Support the news
WBUR was wrong to declare a statistical tie between Senate candidates Martha Coakley and Scott Brown after seeing the results of a recent poll.
"One of the most misunderstood things in polling is the margin of sampling error," said independent pollster Scott Rasmussen, who conducted the poll in question.
"Some people believe that when you say the sampling error is plus or minus 4.5 points, that any result is equally probable within that range", Rasmussen said, "and so what they do, and apparently what you (WBUR) did is say: 'OK, it could be 4.5 points better for the trailing candidate, 4.5 points worse for the leading candidate. You've got a tie, that makes it a statistical tie.'
"In reality, in polling theory, in statistical theory, when we release a result, the most likely outcome is exactly the result that we produce. If we were to do another survey, in theory, the most likely outcome is that we'd get the exact same number, but we report results with a margin of error that we say has a 95 percent level of confidence.
"That means that 19 times out of 20, the numbers will be within 4.5 points of what we are reporting. When you work your way through all the numbers, yes, it is theoretically possible that our poll could produce a result that would show the numbers identical, but it's a pretty long shot."
Rasmussen says "there is a meaningful difference" in the level of support for each candidate.
"If it's 50 to 41, the support for Coakley could be, theoretically, from 45.5 to 55.5, whereas for Brown it could be from 36.5 up to 45.5, so those are two entirely different ranges."
So what is a statistical tie?
"If we had found the candidates a single point apart," Rasmussen said, "well, then that's way too close to call, and I would say for all intents and purposes, that's even.
"When you talk about the nine-point gap, we're probably talking about a less-than-5 percent chance that these numbers would actually turn out to be equal. When you talk about a one-point distinction, it's pretty much a tossup. If it was 50-49 in one direction, it could just as likely be be 50-49 in the other. There is no scenario within our margin of error, even the most extreme example, that would put Brown ahead in this case."
Rasmussen said he is planning to poll this race again. He expects the results will be out next week.
This program aired on January 7, 2010. The audio for this program is not available.
Support the news