|Determining the Optimal Number of Interview Waves in a Panel Survey with Application to the NCVS: Evaluation and Recommendations|
|Interviewing Conditions in the National Crime Victimization Survey, 1993-2013|
|Evaluation of Direct Variance Estimation, Estimate Reliability, and Confidence Intervals for the National Crime Victimization Survey|
|NCVS Task 4 Report: Summary of Options Relating to Local Area Estimation|
|MORE LATEST RELEASES|
|On This Page|
|About this Topic|
The National Crime Victimization Survey (NCVS) was instituted in 1972 to produce national estimates of the levels and characteristics of criminal victimization in the United States, including crime not reported to police departments. Along with the FBI's Uniform Crime Reporting Program (UCR), the NCVS constitutes a key component of our nation's system to measure the extent and nature of crime in the United States.
BJS has invested in the continuous examination of the technical and substantive issues related to the collection and dissemination of victimization statistics.
Though the NCVS was originally designed to provide national level estimates of criminal victimization, BJS has recognized an increasing need for victimization data at the state and local level. The three major reviews of the NCVS program (Penick and Owens, 1976; Biderman et al., 1986; Groves and Cork, 2008) all point to the demand local criminal justice administrators have for empirical information to shape policy. Read more >
The NCVS Instrument Redesign and Testing Project is a major multi-year effort to overhaul the existing survey instrument. The overarching objective of the project is to provide scientific and technical support for the redesign and testing of the National Crime Victimization Survey (NCVS) roster control card, crime screener (NCVS-1), and crime incident (NCVS-2) instruments in support of BJS's efforts related to increasing the efficiency, reliability, and utility of the NCVS. Through the project, BJS aims to evaluate and modernize the organization and content of the NCVS; improve the efficiency of the instruments and the current core-supplement design; and develop a procedure for introducing routine improvements to the survey in order to capture emerging crime types and time-relevant topics. Read more >
BJS has initiated projects to identify, develop, and test the best methods for collecting self-report data on rape and sexual assault. In June 2011, BJS charged an expert panel from the National Research Council's Committee on National Statistics (CNSTAT) to examine conceptual and methodological issues surrounding survey statistics on rape and sexual assault and to recommend to BJS the best methods for obtaining such statistics on an ongoing basis. In September 2011, BJS made a competitive award to Westat, Inc., to develop and test two different survey designs for collecting self-report data on rape and sexual assault. One design is to be an optimal design identified by the CNSTAT panel and the other will be similar to designs used in the public health approach for measuring rape and sexual assault. Estimates from these two designs will be compared to data from the NCVS. Read more >
The central task of the NCVS Historical Trends (NHT) project is to integrate cross-sectional concatenated data from the historical National Crime Survey (NCS) and the contemporary NCVS into a single NHT data file. Although the victimization survey data is publicly available, the data files can be difficult to work with, making it difficult for most persons to use the files or generate historical estimates.
The primary purpose of the NHT data file is to provide BJS analysts, external researchers, and the broader community of BJS stakeholders with easy or more readily available access to the entire series of victimization survey data, which began in 1973. To date there has not been a single location where researchers or the interested public could quickly access information on trends or retrieve data they could then use in their own analysis. Another element of the NHT project is the development of a Historical Trends Tool (HTT), which will work in combination with BJS's current NVAT, allowing users to generate more than 40 years of trend data for select variables.
This project will develop a NCVS supplement focusing on the crime of financial fraud. There is consensus that fraud is an emerging and diverse crime type affecting millions of victims and their families. Currently, fraud is not directly measured on the NCVS. The lack of information on the prevalence and circumstances of fraud limits our understanding of the mechanisms, consequences, and assistance to individuals who experience these crimes. To address this problem, BJS has been working closely with the Financial Fraud Research Center at Stanford Center on Longevity and the FINRA Investor Education Foundation (FINRA Foundation) to develop a standardized fraud classification scheme. The purpose of the taxonomy is to group and organize fraud types meaningfully and systematically into a hierarchical framework that can be translated into survey questions and administered as a supplement to the core NCVS. This future supplement will be used to determine the annual prevalence of specific types of financial fraud and how prevalence rates change over time.
The project goal is to conduct an assessment of how administrative data from Adult Protective Services (APS) agencies can be used to provide data to develop uniform, national statistics about elder abuse, and how APS data may augment currently available crime and victimization statistics. The first step to achieving the project goal is to determine key statistics and to develop a comprehensive taxonomy that describes the various dimensions of elder abuse. The taxonomy defines elder abuse along three dimensions: (1) the acts that constitute elder abuse, (2) the characteristics of the victim, and (3) the relationship between the victim and perpetrator. The second step is to determine if the key statistics can be generated based on the information currently available in APS systems.
This project is a collaborative effort between BJS and the Office for Victims of Crime (OVC). The key objective of this project is to refine and pilot test a Victim-Offender Overlap Survey (VOOS) instrument with a sample of approximately 2,000 jail inmates to better understand the relationship between victimization and offending, and to assess the cost, resources, and challenges associated with implementing the VOOS with a national sample. Jail inmates offer an important opportunity to examine the relationships between victimization and offending because such persons are at high risk for both types of events but are incarcerated for shorter periods than prison inmates, which alleviates potential recall problems when interviewed about a reference period prior to incarceration.
This project developed measures of economic well-being in the NCVS in relation to criminal victimization. One of the first tasks in this project was to provide an assessment and recommendations for appropriate imputation strategies for missing income responses in the NCVS. Beginning in 2015, an imputed income variable will appear on the public use files. In addition, broader measures of poverty and socio-economic status (SES) and their utility and value towards understanding criminal victimization were examined.
This project conducted an evaluation of the use of generalized variance parameters (GVFs) versus direct estimation for producing standard errors around point estimates. The evaluation considered different types of estimates produced from the core NCVS and NCVS Supplements, including annual and multi-year aggregated counts, rates, and percentages for major crime types, various subpopulations, and incident characteristics. The assessment compared the utility of GVFs versus Taylor series and replicate weights for making year-to-year and within year comparisons across estimates. The assessment also considered the implications of moving from GVFs to direct estimation in terms of software needs, public use file configuration and necessary variables, the NCVS Victimization Analysis Tool, and prior BJS analyses and reports.
These two projects examined how BJS should deal with both telescoping and fatigue with the NCVS. The first examined how BJS should handle unbounded interviews, sample fatigue, and attrition or nonresponse when generating point estimates. While the time in sample (TIS) 1 adjustments have increased over time, no adjustment has ever been made to the unbounded rates for replacement households. Because the NCVS follows households and not families, replacement households are households where new families move in TIS 2 through TIS 7 due to change in ownership or renters. The replacement households account for about 15% to 20% of the sample and are as likely as the households in TIS 1 to over-report due to telescoping. In addition to whole household replacement, many individuals move into the sample in TIS 2 through TIS 7 because they age into eligibility (turn 12 years of age) or are added to existing households through marriage or other change in family structure. Additionally, little research to date has examined the impact of respondent fatigue, which could result in under-reporting and attrition in TIS 2 through TIS 7.
The second project examined alternative designs to reduce the time NCVS households are in sample. The NCVS is administered to all eligible respondents within sampled households every 6 months for 3.5 years, totaling 7 interviews. Although this longitudinal design offers a number of benefits, recent evidence suggests that respondents may experience survey fatigue, which could affect the validity of their responses in later interviews. The project assessed the cost and data quality issues associated with moving from a 7-TIS design to a 5-TIS, 4-TIS, 3-TIS, or 1-TIS design.
BJS has investigated ways to include high-frequency repeat victimizations, or series victimizations, in estimates of criminal victimization. Including series victimizations would obtain a more accurate estimate of victimization. A report summarizes the research results and describes changes in BJS's enumeration practices regarding the treatment of series victimizations when estimating annual victimization rates.
The NCVS currently interviews adolescents ages 12 to 17 but excludes children ages 11 or younger. This coverage error results in a direct underestimation of the annual number of crimes occurring in the United States and it may also result in a distortion of the distribution of crime types. For example, children may suffer disproportionately from sexual violence. However, before considering ways to address this coverage error, BJS first needs to assess the reliability of its current operations collecting information from adolescents age 12 to 17.
To accomplish these aims, the project's methodological work focused on reviewing and describing the current approaches to interviewing adolescents age 12 to 17 in the NCVS. The project used current NCVS data files (1993-2012), data from the School Crime Supplement (SCS), which is administered every other year to juveniles enrolled in school, and paradata to analyze both substantive and technical aspects of the data collection operation. The assessment also included a discussion of the advantages and disadvantages to various sample selection options, data collection modes, instrument development, ethical concerns, and victimization prevalence and incidence estimation procedures.
Virtually all data collection efforts experience the challenge of missing data, and the National Crime Victimization Survey (NCVS) is no exception. The NCVS is a panel data collection effort with three possible types of nonresponse: 1) unit, 2) wave, and 3) item. BJS uses weight adjustments to account for unit and wave nonresponse. However, little is done to address item nonresponse. This missing data may result in a potentially biased estimate. This is particularly problematic when it comes to the measurement of household income—one of the NCVS items that has historically suffered from high rates of nonresponse. As BJS works to continually improve the utility of the NCVS for understanding changes in crime and its correlates over time, the high levels of missing income data have to be addressed to accurately establish associations between victimization and socioeconomic factors, including income.
To accomplish these aims, the project's methodological work focused on reviewing and describing the current approaches to interviewing adolescents ages 12 to 17 in the NCVS. This report describes the problem of missing income data in the NCVS, details and assesses several potential approaches to imputing the missing data, and makes recommendations on the use of the single imputation hot deck method for imputation of NCVS household income data, or any other variables, when there is item nonresponse.
The purpose of the project was to explore survey methods that increased survey participation while maintaining affordable costs. The objective was to examine the use of interactive voice response (IVR) as a complementary mode of data collection to the interviewer-based methods currently used in the NCVS. As a mode of data collection, IVR has the potential to collect better information on the more sensitive items, as well as offering a less expensive mode of collection that could be applied to the core NCVS or to a companion survey designed to obtain city or state level estimates of victimization. If feasible, use of IVR could have a significant effect on the resources available for other components of the NCVS
This research examined self-administered survey methods with the potential to increase survey participation while maintaining affordable costs and data quality. This included providing respondents with more options for participation and testing whether nominal incentives increase subsequent survey participation when self-administration modes of inbound CATI and Web were utilized. The objective of the study was to examine the use of Inbound CATI and Web modes as complementary forms of data collection to the interviewer-based methods that are currently used in the NCVS. Inbound CATI and Web modes have the potential to increase survey participation by increasing the ease with which survey respondents participate by allowing them discretion as to when and where they respond to the survey. Self-administered modes also have the potential to collect better information on the more sensitive items, as well as offering a less expensive mode of collection that might be applied to the core NCVS.
This research examined the use of Enhanced Contextual Priming (ECP) as a method for improving event recall and assessed respondent burden when ECPs were incorporated into NCVS interviews. The field test featured two screeners: (1) the current NCVS control screener with a 6-month reference period, and (2) an enhanced screener that included attitudinal and behavioral priming questions that served to provide context to the screener, also with 6-month reference period. BJS used findings from this research to decide whether the incorporation of contextual questions was viable in the current NCVS program. The 1992 NCVS redesign improved the screener questions by incorporating short cues designed to trigger memories in differing contexts. The contextual priming questions examined in this research expanded the approach by providing a broader range of information that could facilitate retrieval of events that are relevant to the NCVS.
BJS conducted an assessment of the NCVS screening questions. Analysis of the response patterns associated with redesigned screening questions and the effectiveness of each question in eliciting recall of victimization had not been assessed since 1993. As a result, BJS deemed that assessing the performance of the screener was a critical first step in the process of overall questionnaire streamlining.
Major consequences of increasing nonresponse rates include an increase in survey costs associated with attempting to complete interviews and the potential introduction of biases in survey estimates for sub-groups with high nonresponse rates. BJS examined this issue in greater detail in an Analysis of Possible NonResponse Bias: Final Report.
|Data Collections & Surveys|
|Publications & Products|
|Determining the Optimal Number of Interview Waves in a Panel Survey with Application to the NCVS: Evaluation and Recommendations Presents the results of a study that evaluated the data quality and cost of changing the National Crime Victimization Survey (NCVS) from the current longitudinal design in which household members are interviewed every 6 months for a total of 7 times over a 3.5-year period to designs in which respondents are interviewed 1, 3, 4, or 5 times.|
|Evaluation of Direct Variance Estimation, Estimate Reliability, and Confidence Intervals for the National Crime Victimization Survey Examines the feasibility of using direct variance estimation for the National Crime Victimization Survey (NCVS).|
|NCVS Task 4 Report: Summary of Options Relating to Local Area Estimation This report, generated by Westat, reviews three related approaches for generating Local Area Estimates (LAEs) for the National Crime Victimization Survey (NCVS).|