Police often attempt to restore order and prevent illegal activity by calling on citizens to alter their behavior. Achieving compliance in these circumstances is an important test of officers' skills and an essential element of effective governance. This article assesses citizen compliance with specific police requests for orderly, legal behavior in 346 encounters observed in Richmond, Virginia. Citizens were compliant in 78% of the encounters. The effects on compliance of several potential influences are estimated in a logistic regression model: instrumental factors concerning the calculation of outcomes for the citizens, factors about the legitimacy of the police intervention, personal characteristics reflecting the citizens' social status and predisposition to compliance, and the officers' skill and work orientation. Significant effects are found in each category of variables, but the results are in some cases contrary to expectations. Legitimating factors, citizens' social status, and police skill and work orientation show particularly strong effects. Implications of these findings for future research are discussed.

Community policing creates the expectation that oficers will become more selective in making arrests and that those decisions will be influenced more by extralegal considerations and less by legal ones. Data on 451 nontraffic police‐suspect encounters were drawn from ridealong observations in Richmond, Virginia, where the police department was implementing community policing. The arrest/no arrest decision is regressed on variables representing legal and extralegal characteristics of the situation. Legal variables show much stronger effects than extralegal ones, but that depends upon the officer's attitude toward community policing. Supporters of community policing are, as predicted, more selective in making arrests and much less influenced by legal variables than are officers with negative views. However, pro‐community‐policing officers are like negative officers in the extent of influence exerted by extralegal factors. There are some differences between the two groups of officers on the strength and direction of effects of predictor variables taken individually, but only 1 of 17 is significant. Thus, in a time of community policing, officers who support it do manifest some arrest decision patterns distinguishable from those of colleagues who adhere to a more traditional view of law enforcement.

This study examines how patrol officers respond to citizens' requests that officers control another citizen—by advising or persuading them, warning or threatening them, making them leave someone alone or leave the scene, or arresting them. Data are drawn from field observations conducted in Indianapolis, Indiana, in 1996 and St. Petersburg, Florida, in 1997. Officers granted the request for the most restrictive form of control requested by the citizen in 70% of the 396 observed cases. Several factors were modeled to determine their influence on officers' decisions to grant or deny the most restrictive request. These factors include legal considerations, need, factors that attenuate the impact of law or need, the social relationship between the requester and target of control, and personal characteristics of the officer. Multivariate analysis shows that the most influential factors were legal considerations. When citizens requested an arrest, the likelihood that the police would be responsive dropped considerably. However, as the evidence of a legal violation against the targeted citizen increased, so did the odds of an arrest. Officers were less likely to grant the requests of citizens having a close relationship with the person targeted for control, disrespectful of the police, or intoxicated or mentally ill. The race, wealth, and organization affiliation of citizen adversaries had little impact on the police decision. Male officers, officers with fewer years of police experience, and officers with a stronger proclivity to community policing, had significantly greater odds of giving citizens what they requested. The implications of the findings for research and policy are discussed.

This article drew on expectancy theory in industrial/organizational psychology to explain arrest productivity for driving under the influence (DUI) in a sample of Pennsylvania police officers. Expectancy theory is a cognitive model of motivation and performance based on workers' perceptions of their situation. Its major elements are estimated in a regression model: the officer's capability and opportunity for DUI enforcement (performance-reward expectancy), the instrumentality of DUI enforcement behavior for the officer, and the reward-cost balance associated with making DUI arrests. These factors account for 26% of the residual variance in the number of DUI arrests made annually once organizational effects have been removed. The relationships revealed are as expectancy theory predicts, except for instrumentality variables, which show a negative relationship to arrest productivity. This is due largely to the orientation of a small number of “rate busters,” whose exceptionally high arrest rate and negative attitudes toward their peers and the department hierarchy make them a distinct group accounting for a disproportionate number of arrests.

No abstract