Related to gathering content-based validity evidence. It involves a structured, logical review of an instrument soon after it is developed to check whether the questions/tasks/behaviors on it are a representative sample of the observable indicators that define the construct.