Wellfleet, Massachusetts is a tranquil town in the Cape Cod area |
There's a saying that there are three kinds of lies: lies, damned lies, and statistics. This saying describes the power of statistics to bolster weak or false arguments. We must always be wary when confronted with statistical data. Statistics are often presented as fact, and numerical data can appear factual even when misleading or outright deceptive. Individuals or organizations with bad intentions can use numbers to deceive us, so we should always consider whether or not the sources of data have a hidden agenda. Students are taught that they should seek out reliable sources when analyzing data. But even sources considered reliable can lead us astray.
It must have been a shock to residents when the Associated Press (AP) claimed that their idyllic town, known for oysters, artists, and tourism, was the murder capital of Massachusetts with a rate double that of Boston. The rate was 40 killings per 100,000 residents versus 17 per 100,000 residents in Boston. It was also a shock to the Wellfleet police, who couldn't remember any homicides ever happening in the town. One of those officers had lived there for 48 years.
So, how could a town with no murders have the highest rate for 1993? The first thing you may notice is the rate is for every 100,000 residents. Wellfleet had only 2,491 residents in 1993. Boston, in comparison, had a population of 574,823 in 1990. That's about 230 residents in Boston for every resident of Wellfleet.
In 1993, a man accused of committing murder turned himself in at the police station in Wellfleet. The crime itself had happened 20 miles away in the town of Barnstable. The Associated Press misinterpreted the Wellfleet arrest as a Wellfleet murder. Adjusted for population, this mistake translated into 40 murders per 100,000 residents, more than double the rate for Boston.
Professor of Economics Gary Smith, the author of Standard Deviations: Flawed Assumptions, Tortured Data, and Other Ways to Lie with Statistics, says this "shows how a statistical fluke can make a big difference if the base is small." He goes on to explain that a misrecorded murder in Boston has little effect on the overall rate while a mistake in a small town like Wellfleet puts it on par with the city of Detroit. Smith says the way to deal with a small base is to use data collected over a period of several years.
Standard Deviations by Gary Smith |
Nowadays, we are flooded with information, data, and opinion. Gary Smith points out that hundreds of claims are made in articles every day and that many are based on meaningless data. Bad data can lead to bad decision-making. We shouldn't automatically believe what we read or hear because a rush to judgment can lead to bad decision-making. We should consume information from multiple sources before taking action. We should take a wait-and-see approach with data that makes new or shocking claims to see if it can be verified by future research. We should pay attention to Twyman's law which states that "the more unusual or interesting the data, the more likely they are to have been the result of an error of one kind or another." The Wellfleet incident is proof of that.
This article contains Amazon affiliate links. Clicking on these links doesn't cost you anything extra, but it helps to support this blog.
0 comments:
Post a Comment