In this paper, we consider existing approaches to ontology definition and validation. Popular techniques include the use of domain experts or reliance on formal logic. We consider these contemporary techniques, their motivation and limitations, and then suggest an empirical approach that statistically identifies knowledge ontology within contextual databases using factor analytic techniques. We find that this method improves upon the process of identifying existing, codified knowledge ontology, and that it can be integrated into other methods to improve upon the efficiency of knowledge ontology identification, validation, and evolution. It can facilitate collaboration and inter-organizational progress by providing a common foundation, empirically supported.