POLLING EXPERTS

Posted in Rants on May 12th, 2011 by Ed

I don't have the patience or energy to tear into this in depth, but today's Right Wing Histrionic Talking Point of the Day is less amusing and more infuriating than usual. Don't you know, the AP poll that gave Obama a 60% approval rating is a result of "poll cooking" by the AP. According to the polling experts at Newsbusters. And comment sections on your local newspaper website.

Apparently the fact that the sample in the AP poll (n = 1001) did not contain the exact same number of Republicans and Democrats – that's what polls are supposed to have, after all! – is de facto evidence that AP is rigging the numbers. The poll in question had 46% Democratic identifiers and 29% Republicans. The chicanery is obvious. Pant-shitting rage ensues.

The response is unsurprising, since that is exactly how someone who knows dick about polling would interpret this.

The 46% is just about what we would expect for Democrats + Leaners. The 29% for Republicans + Leaners is lower than I would expect. In a different AP poll conducted this spring, the split was 43% D to 40% R. That's the kind of variance one gets when taking a random sample of 1000 people in a country with 200,000,000 adults. Random sampling is absolutely fundamental to polling, and the word "random" means that sometimes we will get results that are not precisely what one would expect.

Let's say there are 100,000 poker chips in a bag – some unknown mix of red and black. We mix them up so they are completely random. We don't want to count all of them so we select 500 at random to estimate the red-black mix. We find 300 black and 200 red. Therefore we conclude that the bag contains 60% black and 40% red, plus or minus a margin of error of about 5% (i.e. we are 95% certain that black chips are between 55% and 65%, but the best guess is around 60%). That is how a random sample works.

Now let's say that you call 1000 people using a random phone dialer, 600 of whom are "red" and 400 of whom are "black". But you're convinced that the bag actually contains more black chips, so you pull out a few hundred more chips, throw out all the new red ones, and add more black chips to the original sample. That is how a manipulated, biased, and egregiously fucked-with sample works.

If 35% of the population is Republican and 40% are Democrats, we would EXPECT a random sample to have 35% R / 40% D. Right? Right. And if I flipped a quarter 100 times we would EXPECT to get 50% heads, right? Right. But will I get exactly 50 heads every time I flip a quarter 100 times? Nope. I might get 51. Or 47. If I do it enough times I might even get a wacky result like 37 heads. It's random. Raaaaaaaaaaandom. That means each outcome is totally independent of the one prior. A probability is merely an expectation. A random sample of 1000 people in an electorate with 200,000,000 voting-age adults is going to vary from our expectations, sometimes considerably.

That's why polls have…margins of error. The AP poll in question has a MoE of +/-4.4%. The MoE creates a 95% confidence interval, so the data actually tell us that we are 95% certain that Obama's approval is between 56% and 64%. There is only a 5% chance that his rating is higher or lower than that. Now, if reality is on the low end of that interval, would 56% really be so surprising? He has been hovering at 50% for a while, occasionally popping up to 51-52 or dropping to 45-48. A little bin Laden bump to 55-58 wouldn't be so shocking, would it?

It would be if you're Newsbusters and, among other things, you don't understand a random sample or margins of error.

Idiots. The lot of them.

Tags:

THE INNOCENCE OF A CHILD

Posted in Rants on May 25th, 2010 by Ed

On my last visit I noticed that my 8 year-old nephew had done an "If I was President, I would…" assignment that was proudly pinned to the wall as one commonly does with a child's school work. It was as excellent as I would expect from him, starting with a guarantee of solar powered jet packs for everyone (CAVEAT: Except for Bad Guys) before noting that "taxes should be cut 75% and everyone should get everything they want."

Adults love this kind of thing. There is something inherently amusing, perhaps even heartwarming depending on how much one likes children, about watching kids respond to things they don't completely understand with their imaginations. It's the reason people always ask small kids how old they think Daddy is. Knowing full well that they lack this information and probably do not grasp the concept of age very well, we nonetheless laugh heartily when the child says "Daddy is seven" or, alternatively, "Daddy is 200."

There is quite a bit of this going on when a teacher gives second-graders an assignment such as this one. Kids don't understand politics and we know their responses will lie somewhere among cute, funny, and incomprehensible. It is very easy, therefore, to read my nephew's response and chuckle. "Ha ha! Kids say the darndest things." Unfortunately, this is usually the exact same response that voting adults give to similar, albeit more specific questions. Minus the jet packs. Adults rarely bring up that part.

The fundamental problem with our government's balance sheet, which as we all know is deeply in the red, is that decision-makers respond to electoral incentives, which in turn means that they are responsive to constituent preferences (although the degree to and conditions under which they are responsive is hotly debated in political science). And constituent preferences make absolutely no sense collectively. Everyone wants more stuff from the government, lower taxes, and a balanced budget. Come to think of it, throw in a damn jet pack while we're at it. If we're making shit up, we might as well go hog wild.

The following data are from a handful of recent polls. Note that they are not all from the same poll, but each is based on a nationwide random sample:

  • "Do you think Congress has done enough to help create jobs, or don't you think so?" (CNN/Opinion Research Feb 12-15)
    14% Has done enough
    84% Don't think so
  • "Do you approve or disapprove of the way Barack Obama is handling the budget deficit?" (USA Today/Gallup March 26-28)
    37% Approve
    61% Disapprove
  • "Which of the following comes closer to your view of the budget deficit? The government should run a deficit if necessary when the country is in a recession and is at war. OR, The government should balance the budget even when the country is in a recession and is at war." (Bloomberg Nov. 2009)
    30% Run a deficit
    67% Balance the budget
  • Making the expiring Bush tax cuts permanent (CNN/Opinion Research Apr. 9-11)
    60% Favor
    33% Oppose
  • So, there you have it. Welcome to your new job as President, kid. "Create jobs" somehow (without spending money), keep cutting everyone's taxes, and balance the budget while you're at it. Oh, by the way, don't touch any entitlement programs. This collection of preference is nowhere near as eloquent, colorful, or amusing as the policies expressed by my 8 year-old nephew yet they manage to communicate the same idea. The only difference is that adults have a lot of rationalizations that purport to make this possible; for example, we can cut taxes and balance the budget by "cutting spending", usually on something like "earmarks" or "welfare" or some other $100 million chunk that means absolutely nothing in the yawning chasm of a multi-trillion dollar deficit. And some people express wonder that we keep borrowing, year after year.

    Pardon me for being subjective, but this is a lot less cute when eligible voters do it. I prefer the apple-cheeked eight year-old version. It allows me to plausibly claim that he will mature, something that the average American will not do in adulthood and probably skipped in adolescence.

    Tags:

    SAINT RONNIE vs. SECRET MUSLIN

    Posted in Rants on December 15th, 2009 by Ed

    I took a few minutes to assemble some data on presidential approval ratings at the 10-month mark, the current President having crossed that threshold about two weeks ago. I was immediately struck by the similarities between Obama and Reagan at the equivalent point in their first terms:*

    RR-BO

    The similarity is remarkable. Where did Reagan go from here? Well, his low ratings persisted throughout 1982 and accordingly the GOP results in the midterm Congressional elections were mediocre. The majority Democrats added 27 additional seats to their House delegation while in the Senate, the Democrats remained in the minority but picked up one additional seat, leaving them at 46 (they would make additional gains in 1984 before taking the Senate back in 1986). How then did Reagan end up virtually canonized by 1988 after winning a coronation-style re-election in 1984? Well, the simple answer is that his approval ratings took a dramatic swing upward in 1983. Hmm…

    RR-vs-Unemp

    Coincidental correlation? Maybe. But the link between economic conditions and presidential approval is well-established in political science literature.*** While I recognize – and in fact base my entire Presidency course around – the fact that the President does not have a magic button on his desk labeled FIX ECONOMY, these data should suggest that achieving some tangible improvement in general economic conditions should be Obama's first and only goal at the moment. For some strange reason voters don't seem to worry about the deficit as much when they have jobs.

    * Approval ratings are now available daily, whereas in 1981 data were collected monthly. I combined the data by noting the date of each Reagan approval rating and choosing the corresponding Obama rating from that date.

    ** Unemployment data is the monthly rate calculated by the Bureau of Labor Statistics.

    ***see Erikson, MacKuen, and Stimson 2000 or Edwards, Mitchell and Welch 1995 to name just two.

    Tags:

    ON THE ROAD TO NOWHERE

    Posted in Rants on February 10th, 2009 by Ed

    One of the most frustrating aspects of being an elected official has to be the fact that the American public simply has no earthly idea what it wants or, conversely, that it wants something that makes absolutely no sense. Imagine being the chef at a restaurant with a ridiculously vague menu ("Something kinda Asian with some sort of meat in it") and diners who make requests divorced from all rules of logic ("I want something vegan but full of pork – kosher pork"). Now imagine that the diners can fire you whenever they feel that their eminently reasonable desires are not being addressed.

    The classic dilemma in American politics is that voters want more government services coupled with lower (or preferably no) taxes. They want unlimited rights combined with total security. Free trade and job security. Deregulation and effective regulation. In short, they want all sorts of things that make no sense together. The new President, if his actions during the recent economic stimulus debate are any indication, responds to these unrealistic expectations in a manner that virtually guarantees his failure.

    Many politicians make the fatal error of taking seriously public calls for cooperation or bipartisanship. Yes, opinion polls indicate that bipartisanship is popular. Name a vague positive term that doesn't poll well. What if we polled the public and asked if they like politeness? Happiness? Prosperity? Justice? Elected officials who are the bestest of friends and go on group camping trips together? Of course the public is going to respond overwhelmingly in favor of a ludicrous question about whether it wants to see the two major parties cooperate and get along.

    What is misleading about this line of questioning is the failure to place it in context of the public's dozens of other competing and contradictory wants. Sure, everyone wants "bipartisanship." But how much do they want it? Do they want it more than they want tax cuts? Is it important enough that they're willing to see Congress gridlocked to preserve it? Enough that they're willing to have Congress churn out mediocre, watered-down legislation that is little more than a monument to appeasement and committee thinking?

    Yes, the desire to play nice is inversely related to Congressional productivity and the effectiveness of legislation. If you don't understand why, try this: invite 75 people to your house and order pizza. Give everyone in the room an equal vote and unlimited input on the choice of pizza toppings. Ensure that nobody takes control of the process and says "OK, here's what we're gonna do." What happens? Well, first it will take days to make a decision – long past the point at which everyone started to starve. More importantly, the end result, the Pizza that Attempts to Please Everyone, will of course please no one. The group will either revert to the lowest common denominator, the pizza that offends no one (i.e. plain, with no toppings), or it will end up ordering the Omnibus Pizza with 75 different toppings. Either result will be a tremendous disappointment, leaving many to wonder "If this is the shit we end up with, why bother ordering pizza?"

    As Paul Krugman notes, President Obama's desire to be a "centrist" or Mr. Bipartisanship is going to mortally wound his administration in a hurry if he does not learn to curb it. Yes, people want bipartisanship. But they also want Congress to get things done, and moreover they want said things to be effective solutions to real problems. Nobody loves bipartisanship so much that he or she is willing to endure a $700 billion piece of legislation that attempted to please everyone and hence accomplishes nothing. He needs to rapidly distance himself from the saccharine campaign rhetoric, the let's-all-get-along stuff which he believes the public genuinely wants. In reality the President has a single job: get shit done. He is going to bear responsibility for whatever happens anyway. The buck stops here, remember. In a couple of years the public will not give a flying crap whether or not President Obama reached out and worked with members of the other party in crafting legislation. They will, conversely, care about why the stimulus package spent a metric pantload of money but didn't accomplish a damn thing. It is never going to be acceptable to claim, "Well, the stimulus sucked but we all got along really well and played nice while we created it, and that's all that matters!"

    No, it doesn't matter. Regardless of how many opinion polls indicate that the public wants to see everyone play nice, it will never be more important than productivity and, more importantly, success. It will be far better to craft a piece of legislation that works while telling the GOP to blow it out its collective ass than to be nice and end up with garbage. If the bill succeeds in alleviating the crisis, the overwhelming majority of the public will be too pleased to give a shit that Obama had to be mean to Richard Shelby in order to get it passed. And if it fails, the public will hardly remember or care that the bill's passage involved a heartwarming display of bipartisan cooperation.

    Tags:

    HALF EMPTY

    Posted in Election 2008, Rants on November 11th, 2008 by Ed

    I am just going to say this without applying a sucrose coating: the left blew some opportunities in this election. The "Dewey Defeats Truman" pessimism that, no matter what, Obama would find a way to lose passed from prudent to quaint to irritating to counterproductive over the final months. I and everyone else who voted for Obama wasted time and energy fretting on a race that was, for all intents and purposes, over six to eight weeks ago. We saw the numbers, refused to believe them, and kept our focus on a blowout race while marginal seats in Congress and ballot measures could have been pushed past the tipping point with our all-out effort.

    It seems odd to say that such an overwhelming Democratic win involved missed opportunities, but they were there. And they stem directly from the fact that Obama supporters stubbornly refused to accept reality – that their candidate had a statistically significant lead in nearly every state required for an Electoral College victory. Just like die-hard McCain supporters but for very different reasons, many Obamans were convinced that all that data was simply wrong. It was not the finest hour for logic. After so many years of losing, winning becomes inconceivable. Of course some degree of skepticism is healthy. It would have been irresponsible for Obama supporters to say "Fuck it, we got this one" or simply take a victory as a given. But there are opportunity costs to the singular focus on the presidential race.

    For example.

    In California, the presidential race was not competitive. Prop 8 certainly was. It would have made sense for left-wing activists to put Obama and McCain out of mind and devote full attention to state and local races of import. So what was the Bay Area Obama campaign doing? Rounding up busloads of volunteers and driving them to Nevada to campaign for Obama. Was that logical? Well, only if you managed to convince yourself that Obama's 8-to-10-point lead in the Nevada polls was more in need of elbow grease than the uphill battle to push Prop 8 below 50%.

    Now, I don't want to suggest that every California Obaman behaved identically, that no one thought of this before the election, or that no one gave Prop 8 their all. But this is anecdote is just one example of the consequences of failing to see the line between prudence and irrationality. Nevada simply wasn't that close – certainly not close enough to import volunteers across state lines. There is always a race right where you live that could benefit from your attention. I can't help but wonder, for instance, if the amazingly tight Senate races in Oregon and Minnesota would have benefitted from the attention of Obama voters in those uncompetitive states. Prop 8 – largely a victim of ignorance and a terrible campaign on the "no" side – could have. Perhaps rather than jumping to Americans' favorite political conclusion ("Let's scapegoat the blacks!" Stay classy, Dan Savage!) the "no" supporters should ask themselves if they did all they could and presented their case effectively or if they wasted a lot of time worrying that McCain would employ some manner of sorcery to forestall the obvious.

    Let's take a deep breath and get it through our (apparently) thick skulls – the polling was essentially correct and we can in fact win one every once in a while. Obama led where the data suggested he led. If anything, Rassmussen, Zogby, Strategic Vision, and other right-leaning pollsters caused the aggregate polling to understate Obama's lead. The key in future elections will be to strike a careful balance between overconfidence and neurosis. Lapsing into the latter was costly. Success can never be taken for granted, but we must do a better job of asking ourselves if we are putting our scarce time, energy, and money to the best possible use or if we are hurling them at an uncompetitive race out of paranoia. What did that last $50 million raised by Obama do for his campaign? Nothing. What could it have done for Prop 8? We don't know, but now we have to wonder.

    Tags:

    PUTTING IT ON THE LINE, AGAINST JUDGMENT

    Posted in Election 2008 on November 3rd, 2008 by Ed

    Making predictions is not my favorite thing. We know where people stand but not who's going to show up on Tuesday. It's simple to sit here and tell you which way the public is leaning and very difficult to predict how that will translate into electoral results. But I talk too much about these races to do any less than offer predictions which can be held against me at a later date. So, for your mocking pleasure, I give you the Senate races, on which I did not do half-badly in 2006, and the big race. The current Senate, for reference, is 50 D, 49 R, and one ass clown.

    Easily defended seats (22)

  • Idaho (Open): Jim Risch
  • Tennessee: Lamar Alexander (i)
  • Wyoming: Mike Enzi, John Barrasso (both incumbents)
  • Mississippi 1: Thad Cochran (i)
  • Alabama: Jeff Sessions (i)
  • Kansas: Pat Roberts (i)
  • Oklahoma: Jim Inhofe (i)
  • South Carolina: Lindsey "Chickenhawk" Graham (i)
  • Maine: Susan Collins (i)
  • Nebraska: Mike Johanns
  • Arkansas: Mark Pryor (i)
  • Montana: Max Baucus (i)
  • Rhode Island: Jack Reed (i)
  • West Virginia: Jay Rockefeller (i)
  • Massachusetts: John Kerry (i)
  • Illinois: Dick Durbin (i)
  • Delaware: Joe Biden (i)
  • Iowa: Tom Harkin (i)
  • Michigan: Carl Levin (i)
  • South Dakota: Tim Johnson (i)
  • New Jersey: Frank Lautenberg (i)

    I would bet a lot of money on these races (6)

  • Virginia: Mark Warner over Jim Gilmore (Pickup – D)
  • New Mexico: Tom Udall over Steven Pearce (Pickup – D)
  • Colorado: Mark Udall over Bob Schaeffer (Pickup – D)
  • Louisiana: Mary Landrieu (i) over John Kennedy (Retained – D). Remember when the GOP thought this was a pickup?
  • Alaska: Mark Begich over Ted Stevens (i) (Pickup – D). I don't see how Tubes can survive seven felony convictions in a race he was already trailing.
  • Texas: John Cornyn (i) over Rick Noriega (Retained – R). Noriega made some noise but failed to gain enough momentum.

    Confident, but not enough to bet money (4)

  • North Carolina: Kay Hagan over Elizabeth Dole (i) (Pickup – D). Dole's recent wingnut "godless" ad shows real desperation. As Jesse Helms' former strategist said, "The next sound you'll hear is the roof caving in on Liddy Dole."
  • New Hampshire: Jeanne Shaheen over John Sununu (i) (Pickup – D). A lot more competitive than I thought, but Sununu isn't going to hang on.
  • Oregon: Jeff Merkely over Gordon Smith (i) (Pickup – D). A very pro-Obama state will have enough carryover to give the unknown challenger a narrow win.
  • Kentucky: Mitch McConnell (i) over Steve Lunsford (Retained – R). Wishing McConnell will lose can't make it so. He hangs on by a thread.

    I am not confident, but I have a reasonable guess (1)

  • Georgia: Jim Martin over Saxby Chambliss (i) (Pickup – D). This is not a brilliant call given that Martin has never led, but he closed a large gap in a hurry and has a ton of last-minute momentum. Here is a state in which higher black turnout – which I earlier stated can only boost Democratic results by about 1% – will make a difference. It is going to be extremely close regardless, so I'll buy the "new registrants" argument here.

    I wouldn't even bet someone else's money on these races (2)

  • Mississippi 2: Roger Wicker (i) over Ronnie Musgrove (Retained – R). This is a total wild card. More than one in five eligible voters in MS is African-American. Musgrove led early, but Wicker has consistently held a small lead for several weeks. Wicker hangs on, although if GOP voters really do throw in the towel on McCain the stay-homes will really affect races like this one.
  • Minnesota: Al Franken over Norm Coleman (i) and Dean Barkley (Pickup – D). Three way races are impossible to predict. It's great that voters have a non-mainstream choice but…Barkley isn't going to win. He'll pull about 15%, which accomplishes nothing for him. How that 15% affects the Coleman/Franken balance is anyone's wild guess. Franken has not run a good campaign but the Ventura/Barkley/Reform candidates in MN take positions that are more conservative than liberal. In other words, if I have to pick I will guess that Barkley's futile campaign takes more votes from Coleman than Franken. Without Barkley, Coleman would hold his seat given Franken's flat campaign.

    Wednesday morning split: 59 D, 40 R, and that male hooker from Connecticut.

    And now the big race. It would be lazy and easy to say "Obama wins" because it is looking about 95% likely at this point. But we can put a finer point on things. Let me be clear that I am intent on missing low this year. Giving McCain every benefit of the doubt – the Mountain West, Ohio, Florida, and Missouri – he still cannot make the math work. So I will make a "Best case McCain" and "Best case Obama" map. Reality will probably fall somewhere between the two.

    The best that McCain can do, in my opinion, is Obama 306, McCain 232. In this best-case scenario I am going with Obama in an extremely tight NC race but McCain in Ohio and Florida.

    If everything goes as Obama hopes – cascading waves of excited voters swamp the polls while McCain's followers give in to despair – it's Obama 378, McCain 160. If the race is any more lopsided than this it will require Obama to win in some pretty unbelievable places.

    So for the econometricians, the 95% confidence interval is (306, 378). That is, the odds that Obama does better or worse than that are a combined 5%. I know that pessimism does and will forever reign in the Democratic Party, but if Obama does not win this race then everything we think we know about elections is utterly wrong and I will have more to worry about that this poor prediction.

  • Tags: , ,

    PLAUSIBILITY, OR: NORTH DAKOTA?

    Posted in Election 2008 on October 24th, 2008 by Ed

    The American public has the long-term memory of a fruit fly desperately trying to escape from a blazing bong. It is also, as we are all aware, light on facts. These two things, combined with a healthy dose of denial among half the population, leads to some very curious interpretations of what is or is not plausible in the context of this election.

    It has become very fashionable lately for polling numbers to be rejected out of hand because, well, obviously the results are ridiculous. Montana? North Carolina? Georgia? North Dakota? Pffffft. That's retarded. Anything which puts Obama ahead, or even competitive, in those states surrenders credibility immediately.

    When did the 2000/2004 incarnation of the electoral map become the alpha and omega of American political geography? Maybe, just maybe, there was a constant in those races (Our Leader) and two opponents who had limited appeal. In 1996 – as if ancient history like twelve years ago could ever be relevant! – Clinton/Gore won states in the deep south and 51% of the vote in West Virginia, where Obama's recent polling competitiveness has been the subject of mockery. The Democrats also carried Arizona, Nevada, and New Mexico in that race, proving more than competitive in the mountain west. Bob Dole won South Dakota that year – by 3%. Raise your hand if you realized any of this. The point is that it wasn't all that long ago that the states we now definitively classify as "red" or "blue" were competitive – not quite the Verdun-like fortresses of partisanship they are now made out to be.

    Montana? Can Obama really be competitive in Montana? Well, Montana has a Democratic Governor (Brian Schweitzer, a finalist in the VP search), two Democratic Senators (Max Baucus and future leadership-appointee Jon Tester), and a Democratic majority in the State Senate. Frankly, I'd be more suspicious if the polls indicated that Obama had no shot. Is he the favorite? No. Are polls showing the state to be competitive completely off base? Up to you, but it does not appear to be an outlandish idea.

    North Dakota? Two Democratic Senators and a Democrat in its At-Large House district. Bill Clinton had some traction there. Again, you'd be foolish to call Vegas and put money on Obama, but a poll indicating competitiveness shouldn't be rejected out of hand.

    North Carolina and Virginia both have huge African-American populations and young, growing populations overall. Northern Virginia and the Research Triangle aren't exactly backwoods GOP country. Warner is winning his VA Senate race in a laugher (after a narrow win by Jim Webb in 2006). It's not much of a stretch to see a weak GOP Presidential candidate struggling, or even trailing, in these environments.

    We will know the outcome of this election for certain in just 11 days, but pieces of data suggesting that our electoral map won't look like 2004 aren't cause for skepticism. The Bush years are over. Anyone who lived through them is likely to have a hard time believing that. But it's true. We respond differently as a nation to different candidates and, as McCain is quick to remind you, George W. Bush isn't running. Compare 1984 to 1996, 1996 to 2000. You'll see significant differences. Hell, 20 years ago California was GOP country and West Virginia was one of the mere eight states that Michael Fucking Dukakis won. What you see in 2008 simply isn't going to look like 2004, regardless of who wins. Different times, different issues, different voters, and different candidates. If a Democrat can get elected to Congress in rural Utah and Hawaii chooses a female Jewish Republican Governor, there aren't too many things that should be considered geographically implausible in American politics.

    Tags:

    DON'T QUIT YOUR DAY JOB

    Posted in Election 2008, Rants on October 21st, 2008 by Ed

    There is a divide among political scientists between those who treat polling or survey data as sacrosanct ("Of course it's reliable, look at how scientifically we collected it!") and those who consider it slightly more accurate than flipping a coin. I fall somewhere in the middle. Polling is riddled with issues that aren't easy to explain away or "correct" with post-measurement methodological voodoo (social desirability and question-order effects, for example) but a dozen polls all pointing in the same direction are a reliable indicator of a trend. I suppose I could describe myself as a believer in Zaller's "Miracle of Aggregation" theory with respect to polling – any one is of limited value, but in quantity they paint a useful picture.

    My attititude suggests, therefore, that I believe Obama is going to win. Why? Because it is essentially impossible to find a poll that says otherwise right now. He has won every nationwide poll since Palin opened her mouth and he is the clear trend leader in every important battleground state. His electoral vote total will range between 313 and 375 – a crushing victory – based on aggregated single-state polls. Polling has him ahead in utterly improbable places like North Dakota and North Carolina.

    In short, and I say this with due respect to my pollster colleagues, if Obama loses this election the entirety of the contemporary polling industry should be ridiculed into oblivion. Now that literally every single poll is pointing squarely at a solid Obama victory, his defeat would not mean simply that the polls "got it wrong." It would mean that they got it so utterly, overwhelmingly, and inexcusably wrong that the entire art, science, and industry of measuring public opinion will have to be blown up and rebuilt from scratch. This would not be "getting it wrong" like some journalist who picked the Red Sox over the Rays. This would be Dewey Defeats Truman wrong. Maginot Line wrong. They'll Hail Us As Their Liberators wrong. Coke II wrong. Historically, epically wrong.

    Could they really be that far off? Well, there are two ways to be wrong in this game – missing high and missing low. Here are a pair of logical, ostensibly plausible scenarios that illustrate how.

  • Scenario 1: McCain Wins - Let's say that there is some characteristic about likely McCain voters that makes them unwilling to admit their support. Maybe they're embarrassed or maybe they just like fucking with the librul media and its polls. Whatever the reason, they're saying "Undecided" when their preference is McCain. So in every state where the polls split along the lines of Obama 47, McCain 45, McCain will come out on top because the 8% of respondents indicating "Undecided" or "Don't Know" are really his supporters.
  • Scenario 2: Obama Hits 400 EV - Polls are often accused of undercounting young, black, and low-income voters (more on that later this week). They also under-represent cell phone users in most cases, although good organizations are correcting for that in their samples now. But for the sake of this argument, suppose that turnout among (overwhelmingly Democratic) college-aged and black voters positively dwarfs anything we've seen before. Both demographics turn out in droves, far in excess of the rate at which they are sampled in polls. Obama not only wins everything he is currently predicted to win but pulls a few "holy shit!"-style upsets in places like Tennessee, Louisiana, and Georgia.

    Is either scenario likely? We can only speculate at this point. I know enough about the guts of big polling operations – and some of the folks involved – to be certain that they have thought of these issues. Gallup et al employ high-level statistical wizards and experts in polling methodology to correct for or avoid such landmines. I have confidence in my colleagues. What I don't have confidence in is the efficacy of quantitative ways to "correct" the inherent limitations of survey-based research. When shove comes back to push, we are still basing conclusions about an electorate of over 180 million eligible voters on the responses of ~800 yahoos who are lonely enough to sit on the phone talking to a pollster (or worse, a robo-dialer) for 15 minutes.

    The error and obstacles inherent in this process means that we shouldn't be shocked if polls are wrong – we should be amazed that they're ever right. But this year, with every single indicator pointing in the same direction, there will be consequences for being wrong. The entire industry can't just chuckle and say "Well, nature of the beast!" Heads will roll, souls will be searched, and we will have to go back to the drawing board. The Smooth Jimmy Apollo excuse from The Simpsons ("When you're right 52% of the time, you're wrong 48% of the time!" "OK Jimmy, you're off the hook.") isn't going to cut it. It's not possible to blow something this badly and simply go back to business as usual.

  • Tags: ,

    BIGFOOT THE UNDECIDED VOTER

    Posted in Rants on October 1st, 2008 by Ed

    When people in my field or the broader political discourse argue over undecided voters it is rarely productive because the subject is mischaracterized. We lionize the Undecided Voter. Culturally, we believe that they are better people than the rest of us, for they are doing the idealized version of our Civic Duty: giving both candidates a fair hearing, earnestly debating the merits of each, and not deciding based on partisanship. We believe undecided voters are undecided because they have absorbed both candidates' messages and honestly can't indicate a preference.

    Cute story. Too bad that the vast majority of undecided voters are actually "the biggest idiots on the planet" as Brian Griffin said. Sure, the Mayberry image accurately describes some. There are people out there with mixed ideological beliefs who really can't decide. But the remainder, the people who essentially determine every election that isn't a blowout, are undecided because they haven't paid the slightest bit of attention to the race(s).

    Let me regale you with an anecdote. In 2004 I was asked by the local paper to sit with a group of "undecided voters" as they watched the first Bush-Kerry debate. At their request I said a few words beforehand, then I fielded my first question:

    "Who's the guy running against Bush?"

    Followed by, from a different person:

    "Which guy is which party?"

    It went downhill from there. Anecdotal evidence, but it left a powerful impression on me. Watch any of these "interactive" let's-see-what-the-normal-folk-think circle jerks that the media love so much during the conventions, debates, and so on. Excepting the minority of truly undecided voters, these people simply have no idea what in the hell they're talking about.

    Think of the electorate this way. Both parties start out with a baseline of support (40% GOP, 40% Democrat). No major party candidate has failed to win more than 38% of the vote in a two-way race. So Obama and McCain start at 40%. Each gets another 5% of "leaners" who haven't carved a decision in stone but essentially know what they're doing. Then, squared off 45% to 45%, the campaigns battle for the remaining 10% of likely voters.

    Here's the rub: most of that determinative group are essentially deciding at random. Their opinions differ based on whatever happened most recently (a debate, announcing running mates, conventions, etc), hence the day-to-day variance in polling. Barring a major gaffe, everything that will move public opinion in the next 5 weeks will be forgotten by election day. Alan Abramowitz popularized (and provided data to support) this "minimal effects" theory of campaigns in the mid-80s. He argued that looking at economic indicators and incumbent presidential approval ratings in May has more predictive power than pre-election polling. The effects of any campaign event are forgotten almost immediately. Do you remember anything from the first Obama-McCain debate? It was, by design, entirely unmemorable. The goal is simply to avoid shooting oneself in the foot; the potential benefits are negligible, so the candidates play defense. But if you're an "undecided" (i.e., clueless) it is the only information on which to base a preference. Until the next one comes along in a few days.

    In short, because undecided voters move whichever way the wind is blowing, polling creates a misleading impression of campaign effects. Think of it as a flat line with numerous peaks and valleys, each followed by a return to the mean as the causal events fade from memory. Now here's the rub.

    There is absolutely no logic to how these uninformed/undecided voters will make up their minds (if they bother to vote). Lacking substantive knowledge, they'll choose at the last minute based on whatever fragmentary piece of information they can recall on the way to the voting booth. Or they'll pick the candidate they find more attractive (seriously). Or they'll pick the one whose spouse seems nicer. Or they'll pick the one with the best hair. Or maybe they'll flip a coin. This is how the 5-to-10 percent of the electorate that will determine the outcome of the race makes up their minds. This is why the candidates are so god-awfully repetitive with their thematic talking points ("Hope" or "Change" or "America First" or "Bunnies"). They know that 80-90% of us have paid attention and made up our minds. Some of the remaining 10-20% are leaning one way or legitimately unsure (but informed). After that, it's a parade of the lame, the halt, and the ugly.

    Yes, the Undecided Voter we'd like to imagine exists, but they are like Bigfoot – rarely sighted, their existence always treated with great skepticism. I'd like to take the analogy one step further to suggest that their respective intellectual abilities are likely to be similar as well.

    Tags: ,

    FLEX YOUR HEAD

    Posted in Election 2008 on September 18th, 2008 by Ed

    It's a good day when I can introduce talk of polling with Minor Threat lyrics.

    If you need to crystallize contemporary presidential politics you could do worse than pointing to the recent (9/10-9/11) Newsweek/Princeton survey questions about Sarah Palin:

    "Based on what you have seen or heard about Sarah Palin so far, please tell me whether or not you think each of the following phrases describes Palin. What about [see below]? Does this describe Palin, or not?"
    "Has taken on her own party to fight corruption in the Alaska state government"
    "Has a record of opposing wasteful earmarks or 'pork barrel' government spending"
    "Shares your views on the abortion issue"
    "Shares your views about environmental policy and climate change"

    Keep in mind, this poll is asking Americans (and we know how much substantive political information the man-in-the-street has) questions about someone they had never heard of five days prior. In short, aside from the abortion issue – on which her position was made front and center – an individual would have to be a voracious political junkie to answer any of these questions with information beyond the campaign PR that accompanied her nomination.

    This reduces the system to the worst, most cynical brand of Newspeak: just introduce her as Sarah the Reformer and it'll stick more often than not. Sure, that strategy punts on the 25% of the population who will do some research to determine the veracity of that claim, but that's an acceptable consequence of firmly planting the idea in the remaining 75%. Don't bother finding a candidate who is a reformer or a feminist or whatever. Just nominate whoever you want and stick the label on 'em.

    The question, in essence, is not asking "Do you think Sarah Palin fights corruption?" What the folks at Princeton and Newsweek are really asking (probably unwittingly) is "Of the marketing slogans hurled at you over the past week relative to Sarah Palin, which ones managed to stick?"

    N.B. the Red Flag Polling No-No of prompting responses with information embedded in the questions. If they asked "What word comes to mind when I say Sarah Palin?" I wonder how many people would say "reformer?" On the other hand, I don't need to wonder how many will agree when the question is phrased, "Sarah Palin may be a reformer. Do you think she is?"

    Tags: