Integrated Multi-Criteria Decision Making – Data Envelopment Analysis Models In Efficiency Analysis Of Sponsored Research
Loading...
Date
2018-09
Authors
Noor Saifurina Nana Khurizan
Journal Title
Journal ISSN
Volume Title
Publisher
Universiti Sains Malaysia
Abstract
In the context of higher education administration, specifically for sponsored research, efficiency analysis has become an essential process for both sponsors and researchers. The evaluation process will provide the information on research productivity in fully utilising the resource to produce the desirable outcome. It is easier to evaluate the sponsored research’ performance when its efficiency is represented by a single score. Over the years, many methodologies have contributed into assessing the performance of sponsored research. Among the most popular ones are Multi-Criteria Decision Making (MCDM) Method and Data Envelopment Analysis (DEA). MCDM and DEA method both deal with problems concerning multiple inputs and outputs. This factor is crucial in efficiency analysis as most case studies are presented with more than one resource and outcome. The main objective of this study is to evaluate the efficiency of sponsored research in a university subject to its data availability and number of Decision Making Units (DMUs). The integrated MCDM-DEA methods are presented in order to fulfill the requirement of the main objective. The first case study presents a novel way of integrating three different MCDM and DEA model to evaluate the performance of a sponsored research from a much simpler form of data in a small size under the same grant. The integrated model has managed to handle the subjective data and objective data in producing the single score for each sponsored research. The issue of low discrimination power of DEA is also tackled using the three-stage model. This study continues to further investigate a data set containing larger number of DMUs from one grant for the purpose of investigating the sponsored research performance according to its discipline of study. During evaluation process, the issue of non-homogeneity characteristics of the sponsored research has also come to light especially regarding its research’ discipline which is supported by two statistical analysis. Hence a refined algorithm, properly defined in six steps, is suggested to tackle the issue. The algorithm is refined such that it is applicable to similar case studies. The issue of non-homogeneity characteristics of sponsored research may raise the need into supplementary analysis on the effect of multiple environmental variables on the research’ performance. The presence of multiple environmental variables are investigated when the data set is from different types of grants that operate under multiple environments. An integrated MCDM-DEA method is proposed to tackle the need of evaluating the sponsored research performance under the existence of multiple environmental variables. The method suggested also managed to rank the university’s school based on the performance of its sponsored research. The school’s academic strength on research can be identified and may be of use in evaluating its overall performance. Each distinctive integrated model presented in this study provides a platform to evaluate the sponsored research’ efficiency in accordance to the sizes and information availability of its data. One may select the best model to accommodate the characteristics of the data.
Description
Keywords
Mathematics