NCJ Number
63203
Date Published
January 1979
Length
8 pages
Annotation
PANEL MODELS FOR CRIME RATES AND ARREST RATES, USED TO SEPARATE THE EFFECT OF LAW ENFORCEMENT FROM SEVERAL OTHER PROCESSES ADVANCED AS POSSIBLE DETERMINANTS OF AN ENFORCEMENT-CRIME RELATIONSHIP, ARE DISCUSSED.
Abstract
MOST STUDIES OF THE EFFECT OF LAW ENFORCEMENT ON CRIME HAVE EMPLOYED CROSS-SECTIONAL DESIGNS. THE MORE SOPHISTICATED OF THESE STUDIES HAVE USED SIMULTANEOUS EQUATION METHODS FOR SEPARATING THE EFFECT OF CRIME ON PUNISHMENT FROM THE EFFECT OF PUNISHMENT ON CRIME. HOWEVER, IN THESE ANALYSES, LAGGED EFFECTS ARE ABSORBED INTO LATER MEASURES OF ENFORCEMENT AND CRIME. ANOTHER APPROACH USES PANEL MODELS WHERE DATA ARE COLLECTED FOR A NUMBER OF SPATIALLY DISTRIBUTED UNITS, E.G., CITIES, STATES AT MORE THAN ONE TIME, AND THE RECIPROCAL SHORT TERM AND LONG TERM EFFECTS OF CRIME AND ENFORCEMENT ARE ESTIMATED. GIVEN THE PRESENT STATE OF KNOWLEDGE, THE MULTI-WAVE PANEL MODEL SEEMS HIGHLY PLAUSIBLE, AND OFFERS THE MAJOR ADVANTAGE OF ALLOWING THE DATA TO PLAY A LARGER ROLE IN ESTIMATION. THIS ANALYSIS IS BASED ON REPORTED INDEX OFFENSE RATES AND ARREST CLEARANCE RATES FOR 98 CITIES FOR THE YEARS 1964 THROUGH 1970. DATA ON POPULATION SIZE, NUMBER OF OFFENSES KNOWN TO POLICE, AND NUMBER OF CRIMES CLEARED BY ARREST WERE SUPPLIED BY THE FBI. FOR EACH CITY, YEAR, AND FELONY, CRIME RATE WAS DEFINED TO BE THE NUMBER OF OFFENSES KNOWN TO POLICE FOR THAT YEAR DIVIDED BY THE POPULATION ESTIMATE; CLEARANCE RATE WAS DEFINED AS THE RATIO OF OFFENSES CLEARED BY ARREST TO THE NUMBER OF OFFENSES KNOWN TO POLICE FOR THAT FELONY THAT YEAR. EXAMINATION OF THE 1, 2, AND 3-YEAR TIME LAG MODELS SHOWED NO MEANINGFUL RELATIONSHIP BETWEEN ARREST RATES AND CRIME RATES. ALTHOUGH THERE ARE A NUMBER OF WAYS TO INTERPRET FINDINGS CONSISTENT WITH THE DETERRENCE DOCTRINE, THE MOST PLAUSIBLE INTERPRETATION IS THAT THE DETERRENCE AFFECT IS NEGLIGIBLE. FOOTNOTES, REFERENCES, TABULAR DATA, AND FIGURES ARE INCLUDED IN THE ARTICLE. (LWM)
Date Published: January 1, 1979