Viewing Crime through the Prism of Statistics
While the measures of crime provide us with standardized data on crime, crime rates, and comparative data for states and cities, statistics have their limitations. Perhaps the greatest limitation is that the human factor is missing.
Looking at statistics regarding homicides, burglaries, and sexual assault gives us some idea of the number of these crimes taking place in a location or in the country, but statistics tell us nothing about human suffering. We can learn about the economic costs of crime through statistics and certain kinds of research; on the other hand, we can learn nothing from the annual totals of the UCR Program or NCVS about the emotional toll on families and survivors.
For the crime analyst, there are more immediate concerns: who the offender is, what his or her patterns are, what the offenders motivation is, and how this offender will best be neutralized or apprehended.
Utilizing Statistics as a Form of Accountability: CompStat
CompStat, or comparative statistics, is an accountability process officially adopted by the New York City Police Department in 1994. By and large, CompStat, as used by law enforcement, is an in-house process that holds upper management accountable for crime reduction within their respective areas of patrol.
By utilizing statistical data, with the assistance of GIS technology, police agencies can track crime patterns and view these patterns in a variety of ways. For example, hot spot analysis of specific crime patterns can display spatial clusters relating to hot spot areas for these crimes and then designate the values of these clusters as high and low, with high indicating a greater area of propensity toward the particular crime pattern—for example, robbery; and low indicating a cold spot, or no area of crimes relating to that same pattern.
The basic principles of CompStat are based on the idea of utilizing accurate and timely intelligence about underlying crime conditions. This intelligence would then directly affect tactical decisions that relate to the rapid deployment of personnel and resources needed to relentlessly address, follow up, and reassess the problem.
The Crime Analyst and Statistics
We have indicated that statistics and the various measures used to determine the amount of crime in U.S. society should be viewed with caution, and you should always keep the flaws and drawbacks of each measure in mind. And we think it is not a bad idea to remember the remark that Mark Twain made in Chapters from My Autobiography: “There are three kinds of lies: lies, damned lies, and statistics” (Twain, 1907, p. 471). Although this statement probably did not originate with Twain, he made it popular. And since the late nineteenth century, the phrase has been used over and over again—even appearing in the title of several books. However, the caution is still apt today. Depending on the way you arrange numbers and statistics, you can prove just about anything. And often, statistics don’t mean much of anything when viewed out of context.
Nevertheless, crime analysts—just like the rest of us—must utilize statistics at times, and sometimes even depend on them in critical situations. Most of the time, though, statistics give us averages and ranges; they do not tell us about the personal side of crime and criminals. Typically, the tactical crime analyst must gather intelligence about individuals, and that’s when statistics tend to be useless.
Take, for example, an armed gunman who is holding a child hostage after a bank robbery attempt that has gone wrong. What do the statistics tell us about the risk to children held hostage by a stranger in a failed bank robbery? Does it matter if the analyst can quickly find information that tells him or her that 78% of the time, children are released unharmed by gunmen?
Of course, the answer to that is no, it doesn’t matter. What is important is what the analyst can quickly learn about this man, his background, his violence history, his motive, and his current state of mind. The statistics don’t mean a thing to the worried parents of the child, nor should they mean anything to the hostage negotiator who must find a way to make sure that this child is released without injury.
Statistical Data and Law Enforcement: From Incident to Evaluation
Before we talk about the data collection process, we need to take a brief look at the crime analysis process, the mindset if you will, practiced by crime and intelligence analysts. This process defines the why and how related to the analyst’s use of data.
Data collection and storage, as important as those processes are, would bear no meaning if the data were left to pile up without any forethought as to how to utilize the data. Upon collection, data must be examined for any sort of constructive purpose for law enforcement and for how law enforcement will address any needs related to the collection and examination of the stored data. This process is referred to as the crime analysis process (Figure 2.2).The crime analysis process follows five steps: (1) data collection, (2) data collation, (3) analysis, (4) dissemination of results, and (5) incorporation of feedback from users of the information (Santos, 2013).
Before we examine the analysis process, however, we first need to review how data are collected and stored before they can be extrapolated for analysis purposes.
