Saturday, September 27, 2008

Information Overload 1: Polling


Screw what I said in my last post: too much information is awesome! Makes me feel so smart to be more informed than other people. Even better, with the internet I don't even need to think about anything, I just need to find someone who already did the thinking, all the research and analysis. It is like being omniscient. 

This week, I introduce a new feature dedicated to these people who think about the little things so we don't have to. For first installment, I introduce Mark Blumenthal whose information will allow you to speak intelligently on political polling, a currently hot topic based heavily in math and other boring stuff which none of us are going to take the time to fuck with. This topic is great in conversation because it allows you to talk about politics without pissing anyone off. Talking intelligently about the foundations of assumptions makes you sound smart to both sides. Of course, you can use it to make a fool of someone you disagree with, should you so desire.

Blumenthal writes columns all over the place, but the lion's share of his wisdom is available at his site, Pollster.com, and the best insider info is available on the blog there. The general public understanding of polls is that you can find one that will say whatever you want, and that seems to be generally a truism. But the fact of the matter is that we are presented with summaries of polls in the form of news reports on the polls. The data is there, deep and valuable data rife with meaning and import. But all we get is "Obama and McCain in statistical dead heat". 

Pollster.com helps tweak out the important shit. Just reading the daily update over there will make you far more informed than the average American. Even better is that Blumenthal does not limit his analysis to high political drama, he has recently discussed the public opinion polls on the government bailout and, my personal favorite, "Cell Phone Only Households by State" and discusses the 2-3% affect that has on presidential preferences. The "cell phone only" crowd is a great smart talking point, because it messes up their numbers because they need to figure out who it is that they are missing:
Pollsters have long understood that the cell phone only population -- those who have cell phone but no landline telephone service -- tend to be younger, and that the growth of that population has made it more difficult to reach 18-29-year olds. However, the conventional wisdom among pollsters has held that weighting by age could mostly alleviate any potential bias, as they did they did in 2004.
Pollster.com will help you understand what these discrepancies are and how they polls are compensating. In an election this close, these things could be the difference between a dead-heat poll that results in a landslide election.

No comments: