This is James's Typepad Profile.
Join Typepad and start following James's activity
Already a member?
Update has been hidden from all public facing feeds in Typepad
Excellent post! I think between Andrew's post and the examples you provide here, it captures the spirit of what analysts go through every day. One point I would make specific to your P&G example: while I wholeheartedly agree that the reverse causal questions result in only "approximate answers," I think that business leaders in general view this in the opposite way. One of the reasons that I suspect business leaders often frame problems in a reverse causal framework is that they *know* that forward causal problems can yield only approximate answers (after all, no one can predict the future with complete accuracy), but there is a belief that reverse causal questions, since they rely so heavily on historical data and hindsight, will yield answers that are both precise and accurately fit the observed situation. I'd be curious if you have thoughts on how to combat this mentality. There is a level of uncertainty that exists regardless of the direction with which one approaches causality, but experience suggests that folks perceive that reverse causal "answers" are more definite--and therefore more reliable--than their forward causal cousins.
Commented Jul 19, 2013 on
Big Data, Plainly Spoken (aka Numbers Rule Your World)
Andrew Gelman, Columbia professor, wrote an important post about causal thinking (link) that I highly recommend reading. While he approaches the topic from a researcher's perspective, his framing of the issue is very practical, as I will demonstrate in this post. Gelman's main point is the two ...
James is now following
The Typepad Team
Jul 19, 2013
Subscribe to James’s Recent Activity
View all »
Around The Web
All Rights Reserved.
Terms of Service