Here's one question that I think needs to be asked in the wake of the Massachusetts special election: did media outlets that reported at length on the Boston Globe poll showing Martha Coakley with a 15 point lead while ignoring the ones from Rasmussen and us showing a toss up really do their readers/viewers a service?
Maybe the Globe poll was correct when it was conducted but there's no question the one we put out the night of Saturday the 9th showing Brown up by a point and then the one Rasmussen put out Tuesday the 12th showing Coakley with just a two point lead gave a more accurate picture of the race. Yet some media outlets kept on talking exclusively about the Globe poll for five days, until Suffolk came out showing Brown in the lead.
For the most part this happened not because of liberal bias in the media but because some outlets are still sticking to 20th century policies against reporting automated polls, despite the fact that their predictive accuracy is proven one election cycle after another.
What was perhaps most amusing about this particular election is that several outlets, including the New York Times, actually talked about our polls and Rasmussen's in their pages generally but refused to name us or print the actual numbers. I will be interested to see if any of the outlets who gave their audiences an unrealistic picture of the race at this time last week by reporting on the Globe poll and ignoring the others out there will be transparent with their readers about why those decisions were made and either change their policies in the future or defend them in a way based on quantitative data and not just emotion or conventional outdated wisdom about what does and does not make an accurate poll.
But I'm not holding my breathe.