Thursday, April 19, 2007

Being honest with polls

I was very impressed with the two guest speakers at the Civitas luncheon, freshman Representatives Ric Killian (R-Mecklenburg) and Larry Hall (D-Durham). They were insightful, engaging and knowledgeable of the issues that were raised during the discussion. Representative Hall made one point that I thought deserved some more discussion. He said that we all needed to be honest about what the polls really mean and how we use them to justify our arguments.

He gave a good example from the Civitas poll. One question asked: “One-third of North Carolina students who start high school do not graduate. Would you support or oppose an increase in the number of charter schools that provide extensive career and technical training as one way to increase the number of students who stay in school and graduate?” 75% support, 16% opposed

That question could be used to justify arguments in favor of more charter schools, but would the results have turned out any differently if we switched the word charter for public? Are not people actually in favor of more extensive career and technical training? Larry Hall thinks so, and I do too.

Pollsters are often in precarious situations like this. While it is our job to measure public opinion, often times we are used by clients and others to help make justifications for their own arguments. To show that the public is behind your proposal can be a powerful lobbying and public relations tool.

PPP and every other pollster are caught up in this. And it’s important for people who are trying to interpret our poll results to be honest about what the results are really saying.

No comments:

Web Statistics