Relevant Ingredients for Identifying Factors With Significant Impact on Research Structures Efficiency in Higher Education

Abstract

This work proposes relevant ingredients to highlight the factors with significant impact on efficiency of research structures in higher education. The ingredients in question include methods and their implementation taking into account the choice and operationalization of factors as well as options of the retained methods. The methods employed in this work are Data Envelopment Analysis and Multiple Linear Regression. The originality of this work lies in its intermethodology, as efficiency scores which are determined by Data Envelopment Analysis are considered as the dependent variable of the Multiple Linear Regression. Its originality resides also in dealing with the Multiple Linear Regression methods that are little approached in the works which use this technique and in combining parametric and nonparametric methods. This work has social and research implications, as research structures feel that research subsidies they receive depend on their outcomes and leads them to privilege factors with great impact on their efficiency. This work has shown that to determine the factors that have a significant impact on the research structures efficiency, research activities proposed by the international research institutions and contingency factors of the research structures must be taken into account. On the other side, the regression methods used to highlight the aforementioned impact become outmoded and needed some improvements. Finally, for an appropriate measure of research activities, the ideal is to integrate Data Envelopment Analysis and Multiple Linear Regression, especially as linear programming has already been integrated in Multiple Linear Regression.