Monday, April 22, 2013

Double chekc your results again...

Otherwise it may lead to economic ruin, or at least bad government policy!
Well not for scientists.
Earlier I posted that MIstakes happen in research, describing one of my recent ones.

Paul Krugman has a nice New York Times op-ed piece The Excel Depression which describes a flawed paper by two Harvard economists that led to a widely quoted claim that once government debt exceeds 90 per cent of Gross Domestic Product (GDP) economic growth declines sharply.

One of the mistakes the authors made was an Excel spread sheet coding error that meant they left five countries (including Australia!) out of their analysis. You can read a more detailed discussion here.

4 comments:

  1. Aside: It's odd that so much attention is paid to the Excel mistake when it was such a minor error relative to the others.

    Perhaps there's another lesson to be learned from this newscycle: narratives sell better than facts.

    ReplyDelete
    Replies
    1. I think your Aside is a very good point.
      The problem may be the "common person" can easily understand an Excel mistake but struggles to see clearly that dubious assumptions including "cherry picking" data are actually worse.
      I say worse, because an Excel mistake is more of a "sub-conscious" mistake whereas a dubious assumption that is stated is conscious and explicit.

      Delete
  2. Mistakes in research are inevitable, and though we should work to decrease them, we can't eliminate them.

    I think the larger problem is that people tend to believe what suits them, and portray the evidence as favouring this, whatever the reality.

    ReplyDelete
    Replies
    1. I agree. This is a good example of Kauzmann's maxim
      "People tend to believe what they want rather than what the evidence before them suggests they should believe."
      Scientists can be just as bad at this at non-scientists.

      http://condensedconcepts.blogspot.com.au/2009/03/walter-kauzmann-2009-master-of.html

      Delete