The subscriber-base of Banker & Tradesman surged by about 150,000 additional readers when the editor added my photo to the column.
Were the new readers all women? Am I just another pretty face? Or, is my sweet, sensitive prose doing the trick? Or, could it just be more guys who know a man’s man when they see and read one?
Correlation and causation are the curse of the analytical world; it isn’t easy to determine exactly what caused what – assuming your data is accurate to begin with.
Advertising and marketing executives make themselves crazy attempting to answer the age-old question: How many additional units of “x” will be sold because of the next dollar spent on advertising? Is it the advertising? Is it the pricing? Is it the color of the carton? Or, are you just having a lucky/unlucky month?
There are few collections of data points that prompt more hilarity among academics than “crime” statistics – mixed as they often are in a heady brew of cause-and-effect with levels of police staffing and socio-economic trends and other Partridges in a Pear Tree.
There it was last month in Lawrence, where a “surge” in crime was attributed to a cut in police positions. Since the town whacked 41 police positions last year, crime statistics in the last year have risen at double-digit levels – from auto thefts to assaults to robberies.
It makes for good headlines – and it certainly gives the police chief something to whine about at budget meetings. But the universal question remains: To what degree was the cut in police responsible for the jump in crime?
If the city added 10 cops, would that contribute to a measurable drop in the crime rate? Should cops be “on the street” (the politicians love that one), or should the detective division be goosed up, instead, to seek out the bad guys who have so far escaped detection?
About 25 percent of Lawrence residents have incomes below the poverty level. If a blimp showered the town with $5 bills, or if the economy rebounded, would the crime rate drop, quite aside from how many cops were on patrol?
Measuring Results
And how meaningful are the crime statistics being used to make the case for more police? Urban police logs are filled with crime that no police force or army could prevent.
In Hartford last month, a guy “hanging out” on a front porch with a group of friends got into a squabble with one of them. He ended up in the hospital with a gash on his cheek from a broken bottle. A silly, frivolous, ridiculous kind of “crime?” Sure – and there aren’t enough police in all of creation to stop that kind of thing.
The staffing levels of law enforcement personnel in urban settings are often less important than where their presence is felt. The priorities for prevention are based to an embarrassingly large extent on the attractiveness of the victims.
It is a squeamish subject for the politicians to talk about, but if the bad guys “stick to their own kind,” there is less of a public imperative to add to the police payroll. As brave Professor Alfred Blumstein of Carnegie Mellon University put it a few years ago: “When you see a crime spike in a city, it’s often attributable to young black males attacking other young black males. The duration depends on how fast the city reacts…”
But, to “react” too quickly to quirks in the crime statistics doesn’t always pay off, either. It was late 2008 and early 2009 when poor Boston Police Commissioner Edward Davis woke up one day with a great idea to unleash plainclothes cops on tough neighborhoods – plainclothes cops who were actually uniformed cops, but would freshen up their wardrobe for the crime fighting experiment. Six months later? Never mind. Back to the uniforms.
Such experiments can die for many reasons, not the least of which is the lack of decent data to determine whether or not they actually “worked.”
Academic constipation aside, the good folks of Lawrence, like almost everywhere else, would “feel better” with more cops roaming the mean streets. The research to prove it? Oh, who cares?
