A decomposition of productivity measures for Bangladesh universities

Title
A decomposition of productivity measures for Bangladesh universities
Publication Date
2010
Author(s)
Zaman, Parisa
Villano, Renato
( author )
OrcID: https://orcid.org/0000-0003-2581-6623
Email: rvillan2@une.edu.au
UNE Id une-id:rvillan2
Fleming, Euan
Type of document
Conference Publication
Language
en
Entity Type
Publication
Publisher
International DEA Symposium website
Place of publication
Online
UNE publication id
une:9368
Abstract
The efficiency of universities has attracted strong attention in recent years, not only due to the ever-increasing public funding made available to the universities, but also because universities play a significant role in shaping the future of any nation. In this study, we measure the performance of forty universities in Bangladesh for the period 2003 to 2008 using data envelopment analysis. Using the DPIN program, productivity measures are decomposed into changes in technology, output and input technical efficiency, output and input scale efficiency, and output and input mix efficiency. Total factor productivity was found to have varied widely across the sample. Our preliminary results confirm that universities achieved mean output and input technical efficiency measures for the whole period of 0.72 and 0.78, respectively. Mean output and input scale efficiency measures were 0.83 and 0.76, and mean output and input mix efficiency measures were 0.88 and 0.42, respectively. The relatively low mean input mix efficiency score probably reflects the inflexibility that universities have in adjusting their input mixes to changing student enrolments. Our results also suggest that public universities performed better on average than private universities on all efficiency scores over the study period. This study constitutes a preliminary analysis of efficiency change of Bangladeshi Universities, and we have some concerns with data quality and the availability of data on output variables that are currently being examined with the aim to provide a more suitable data set for analysis in the future.
Link
Citation
International DEA Symposium Program and Abstracts, p. 12-12
Start page
12
End page
12

Files:

NameSizeformatDescriptionLink