Measures of Gun Ownership Levels for Macro-level Crime and Violence Research

• NCJ Number: 203876
• Journal: Journal of Research in Crime and Delinquency Volume: 41 Issue: 1 Dated: February 2004 Pages: 3-36
• Author(s): Gary Kleck
• Date Published: February 2004
• Length: 34 pages
• Publication Format: Article
• Publication Type: Measurement/Evaluation Device
• Language: English
• Country: United States of America

Annotation

This study assessed the criterion validity of research instruments that measure levels of gun ownership.

Abstract

Macro-level research on levels of gun ownership is crucial to establishing the impact of gun ownership on both increasing violence and decreasing violence. Researchers have speculated that increased levels of citizen gun ownership may deter criminal activity because the costs of engaging in that activity are too high, while other researchers have argued that the ease of availability of guns increases violence. However, many of these researchers are basing their assertions on research measures that have not been empirically validated. As such, the author analyzed city, State, cross-national, and time series data to establish the criterion validity of 25 macro-level indicators of gun ownership levels. The criterion measures included primarily direct survey measures of household gun ownership, such as the General Social Survey which polls households on numerous indicators. Results of statistical analyses reveal that most measures utilized in past research have poor validity, rendering past research findings uninterpretable. The findings also indicate that the best measure for cross-sectional research on gun ownership is the percentage of suicides committed with a gun. No measures were found to be valid indicators of trends in gun levels, which makes longitudinal studies on gun ownership currently impossible. Past studies in which researchers have claimed to have found a significant effect of gun ownership on crime or violence rates suffer from two critical flaws: first, they rely on invalid measures of gun ownership and second, they fail to resolve casual order issues regarding gun levels and crime or violence rates. If all this research is to be discounted, the best available research using valid measures of gun ownership indicates that there is no effect of general gun ownership on violence rates. Tables, references