Jump to content

Weird...obama Has Huge Surge In The Polls (For Now) According To Gallup.

Recommended Posts

  • Replies 202
  • Created
  • Last Reply

Top Posters In This Topic

Fair to say that Rasmussen has excellent sampling methodology but suspect weighting methodology?

Do you think next time they will just follow Lord Silver's advice and drop the party ID weighting?

Btw, retreading these threads is a collosal waste of time but an interesting history of the recent past.

I think that's right. And it was interesting to read through the thread.

Link to post
Share on other sites

I don't have a huge problem with Rasmussen as long as you consider the constant bias (statistically speaking, not a pejorative) in his polls. That bias is the result of methodological issues, not him cooking his numbers. IOW, I have problems with his methodology and I think this election once again proved those problems, but it's not like the Zogby polls or other polling firms that are just not credible.

Gallup was the one that had me most perplexed, but in the end they also were not terribly far off the actual results. I'm curious what would have occurred in their polls had they not taken a hiatus because of Hurricane Sandy.

My suspicion is that the same issues would have occurred. There were preconceived notions that got baked into the polling. We know from the post-election data mining that faulty presumptions were made about the voting composition of the electorate. Anecdotally, I wound up polled on half a dozen occasions. The same two organizations kept calling me back. One of them was Rasmussen. I suspect that a person's willingness to subject him/herself to polling has grown rare in the smartphone era. The ability to collect information on potential voters, on the other hand, has expanded exponentially. So I wonder about the validity of most old school polling methodology. I do not believe it is coincidence that Rasmussen and Gallup finished tied for 24th out of 28 polling organizations.

Finally, and this isn't directed at your post or anything specific, I believe the biggest victory last night was for the quantitative study of politics and elections. I thought Nate Silver would be a little off and I was wrong. I suspect some small part of that was luck, but ultimately his model captured something that others missed. Whatever the case, last night once again showed that polls (on average) and forecasting models can give legitimate insight into what's likely to happen.

What has always impressed me about Silver (and I knew him from Baseball Prospectus) is that he doesn't let his biases get in the way of his calculations. His argument against the value of Gallup polling proved to be the most acurate of any articulated. And I would add that while luck is obviously a small factor in this, he has now predicted 99 out of a possible 100 states over the past two presidential elections. 99% accuracy is the way to build tremendous trust. It is a key reason why his book jumped all the way up to #2 on Amazon yesterday. Suddenly, people want to know more about what is under the hood in those quantitative studies you mention.

Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

  • Create New...