In order to better understand what structural and functional brain components changes are associated with schizophrenia, various investigations have been conducted. Functional Network Connectivity (FNC) generally interpreted as an indirect measure of brain activity, measures the functional component, and Structural Based Morphometry (SBM), an indirect measure of concentration of Gray Matter (GM), assesses the structural component. This work investigates the possibility of performing an automatic diagnosis of schizophrenia using FNC or SBM, considering each component individually and also the correlation between them. The best classification obtained for the diagnosis of schizophrenia was based on a Naïve Bayes classifier with an accuracy of 83.7%. In general, the accuracy of classifiers varied between 62% to 84% in both FNC and SBM attributes. The components that have experienced higher correlation were those related to the basal ganglia, the posterior components, motor and media components and frontal components. But we also show that the best results are obtained by considering both FNC and SBM at the same time.