Get Over Newspaper Polling

I think it’s borderline unethical for news organizations to conduct polls and then report on them “exclusively.” If they want to pay for a poll, fine, but then open up the results to everyone to read for themselves, including other news organizations, bloggers, professional researchers — whoever. If the sponsoring organization wants to write a news story about the poll, let them do it knowing that every other news organization has simultaneous access to the same data, and will be publishing their own interpretations.

I respect the science of public opinion. I’ve been fortunate enough to work with professionals who conduct surveys that will influence big decisions by clients. Their polls are carefully worded and structured to make sure no bias creeps into the way questions are phrased. The reports on the polls, likewise, assess the data dispassionately and with a precision that clients sometimes find cruel. But useful.

Not so with newspaper polls. Oh, the polls themselves might pass muster, but you and I don’t really see the poll. We see a news article based on the poll, for which the results have already been mined for “news value.” That’s inherently unscientific: The reporter is trying to figure what will make their readers say “a-ha!” or what will make people buy the newspaper, or what adds to the sum of human knowledge…or, as many political ideologues suspect, what will help elect someone the reporter favors and hurt someone the reporter disfavors.

It doesn’t matter which motive is at play. Plugging the juicy stuff is not what professional opinion researchers do, and using that criteria alone almost always leads to a misrepresentation of the information.

According to the conservative site Powerline, that’s what happened when, earlier this month, the LA Times topped the news with a story suggesting that a high percentage of voters, 37 percent, would not vote for a Mormon candidate, e.g. Mitt Romney, the governor of Massachusetts, a Republican who has talked about running for president in 2008.

The story described Romney’s religious background in depth, and quoted a Southern Baptist Convention website as associating Mormonism with “cults, sects and new religious movements” — all of which became relevant to a political news story only because of the poll findings. The incendiary article startled Republicans, and clearly made Romney appear to be a less credible candidate. Undoubtedly, it was circulated by other Republican candidates and influenced fundraising and endorsement decisions.

But the story got the poll results wrong, Powerline’s Paul Mirengoff says:

…the underlying report shows the Times’ July 3 story to be misleading, in my view. The question posed by the pollsters was, “Just thinking about a candidate’s religion, do you think you could vote for a Mormon [or Jewish, or Catholic, or Evangelical, or Muslim] candidate.” Thus, contrary to what the Times reported, the poll does not show that 37 percent of those questioned would not vote for a Mormon candidate; it shows that 37 percent of those questioned would not vote for a Mormon candidate if they thought only about that candidate’s religion. Indeed, the report (but not the story) acknowledges that “there is nothing to indicate that numbers such as these, while certainly indicative of a basic level of resistance, are a real barrier to legitimate candidacy.” In addition, the report (but not the story) states that there is no evidence “to infer that a candidate’s religion would trump other important voter criteria such as trust, charisma, shared values. . .or the candidate’s stand on [issues].”

The story also neglects to mention that, while half of the Democrats who expressed an opinion said they would not vote for a Mormon if all they thought about was religion, independents and Republicans showed less prejudice. About 60 percent of independents who expressed an opinion, and more than 70 percent of opining Republicans, were prepared to vote for a Mormon even if they thought only about his religion. Thus, Romney’s religion would appear to be less of an obstacle to his nomination than one might infer from the Times’ story, which quotes a political science professor who states that religious-based resistance to Romney “among Southern Baptists” could be a “huge problem.”

Romney may or may not have a “huge problem” due to his religion. In either case, The LA Times seems to have had a problem reporting on its own poll.

My suggestion to the Times: Stop doing polls like this. Go out and find news. If you want to demonstrate that Mitt Romney’s Mormon faith will block his candidacy, find actual people who will say that to you. (The only quote of that nature came from Emory University’s political science professor Merle Black, who was reacting to the reporter’s questions and description of the poll, and nothing else.) But don’t hide that assertion behind the alleged science of a poll — especially if you aren’t willing to assess the poll accurately.

Truly, if all the big news organizations stopped polling now, we wouldn’t be deprived. There are lots of research companies issuing their poll results directly over the Internet now. If there was ever a time when we needed newspapers like the Times to get information from the science of opinion polling, that day has passed. In the meantime, the journalism associated with newspapers reporting on their own polls is so suspect as to have lost all credibility.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s