Readers have been lining up to holler about a recent op-ed this newspaper published a few days ago. It was written by Danny Morrison, a Bakersfield man who described his transformation from a fiscally conservative, socially moderate not-quite-Republican to an exasperated, disillusioned may-never-be-Republican.
The objections generally fell into two categories. One group found fault with Morrison’s conclusions, including his assertion that the widespread Republican malaise he described was in part an outgrowth of old-timers’ frustration with the browning of America. “And now,” wrote Morrison, who is black, “we have a black president. Ouch.”
Some perceived an implied accusation of racism — or at least discomfort with the country’s changing racial makeup.
Some objections fell into a second category: Morrison’s facts, they said, aren’t facts at all.
Morrison, making the point that the Republican establishment is out of step with prevailing attitudes, wrote: “The majority of Americans now support gay marriage, legalizing marijuana, abortion rights, raising the minimum wage, health care system changes, and immigration reform.” And most, he noted, gave the tea party primary blame for the recent government shutdown fiasco.
The Gallup organizationhas polled Americans on just about every one of these topics, and Morrison is right. The margin of approval is decisive on all but abortion: Americans still narrowly support Roe v. Wade, according to a Gallup poll from January 2013, but that support is nuanced and conditional. And support for health care reform has taken a hit since Morrison wrote his article.
But facts are no longer facts in this world of agenda-sponsored truth. Media companies that maintain journalistic fact-check teams such as the Washington Post and the Tampa Bay Times, with its Pulitzer Prize-winning Politifact, are dismissed as tools by partisans who don’t like their verdicts.
Morrison’s article inspired a second line of criticism I had heard very little of before: Opinion polls are not to be believed.
Well, some can’t, of course. Political organizations come up with phony surveys all of the time. They’re called push polls, and they’re intended to influence your opinion, not guage it. Even among legitimate polls, some are more reliable than others. A handful are rock solid, though: Washington Post/ABC News, Wall Street Journal/NBC News, Pew, Harris and Gallup, to name a few.
But how, asked one reader, can anyone expect a poll with 1,000 respondents to accurately reflect the opinions of 300 million Americans? Gallup says part of it is the principle of equal probability of selection: If everyone in the country has an equal chance of being included in a sample, that sample will be representative of the population. If you doubt the validity of that principle, go online sometime and check out Gallup’s track record in presidential elections.
Which came first, though, the opinion or the opinion poll? Unlike that other conundrum involving fowl, the answer is pretty clear. Humans had opinions long before other humans started interrupting dinner to ask them what those opinions were.
But can survey results influence public opinion? Might the opinion poll occasionally come before the opinion? I think so. Public opinion is fueled by media-driven momentum as much as by our considered response to actual events. That explains why Republicans, dragged down by zealots, can be lurching toward Whiggish irrelevance one minute and swaggering with political fertility the next as the Democrats implode.
Morrison may have his facts straight for now, but somewhere in this country, at this moment, a pollster is interviewing an ordinary American, and that American isn’t feeling the same way about things as he did last month.