LectureCompStat

Hello PLC 216 and welcome to week number four. This is going to be our last set of lectures. The remainder of the course will be for you to work on the final project. I will try to keep the lectures short and to the point. For this lecture, we're going to talk a little bit about problem-oriented policing, SARA, and CompStat.

Let's start off with problem-oriented policing. According to Ratcliffe, there are four components. Targeting of offenders, which usually means targeting of active criminals through over and covert means. The management of crime and disorder hot spots, the investigation of linked series of crimes and incidents, and the application of preventive measures, including working with local partnerships to reduce crime and disorder. So, basically, the mapping techniques you have learned so far help us put into play the idea of problem-oriented policing.

Problem-oriented policing is similar to community policing. Um, they are thought to be the same, they're not. The main premise of problem-oriented policing is that the focus should be more on the problems that cause the crime incidents, rather than the criminal incidents in and of themselves. Problem-solving is the key to reducing criminal incidents versus just locking people up, just having police patrols. Herman Goldstein proposed a problem-oriented approach to violence, and our data set has been dealing with shooting shooting violence in New York City. Responding to the incident, that's only the first step in the strategy. Finding permanent solutions to the problems that lead to the to the incidents, and as we discussed, economics, um, families below poverty. We see that in our first map. There is a definite correlation there between high counts of poverty and high numbers of shootings in a particular area. Poverty in and of itself is but one symptom. You also have what goes along with poverty, substance abuse, lack of educational opportunities, etc. So, we look at these underlying conditions. As these conditions continue to degrade, it leads to the crime incidents. And while crime may appear to be isolated, I think we by now have learned crime is related to location or space, and what's going on in that location in or space a lot of times has to do with the underlying conditions.

So, part of problem-oriented policing is data collection, which means we need a routine method for identification of problems, analyzing those problems, responding, and then evaluating how effective our response was. And again, using the crime mapping software, we're able to measure all of this. With problem-oriented policing, we look at the crime event. But that crime event always has three factors. We call this the crime triangle. Perpetrator, victim, and of importance to us, location. Reviewing the crime theories we discussed earlier in the semester, we know that location plays prominently into crime events. And many times, the locations are locations that have societal problems, usually related to economics. We expect with problem-oriented policing to hopefully eliminate the problem entirely. However, that is normally not the case. Therefore, if we can't eliminate entirely, we try to reduce the number of occurrences. Um, or we try to reduce the degree of harm caused by the problem. We also look for ways to improve the way the problem is being dealt with. Or perhaps, we try to change the environment through a process called crime prevention through environmental design. Now, one thing to note with our crime maps, the numbers may have decreased. The areas, the locations have not changed. So again, I can show you shooting map data from 10 years ago, 15 years ago, 20 years ago. The maps are going to look virtually the same as those you created for this course.

Problem-oriented policing looks for those underlying conditions. So, police officers often deal with the symptoms of the problem. Crime being the symptom. Um, the characteristics of people who live or enter a neighborhood. How people feel about their neighborhood and the overall condition of the neighborhood. It's not as simple as rational choice theory makes it out to be. In that the bad guy makes a decision. Yes, they're making a decision. But why are they making that decision? What is going on in that area? Looking at these underlying conditions, we realize that crime events can arise from a single common source. The most common source, again, poverty. Uh, and then police officers only deal with the most obvious symptom of the problem. So, they see the drug use and abuse. They see the vandalism. They see the burglaries, the robberies, they see the homicides. The idea is, however, if we're dealing with the problems of economic inopportunity, poverty, substance abuse, by dealing with those problems, we can start curing the symp- uh, symptoms.

Now, part of this discussion in terms of problem-oriented policing and crime mapping revolves around a process called SARA, SARA. SARA is shorthand for scanning, analysis, response, and assessment. Scanning, identify neighborhood crime and disorder problems. Analysis, understand conditions that cause problems to occur. Response, develop and implement solutions, and then assessment, determine the impact. I think if you go back and review the maps we have already created, we've been doing this all along. However, I like for you to work with the software first and get your map-making routine down, then understand the practicality behind what you're doing. Part of this process involves something called risk terrain modeling. This was developed by criminologists Kaplan and Kennedy to identify emerging areas of concern. It uses multiple GIS layers or factors correlated with past crime to determine a geographic area where there is a high um, a higher risk for new crimes. Um, and you may recall I mentioned something early on called the mosaic effect. This is the result of combining multiple data sources so that we can get a clearer picture. We also look at whether or not crime in an area is spatial coincidence, or do the occurrences um, defy probability of coincidence? So, although I did not assign a risk terrain map just because of the short length of the semester, I do assign these maps during a normal semester. I want you to have an understanding of how they work. So, you can just follow through with this. If you feel like creating it for practice, fine. However, it's not an assignment. But, I want you to understand the process.

I'm going to create a risk terrain map of family violence related 911 calls. I've already extracted the 911 calls from the call data set, and you're going to be using this particular data set for your final projects, by the way. And I want to get an idea of where I'm likely to see these calls in relation to poverty. So, I'm test I have to test two variables. The first thing I'm going to need to do is have the software count up the number of calls in each neighborhood. So, I'm going to use a tool called summarize within. Where my neighborhoods are the polygons, and I'm summarizing the 911 calls. It's a pretty easy tool to use. And what it's going to do is it's going to create a new polygon for me. And in that polygon, I'm going to shut off my other layers here. I'll open it up. I'll open up the attribute table. And you'll see it created a new column of all the counts for each neighborhood. Now, it's interesting to me to see where these high counts intersect with high counts of families below poverty. So, using the symbology tool and a symbology called bivariate colors. Bivariate colors tests two different variables. Of course, the one variable I wanted to test are my count of the points and my other variable, the families below poverty. I'm going to select a 4 by 4 grid size. And again, this is you don't have to worry about creating this. I just want you to have an idea of how it works. Note your legend. Families below poverty is the blue color. Count of the number of 911 calls is the pinkish color. Where they merge, high counts of both, you get that purplish, dark blue, violet blue, whatever color you want to call it. You see that here. Not so coincidentally, when you look at the precinct assignments for the final project, they're going to fall within these high count neighborhoods. But that's a risk terrain map. We can get an idea of where the problems are and where we're likely to see continued problems based on the poverty levels and the number of 911 calls coming in. These are very specific 911 calls, by the way, um, for violence, for crimes in progress, where people are at risk of being harmed. So, again, risk terrain mapping is combining multiple data sources. We had 911 call data, we had poverty data, and we can do this for a variety of sources.

For the last segment of this lecture, I want to talk a little bit about CompStat and how CompStat came into being. This is an article from the New York Times circa 1990 that is titled to restore New York City, first reclaim the streets. Note the language that's used. New York City is staggering. The streets already resemble a new Calcutta bristling with beggars and sad schizophrenics tuned to tuned into inner voices. Crime, the fear of it as much as the fact, adds overtones of a new Beirut. So far, not a very nice leave written article. Many New Yorkers now think twice about where they can safely walk. In a civilized place, that should be as automatic as breathing. And now the tide of wealth and taxes that helped the city make these streets bearable has ebbed. Safe streets are fundamental. Going out on them is the simplest expression of the social contract. A city that cannot maintain its side of that contract will choke. So, crime, as well as the perception of crime, is at the forefront of most of our thinking when we are traversing in New York City, I think. This has been an ongoing problem, and I cite this article because the language that's used, if we fast forward to 2026, we get we we You similar language about crime in New York, even if we can prove that crime in New York is not as high as the media makes it out to be.

CompStat got its start from a cop by the name of Jack Maple. He is basically credited with coining the term CompStat. He was a transit police officer. He rose from an undercover detective patrolling MTA locations to the rank of lieutenant in the New York City Transit Police. This is at a time when robberies were prevalent in the transit system. He purchased a computer from Radio Shack. For you youngins, Radio Shack was a tech store that us old-time computer geeks could go buy hobby computers. So, he basically buys a hobby computer and hence the name CompStat was born. Uh, he said that CompStat was a word invented as a prototype name for the computer which compiled and stored the first sets of crime numbers. So, CompStat may mean computer statistics. It can mean comparative statistics. Um, New York City has CompStat meetings and New York City has a CompStat page that the public can access and look at real-time crime numbers. Maple called crime maps the charts of the future. He used them to discern underground crime patterns and then dispatch police officers appropriately. Now, what he noticed is by placing officers at locations the robberies were being moved to other locations. They were being displaced. And we talked about this. The answer is not always just more cops because once you have more cops in one area, crime moves to another area. Um, and he dispatched officers in what he called a rapid response.

CompStat is a process, it's a philosophy, and make no mistake about it, it is a performance management system, and cops have to an- answer for it. Um, and I know New York City, we don't have quotas, we don't have quotas, we don't have quotas. Well, you have quotas, we just don't call them quotas. They're performance management measures. Uh, CompStat is meant to stress accountability and the joint participation of units within the police department. Um, and this is performed during regular meetings, which sometimes, uh, you know, precinct commanders during these meetings are hung out to dry sometimes, based on their CompStat numbers. Ideally, CompStat should be used to reduce crime and achieve other police department agendas. The emphasis is on the sharing of information, responsibility and accountability, and then using that information to improve effectiveness. There is generally four recognized core components. Timely and accurate information or intelligence, so real-time crime data, that's good data, it's been recorded correctly, rapid deployment of resources, effective tactics, and then relentless follow-up. It's not enough to put a measure in place, we have to know how it's working. We can use the crime mapping software for that. Um, Bill Bratton described the earliest version of CompStat as a system to track crime statistics and have police respond to the statistics rather than just the crime, if that makes sense. It's about the numbers. CompStat has pros and it has cons. It can be very political.

So, to conclude when you ask, "Professor, why did we have to create all those maps?" Um, and even though this was a short session, when you take into account the maps you're creating for the final project, there's four of them. You've gotten your practice in. Um, albeit in a short time frame. Now you know why I want you mapping. The theories I think make much more sense when you can visualize the data, when you can work with the data. Policing is an information endeavor. Some of This is something many of you alluded to with the discussion board question where you introduced yourselves. Um, and you made some mention about learning the information aspects of policing. Spatial analysis, of course, is a significant component of the policing and crime prevention process. So now, as you begin work on your final projects and maps hopefully you're creating them with a new eye on how to use the data, how it's used in the real world. All right, like I said, I was going to keep this lecture short and sweet. Um, I'll see you for the next lecture. And we, I think, have successfully concluded the course other than the final project. As always, if you have any questions, please reach out to me, and I look forward to speaking to you real soon.