abstract
- This study aims to evaluate the innovation system of the metropolitan statistical areas (MSAs) in the USA through the application of a two-stage data envelopment analysis model. Our two-stage model permits analyzing the process of creating solutions and transforming those solutions into an economic benefit sequentially; then, we applied the K-means algorithm, identified similarities among the MSAs under analysis, and categorized them according to their results in each model stage. This allowed us to determine the focus and opportunities for each region depending on their category. The analysis revealed that most MSAs with the best innovation efficiency are located on the West Coast, the Northeast, the Midwest, and Texas. Furthermore, by comparing the results from the proposed data envelopment analysis model to other studies¿ outcomes in the context of innovation in the USA, we were able to test the robustness of the model itself. The evaluation indicates that the results are not distant from other methods¿ outcomes, and the model could serve to analyze and compare regional innovation systems globally. © 1988-2012 IEEE.