Feels like Citizens United redux, eh? Get that finger on the delete key ...
We're trying to resuscitate some of our graphics and figures from previous years. I think Nate has veered too far in the other direction now, claiming that there is NOTHING that you can learn from early turnout. You can learn something, you just have to be modest with what you can infer at this point.
In the mea culpa, he mainly focuses on the problems with comparing early voting turnout figures to overall registration figures--this was the gist of PART of mine and Michael's response. But he doesn't acknowledge another flaw in the original analysis: lumping all "early voting" as if it is the same, even though the literature (me, Michael, Bob Stein at Rice, Barry Burden et al at Wisconsin, Kate Kenski at Arizona, and others) have shown again and again that different "modes" display different patterns of turnout).
But the mea culpa goes too far in the opposite direction. Silver writes:
There is no rule of thumb about what early voting figures “should” be. The early voting advantage is presumably some function of: (i) demographics, i.e., older voters are more inclined to vote early; (ii) enthusiasm, which is peculiar to each particular election; (iii) the extent to which each party emphasizes early voting; (iv) whether or not the year is a midterm. Suppose we were trying to fit a four-variable regression model to predict early voting. You can’t really fit a four-variable model on only four data points (e.g. 2008, 2000, 2004 and 2006). It just doesn’t work, statistically. You can’t even hazard a guess. ... We really don’t know. I mean, we really don’t know. Early voting is quite new in most parts of the country, and something most voters weren’t taking advantage of until quite recently.
Well, no, we really DO know much more than that.
We also know how many ballots came back early: 15% of ballots in 2000, 20% in 2004, 25% in 2006, and 33% in 2008.
That does sound like "most", I suppose, but in many states, there is a long history of early voting, it is not new at all. Silver's regression example is incomplete--it is not as if the only way you can model early voting behavior is to example NATIONAL early voting rates for four NATIONAL elections. That is the only way I can imagine we can come up with four data points.
In Texas early voting has been around since the early 1980s. In Oregon, it has been in place for all voters since 2000. California, Washington, and Oregon have allowed no-excuse balloting since the late 1970s. Tennessee went no-excuse in the early 1980s. The examples go on and on. A good model of early voting turnout must be estimated at the state level (as we did in our article in Democracy in the States) so that you can include precisely the kinds of measures Silver points to in his blog post. His mistake is to assume the measures he lists (demographics, enthusiasm, campaigns, midterm/not) are only measurable on a national basis.
Of course, they are not. Here in Oregon, we do not have a competitive Senate race, a competitive gubernatorial race, and only one competitive House race. California is much more competitive at all three levels. Prior research has shown that this will make a notable difference in early voting turnout. We *CAN* model this in these states, and we *CAN* model this at the individual voter level, using voter history files. Far from a regression with just four observations, you can run an obscenely complex regression with millions and millions of cases, as I'm sure the national GOTV operations for the Democrats and Republicans are doing every day.
Keeping that in mind, I do think it is possible to compare 2006 rates to 2010 with appropriate caveats. If a state has had early voting for a long period of time (
---
Paul Gronke Ph: 503-517-7393
Fax: 734-661-0801
Professor, Reed College
Director, Early Voting Information Center
3203 SE Woodstock Blvd
Portland, OR 97202
EVIC:
http://earlyvoting.netOn Oct 25, 2010, at 9:21 AM, Douglas Johnson wrote:
Mr. Silver's response still fails to note why all the data he and the Politico article analyzed tell us nothing (and, in the process, reveal the failure in usefulness of official voting data): one cannot draw any accurate conclusions from comparisons of early voting data at this point in the 2010 election to TOTAL early voting data in past elections, as Mr. Silver tries to do. There's a reason Dr. McDonald does not do that -- even though the data set Mr. Silver uses comes from Dr. McDonald. The ONLY accurate comparison is between early voting data at this point in the 2010 election to early voting data at the same point in past year election cycles.
Many elections officials have that data, but they either don't disclose that they have it or, more often, they only distribute it in a format that only the professional voting registration list vendors can draw any useful data out of it (and those vendors are only doing so for paying [political party] customers). For example, in California counties will provide a list of voter ID numbers and the day a ballot is received. The parties are obviously tracking this information by the day, but they only reveal it to the public when it suits their purpose and there's no funding for more public organizations to do what's needed to process the raw data and get any useful information.
- Doug
Douglas Johnson
Fellow
Rose Institute of State and Local Government
Claremont McKenna College
o 909-621-8159
m 310-200-2058
douglas.johnson@cmc.edu
www.RoseReport.org
From: election-law-bounces@mailman.lls.edu [mailto:election-law-bounces@mailman.lls.edu] On Behalf Of Paul Gronke
Sent: Monday, October 25, 2010 8:14 AM
To: Michael McDonald
Cc: Scarberry, Mark; Law Election
Subject: Re: [EL] Nate Silver on early voting stats
Michael
Thanks for the reply. I note that Nate has "replied", sort of. It's good to see the mea culpa.
---
Paul Gronke Ph: 503-517-7393
Reed College and Early Voting
Information Center
http://earlyvoting.net
On Oct 25, 2010, at 1:34, Michael McDonald <mmcdon@gmu.edu> wrote:
Here is a link to my response.
http://www.huffingtonpost.com/michael-p-mcdonald/does-early-voting-show-re_b_773236.html
I lay out three main errors in Nate Silver's analysis and Molly Ball's Politico story:
• Nate and Molly assume that overall voter turnout rates are typically similar for registered Democrats and registered Republicans. This is not the case. Registered Republicans tend to vote at higher rates.It is invalid to compare early voters to all registered voters in the current election.
• Nate and Molly assume that early voters are the same as Election Day voters, when there is plenty of evidence that they are not - except for the all-mail ballot states. Nate's biggest pro-Republican state, Pennsylvania, exposes this poor assumption. There is a simple explanation for the apparent doom that Nate spells for Joe Sestak. Pennsylvania has a relatively small number of returned ballots because the state requires an excuse to vote an absentee ballot. There is a mountain of survey evidence that excuse-required absentee voters tend to be Republican (think: retirement homes and traveling businessmen). A valid comparison is early voters in the current election to early voters in a previous election, not registration in the current election -- or exit polls in 2008, which Nate also analyzes.
• More generally, the first early voting numbers a state reports tend to heavily composed of absentee mail ballots, like those in Pennsylvania. As Election Day draws closer, the numbers have tended to shift in a Democratic direction. A more valid comparison is early voters now to early voters at a similar point in time in a previous election.
============
Dr. Michael P. McDonald
Associate Professor, George Mason University
Non-Resident Senior Fellow, Brookings Institution
Mailing address:
(o) 703-993-4191 George Mason University
(f) 703-993-1399 Dept. of Public and International Affairs
mmcdon@gmu.edu 4400 University Drive - 3F4
http://elections.gmu.edu Fairfax, VA 22030-4444
From: election-law-bounces@mailman.lls.edu [mailto:election-law-bounces@mailman.lls.edu] On Behalf Of Paul Gronke
Sent: Sunday, October 24, 2010 11:43 PM
To: Scarberry, Mark; Law Election
Subject: Re: [EL] Nate Silver on early voting stats
ARRGH! This comparison is INVALID folks! Here are my comments on the Silver posting:
Nate YOU CAN'T MAKE THIS COMPARISON!
This is so frustrating--if Molly had even read a bit of the research in this area, she, and you, would have learned that Republicans have LONG cast the "earliest" of the early votes.
GOP voters have long leaned toward absentee ballots, and many of the states listed above (CA, AZ, CO, IA) are almost complete no excuse absentee ballots, which you (and Molly) are lumping in with early in person ballots.
We have no documentary evidence that compares the early voter turnout measured more than a week before the election with the eventual early voting turnout, the final turnout, or the registration numbers. Michael McDonald's quote in the Politico story is dead on: "The large gap could be a red herring ... we are in uncharted territory."
There is no "value" in looking at these early numbers if they are misinterpreted as indicating more than anyone can possibly interpret. The only meaningful piece of information in the Politico story is the comment about the GOP in Florida leading in the EARLY IN PERSON ballots. That is a real piece of news. The rest of this is really just unjustified speculation.
---
Paul Gronke Ph: 503-517-7393
Fax: 734-661-0801
Professor, Reed College
Director, Early Voting Information Center
3203 SE Woodstock Blvd
Portland, OR 97202
EVIC: http://earlyvoting.net
On Oct 24, 2010, at 7:40 PM, Scarberry, Mark wrote:
Nate Silver on early voting stats:
http://fivethirtyeight.blogs.nytimes.com/2010/10/24/early-voter-enthusiasm-gap-appears-consistent-with-polls/?hp
Mark Scarberry
Pepperdine
_______________________________________________
election-law mailing list
election-law@mailman.lls.edu
http://mailman.lls.edu/mailman/listinfo/election-law