The study's focus was on the correlates of police officer and citizen discretionary decisionmaking in the course of their interactions. The dependent variables measured pertained to officer decisions to exercise techniques of order maintenance, citizen decisions to comply with officer directives, and officer decisions to make arrests. Independent variables focused on three broad levels of analysis: individual-level correlates, situation-level correlates, and community-level correlates. Data for the study were obtained from systematic social observations of interactions with citizens by beat officers and community-policing officers in Cincinnati, OH; census data; and crime data. Within the Patrol Bureau of the Cincinnati Police Department there were seven separate sections. Included in the Patrol Bureau was the Community Policing Coordinator, a lieutenant whose chief responsibilities included overall coordination of community-policing (COP) officers and department-wide planning and staffing of COP officers. All officers who participated in the study were assigned to the Patrol Bureau at the time of observation. One type of officer observed was the beat officer. Beat officers typically performed all duties associated with traditional line-level officers, particularly responding to calls for service. Their tasks included traffic enforcement and traffic accident investigations, criminal investigations and arrests of persons believed to be engaged in unlawful activity, the completion of crime reports, security checks, and the enforcement of laws related to the operation of public and licensed places. COP officers, on the other hand, were assigned to specific communities to become acquainted with citizens of their assigned neighborhoods, identify neighborhood problems, forge partnerships with citizens to develop solutions to neighborhood problems, network with local service agencies to assist in problem solving, attend community meetings, conduct security surveys, and develop initiatives to improve youth development in the neighborhood. Observations were conducted over a 1-year period and involved 33 COP officers and 18 beat officers. Data were analyzed by using two-stage weighted least square regression. The findings indicated that individual-level and community-level correlates offered little explanatory value regarding discretionary decisions in police-citizen interactions; however, some variation was observed between beat and COP officers on several situational-level correlates. Citizens were significantly more likely to comply with the demands of COP officers than the demands of beat officers. When the influences of individual, situation, and community-level correlates were controlled, citizens were more likely to acquiesce to COP officers. Individual officer characteristics exerted little influence on whether citizens complied with officer requests. Further, the influence of citizen demeanor was more significant for beat officers in the decision to make an arrest. Citizen demeanor was less important for COP officers. Thus, the analysis indicates that community-policing officers did act somewhat differently than beat officers across situational-level correlates. 38 tables, 210 references, and appended study instruments and supplementary data-analysis details