Time Aggregation and Time Lag in Macro-Level Deterrence Research

• NCJ Number: 139873
• Journal: Criminology Volume: 30 Issue: 3 Dated: (August 1992) Pages: 377-395
• Author(s): M B Chamlin; H G Grasmick; R J Bursik Jr; J K Cochran
• Date Published: 1992
• Length: 19 pages
• Publication Format: Article
• Publication Type: Research (Theoretical)
• Language: English
• Country: United States of America
• Note: Revised version of paper presented at the annual meetings of the American Society of Criminology, 1991, San Francisco

Annotation

Recent arguments in research on individual-level deterrence suggest that the effect of perceived sanction risk on illegal behavior may occur over a shorter time period than yearly lags typically incorporated in panel studies, and this study raises the same issue for macrolevel deterrence research by suggesting that panel studies using yearly data may have failed to capture the deterrent effect.

Abstract

The analysis uses ARIMA (autoregressive integrated moving average) models, with data aggregated monthly, quarterly, and semiannually to estimate reciprocal effects between arrests and crimes for robbery, burglary, larceny, and auto theft in Oklahoma City from 1967 through 1989. Findings do not indicate an effect of crimes on arrests, although significant effects of arrests on crimes appear for three of the four offenses in shorter temporal aggregations. These results suggest the need to reconsider conclusions based on panel studies that have used time aggregations and time lags that may be too long to uncover deterrent effects. Ultimately, deterrence must be examined in appropriate spatial and temporal aggregates over appropriate time lags for meaningful behavior categories. 53 references and 2 tables