Lee,
One of the first decisions we made after invalidating the Web-only poll was that its replacement would be a statistically valid representation of our readers. That's why we hired an an outside research company and those are the instructions we gave them. They determined the number of completed interviews required.
They worked from our database of print subscribers because Electric-only readers don't supply as much demographic information about themselves for us to be able to determine which of them are IS professionals with the certainty that we can with the print list.
I guaranteed that the new poll would reflect the opinions of InfoWorld readers, not InfoWorld Electric-only readers. Sixty percent of the poll's respondents are Electric readers as well so we clearly didn't exclude Electric readers.
By next year's awards issue, I hope we will be able to conduct a statistically valid poll that includes Web voting and doesn't rely solely on phone surveys.
Sandy Reed
For questions or comments, email the Forums Editor