Book Review: Lynch, J. P. and Addington, L. A. (Eds.) Understanding Crime Statistics: Revisiting the Divergence of the NCVS and UCR New York: Cambridge University Press, 2007. xiv pp., 340 pp

• Published date: 01 March 2010
• Date: 01 March 2010
• DOI: 10.1177/0734016809349167
• Subject Matter: Articles

Book Reviews

Lynch, J. P. and Addington, L. A. (Eds.)

Understanding Crime Statistics: Revisiting the Divergence of the NCVS and UCR New York: Cambridge University

Press, 2007. xiv pp., 340 pp.

Reviewed by: John M. Roberts, Jr., University of New Mexico

DOI: 10.1177/0734016809349167

The papers collected in Understanding Crime Statistics discuss issues in the quality and reporting of

aggregate crime data obtained from the National Crime Victimization Survey (NCVS) and the Uni-

form Crime Reporting System (UCR). In particular, analyses address various aspects of the similar-

ity or dissimilarity of the two sources’ national-level measures of the extent of non-homicide crime,

and the convergence or divergence of trends from each. While the two systems nominally measure

the same phenomena, they differ in many ways. The most fundamental difference is that the NCVS

elicits self-reports of respondents’ victimization experiences, while the UCR tabulates incidents

known to, and considered crimes by, police. Victims’ nonreporting of crime incidents to the police

would therefore contribute to differences in national crime totals produced by the two systems, as

would incidents perceived as crimes by respondents but not by police. Much research has considered

these and other reasons for differences in the two data sets’ implied national crime rates; the selec-

tive list presented here by McDowall and Loftin includes over 50 studies from the past 40 years.

Understanding Crime Statistics contributes to this literature with analyses of recent data and new

approaches to investigating differences between the NCVS and the UCR.

McDowall and Loftin take up the question of how convergence in crime rates from the two sys-

tems should be defined. Chapters by Catalano, Planty, Addington, and Rosenfeld explore various

(and sometimes rather subtle) features of the two systems that may account for discrepancies in

national totals from the two series, while Cohen and Lynch consider an analogous comparison of

the total number of emergency room visits due to assault implied by the NCVS and by surveys

of emergency room records. Maltz examines the extent of missing data in the UCR, and there are

also introductory overviews of the history and execution of the NCVS (Rennison and Rand) and

UCR (Barnett-Ryan). Finally, the editors advance the idea of ‘‘complementarity’’ in the introduction

and conclusion, arguing that discrepant national estimates do not necessarily mean that either system

is incorrect in some absolute sense.

Chapters on the impact of the two systems’ methodological features include various interesting

strategies to estimate how these features affect the systems’ comparability, and accordingly adjust

figures from one or both systems. Some steps in this regard have been taken before (such as adjust-

ment for the NCVS restriction to respondents age 12 years and older, or for acknowledged nonre-

porting to police by NCVS respondents), but these papers extend the scope and rigor of previous

work. Adjustments attempt to account for such factors as the treatment of repeat (series) victimiza-

tion in the NCVS, the UCR hierarchy rule for counting incidents with concomitant offenses and the

inclusion of commercial or organizational victims in the UCR statistics (assessed via the greater

detail available in National Incident Based Reporting System data), and the impact of the 1992

NCVS redesign on reporting of simple and aggravated assault.

Criminal Justice Review

35(1) 115-130

ª2010 Georgia State University

Reprints and permission:

sagepub.com/journalsPermissions.nav

http://cjr.sagepub.com

115