Here’s today’s podcast with me, Laurie Buckley, Dave Plunkett and host Bill Bronner:
The panel discuss the GOP debate, the “War on Women” and more.
Recent polls suggest that 23% ofpollsters don’t believe in polls. This is shocking and I called up severalpeople to talk about it.
A sampling of this group did notknow they were being sampled as part of a group. I point this out because pollshave indicated that responders behave differently depending on whether they arealone or part of a group. Some polls show that once people have been told theyare part of a group, they act like the group, even if they are physicallyseparated from the group.
One pollster was so fascinated by the statistics of theseresults that he decided to identify a random sampling of people that he designatedas a group, but only behind their backs. The people in the group were isolatedduring questioning and never met any of the other members in the group. At nopoint were they told they were part of the group. Even so, the results ofasking the same questions from the original survey produced similar results.What can we deduce from this? I’m not a statistician. And besides, I’m makingthis up anyway.
This brings up the troubling point that sometimes pollstersjust make the answers up. This can lead to some interesting surveys and weshould probably take a poll to find out what happens when people believe a madeup poll.
I have the results here. Sixty seven percent of peoplepolled believed that the made up poll was real. Eighty-two percent of those polledbelieved the previous statistic.
Interestingly, one of the respondents from the isolatedgroup survey suffered for years from separation anxiety that doctors could notidentify. Perhaps we should ask them some questions to see if it is areasonable hypothesis to conclude that being separated from a group that you donot know you are part of can cause the symptoms that were displayed. But Idigress, and I see that we are out of time.