Comments on: Sabato’s Crystal Ball: TRACKING THE 2008 PRESIDENTIAL ELECTION http://themoderatevoice.com/21306/sabatos-crystal-ball-tracking-the-2008-presidential-election/ An Internet hub for moderates, centrists, and independents, with domestic and international news, analysis, original reporting, and popular features from the left, center, and right Mon, 23 Nov 2009 11:24:33 -0500 http://wordpress.org/?v=2.8.4 hourly 1 By: DLS http://themoderatevoice.com/21306/sabatos-crystal-ball-tracking-the-2008-presidential-election/comment-page-1/#comment-120693 DLS Fri, 25 Jul 2008 23:38:21 +0000 http://themoderatevoice.com/politics/john-mccain/21306/sabatos-crystal-ball-tracking-the-2008-presidential-election/#comment-120693 Meanwhile, in Denver -- outdoing what was done in Berlin<br><br><a href="http://thecaucus.blogs.nytimes.com/2008/07/24/the-price-to-pay-for-obama-stadium-tickets/">http://thecaucus.blogs.nytimes.com/2008/07/24/t...</a> Meanwhile, in Denver — outdoing what was done in Berlin

http://thecaucus.blogs.nytimes.com/2008/07/24/t...

]]> By: Silhouette http://themoderatevoice.com/21306/sabatos-crystal-ball-tracking-the-2008-presidential-election/comment-page-1/#comment-120692 Silhouette Fri, 25 Jul 2008 18:27:35 +0000 http://themoderatevoice.com/politics/john-mccain/21306/sabatos-crystal-ball-tracking-the-2008-presidential-election/#comment-120692 I ignore polls by and large. Or rather, I pay attention to actual trends first and weigh in polls only as a last "iffy" component before I decide what's what.<br><br>I had some statistics in college and know that numbers can be manipulated.<br><br>And wording can be manipulated too in polling. That's why I gave the example of polling a bunch of local people in a bar on if they think drinking before noon is socially acceptable. Then produce results that reflect 90% of "local" (omit pub attendees) people feel drinking before noon is socially acceptable.<br><br>Polls are the favorite tools of spindoctors. I ignore polls by and large. Or rather, I pay attention to actual trends first and weigh in polls only as a last “iffy” component before I decide what's what.

I had some statistics in college and know that numbers can be manipulated.

And wording can be manipulated too in polling. That's why I gave the example of polling a bunch of local people in a bar on if they think drinking before noon is socially acceptable. Then produce results that reflect 90% of “local” (omit pub attendees) people feel drinking before noon is socially acceptable.

Polls are the favorite tools of spindoctors.

]]>