It was just two red-white-and-blue weeks ago that we learned, via some Hahvard dons, that attending Fourth of July parades was likely to make one develop Republican sympathies. This finding was widely reported in the press, but it was clear that no reporters bothered reading past the abstract of the dons’ paper.
Because if they did, they would have discovered that the method the professors used to “prove” their contention was absurd. They did not measure actual parade attendance and actual voting behavior, they instead examined weather records and noted whether it rained on the Fourth when the interviewees were tykes, and then asked the interviewees how they felt about political parties as adults. The dons assumed that if it rained, parade attendance was unlikely and therefore one was likely to turn into a Democrat (sort of an anti-full moon effect).
It’s happening again. Just yesterday we began hearing of a new study, this one purporting to show that even a brief glance at Old Glory was enough to tug heart strings to the right, even nigh unto eight months after being exposed to the American tricolor.
That sound plausible to you? Me neither. But the peers of Psychological Science who reviewed the paper of Carter, Ferguson, and Hassin were evidently convinced.
Here is what reporters saw (from the abstract):
We report that a brief exposure to the American flag led to a shift toward Republican beliefs, attitudes, and voting behavior among both Republican and Democratic participants, despite their overwhelming belief that exposure to the flag would not influence their behavior. In Experiment 1, which was conducted online during the 2008 U.S. presidential election, a single exposure to an American flag resulted in a significant increase in participants’ Republican voting intentions, voting behavior, political beliefs, and implicit and explicit attitudes, with some effects lasting 8 months after the exposure to the prime. In Experiment 2, we replicated the findings more than a year into the current Democratic presidential term. These results constitute the first evidence that nonconscious priming effects from exposure to a national flag can bias the citizenry toward one political party and can have considerable durability.
Sounds persuasive enough. Not let’s see how it was done.
The trio asked 396 people to participate in Experiment 1, which was of four sessions, from before the election to eight months after. Only 197 made it through Session 2, but just 71 made it to the end—and 8 participants were excluded because, apparently, they fibbed their on-line survey answers (p. 2, bottom). That, dear readers, leaves just 63 out of the original 396 from which to draw conclusions.
Even stranger was that these 63 were pre-selected to have polarized views. The researchers only used those folks “who planned to vote in a state where polling indicated that a significant margin separated Obama and McCain.” Good grief! And this was thought kosher by the peer reviewers.
But never mind. The trio asked folks a bunch of questions (such as the “Patriotism and Nationalism Scale”, etc.). They also asked for whom these folks would vote, how they felt about that, and so forth. Some of the 63 got a “small picture (72 × 45 pixels) of an American flag…in the top left corner of the survey” and some others did not. By Session 4, the folks were asked how President Obama did, etc.
It is at this point where decisions of a statistical sort have to be made. Do we look at the rate of those who voted for McCain or Obama and whether they saw the flag? Or perhaps we tally the responses of the questions by flag “exposure.” Seems simple enough, and it would tell us directly if we’re on to something.
But this is science, where nothing is easy! Instead of doing it the plain way, the researchers added a twist:
We created composite measures of voting intentions for both Sessions 1 and 2 by calculating the difference between intentions to vote for McCain and intentions to vote for Obama; higher numbers indicate a greater intention to vote for McCain than for Obama. We then regressed the centered Session 2 intentions on centered Session 1 intentions and used the residuals from this analysis as our main measure of voting intentions.
In other words, toss out the actual answers and replace them with residuals from some weird linear regression model. They go on to create many other “composite” measures, by adding the result from this question to the result from that one, and so on.
And, lo, they discovered that those of the 63—minus, I think, those who did not actually vote for either McCain or Obama, leaving how many we don’t know—who saw the flag tended to disapprove of Obama’s eight-month-old presidency.
They did another experiment of the same sort, just as odd and convoluted as the first, and, after much manipulation, came to the same conclusion as before. Which is that a brief, one time viewing of a 72 × 45 pixel picture of an American flag is apt to turn one into a raving Republican.
See the picture above. Boo!