The study population consisted of mental health employees who worked in Western Pennsylvania. The population, as defined by the Department of Public Welfare-Office of Mental Health under Title 55 of the Pennsylvania Code, was composed of mental health workers who were employed in licensed mental health facilities, including outpatient, inpatient, partial hospitalization, crisis, family-based, long-term structured residence, and community residential rehabilitation. Based on the sampling frame, facilities were limited to those that were licensed, which excluded employees who worked in private practice. The final sample size was 449, with 162 working in nonresidential facilities, 157 in residential facilities, and 130 in crisis facilities. In order to test the hypotheses, a questionnaire was designed to obtain information on employees' routine activities and patterns of criminal victimization. The questionnaire was administered to determine the relationship between the three central elements of routine activities theory -- exposure to potential offenders, guardianship, and target suitability -- and victimization at work. Verbal aggression, verbal threats, and physical attacks were examined over the past 12 months, with attention to the frequency as well as the nature of the aggression. Findings show that those employees who viewed more of their weekly client contacts as dangerous were more likely to experience victimization. Type of mental disorder, number of weekly client contacts, and setting provided weak or no support for the alternative hypotheses, in that the variables did not have a significant impact on victimization. Those who worked in crisis facilities were more likely to experience victimization than those who worked in residential and/or nonresidential facilities. Working evening or night shifts was associated with an increased risk of victimization. Males were more likely to be victimized than females, except in types of sexual assault. Some of these findings provide strong support for routine activities theory, as well as rich data on violence in the field of mental health. Several policy or practical implications are drawn from the study. 14 tables, 143 references, and a subject index