EXPERIMENTAL DESIGN AND CAUSAL ANALYSIS IN CRIMINAL JUSTICE EVALUATION

• NCJ Number: 53929
• Author(s): J G HEILMAN; M T MILLER
• Date Published: 1977
• Length: 14 pages
• Grant Number(s): 76-FIS-21
• Sponsoring Agency: Alabama Law Enforcement Planning Agency
• Corporate Author: Auburn University
• Publication Format: Document
• Publication Type: Report (Study/Research)
• Language: English
• Country: United States of America
• Note: PRESENTED AT THE NATIONAL CONFERENCE ON CRIMINAL JUSTICE EVALUATION

Annotation

A QUASI-EXPERIMENTAL RESEARCH DESIGN TERMED AN 'EX POST FACTO, STAGGERED, TIME SERIES DESIGN' IS DESCRIBED AND JUDGED APPROPRIATE FOR PRACTICING EVALUATORS CONSIDERING MOST CRIMINAL JUSTICE PROGRAMS.

Abstract

THIS PAPER WAS WRITTEN SPECIFICALLY FOR PRACTICING EVALUATORS PAID BY A GOVERNMENTAL AGENCY TO EVALUATE A RANGE OF PROJECTS AND PROGRAMS ON AN ONGOING BASIS, WITH RELATIVELY LIMITED FUNDING. THESE EVALUATORS NEED TO OBTAIN SUITABLE DATA AND AN APPROPRIATE ANALYTIC DESIGN WITHIN A RESTRICTED PERIOD OF TIME WITH LIMITED RESOURCES AND MOST OFTEN AFTER PROJECT ACTIVITIES HAVE COMMENCED. TO PROVIDE A SETTING FOR THE PRESENTATION, RESEARCH DESIGNS AND POSSIBLE THREATS TO THEIR VALIDITY ARE DISCUSSED. CERTAIN CHARACTERISTICS OF GOVERNMENT-FUNDED CRIMINAL JUSTICE PROGRAMS ARE FOUND TO ENCOURAGE THE USE OF QUASI-EXPERIMENTAL DESIGNS: (1) CONTINUATION AND FUNDING OF PROGRAMS FOR SEVERAL YEARS; (2) THE INTRODUCTION OF THE SAME OR SIMILAR PROGRAMS AT DIFFERENT TIMES IN VARIOUS LOCALES; AND (3) MUCH OF THE DATA NEEDED FOR EVALUATION IS AVAILABLE INDEPENDENTLY, APART FROM THE EVALUATOR'S RESEARCH EFFORTS. EX POST FACTO TIME SERIES ANALYSIS CAN BE USED ON THOSE PROGRAMS THAT HAVE BEEN OPERATING FOR A PERIOD OF TIME. THE FUNDING OF SIMILAR PROJECTS IN DIFFERENT LOCATIONS AT VARIOUS TIMES PERMITS THE USE OF STAGGERED EXPERIMENTAL GROUPS, WHICH (COMBINED WITH A CONTROL GROUP RECEIVING THE TREATMENT) PERMITS CONTROL OF HISTORICAL EVENTS AS VALIDITY THREATS. MUCH OF THE DATA REQUIRED BY EVALUATION ALSO HAS BEEN RECORDED ROUTINELY BY PUBLIC AGENCIES OVER A PERIOD OF YEARS PREDATING ANY EVALUATION EFFORT. IT IS CONCLUDED, THEREFORE, THAT AN EX POST FACTO TIME SERIES DESIGN, USING STAGGERED EXPERIMENTAL GROUPS WITH A CONTROL GROUP AND EMPLOYING DATA ALREADY GATHERED, IS AN APPROPRIATE EVALUATION DESIGN WHICH USES TO ADVANTAGE THE GIVEN CHARACTERISTICS OF GOVERNMENT-FUNDED CRIMINAL JUSTICE PROGRAMS. THE APPLICATION OF THE DESIGN TO A PROGRAM FOR IMPROVING THE PROSECUTORIAL EFFICIENCY OF DISTRICT ATTORNEYS IS DESCRIBED. A BRIEF BIBLIOGRAPHY IS PROVIDED.