All polls have to be taken with a grain of salt, but some experts suggest this time of year — between Christmas and New Year’s — is the worst time of the year to conduct a poll. Response rates — predictably — are typically lower, increasing the challenge of accurately capturing public opinion. Other pollsters argue cell phone samples and fixed opinions about the impending caucus diminish the confounding effect of the time of year. The New York Times reports:
“I don’t think it is a solid rule of polling,” said Cliff Zukin, a professor of public policy and political science at Rutgers. “But it sure was conventional wisdom. Obviously fewer people are home and they have less time and inclination to talk during the holidays.” …
“We’ve often polled through the holiday period,” said Scott Keeter, the director of survey research at the Pew Research Center. “We do have a concern that some people may not be accessible during that time because of travel, but that may be diminishing as a concern because of the routine use of cell phone samples now.”
Dr. Zukin has less concern about polls in Iowa and New Hampshire than national polls. “Folks there are more likely to have firmly based opinions, what with all the attention and thus a more knowledgeable public. I suspect the more firmly based the opinions, the less one worries about the methodology adding its own voice to the mix,” he said.
Zukin’s theory about Iowa doesn’t strike me as particularly sound, given the conventional volatility of opinion in the state and the general state of indecision that has presided there this primary season. Bottom line: Don’t take any poll results you read this week too seriously.