It's always important to me to make a distinction between data, and information. Data is defined by Dictionary.com as "individual facts, statistics or items of information." Information is defined as "knowledge gained through study."
I think this is an important distinction to make because information trumps data...and yet too often, people make decisions based on data, not information.
Merrit Clifton's data is a perfect example of the difference. Clifton's data is starting to gain popularity in city council meetings. Just yesterday, I linked to two stories that mentioned Clifton's data.
Clifton's data is compiled information of media reports of dog bites that were considered either "deaths or maimings" over a 24 year timeframe listed by "breed". Clifton uses this data to try to "prove" that pit bulls are more dangerous than other "breeds".
What's interesting though, is if you research other sources, such as Dog Bite Law, you'll note that according to their data, nearly 800,000 dog bites per year are serious enough to require medical attention. Approximately 368,000 victims require an emergency room visit.
So during the 24 year timeframe in which Clifton's data is pulled, there have been approximately 8.8 million dog bites that have required an emergency room visits and a whopping 19 MILLION bites that require medical attention. Clifton's data includes 2,209 bites....or .025% of all dog bites that require a trip to the emergency room.
Clearly Clifton's data involves a statistically insignificant number of dog bites that an ER visit. Why so few recorded?
Well, in many cases the news reports that Clifton reads don't include the breed of the dog. Many are thrown out based on Clifton's decision that the attack wasn't severe enough to include (based 100% from how the newspaper reported the accident, Clifton never visits any of the victims to get a true feel of how "disfiguring an attack is, and I have all kinds of examples of newspaper reports that have over-exaggerated the devastation of an attack, here's one for an example). Additionally, not nearly every dog attack that occurs is reported by a media outlet -- in fact, just a tiny fraction of a percentage of them are.
There are plenty of other reasons why using media reports are misleading...which I'll attempt to get to later in this series.
Clifton's data is clearly represents just a small fraction of the overall significant bites. I would suggest that the reason pit bulls show up in such high numbers on his list is due to selective inclusion. Not only do I feel strongly that the media reports pit bull attacks with more regularity, I also feel as if Clifton's bias against this group of dogs causes him to include more as "maimings" and dismiss those of other breeds more readily. In this case, the data is not even information, is mis-information. Even worse.
Tomorrow, why using just a fraction of the information available will get you in trouble.