Using Calls For Service Data To Reduce False Burglar Alarms (And More)
Open police data is often at its best when it is big, messy, and gives an unfiltered view into the complexities of policing, which is why Calls for Service data can be so wonderful.
Calls for Service data reflect incidents captured in an agency’s Computer-Aided Dispatch (CAD) system. Each row represents one incident that law enforcement, the fire department, and EMS respond to ranging from reports of a suspicious person to murder and everything in between.
Often times these data sets capture information on the minutiae of crime and policing in a jurisdiction. This includes an officer’s response time, where the call took place, what priority the call was, and the incident’s disposition, or what happened when an officer arrived. The beauty of Calls for Service data is in how citizens and departments alike can evaluate a wide array of trends in near real time.
Take burglar alarms, for example. Burglar alarms are important to respond to in case something is wrong, but the vast majority are false alarms, which drain thousands of hours of working time.
The New Orleans Police Department (NOPD) has been publishing calls for Service data since 2015, and when they used that data to analyze burglar alarms, they found the department responded to 48,000 false alarms in 2014. With this compelling data in hand, the City of New Orleans successfully advocated for a false alarm ordinance that went into effect in June 2017. The ordinance is designed to reduce false alarms by requiring owners to register their alarms and handing out fines for false alarms.
But the role of open data doesn’t stop now that we’ve got the ordinance! Thanks to the Calls for Service data available on New Orleans’ open data portal, we also have a way to measure the results.
The New Orleans open data portal provides information on over 300,000 burglar alarm incidents from 2011 to present. I downloaded all burglar alarm Calls for Service over that timeframe and counted the number of incidents per day. Over the first six months of 2015, just before the ordinance was passed, there was an average of 138 burglar alarm calls per day. In the six months before the ordinance took effect, there were 108 burglar alarm calls per day. Since the ordinance has been in effect, there has been an average of 89 burglar alarm calls per day. This decline suggests that the ordinance is achieving its intended effect.
But wait, there’s more that can be done with the Calls for Service data.
New Orleans’ Calls for Service data include the time of dispatch as well as the time an incident was closed. Using those two measures, I calculated that the average burglar alarm call for service takes approximately 19 minutes of officer time. Relative to 2015’s total, we can expect a 35 percent reduction in burglar alarm responses if the current decline continues. Combining those two facts works out to a remarkable savings of 5,806 hours -- or 242 days -- of work for police officers.
Calls for Service data is available for 28 of the PDI’s current 136 jurisdictions, with Cincinnati, Detroit, Orlando, and Richmond (California) providing some of my favorite examples. Detroit’s, for example, breaks down how long each incident took at every step along the way - intake, dispatch, officer travel time, and time on scene handling the incident. There is a ton that can be done with richly detailed information like that!
Calls for Service data represent the front line of police interactions with the public, making these data sets especially useful for informing constructive discussions about accountability, priorities and resources.
SIDE NOTE: When publishing calls for service, take care to protect victim privacy, even when state and local laws might allow for the publishing of a great deal of detail. Review this article from the Police Foundation for further information.
Are you collaborating with your library to publish police data or engage with the public? Tell us more at PDI@policefoundation.org or find us on Twitter at @PoliceOpenData
If you liked this post, please share!