Sex differences in disruptive behavior and sensitivity to social context are documented, but the intersection between them is rarely examined empirically. This report focuses on sex differences in observed disruptive behavior across interactional contexts and diagnostic status. Preschoolers (n = 327) were classified as nondisruptive (51%), clinically at risk (26%), and disruptive (23%) using parent and teacher reports on developmentally validated measures of disruptive behavior and impairment. Observed disruptive behavior was measured with the Disruptive Behavior Diagnostic Observation Schedule, a developmentally sensitive observational paradigm characterizing variation in preschoolers' disruptive behavior across two interactional contexts (parent and examiner). Repeated measures analyses of variance revealed a three-way interaction of child sex by diagnostic status by interactional context (F = 9.81, p less than 0.001). Disruptive boys were the only subgroup whose behavior was not sensitive to interactional context: They displayed comparable levels of disruptive behavior with parents and examiners. In contrast, disruptive girls demonstrated the strongest context effect of any group. Specifically, with the examiner, disruptive girls' behavior was comparable to nondisruptive boys (though still more elevated than nondisruptive girls). However, in interactions with their mothers, disruptive girls displayed the highest rates of disruptive behavior of "any" subgroup in any context, although the difference between disruptive boys and disruptive girls in this context was not statistically significant. Findings suggest the importance of sex-specific conceptualizations of disruptive behavior in young children that take patterns across social contexts into account. (Contains 2 tables, 2 figures and 2 footnotes.)