Randy Shannon – 9/27/04

This page was last updated on September 27, 2004.

Look out for dirty tactics; Randy Shannon; Beaver County Times; September 27, 2004.

Below is a detailed critique of the subject letter.

“Last week, the Gallup Organization released a poll showing Bush ahead of Kerry 55 percent to 42 percent, contradicting most polls that showed a dead heat. This was given big news coverage by USA Today and CNN, among others.

[RWC] Mr. Shannon is not correct. Most polls taken after the Republican convention showed Bush/Cheney ahead by four to seven percent, not a dead heat. A couple of polls, including Gallup, showed a double-digit lead.

“In the last two presidential elections, the actual voters consisted of 39 percent Democrats, 34 percent Republicans and 27 percent Independents. You would think a respected polling organization would sample their likely voters in a ratio similar to their voting patterns.

“Wrong.

“The CEO of Gallup is a big contributor to the Republican Party. For its poll, it interviewed 33 percent Democrats, 40 percent Republicans, and 28 percent Independents. Gallup is trying to convince the public that Bush is way ahead in order to influence the election.”

[RWC] The source of Mr. Shannon’s information appears to be “the left coaster.” It is a web site dedicated to bashing President Bush, supporting the DNC, and electing John Kerry. I’ll leave it up to you to decide if you should accept info from a partisan source as credible.

James Clifton is CEO of Gallup. When I looked up his Federal Election Commission contribution records, I found only one contribution of $2,000 for the 2000, 2002, and 2004 election years. It was to the 2004 campaign of Herman Cain (R-GA) for the Senate. Does this sound like “a big contributor to the Republican Party?”

Regarding the voter demographics, all polls try to estimate participation in the subject election in order to get a more accurate result. For various reasons, the voter breakdown is not the same for each election so simply using demographics from the last election doesn’t usually make sense. Each poll uses its own methods to estimate participation in an attempt to be more accurate than the competition. There is no incentive for any pure polling organization to skew their results. After all, their livelihood depends on their accuracy. Does anyone believe a polling organization would risk its life to skew results in favor of a particular candidate? I suppose it is possible, but not likely.

Earlier this summer it was reported a poll – the Los Angeles Times, I believe – overloaded its demographics with Democrats; I don’t remember seeing a letter from Mr. Shannon about that incident.

One point easy to miss is that Mr. Shannon doesn’t acknowledge that a significant chunk of independents must have gone to Bush regardless of the Democrat/Republican split. Let’s assume the Democrats and Republicans vote for their party’s candidate. For Bush to get 55%, he got 54% of the independent vote. Kerry received only 32%. The proportion of independents voting for each candidate would be the same regardless of the Democrat/Republican split.

“Look out for dirty tricks. It’s Bush’s only chance to steal the White House again.”

[RWC] When will folks like Mr. Shannon give up their 2000 election myth? If these guys were alcoholics, they couldn’t complete step one of the 12 steps.