How to Lie With Statistics example number 12,498,122

This post is by Phil Price.

Bill Kristol notes that “Four presidents in the last century have won more than 51 percent of the vote twice: Roosevelt, Eisenhower, Reagan and Obama”. I’m not sure why Kristol, a conservative, is promoting the idea that Obama has a mandate, but that’s up to him. I’m more interested in the remarkable bit of cherry-picking that led to this “only four presidents” statistic.

There was one way in which Obama’s victory was large: he won the electoral college 332-206. That’s a thrashing. But if you want to claim that Obama has a “popular mandate” — which people seem to interpret as an overwhelming preference of The People such that the opposition is morally obligated to give way — you can’t make that argument based on the electoral college, you have to look at the popular vote. That presents you with a challenge for the 2012 election, since Obama’s 2.7-point margin in the popular vote was the 12th-smallest out of the 57 elections we’ve had. There’s a nice sortable table at Wikipedia if you would like to look at the numbers.

How do you make a narrow win in the popular vote look like a mandate? Time to get creative. You need to choose some metric by which Obama’s victory seems historically large rather than historically small. You need to cherry-pick. Start looking for cherries.

Well, let’s just look at re-elections instead of all elections; does that help? It’s quite a fair thing to do: there’s a good argument to be made that if people re-elect a president, that should provide more of an endorsement than electing him in the first place because in a re-election you have a lot more information about how the guy will act if he wins. So, how does Obama’s popular-vote margin compare to other re-elected presidents? W won re-election by only 2.5 points in the popular vote, so that’s a promising start. But Clinton won re-election by 8.5 points, Reagan by 18, Nixon by 23, Eisenhower by 15, FDR by 7.5 in his worst re-election and by 24 in his best (recall, back then presidents could serve more than two terms). In fact, Obama’s popular-vote margin was the second-smallest re-election in history, second only to W. Evidently one cannot base a case for Obama’s mandate on the size of his re-election margin.

Well, what about the percent of the vote? Thanks to third-party candidates, the popular vote margin of victory is not perfectly correlated with the percentage of the vote that a candidate receives. That’s how Clinton won by substantial margins with only 43 and 49% of the vote. Maybe if we just look at the percent of the popular vote in re-elections? Obama 50.6, W 50.7, Clinton 49.2, Reagan 58.8, Nixon 60.7, Eisenhower 57.3, Roosevelt 60.8  54.7  53.3, Wilson 49.2…OK, that didn’t work either, we’ve already gone back 100 years and Obama is ahead of only Clinton and Woodrow Wilson.  But hey, wait a second, what if we include the first term too? This really doesn’t make any sense, or at least not much, but it does some magic to our list: We can knock off W (who won with less than 50% thanks in part to a Nader candidacy), and Nixon (strong third-party challenge from Wallace).  Only 8 presidents in the past 100 years have been re-elected, counting Obama; applying a 50% filter on the second election knocks off Clinton and Wilson, and applying a 50% filter to the first one as well knocks off W and Nixon. We are down to four. Genius! We have found a way to make Obama’s victory seem like a historic mandate…kind of. If you don’t think about it. It’s a great example of how to lie with statistics.

[To try to anticipate the comments some people might otherwise make, I’ll say here that I (Phil Price) greatly prefer Obama to Romney and that I overwhelmingly prefer the Democratic platform to the Republican platform. Obama won and, as W famously said, “elections have consequences.” But that doesn’t make this “only four presidents…” trivia any less laughable.]

16 thoughts on “How to Lie With Statistics example number 12,498,122

  1. From Xkcd to Bill Kristol. That’s a step down! Bill Kristol’s like Niall Ferguson without the charisma, or Gregg Easterbrook without the cleverness, or John Yoo without the colorful C.V.

  2. “But if you want to claim that Obama has a “popular mandate” — which people seem to interpret as an overwhelming preference of The People such that the opposition is morally obligated to give way — you can’t make that argument based on the electoral college, you have to look at the popular vote.”

    Fair enough, if you define a mandate as arising from the preference of The People. I don’t because the President is determined mainly by electoral votes. By that measure, which is more important in American presidential politics than the will of the people, Mitt Romney and the Republicans he represented took a beating, and this is the second time for the Republicans. So Bill Kristol and his very strange bedfellows the frothy-mouthed Democratic partisans might be wrong about this being an historical mandate. But in the here and now, things do not look good in the long run for the Republican party. The electoral college results are only one symptom among many.

      • The point of the blog post was to point out that Bill Kristol is cherry-picking history to argue that Obama has a bargaining advantage. I didn’t miss the point, I just caromed off of it to another, admittedly tangential one. I did so because I felt that some readers would interpret the post as refuting what I believe to be the truth, which is that Obama does have a bargaining advantage. It’s just that Kristol cites poor evidence for it. :-)

        • The point of the blog post was to further illustrate, with yet another example, how statistics can be molded to fit an argument. The content of the example is currently relevant, but it was not the focus of the post.

  3. I think it was actually Obama who said “elections have consequences.”

    George W. Bush said “we had an accountability moment” in 2004, referring to the election.

  4. Although I don’t think there is any evidence that Obama has strong popular support, the low voter turnout makes it hard to say; the Electoral College numbers may truly represent his mandate, because states that are locked Democratic (like NY) or Republican (like Texas) probably have lower turnout than states that are swing (like Ohio) — oh wait: do they? (I guess that’s known, but not by me.)

    What I am imagining is that voters in the locked states are less incentivized to vote, so they don’t come out in as large numbers. Since Obama won more states, with more population, maybe this effect means he has more of the “I would have voted if I lived in a swing state but I don’t so I didn’t” vote. If you know what I’m sayin’.

  5. Since Reagan, it’s been pretty much mandatory for the victorious party to declare a “mandate”, regardless of how close the election was. It’s up to the media to backfill (since these days going from evidence to conclusion is tiresome and unfashionable, much easier to start with the conclusion you want). I’m surprised that Kristol is pitching in. He must be really disillusioned at where his party has gone.

  6. Phil – I completely agree with you. Bush had no mandate back in 2004 and a 2.7% gap (possibly larger, eventually) now is no mandate either. Kristol is clearly cherry picking. ESPN does the same thing when they say, “player A is only the 3rd player in history to have more than 31 homers, 18 stolen bases, 17 outfield assists, and 100 RBI’s in one season”. I don’t know if I would call it “cherry picking” however. I think it’s more of a misleading way of framing the statistics. I think of cherry picking as just picking out one or two data points.

    • ESPN does that so often and it’s unfortunate, although it usually makes me laugh. The most entertaining is when there are only a handful of players that have ever achieved a statistical feat and a portion of those players are very obscure (certainly not in the HOF). What then is the interpretation, since the feat can no longer be used to argue potential “greatness”? Stats are fun.

  7. Perhaps you folks can’t agree on whether there was a “statistical” mandate because a mandate is not solely a statistical phenomenon. There’s no generally recognized way to define it statistically. When Republican thought-leaders are instantly conceding on important issues debated during the election season (path to citizenship, letting the Bush tax cuts for the wealthy expire, etc), debating whether Obama has a mandate for them, statistically speaking, is kind of beside the point.

    • I’ve got no objection to people claiming that Obama has a mandate, as long as they don’t use bogus statistics to make the case!

  8. To the very limited extent to which a mandate in a presidential election has any meaning at all—and see the piece by Ryan Lizza, a staff writer for The New Yorker magazine in his piece titled “The Second Term” (June 18, 2012)—it has to refer to a specific area or two of legislation that a president runs on in his winning campaign. Since Obama ran with a strategy of making the smallest possible target—devoid of any substantive discussion of issues—backed up with an unprecedented ground game, any talk of a mandate arising from his reëlection is, I think, without any meaning politically.

    • Gestur, of course there are always “low-information voters” who don’t know nuthin’, but anyone who paid any attention to the campaigns knows that (1) Obama wants to increase taxes on the wealthy, and (2) Romney wanted to repeal “Obamacare.” I’d say these were the signature planks in their respective platforms.

      Personally, I don’t think that, even if you do base your whole campaign on one thing, you get to claim a “mandate” based on the fact that an extra 3% of the voters prefer you to the other guy. But as far as this post is concerned that’s neither here nor there, if someone wants to say “Obama made it clear that he wants to increase taxes on the wealthy, and he won, therefore he has a mandate to increase taxes on the wealthy” I might agree or disagree but at least I don’t think “he won” is misleading, so as far as I’m concerned it is fine to say this. Replace “he won” with “he won over 50% of the vote in both of his elections, an event equaled only three other times in the past century” is misleading and therefor not fine.

      • Phil, referring to your: “I’d say these were the signature planks in their respective platforms”, I don’t disagree. But to say that it was in the Democratic platform and that Obama ran on it in a major systematic way, is quite another thing. And it’s that latter magnitude and systematic nature of campaigning effort—that Ryan Lizza and others who have written on the concept of mandate—use as the only way a large vote margin can be called a mandate politically. In effect you have two necessary conditions for there to be a real political mandate. And you understand that when you write: “Personally, I don’t think that, even if you do base your whole campaign on one thing, you get to claim a ‘mandate’ based on the fact that an extra 3% of the voters prefer you to the other guy.” And I’m guessing most politicians understand that at some level as well, but that doesn’t stop them or pundits from saying they won a mandate on any issue they try to champion, once reelected. But don’t get me wrong: I personally hope that Obama does stick to his position to let the Bush era tax reductions expire. It’s just that he doesn’t have a mandate on this one, and for sure on anything else of real importance to me (putting a price on carbon).

Comments are closed.