NCJ Number
10357
Date Published
1972
Length
202 pages
Annotation
DEVELOPMENT AND POTENTIAL USES OF A STATISTICAL MODEL FOR ESTIMATING CRIME LEVELS IN ONE SECTION OF A CITY FROM THE LEVELS IN OTHER STATISTICALLY SIMILAR AREAS OF THAT CITY.
Abstract
THE AUTHOR POSTULATES THAT SINCE THERE ARE A NUMBER OF CRIME-RELATED INFLUENCES WHICH OPERATE IN A CITY AS A WHOLE, THE LEVEL OF CRIME IN ONE AREA OF THE CITY MIGHT BE THE FUNCTION OF THE CRIME LEVEL WITHIN ANOTHER AREA. DATA WAS GATHERED ON THE MONTHLY INDEX CRIME LEVELS OVER A THIRTY-MONTH PERIOD FOR EACH REPORTING AREA IN WASHINGTON, D.C. RESULTS OF THE DATA ANALYSIS INDICATED THAT TWO AREAS COULD BE HIGHLY CORRELATED WITH ONE ANOTHER WITH REGARD TO CRIME LEVELS. THE ESTIMATES OF CRIME DERIVED FROM THE 'CRIME-CORRELATED AREA MODEL' CAN BE USED AS A BASE LINE AGAINST WHICH TO EVALUATE IN A RETROSPECTIVE FASHION THE EFFECT OF DIFFERENT PATROL STRATEGIES. THE MODEL IS ALSO A POTENTIAL AID IN EVALUATING CRIME CONTROL PROGRAMS INCLUDING MANPOWER STUDIES, HELICOPTER PATROL PROGRAMS, AND HIGH INTENSITY STREET LIGHTING. (AUTHOR ABSTRACT)