U.S. flag

An official website of the United States government, Department of Justice.

NCJRS Virtual Library

The Virtual Library houses over 235,000 criminal justice resources, including all known OJP works.
Click here to search the NCJRS Virtual Library

BOX-TIAO TIME SERIES MODELS FOR IMPACT ASSESSMENT - A COMMENT ON THE RECENT WORK OF DEUTSCH AND ALT

NCJ Number
57764
Journal
EVALUATION QUARTERLY Volume: 3 Issue: 2 Dated: (MAY 1979) Pages: 277-314
Author(s)
R A HAY; R MCCLEARY
Date Published
1979
Length
38 pages
Annotation
THE METHODOLOGY OF A STUDY THAT ATTRIBUTED REDUCTIONS IN GUN-RELATED CRIME TO PASSAGE OF A GUN CONTROL LAW IS REVIEWED CRITICALLY.
Abstract
THE STUDY IN QUESTION USED A TIME-SERIES QUASI-EXPERIMENTAL TECHNIQUE TO ASSESS THE IMPACT OF THE 1975 MASSACHUSETTS GUN CONTROL LAW ON ARMED ROBBERY, ASSAULT, AND HOMICIDE RATES IN BOSTON, MASS. IT CONCLUDED THAT THE LAW HAD RESULTED IN STATISTICALLY SIGNIFICANT DECLINES IN ROBBERY AND ASSAULT, BUT NOT IN HOMICIDE. REANALYSIS OF THE STUDY'S DATA INDICATES THAT THESE CONCLUSIONS ARE TENUOUS AND POSSIBLY THE PRODUCTS OF METHODOLOGICAL ARTIFACTS. THE MAJOR CRITICISM OF THE STUDY HAS TO DO WITH HOW THE TIME SERIES WERE MODELED. WHEN DIFFERENT MODELS WITH 'BETTER STATISTICAL PROPERTIES' ARE APPLIED TO THE SAME DATA, THE RESULTS DIFFER FROM THOSE OBTAINED BY THE ORIGINAL RESEARCHERS. A SECOND CRITICISM IS THAT THE STUDY WAS CONDUCTED ONLY 6 MONTHS AFTER THE NEW LAW WAS PASSED, TOO LITTLE TIME FOR THE INTERVENTION EFFECT TO BE MEASURED. IN ADDITION TO INDICATING THE INCONCLUSIVENESS OF THE RESEARCH IN QUESTION, THE CRITIQUE HAS SEVERAL IMPLICATIONS FOR THE USE OF TIME-SERIES ANALYSIS IN GENERAL. ONE SUCH IMPLICATION IS THAT SOPHISTICATED TIME-SERIES METHODS CANNOT OVERCOME INADEQUACIES OF DATA. THE CRITIQUE IS COUCHED IN STATISTICAL TERMINOLOGY. SUPPORTING DATA AND A LIST OF REFERENCES ARE INCLUDED. SEE NCJ-57765 FOR A RESPONSE TO THE CRITIQUE. (LKM)

Downloads

No download available

Availability