Although the measurement of sexual harassment is still in its infancy, certain recommendations can be made based on what has been learned so far. Instruments attempting to measure sexual harassment should demonstrate the same characteristics of reliability that are required of any other psychometric technique, and validity data should be reported. It appears that asking respondents about the intent of a behavior they have experienced is useful in examining attributions or perceptions but may be problematic when collecting incidence data. The author points out that most definitions of sexual harassment are a priori in nature; that is, rather than being explicitly based on data, they are derived from theoretical propositions. A more inductive definitional strategy, one generally employed by researchers than by legal theorists, is to ask women directly if they have ever been sexually harassed and to describe their experiences. The value of the Sexual Experiences Questionnaire in measuring gender sexual harassment, seductive behavior, sexual bribery, sexual coercion, and sexual imposition is discussed. A definition of sexual harassment is suggested that is based on the sexualization of an instrumental relationship through sexist or sexual remarks, requests, or requirements in the context of a formal power differential. 40 references, 1 note, and 2 tables