The personal space of strangers who are not attempting to influence each other has been extensively studied using single dependent measures. Multivariate personal space can be defined by three dimensions: relational, locational and interactional, and operationalized by the measures previously used singly. The present experiment tested the relation of multivariate personal space defined by shoulder orientation, seating position at or away from an occupied table, and interpersonal distance to psychological stress in the form of status differences among strangers. Results indicate adjustments in the magnitude of multivariate personal space are a function of stranger status level and that these adjustments occur in all three dimensions of multivariate personal space. The complex nature of human social behavior suggests multiple behavioral measures should be used in future research to minimize experimenter bias, maximize generalizability of results, and maximize power in hypothesis testing. (Author)