NCJ Number
46360
Date Published
1976
Length
13 pages
Annotation
REEXAMINATION OF A TIME-SERIES STUDY BY ISSAC EHRLICH DEMONSTRATES THAT THE TECHNICAL APPROACH AND DATA USED DO NOT PERMIT ANY INFERENCE ABOUT THE DETERRENT EFFECT OF THE DEATH PENALTY.
Abstract
IN 1975 ISSAC EHRLICH CLAIMED TO HAVE DISCOVERED A NEGATIVE RELATIONSHIP BETWEEN REPORTED MURDER RATES AND EXECUTION RATES OF CONVICTED MURDERERS. THE FIRST SECTION OF THIS PAPER IS DEVOTED TO A DETAILED EXAMINATION OF THE MATHEMATICAL MODEL EHRLICH USED. THEORY SUGGESTS THAT THE STATISTICAL FORM CHOSEN SHOULD CLOSELY CONFORM TO BEHAVIORAL EXPECTATIONS. IN GENERAL, LINEAR PARAMETERS ARE USED TO MAKE IT EASIER TO INTERPRET THE STATISTICAL PROPERTIES OF THE ESTIMATOR WHILE THE LOGARITHMIC FORM IS CHOSEN TO INTERPRET LINEAR PARAMETER ESTIMATES AS PARTIAL ELASTICITIES. THE LOG-LINEAR FORM IS USUALLY JUSTIFIED BECAUSE DATA IS EXPECTED TO EXHIBIT CONSTANT ELASTICITIES IN THE RELEVANT RANGE OR THE DATA IS MORE CONSISTENT WITH THIS FORM. IN THE EHRLICH STUDY NEITHER OF THESE JUSTIFICATIONS ARE VALID. IN FACT, THE JUSTIFICATION BECOMES ABSURD WHEN THE ESTIMATED PROBABILITY OF EXECUTION DROPS TO ZERO IN 1968. A SERIES OF EQUATIONS ARE PRESENTED TO SHOW THAT CHOICE OF THIS FORM RESULTS IN DATA WHICH INDICATES UNUSUAL SENSITIVITY BETWEEN CAPITAL PUNISHMENT AND CRIME RATES, DATA NOT CONFIRMED BY ACTUAL EXPERIENCE. ALSO, THERE IS NO CORRECTION FOR ERROR KNOWN TO EXIST IN FEDERAL BUREAU OF INVESTIGATION CRIME DATA COLLECTED BEFORE 1960. FOR THESE REASONS IT IS CONCLUDED THAT THE EHRLICH RESEARCH CONTRIBUTES LITTLE TO THE CAPITAL PUNISHMENT DETERRENCE ISSUE. TABULAR DATA, GRAPHS, AND EQUATIONS ARE INCLUDED. AN APPENDIX LISTS THE VARIABLES USED IN THE DATA SOURCES. A BIBLIOGRAPHY AND NOTES ARE PROVIDED. (GLR)