INTERPRETING OFFICIAL RECORDS AS INDICATORS OF RECIDIVISM IN EVALUATING DELINQUENCY PREVENTION PROGRAMS

• NCJ Number: 57907
• Journal: Criminology Volume: 15 Issue: 3 Dated: (NOVEMBER 1977) Pages: 397-424
• Author(s): J D HAWKINS; C H CASSIDY; N B LIGHT; C A MILLER
• Date Published: 1977
• Length: 29 pages
• Grant Number(s): 1449
• Sponsoring Agency: US Dept of Justice; King Cty Dept of Planning and Community Development
• Corporate Author: Sage Publications, Inc
• Publication Format: Article
• Language: English
• Country: United States of America

Annotation

PROBLEMS IN USING POLICE AND COURT RECORDS AS SOURCES OF RECIDIVISM INFORMATION IN THE EVALUATION OF CRIME REDUCTION AND PREVENTION PROGRAMS ARE ILLUSTRATED IN A STUDY OF A KING COUNTY, WASH., DELINQUENCY PROGRAM.

Abstract

THE SUBJECT OF THE STUDY WAS A SYSTEM OF EIGHT YOUTH SERVICE BUREAUS IN KING COUNTY. A RANDOM SAMPLE OF 50 PERCENT OF THE BUREAUS' CLIENTS (510) WAS SELECTED, AND DATA ON THE SAMPLE'S POLICE AND COURT CONTACTS WERE OBTAINED FROM THE COUNTY DEPARTMENT OF PUBLIC SAFETY, THE JUVENILE COURT, AND SIX POLICE DEPARTMENTS WITH JURISDICTION OVER THE TARGET AREAS OF THE BUREAUS. DEPENDING ON WHICH SOURCE OF OFFICIAL DATA WAS USED, WHAT MEASURE OF RECIDIVISM WAS CHOSEN, AND HOW THE FINDINGS WERE PRESENTED, CONCLUSIONS REGARDING HOW SUCCESSFUL THE BUREAUS WERE IN PREVENTING DELINQUENCY VARIED WIDELY. FOR EXAMPLE, IT WAS POSSIBLE TO PRESENT STATISTICS SHOWING THAT FROM 0.2 PERCENT TO 12.7 PERCENT OF THE SAMPLES WERE RECIDIVISTS. FOCUSING ON RECIDIVISM RATES, THE EVALUATORS COULD SHOW RATES RANGING FROM 9.7 PERCENT TO 34.2 PERCENT. IF THE FOCUS WAS ON TOTAL POST-INTAKE DELINQUENCY OF CLIENTS (REGARDLESS OF PREINTAKE DELINQUENCY), IT COULD BE SHOWN THAT FROM 1.2 PERCENT TO 25.4 PERCENT OF THE CLIENTS WERE DELINQUENT AFTER INTAKE. INTERPRETATION AND PRESENTATION CHOICES ALSO DETERMINED WHICH SOCIODEMOGRAPHIC AND SERVICE-RELATED VARIABLES APPEARED TO BE RELATED TO DELINQUENCY. ONLY SEX AND SOURCE OF REFERRAL WERE CONSISTENLY RELATED TO RECIDIVISM REGARDLESS OF THE MEASURE USED. IT IS CONCLUDED THAT, IF DELINQUENCY REDUCTION PROGRAMS ARE TO BE COMPARED ON THE BASIS OF OFFICIALLY RECORDED RECIDIVISM, EVALUATORS OF ALL SUCH PROGRAMS MUST EITHER USE MULTIPLE MEASURES OF RECIDIVISM OR ADOPT A CONSISTENT MEASURE AND APPLY IT ACROSS THE BOARD TO ALL OFFICIAL RECORDS. OTHERWISE COMPARISONS OF RECIVIDISM RATES AMONG DIFFERENT PROGRAMS WILL BE LITTLE MORE THAN COMPARISONS OF RESEARCHERS' METHODS. SUPPORTING DATA AMD A LIST OF REFERENCES ARE INCLUDED. (LKM)