Jeff Miller of Dash of Insight wrote a brilliant post on common research mistakes. It can be found here.

He breaks it down into two things:  Selecting the Right Data and Comparison.

Selecting the Right Data.  As the saying goes, if you torture the data for long enough it will tell you anything.  There is so much data out there that you can make it say whatever you want.  The most important thing about finding the right data is accepting the limitations of it and understanding it.  Look at this way, I like a fast car but insurance and gas will cost more. Does that justify the tradeoffs for me?  Yes.  You?  Maybe.  All data is local.

Comparison.  Mr. Miller calls is comparison but I prefer the word context.  Most anyone can look at a data or chart and come to a reasonable outcome.  Data records it does not tell the story.  It can can tell a story but for the most part they are without context.  Yes you can throw a couple more pieces of data in and get closer but who knows.  Think about how ridiculous it would sound if I showed you a picture from 1950′s Chicago and asked you what it was like.

What Mr. Miller didn’t mention, probably because of time or my inabilities, is that data is secondary to what we do with it.  For the most part we are untrained at using it.  We are 14 year olds at the Playboy mansions. We get data F@#%&ed, data blocked, data complacent or whatever you want to call it.