Analysis of early voting data

[Fiveyearlurker]

I’m getting to the point where I’m ready to stop trying to read tea leaves and just wait to see what happens!

[Moriarty]

My working theory of this election is that early voting broke heavily for Hillary Clinton (Note that early voting began before Comey’s letter was written), which is why she’ll survive the narrowing of the polls in these last few days.

It also seems to me that early voting has become much more commonplace than it ever was. When I was a kid growing up in the 80s, my recollection is that everyone voted on election day - early voting wasn’t an option. You generally needed a reason that you couldn’t show up on election day in order to be able to request an absentee ballot.

Now, entire states do vote by mail. You can usually get an absentee ballot just because you want one. I heard one prediction that 2/3 of Floridians who will vote will have cast their ballot before election day. It makes vote counting easier (which means results will be available earlier in the evening on Nov. 8th!), and I think it encourages turnout.

I also think that much of the early vote consists of the “silent majority” of Americans who are horrified by the prospect of Trump as President. They may not have all gone for her (probably a lot of 3rd party picks), but I really doubt he did spectacular in the early returns. And with a lackluster GOTV operation, I’m hopeful that Trump is toast.

[DSeid]

@CarnalK

Hmm, the numbers on my chart are noticeably different in that they show black turnout beating white turnout in both '08 & '12. It still shows a steady rise though through the shorter time period the Cook Report covered. I gave the raw numbers page but they graph it here:
http://www.electproject.org/home/voter-turnout/demographics

Yeah it does not change the point but nerding out here a bit I see the reason.

There are two sets of numbers in their linked data files both that start with the numbers from the Current Population Survey (CPS).

Cook Political Report uses the set labelled “census weight”, which the actual raw data. Election Project uses the set labelled “census weight for vote overreport bias correction”, which is the raw data that they’ve attempted to clean up for people who say they’ve voted but did not.

Here’s their page about overreport bias and also for the issue of non-response to voting questions in the survey. The set they use not only attempts to correct for overreport bias (as it is labelled) but also attempts to correct for variable nonresponse rates. Problem is they screwed it up.

The CPS and Vote Over-Report Bias

Another attractive feature of the CPS is it’s low vote over-report bias. Pollsters have long noted that poll respondents overstate their voting participation. A reason – although not the only one – is that people like to think of themselves in a favorable light within social norms, in this case presenting themselves to interviewers as voters even if they did not vote. While some election surveys’ vote over-report bias can be ten percentage points or higher, the CPS is quite low. For example, the Census Bureau reported a 2012 CPS turnout rate of 61.8% whereas the 2012 VEP turnout rate was 58.6%, a difference of 3.2 percentage points.2 In comparison, the venerable American National Election Study reported a 2012 turnout rate of 78%.
The CPS and Non-Response

CPS respondents participate in the lengthy employment questionnaire, but when asked voting and registration questions, they are less willing to provide responses. The entire trend over the history of the CPS is presented in Figure 1. For example in 2012, 12.8% of CPS respondents either were never asked the voting and registration supplement questions (No Response), refused to answer the questions (Refused), or did not know an answer to the voting question (Don’t Know). …

… The CPS’s over-report bias is not constant across states. Some states have higher levels than others, as measured by subtracting the 2012 VEP Turnout Rates from the CPS’s Citizen Voting-Age Population turnout rates, as is done in the figure below. These errors are systematic over several years, in that the same states tend to have the highest and lowest levels of vote over-report bias. … Aram Hur and Christopher Achen suggest excluding the non-responses to the CPS and reweighting the resulting CPS state level voter turnout rates to the VEP turnout rates disseminated on this website. They did not provide their weights. To assist the survey research community in making this correction, the associated Stata files for DIY are provided at the top of this page. Selected statistics using these corrected weights are reported here.

That last “here” links to the page you cited.

It seems that those attempts to correct for overreport bias and nonresponse though resulted in some mistake.

In 2012 White turnout raw was 64.1 and dropped to 61.8 corrected (a drop of 2.3) and Black turnout went from raw of 66.4 to 65.4 (a drop of only 1.0). Okay, that can make sense.

In 2008 the raw had White turnout on top at 66.1 and saw it drop by 0.9 to 65.2. But Black turnout raw was 65.2 (less than the raw number for White turnout but close) and went up by correction for overreport and nonresponse to 69.1, up 3.9 !?! That is simply wrong.

I think we need to go with the raw numbers as graphed by Cook.