Comments on: Request for help

By: Brian Weatherson

Brian Weatherson — Thu, 25 Sep 2003 19:13:12 +0000

I thought Atrios’s point was that web-generated samples are always awful. At least he always said the point was to get Blitzer to stop taking these polls because of the sampling issues.

By: critic

critic — Thu, 25 Sep 2003 05:30:55 +0000

Looks like Eszter beat me to it. And a little more diplomatically.

By: critic

critic — Thu, 25 Sep 2003 05:29:47 +0000

Uh, what possible statistical relevance does this have? You are self selecting “people that read our blogs” to find out the use of blogs in the media. It reminds me a little of atrios’ silly “Torture Wolf Blitzer” campaign…

By: Henry Farrell

Henry Farrell — Wed, 24 Sep 2003 16:22:10 +0000

Eszter – precisely because of that problem, I suspect that we’ll be steering clear of any forthright aggregate judgements about the popularity of specific blogs with media types. As soon as the paper is fit for human consumption, we’ll be posting it for comment – my best guess is sometime around the end of next month for the first really solid draft. And comments will be welcomed – especially from you! Bob – thanks for the lead. I suspect that we’re on the verge of an explosion of academic work on blogging …

By: bob

bob — Wed, 24 Sep 2003 16:12:02 +0000

http://radio.weblogs.com/0106327/2003/09/23.html#a61 Next thing we’ll need a survey to determine which journalism professors are doing surveys about blogs that journalism professors read!

By: eszter

eszter — Wed, 24 Sep 2003 15:53:25 +0000

I foresee leading popularity for CT in the results.;-) Will you post more on what this paper is about or is it too soon for that? I’ve stayed away from exploring blogs as a topic of inquiry in my research partly b/c of the sampling issues.. it’s a tough one! Perhaps you could post on that sometime as well.

By: Henry

Henry — Wed, 24 Sep 2003 12:39:27 +0000

Note to the interested – an anonymous comment making a gratuitous slur has been deleted.

By: Henry

Henry — Wed, 24 Sep 2003 04:32:51 +0000

Stentor – Indeed, we’re gathering other information besides. This is closer to a pilot study than a comprehensive analysis obviously, and we’ll likely avoid using aggregate results because of the obvious sampling issues. But it still would be nice to have some data on a set of relationships that is the subject of a lot of speculation, but not much proper investigation.

By: Stentor

Stentor — Wed, 24 Sep 2003 04:22:58 +0000

I’m curious as to your strategy for collecting data. I presume this post is not your only method, since that strikes me as seriously unscientific (though perhaps adequate for a pilot study that would help you develop a more sophisticated survey instrument for the real thing, or possibly if the survey portion is only a small fraction of your paper).

By: Brad DeLong

Brad DeLong — Wed, 24 Sep 2003 02:51:13 +0000

Are you going to cite Donald Ruskin, Stephen Moore, or Larry Kudlow as informed or neutral sources? Brad DeLong

By: sq

sq — Wed, 24 Sep 2003 01:18:30 +0000

2: fewer. (sorry)