Corrected normalized additive analysis as alternative method for easy measurement of efficiency

Various best-practice benchmarking methods for measuring the efficiency of decision making units are intensively discussed in the literature. Compared with such methods as Data Envelopment Analysis (DEA), the comparatively straightforward Normalized Additive Analysis (NAA) has received little attention so far. Nevertheless, a recent study revealed that NAA performs similarly to established methods regarding the accurate determination of the ranking of decision making units based on their efficiency scores. However, NAA systematically underestimates the true efficiency scores with respect to the underlying production system and the chosen Farrell efficiency. To counteract this underestimation, we present two modifications of NAA in the sense of a Corrected NAA (CNAA). We conduct a Monte Carlo simulation to analyze the performance of these modifications. For our simulation design, we show that these CNAA specifications estimate the efficiency scores more accurately than NAA while determining the ranking of decision making units at an equally good level. To further analyze the estimation quality of CNAA, we use the well-known basic DEA models with constant and variable returns to scale as benchmarks. It is shown that the performance of the CNAA specifications is largely comparable to that of the DEA models. Hence, CNAA can be considered to be a suitable method.