How to avoid common traps when looking at data

A while ago I heard Bill Hybels say that, “Facts are your friends”.  He was talking about the output of a large congregational survey they had undertaken and explaining that no matter how good or bad the results, at least you know how things are.  That is, you are not left guessing or even worse making wrong assumptions.

This is of course true, but the difficult task can be working out what the facts are, based on the data.  This is because there is a task of interpreting information in order to work out what it is saying.  Different results can be drawn from the same numbers and whether unwittingly or deliberately, false conclusions reached.  As the famous phrase, often attributed to Disraeli, but more likely from Mark Twain, says, “There are three kinds of lies: lies, damned lies, and statistics.”

Nate Silver shot to fame as a skilful statistician, accurately predicting the results of US Presidential elections, and has some good advice for those wanting to interpret data.  Inc.com has summarised an interview with him and come up with 5 key lessons for non-statisticians who want to be more careful in their handling of information.

Here is the summary:

  1. Correlation and causation are not the same thing – just because shark attacks and ice cream sales go up and down at the same time of year does not mean that one causes the other
  2. Averages can be useful, if you know their limitations – single results that are significantly different from the rest can shift an average and make it misleading
  3. Be wary of intuition – it can be good, but it can lead to taking the wrong direction from the start
  4. Look for the truth – rather than using the numbers to justify what you think you already know
  5. Make predictions – these can then be tested in the future and you can see how accurate your conclusions are

Here is a link to the original post.

If you want to read more on the subject, here is a link to Nate Silver’s book:

The Signal and the Noise: The Art and Science of Prediction

You can leave a response, or trackback from your own site.

Leave a Reply

Powered by WordPress