Washington Post Misleads with its Poll
Just when Obama and Clinton need the most help maintaining the confidence of their voters the Washington Post releases a poll showing both Democrats with commanding leads over John McCain. Unfortunately for Obama, Clinton and the Washington Post’s polling credibility, their poll artificially inflates the Democrats’ numbers and their method of polling leaves no doubt as to their intention, to undermine confidence in McCain. The Post accomplishes this by polling all adults as opposed to narrowing their sample to registered or likely voters.
Every pollster, pundit and political junkie knows that polling all adults gives Democrats a boost, and usually a big one. To be fair one should consider what possible reasons the Washington Post might have to poll all adults as opposed to registered or likely voters.
Might polling all adults provide insight that could be critical to the election that a poll of voters would not? Perhaps voter turnout is inching toward one hundred percent? Maybe the distribution of Republicans and Democrats in the general population parallels that of the electorate?
Of course none of these possibilities come close to the truth. The Washington Post, as will the New York Times and CBS News in short order, put out a poll of adults showing large leads for both Hillary and Obama over McCain to give Democratic voters jolts of confidence and for no other reason.
You may wonder why polls of adults inflate Democratic numbers. This occurs because those with views sympathetic to Democrats always outnumber Republican sympathizers in the general population but they do not turn out to vote as reliably. This tendency virtually always holds true. Pollsters know this. When they poll only adults in a political horserace context, therefore, they engage in intentionally biased polling.
These polls appear every four years and you can expect to see them throughout this election season. This particular penchant of establishment (liberal) media represents the biggest reason why PoliticalRef.com keeps poll averages that exclude polls of all adults.
The next time you read a political horserace poll look at the sample. If the poll failed to narrow its sample to registered or likely voters it is not a political poll but a survey of general public opinion. When polling a political race one should use a political poll. General public opinion polls have their place, but that's not in the context of a political horserace.
If you disagree just ask yourself whether political candidates rely most on polls of all adults or of likely voters in their own surveys. Of course they do not rely on polls of all adults because it tells them nothing about their place in the race.
Read the story: Both Obama And Clinton Hold Edge Over McCain
The Ref | 3/6/08 | Permalink