In an effort to measure the effectiveness of crime deterrents and to estimate crime rates, calls for assistance placed to police in Oklahoma City over a two-year period were enumerated. This type of call was studied in order to circumvent problems such as "interviewer's effect" and sampling errors that occur with other methods. The telephone calls were stratified by police district, allowing for analysis on the neighborhood level to determine whether deterrence operates ecologically--that is, by neighbors informing one another about arrests which took place as a result of their calls to the police. In measuring deterrence, only the calls that concerned robbery were used. To estimate crime rates, calls were tallied on a monthly basis for 18 types of offenses: aggravated assault, robbery, rape, burglary, grand larceny, motor vehicle theft, simple assault, fraud, child molestation, other sex offenses, domestic disturbance, disorderly conduct, public drunkenness, vice and drugs, petty larceny, shoplifting, kidnapping/hostage taking, and suspicious activity.