The article presents a typology of statements that can be made about discourse data. The classification is based on two parameters: level of generality and level of interpretation. Each of these is operationalized into three discrete levels, to yield a nine-cell table containing types of statements going from the presentation of a single case to the elaboration of complex general theories. The discussion is based on an extensive review of published research and on an empirical study on other-repetition in child second language discourse. The proposed classification aims at clarifying a number of questions and possible misunderstandings about issues of interpretation, generalization, and quantification. It is shown that in some cases apparently opposed approaches actually make the same type of statements, simply using different terms and methodologies, whereas in other cases, the dispute is not on different methodological options, but rather on different types of statements altogether. Researchers are invited to be more explicit as regards their position in order to promote interdisciplinary dialog and facilitate cross-method comparisons.