Brendan Nyhan

Last week, I criticized Matt Bai’s claim that it was an “ominous sign” for Democrats that President Obama’s approval rating is under 50%.

Writing in The Hill, top Democratic pollster Mark Mellman goes even further, calling 50% approval a “magic number” for midterm elections:

[P]erhaps there are some magic numbers after all. 

Take the effect of presidential approval on midterms. We graph the relationship between a president’s approval rating and his party’s gains and losses in midterm elections, thinking of the result as a smooth relationship. The lower the president’s approval rating, the more seats his party loses.

But the pattern is not really so linear after all. There is a sharp discontinuity at 50 percent. Presidents whose approval rating is at 50 percent or above have lost, on average, just 11 seats in the House, while presidents under the 50 percent mark have lost an average of 33 seats.

Averages can obscure as much as they reveal, so pick apart the numbers. No president with an approval rating under 50 percent has lost fewer than 15 seats. The next smallest number is 26. Even a president just below 50 percent can lose a lot. When Democrats were punished with the loss of 52 House seats in 1994, President Clinton’s approval rating rested just under the 50 percent threshold, at 48…

So where does President Obama stand? Last week Gallup put his approval at 45 percent, this week at 49. The Pollster.com weighted average of all polls says 46 percent. In short, for now, the president is hovering just below what may prove to be a magic number for Democrats in 2010.

Unlike Bai, Mellman brings historical evidence to the table, so let’s consider his argument. He acknowledges the seemingly linear relationship between presidential approval and changes in House seats for the president’s party in midterm elections. However, he argues that there is a “sharp discontinuity at 50 percent” in which “[p]residents whose approval rating is at 50 percent or above have lost, on average, just 11 seats in the House, while presidents under the 50 percent mark have lost an average of 33 seats.”

First, let’s replicate Mellman’s numbers. (He appears to be excluding the replacement presidents — Truman in 1946 and Ford in 1974 — so I do the same here.) Using a cutpoint of 50 percent, I find that presidents with greater than 50 percent approval in the most recent Gallup poll before Election Day lose an average of ten seats and those below 50 percent lose an average of 33.5 seats (these slight discrepancies are likely the result of how different sources calculate seat change).

The problem is that the choice of 50 percent is arbitrary. For instance, presidents with approval ratings above 45 percent lost an average of 15 seats, while those below 45 percent lost an average of 33 seats — results that aren’t that different from Mellman’s original numbers. Going the other direction, any cutpoint from 51 percent to 56 percent will yield the same results as 50 percent because there are no presidents who had approval ratings in that range in the data.

If you prefer graphical evidence, here is the data with a standard linear fit:

If we instead use a more flexible local polynomial fit to allow for nonlinearity, the predicted values show an inflection point around 50 percent approval, but the 95% confidence intervals reveal a great deal of uncertainty in that estimate — hardly enough to justify a claim of a “sharp discontinuity”:

With so few data points, it’s very difficult to demonstrate a non-linear relationship. Absent further evidence, we can’t be confident that a discontinuity exists at 50 percent.

It’s also hard to believe the claim of a discontinuity in the context of the upcoming midterm elections. Given the state of the economy and the generic ballot, it’s clear that Democrats are likely to lose a substantial number of seats regardless of whether Obama’s approval rating is 49 percent or 51 percent on Election Day. Does Mellman believe otherwise?

Update 6/24 4:16 PM: To illustrate the point a bit further, I created a simple simulation of a linear relationship between approval and seat change that produces data approximately similar to what we observe above:

Over 1000 iterations of the simulation, the average outcome for presidents below 50 percent approval was a 12 seat loss while the average outcome above 50 percent approval was a 29 seat loss. In other words, it is very easy for a linear relationship to produce the sort of outcomes that Mellman describes.

[Cross-posted to Pollster.com]

In a post yesterday, NYU journalism prof/blogger Jay Rosen argued for more fact-checking in political journalism, citing a Greg Sargent post reporting that the AP’s fact-checks are the most popular articles that the wire service publishes:

Fact checking is good journalism. Journalists should take a lesson from the success of the fact-checking site, Politfact.com. I have already written extensively about this one, so there is no need to repeat myself.

But don’t do it unless you are willing to do what Politifact does: tell us when a political actor is lying, or speaking falsely. Drop the pretense that there must be deception in equal measure on both sides of the partisan ledger—a lie for a lie, and untruth for an untruth—just because we, the journalists, need to show how even handed we are. The AP has started doing it, and as Greg Sargent reported, “Their fact-checking efforts are almost uniformly the most clicked and most linked pieces they produce. Journalistic fact-checking with authority, it turns out, is popular.”

I agree with the sentiment, but Rosen is overstating the case — if fact-checks were unambiguously popular, they’d be a far more common practice. However, based on my experience with Spinsanity and research on correcting misperceptions, I’d expect that the online popularity of fact-checks is in large part driven by bloggers and readers who want to see their political opponents be discredited. As Jason Reifler and I showed (PDF), people aren’t generally receptive to corrections of politicians on their side of the political spectrum. This isn’t a direct problem for the AP, which provides content to other outlets, but a newspaper or website runs the risk of antagonizing half of its audience with each fact-check. That’s the principal reason the practice is so rare.

(For more on how economic incentives shape news content, see Jay Hamilton’s excellent book All the News That’s Fit to Sell.)

Reading the opening of David Remnick’s book The Bridge
last night, I was struck by how small a role race has played in Barack Obama’s presidential campaign and time in office relative to what we might have expected a few years ago. His popular vote total in 2008 was very close to what the leading models forecast, and his approval trajectory in office is very similar to his predecessors. Obama’s status as the first black president has of course manifested itself in all sorts of ways, but the basic electoral mechanics of presidential voting and approval are operating seemingly unchanged. That normalcy is itself a triumph for our democracy.

Mark Blumenthal quoted me in a blog post at Pollster.com (where I frequently cross-post) commenting on Nate Silver’s pollster ratings — here’s what I wrote:

It’s not necessarily true that the dummy variable for each firm (i.e. the “raw score”) actually “reflects the pollster’s skill” as Silver states. These estimates instead capture the expected difference in accuracy of that firm’s polls controlling for other factors — a difference that could be the result of a variety of factors other than skill. For instance, if certain pollsters tend to poll in races with well-known incumbents that are easier to poll, this could affect the expected accuracy of their polls even after adjusting for other factors. Without random assignment of pollsters to campaigns, it’s important to be cautious in interpreting regression coefficients.

Silver has expressed bewilderment at the criticism he has received and the lack of competing measures, writing that “the task is not inherently more daunting than, say, trying to determine which are the most skilled baseball players are based on their batting averages.” As the comment above suggests, I think this is incorrect.

The problem is that polling isn’t like baseball. The performance of hitters is measured hundreds of times per year in a wide range of contexts that the hitters generally do not select (setting aside platoons, pinch-hitting, etc.) and it is relatively easy to adjust for external factors such as stadiums, opposing pitchers, defense, etc. By contrast, pollsters must be hired by candidates or assigned by staff to poll in specific races. In addition, their performance (as measured by forecasts of election outcomes) is measured very infrequently — only fifteen firms in Silver’s ratings have a sample size of more than fifty polls — and is subject to inherent sampling error. For all of these reasons, it’s not clear that we can distinguish between the performance of most pollsters without a heroic level of belief in our modeling assumptions and a neglect of the error associated with the projected ratings.

From my Twitter feed:
-A high point in the history of fact-checking: Politifact investigates Snooki’s claims about the health care reform tanning tax (for the record: true)
–TNR’s Jon Chait on the “postmodern scandal” of BP’s CEO watching a yacht race — an entirely image-driven faux controversy
-I’m not sure why Eliot Spitzer is seen as a “potentially tangy” choice to host an 8 PM show on CNN — won’t the scandal make him more risk-averse?
-CJR’s Greg Marx provides the missing political science for McClatchy’s story on Trent Lott and polarization in Congress
-Kevin Drum cites public opinion in a post titled “Why Congress is at a Standstill,” but the filibuster is more important
-It’s horrifying that Utah’s Attorney General live tweeted an execution
-Politico details Darrell Issa’s aspirations to be the Dan Burton of the next Congress if the GOP takes back the House

New York Times columnist Ross Douthat correctly diagnoses the surge of Green Lantern-ism among liberals noted here a few days ago:

At work in this liberal panic are two intellectual vices, and one legitimate fear. The first vice is the worship of presidential power: the belief that any problem, any crisis, can be swiftly solved by a strong government, and particularly a strong executive. A gushing oil well, a recalcitrant Congress, a public that’s grown weary of grand ambitions — all of these challenges could be mastered, Obama’s leftward critics seem to imagine, if only he were bolder or angrier, or maybe just more determined.

This vice isn’t confined to liberals: you can see it at work when foreign policy hawks suggest that mere presidential “toughness” is the key to undoing Iran’s clerical regime, or disarming North Korea. But it runs deepest among progressives. When Rachel Maddow fantasized last week about how Obama should simply dictate energy legislation to a submissive Congress, she was unconsciously echoing midcentury liberal theoreticians of the presidency like Arthur Schlesinger Jr., who often wrote as if a Franklin Roosevelt or a John F. Kennedy could run the country by fiat. (They couldn’t.)

For those who aren’t familiar with the Green Lantern Theory of the Presidency^TM, here’s my original definition:

During the Bush years, [Matthew] Yglesias coined the Green Lantern Theory of Geopolitics* to mock conservatives who believed that “[t]he only thing limiting us is a lack of willpower” in foreign policy. What he identifies here is nothing less than a Green Lantern theory of the presidency in which all domestic policy compromises are attributed to a lack of presidential will. And, like the Green Lantern theory of geopolitics, this view is nonfalsifiable. Rather than learning from, say, the stimulus vote that Obama faces severe constraints in the Senate, liberal GL proponents have created a narrative in which all failure and compromise is the result of a lack of presidential willpower. (Hamsher, for instance, claims that “The failure to establish a public option to control medical costs and increase competition is President Obama’s failure alone.”) It’s a fantasy world…

* Here’s Yglesias explaining the Green Lantern reference:

[T]he Green Lantern Corps is a sort of interstellar peacekeeping force set up by the Guardians of Oa to maintain the peace and defend justice. It recruits members from all sorts of different species and equips them with the most powerful weapon in the universe, the power ring.

…[The ring] lets its bearer generate streams of green energy that can take on all kinds of shapes. The important point is that, when fully charged what the ring can do is limited only by the stipulation that it create green stuff and by the user’s combination of will and imagination. Consequently, the main criterion for becoming a Green Lantern is that you need to be a person capable of “overcoming fear” which allows you to unleash the ring’s full capacities…

As liberal discontent grows toward President Obama and Congressional Democrats, TNR’s Jon Chait has been documenting a resurgence of what I call the Green Lantern theory of the presidency. In this fantasy world, all legislative obstacles can be overcome through the sheer exertion of presidential will. As such, when Obama fails to overcome the sixty vote requirement for legislative action in the Senate, Green Lantern-ites conclude that he must not have tried hard enough. If you accept the false premise that the president is all-powerful, it’s totally logical!

From my Twitter feed:
-TNR’s Jon Chait slams Peggy Noonan’s retreat to the “familiar embrace of mysticism” as an explanation for presidential popularity
-Emory’s Alan Abramowitz shows that Democrats have fewer marginal seats to defend than in 1994
–NYT’s Ross Douthat on how members of the cult of the presidency demand the appearance of action from Obama on the Gulf spill
–As predicted, repeal of health care reform is a non-starter so far — proposal to roll back individual mandate loses 230-187 in House

-Kevin Drum takes down an insipid psychoanalysis of the Tea Party movement — highly recommended
-Things I hate: Disproportionate coverage of televised speeches like Obama’s — they don’t matter (much)
–Great news for making gay rights a bipartisan issue
-John Boehner — yet another high-ranking Republican who denies that tax cuts reduce revenue and increase deficits
-If you went to a college like Swarthmore and you’re not following feministhulk on Twitter, you are missing out
-People are way too carried away with conspiracy theories about the Democratic Senate primary in South Carolina — remember all the nonsense about Ohio and Florida in 2004? Time to slow down

Two important points of followup on Tuesday’s post about how Matt Bai overhyped President Obama’s approval rating as “ominous” for Democrats:

1. First, as Emory’s Alan Abramowitz correctly pointed out in an email to me, “Seat exposure and the midterm dummy variable predict substantial Democratic losses regardless of what happens to either the generic ballot or Obama approval.” See, for instance, Abramowitz’s statistical model of House seat swings, which predicts a 38 seat loss for Democrats on the basis of those two factors alone.

2. Second, Rahm Emanuel seems to have bought into the hype about the president’s approval rating as the overriding factor in midterms. In Bai’s article, he is paraphrased as follows:

For every point that Obama’s approval rating dips below 50 percent, Emanuel said, there are probably four or five more House districts that will swing into the Republican column, and vice versa.

These results are not corroborated by the statistics. Controlling for other factors, Abramowitz’s model predicts that a one point decrease in Obama’s net approval (approval-disapproval) is associated with a .22 seat shift toward Republicans. (He indicates by email that the coefficient for raw approval is less than 0.5.) Even the simple slope in a bivariate plot is far less than 4-5 seats per point of approval. Unless there’s some massive non-linearity around 50% approval, Emanuel’s estimate is off by an order of magnitude.

[Cross-posted to Pollster.com]

In the wake of President Obama’s speech to the nation about BP and the Gulf last night, it’s worth noting that his approval ratings have not been affected by the spill so far. The speech is unlikely to have a significant effect either.

I’m laying down a marker on these two points because of the likelihood that a post hoc narrative will be created in which the Gulf spill and/or the speech played a major role in Democratic losses in the 2010 midterms or a subsequent Obama defeat in 2012. This is precisely what happened after Jimmy Carter’s so-called “malaise” speech, which is frequently used to “explain” Carter’s subsequent defeat in 1980. Here, for example, is a random Omaha World Herald editorial from 1995 I found in Nexis:

A few commentators seized on Clinton’s rumination and compared it to Jimmy Carter’s 1979 “malaise” speech. Carter’s call for America to shake itself out of a national malaise and regain its self-confidence was perceived by some people as a whiny effort by a president searching for something to blame for his low ratings in the polls. It’s believed that in appearing to rebuke the public, Carter alienated voters, contributing to his election defeat in 1980.

In reality, Carter’s approval ratings after the speech, while low, were generally stable until the Iranian hostage crisis. His defeat can easily be explained by the state of the economy, which was terrible:

Remember, dramatizing an event (e.g., Carter’s defeat) is not the same thing as explaining it. It’s not clear that the malaise speech had a significant effect on Carter’s fortunes, and so far there’s no evidence that the oil spill or last night’s speech have had a significant effect on Obama’s.

Update 6/17 3:49 PM: See also Greg Marx’s take at CJR.

[Cross-posted to Pollster.com]