Added: July 12, 2014 – Last updated: April 25, 2015


Author: Sami Ansari

Title: Understanding and Modeling the Convergence of the UCR and NCVS

Subtitle: A Time Series Analysis

Thesis: Ph.D. Thesis, Northeastern University

Year: August 2010

Pages: 240pp.

OCLC Number: 829919535 – Find a Library: WorldCat

Language: English

Keywords: 20th Century, 21st Century | U.S. History | Prosecution: Statistics


Link: IRis (Free Access)


Author: Sami Ansari, Department of Criminal Justice, Salem State


»The initial discrepancies between the UCR and NCVS data sets decreased with time, and the two data sets are assumed to have converged for most of the crime categories. Different definitions and methods have been used by the studies that have tested and explained the convergence between the two data sets, and those studies often reported different results. The two objectives of this study include understanding and explaining the convergence between the two data sets.
The data for the study have been drawn from multiple sources. A multiple analytic strategy is used to test the convergence, and autoregression models with relevant predictor variables are estimated to explain the convergence. Graphic and correlational analyses support the convergence between the two series for all categories. However, the cointegration test indicates that the series are cointegrated for burglary and are in the process of converging for robbery and violent crime. The rate differences between the two data sets have been greatly affected by the percentages of reporting crimes to the police. Therefore, the convergence tests were repeated after adjusting the UCR rates for reporting, but the results did not differ substantially. The results of autoregressive models show that the increase in number of police officers and the methodological changes in the NCVS in 1992 are significant factors that reduced the divergence between the UCR and NCVS data sets.
The determining convergence largely depends on the definition of convergence. In this case, a perfect convergence, in which the two series overlap and move together, is neither possible nor desirable because the UCR and NCVS use nonidentical measurements to measure nonidentical sets of crimes. The study provides important research, policy, and methodological implications and suggests future research directions on the subject.« (Source: Thesis)


  Abstract (p. 2)
  Acknowledgments (p. 4)
  List of Tables (p. 8)
  List of Figures (p. 11)
  Chapter 1: Introduction (p. 12)
    Problem Statement (p. 15)
    Purpose of the Study (p. 17)
      I. Data and method (p. 18)
      II. Contributions to the literature and rationale for the study (p. 19)
      III. Limitations (p. 21)
    Organization of the Dissertation (p. 22)
    Summary (p. 23)
  Chapter 2: Literature Review–UCR and NCVS: Historical and Methodological Review (p. 24)
    Uniform Crime Reporting (UCR) (p. 25)
      I. National Incident-Based Reporting System (NIBRS) (p. 27)
    Reporting Crime to Police (p. 28)
    National Crime Victimization Survey (NCVS) (p. 31)
      I. Historical overview and objectives (p. 31)
      II. Current survey methodology (p. 33)
        A. Sampe and design (p. 34)
        B. Data collection and instruments (p. 36)
      III. Methodological changes in the NCVS (p. 38)
      IV. Methodological changes in 1992 and its effects (p. 39)
      V. Changes beyong 1992 and future directions (p. 41)
    Summary (p. 42)
  Chapter 3: Literature Review–Conception, Exploration, and Explanation of Convergence Between the UCR and NCVS: Analytical Framework and Major Hypotheses (p. 45)
    UCR and NCVS: Why Do They Diverge? (p. 46)
    Convergence between the UCR and NCVS (p. 49)
      I. Debate on methodology explaining the convergence (p. 52)
      II. Meaning of convergence (p. 55)
        A. Cross-sectional studies (p. 57)
        B. Time series studies (p. 58)
    Factors Leading to the Convergence between the UCR and NCVS (p. 60)
      I. Police productivity (p. 61)
      II. Population characteristics (p. 65)
      III. Changes in measurement (p. 66)
      IV. Attitudes of people toward crime and police (p. 66)
      V. Mobile phone use (p. 67)
    Research Hypotheses (p. 70)
    Summary (p. 73)
  Chapter 4: Data, Measurement, and Methodology (p. 77)
    Data (p. 77)
      I. Uniform Crime Reports (p. 78)
      II: National Crime Victimization Survey (p. 78)
        A. Adjustment with the pre-1993 victimization data (p. 79)
      III. Census (p. 79)
      IV. Law Enforcement Management and Administrative Statistics (p. 80)
      V. The General Social Survey (p. 80)
      VI. Cellular Telephone Industries Association (p. 80)
    Measures (p. 81)
      I. Dependent Variables (p. 81)
        A. Reasons for using only six Part I crimes and aggregated measure crime (p. 82)
        B. Reasons for using an aggregated category of property crime (p. 83)
      II. Independent variables (p. 85)
    Descriptive Statistics (p. 85)
    Analytic Strategy (p. 91)
    Summary (p. 97)
  Chapter 5: Are the UCR and NCVS Converging? (p. 98)
    Graphic Analysis of Convergence (p. 99)
    Correlational Analyses of Convergence (p. 107)
    Convergence Test through Regression of UCR and NCVS Rate Differences on Time (p. 114)
    Cointegration Analyses of Convergence (p. 116)
      I. Rate spread plots (p. 119)
      II. The ADF test of logged differences between UCR and NCVS rates (p. 119)
      III. Cointegration: Engle-Granger test (p. 121)
        A. Nonstationarity test (p. 122)
        B. ADF test of residuals of cointegration regressions between UCR and NCVS rates (p. 124)
        C. Error correction models (p. 127)
    Convergence Analysis with Adjusted UCR Rates (p. 129)
        I. Graphic analysis (p. 134)
        II. Correlational analysis (p. 134)
        III. Convergence test through regression of rate differences on time (p. 137)
        IV. Cointegration analyses of convergence (p. 138)
        A. Rate spread plots (p. 139)
        B. The ADF test of logged differences between adjusted UCR and NCVS rates (p. 139)
        C. Cointegration: Engle-Granger test (p. 141)
    Result Summary and Hypothese Testing (p. 143)
    Summary (p. 146)
  Chapter 6: Modeling Convergence: Investigation of Factors (p. 147)
    Assumptions and Data Issues (p. 148)
      I. Multicollinearity (p. 149)
      II. Autocorrelation (p. 150)
    Multivariate Results (p. 152)
    Results Summary and Hypotheses Testing (p. 157)
    Summary (p. 159)
  Chapter 7: Discussion and Conclusions (p. 185)
    Background and Summary of the Study (p. 185)
    Discussion of Findings (p. 187)
    Implications (p. 195)
      I. Policy implications (p. 196)
      II. Methodological implications (p. 198)
    Limitations (p. 199)
    Future Research Directions (p. 201)
    Conclusions (p. 202)
  Appendix A: Comparison Between the UCR and NCVS in Research (p. 205)
  Appendix B: List of Variables Explored and Used in the Study (p. 220)
    Dependent Variables (p. 220)
    Explanatory Variables (p. 223)
      Demographic Variables (p. 223)
      Police Organizational and Operational Variables (p. 224)
      Social Attitude Variables (p. 226)
      Other Variables (p. 227)
  Appendix C: Stationarity Check and Transformation (p. 228)
    Checking For Stationarity (p. 228)
    Making a Time-Series Stationary (p. 230)
  References (p. 232)

Wikipedia: National Incident Based Reporting System, Uniform Crime Reports