Democrats & Liberals Archives

The Magic of Polling

It is well known that the specific question asked in a poll can greatly affect the results of the poll. For example, in abortion polling in 1985, over 50% said “yes” when asked if abortion is murder, but over 90 percent thought abortion was sometimes the best way out of a bad situation.

In light of these known variances, I’m confused by recent results in polling between Bush and Kerry.

Over March 3-4, FOX News asked the following question and received 44% for each Bush and Kerry:

"If the election for president of the United States were held today, for whom would you vote if the candidates were Republican George W. Bush and Democrat John Kerry?"

In constrast, over March 5-7, CNN/USA Today/Gallup found a 52% to 44% edge for Kerry asking the following question:

"If Massachusetts Senator John Kerry were the Democratic Party's candidate and George W. Bush were the Republican Party's candidate, who would you be more likely to vote for: John Kerry, the Democrat, or George W. Bush, the Republican?"

Similarly, over March 4-7, a ABC News/Washington Post Poll found a 53% to 44% edge for Kerry asking this question:

"If the 2004 presidential election were being held today, would you vote for George W. Bush, the Republican, or for John Kerry, the Democrat?"

That's an eight- and nine-point difference over questions that don't seem biased to me. Also, according to PollingReport, the names were rotated for both polls, so being first or second in the question is not an issue.

What accounts for the disparity? Was there a change from March 3 to March 5? Is it just the Margin of Error (5% for Fox and 3% for CNN/USA Today/Gallup, not reported for ABC News)? Does it matter that right-leaning Fox found more support for Bush than left-leaning CNN?

Posted by LawnBoy at March 12, 2004 10:01 AM