The effect of principal components analysis improving discrimination power on data envelopment analysis process.
Data Envelopment Analysis (DEA) is one of the analysis techniques widely used in the evaluation of Decision Making Units performance based on multiple inputs and multiple outputs. Besides to its many advantages, the discrimination power of the analysis declines in DEA in case of the number of total input and output is relatively higher in comparison with the decision unit number. For a particular number of decision units, efforts to increase sensitivity by removing some input and outputs from the model cause to loss of information that those input or output have. Instead of this, the use of Principal Component Analysis (PCA) as one of the multivariate statistical analyses techniques for data reduction makes great contributions in analysis process. In this study, the role of PCA in DEA analyses will be presented with the case of efficiency measurement at the foundation universities in Turkey.