NCJ Number
197997
Journal
Prison Journal Volume: 82 Issue: 4 Dated: December 2002 Pages: 423-439
Date Published
December 2002
Length
17 pages
Annotation
This article discusses the use of data envelopment analysis (DEA) to compare technical efficiency among juvenile justice facilities in Florida.
Abstract
DEA is becoming an increasingly valuable tool for comparing the performance (often termed "benchmarking") of government organizations. DEA uses the term "decision-making unit" (DMU) as the basic unit of analysis. This can be any organizational entity as defined by the analyst. In this article, efficiency is defined as technical efficiency, i.e., the conversion of inputs into outputs. A DMU is efficient if no other DMU or some combination of DMU's can produce the same amount of outputs with fewer inputs. DEA uses deterministic linear programming to construct an efficiency frontier through the use of pair-wise comparisons of the performance of various DMU's. The primary output of the DEA model is a single scalar measure (between 1.0 and 0) of the relative efficiency of each DMU based on all the weighted input and performance variable ratios. A case study of the application of the DEA model is presented for the determination of the efficiency of juvenile halfway houses in Florida. The steps described in the analysis are the designation of the input, output, quality, and outcome variables; the inclusion of a nondiscretionary input variable and determination of the programs to be included in the study; and presentation of the study results. 2 figures, 4 tables, and appended DEA mathematical formulation