A SoapBlox Politics Blog
[Mobile Edition]
- About Us
- Email Us (news/tips)
- Editorial Policy
- Posting Guidelines
- Advertise Here

Subscribe to Michlib daily email summary. (Preview)
Enter address:

Michigan Political Blog Ad Network

Advertise Liberally

50 State Ad Network

One new idea under the sun: Poll that asks "Who will win?"

by: Grebner

Mon Feb 25, 2013 at 21:48:56 PM EST

I've just run into one of those simple ideas that make me wonder why I didn't see it myself, and makes me regret all the effort I wasted before I learned about it.

In a horserace poll, we want information that will predict who will win the election, or some closely related measurement like the size of the lead, whether the lead is growing or shrinking, and so on.  We've always assumed the best question to ask our respondents is "Who will you vote for?"  We use statistical methods to combine all the individual answers we receive into a prediction of what will happen on election day when a much larger number of people cast their ballots.  The fundamental idea is the extrapolation from a collection of YES/NO votes to the total the election Clerk will announce a few hours after the polls close.

But what if we ask a completely different question?  What if we ask, "Who do you expect will win?" 

Grebner :: One new idea under the sun: Poll that asks "Who will win?"

If we put aside our common sense idea of how a poll "should" be conducted, which simply arises from our experience with doing it the same way every time, we realize we have no way to guess whether "who will win?" is a better or worse predictor than "who will you vote for?".

We can think of arguments that point in either direction.  "Who will you vote for?" at least asks a question to which the voter should KNOW the answer, while "who will win?" requires them to guess about something they can't really know.  And the answer to "who will win?" is bound to be tainted by superficialities like media exposure or irrelevant results of previous elections.

But "who will win?" may free shy voters to whisper their secret opinions, since they're supposedly only talking about other people's opinions.  And it might possibly serve to increase the effective sample size of the poll, by asking each respondent to tell us not how ONE person will vote (the respondent) but how TEN or TWENTY people will vote (their neighbors, relatives, co-workers, and friends).

One way to settle this argument is to pile up opinions and war stories, and then divide into conflicting schools of thought, which each raise doubts about the other's competence and integrity.  If we employ this approach to deciding between them, we'll either see the dead hand of history triumph, or at least battle the new idea to an inconclusive draw.

Instead, we might TEST the two, by asking both questions of all the respondents in a large series of polls, to see which successfully predicts the election result more often.  In a wide range of high-visibility elections, "who will win?" turns out to be considerably more accurate.

From the paper by Rothschild and Wolfers, it appears "who will win?" works best within say 90 days of an election.  And most of their comparisons involved high level offices (president, US Senate, Brittish Parliament) so we don't have strong evidence about the use in local races.  I suspect the superiority of "who will win?" depends on having an election contest that is salient enough that the poll respondent might have actually talked to other people about it. So it may not work if applied to an obscure election, or one that is far in the future.

I've only started testing the new method, but it seemed to work reasonably well in predicting Lon Johnson's win over Brewer - keeping in mind that wasn't a genuine poll. For the forseeable future, I intend to ask the horserace question BOTH ways, in order to develop a better feel for the method.  According to the paper cited above, combining the two versions gives a slightly more accurate result than using "who will win?" alone.

"Who will win?" has two major advantages over "who will you vote for?".  First, it makes the most of a small sample.  To take the Johnson/Brewer race for example, assuming it was an actual poll of a randomly selected sample, imagine I had asked "who do you support?", and gotten a result of 26-to-14, which would be non-significant.  But imagine that four of Brewer's supporters in the poll had realized from what they'd been hearing, that Brewer was in trouble, and had switched their vote when answering "who will win?", yielding the 30-to-10 result we actually saw - a statistically significant margin.

Second, although the answers to the two questions are strongly correlated (meaning that the answers to "who will win?" are biased by the voter's personal preference) it turns out to be easier to correct for the bias.  As a result, even from a small sample in which one political party is over-represented, it is possible to get a fairly accurate reading of the overall state of the horse-race. 

Tags: , , (All Tags)
Print Friendly View Send As Email

I would find it exceptionally interesting to see what results
you get from that question in truly local races -- county, city, township -- I'm guessing that many will say "don't know", which is OK -- those that have an opinion, that opinion may be based on more than their preference, as in higher level races. So the advantage over the preference question may persist.  

There's a way to answer that question: research.
I plan to start testing the two formulations side-by-side.  As I said, I'm betting that "who will win?" only really works when the public begins to form a shared opinion.  But we'll see.

[ Parent ]
Local would be difficult
Having run for both a local (city) office and State Rep, I can tell you that the lack of name ID is far to big a problem for this method to be effective.

I believe that the State Rep level race would probably be even worse than the local level race, especially in districts like mine that cover a large geographical area. If the candidate doesn't have extremely good name ID before entering the race then many voters won't know who they are, even right before the election.

I think this will be effective for a very limited number of races. However, for those races (such as President, Governor and high-profile US Senate races) this could prove to be a very strong connection. In a way it would rely on the respondents to do much of the work (which can be dangerous). If someone talks to 10 friends about a race and comes to the conclusion that a certain candidate will win, then that one response is somewhat representative of 11 people instead of just one. It provides something of a multiplier effect. However, this could potentially add a multiplier to how wrong the prediction is, but my guess is that in a high profile race it won't be.

"The fool doth think he is wise, but the wise man knows himself to be a fool." - Shakespeare  

[ Parent ]
I think it would work in many situations.
I don't have enough experience with the new formulation to be sure, but I expect it will work pretty well wherever the candidates are well enough known that a pollster wouldn't receive blank looks when asking about a race.  And keep in mind that for many races, the critical electorate isn't defined by even-year November, but by August primaries or odd-year municipal turnouts.  So the survey isn't conducted among the clueless, but the relatively clued-in primary voters, or regular municipal voters.

If we ask Democratic primary voters, they may have a pretty good idea who will win the upcoming Democratic primary for state representative, or even for drain commissioner.  This knowledge may not be based on talking to other voters about the upcoming election, but upon knowledge of the candidates' strengths and weaknesses among demographic or geographic groups.  It may also tap into voters' hunches about how their neighbors will vote on ballot proposals.

In short, we'll find out just how far the new tactic can be taken.  I'm reasonably confident that it will improve accuracy in at least some situations.

[ Parent ]
Good points
I'll be very interested to see what your findings are from testing this out. I also believe that it will improve accuracy in some situations.  

"The fool doth think he is wise, but the wise man knows himself to be a fool." - Shakespeare  

[ Parent ]
Getting into the Weeds...
How do you get from the responses to a determination of who will win?

If we ask the poll question the "usual" way ("Who will you vote for?") we get an estimate of the proportion of voters who will vote for Brewer or for Johnson. The connection between that estimate and predicting who will win makes intuitive sense.

If we ask the poll question the "new" way ("Who do you think will win?") we get an estimate of the proportion of voters who think Brewer will win or think Johnson will win. How do we translate that into a sound prediction about who will win?

Also, another thought... campaigns already maintain a message of confidence even when their real chances look hopeless. If this kind of polling starts to catch on, will campaigns start manipulating public impressions even more than they do already through orchestrated demonstrations of support, whether mass rallies, staged photo ops, or whatever else they think of. (Yes, they do such things even now, but this gives campaigns a bigger incentive.)

Good questions.
On the second question, there's undoubtably some danger campaigns will try to manipulate the results, but of course winning in a poll isn't their real goal.  Unless deceiving the polls helps you win the election (see: Romney, Rasmussen) there isn't any point in wasting resources to do it.  But it's certainly something to keep an eye on.

On the first question, there are several levels at which an answer can be posed.  First, and most basic, we simply consider the marginals we get from our respondents.  On the average, we get a bigger difference between candidates when we ask "who will win?" than we get from "who will you vote for?".  And because the confidence interval is almost exactly the same for the two questions, there's a better chance of finding a statistically significant lead when we ask "who will win?".

But we can go beyond that level, if we're working from a file where we have good party coding.  By comparing the partisan composition of the people whose responses we gather in a poll to the composition of the entire electorate, we can see whether our poll over-represents one side or the other.  By re-weighting the responses (counting the under-represented side more in order to make up for their scarcity) we can fix the bias.  The authors of the paper cited showed this method worked surprisingly well, when they applied it to hundreds of small, biased samples.

But again, it's a question which bears watching and begs for actual experience.

[ Parent ]
I believe this question
was asked by one of the national polling outfits regarding the last presidential election. They discovered that a fair number of those who planned to vote for Romney, believed that Obama would win.    

Progressive Blogroll
For MI Bloggers:
- MI Bloggers Facebook
- MI Bloggers Myspace
- MI Bloggers PartyBuilder
- MI Bloggers Wiki

- Blogging for Michigan
- Call of the Senate Dems
- [Con]serving Michigan (Michigan LCV)
- DailyKos (Michigan tag)
- Enviro-Mich List Serve archives
- Democratic Underground, Michigan Forum
- Jack Lessenberry
- JenniferGranholm.com
- LeftyBlogs (Michigan)
- MI Eye on Bishop
- Michigan Coalition for Progress
- Michigan Messenger
- MI Idea (Michigan Equality)
- Planned Parenthood Advocates of Michigan
- Rainbow Mittens
- The Upper Hand (Progress Michigan)

Upper Peninsula:
- Keweenaw Now
- Lift Bridges and Mine Shafts
- Save the Wild UP

Western Michigan:
- Great Lakes Guy
- Great Lakes, Great Times, Great Scott
- Mostly Sunny with a Chance of Gay
- Public Pulse
- West Michigan Politics
- West Michigan Rising
- Windmillin'

- Among the Trees
- Blue Chips (CMU College Democrats Blog)
- Christine Barry
- Conservative Media
- Far Left Field
- Graham Davis
- Honest Errors
- ICDP:Dispatch (Isabella County Democratic Party Blog)
- Liberal, Loud and Proud
- Livingston County Democratic Party Blog
- MI Blog
- Mid-Michigan DFA
- Pohlitics
- Random Ramblings of a Somewhat Common Man
- Waffles of Compromise
- YAF Watch

Flint/Bay Area/Thumb:
- Bay County Democratic Party
- Blue November
- East Michigan Blue
- Genesee County Young Democrats
- Greed, Eggs, and Ham
- Jim Stamas Watch
- Meddling Outsider
- Saginaw County Democratic Party Blog
- Stone Soup Musings
- Voice of Mordor

Southeast Michigan:
- A2Politico
- arblogger
- Arbor Update
- Congressman John Conyers (CD14)
- Mayor Craig Covey
- Councilman Ron Suarez
- Democracy for Metro Detroit
- Detroit Skeptic
- Detroit Uncovered (formerly "Fire Jerry Oliver")
- Grosse Pointe Democrats
- I Wish This Blog Was Louder
- Kicking Ass Ann Arbor (UM College Democrats Blog)
- LJ's Blogorific
- Mark Maynard
- Michigan Progress
- Motor City Liberal
- North Oakland Dems
- Oakland Democratic Politics
- Our Michigan
- Peters for Congress (CD09)
- PhiKapBlog
- Polygon, the Dancing Bear
- Rust Belt Blues
- Third City
- Thunder Down Country
- Trusty Getto
- Unhinged

MI Congressional
District Watch Blogs:
- Mr. Rogers' Neighborhood (CD08)

MI Campaigns:
MI Democratic Orgs:
MI Progressive Orgs:
MI Misc.:
National Alternative Media:
National Blogs:
Powered by: SoapBlox