What does real data analysis look like? Hilary Mason and Chris Wiggins set out these stages, as the OSEMN model ("which rhymes with possum"):

* Obtain;
* Scrub;
* Explore;
* Models;
* iNterpret.

http://www.dataists.com/2010/09/a-taxonomy-of-data-science/

In a 2013 talk, Mason notes that these stages may seem obvious now, but that no-one had previously thought to write them down: https://www.youtube.com/watch?v=fZuDwiM1XBQ

My feeling is that standard statistics courses for scientists talk a lot about models and interpretation, but very little about the other stages. Even for the Models phase, we don't teach much about selecting or designing models.

The Google+ URL for this post was https://plus.google.com/+MatthewBrett/posts/WV8SvKd9fuD

Share on: TwitterFacebookEmail



Published

Category

G+ archive

Atom feed