Very interesting article in Le Devoir this morning by Jean-Marc Leger, the guru of polling in Quebec. It's worth trotting out your high school French if you're not bilingual to read what he says about the reliability of internet polling.
Among other things, he says that the rates of response are much higher on internet polls, that a public that telephone surveys are not reaching (the growing percentage who don't have landlines) can be polled, the rate of undecideds is much lower because people can answer when they want and are the respondants are paid, not the pollsters.
This last is particularly important: Leger Marketing has 315,000 on its "panel" of people who have agreed to answer their polls. Not everyone is polled all the time, but because the people on the larger panel have provided certain demographic data, groups can be pulled from it to be questioned on different subjects. And the respondants answer because they get something out of it, not because they've been bugged to respond to a telephone call.
Leger ends his piece by saying that comparisons of results obtained by electronic polling has proved more accurate than telephone polling when it comes predicting election results.
Don't know quite what this means for the democratic process. Certainly there will be fewer telephone solicitor jobs available in the future. Can one volunteer for these polls, I wonder. A file to be watched as we move deeper into the cyber age.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment