It is customary in the empirical trade literature to analyze specialization patterns of countries using Revealed Comparative Advantage (RCA) measures. This paper explores the pros and cons of the most commonly used RCA index: the Balassa (1965) Index of RCA. After describing the properties of the Balassa Index, we analyze two different normalizations of the original Index recently proposed by Laursen (2000) and Proudman and Redding (1998) to remedy some of the shortcomings of the original index, and we propose an alternative and more traditional strategy in order to analyze the dynamics of specialization.