Earlier this morning, I called Professor Darrell West, Director of the Taubman Center for Public Policy at Brown University, and asked why Rhode Island Republican Primary Senate results were not included in the February 2006 State Survey. Professor West answered that it is too hard to predict where independents will go.
If you look back at 2002, there were about 240,000 ballots cast for Governor in the general election, while only 25,000 people voted in the contested Republican Gubernatorial primary. That means the poll sample of 785 probably contains around 80-90 likely Republican primary voters.
To have any chance at all of giving an accurate snapshot of the race, the Brown University pollsters would have to figure out some way to identify independents likely to vote in a Republcan primary (when many independents themselves probabaly haven’t decided which primary they’re going to vote in!) and then conduct a second poll to get a big enough sample of Republicans + Republican-voting independents.