Polls must be taken on their own terms
Monday, January 14th, 2008One of the big aftershocks of the New Hampshire primary election for President was the failure of the polls to “predict” the results.
Nevermind that pollsters steadily insist they are not in the business of predicting. Rather, they are in the business of describing how people think at a particular time on a particular topic.
It is abundantly clear from conversations with folks just before and after the voting in New Hampshire that many voters made up, changed, re-changed and then changed their minds again at the very last minute. Many said they came to their decision only in the voting booth.
Thus questions about which way they were “strongly leaning” a couple of days earlier had little predictive value. In addition, the polls focused on people with opinions, not the great numbers who were steadfastly undecided.
There were other issues as well, including the fact that a fair number of folks voted absentee and thus missed the survey. And events on the campaign trail were red hot in the final 24 hours or so.
So, the polls might have been “right” about how folks were thinking leading up to the voting, but wrong about how they actually decided.
Does this mean we should ignore polls? Hardly. They help measure public sentiment and give us a good cross-check on the election outside of the spin machines of the campaigns themselves.
So whether the polls are local in Hawaii or national, take them for what they are. Use them as part of your decision-making process, but don’t let them become a substitute for your own critical thinking.
For sure, they won’t go away. No matter what anyone says, you can be sure the campaigns themselves are polling. And they are using those poll results to guide decisions and actions.
Why shouldn’t the public be in on the action?








