Polls

POLLS….This is pretty much unrelated to anything, but I just wanted to point out a tidbit from that Hotline poll I linked to yesterday:

This is one of my favorite examples of why poll results should be taken with many large grains of salt: respondents lie. The actual turnout for the 2004 election among registered voters was about 70%, which means that a full 20% of the respondents to this poll flatly lied to the polltaker. This dwarfs anything related to sampling error, the only kind of error that’s ever routinely cited in news reports on polls, so instead of the usual “this poll has a margin of error of 3.5%,” how about the following standard disclaimer instead:

Due to well known problems of statistical error, question wording, question order, respondent dishonesty, and poor survey design, the results of this poll are probably not accurate to within more than 10 percentage points. Caveat emptor.

Just a thought.

UPDATE: Apologies. The Hotline poll was conducted only among registered voters, not all adults. I’ve corrected the numbers to reflect this.

Washington Monthly - Donate today and your gift will be doubled!

Support Nonprofit Journalism

If you enjoyed this article, consider making a donation to help us produce more like it. The Washington Monthly was founded in 1969 to tell the stories of how government really works—and how to make it work better. Fifty years later, the need for incisive analysis and new, progressive policy ideas is clearer than ever. As a nonprofit, we rely on support from readers like you.

Yes, I’ll make a donation