NCJ Number
13543
Date Published
1974
Length
21 pages
Annotation
CONSTRUCTION AND VALIDATION OF AN EXPERIENCE TABLE USED BY THE UNITED STATES BOARD OF PAROLE TO ASSESS AN APPLICANT'S PAROLE PROGNOSIS.
Abstract
THE THREE SAMPLES USED TO CONSTRUCT THIS INSTRUMENT INCLUDED A CONSTRUCTION SAMPLE CONSISTING OF 25 PERCENT OF ALL PERSONS RELEASED FROM FEDERAL PRISONS BY PAROLE, MANDATORY RELEASE, OR EXPIRATION OF SENTENCE DURING THE FIRST SIX MONTHS OF 1970. TWO VALIDATION SAMPLES WERE ALSO USED, THE FIRST CONTAINING AN ADDITIONAL 25 PERCENT OF THE PERSONS RELEASED DURING THE SAME TIME PERIOD AND THE SECOND CONSISTING OF A SIMILAR 20 PERCENT SAMPLE OF PERSONS RELEASED DURING THE SECOND SIX MONTHS OF 1970. BACKGROUND DATA INCLUDING INFORMATION ABOUT OFFENSES, AGE, EDUCATION, PRISON CONDUCT, AND EMPLOYMENT RECORD WERE COMPLETED FOR EACH INDIVIDUAL. IN ADDITION, INFORMATION ABOUT PERFORMANCE AFTER RELEASE WAS CODED, USING A TWO YEAR FOLLOWUP FROM DATE OF RELEASE FOR EACH INDIVIDUAL. EACH WAS MEASURED IN TERMS OF FAVORABLE OR UNFAVORABLE OUTCOME. NINE ITEMS SELECTED TO PREDICT PAROLE OUTCOME WERE PRIOR CONVICTIONS, PRIOR INCARCERATIONS, AGE AT FIRST COMMITMENT, AUTO THEFT, PAROLE REVOKED, DRUG HISTORY, GRADE CLAIMED TO HAVE BEEN COMPLETED, EMPLOYMENT, AND LIVING ARRANGEMENT. THIS INSTRUMENT WAS USED TO PREDICT A SCORE FOR EACH INDIVIDUAL IN THE CONSTRUCTION SAMPLE AND THEN PRESENTED TO THE RESEARCH COMMITTEE OF THE PAROLE BOARD AND TO THE FULL BOARD FOR OPERATIONAL USAGE WITH SEVERAL MINOR MODIFICATIONS.