Studies of early schizophrenia offer several advantages for efforts to unravel the biology, treatment, and outcome of this disorder. A review of available first-episode studies, however, suggests marked variability in findings, frequently attributed to the presumed heterogeneity of the disorder. Another cause may be variation in definitions and criteria used. Thus, attention has turned to the study of patients experiencing their first episode of psychosis. Although this strategy will not control for the true underlying heterogeneity of the illness, it provides a valuable method for homogenizing variability due to course. For this strategy to be effective, definition and operationalization of criteria for such studies are crucial. In this article, we reviewed assessment methods used in first-episode studies in the recent literature and found marked inconsistencies. We offer a framework for definition of patient populations for studies of early schizophrenia and outline some key variables that may serve as the basis for fuller characterization of the disorder. It is important to provide data on such variables to enhance comparability among studies of early schizophrenia and to facilitate meaningful interpretation of data.