Nate was right in 2012.
But he hasn’t been right about Trump even once.
People coming out of slumber to vote again or for the first time do not figure to Nate. Neither do democrat cross-overs.
Silver decries messing with polls, and then he says to add 6 points to this poll.
What a numbnutz.
“people often misstate or misremember their vote from previous elections.”
[citation needed]
Prediction:
The polls will converge as we get closer to Election Day. The Nate Silver Projections and these polls will grow closer and closer to the USC Poll as we move through September and October. Nate Silver will stop talking about his phony 6 point “error” in the USC Poll.
By the night before Election Day we should know who is going to win. I think it will be Trump. Hope I’m right.
The methodology of the USC poll might be considered experimental, but we are in an age of experimental polls.
In the old days, you’d call (landline) phones. Today, you simply have to also call a certain percentage of cellphones, or else make some other adjustment for the fact that a lot of people today don’t have landlines. Among the phone-based polls are some that use an automated voice and others that use a live caller. Also, the length of the interview varies greatly. I know a poll-taking firm that buys lists of registered voters with all the demographic information pre-coded. His interviews consist of 3 to 5 questions, and take only a minute or so. Some other interviews take 20 to 30 minutes to complete. You get very different results, nowadays, in live-caller versus automated-caller polls, and in long interview polls versus short interview polls.
Along with phone-based polling, there is “internet” polling. Firms assemble “panels” of respondents, and recurrently ask questions of them. Basically, the same pool of people are asked, and asked again repeatedly through the course of a year. The firms try to keep the pools representative of the American population, but mostly they rely on weighting responses by demographic data.
The USC poll is of the second variety, with some twists unique to it. As to whether it has a bias as compared to the polls that Nate Silvers likes, isn’t the real question. The real question is why do live-caller polls favor Hillary by an average of something like 6 points, relative to automated-caller polls?
I guess he doesn’t like the UPI poll either (Trump +1), or the YouGov poll (Clinton +3), or the Zogby poll( Clinton +2). I’m sure he despises the Pew poll (Clinton +4).
Poor Nate is running out of polls to use.
Is this the same guy who has been wrong about Trump at every step of the way?
Is this the same guy who has been wrong about Trump at every step of the way?
Is this the same guy who has been wrong about Trump at every step of the way?
“hat’s probably a mistake, because people often misstate or misremember their vote from previous elections.”
Huh, what? Talk about stupid or desperate...
Any time you start censoring data points from a study (such as a poll) you threaten accuracy. There are already so many types of bias inherent in polling, things like confirmation bias, proper sample distribution, selection bias etc as to make them an iffy proposition. To then go back and censor 5% or 6% of the respondents because you think they were over-represented is a perilous endeavor. Your reasoning may be correct (or maybe not), but your sample size is still relatively small in the grand scheme of things so you cannot really be sure you captured a sample that really is overweighted. Plus your assumption ignores unknown variables such as whether and how this election may differ from previous elections. Just because 6% may have misremembered who they voted for (or intentionally lied and said they voted for the winner) does not mean that they are lying about their current intentions. Obama may have had a 4% overall lead in popular vote in 2012, but Obama did not win every state nor did he do so by 4% in every state he won. North Carolina for example went to Romney by 2%. How do you adjust a national poll by 6% when you know that this 6% number did not apply in a large % of the voting districts, and in fact went the other way in say 40% of the districts or an 8% swing from your hypothesis? His idea is fraught with problems.