NCJ Number
14083
Journal
University of Chicago Law Review Volume: 41 Issue: 2 Dated: (WINTER 1974) Pages: 224-241
Date Published
1974
Length
18 pages
Annotation
A REANALYSIS OF THE EVALUATION OF THE VERA INSTITUTE OF JUSTICE MANHATTAN COURT EMPLOYMENT PROJECT, INDICATING SERIOUS WEAKNESS IN THE PREVIOUS EVALUATIONS OF THAT PROJECT AND OFFERING AN ALTERNATIVE EVALUATION METHOD.
Abstract
TWO EVALUATION METHODS ARE OUTLINED - THE RANDOM ASSIGNMENT WHICH USES AN EXPERIMENTAL CONTROL GROUP, AND THE MATCHING METHOD, IN WHICH THE RESEARCHER COMPARES THE EXPERIMENTAL GROUP TO A GROUP OF UNTREATED INDIVIDUALS WHOSE CHARACTERISTICS APPEAR TO MATCH THOSE OF PERSONS IN THE TREATMENT GROUP. THE ORGINAL EVALUATION OF THE MANHATTAN COURT EMPLOYMENT PROJECT (CEP) EMPLOYED THE MATCHING METHOD. DEFICIENCIES IN THE MATCHING OF THE EXPERIMENTAL GROUP TO THE NON-TREATMENT GROUP ARE EXPLAINED, AND STATISTICS ARE PROVIDED WHICH COMPARE RESULTS OF THE TREATMENT GROUP TO A DIFFERENT AND MORE ACCURATELY MATCHED NON-TREATMENT GROUP. THE RESULTS OF THIS COMPARISON ARE LESS POSITIVE THAN THE RESULTS OF THE ORIGINAL EVALUATION. TO AVOID THE DIFFICULTIES OF THE MATCHING EVALUATION METHOD, IT IS STATED THAT THE EXPERIMENTAL METHOD SHOULD BE EMPLOYED. ETHICAL OBJECTIONS TO THIS ARE COUNTERED, AND CRITICISMS OF THE TENDENCY TO TREAT ONLY 'BEST RISK' CASES ARE PRESENTED.