Princeton Election Consortium

Innovations in democracy since 2004

Outcome: Biden 306 EV (D+1.2% from toss-up), Senate 50 D (D+1.0%)
Nov 3 polls: Biden 342 EV (D+5.3%), Senate 50-55 D (D+3.9%), House control D+4.6%
Moneyball states: President AZ NE-2 NV, Senate MT ME AK, Legislatures KS TX NC

Covert decisionmaking and the Bradley effect

October 28th, 2008, 11:20am by Sam Wang

Today’s New York Times is a target-rich environment for a brain geek. In addition to my own op-ed on the brains of undecided voters, there’s an op-ed by David Brooks on irrational decision-making in financial markets. All decisions, all the time. And for my second geek, there’s the piece on polling analysis (in which I am quoted). There are a few points that didn’t fit in our article.

The undecided v. persuadables. Seemingly paradoxically, the undecided voters are not necessarily the ones that need to be won over at this point. Charles Franklin has data showing that considerable swings in the race can occur rapidly, while the number of undecided voters declines only slowly. Focus groups of undecided voters do not, on average, jump on or off bandwagons quickly. These observations are consistent with the idea that a large share of movement in a race comes from voters switching allegiances, i.e. from McCain to Obama, or Obama to McCain.

I suspect that “undecided” voters are not easily distinguishable from “decided” voters in what moves them. In past elections they have broken unevenly between the candidates. But they may have biases as strong as those of decideds. In other words, what you see in decided respondents is what you get. In general, campaigns should really think of all voters as committed, and work on persuading them.

The Bradley effect. Undecided voters may differ from the decided mainly in their awareness of their own voting tendency. Although I am unaware of data to speak to this point, this could be relevant to the Bradley effect. Poll consumers have speculated that respondents lie about their preference. But what if the respondents are in fact being as truthful as they can, and realize that their preference is different once they are in the voting booth?

I consider this hypothesis to be more likely than the idea that they feel constrained to lie to a stranger on the telephone. Lying makes little sense since respondents can give any reason, or none at all, when expressing a candidate preference.

In other words, respondents may be simply unaware that they have a race-based bias. Such a bias is well-documented. For a popular description, see Malcolm Gladwell’s book Blink.

Tags: 2008 Election

45 Comments so far ↓

  • Ken Hundzinski (use the name realtime)

    I am curious if anyone has studied whether or not the undecided voters fit into the category of being a “low information voters” (not necessarily stupid, but uninformed or not that interested). Intuitively, it feels that the less that is known the more middle of the road one can be. I would also like to know if some of these voters fit into a category of being the type of people that like underdogs and are waiting to find out who the underdog really is.

  • Rachel Findley

    I’m generally not a low-information voter in Presidential races. When I’ve been undecided (in primaries), it’s been because one candidate seems closer to some of my values, the other candidate appeals to other values, and both seem equally electable. I use the time to dig out more information about their policy advisors and records, and wait for some news that may show which one truly has a better chance at being elected.

    I’ve never been undecided in a Presidential general election, but I can imagine some voters, serious evangelicals for example, reading all the commentators on health care policy and weighing abortion and homophobia against care for the poor, human rights, and ending the war in Iraq.

    Maybe I really didn’t change my mind on the way to the polls in January. Maybe those conflicted voters really have their inner minds made up already. But when different values point in different directions, it’s hard to be sure which way to go.

  • Frank

    Interesting that “a large share of movement in a race comes from voters switching allegiances.”

    For a voter with perfect information who wants a divided federal government — the scenario that McCain is now laying out — the decision how to divide one’s votes (Democrat for President and Republican for Senate/House or vice versa) would rationally depend on both the state and national probabilities in each contest at the time.

    Re the Bradley effect (assuming there is or even was such a thing), I wonder whether persons are convinced that polls are anonymous and whether persons have become more convinced over time, which would explain the apparent disappearance of the effect.

  • Helen Good

    Norman F. Dixon’s 1976 book ‘On the Psychology of Military Incompetence’ gives early decision making as one cause of military incompetence. He gives examples of intelligence reports being ignored because a decision had already been taken. He describes a psychological inability to hear new information that corrects the defective information on which the decision was based.

  • Eddie

    So does anyone here think there is a 1:1 chance of a fort of Bradley Effect within the undecideds? Do we have any more undecideds in this election than last(s), who could be an undercover McCain crowd? This could perhaps swing FL and/or OH, which might not matter in the count to 270.

    This has probably already been brought up a lot, but my feeling isn’t that, if we saw Bradley Effect, it wouldn’t be just folks going to McCain for fear of a Black president, but also folks polling undecided/for-Obama for fear of appearing racist.

    If I wanted to vote for McCain, I would wrestle with concern over potentially-serious social problems from a Black candidate losing. Thankfully, McCain picked Palin for VP, so I don’t have to struggle with this dilemma.

  • Hans

    I would like to magnify on Rachel Findlay’s point, above, regarding the idea that undecided voters are trying to make “lesser of two evils” type choices, because they have strong political beliefs that do not map easily to a particular candidate. Rachel provides the example of a socially conscious pro-life evangelical voter (which pretty much describes me to a tee). But I could see others as well, for example, someone who is both a civil and fiscal libertarian in the Barry Goldwater-type mould.

    I wonder if there has ever been any kind of factorial analysis on polling data that would identify whether there are specific groupings of these conflicted “undecided” voters.

  • blair alef

    Pew Research’s latest poll – today – has some interesting information on the demographics of those who still remain undecided. In general, they tend to be:

    Older 65+
    Lower education level
    Lower income
    In the North East & Midwest
    Evangelical Christian

    Full details at:

  • Glenn


    An article posted to Huffington revealed that Bradley was not as far ahead in the polls as currently thought, and, during the final two weeks of the election, he ceased to campaign, effectively anyway, altogether.

    Here’s the link to the article, which includes an interview with Dan Walters, reporter with the Sacramento Bee and first hand witness to that election as press following opponent Deukmejian’s:

  • Steve Nelson

    I tend to think that “effect-based” switching (Bradley or otherwise) has more to do with entering the voting booth with a sudden mindset conversion that resolves, as you say, in a blink. The switch from a position given to pollsters has less to do with deceiving the pollster than an initial deception of oneself, when one rehearses what seems to be a self-congruent position with the pollster acting as a surrogate for the actual polling place.

    I can just as easily see the effect working against McCain – Republicans telling pollsters in a congruent rehearsal that they’re going to vote for McCain, but who blink at the last minute based on a variety of converging factors.

  • Walter Freeman

    I think that changes-of-heart at the voting booth will favor McCain if they favor anyone.

    The arguments for Obama rely more on reason, logic, facts, etc., while McCain is trying to persuade voters via appeals to emotion, fear, and a vague patriotism.

    If people change their minds at the last minute, it’s more likely to be because of a sudden attack of pathos (fear, “patriotism”, etc.) rather than a sudden attack of reason or analysis, simply because the latter is less inclined to be sudden.

  • Frank

    Prof. Freeman: Or maybe a sudden SUPPRESSION of Republican pathos will allow reason to dominate. But there is also Democratic pathos. In my own case, although reason tells me that Obama would be the far better president, I also recognize that in some ways he’s an ordinary politician (he has to be), and what carries me beyond this is the high symbolic value of having our first African American president.

  • Bruce W

    Is there any reason to believe that undecided voters are a fairly constant (though decreasing with time) group? It seems likely to me that if a number of voters are switching allegiances, then they are probably moving back and forth through undecided territory. So maybe the undecided group actually contains more people than just those who are undecided when they are polled, but also those who swing back and forth, but are at one side of a swing or the other when they are polled.

  • Oz Observer

    In Australia we have a different set of circumstances, where we have compulsory voting, plus a “preferential”voting system, in which a voter places the candidates in order of prferences, and the candidate with the least number of first preferences is eliminated, and his voters’ second preferred candidate gets his vote, and so on until someone has 50% +1.
    Compulsory means that a truly undecided voter has to at least cast a ballot, which might be “informal” and not counted, or might be effectively random. In America an undecided can choose not to vote, and it seems to me that more Republicans in this in this election might take this option, rather than put Ms Palin near the red button.
    In preferential votong, a voter with some concerns about a candidate can cast his vote for a third party, indicating his concerns, but eventually electing the candidate closest to their position> For example in Australia staunch enveronmentalists can vote Green party, but give their preferences to the Labor party, ensuring the exclusion of conservative candidates.
    In America in 2000, Nader supporters would not have ensured that Bush got in, if the preference sytem had applied.
    I might also add that the control of the voting is in the hands of a reasonably independent FederalAuthority, so that confidence in the integrity of the system is much higher than it seems to be in the USA. Compulsory voting also helps here, so that we are a little bemused by beat ups like the “ACORN”.
    Anyhow like many foreigners I’m firmly hoping for an Obama landslide, so that I can say again “God bless America”

  • Dave Kliman

    I know of at least 5 republicans who have confided in me that they will be voting for Obama, but they do not want me to tell any of their friends or family.

    If I know of five like that, I wonder how many more there are nationwide…

    If only there were no election fraud, then we’d get to see the real numbers.

  • Bruce

    I appreciated the comments from OzObserver. Instant Runoff Voting is what it’s often called elsewhere. I seriously wonder if American voters would be able to understand it, though. (We still cannot even understand the metric system.)

    It seems odd that the American government can claim to have lost track of all of its citizens every 2 or 4 years. Motor-Voter and similar bills attempted to address this but the situation seems to have gone backwards under Bush and the Help America Vote Act of 2002. In Japan the Family Register system goes back centuries. Most citizens register births, deaths, and changes of address within weeks, so the system is always up to date. I’ve never heard of any need to “register” to vote here in Japan.

    I have been waiting for your comments on this year’s situation as promised in On the Track Record of Simple Poll Aggregation. Forgive me if you’ve already addressed this issue, but if poll aggregation is so effective, how do you account for the disparity in the polls preceding the New Hampshire Democratic primary? Sometimes ALL the pollsters get it wrong, as Nate Silver said in a recent interview on Democracy Now. I’m wondering if the answer is as simple as that poll aggregation gets rid of most of the noise, most of the time. I’m wondering if you could address this or point me to where you do or have.
    Thank you and keep up the good brain work and poll work!

  • Ben


    You bring up a set of excellent points. I too had been wondering about the reasons that the United States has problems every election cycle with registrations.

    Do you feel that it is tied to the age-old “frontiersman” conditions that many Americans want to typify? The need to be “free and without restraint”, and thus, are hesitant to keep their address and affairs in order on a reoccuring basis?

    Or is it more indicitive in a poorly run government programs to keep on top of voter migration?

    I have never understood why a country such as ours, with the technological resources and abilities available, cannot institute something of a national voter program to cut down on the issues that crop up every four years.

    Have there been attempts in the past to tie voting to tax rolls? Whether you paid that year or not (meaning no requirement, not the case of someone who fails to pay) isn’t the important part, but having the tax ID, and thus, the eligibility to vote would be.

    I know this needs to be developed a lot further before it’s really a coherent thought, but the next four to eight years really should be spent attempting to solve the problem.

    Thank you Sam for keeping the site active, it is the first stop in my daily browsing!

  • Sam Wang

    Bruce W – I wrote about the dynamics of how voters switch when I made my initial post on the neuroscience of being undecided. The evidence against “jumping on the fence, then off the fence” is that (a) the two-candidate margin changes much faster than the number of undecided voters, and (b) focus groups of undecided voters do not show sudden movement of a majority of the group.

    Bruce – The New Hampshire primary is an interesting case. Primary voters are high-information and therefore sensitive to events. New Hampshire’s experience in Presidential elections makes their citizens exceptionally engaged. Recall Hillary’s moment of emotion right before that election.

    In any event, focusing on a single example is cherry-picking, the kind of thing that leads people to believe that there is a Bradley effect. But it does give poll geeks something to talk about! On average, my statement is very well supported by empirical evidence.

    Oz Observer and Bruce – Instant Runoff Voting is an excellent idea. For it ever to take hold we would need a wave of good-government sentiment. Such feelings are more commonly found in the West and in places like Minnesota and Wisconsin. I think a sustained collapse of the balance between the two major parties might also set the stage for such reform.

  • William

    Instant Runoff voting *seems* like a good idea, but it has the major problem that a person can be preferred over every other candidate by a majority and yet still not win. This is called the Condorcet criterion and there are voting systems that follow it.

  • Snowball

    Naive question: Why isn’t proportional representation used by states to assign their electoral college votes?

    It seems like the fairest thing to do, opening up the race to other parties (could this be why it’s not used?). As an independent liberal, I want my voice heard, WITHOUT being ‘blackmailed’ into voting for a Democrat (who doesn’t accurately represent my views).

    It would also serve to radically transform campaign strategy. Personally, I find it comical that the candidates have to pander to a relatively small percentage of the population in the battleground states. I feel we are being held hostage by a minority!

  • Vicki Vance

    Hi Sam,

    Below is a quote from McCain’s campaign with their prediction that the election is going to be too close to call and their reasoning. What do you think about their analysis?

    Despite widespread polling to the contrary, McInturff wrote that “the campaign is functionally tied across the battleground states … with our numbers improving sharply over the last four tracks.”

    The pollster said that the number the campaign is watching is “Sen. Obama’s level of support and the margin difference between the two candidates.”

    “As other public polls begin to show Sen. Obama dropping below 50 percent and the margin over McCain beginning to approach margin of error with a week left, all signs say we are headed to an election that may easily be too close to call by next Tuesday,” he said.

  • William

    Technically, Electoral College decisions are made by the states. Proportional EVs are in place in Maine and Nebraska(although two each are still winner-takes-all), and there were plans to implement it in North Carolina by the local Democratic government there but Howard Dean had it called off because it was driving something similar in California.

  • Bruce (B)

    I double-checked about Japan. It seems every eligible voter is sent a postcard. It tells you your voting place. Take it there. No ID required– the postcard is the ID. Vote. End of story. I haven’t heard of any problems, but I suppose if someone took your card from your mailbox you might see on the news/campaign posters that you were supposed to get it by now, and could go to the local city/ward office to report it missing, have that number cancelled, and get a replacement issued.

    State governments in the United States DO use public data, not to *register* voters automatically, rather to systematically *exclude* them.

    Why? You only need to look back at history. The history of American democracy began with white male landowners electing representatives indirectly. This has expanded to all races and women with some resistance by those who held more power, right up to the 1960s and voter suppression, voter exclusion is continuing in an attenuated form up to today. Many people in the USA seem to think voting is a privilege, not a right. In contrast, democracies that came of age later or in different conditions, such as Australia, have written voting into law as a required duty.

    If you have a driver’s license or state ID, have a passport, have applied for any public services, have a social security number, answered the census, etc you should be registered to vote already. Perhaps this function could be outsourced to Google. They will have the voter lists drawn up by Friday. ;-)

    Instant runoff voting may have some flaws, as no system is perfect, but it is an improvement over 2 choices, winner-takes-all.

    In my opinion, proportional representation tends to produce parliaments with all kinds of factions. In the US, if you tried a national proportional representation, it might produce many small, ethnically based parties, as well as a variety of tiny fascist, communist, socialist, Marxist, Trotskyist, and other narrow ideological parties, who would then have to reach a common ground and pass laws in the legislative chamber. There would be no party discipline, just unstable coalitions, and fewer shared assumptions needed to have a conversation. (It would be Italy.)

    Having to compete to get the most votes drives the number of parties down to two or low single digits. If 2 parties continually got 30%, losing to another party that got 40%, they would have pressure to unite and resolve their differences before they could get into the legislature. This drives the creation of large, multiethnic, broadly based mass parties like the Republicrats and the Democans, competing for the vote of Mr and Mrs Fifty-one Percent. It’s probably actually a good thing. The reason small/third parties are shut out more than they naturally would be is because of restrictive laws. The instant runoff voting system would help to crack open the two-party system and maybe give us 3 to 5 viable parties like Canada has. (Sorry to write too much and go off the undecided brain topic.)

  • William

    As a sidenote, having such a plan would disproportionately benefit small states which tend to lean very Republican.

  • Michael

    Have there been attempts in the past to tie voting to tax rolls? Whether you paid that year or not (meaning no requirement, not the case of someone who fails to pay) isn’t the important part, but having the tax ID, and thus, the eligibility to vote would be.

    I am not a lawyer, and I don’t play one on TV, but I’m pretty sure this would be illegal under the voting rights act. Even if there is no intent to institute a poll tax, any attempt to link voter rolls to tax rolls is going to beg to be litigated.

    “As other public polls begin to show Sen. Obama dropping below 50 percent and the margin over McCain beginning to approach margin of error with a week left, all signs say we are headed to an election that may easily be too close to call by next Tuesday,” he said.

    Of course, in most of these key battleground states Obama has not dropped below 50% and the margin has not fallen to within the MOE, and McInturff is careful not to say that it has. He just implies that things might be trending that way. But what do you expect McCain’s chief pollster to say? I don’t think, “We’re toast.” is an option for someone in his position.

  • Bruce (B)

    Even if there is no intent to institute a poll tax, any attempt to link voter rolls to tax rolls is going to beg to be litigated.

    I think it could work as a pre-registration system. If you filed an income tax last year and have not moved since, you are registered to vote. If you have a valid drivers license/ID, are over 18, and have not moved, you are registered to vote. If you have a passport and haven’t moved since you applied for it, you are registered to vote. Perhaps there could even be a box or attachment on your census form where you register to vote. Thus, 95% of the voters could already be registered and be encouraged (if not required) to vote. Only if you have moved recently, might you need to make it a point to register.

    I don’t think there are hundreds of thousands or millions of people voting twice, but there are hundreds of thousands of people being excluded, prevented from voting. Exclusion is the worse crime.

    There’s always the dip-your-finger-in-indelible-purple-ink method if we can’t handle high-tech.

  • William

    By the way, we already have a system that attempts to seek out every single person in the United States. It’s called the Census.

  • Todd S. Horowitz

    Assigning electoral votes proportionally does not solve the problem of being blackmailed into supporting a Democrat when you really want to vote for the Rainbow Granola Party (RGP) candidate. It merely transfers to the level of the Electoral College the problem we saw eight years ago in Florida. Let’s say that, if forced to choose between the Republican and the Democrat, 51% of the people would vote for the Democrat and 49% for the Republican. Of course, given today’s geographical distribution of voters, the extra influence that the Electoral College gives to small states would magnify the Republican vote and shrink the Democratic vote, but let’s forget about that. Now assume that all the states divide up their electoral votes proportionally, and 5% of voters who would prefer the Democrats to the Republicans now vote for the RGP. Now the Republicans are on top, with 264 electoral votes to the Democrats’ 247, with 27 for the RGP. Technically, since the electoral college requires a majority, this would then throw the election into the newly-elected House of Representatives, which would make the outcome unpredictable.

    The point is that the presidency is a winner-take-all race, whether the state electoral votes are assigned in a winner-take-all fashion or in a proportional fashion, or even if we had a straight popular vote. Unless we adopt some version of “instant runoff” voting, as proposed by Oz Observer, voting for a “third” party will necessarily function as an abstention from the two-party race, so voting for the RGP instead of the Democrats functions to strengthen the Republicans.

    Note that third parties can win electoral votes under the current system, as long as they are regionally strong. Most examples of this have been southern parties, such as the” Dixiecrats”.

  • Ben


    If the Census was conducted yearly, I would agree with you.

    Does anyone know if in the National ID proposals that were pitched a few years back, was there a tie in to elections and voter registration?

  • Snowball


    Let me put it a different way. With the current system, if I vote for the Rainbow Granola Party (RGP), my vote is almost completely wasted (unless, of course, I use the loophole of persuading enough voters in a small southern state- A perverse way of doing things in a democracy, don’t you think?).

    My only option as a true believer in ‘Granola for All’, is to either stay at home on election day, or vote for the lesser of two evils (i.e. the Democrats).

    On the other hand, proportional representation means that my vote is (almost) never wasted. Incredibly, I have the opportunity to actually elect someone who represents me into office! I’m not talking about government- I’m talking about getting my voice heard on important issues.

    The best example I can think of is Iraq. I am incredibly angry at the Democratic party for voting to give Bush the power to go to war Forget about the ‘we were duped’ BS. EVERYONE knew Iraq was not about terrorism and did not have WMDs – Either that, OR they were incredibly stupid and believed the blatant lies. Either way, I cannot vote for people like this.

    I therefore want a way of expressing my anger. How can I do this when BOTH parties of the 2-party system supported the war in Iraq?

    Stay at home? Vote for someone who doesn’t represent me? This isn’t my idea of democracy.

    On a side note- Proportional allocation of EVs would also make sure that winning the popular vote wins you the election (well, with a much higher probability anyway).

  • Snowball

    I forgot to mention something obvious, but important-

    Some countries with proportional representation will only allow candidates to be elected to parliament from parties that reach a certain threshold of the national vote (e.g. 5%). So a candidate with strong localized support -for example- cannot be elected. This also ensures that you don’t end up with a dozen parties struggling to pass legislation.

  • Todd S. Horowitz

    The system of proportional representation you propose works for a parliamentary system. In theory, states in the US could allocate their representation in Congress in this fashion. However, my point was that it won’t work for the Electoral College. Even if states divided up their electoral votes for president according to the proportion of votes for each (qualified) party, your votes for the Rainbow Granola Party are still “wasted” (except insofar as they convince the Democrats to go for the granola vote in the next election).

    If we switched to a parliamentary system, the problem doesn’t go away entirely. Now you have someone who represents you ideologically (though not geographically, since there’s a tradeoff between having a local rep and having proportional representation), but since governments in such systems typically require coalitions of several parties, you have a less direct impact on who actually governs the country, since the coalitions have to be worked out in “backroom deals” between the parties.

  • Bruce W

    Sam, thanks for your response. I think the only way to be sure about the allegiance-switching behavior would be to poll not only people’s current choice, but the recent history of their preferences (though it would probably be more accurate to just poll the same people multiple times to see how many change their preference and how often).

    On another note, I wonder if there is something analogous to the Bradley effect when some people are questioned about their previous voting behavior and current intent to vote. There may be people who tell the pollster that they voted in the last election and they intend to vote in this one because they think they are supposed to vote and there is a certain stigma associated with not voting. So maybe they’re really not likely voters at all. The recent Pew poll indicates that they tend to be little old ladies living in the northeast who aren’t well educated and go to church regularly — if that group can’t decide between Obama and McCain by now, they’re probably not going to vote at all.

  • Todd S. Horowitz

    @ Snowball

    “On a side note- Proportional allocation of EVs would also make sure that winning the popular vote wins you the election (well, with a much higher probability anyway).”

    This would only be true if there were no third parties involved. As I noted earlier, if a sufficiently popular third party denied the leading candidate a majority of the electoral votes, then the election is determined by the House of representatives, which is if anything a less representative vote than the electoral college. Also, proportional representation would not eliminate the distortion caused by allocating proportionally more electoral votes to small states than to large ones. If you’re interested in ensuring that the President is the winner of the popular vote, then you should just endorse a popular vote.

    However, the proportional allocation system would have one important advantage: it would eliminate the “battleground state” phenomenon, where candidates are campaigning in (and therefore paying attention to the needs of ) only states where the outcome is not in doubt. Unde the current system, my vote in Massachusetts is largely a pointless gesture, and no one is going to try to get me to vote for them. Under a proportional system, McCain would have an interest in enlarging his share of the Massachussetts vote, even if it were less than 50%, while Obama would have an interest in denying votes to McCain. In that case, public transit funding might actually be discussed in the campaign!

    In any case, I don’t see any advantage to proportionally dividing up the electoral votes that would not accrue from simply abolishing the electoral college and going for a straight popular vote.

  • NeuvoLiberal

    Sam, just came across your site. Nice work! I like your method/approach for getting the precise probability distribution profile for electoral college outcomes.

    Do you have scilab versions of your matlab scripts? Since scilab is free, everyone can run the scilab version of the scripts on their computers, if available . I’ll try to find some time to try and convert the scripts and I may drop you an email on this. — best, NL

  • Bruce (B)

    It has its merits, but I think the time for the Electoral College may have passed a century or two ago. It’s like old MS-DOS code stuck in your OS. I don’t think you will ever make a radical and sudden change to the system, though. Amending the Constitution takes time and may face opposition, but there is an easier way. The simplest solution is a National Popular Vote.
    Under the U.S. Constitution, the states have exclusive and plenary (complete) power to allocate their electoral votes, and may change their state laws concerning the awarding of their electoral votes at any time. Under the National Popular Vote bill, all of the state’s electoral votes would be awarded to the presidential candidate who receives the most popular votes in all 50 states and the District of Columbia. The bill would take effect only when enacted, in identical form, by states possessing a majority of the electoral votes—that is, enough electoral votes to elect a President (270 of 538).

  • Frank

    John (the) Zogby on C-SPAN just said that last night’s media blitz was “optimally” timed, in terms of days to election, for the undecided voter.

  • Michael

    The chances of the National Popular Vote initiative ever succeeding are only slightly better than the chances of a constitutional amendment passing. Either one requires a large number of state legislatures voting against what their constituents view as their states’ interest. This aproach changes the arithmetic, but not the dynamic.

    Also, do we really think that this approach would do any good in a scenario like the 2000 election, where the national popular vote was decided by less than 544,000 ballots out of more than 100 million cast? Republicans would have gone to court demanding a nation-wide recount which would likely have dragged on even longer and involved even more litigation than what we actually witnessed. And of course, the SCOTUS would have taken their side, 5-4.

  • Todd S. Horowitz

    @Bruce (B)

    I don’t see that the Electoral College has any merits.

    The National Popular Vote bill is a clever idea, but it’s a kludge. If you can persuade state legislators in states with a majority of electoral votes to enact this bill, you can presumably persuade them to enact a bill amending the constitution to abolish the electoral college. The problem with the NPV bill is that if only a few states pull out of the compact, it falls apart. So you could have a situation where the popular vote determines the Presidency for an election or two, and then we revert to the current system when popular vote opponents take over enough state legislators. By comparison, if you can take advantage of the political climate to change the constitution in favor of the popular vote, that would be very hard to undo.

  • ak

    Ben, Bruce et al., one reason the business of tying electoral rolls to tax IDs or census rolls won’t work is that there are plenty of perfectly legal taxpayers and or residents that are not eligible to vote because they are not citizens. The census of course specifically stays far away from the issue of the legality of a persons residence in order to get a fair estimate. Actually that’s sort of true of the tax system too. Social Sec calculations for example assume a fair chunk of change from people that are never going to claim a check…

    The mechanics of the way elections are run by partisan hacks at local levels is what really amazes me each cycle. That, and the need for the most sophisticated voting gizmo instead of the most effective one. Note that elections in India, with large numbers of illiterate voters, are all electronic with deliberately obsolete technology:

    Going a bit further off topic I think the fears about unstable govts. due to multiple parties and comparisons with Italy are a bit overdone. Take a look at the ridiculous number of registered party in India for example:

    For the longest time there was basically one dominant party. When that started falling apart there were all sorts of dire predictions of unstable governments. That did happen the first couple of times, but then the self-interest of politicians not wanting to face continuous elections took over and the last two coalitions have lasted the entire 5 year term. This does lead to inordinate power to some smallish players, and some that appeal to fairly fringe elements. But as we have seen in the last few weeks the two party system is perfectly capable of pandering to and promoting the fringe.

  • Oz Observer

    when thinking about how elections are held..and won, I think the “Wigwam”convention of 1860 is a fine example of an exciting drama which happened to produce a miraculously good result. I’ve always thought of the many parallels between Obama and Lincoln (Illinois, one term in Congress, opposed a jingoistic foreign war etc) and if you compress the time frame, Obama winning over the Super delegates is much the same as the Lincoln team’s victory in Chicago. Let’s hope his victory doesn’t provoke an Alaskan Secession!

  • Andre Washington

    At this point there are no undecided voters. Just undeclared voters. These are the Bradley voters that the experts say will break heavily to McCain.

  • Philip Stoddard

    Sam, I enjoyed your paper in Brain Behavior & Evolution. I had no idea you were the same Sam Wang. Now I’m doubly impressed. When do you sleep?

  • Andy Daniel

    I wanted to comment on the Bradley effect and then ask if anyone had verified some data…

    I grew up in Quebec, where in 1976 a separatist government swept into power on a platform of separating Quebec from Canada. They had run several times on this platform before and lost, they then softened it to holding a referendum first and won handily.

    The referendum was the dominant political event of 1980. The government of Quebec openly advocated the “Oui” (yes – separate) position, supported by throngs of young French Quebecers with large blue “Oui” bumper stickers. The English and older Quebecers had very small round red “Non, Merci” stickers (no – keep Canada together). The environment made it uncomfortable to publicly support the No position. Quebec was a sea of “Oui”.

    Polls showed the referendum passing slightly – by 3-4% if I recall. Even as a 20-year old, I remember reassuring my worried mother saying that it wasn’t enough – the real vote would be against separation – and I was correct.

    I think the Bradley effect is real, but I wonder if someone could analyze state-by-state data on it. I would assume that, for instance, on same-sex marriage, polls would overstate support in an openly liberal state like California, but understate support in a conservative state like Alabama.

    I also think that the Republican’s organized anti-Obama rhetoric, while possibly effective in gaining votes, also creates the same effect where it is a little tougher to publicly state support for Obama then for Romney, who has been the subject of negative portrayals of course, but less vitriolic than those against Obama, and for a far, far shorter time.

  • Andy Daniel

    My frustration with the Electoral College is not so much that it sometimes produces a result different than the popular vote, it’s that it forces voters – by virtue of being alive – to often cast a vote against their conscience. In that respect it is worse than disenfranchisement, which is the loss of half a vote (same as choosing not to vote).

    In an admittedly ridiculous example, if every democrat in Florida had gotten together just before the last census and, deciding that the Florida election would go to the Republicans, chosen instead to commit mass suicide (or move to Georgia), Al Gore would have become president. I’m obviously not recommending suicide here, just making a mathematical point.

Leave a Comment