More

Robert Naiman

Robert Naiman

Posted: May 22, 2008 12:02 PM

Using Statistics to Estimate the Michigan and Florida Elections That Didn't Happen


With all the ink that's been spilled over the question of how Michigan and Florida will be represented at the Democratic convention, with a much-anticipated meeting of the Democratic rules committee at the end of this month to consider this question, and with another round of speculation in the last day or so about whether Hillary would "carry the fight to the convention" over Michigan and Florida, I'm surprised that I haven't seen anywhere anyone trying to answer the following basic question:

If Michigan and Florida had voted when they were supposed to according to DNC rules, and if both candidates had campaigned there, what would have been the result?

Of course, the true answer to this question is an unknowable counter-factual. But, as any economist or social scientist could tell you, economists and social scientists answer such questions all the time, often on the basis of considerably less data than exists in this case.

Since January, 48 states and DC have voted, and we know the results of these votes. Throw out, perhaps, New York and Illinois as outliers. We also have detailed demographic data on who voted, in the form of exit polls, for example, here.

We also have detailed demographic information on who lives in Florida and Michigan - the US census.

We should also include a right-hand variable for days since Iowa to allow for a time trend. Assume that Michigan and Florida held their elections on the earliest date that would have been allowed. Also include an indicator variable for whether the contest was a primary or a caucus.

Run a multiple linear regression of the results against the demographic data, time, and the primary/caucus indicator variable. Use the estimated relationship to "predict" what the results would have been in Michigan and Florida. Anyone with the data can do this using Excel (you may have to install the statistics add-in.)

Suppose, as a first pass, that we do the following: ignore other candidates, make the left hand variable Obama's share of the Obama/ Clinton vote in terms of delegates, ignore the internal dynamics of delegate apportionment within the state, pretend that the demographics of the Florida and Michigan votes matched their demographics in the US Census (this last one should be straightforward to correct by estimating the demographics of the turnout first, but that would take another round of entering data from the census for each state.)

This would be a great project for college statistics classes that use Excel, or for any group of motivated people working collaboratively. It would take a little effort to enter the data into the spreadsheet, but if a group were working on it, they could easily divide up the task using for example a shared spreedsheet under Google Docs. Folks could publish their results on the web, so that everyone could see what the right hand variables were, and examine and compare different models (another interesting exercise.)

It would also be a nice project for some enterprising journalists.

 
 
  • Comments
  • 15
  • Pending Comments
  • 0
  • View FAQ
Comments are closed for this entry
View All
Recency  | 
Popularity
09:47 AM on 05/23/2008
That's a fun idea and even, in terms of academic analysis, a smashing one. But:

Unless the result favors Hillary Clinton, she won't accept it (and Obama doesn't have to, what with his already having won an absolute majority of the elected delegates and now coming close to turning his current majority of the declared superdelegates into an absolute--and conclusive--majority of them, too).

Still, its a fun idea.
07:04 PM on 05/22/2008
And we'd likely see a result that reverts to the mean of the national results - 51% Obama and 49% Clinton. Split the delegates....Mr. Obama wins....

While I voted for Senator Clinton, her claim to win Michigan and/or Florida in accordance with votes cast is absurd. She agreed to a set of rules along with all the other candidates; now she should play by them.

If she continues to press the issue, supporters like me will abandon her the next time around.
HUFFPOST SUPER USER
NABNYC
06:01 PM on 05/22/2008
Florida and Michigan delegates should not be allowed to vote until Hillary is officially out of this race.

Why do incumbents win? Well yes, they do receive enormous bribes to stuff their campaign piggy banks. But in addition, they are known, they have name recognition.

Everyone knew who Hillary Clinton was since 1992. A year ago, nobody even knew Obama. Any election which took place without campaigning has an enormous and unfair bias towards Hillary Clinton, the known person, and against Barack Obama, the unknown person. It is absurd to suggest the Florida and Michigan primary represents anything other than absolutely nothing.
05:52 PM on 05/22/2008
DOn't leave out of the equation that who voted is NOT representative of anything that happened in any other state...IN EVERY OTHER PRIMARY -the number of Democrats outnumbered Republicans EXCEPT in FLORIDA and MICHIGAN...IN MICHIGAN the Republicans had almost 50% higher turnout than the Democrats..That is a HUGE OUTLIER and tells how completely unreliable any result in Michigan is...who knows how many of the nonvoters (or Republican voters) would have voted Democratic if they thought it would have counted and how they would have broken down....It is fair to say that it would not have been the same as it ended up....
05:18 PM on 05/22/2008
Bob- why not just ask the famed Poblano, who has called most of the primaries better than the pollsters using essentially this technique?
03:49 PM on 05/22/2008
I was thinking that having a poll makes more sense (since we have so many people with so much experience in polling). Of course you couldn't count each vote, but the percentages would be close enough to assign delegate counts. Then the DNC could perhaps cut the delegates to a half vote each as suitable "punishment" for their states sins.
Probably you'd need several pollsters and average together the results. And of course each candidate would need to approve the list of pollsters. Hmmm that could be difficult with Hillary not approving anything that has any appearance of fairness.
This would not pass constitutional voting rights muster, but it doesn't need to! People tend to forget is that the political parties can run their own primaries and candidate selection however they want.
photo
HUFFPOST SUPER USER
RobertNaiman
Policy Director at Just Foreign Policy
06:53 PM on 05/22/2008
I totally agree with you that the parties can do whatever they want - subject of course to the constraint that the decision-makers agree. Certainly a poll today, if the decision-makers agreed, could accomplish the same result of settling the controversy. It wouldn't directly answer the question I posed: what would have happened if the elections had been held according to the rules.

Personally, FWIW, I don't see a need to cut the delegations if the allocation is based on a fair estimate of true sentiment. The states have already received the biggest punishment there is - the nominee was determined without their input. As a result of trying to "jump the queue" to have more influence, they had less influence than everyone else who stayed in line. Isn't that punishment enough?
03:30 PM on 05/22/2008
Well enterprising journalist, get started. I am eager to know.
03:23 PM on 05/22/2008
This is a very interesting exercise and certainly could prove some points about voter suppression -- less than 600,000 total votes were actually cast in Michigan in contrast with other states where voters were not discouraged by the rules, there were actual campaigns and get out the vote drives, and there was record turnout.
The actual results, even given all of that, don't really support Clinton. She won less than 50% of the votes cast in Florida, and 55% in Michigan where she was the only major candidate who chose to leave her name on the ballot. Her popular vote margin in the two states is less than impressive 51%,
03:02 PM on 05/22/2008
Sound like a plot from CBS's Numb3rs. Where's David Krumholtz when you need him?
02:40 PM on 05/22/2008
Hell, I'm all for it, if we decided the vote from that it would be a hell of a lot more fair and accurate than Hillary's way
02:36 PM on 05/22/2008
Where's Chuck Todd or David Sirota when you need them?!
photo
HUFFPOST SUPER USER
tcagle
Solar and wind energy consultant
12:31 PM on 05/22/2008
Ah, an idea man. Where's the work on this?
photo
HUFFPOST SUPER USER
RobertNaiman
Policy Director at Just Foreign Policy
01:32 PM on 05/22/2008
Give me the excel spread sheet, and I will give you the regression.
05:13 PM on 05/22/2008
You don't have your own?