The Case Against Dial Groups
By Justin Gardner | Related entries in 2008 Election, Barack, McCain, Media, Polls
Dial groups are bad enough, but actually putting this madness on television as a verdict of some kind is reckless. I said as much at the beginning of my segment. First, the sample from one group is far too small to mean much. Second, turning this voodoo into a television spectacular completely distorts whatever limited research value a group might provide. Research technique is supposed to leave respondents alone and unmolested, not plopped down in front of live TV cameras. No wonder the respondents in these groups are really thinking about their key lighting and asking how you get an agent. Their minds are on anything but what they really think about the candidates.
Nate Silver illuminates the problems…
“The problem is that the squigglys may give thirty random strangers from Bumbleweed, Ohio just too damned much power to influence public perception. The squigglys influence the home viewers, the home viewers participate in the snap polls, the snap polls influence the pundits, the pundits influence the narrative and — voilà ! — perceptions are entrenched.”
Me? I watch PBS.
But yes, I agree. If the nets want to do dial groups, that’s fine, but if they want to publish that information in real time they need to have dial groups in all of the swing states, not just one.
This entry was posted on Saturday, October 18th, 2008 and is filed under 2008 Election, Barack, McCain, Media, Polls. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.
