Making decisions from limited data

24 May


Daniel Kahneman was profiled on NPR’s Science Friday Desktop Diaries segment last week. In this segment Flora Lichtman interviews scientist / researchers and used desktop trinkets to get a personal view of the human being behind the science. Daniel Kahneman, who won the 2002 Nobel Memorial Prize in Economic Sciences for his work in psychiatry including 1974’s Judgment under uncertainty: Heuristics and biases published in Science, which helped establish an understanding of people’s predictable irrationality (sounds like another book I enjoyed) in decision making stemming from the belief that severely limited data can provide predictive power. In his own words, he describes this as, “regressive prediction, that is that people make absurdly, extreme predictions on the basis of very weak evidence. If I tell you about this graduating senior and I call her Julie that she read at age four. And I ask you what’s a GPA? You have an answer. An answer comes to mind. I mean, you know, that’s ridiculous. And somehow, it’s a very narrow range of answers, and it’s that sort of answer that comes to everybody’s minds.”

I was struck by something in that and have been trying to find some data on it for a couple hours. The problem is that I’m not sure how to search for it. What I want to know is, ‘is this really so ridiculous?’ That is, has anyone ever gotten data on this? And if so, do we know that it doesn’t correlate?

Because if it does, maybe we (all humans, apparently) are not so silly for believing this. His assumption only makes sense if he is sure that what we are jumping to a conclusion about it wrong. If it’s right, maybe the connection isn’t so spurious.

Perhaps I am making too much out of a short remark, but it did interest me to know whether or not there has been data collected that could answer this question at all. 

My wife has recommended his book, Thinking Fast and Slow, to me several times. Perhaps I should pick it up. 

Please, if you know of any data about this one question (reading age vs academic success) please forward it to me here and I will take a look at it. Or, better yet, write it up on your own blog and we can link to one another.Image

1 Comment

Posted by on May 24, 2013 in Uncategorized


Tags: , , , , , , ,

One response to “Making decisions from limited data

  1. downhousesoftware

    May 24, 2013 at 10:31 pm

    Well, just after finishing this post, I did find some work by Sebastian Suggate. He found that in schools where reading was taught at age five vs those where it is taught at age seven, ability scores converge at age ten and have no lasting trend afterward.
    You can find one of his (review) papers at I’m not sure that this is the same question, but at least it approaches it.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: