Tuesday, July 31, 2012

60 Steps Revisited: Step 30 - Consider Repeat Offending

Back in 2009, I did a series of posts covering the excellent book Crime Analysis For Problem Solvers. The book is published by the US DOJ's Problem Oriented Policing Center (POP Center). Because of the value I think this book has for crime analysts, and policing in general, I am going to re-post this series on here on the blog.

Last time we looked at repeat victimization, in this post on our journey through Crime Analysis For Problem Solvers we're going to look at Step 30 - Consider Repeat Offending. Back in Step 18 we covered the 80 - 20 rule, the old adage that 80% of crimes are committed by 20% of offenders. Now, we know that the numbers aren't often literally 80 - 20, but it is true that a minority of offenders commit a majority of the number of offenses.

The authors discuss a couple of explanations as to why this is true from the point of view of the offender. While this is well and good I'm not sure that most police departments can do anything to prevent weak social attachments in impulsive individuals. However, in order to craft appropriate prevention strategies, we should consider repeat offender's objectives and motivations.

The authors state that there are three ways that successful offending can lead to more offending. They are:

  • Offenders, like others, learn from doing. A successful crime teaches important lessons. This can lead to the offender attacking the same target again (see box). But offenders, like everyone else, can generalize. So they learn that they may be successful if they attack similar targets (see Step 29).
  • Offenders learn from each other. Information can spread through individuals working in small groups, group breakup and new group formation. This underscores the need to understand offender networks. Police can use networks to spread information that enhances offenders' perceptions of risks or of the undesirability of the target or place. Part of the effort to reduce juvenile homicides in Boston, Massachusetts for example, involved highly targeted messages to gang members.
  • Successful offending can erode prevention, thus making subsequent offending easier. A small break in a fence, for example, will become larger with use. If the influx of offenders and offensive behaviors is faster than the responses of guardians and place managers, then a small problem will become worse.
If you have been in law enforcement for any time, I am sure that you can think of examples for each of these three. The important thing to take from this is that if you concentrate your efforts on targeting these prolific offenders, you'll get the most bang for the buck when it comes to impacting your crime problem. Your analysis should take into account the Crime Triangle to determine the best strategy to use. The authors do have an interesting caution though. They state:
Conversely, creating crime opportunities to catch offenders can make things worse. In the late 1970s and early 1980s, a number of U.S. police departments experimented with "sting" operations in which they created fake markets for stolen goods, documented who sold such goods to them, and then arrested many thieves. A number of these operations were evaluated. There is no evidence that these operations reduced crime. There is some evidence that they may have increased crime by providing lucrative and convenient ways to sell stolen goods. Throughout this manual we have noted the strong influence facilitating environments can have on promoting criminal behavior. So one should be very cautious about creating artificial crime opportunities to round up unknown prolific offenders.
Now that's an interesting assertion. How many times at your agency has the response to a specific crime problem been met with a suggestion to set up a "sting operation"? While some sting operations may not have the dynamic the authors mentioned above, we should conduct our analysis and develop strategies that actually target our prolific offenders and not just create an environment that an opportunist may take advantage of.

Next time we'll look at Step 31 - Know The Products That Are CRAVED By Thieves.

Monday, July 30, 2012

LAPD Struggles With Crime Lab Backlog

There was a sad story last week from the Los Angeles Times that looked at the huge backlog in fingerprint analysis by LAPD's Latent Print Unit. Things have gotten so bad that LAPD has instituted a plan to ration the number of cases where fingerprint analysis will be done.
Under the plan, which the department will roll out in coming months, each of the LAPD's 21 police stations and specialized divisions will be allotted only 10 cases each month in which fingerprints will be analyzed promptly, Deputy Chief Kirk Albanese said. All other cases will be placed on a waiting list. In addition, said Albanese, a handful of officers will be trained to collect prints at crime scenes in order to allow the print unit to spend more time in the lab analyzing prints.
Unlike TV show detectives, real police agencies struggle with a mountain of evidence to process and analyze. Given California's budget woes, I don't expect the situation to change out there anytime soon.

Friday, July 27, 2012

60 Steps Revisited: Step 29 - Be Ready For Repeat Victimization

Back in 2009, I did a series of posts covering the excellent book Crime Analysis For Problem Solvers. The book is published by the US DOJ's Problem Oriented Policing Center (POP Center). Because of the value I think this book has for crime analysts, and policing in general, I am going to re-post this series on here on the blog.

In this post in our continuing walk through the excellent book Crime Analysis For Problem Solvers we're going to look at Step 29 - Be Ready For Repeat Victimization. Have you ever heard the old saw "Lightning never strikes in the same place twice"? Well, at least where crime is concerned this is not true. Any cop can probably chime in with a certain business or a particular house that they always seem to be responding to over and over again.

The authors list two different kind of repeat victimizations:

  1. Boost accounts explain repetitions in terms of positive experiences at the initial offense. A burglar, for example, learns a great deal about a home during a break-in. This knowledge may encourage him to come back for another break-in. A burglar may also tell others about goods he left behind, leading to subsequent break-ins by other burglars.
  2. Flag accounts explain repetitions in terms of the unusual attractiveness or vulnerability of particular targets that result in their victimization by a variety of offenders. Some occupations have much higher victimization rates than others (taxi drivers, for example) and people who spend time in risky facilities (such as convenience store clerks) are also more prone to repeated victimization. Finally, the ownership of hot products, such as cars attractive to joyriders (Step 31), will also increase the probability of repeat victimization.
Related to repeat victims is the phenomenon of "near" repeat victims. This refers to victims that have similar geographic or other characteristics as the original victim. For instance, a neighboring apartment may be burglarized after the initial burglary. 

When repeat victimization occurs, we can use this to focus our efforts on these high risk victims. Way back when we looked at Step 8 - Use The Problem Analysis Triangle, we learned that some repeat victimizations lend themselves to effective strategies to reduce victimization. 

Next time, we'll look at a related chapter Step 30 - Consider Repeat Offending.

Thursday, July 26, 2012

Accurate Crime Stats Important To Build Trust In Police

Last week there was a story in the Milwaukee Journal Sentinel about the Milwaukee Fire and Police Commission hiring an independent auditor to audit the Milwaukee Police Department's crime reporting system. This comes after a kerfuffle that uncovered quite a number of police reports of assaults that had been wrongly categorized as lesser assaults. The story also touched on the steps MPD was taking to correct the problems with their crime stats reporting. 

Inside the story was this great quote from Mike Tobin, the head of the commission:
"The issue has brought about a certain degree of community questioning about whether the Police Department is recording the statistics accurately," Tobin said. "One of our jobs is to promote community trust in the Police Department, and the only way we are going to help improve community trust is to assure our community members that the statistics are being reported correctly."
 I couldn't have said it better.

Wednesday, July 25, 2012

60 Steps Revisited: Step 28 - Identify Risky Facilities

Back in 2009, I did a series of posts covering the excellent book Crime Analysis For Problem Solvers. The book is published by the US DOJ's Problem Oriented Policing Center (POP Center). Because of the value I think this book has for crime analysts, and policing in general, I am going to re-post this series on here on the blog.

 In this post examining Crime Analysis For Problem Solvers we're up to Step 28 - Identify Risky Facilities. In law enforcement, there is an analogy known as the 80-20 rule. The rule states that 80% of crimes are likely committed by 20% of offenders. While the ratio is not really set in stone at 80% or 20%, the idea is that most of the crimes reported are committed by a small number of offenders. In fact, Step 18 in Crime Analysis For Problem Solvers covers this phenomenon is a section titled Learn If The 80-20 Rule Applies.

When applied to locations, it's not unusual that a majority of crimes occur and a minority of facilities. An number of examples of this was given by the authors. One of them was this:

A national survey found that 6.5 percent of convenience stores experience 65 percent of all robberies.
If we can identify these risky facilities we can then attempt to identify the factors that make these places such a nuisance and a drain on our resources. The authors list eight reasons why facilities are "risky". They are:

  1. Random variation
  2. Reporting practices
  3. Many targets
  4. Hot products
  5. Location
  6. Repeat victimization
  7. Crime attractors
  8. Poor management

The authors include definitions for those eight reasons. I encourage you to hit the link and read them. The definitions tie many of these reasons to other steps in Crime Analysis For Problem Solvers that we have either already covered, or will cover before we get to the end of this series. In fact we're going to look at one of the reason in my next post when we cover Step 29 - Be Ready For Repeat Victimization.

Tuesday, July 24, 2012

Illegal Pot Operations Creating Ecological Devastation

This is sad: A story from NPR states that scientists in California studying an endangered species known as a "fisher" have attributed a number of deaths of the animals to poison.
The "likely source of AR exposure to fishers is the emerging spread of illegal marijuana cultivation within California public and private lands." Growers put out the poison to keep animals from eating their crops.
California has really struggled with illegal marihuana growing on public lands. In addition to the ecological issues with these operations, there have also been incidents where persons tending these "grows" have assaulted or threatened hikers and others using these public lands.

Monday, July 23, 2012

60 Steps Revisited: Step 27 - Know How To Use Rates and Denominators

Back in 2009, I did a series of posts covering the excellent book Crime Analysis For Problem Solvers. The book is published by the US DOJ's Problem Oriented Policing Center (POP Center). Because of the value I think this book has for crime analysts, and policing in general, I am going to re-post this series on here on the blog.

This post in my series examining Crime Analysis For Problem Solvers is going to cover Step 27 - Know How To Use Rates and Denominators. A question I get often from people is "What is the crime rate?" Most often this question comes from citizens who aren't really looking for rate but instead are looking for something else. In fact, the confusion occurs so often I tried to explain it with this post. But in this step, we're going to look at determining rates on a smaller scale to help you identify problem areas.

If you were to look at a crime map of the city where I work you'd likely see a lot of points concentrated in one area of town. In fact, you might even think this area of town had a serious crime problem. However, if you compare that map to a map of population density, the same area of town with the most crime points is also the most densely populated. This makes sense, where you have more people, you are going to have more crimes. In order to identify problem areas, we need to know where the crime does not correspond with the population, specifically where the number of crimes for a given population (crime rate) is greater than other areas.

But to be even more useful, the authors suggest determining rates based on the target of the crime. For example, instead of determining the number of vehicle burglaries per X number of persons, determine the number per vehicles registered for this area. The authors included a great example of this in the book.

Calculating rates can be very helpful in finding risky facilities (Step 28). Karin Schmerler and her colleagues in the Chula Vista, California, Police Department investigated calls from the city's motels. The 10 national chain and 16 local independent motels generated similar numbers of calls, but the national chains contained more rooms.

When they added up all the calls for the local independents and divided this by the rooms in these motels, Schmerler found that the average call rate for the independent motels was 1.8 per room. Doing the same for the national chains yielded a call rate of 0.5. Clearly, the local independents generate many more calls per room.
Calculating the rates of crime per target may require some outside of the box thinking to calculate an appropriate target value. In fact, you may not be able to determine the number of actual targets and instead have to use a proxy for this. It may not be possible for you to determine the number of vehicles in a shopping center parking lot, but you might be able to use the number of parking spaces as a proxy for this. The authors caution that you make sure your proxy is relevant to the target of your crimes.
Should you put more emphasis on high numbers or high rates? If your objective is to reduce the volume of crime, then focusing on numbers may the best choice. But if your objective is to reduce the chances of harm, then focus on rate.
The effort required to calculate rate may not always be worth it, but it is another tool to put in your toolbox for when the need arises.

Next time, Step 28 - Identify Risky Facilities.

Friday, July 20, 2012

60 Steps Revisited: Step 26 - Take Account of Long Term Change

Back in 2009, I did a series of posts covering the excellent book Crime Analysis For Problem Solvers. The book is published by the US DOJ's Problem Oriented Policing Center (POP Center). Because of the value I think this book has for crime analysts, and policing in general, I am going to re-post this series on here on the blog.

 This post in Crime Analysis For Problem Solvers is going to cover Step 26 - Take Account of Long Term Change. Law enforcement practice has come a long way. Agencies now regularly use sophisticated analysis tools to uncover trends and identify problems. Charts and graphs that were once only regularly found in the corporate world are now making regular appearances at police department staff meetings. It's also not uncommon to see crime analyst job postings that require college or even graduate level education in statistics. All this is because departments are applying business intelligence techniques to their workflow.


Time series analysis is measuring the number of crimes or other events over time. This step in Crime Analysis For Problem Solvers discusses three important ideas in time series analysis. They are:
  • Overall trend
  • Temporal cycles
  • Random fluctuations
All three of these concepts are important to understand when graphing crimes or events over time. If your graph of events over time exhibits change, you should attempt to identify the reason for these changes in order to better understand what is occurring. 

Overall Trend
This is the one your Chief will likely lose sleep over. Did your crimes increase or decrease this year? Overall trend, the authors state: "shows whether the problem is getting worse, better, or staying the same over a long period."

Temporal Cycles
Temporal cycles are those changes that occur regularly due to the rhythms found in normal life. A daily cycle might be the increase in traffic accident calls each workday around the morning and evening commutes. A weekly one may be an increase in DWI arrests on weekends. At my agency we have a yearly decrease in calls for service around January and February when the weather is the coldest.

Random Fluctuations
These are the hardest to explain. They may be caused by a one or two day increase in crimes committed by a lone offender or just minor changes in the "noise floor" of crimes. The authors describe the cause as "a large number of minor influences". 

The authors state the importance of understanding and using time series analysis comes from using it as a gauge of the effectiveness of your response to problematic crime.
Time series analysis is a powerful tool for evaluating the effectiveness of a response. The basic principle is to obtain a good idea of a problem's natural trends, cycles, and variation before the response is implemented, using the techniques just discussed. This tells you what you can expect from the problem in the future, if you did nothing about the problem. This provides a basis for examining time frames after the response. Changes in the trend, cycles, or even the random fluctuation suggest the response had an impact. The longer the time frames before and after, the greater the confidence you can have in your conclusions.
Next time we'll look at Step 27 - Know How To Use Rates and Denominators.

Thursday, July 19, 2012

The Cultural Shift Of Police Getting Out Of Their Cars

Last week I saw this bit over at The Seattle Times that I think is worth commenting on. Seattle Police recently announced a crime fighting initiative aimed at reducing crime in identified crime hot spots in Seattle.
Armed with crime data, commanders in each of the Seattle Police Department's five precincts are identifying "hot spots" and directing patrol officers to get out of their cars in those areas, interact with residents and business owners to gain information, and deter crime and reduce fear by increasing their visibility.
The one quote in the story that really caught my eye was this one:
"We typically work from a random patrol standpoint," he said. "Getting out of the patrol car is a cultural change for us."
Seattle is probably not the only place where having police officers get out of their cars to interact with the citizenry is a "cultural change". However, it's very hard to meaningfully interact with people when you're whizzing by in an air conditioned patrol car at 30 miles per hour with the windows rolled up and the good times radio blasting out your favorite tunes. It's a shame that in many cities, police work has devolved to officers driving from call to call with little discretionary time to get out of their cars. 

It's also unfortunate that "random patrol" is thought of as a crime fighting strategy by so many agencies. You might as well say that your plan is "driving around aimlessly hoping to stumble onto crime". The good thing is that most officers' "random patrol" is hardly random. They know what neighborhoods have crime problems and likely are spending more time there trying to reduce crime even if they haven't been directed from above.

As crime analysts we can help those officers and the agency brass understand your community's crime problems more clearly and help formalize a strategy that is anything but random. What are you doing to help direct your officers' discretionary time to where it's most likely to be effective?

Wednesday, July 18, 2012

60 Steps Revisited: Step 25 - Pay Attention To Daily And Weekly Rhythms

Back in 2009, I did a series of posts covering the excellent book Crime Analysis For Problem Solvers. The book is published by the US DOJ's Problem Oriented Policing Center (POP Center). Because of the value I think this book has for crime analysts, and policing in general, I am going to re-post this series on here on the blog.

We're up to Step 25 - Pay Attention To Daily And Weekly Rhythms in our walk through Crime Analysis For Problem Solvers. There is an old adage that "everything in life is cyclical". This is true not only in the change of seasons, but also in crime problems. Many crime problems revolve around the cyclical patterns of the lives of both offenders and their victims.

In the jurisdiction where I work, we often see the frequency of certain types of incidents increase in cyclical patterns. For instance, disturbances involving juveniles will increase immediately after the end of the school day when unsupervised children begin their journey home. Large public disturbances or fights in the parking lots of night clubs usually occur around closing time for the bars. Traffic accidents are more common during the morning and afternoon commutes.

The cyclical nature of these types of events make them relatively easy to predict and deploy solutions to combat them. In the example above involving errant school children, my agency increased staffing by detailing extra officers in the neighborhood near the school during the time these teens were making their way home. These officers were then able to respond to these incidents and take action rather then responding from other areas of town.

The authors state that there are three forms of temporal clustering.

  1. Diffused - Events relatively evenly spread over the entire day
  2. Focused - Events clustered around rush hours
  3. Acute - Events tightly packed within small periods
The authors state:
Focused and acute patterns immediately suggest temporal cycles that should be investigated.
Conducting temporal analysis of problems will often reveal such patterns. Most spreadsheet applications can create some very nifty charts that will help you to easily identify temporal patterns. The authors suggest performing an analysis of both time of day and day of week together rather than separately. The reason for this is that it is not unusual that if the frequency of events is calculated separately and then the results are combined you will end up with a misleading analysis. A surface or contour chart is a neat way to graph two variables together and determine the day/time that has the most events clustered around it.

Next time we'll look at Step 26 - Take Account of Long Term Change.

Tuesday, July 17, 2012

Penn State Report Also Highlights Clery Act Non-compliance

Last week the news was dominated by the release of the Freeh Report. This report was commissioned by Penn State University to look into the scandal surrounding former Penn State coach and convicted child rapist Jerry Sandusky and the Penn State's handling of the whole sordid affair. 

I snagged a copy of the Freeh Report after it was published and the part I want to touch on here is the issue of Penn State's failure to abide by the Clery Act. This federal law requires colleges and universities to report and publish information about crimes occurring on campus. At the police department where I work we regularly get requests from our local colleges who are looking for crime information that are reported on satellite campuses out in our sleepy little burg and not on the main campus itself which has it's own police department.

The Freeh Report devotes an entire chapter to the problems with Penn State's compliance with the Clery Act. In Mr. Freeh's remarks at the press conference where he announced the release of the report he puts it this way:
As you will read in our report, Penn State failed to implement the provisions of the Clery Act, a 1990 federal law that requires the collecting and reporting of the crimes such as Sandusky committed on campus in 2001. Indeed, on the day Sandusky was arrested, Penn State’s Clery Act implementation plan was still in draft form. Mr. Spanier said that he and the Board never even had a discussion about the Clery Act until November 2011.
Another thing I found interesting is that in the US Department of Education's The Handbook for Campus Safety and Security Reporting they specifically indicate that certain persons they call Campus Security Authorities or CSA'a are required to report crimes brought to their attention to comply with the Clery Act. Under the section giving examples of persons required to report they list:
A director of athletics, a team coach or a faculty advisor to a student group.
In this instance, the late Penn State coach Joe Paterno meets the definition of CSA and would be required to report crimes brought to his attention in keeping with the Clery Act. It also appears that others involved in the apparent coverup at Penn State would also meet the definition of CSA and would also be required to report crimes to comply with the Clery Act.

I have a sneaking suspicion that worried college and university administrators all across the country are reviewing their own Clery Act compliance after the release of the Freeh Report. This is probably a good thing. Crime statistics are only useful if they are reasonably accurate.

Monday, July 16, 2012

60 Steps Revisited: Step 24 - Know When To Use High-Definition Maps

Back in 2009, I did a series of posts covering the excellent book Crime Analysis For Problem Solvers. The book is published by the US DOJ's Problem Oriented Policing Center (POP Center). Because of the value I think this book has for crime analysts, and policing in general, I am going to re-post this series on here on the blog.

In this post in our journey through Crime Analysis For Problem Solvers we're going to cover Step 24 - Know When To Use High-Definition Maps. As computers become more powerful and mapping software becomes more ubiquitous, we're seeing an explosion of crime mapping. However, crime mapping does have some limitations.

Most Department's crime mapping is done based on address data found in their Records Management Systems. One common problem is due to the way that crimes locations are recorded. For instance, there is a large indoor shopping mall in the jurisdiction where I work. This mall, also has a number of outbuildings around it's perimeter. In spite of this very large area, with numerous buildings, they all have the same physical address. When the crime is entered into the Computer Aided Dispatch (CAD) system with the same address regardless of where it occurred on the complex. In fact, incidents happening on one side of the property are about one half mile away from incidents happening on the other end.

If you have problematic crimes occurring at a similar large facility in your city, you may need to use high-definition maps to adequately study the problem. The authors of Crime Analysis For Problem
Solvers
put it this way:

Mapping might therefore suggest that a particular building or facility has a crime problem, but this may only be because it is so large. When account is taken of the many people working in the building or using the facility, it could prove to be relatively safe. For example, George Rengert showed that a parking garage in central Philadelphia identified as an auto crime hot spot actually had a lower rate of auto crime than the surrounding streets, once account was taken of the large number of cars that could be parked in the facility.
A solution to this problem is to use high definition maps to precisely locate these incidents on a map of the facility. The authors give an example of a college campus where an effort to precisely identify crime locations led to this interesting finding.
Crimes recorded by the campus police were then plotted exactly where they occurred, allowing them to be related to environmental features such as poor lighting or a blind corner allowing the attacker to lie in wait.
For most Departments and most crime problems, this kind of high-definition mapping is likely to be a bit of overkill. But it should be another technique to put in your crime analyst's toolbox should the occasion for it's use arise.

Next time, we'll cover Step 25 - Pay Attention To Daily And Weekly Rhythms.

Friday, July 13, 2012

Even Victims Of Minor Crimes Face Big Fears

After working so long in law enforcement, I sometimes wonder if we loose a little of our empathy for crime victims. It takes a story like this one over at The New York Times every now and then to remind me of why we chose the profession we did.

This good news was lost on Eric. He does not inflate what happened, and he knows that much worse things happen to people all the time, and that being relieved of cash and electronics is practically a rite of passage in some neighborhoods. But the psychological aftermath of the robbery had a deep impact.

“Mentally, everything was different,” he said. “I’d turn around to check my back every couple of steps. I went from knowing everybody to having a bull’s-eye on my back.”

He all but stopped walking outdoors alone. He said he would sleep late on purpose, so that his mother had to drive him to the subway. He missed the Explorers classes and church nights. Finally, a supervisor with the Explorers promised to drive him home from class.

Don't ever lose sight of the fact that for a victim, there are no minor crimes.

Thursday, July 12, 2012

60 Steps Revisited: Step 23 - Diagnose Your Hot Spot

Back in 2009, I did a series of posts covering the excellent book Crime Analysis For Problem Solvers. The book is published by the US DOJ's Problem Oriented Policing Center (POP Center). Because of the value I think this book has for crime analysts, and policing in general, I am going to re-post this series on here on the blog.


I know that it's been a while since I posted in my series covering Crime Analysis For Problem Solvers. Now that the holidays are over and we're all back in the office regularly, it's time to look atStep 23 - Diagnose Your Hot Spot. With the increase in mapping capabilities in law enforcement software packages, there is an increase in these programs offering "Hot Spot" tools to designate a geographic location as a "hot spot" based on crime data. But what is a "hot spot" and why are they important?

The authors of Crime Analysis For Problem Solvers define three different types of Hot Spots.
  • Hot dots are locations with high crime levels. These show crime concentrated at facilities or at addresses of repeat victims (see Steps 28 and 29). Multiple crime events at places are represented by dots.
  • Hot lines are street segments where crime is concentrated. These might occur, for example, if vehicles parked along particular streets suffer high rates of break-ins. Multiple crimes along street segments are shown with lines.
  • Hot areas are neighborhoods where crime is concentrated. Hot areas arise for a variety or reasons. Area characteristics may give rise to crime. Or a hot area may contain many separate and discrete problems. On maps, hot areas are shown as shaded areas, contour lines, or gradients depicting crime levels.
Geospatial analysis to identify Hot Spots will help you in identifying problematic areas. In my workflow, after I pull Calls and Offense data from our records management system and into our GIS, I'll then run these crime layers through CrimeStat, a spatial statistics analysis tool, to create additional Hot Spot layers for my crime maps.

Running a Hot Spot tool to identify Hot Spots on a map should not be the end of your analysis. Just knowing that there is a problem in a particular geographic location is not enough. This is alluded to in this chapter by the authors who conclude the chapter with this:
Hot spot analysis can be a valuable tool early in the problem-solving process, but having discovered hot spots, you need to ask why some spots are hot and others are not. Stopping analysis after the discovery of hot spots can result in superficial analysis and the implementation of ineffective responses. If there is no geographical component to the problem, hot spot mapping has little utility and you must use other analytical approaches.
Hot Spot analysis identifies geographic areas that need further analysis to identify specific problems and develop solutions to those problems. However, it's not the end of your analysis but instead it's just a beginning.

Next time, we'll cover Step 24 - Know When To Use High-Definition Maps.

Wednesday, July 11, 2012

Crime's Down But Not Everywhere

There were a couple of interesting crime statistics related news stories last week. The first one from over at the Austin American Statesman showed that overall, the crime rate in Texas is down.

Compiling data from the FBI's Uniform Crime Report, DPS officials said that the state's overall crime rate — the number of crimes per 100,000 people — decreased by 8.3 percent in 2011 compared with 2010.

Texas saw significant declines in violent crimes between those years, including a 14.3 percent drop in murders, a 15.4 percent drop in robberies and a 4.3 percent drop in rapes, according to the data. In addition, property crimes, including burglary, theft and motor vehicle theft, dropped between about 8 percent and 9 percent each.

While there was good news in Texas, the news isn't so good in New Orleans. NPR had an interesting piece about the surge in murders in the Big Easy.

New Orleans now has the highest per capita murder rate in the country. Most of the killings are concentrated in the city's poorest neighborhoods — places like Central City, just a few blocks north of the stately mansions that line St. Charles Avenue.

One part of the piece that struck me was how the lack of trust of NOPD by community contributes to the cycle of retaliatory violence when victim's families take matters into their own hands because they can't count on the criminal justice system.

Tuesday, July 10, 2012

60 Steps Revisited: Step 22 – Examine Your Data Distributions

Back in 2009, I did a series of posts covering the excellent book Crime Analysis For Problem Solvers. The book is published by the US DOJ's Problem Oriented Policing Center (POP Center). Because of the value I think this book has for crime analysts, and policing in general, I am going to re-post this series on here on the blog.

In the last post in my series covering the book Crime Analysis For Problem Solvers we looked at the necessity to collect our own data. Now that we have the data we need to start our analysis by Step 22 - Examine Your Data Distributions.
I have a confession to make, I never did well in math in school. In fact, if you were to ask my high school math teachers what kind of job I would have as an adult, they definitely would not have guessed one that involves any sort of mathematical competency. Yet, as a crime analyst I spend a lot of time crunching numbers. I like to tell people, I have a calculator with a lot of buttons and I have to use them all.
The authors state:
“After collecting your data you need to know what it is telling you.”
Analysis is the art of learning what the data is telling you. A good way to analyze the data is with statistics. I can hear you all groaning right now. You probably didn’t like statistics when you took it in high school or college. One thing I found is that while I didn’t like math in school because didn’t see the point, now that it can do some very useful stuff and make my life easier, I think it’s pretty neat. What’s even better is that you don’t have to worry about the whole theoretical framework behind how and why statistics work and just concentrate on learning what is useful.
The authors discuss a few basic statistical tools for analyzing data distributions. These tools analyze the average case or the spread of cases. To analyze the average case you can use:
  • Mean
  • Median
  • Mode
To analyze the spread of cases, use:
  • Range
  • Inner Quartile Range
  • Standard Deviation
Another set of tools is used to measure the scale of the distribution. This can be done using:
  • Nominal scales
  • Ordinal scales
  • Ratio scales
I’m not going to try to explain these, but instead will encourage you to hit the link and read them for yourself. Understanding these tools and what they can tell you about your data is really important if you are going to be a problem solving crime analyst. The authors even include a couple of hyperlinks to some websites with additional information about statistics.
Next time we’ll cover Step -23 Diagnose Your Hot Spot.

Monday, July 9, 2012

Is Predictive Policing A Crime Analyst In A Box?

There has been quite a lot of interest lately in technology that's been called "Predictive Policing". In fact, I've covered quite a few stories about this here on The Crime Analyst's Blog. Just last week there were these stories on the topic:
At least one of those stories is an AP News piece so it's probably popped up in quite a few other publications that subscribe to AP stories.

These stories sparked a discussion over at the International Association of Crime Analysts mailing list.This list is a members only list that has quite a number of participants from academia to practitioners and all ranges in between. On thread on the topic I thought was interesting centered around the discussion of whether this type of technology could replace a crime analyst either now or in the future.

Years ago, in what sometimes seems like another life, I was a hard charging street cop in a mid sized agency of around 200 officers. I worked hard and got on our department's SWAT team and was a tactical officer for about seven years. Our agency's team regularly participated in the Texas Tactical Police Officers Association. We'd usually send several officers to the TTPOA conference every year. Like any good conference, there was always a vendor expo with vendors hawking all kinds of kit they though every tactical operator needed.

Many vendors had great gear. However, there were some vendors that weren't so great. In fact, it became kind of an inside joke among operators that for these less stellar vendors, they would take an ordinary product, paint it black, stick some velcro on it, and call it "tactical".

Often times vendors who sell things to government entities will hear a buzzword like "predictive policing" and then decide to jump on the buzzword bandwagon. In the field of crime analysis, the buzzword du jour seems to be "predictive policing". As news stories about predictive policing proliferate, we'll likely see an increase in vendors hawking their latest product that does predictive policing.

The problem with this is that predictive policing is a definition that's a broad as a barn door. Is predictive policing only valid if it's something like this: Self-Exciting Point Process Modeling of Crime? Or is predictive policing as simple as saying the guy that has robbed five pizza joints in your sleepy little burg almost always does it on Wednesdays and Thursdays between 9PM and midnight so I'll make a prediction that "I think he's going to hit again on Wednesday or Thursday from 9PM to midnight"?

Even more problematic is the idea that someone could get the idea that you can replace crime analysis with the latest whiz bang software. You might make your analyst's life easier with this kind of tool but I doubt that you're going to be able to replace an analyst with software, especially since the definition of what makes up predictive policing is not settled.

Just because you paint something black and stick velcro on it doesn't make it tactical. Neither does sticking a "predictive policing" label on the box make a crime analysis tool worthwhile.This is an emerging technology and I think it's going to be a while before it's a "crime analyst in a box".


Friday, July 6, 2012

60 Steps Revisited: Step 21 - Collect Your Own Data

Back in 2009, I did a series of posts covering the excellent book Crime Analysis For Problem Solvers. The book is published by the US DOJ's Problem Oriented Policing Center (POP Center). Because of the value I think this book has for crime analysts, and policing in general, I am going to re-post this series on here on the blog.


In this post we are going to cover Step 21 - Collect Your Own Data. For a while we've been making our way through the excellent book Crime Analysis For Problem Solvers. The book is published by the Center For Problem Oriented Policing and is a must read for any crime analyst. I've been posting about each step is order to pique your interest in reading the book on your own. The thing that makes a crime analyst a professional and not just a job is the constant need to learn and improve our skillset. A professional crime analyst should never be satisfied that he or she knows all there is to know about their profession. Going through Crime Analysis For Problem Solvers is one good way to add tools to your analyst toolbox.

Before I became a crime analyst, I was a working police officer for about 14 and one half years. There was a change in mindset in going from a police officer to be an analyst as instead of jumping into a city car and going out to investigate, I was stuck behind a desk staring into a computer screen. This chapter will give you permission to get out from behind your desk, at least every now and then.


The reason being, is that in Problem Oriented Policing, not all the data you may need to solve a problem is to be found in your database. Some of it may be in a database, but it may not be one you have immediate access to such as one maintained by another City department or even another agency. In several examples given by the authors, the data likely didn't exist in any database and required the analyst to get out and do the "fieldwork" required to collect it.

The authors list a number of additional benefits to collecting your own data.
  1. Getting into the field can give you an understanding of the problem that you would never get from sitting in front of your computer, however rich the data that you manipulate.
  2. Designing a data-collection instrument can force you to think very hard about the nature of the problem, about the kind of responses that might be effective, and how best to evaluate your efforts.
  3. Involving police officers in data collection (and in the design of the exercise) provides a valuable opportunity to train them in the need for a rigorous, systematic approach in a problem-oriented project.
  4. Undertaking your own data collection gives you the opportunity to hone your research skills and be genuinely creative. Source: Crime Analysis For Problem Solvers
Next time, we'll look at Step 22 - Examine Your Data Distributions.

Thursday, July 5, 2012

Left, Right, Left Applies To Sneaker Thieves Too

We've talked about CRAVED items here before. CRAVED is an acronym that describes things that are often targeted by thieves. One item that often ends up on any CRAVED list is athletic shoes. There was a piece over at USA Today that looked at an interesting trend by sneaker thieves.

The newspaper reports that thieves have figured out which sneaker shops put left-foot shoes on display and which use right shoes, and hit them both to assemble a good pair.

"They go to Foot Locker, which has all right-sided shoes. Then they come here because we display left-sided shoes," Salissou Mohamam, 40, manager of Michael K., a SoHo sneaker store, tells the Post.

It's not unusual for thieves to adapt to theft prevention strategies. Any strategy needs to be constantly evaluated to make sure that thieves haven't figured a way around it.

Wednesday, July 4, 2012

Happy Independence Day!

It's Independence Day here in the United States and most everyone is on holiday including me. It's my hope that you're fortunate enough to spend your holiday with friends and family which is what I plan on doing today.

I'll be back to a regular post tomorrow. Happy 4th everyone!

Tuesday, July 3, 2012

Are You Fighting Crime With Social Media?

There was an interesting piece over at USA Today that looked at the use of YouTube and Twitter by police departments. Many agencies are using these social media sites to help solve crimes.

The Kansas City Police Department has posted at least 46 surveillance videos on their YouTube channel hoping for tips.

"It's been very successful, especially for our robbery unit," spokeswoman Sarah Boyd says. "Detectives are much more likely to put the video up because they know it's so easy."

Posting surveillance videos to YouTube gives the media instant access so they can spread the footage quickly, Boyd says.

With the proliferation of video surveillance technology, even relatively minor crimes often have video evidence. Tools like YouTube make it much easier to share these videos worldwide with just a few clicks.

Is your agency using YouTube or similar sites to distribute surveillance videos? If not, why not?

Monday, July 2, 2012

60 Steps Revisited: Step 20 - Formulate Hypothesis

Back in 2009, I did a series of posts covering the excellent book Crime Analysis For Problem Solvers. The book is published by the US DOJ's Problem Oriented Policing Center (POP Center). Because of the value I think this book has for crime analysts, and policing in general, I am going to re-post this series on here on the blog.

We're up to Step 20 - Formulate Hypothesis in Crime Analysis For Problem Solvers. Imagine that your Chief comes to your office. He tells you that City Hall is getting complaints about a particular crime problem and tasks you with coming up with a solution. What do you think his reaction would be if you tell him that you will take a guess at a solution? I imagine he'll say something like "You're an analyst. I don't pay you to guess. I want a solution." In actuality, they do pay you to guess, it's just dressed up and respectable when we call it a "hypothesis". Much of what we do in law enforcement is guess work. We base our guesses or hypotheses our training and experience but in the end it is still a guess.

The authors have some guidelines about this guess work:
  1. clearly state your hypotheses
  2. not be wedded to them
  3. use data to objectively test them (Source: Crime Analysis For Problem Solvers )
These guidelines are what differentiates a guess from a hypothesis. As we analyze our crime problems using the strategies outlined in the previous 19 steps we've covered in Crime Analysis For Problem Solvers we should be working towards a hypothesis. This hypothesis is our attempt at understanding the dynamics of the problem so we can then develop a possible solution. The analysis should always move us in the direction of a solution. It should not, as the authors caution lead to "Paralysis by Analysis".
The lack of explicit hypotheses can lead to "paralysis by analysis," collecting too much data, conducting too much analysis, and not coming to any useful conclusion. Source:Crime Analysis For Problem Solvers
It is important to realize that we will never have all the information we need to be absolutely certain our hypothesis is correct. There comes a time to be decisive and put our hypothesis out there. We don't want our crime analysis units to be thought of as a "black hole" where data comes in but information never escapes. 

Next time, we'll cover Step 21 - Collect Your Own Data.