Spinning the polls

With the tide of good news sweeping President Bush and the Republicans toward the next presidential election cycle, even the New York Times is facing some tough negative spinning challenges.

The difficulty in spinning the news in favor of the Democrats has been compounded by recent events which have increased President Bush’s standing in the public opinion polls. Circumstances have thus forced the Times to reach into its bag of tricks, including a classic technique involving the use of public opinion polls to influence public opinion.

By carefully controlling what questions are asked in an opinion survey, how they are asked and when they are asked, the Times has learned to report not necessarily what the public believes, but what the Times wants the public to believe. And when the poll results fail to reflect the Times’ desired outcome, you can expect the paper to spin the results with the ease and cunning of a paid campaign strategist. Results favorable to Democrats are trumpeted on page one; results favorable to Republicans are smothered with qualifications, besmirched with irrelevancies, buried in the body of the paper, or not reported at all.

And so it was this week when the front page of Monday’s Times reported an increase in President Bush’s approval rating following the capture of Saddam Hussein (“Bush’s Approval Rating Climbs in Days After Hussein’s Capture,” Dec. 17, 2003, reporting on two sets of polls, one taken immediately before Saddam’s capture and one immediately after).

That headline, by the way, was about the only objective aspect of the article – a straightforward summary of the results. By contrast, during the Clinton years, the Times regularly biased headlines to favor the Clinton administration, as it did in “Good News for President in Latest Poll,” Jan. 23, 1996). That the news was “good” was clearly an opinion which the Times had no qualms about expressing in a headline of a straight news article on Clinton’s favorable poll results. Don’t expect them ever to do the same for Bush, but do not be surprised if you should soon see, “Good News for Howard Dean in Latest Poll.”

As for the spin, here’s how the Times reported this week’s good news for Bush:

The capture of Saddam Hussein has lifted Americans’ view of the state of the nation and their opinion of President Bush, while at least momentarily halting what had been a spiral of concern about the nation’s economic and foreign policy, according to the latest New York Times /CBS News Poll.

“At least momentarily halting what had been a spiral of concern”? How does this “spiral of concern” square with what the Times reported just two paragraphs later in the same article:

There was even a slight bump between the two polls in the number of Americans who thought the economy was on the mend, a number that had already been growing in polls since October.

If the number of Americans who thought the economy was on the mend was a number that has been growing since October, there would not appear much of a “spiral” on that score, am I correct? If there had in fact been a “spiral of concern,” the facts to support that assertion were not reported in the article.

A disturbing trend in journalism – one which has slipped into common media practice with little public scrutiny or media criticism about its ethical consequences – has been news organizations reporting, as news, the results of public opinion polls over which those news organizations have complete control of the content, conduct and timing.

You may recall a time when news organizations would report the results of opinion polls conducted by independent polling organizations, such as Gallop. Then, sometime during the ’70s, the New York Times teamed up with CBS, the Wall Street Journal with NBC, and the Washington Post with ABC, each crafting their own polls and reporting the results as “news.” What the public lost with these joint ventures was a level of objectivity and credibility that only an independent polling organization can provide.

Now, a news organization can influence how a poll is conducted, what questions are asked, how the questions are phrased and when they are asked.

Who decided, for example, that it was important to ask the public last weekend, as the New York Times did, “Should Bush attend military funerals of soldiers killed in Iraq?” It was a point on which Bush has been hounded by the Times, not just on its editorial pages, but on its news pages. Never mind that it is rare for U.S. presidents to attend funerals of fallen soldiers – the Times thought it was important to ask and to report the results in a front-page article. It is not likely that an independent polling organization would have taken up the Times editorial crusade against the president and included the question in an opinion poll.

The day after the Times reported these results on its front page, it published another article on the results of another New York Times poll. Buried on page 22 was “Candidates in Presidential Contest Are Failing to Move Democratic Primary Voters, Poll Shows,” Dec. 18, 2003. Remarkably, the writers made this suggestion at the beginning of the article’s second paragraph:

The poll suggests that Democrats clearly have a chance next year.

If you look at how CBS reported the very same poll on its website, you’ll see no such suggestion – perhaps because to make such a suggestion is to state the obvious. Of course, the Democrats have a chance. The purpose of a news article about the results of a poll is to report the results, not to provide hope for one party or another. Providing hope falls within the purview of an editorial.

Of course, if the Times wished to inform its readers of some important results of the survey, it might have stated that, on handling the rebuilding of Iraq, the Republicans were favored by the public nearly two to one over the Democrats. On making the right decisions on terrorism, the Republicans scored an astounding 50 percent over the Democrats’ 22 percent. Asked which party was better able to ensure a strong economy, the public chose Republicans 42 percent to 40 percent over the Democrats.

Details, details. What’s important to readers is the fact that “Democrats have a chance next year” and that the public believes that President Bush should attend a funeral once in a while.

Writing poll questions and reporting the results of the survey is not about what you need to know. It’s about what they want you to believe. We need to keep this in mind whenever we read the Times’ own analysis of poll results based on questions they control.