 months, that is since __________ ______, 20 ____.Was something belonging to YOU stolen, such as -Read each category.(a) Things that you carry, like luggage, a wallet, purse, briefcasebook -(b) Clothing, jewelry, or cellphone -(c) Bicycle or sports equipment -(d) Things in your home - like a TV, stereo, or tools -(e) Things outside your home such as a garden hose or lawn

furniture - (Asked of Household Respondent Only)(f) Things belonging to children in the household -

(Asked of Household Respondent Only)(g) Things from a vehicle, such as a package, groceries, camera, orCDs -OR(h) Did anyone ATTEMPT to steal anything belonging to you?(Other than any incidents already mentioned,) has anyone -Read each category.(a) Broken in or ATTEMPTED to break into your home by forcing adoor or window, pushing past someone, jimmying a lock, cuttinga screen, or entering through an open door or window?(b) Has anyone illegally gotten in or tried to get into a garage, shed,or storage room?OR(c) Illegally gotten in or tried to get into a hotel or motel room orvacation home where you were staying?How many times?36b. SQTHEFTTIMES

RESPONDENT'S SCREEN QUESTIONSAsk only if necessary39a. SQMVTHEFT (Asked of Household Respondent Only)

38. SQTOTALVEHICLES (Asked of Household Respondent Only)

What was the TOTAL number of cars, vans, trucks, motorcycles, orother motor vehicles owned by you or any other member of thishousehold during the last 6 months? Include those you no longerown.Page 3Did any incidents of this type happen to you?Ask only if necessary39b. SQMVTHEFTTIMES (Asked of Household Respondent Only)40a. SQATTACKWHEREDid any incidents of this type happen to you?Ask only if necessary40b. SQATTACKWHERETIMES01234

None - SKIP to 40a1234 or more536Briefly describe incident(s)1 Yes - ASK 39b537Number of times (39b)538

No - SKIP to 40a2Number of times (40b)540Briefly describe incident(s)39c. SQMVTHEFTSPEC (Asked of Household Respondent Only)

No - SKIP to 41a2Yes - ASK 40b153940c. SQATTACKWHERESPECWhat happened?What happened?During the last 6 months, (other than any incidents alreadymentioned,) (was the vehicle/were any of the vehicles) -Read each category.(a) Stolen or used without permission?(b) Did anyone steal any parts such as a tire, car stereo, hubcap, orbattery?(c) Did anyone steal any gas from (it/them)?OR(d) Did anyone ATTEMPT to steal any vehicle or parts attached to(it/them)?(Other than any incidents already mentioned,) since __________ ______,20 ____, were you attacked or threatened OR did you have somethingstolen from you -Read each category.(a) At home including the porch or yard -(b) At or near a friend's, relative's, or neighbor's home -(c) At work or school -d) In places such as a storage shed or laundry room, a shoppingmall, restaurant, bank, or airport -(e) While riding in any vehicle -(f) On the street or in a parking lot -(g) At such places as a party, theater, gym, picnic area, bowlinglanes, or while fishing or hunting -OR(h) Did anyone ATTEMPT to attack or ATTEMPT to stealanything belonging to you from any of these places?How many times?How many times?

RESPONDENT'S SCREEN QUESTIONSPage 441a. SQATTACKHOWDid any incidents of this type happen to you?How many times?Briefly describe incident(s)1 Yes - ASK 41b541Number of times (41b)542

No - SKIP to 42a242a. SQTHEFTATTACKKNOWNOFFDid any incidents of this type happen to you?How many times?Briefly describe incident(s)1 Yes - ASK 42b543Number of times (42b)544

No - SKIP to 43a241c. SQATTACKHOWSPEC42c. SQTHEFTATTACKKNOWNOFFSPECWhat happened?What happened?(Other than any incidents already mentioned,) has anyone attacked orthreatened you in any of these ways -(Exclude telephone threats) -Read each category.(a) With any weapon, for instance, a gun or knife -(b) With anything like a baseball bat, frying pan, scissors, or stick -(c) By something thrown, such as a rock or bottle -(d) Include any grabbing, punching, or choking,(e) Any rape, attempted rape or other type of sexual attack -(f) Any face to face threats -OR(g) Any attack or threat or use of force by anyone at all? Pleasemention it even if you are not certain it was a crime.People often don't think of incidents committed by someone theyknow. (Other than any incidents already mentioned,) did you havesomething stolen from you OR were you attacked or threatened by -(Exclude telephone threats)Read each category.(a) Someone at work or school -(b) A neighbor or friend -(c) A relative or family member -(d) Any other person you've met or known?41b. SQATTACKHOWTIMES42b. SQTHEFTATTACKKNOWNOFFTIMES

RESPONDENT'S SCREEN QUESTIONSAsk only if necessaryAsk only if necessaryPage 5During the last 6 months, (other than any incidents alreadymentioned,) did you call the police to report something thathappened to YOU which you thought was a crime?44a. SQCALLPOLICECRIME 1 Yes - ASK 44b547

No - SKIP to 45a2How many times?CHECKITEM BIf not sure ask:Were you attacked or threatened, or was something stolen or anattempt made to steal something that belonged to you oranother household member?12Yes - ASK 44d

No - SKIP to 45a549Number of times (44d)55044b. SQCALLPOLICESPECSQCALLPOLICEATTACKTHREATWhat happened?44c.43a. SQSEXUALDid any incidents of this type happen to you?How many times?Briefly describe incident(s)1 Yes - ASK 43b545Number of times (43b)546

No - SKIP to 44a243c. SQSEXUALSPECWhat happened?Incidents involving forced or unwanted sexual acts are often difficultto talk about. (Other than any incidents already mentioned,) have youbeen forced or coerced to engage in unwanted sexual activity by -Read each category.(a) Someone you didn't know -(b) A casual acquaintance -OR(c) Someone you know well?43b. SQSEXUALTIMESBriefly describe incident(s)44d. SQCALLPOLICEATTACKTHREATTIMES

RESPONDENT'S SCREEN QUESTIONSAsk only if necessaryPage 6

No - SKIP to 71How many times?If not sure ask:Were you attacked or threatened, or was something stolen or anattempt made to steal something that belonged to you or anotherhousehold member?12Yes - ASK 45d553

• SKIP to 71

45c.During the last 6 months, (other than any incidents alreadymentioned,) did anything which you thought was a crime happento YOU, but you did NOT report to the police?45a. SQNOCALLPOLICECRIME 1 Yes - ASK 45b551245b. SQNOCALLPOLICESPECWhat happened?

No - SKIP to 71Briefly describe incident(s)45d. SQNOCALLPOLICEATTACKTHREATTIMESNumber of times (45d)554Is one or more crimes reported in36b, 37b, 39b, 40b, 41b, 42b, 43b, 44d, or 45d?Yes - SKIP to Crime Incident Report (Item 2a on the NCVS-2)

1 Telephone interview - SKIP to 73555No one besides respondent presentRespondent's spouseHHLD member(s) 12+, not spouseHHLD member(s) under 12Nonhousehold member(s)Someone was present - Can't say whoDon't know if someone else presentWho besides the respondent was present when the screen questionswere asked? (If telephone interview, mark box 1 only.)

123YesNoPerson for whom interview taken not present556Did the person for whom this interview was taken help the proxyrespondent answer any screen questions?2345678

If a Proxy interview,ASK 72,else SKIP to 73

No - SKIP to 81 (ASK INTRO_SOCIO_DEMO)Personal interview - Mark all that apply.

RESPONDENT'S SCREEN QUESTIONS

RESPONDENT'S CHECK ITEMS D, E, AND GSQNOCALLPOLICEATTACKTHREATPROXYHELPPRESENTFORSQSCHECKITEM CCHECKITEM DCHECKITEM ECHECKITEM G169a. DIS_HEARINGAre you deaf or do you have serious difficulty hearing?169b. DIS_VISIONBecause of a physical, mental, or emotionalcondition, do you have serious difficulty:2 NoYes12 NoYes12 NoYes1170a.2 NoYes12 NoYes1170b. DIS_LEAVING_HOME2 NoYes1

82. CITIZENSHIP

Yes, born in the United StatesYes, born in Puerto Rico, Guam, the U.S. Virgin Islands,

or North MarianasYes, born abroad of U.S. citizen parent or parentsYes, U.S. citizen by naturalizationNo, not a U.S. citizen12345

83. ORIENTATION_MALE

GayStraight, that is, not gayBisexualSomething elseI don't know the answer123456 RefusedPage 7(Asked at 1st interview or if never asked before. Asked at subsequentinterviews if no or Don't Know/Refused at prior interview.)Are you a citizen of the United States? That is, were you born in theUnited States, born in a U.S. territory, born of U.S. citizen parent(s), orbecome a citizen of the U.S. through naturalization?(Asked of persons age 16 or older.Asked at 1st, 3rd, 5th, and 7th interview, or if never asked before.

Asked if NCVS-500 roster SEX question = Male)Which of the following best represents how you think of yourself?

RESPONDENT'S SOCIO-DEMOGRAPHIC QUESTIONS

81. INTRO_SOCIO_DEMO

All incident reports must be completed before asking this series of questions.

SKIP to 85(Asked at 1st, 3rd, 5th, and 7th interview, or if never asked before.)Now we have some questions about your demographic characteristics.(Asked at 1st, 3rd, 5th, and 7th interview, or if never asked before.)Are you blind or do you have serious difficulty seeingeven when wearing glasses?(1) DIS_LEARN_CONCENTRATE(Asked at 1st, 3rd, 5th, and 7th interview, or if never asked before.)Concentrating, remembering or makingdecisions?(2) DIS_PHYSICAL_LIMIT(Asked at 1st, 3rd, 5th, and 7th interview, or if never asked before.)Walking or climbing stairs?(3) DIS_DRESS_BATH(Asked at 1st, 3rd, 5th, and 7th interview, or if never asked before.)Dressing or bathing?(Asked at 1st, 3rd, 5th, and 7th interview, or if never asked before.Asked only of persons age 15 or older.)Because of a physical, mental, or emotional condition, do you havedifficulty doing errands alone such as visiting a doctor's office orshopping?

86. GENID_DESCRIBE Male

FemaleTransgenderNone of these1234

87. GENID_CONFIRM

YesNo - SKIP back to 85 and/or 86 to correctRefused

Don't Know1234Have you ever served on active duty in the U.S. Armed Forces,Reserves, or National Guard?Mark one box.Never served in the militaryOnly on active duty for training in the Reserves

or National GuardOn active duty in the past, but not now1234

89. ACTIVE_DUTYWHEN

When did you serve on active duty in the U.S. Armed Forces?Mark a box for each period in which the person served, even if just for part ofthe period.September 2001 or later

August 1990 to August 2001 (including Persian Gulf War)May 1975 to July 1990

Vietnam era (August 1964 to April 1975)

February 1955 to July 1964

January 1947 to June 19501234567

Korean War (July 1950 to January 1955)

World War II (December 1941 to December 1946)8November 1941 or earlier9Page 8

85. GENID_BIRTH Male

FemaleRefusedDon't know1234(Asked of persons age 16 or older.Asked at 1st, 3rd, 5th, and 7th interview, or if never asked before.)What sex were you assigned at birth, on your original birth certificate?

88. ACTIVE_DUTY

(Asked of persons age 18 or older. Asked at 1st interview or if never asked before. Asked at subsequent interviews if Never served in the militaryor Don't Know/Refused at prior interview. Do not ask if age 40 or older and valid response prior interview.)

84. ORIENTATION_FEMALE Lesbian or gay

Straight, that is, not lesbian or gayBisexualSomething elseI don't know the answer123456 Refused(Asked of persons age 16 or older.Asked at 1st, 3rd, 5th, and 7th interview, or if never asked before.

Asked if NCVS-500 roster SEX question = Female or Unknown)Which of the following best represents how you think of yourself?(Asked of persons age 16 or older.Asked at 1st, 3rd, 5th, and 7th interview, or if never asked before.)Do you currently describe yourself as male, female or transgender?(Asked of persons age 16 or older.Asked at 1st, 3rd, 5th, and 7th interview, or if never asked before.Asked if GENID_BIRTH and GENID_DESCRIBE do not match.)Just to confirm, you were assigned (male/female) at birth and now(describe yourself as male/ describe yourself as female/ describeyourself as transgender/ do not describe yourself male, female, ortransgender). Is that correct?The instrument prefills the question wording based on responses to GENID_BIRTH and GENID_DESCRIBE.(Asked of persons age 18 or older.Asked at 1st interview or if never asked before. Asked at subsequent interviews if no or Don't Know/Refused at prior interview.Do not ask if age 40 or older and valid response prior interview.)

• SKIP to 74
• ASK 89

Now on active dutyPage 91 Physician?Nurse?Technician?Other medical profession? - Specify - ASK 76dDid you have a job or work at a business LAST WEEK? (Do notinclude volunteer work or work around the house.)(If farm or business operator in household, ask about unpaid work.)75a. JOBLASTWEEK 12

Yes - SKIP to 76aNo - ASK 75bAsk or verify -Did you have a job or work at a business DURING THE LAST 6 MONTHS?75b. JOBDURINGREFPERIOD1212Yes - ASK 75c

No - SKIP to 80Yes - ASK 76a

No - SKIP to 80Did that (job/work) last 2 consecutive weeks or more?75c. JOBLAST2WEEKS76a. JOBDESCRIPTION

Medical Profession

Mental Health Services Field5 Professional (Social worker/psychiatrist)?Custodial care?Some other mental health services profession? - SpecifyAsk or verify -Which of the following best describes your job?Were you employed in the ...Read each category until respondent says "yes", then enter appropriate precode.Is the respondent 16 years or older? 12Yes - ASK 75a

No - SKIP to 80Medical Profession? - SKIP 76cMental Health Services Field? - SKIP to 76eTeaching Profession? - SKIPto 76gLaw Enforcement or Security Field? - SKIP to 76iRetail Sales? - SKIP to 76kSomething else? - Specify - ASK 76bPlease specify the job not covered in answer categories 1-6 in 76a.123456

Employed in the Medical Profession:As a -Read each category.

Employed in the Mental Health Services Field:Are YOUR duties -Read each category.Please specify employed in the medical profession as a -Please specify duties in the mental health services field.7Transportation Field? - SKIP to 76m

SKIP to 76o

SKIP to 76o

76b. JOBDESCRIPTIONSPEC76c. MEDICALJOB23476d. MEDICALJOBSPEC76e. MENTALHEALTHJOB6776f. MENTALHEALTHJOBSPEC

Specify - SKIP to 76o

Specify - SKIP to 76oASK 76f

Specify - SKIP to 76o576577578

RESPONDENT'S EMPLOYMENT QUESTIONSCHECKITEM H1Page 10

Law Enforcement or Security FieldLaw enforcement officer?Prison or jail guard?Security guard?Law enforcement profession? - Specify - ASK 76j1617181976i. LAWENFORCEJOB

Employed in the Law Enforcement or Security Field:Were you employed as a -Read each category.76j. LAWENFORCEJOBSPECPlease specify employed in the law enforcement or security field as a -

• SKIP to 76o

579

Teaching Profession8 Preschool?Elementary school?Junior high or middle school?High school?College or university?Technical or industrial school?Special education facility?

Employed in the Teaching Profession:Were you employed in a -Read each category.Please specify employed in the teaching profession as a -76h. TEACHINGJOBSPECOther teaching profession? - Specify - ASK 76h

• SKIP to 76o

76g. TEACHINGJOB9101112131415

Retail Sales -

Transportation Field -Convenience or liquor store clerk?Gas station attendant?Bartender?Other retail sales profession? - Specify - ASK 76l20212223Bus driver?Taxi cab driver?Other transportation Field profession? - Specify - ASK 76n24252676k. RETAILSALESJOB

Employed in Retail Sales:Were you employed as a-Read each category.76l. RETAILSALESJOBSPECPlease specify employed in retail sales as a -76m. TRANSPORTJOB

Employed in the Transportation Field:Were you employed as a-Read each category.76n. TRANSPORTJOBSPECPlease specify employed in the transportation field as a -

• SKIP to 76o
• SKIP to 76o

Specify - SKIP to 76o

Specify - SKIP to 76o

Specify - SKIP to 76oSpecifyIf 76a equals 7, enter 27 here., otherwise enter the number of theemployment code entered in item 76c, 76e, 76g, 76i 76k, or 76m.76o.Ask or verify -Is your job with -Read each category.

77. EMPLOYERTYPECURRENT 1 A private company, business, or

individual for wages?580234The Federal government?A State, county, or local government?

Yourself (Self-employed) in your ownbusiness, professional practice, or farm?If 76o = 22

SKIP to 79,Else ASK 78.

78. COLLEGEEMPLOYER 581 1

2YesNoAre you employed by a college or university?While working at your job, do you work mostly in -Read each category.

79. CURRENTJOBMSATYPE 1

234A city?Suburban area?Rural area?Combination of any of these?582

RESPONDENT'S EMPLOYMENT QUESTIONSCHECKITEM H2FIELD REPRESENTATIVE -- (Read to the Household Respondent Only.) If there are any household members under 18, tell the HouseholdRespondent that you will be asking the same questions you just asked him/her.Is this the last household member to be interviewed? Yes - If Household Respondent finish collecting incomeand telephone information, then END interview.Otherwise END interview.No - GO TO question 33a for the next respondent. Seenote below before interviewing next householdmember.

Page 11

RESPONDENT'S CHECK ITEM I

90. HOUSEHOLDINCOME (Asked of Household Respondent Only)

Which category represents the TOTAL combined income of all membersof this HOUSEHOLD during the past 12 months? This includes moneyfrom jobs, net income from business, farm or rent, pensions, dividends,interest, Social Security payments, and any other money incomereceived by members of this HOUSEHOLD who are 14 years of age orolder.Less than $5,000$7,500 to $9,999$10,000 to $12,499$12,500 to $14,999$17,500 to $19,9991234567$15,000 to $17,499$20,000 to $24,9998$25,000 to $29,9999$30,000 to $34,99910$35,000 to $39,99911$40,000 to $49,99912$50,000 to $74,99913$75,000 to $99,99914$100,000 to $149,99915$150,000 to $199,99916$200,000 or more17214(Asked of household respondent.Asked at 1st, 3rd, 5th, and 7th interview, or if never asked before. Asked at subsequent interviews if no or Don't Know/Refused at prior interview.)$5,000 to $7,499CHECKITEM IOMB No. 1121-0111: Approval Expires 11/30/2024Implementation Date: (7/1/2019)NCVS-2

NCVS-2 CRIME INCIDENT REPORT

NATIONAL CRIME VICTIMIZATION SURVEY12Yes (more than 6 months) - SKIP to 3No (6 months or less) - ASK 2c606While living at this addressBefore moving to this address2c. INCIDENTADDRESSDid (this/the first) incident happen while youwere living here or before you moved to thisaddress?

3. INCIDENTDATE

Month Year605CHECKITEM AHas the respondent lived at this addressfor more than 6 months? (If not sure,refer to 33a on the NCVS-1 or ASK.)

4. INCIDENTNUMBEROFTIMES 607 ____________ Number of incidents

CHECKITEM B121-5 incidents (not a "series") - SKIP to 66 or more incidents - ASK 5b608CHECK

ITEM C 12

Similar - ASK 5c

Different (not a "series") - SKIP to 6609If unsure, ask:Are these incidents similar to each other indetail or are they for different types of crimes?How many incidents?

(Refer to 4.)5a.5b.5c.If unsure, ask:Can you recall enough details of each incident todistinguish them from each other?ITEM DCHECK Yes (not a "series")610 12 No (is a "series")(If Box 2 is marked in 5c, read: The following questionsrefer only to the most recent incident.)About what time did (this/the most recent)incident happen?

6. INCIDENTTIME During day

ORDon't know whether day or night98 Don't know what time of nightAfter 12 midnight - 6 a.m.75After 9 p.m. - 12 midnight6After 6 p.m. - 9 p.m.At nightDon't know what time of day4After 3 p.m. - 6 p.m.3After 12 noon - 3 p.m.2After 6 a.m. - 12 noon16122b.INCIDENTSSIMILARRECALLDETAILSNOTE: Questions are listed in the order asked. Skips in question numbering are due to questionnaire changes over time.You said that during the last 6 months :(description of the crime reported in the screen question.)2a. INCIDENTINTROIf unsure, ask:Altogether, how many times did this type ofincident happen during the last 6 months?In what month did (this/the first) incident happen?Encourage respondent to give exact month.1234Outside U.S. - SKIP to 10aNot inside a city/town/village - ASK 8aSAME city/town/village as present residence - SKIP to 97a. INCIDENTPLACEIn what city, town, or village did thisincident occur?613Don't know - ASK 8aDIFFERENT city/town/village from presentresidence - ASK 7b5614County ________________12YesNo6158a. INCIDENTSTATEIn what county did it occur?8c. COUNTYSTATEAsk or verify:Is this the same county and state as yourpresent residence?12YesNo

6339. INCIDENTAIR

Did this incident occur on an American IndianReservation or on American Indian Lands?Please specify the city, town, or village, in which theincident occurred.7b. INCIDENTPLACESPEC SpecifyIn what state did it occur?8b. INCIDENTCOUNTY10a. LOCATION_GENERALDid this incident happen ...Read each category until respondent says "yes", thenenter appropriate precode.In your home or lodging? - SKIP to 10bNear your home? - SKIP to 10cAt, in or near a friend's/relative's/neighbor'shome? - SKIP to 10dAt a commercial place? - SKIP to 10eIn a parking lot or garage? - SKIP to 10fAt school? - SKIP to 10gIn open areas, on the street, oron public transportation? - SKIP to - 10hSome where else? - SKIP to 10i12346578Page 2State ________________12Inside bank24Inside gas station25SKIPto 17c151617SKIPto 17cInside school building1819 On school property (school parkingarea, play area, school bus, etc.) SKIP to 17c20 In apartment yard, park, field,playground (other than school)On the street (other than immediatelyadjacent to own/friend's/relative's/neighbor's home)On public transportation or in station(bus, train, plane, airport, depot, etc.)2122SKIPto 18

Other - Specify23 SKIP to 17cNoncommercial parking lot/garageApartment/townhouse parking lot/garageInside other commercial building, suchas a store26Inside office14Inside factory or warehouse27Inside restaurant, bar, nightclub10e. LOCATION_COMMERCEAsk if necessary:At what type of a commercial place did thisincident happen?10f. LOCATION_PARKING10g. LOCATION_SCHOOLAsk if necessary:In what type of a parking lot or garage did thisincident happen?Ask if necessary:Where at school did this incident happen?10h. LOCATION_OPEN_AREAAsk if necessary:Where in an open area, on the street, or onpublic transportation did this incidenthappen?10i. LOCATION_SPECPlease specify the other location where thisincident occurred.Commercial parking lot/garageYard, sidewalk, driveway, carport (doesnot include apartment yards)SKIPto 18On street immediately adjacent to theirhome11Apartment hall, storage area, laundryroom (does not include apartmentparking lot/garage)109At or in home or other building on theirproperty8Ask if necessary:Where at, in, or near a friend's/relative's/neighbor's home did this incident happen?10d. LOCATION_OTHER_HOMEPage 3616 In own dwelling, own attached garage,or enclosed porch (Include illegal entryor attempted illegal entry of same)In hotel or motel room respondentwas staying in (Include illegal entry orattempted illegal entry of same)43In detached building on own property,such as detached garage, storage shed,etc. (Include illegal entry of same)21Ask if necessary:Where in your home or lodging did thisincident happen?10b. LOCATION_IN_HOMEIn vacation home/second home(Include illegal entry or attemptedillegal entry of same)Apartment hall, storage area, laundryroom (does not include apartmentparking lot/garage)SKIPto 18Own yard, sidewalk, driveway, carport,unenclosed porch (does not includeapartment yards)On street immediately adjacent to ownhome or lodging765Ask if necessary:Where near your home or lodging did thisincident happen?10c. LOCATION_NEAR_HOMESKIPto 11

SKIP to 17aPage 4

11. OFFENDERLIVE 1

23

Yes - SKIP to 19617Did the offender live (here/there) or have aright to be (here/there), for instance, as a guestor a repairperson?NoDon't know ASK 12123

Yes - SKIP to 14618NoDon't know ASK 13

12. OFFENDERINSIDE

Did the offender actually get INSIDE your(house/apartment/room/garage/ shed/enclosed porch)?Did the offender TRY to get in your (house/apartment/room/garage/shed/porch)? Don't know - ASK 14

No - SKIP to 19Yes - ASK 1432

161913. OFFENDERTRY

No - SKIP to 16aYes - ASK 15a21620Was there any evidence, such as a broken lockor broken window, that the offender(s) (got inby force/TRIED to get in by force)?

14. FORCEDENTRY

Window1 Damage to window (include frame,glass broken/removed/cracked)625Screen damaged/removedLock on window damaged/tamperedwith in some way23

Other - Specify4Door5 Damage to door (include frame, glasspanes or door removed)626Screen damaged/removedLock or door handle damaged/tamperedwith in some way67

Other - Specify8Other9 Other than window or door - Specify15a. EVIDENCEWhat was the evidence?Probe: Anything else?Enter all that apply.SKIPto 19

SKIP to 15c

SKIP to 15d

SKIP to 15bSKIPto 1915b. EVIDENCE_SPEC14Please specify what was the other evidence relatedto a window.Specify15c. EVIDENCE_SPEC18Please specify what was the other evidence relatedto a door.15d. EVIDENCE_SPEC19Please specify what was the evidence other than to awindow or door.

• SKIP to 19

Specify - SKIP to 19

Specify - SKIP to 19Page 52 Offender pushed his/her way in afterdoor opened627

Through OPEN DOOR or other openingThrough UNLOCKED door or window341 Let in5 Through LOCKED door or window - Hadkey6 Through LOCKED door or window -Picked lock, used credit card, etc., otherthan key7 Through LOCKED door or window -Don't know howDon't know

Other - Specify8916a. OFFENDERGETINHow did the offender (get in/TRY to get in)?Please specify - how the offender got in/TRIED to getin.16b. OFFENDERGETIN_SPECSKIPto 19

SKIP to 16b

Specify - SKIP to 1917a. RESPONDENTSSCHOOLWas it your school?628 1 Yes

No - SKIP to 17c217b. PARTSCHOOLBLDGIn what part of the school building did ithappen?629 1 ClassroomHallway/StairwellBathroom/Locker roomOther (library, gym, auditorium, cafeteria)2342 Restricted to certain people (or nobodyhad a right to be there)630 1 Open to the publicDon't know

Other - Specify - ASK 17d3417c. RESTRICTEDAREAAsk or verify -Did the incident happen in an area restricted tocertain people or was it open to the public atthe time?17d. RESTRICTEDAREA_SPECPlease specify.123Indoors (inside a building or enclosed space)Both

63118. INSIDEOROUT

Ask or verify -Did it happen outdoors, indoors, or both?SKIPto 18OutdoorsSpecify1 At, in, or near the building containingthe respondent's home/next door63223456A mile or lessFive miles or lessFifty miles or lessMore than 50 milesDon't know how far

19. FARFROMHOME

Ask or verify-How far away from home did this happen?PROBE: Was it within a mile, 5 miles, 50 miles ormore?Enter the code for the first answer category that therespondent is sure of.Ask or verify -Were you or any other member of thishousehold present when this incidentoccurred?You may need to probe to obtain more details todetermine if respondent was present.20a. HHMEMBERPRESENT 12Yes - ASK 20b

No - SKIP to 56634123Respondent onlyRespondent and other householdmember(s)635Ask 21Only other household member(s), notrespondent - SKIP to 59a20b. WHICHMEMBERAsk or verify -Which household members were present?Page 612YesNo

63621. SEEOFFENDER

Ask or verify -Did you personally see an offender?Did the offender have a weapon such as a gunor knife, or something to use as a weapon, suchas a bottle or wrench?123Yes - ASK 23a637NoDon't know

22. WEAPONPRESENT

123456Hand gun (pistol, revolver, etc.)Other gun (rifle, shotgun, etc.)KnifeOther sharp object (scissors, ice pick, axe,etc.)Blunt object (rock, club, blackjack, etc.)

Other - Specify - ASK 23b63823a. WEAPONWhat was the weapon?Probe: Anything else?Enter all that apply.23b. WEAPON_SPECPlease specify the other weapon.Did the offender hit you, knock you down oractually attack you in any way?12

Yes - SKIP to 29aNo - ASK 25

63924. ATTACK

12

Yes - SKIP to 28aNo - ASK 2664012

Yes - SKIP to 28cNo - ASK 27a641

25. TRYATTACK

Did the offender TRY to attack you?

26. THREATEN

Did the offender THREATEN you with harm inany way?SKIPto 24Specify642

SKIP to 27cSKIPto 35cSKIPto 35c123Something taken without permissionAttempted or threatened to takesomethingForcible entry or attempted forcibleentry of house/apartmentForcible entry or attempted forcible entryof carDamaged or destroyed property89

Other - Specify - ASK 27bAttempted or threatened to damage ordestroy property71027a. WHATHAPPENWhat actually happened?Probe: Anything else?Enter all that apply.46Unwanted sexual contact without force(grabbing, fondling, etc.)Harassed, argument, abusive languageUnwanted sexual contact with force(grabbing, fondling, etc.)5Please specify what actually happened.27b. WHATHAPPEN_SPECYou mentioned some type of unwanted sexualcontact with force. Do you mean forced orcoerced sexual intercourse including attempts?27c. SEXCONFORCEPROBE_1 Yes - SKIP to 29a

No - SKIP to 35c12

Specify - SKIP to 35c

SKIP to 24643 1234Verbal threat of rapeVerbal threat to killVerbal threat of attack other than to kill or rape

Other - Specify - ASK 28b14Attempted attack with weapon otherthan gun/knife/sharp weaponObject thrown at personFollowed or surroundedTried to hit, slap, knock down, grab, hold,trip, jump, push, etc.11131264464528a. HOWTRYATTACKHow did the offender TRY to attack you?Probe: Any other way?Enter all that apply.Verbal threat of sexual assault other than rapeUnwanted sexual contact with force(grabbing, fondling, etc.)Weapon present or threatened with weaponShot at (but missed)Attempted attack with knife/sharp weapon578910Unwanted sexual contact without force(grabbing, fondling, etc.)628b. HOWTRYATTACK_SPECPlease specify how the offender TRIED to attack you.You mentioned some type of unwanted sexualcontact with force. Do you mean forced orcoerced sexual intercourse includingattempts?28e. SEXCONFORCEPROBE_2 Yes - ASK 29a

No - SKIP to 35c12SKIPto 35cSKIPto 35c

Specify - SKIP to 35c643 1234Verbal threat of rapeVerbal threat to killVerbal threat of attack other than to kill or rape

Other - Specify - ASK 28d14Attempted attack with weapon otherthan gun/knife/sharp weaponObject thrown at personFollowed or surroundedTried to hit, slap, knock down, grab, hold,trip, jump, push, etc.11131264464528c. HOWTHREATENHow were you threatened?Probe: Any other way?Enter all that apply.Verbal threat of sexual assault other than rapeUnwanted sexual contact with force(grabbing, fondling, etc.)Weapon present or threatened with weaponShot at (but missed)Attempted attack with knife/sharp weapon578910Unwanted sexual contact without force(grabbing, fondling, etc.)628d. HOWTHREATEN_SPECPlease specify how you were threatened.SKIPto 35cSKIPto 35c

Specify - SKIP to 35cHow were you attacked?Probe: Any other way?Enter all that apply.29a. HOWATTACK 646 1234567891011RapedTried to rapeSexual assault other than rape orattempted rapeShotShot at (but missed)Hit with gun held in handStabbed/cut with knife/sharp weaponAttempted attack with knife/sharp weaponHit by object (other than gun) held in handHit by thrown objectAttempted attack with weapon otherthan gun/knife/sharp weaponHit, slapped, knocked downGrabbed, held, tripped, jumped, pushed, etc.

Other - Specify - ASK 29b121314647648ASK 29dASK 29c29b. HOWATTACK_SPECPlease specify how you were attacked.

Specify - SKIP to 30aPage 7

SKIP to 28e

SKIP to 28eSKIPto 30aPage 829c. RAPE_CK129d. ATTRAPE_CK1You mentioned rape. Do you mean forced orcoerced sexual intercourse?If "no", then ask: What do you mean?You mentioned attempted rape. Do you meanattempted forced or coerced sexualintercourse?If "no", then ask: What do you mean?

Yes - SKIP to 30aNo - go back to 29a

Yes - SKIP to 30aNo - go back to 29a123YesNo

Other - Specify - ASK 30b649655 12345678910NoneRapedAttempted rapeGun shot, bullet woundsBroken bones or teeth knocked outInternal injuriesKnocked unconsciousBruises, black eye, cuts, scratches,swelling, chipped teeth

Other - Specify - ASK 31b1165630a. PRETHREATENDid the offender THREATEN to hurt you beforeyou were actually attacked?31a. INJURYWhat were the injuries you suffered, if any?Probe: Anything else?Enter all that apply.30b. PRETHREATEN_SPECPlease specify.31b. INJURY_SPECPlease specify the injuries you suffered.

SKIP to 31a

SKIP to 35c

SKIP to 31c

SKIP to 31dSexual assault other than rape orattempted rapeKnife or stab woundsSKIP to32aSpecify

Specify - SKIP to 32a12Yes - ASK 32b

No - SKIP to 33a65765832a. INJURYNOTGUNAsk or verify-Were any of the injuries caused by a weaponother than a gun or knife?32b. FIRSTINJURYWhich injuries were caused by a weaponOTHER than a gun or knife?Enter all that apply.You mentioned rape. Do you mean forced orcoerced sexual intercourse?If "no", then ask: What do you mean?31c. RAPE_CK2You mentioned attempted rape. Do you meanattempted forced or coerced sexualintercourse?If "no", then ask: What do you mean?31d. ATTRAPE_CK2Sexual assault other than rape or attempted rape23478910Attempted rapeBroken bones or teeth knocked outInternal injuriesKnocked unconsciousBruises, black eye, cuts, scratches,swelling, chipped teeth

Other - Specify11Raped

Yes - SKIP to 32aNo - go back to 31a

Yes - SKIP to 32aNo - go back to 31a12Yes - ASK 33b

No - SKIP to 35c659Were you injured to the extent that youreceived any medical care, including selftreatment?33a. MEDICALCARE

SKIP to 33a

SKIP to 32aPage 912At the sceneAt home/neighbor's/friend's660662Health unit at work/school, first aidstation at a stadium/park, etc.Doctor's office/health clinicEmergency room at hospital/emergency clinicHospital (other than emergency room)

Other - Specify - ASK 33c34567663 ____________ Number of daysWhere did you receive this care?Probe: Anywhere else?Enter all that apply.33b. RECEIVECAREWHERE34a. CAREOVERNIGHTDid you stay overnight in the hospital?34b. CAREDAYHOSPITHow many days did you stay in the hospital?33c. RECEIVECAREWHERE_SPECPlease specify where you received this care.SKIPto33d33d. Is (Box 6) "Hospital" marked in35a?SpecifyCHECKITEM E112Yes - ASK 34b

No - SKIP to 35a$ ____________ .665 Total amount000 No cost35b. MEDICALEXPENSESWhat was the total amount of your medicalexpenses resulting from this incident(INCLUDING anything paid by insurance)?Include hospital and doctor bills, medicine,therapy, braces, and any other injury relatedexpenses.Obtain an estimate, if necessary.At the time of the incident, were you coveredby any medical insurance, or were you eligiblefor benefits from any other type of healthbenefits program, such as medicaid, VeteransAdministration, or Public Welfare?35a. MEDICALINSURANCE3 Don't know2 No1 Yes66412Yes - ASK 34a

No - SKIP to 35a35c. Is (Box 1) "Yes" marked in 24, 25 or26 or are (Box 4 or 5) "Unwantedsexual contact with or withoutforce" marked in 27?CHECKITEM E212Yes - ASK 36a

No - SKIP to 3936a. EMO_TOLL_IMPACT_JOBBeing a victim of crime affects people indifferent ways. Next I would like to ask yousome questions about how being a crimevictim may have affected you.Did being a victim of this crime lead you tohave significant problems with your job orschoolwork, or trouble with your boss,coworkers, or peers?12YesNo36b. EMO_TOLL_IMPACT_FAMILYDid being a victim of this crime lead you tohave significant problems with familymembers or friends, including getting intomore arguments or fights than you did before,not feeling you could trust them as much, ornot feeling as close to them as you did before?12YesNo36c. EMO_TOLL_IMPACT_DISTRESSINGHow distressing was being a victim of thiscrime to you? Was it not at all distressing,mildly distressing, moderately distressing, orseverely distressing?12Not at all distressingMildly distressing34Moderately distressingSeverely distressing971972973Page 10Yes1 2 NoStill thinking about your distressassociated with being a victim of this crimedid you feel any of the following ways for AMONTH OR MORE? Did you feel ...Worried or anxious?

(a) EMO_TOLL_FEEL_WORRIED

36d. Is (Box 1) "Yes" marked in 36a or36b or are (Box 3 or 4)"Moderately or severelydistressing" marked in 36c?CHECKITEM E312Yes - ASK 37

No - SKIP to 39Angry?

(b) EMO_TOLL_FEEL_ANGRY Yes1 2 NoSad or depressed?

(c) EMO_TOLL_FEEL_SAD Yes1 2 NoVulnerable?

(d) EMO_TOLL_FEEL_VULNERABLE Yes1 2 NoViolated?

(e) EMO_TOLL_FEEL_VIOLATED Yes1 2 NoLike you couldn't trust people?

(f) EMO_TOLL_FEEL_MISTRUST Yes1 2 NoUnsafe?

(g) EMO_TOLL_FEEL_UNSAFE Yes1 2 NoSome other way?

(h) EMO_TOLL_FEEL_OTHER_WAY Yes1 2 No37i. FEEL_OTH_WAY_SPWhat other way did being a victim of thiscrime make you feel?Specify37k. EMO_TOLL_SEEK_PRO_HELPDid you seek any kind of professional help forthe feelings you experienced as a result ofbeing a victim of this crime?12Yes - ASK 37l

No - SKIP to 3837l. EMO_TOLL_PRO_HELP_SOUGHTDid you seek any kind of professional help forthe feelings you experienced as a result ofbeing a victim of this crime?Enter all that apply.12Counseling/therapyMedication34Visited a doctor or nurse

Visted ER/hospital/clinic

5 Other - Specify - ASK 37mSKIPto 3837m. HELP_SOUGHT_SPWhat other kind of professional help did you seek?Specify37j. Is (Box 1) "Yes" marked in any of37a through 37h?CHECKITEM E412Yes - ASK 37k

No - SKIP to 38974975976977978979980981982983Yes1 2 NoDid you experience any of the followingphysical problems associated with being avictim of this crime for A MONTH OR MORE?Did you experience ...Headaches?

(a) EMO_TOLL_HAVE_HEADACHES

Yes1 2 NoTrouble sleeping?

(b) EMO_TOLL_TRBL_SLEEPINGYes1 2 NoChanges in your eating or drinking habits?

(c) EMO_TOLL_EATING_PROBSYes1 2 NoUpset stomach?

(d) EMO_TOLL_UPSET_STOMACHYes1 2 NoFatigue?

(e) EMO_TOLL_FATIGUEYes1 2 NoHigh blood pressure?

(f) EMO_TOLL_HIGH_BLOOD_PRESSYes1 2 NoMuscle tension or back pain?

(g) EMO_TOLL_MUSCLE_TENSIONYes1 2 NoSome other physical problem?

(h) EMO_TOLL_OTHER_PHYSICAL38i. OTH_PHY_SPWhat other physical problem did youexperience for A MONTH OR MORE?Specify38j. Is (Box 1) "Yes" marked in any of38a through 38h?CHECKITEM E512Yes - ASK 38k

No - SKIP to 3938l. EMO_TOLL_KIND_HELP_PHYPROBSWhat type of professional or medical help didyou seek?Enter all that apply.12Counseling/therapyMedication34Visited a doctor or nurse

Visted ER/hospital/clinic

5 Other - Specify - ASK 38mSKIPto 3938m. KIND_HELP_PHYOTH_SPWhat other kind of professional help did you seek?Specify38k. EMO_TOLL_SEEK_HELP_PHYPROBSOther than any medical care you received forthe injury(ies) you suffered) did you seek anyprofessional or medical help for the physicalproblems you experienced as a result of beinga victim of this crime?12Yes - ASK 38l

No - SKIP to 39

39. Is respondent female and

between the ages of 18-49?CHECKITEM F112Yes - ASK 40

No - SKIP to 41a960 12YesNo

40. PREGATTIMEOFINC

Research shows that pregnant women may beat a higher risk of being the victim of a violentcrime.Were you pregnant at the time of this incident?Page 11984985986987988989990991992993Page 12668 123456Attacked offender with gun; fired gunAttacked with other weaponThreatened offender with other weaponThreatened to injure, no weapon43a. INJACTIONDid you take these actions before, after, or atthe same time that you were injured?Enter all that apply.43b. INJACTIONHELPDid (any of) your action(s) help the situation inany way?Probe: Did your action(s) help you avoid injury,protect your property, escape from theoffender - or were they helpful in some otherway?12

Yes - SKIP to 42aNo/took no action/kept still - ASK 41b

USED PHYSICAL FORCE TOWARD OFFENDER672 123Actions taken before injuryActions taken after injuryActions taken at same time as injury673669 7

RESISTED OR CAPTURED OFFENDER8 Chased, tried to catch or hold offender9 Yelled at offender, turned on lights,threatened to call police, etc.

SCARED OR WARNED OFF OFFENDER

PERSUADED OR APPEASED OFFENDER67010 Cooperated, or pretended to (stalled, didwhat they asked)11671121314Called police or guard

REACTED TO PAIN OR EMOTION15 Screamed from pain or fear41a. PROTECTSELFDid you do anything with the idea ofprotecting YOURSELF or your PROPERTY whilethe incident was going on?41b. DURINGINCIDENTWas there anything you did or tried to doabout the incident while it was going on?42a. ACTIONSDURINGINCWhat did you do?Probe: Anything else?Enter all that apply.OTHER

16 Other - Specify - ASK 42b42b. ACTIONDURINGINC_SPECPlease specify what you did.Attacked without weapon (hit, kicked, etc.)Threatened offender with gunDefended self or property (struggled,ducked, blocked blows, held onto property)Argued, reasoned, pleaded, bargained, etc.Ran or drove away, or tried; hid, locked doorTried to attract attention or help, warn others(cried out for help, called children inside)SKIPto 42cSKIPto 42c12

Yes - SKIP to 42aNo/took no action/kept still - ASK 47

ESCAPED OR GOT AWAY

GOT HELP OR GAVE ALARMSpecify42c. Was the respondent injured inthis incident? (Are any of theBoxes 2-11 marked in 31a?)CHECKITEM F212Yes - ASK 43a

No - SKIP to 43b3 Don't know2 No1 Yes - ASK 44a666667

SKIP to 45

67545. ACTIONWORSE

Did (any of) your action(s) make the situationworse in any way?Probe: Did your action(s) lead to injury, greaterinjury, loss of property, make the offenderangrier, or make the situation worse in someother way?46a. WORSEHow did they make the situation worse?Probe: Any other way?Enter all that apply.

47. ANYONEPRESENT

Was anyone present during the incidentbesides you and the offender? (Other thanchildren under age 12.)

48. OTHERSACTIONS

Did the actions of (this person/any of thesepeople) help the situation in any way?49a. HOWOTHERSHELPHow did they help the situation?Probe: Any other way?Enter all that apply.676 123456Led to injury or greater injury torespondentCaused greater loss of property ordamage to propertyOther people got hurt (worse)Offender got awayMade offender angrier, moreaggressive, etc.

Other - Specify - ASK 46b44b. HELP_SPECPlease specify how were they helpful.46b. WORSE_SPECPlease specify how the respondent's actions madethe situation worse.49b. HOWOTHERSHELP_SPECPlease specify how they helped the situation.674 123456Helped respondent get away from offenderProtected propertyProtected other people

Other - Specify - ASK 44b44a. HELPHow were they helpful?Probe: Any other way?Enter all that apply.SKIPto 47Helped avoid injury or greater injury torespondentScared or chased offender offSpecify3 Don't know2 No1 Yes - ASK 46aSpecify6773 Don't know2 No1 Yes - ASK 486783 Don't know2 No1 Yes - ASK 49a679 123456Helped respondent get away from offenderProtected propertyProtected other people

Other - Specify - ASK 49bHelped avoid injury or greater injury torespondentScared or chased offender offSpecify680

50. OTHERSACTIONSWORSE

Did the actions of (this person/any of thesepeople) make the situation worse in any way? 3 Don't know2 No1 Yes - ASK 51aPage 13SKIPto 45

SKIP to 47

SKIP to 54c

SKIP to 50SKIPto 50

SKIP to 52Page 14684

0 None - SKIP to 54c

52. PERSONSHARMED

Not counting yourself, were any of the personspresent during the incident harmed (Pause),threatened with harm (Pause), or robbed byforce or threat of harm? (Do not includeyourself, the offender, or children under 12years of age.)

53. PERSONSHARMEDNUM

How many? (Do not include yourself, theoffender or children under 12 years of age.)54a. HHMEMHARMEDHow many of these persons are members ofyour household now? (Do not include yourself,the offender or children under 12 years of age.)____________ Number of persons683____________ Number of persons51a. OTHWORSEHow did they make the situation worse?Probe: Any other way?Enter all that apply.681 123456Led to injury or greater injury torespondentCaused greater loss of property ordamage to propertyOther people got hurt (worse)Offender got awayMade offender angrier, moreaggressive, etc.

Other - Specify - ASK 51b51b. OTHWORSE_SPECPlease specify how they made the situation worse.SKIPto 52Specify6823 Don't know2 No1 Yes - ASK 53Line number(s)If not sure ask:Who are these household members? (Do notinclude yourself, the offender, or childrenunder 12 years of age)Enter the line number(s) of other householdmembers.54b. HHMEMHARMED_NAMES68554c.Did the respondent use orthreaten to use physical forceagainst the offender? (Are any ofthe Boxes 1-6 marked in 42a?)CHECKITEM G12Yes - ASK 55

No - SKIP to 60123RespondentOffender(s)Someone else

68655. FIRSTTOUSEFORCE

Who was the first to use or threaten to usephysical force - you, the offender, or someoneelse?12Yes - ASK 57

No - SKIP to 88687123SuspicionFairly sureCertain688

56. KNOWLEARNOFFENDERS
57. SUREOFINFO

How sure are you of this information? Do youhave a suspicion, are you fairly sure or are youcertain?Do you know or have you learned anythingabout the offender(s) - for instance, whetherthere was one or more than one offenderinvolved, whether it was someone young orold, or male or female?RESPONDENTUSEFORCE

SKIP to 60

SKIP to 54c689SKIPto 881 Respondent saw or heard offenderFrom other member of household whowas eyewitnessFrom eyewitness(es) other thanhousehold member(s)From policeOther person (not eyewitness)Offender(s) admitted itOffender(s) had threatened to do it4567

Other - Specify - ASK 58b3106902Stolen property found on offender'sproperty or in offender's possession8Figured it out by who had motive,opportunity, or had done it before958a. LEARNOFFENDERSHow did you learn about the offender(s)?Probe: Any other way?Enter all that apply.58b. LEARNOFFENDERS_SPECPlease specify how you learned about the offender(s).

Specify - SKIP to 88123Something taken without permissionAttempted or threatened to take somethingHarassed, argument, abusive language691Forcible entry or attempted forcible entryof house/apartment4567Forcible entry or attempted forcibleentry of carDamaged or destroyed propertyAttempted or threatened to damage ordestroy property

Other - Specify - ASK 59b859a. HAPPENWhat actually happened?Probe: Anything else?Enter all that apply.SKIPto 6059b. HAPPEN_SPECPlease specify what actually happened.Specify123

Only one - SKIP to 62

More than one - SKIP to 74Don't know - ASK 61

69260. ONEORMOREOFFENDERS

Ask or verify -Was the crime committed by only one or bymore than one offender?12Yes - ASK 62

No - SKIP to 88

69361. KNOWOFFENDERS

Do you know anything about one of theoffenders?

62. SINGOFFENDERKNEW
63. SINGOFFENDERRECOG
64. SINGOFFENDERHOWWELL

65a. SINGOFFENDERSIGHT65b. SINGOFFENDERSIGHT_SPECWas the offender someone you knew or astranger you had never seen before?703 Knew or had seen before - SKIP to 64StrangerDon't know123Page 15Would you be able to recognize the offender ifyou saw him/her?

SKIP to 65a704 YesNot sure (possibly or probably)

No - SKIP to 67123How well did you know the offender - by sightonly, casual acquaintance, or well known?

SKIP to 66a705 Sight only - ASK 65aCasual acquaintanceWell known123Would you have been able to tell the policehow they might find the offender, for instance,where he/she lived, worked, went to school, orspent time?

SKIP to 67706 123YesNo

Other - Specify - ASK 65b

Specify - SKIP to 67

67. SINGOFFENDERGENDER 698 1

23MaleFemaleDon't knowWas the offender male or female?699 1234Under 1212-1415-1718-20

68. SINGOFFENDERAGE

How old would you say the offender was?56721-2930 or olderDon't know

71. SINGOFFENDERGANG

Was the offender a member of a street gang, ordon't you know?Yes (a member of a street gang)No (not a member of a street gang)Don't know (if a member of a street gang)700 12372a. SINGOFFENDERDRINKDRUGWas the offender drinking or on drugs, or don'tyou know?72b. SINGOFFENDERDRINKORDRUGWhich was it? (Drinking or on drugs?)701 123Yes (drinking or on drugs) - ASK 72bNo (not drinking/not on drugs)Don't know (if drinking or on drugs)702 DrinkingOn drugsBoth (drinking and on drugs)Drinking or on drugs - could not tell which1234Page 1666a. SINGOFFENDERRELATIONHow well did you know the offender? Forexample, was the offender a friend, cousin,etc.?707Spouse at time of incidentEx-spouse at time of incidentParent or step-parentOwn child or step-childBrother/sister

Other relative - Specify - ASK 66b123456RELATIVENONRELATIVEBoyfriend or girlfriend, ex-boyfriend orex-girlfriendFriend or ex-friendRoommate, boarderSchoolmateNeighborCustomer/clientPatientSupervisor (current or former)Employee (current or former)Co-worker (current or former)789101112151617181413Teacher/school staff

Other nonrelative - Specify - ASK 66c66b. SINGOFFENDERRELATION_SPEC_16Please specify the other relative.66c. SINGOFFENDERRELATION_SPEC_28Please specify the other nonrelative.SKIPto 67SKIPto 67

Specify - SKIP to 67Specify

69. SINGOFFETHNICITY 994 1

23YesNoDon't knowWas the offender Hispanic or Latino?

70. SINGOFFENDERRACE

What race or races was the offender? You mayselect more than one. Was the offender...Enter all that apply.White?Black or African American?American Indian or Alaska Native?Don't know123470856Native Hawaiian or Other Pacific Islander?Asian?Was this the only time this offender committeda crime against you or your household or madethreats against you or your household?

73. SINGOFFENDERONLYTIME SKIP to 88

709 Yes (only time)No (there were other times)Don't know123

74. HOWMANYOFFENDERS

How many offenders?710Number of offenders

SKIP to 73Page 17

71875. MULTOFFENDERKNEW

Were any of the offenders known to you, orwere they strangers you had never seenbefore?12All knownSome knownASK 7634All strangersDon't know

76. MULTOFFENDERRECOG

Would you be able to recognize any of them ifyou saw them?719 123YesNot sure (possibly or probably)

No - SKIP to 80a

SKIP to 78a77a. MULTOFFENDERHOWWELL 720 123Sight onlyCasual acquaintanceWell knownHow well did you know the offender(s) - bysight only, casual acquaintance, or wellknown?Probe: Anything else?Enter all that apply.77b. Is "casual acquaintance" or "wellknown" marked in 77a?CHECKITEM H

Yes - SKIP to 79aNo - ASK 78a78a. MULTOFFENDERSIGHTWould you have been able to tell the policehow they might find any of them, for instance,where they lived, worked, went to school, orspent time?722 123YesNo

Other - Specify - ASK 78b78b. MULTOFFENDERSIGHT_SPECPlease specify.

Specify - SKIP to 80a79a. MULTOFFENDERRELATIONHow did you know them? For example, werethey friends, cousins, etc.?Probe: Anything else?Enter all that apply.Spouse at time of incidentEx-spouse at time of incidentParent or step-parentOwn child or step-childBrother/sister

Other relative - Specify - ASK 79b123456RELATIVENONRELATIVEBoyfriend or girlfriend, ex-boyfriend orex-girlfriendFriend or ex-friendRoommate, boarderSchoolmateNeighborCustomer/clientPatientSupervisor (current or former)Employee (current or former)Co-worker (current or former)789101112151617181413Teacher/school staff

Other nonrelative - Specify - ASK 79cSKIPto 80aSKIPto 80a96196279b. MULTOFFENDERRELATION_SPEC_16Please specify the other relative.79c. MULTOFFENDERRELATION_SPEC_28Please specify the other nonrelative.

Specify - SKIP to 80aSpecify

SKIP to 811234All maleAll femaleDon't know sex of any offendersBoth male and female71180a. MULTOFFENDERGENDERWere they male or female?(If only two offenders,

SKIP to 81otherwise ASK 80b)71280b. MULTOFFENDERMOSTGENDERWere they mostly male or mostly female?1234Mostly maleMostly femaleEvenly dividedDon't know963965964SKIPto 80a

SKIP to 77a

SKIP to 80a

81. MULTOFFENDERYOUNG

How old would you say the youngest was?

82. MULTOFFENDEROLD

How old would you say the oldest was?

85. MULTOFFENDERGANG

Were any of the offenders a member of a streetgang, or don't you know?86a. MULTOFFENDERDRINKDRUGWere any of the offenders drinking or ondrugs, or don't you know?86b. MULTOFFENDERDRINKORDRUGWhich was it? (Drinking or on drugs?)Page 18713 1234Under 1212-1415-1718-2056721-2930 or olderDon't know714 1234Under 1212-1415-1718-2056721-2930 or olderDon't knowYes (a member of a street gang)No (not a member of a street gang)Don't know (if a member of a street gang)715 123716 123Yes (drinking or on drugs) - ASK 86bNo (not drinking/not on drugs)Don't know (if drinking or on drugs)717 DrinkingOn drugsBoth (drinking and on drugs)Drinking or on drugs - could not tell which123483a. MULTIPLEOFFENDERETHNICITY 995 123Yes - ASK 83bNoDon't knowWere any of the offenders Hispanic or Latino?83b. MULTIPLEOFFENDERMOSTETHNICITYWere the offenders mostly Hispanic, mostlynon-Hispanic, or an equal number of Hispanicand non-Hispanic?

996 Mostly Hispanic or LatinoMostly non-HispanicEqual number of Hispanic and non-HispanicDon't know123472684a. MULTOFFENDERRACEWhat race or races were the offenders? Werethey...Enter all that apply.White?Black or African American?American Indian or Alaska Native?Asian?123484b. Is more than one box marked in84a?CHECKITEM IYes - ASK 84c

No - SKIP to 85

Mostly White

Mostly Black or African American

Mostly American Indian or Alaska Native

Mostly Asian

Mostly Native Hawaiian or Other Pacific Islander72784c. MULTOFFENDERRACEMOST 12345What race were most of the offenders?Don't know56Native Hawaiian or Other Pacific Islander?Equal number of each raceDon't know67