Concurrent with the public outcry of recent years to improve the quality of America's schools, has come a demand for accountability in public education. This study employs Data Envelopment Analysis (DEA) as a measure of technical efficiency in the allocation of limited resources in eighty-one non-rural public elementary schools in Western Massachusetts. Three major research questions guided the study: (1) How do the selected elementary schools differ as to the degree of inefficiency when compared with each other? (2) What factors may account for differences in expected achievement among relatively efficient schools? (3) What factors may account for differences in relative efficiency scores? Data were collected for four outputs representing student achievement and for sixteen inputs representing a balance of school, student and teaching resources. Preliminary analysis reduced the number of inputs to eight. DEA results indicated that of 81 schools, 37 (or 46%) were found to be efficiently utilizing their resources and 44 (or 54%) were found to be inefficient to varying degrees. DEA provided for each school a relative efficiency index, an identified peer set of efficient schools, optimal weights assigned to inputs and outputs, and estimates of the augmentations in outputs and/or the reductions in inputs (i.e., slack values) that could be attained if efficiency were to be achieved. Since the DEA analysis results identified the sources and degree of inefficiency, the factors could be adjusted to remove these inefficiencies and thus the variables which influenced student achievement could be determined; four inputs were found to be significant. Five inputs (representing three areas of resources) were identified as contributing most to differences in relative efficiency scores by being overconsumed, or underutilized, in a significant number of schools. The study concludes that the strength of DEA lies in its ability to identify empirically-based sources and amounts of inefficiencies in a multiple outputs-multiple inputs settings. Limitations exist primarily in the availability of data for outputs and inputs. Finally, DEA can add significantly to renewal at the school level by providing school decision makers with the tools to make valuable and effective choices.
| Identifer | oai:union.ndltd.org:UMASS/oai:scholarworks.umass.edu:dissertations-7753 |
| Date | 01 January 1990 |
| Creators | Zomorrodian, Mohammad Reza |
| Publisher | ScholarWorks@UMass Amherst |
| Source Sets | University of Massachusetts, Amherst |
| Language | English |
| Detected Language | English |
| Type | text |
| Source | Doctoral Dissertations Available from Proquest |
Page generated in 0.0121 seconds