Between and within-Krugman correlation

Posted on March 8, 2012 11:09 PM by Andrew

I just wanted to point out this comment of mine which applies the principles of analysis of variance to an informal model of political opinions and intellectual stances.

I hate it when my best lines are buried within a comment thread. (I also hate that I respond to blog comments rather than doing real work, but that’s another story. One of the good things about blogging for 538, back when I was doing that, was that the comments were soooo bad I just stopped reading them. Commenters here often have something interesting to say.)

P.S. See also here.

5 thoughts on “Between and within-Krugman correlation”

Rahul on March 9, 2012 2:01 AM at 2:01 am said:

What is / was “538”? Sorry if I am being obtuse.
- Steve on March 9, 2012 9:08 AM at 9:08 am said:
  
  Fivethirtyeight
  http://www.fivethirtyeight.com/
Henry on March 10, 2012 5:38 AM at 5:38 am said:

(I also hate that I respond to blog comments rather than doing real work, but that’s another story. One of the good things about blogging for 538, back when I was doing that, was that the comments were soooo bad I just stopped reading them. Commenters here often have something interesting to say.)

Does that mean I could help you by making enough terrible comments that the average is lowered enough to make you stop reading them?
- Henry on March 10, 2012 5:45 AM at 5:45 am said:
  
  (Notice how that comment was terrible by using the word “average” on a statistics blog).
Henry on March 10, 2012 5:42 AM at 5:42 am said:

Also, finally someone’s said what I’ve thought for a long time: that 538’s comments are bad. Why is the case exactly? Is it just because Nate’s successes attracted a relatively mainstream audience? Maybe because he started at DailyKos and brought partisan chearleaders along with him in the comment culture?

Comments are closed.