Counting blogs! One blog, two blogs, three, three blogs!

The Truth Laid Bear - a guy who definitly knows something about blog traffic - has a look at the stats behind a CNN article that says blogs are far below the MSM in election viewership. He points out that the data is from comScore, which tracks viewers, and sees that a difference in the definition of views vs. unique visitors is the reason for the discrepancy between comScore and the SiteMeter visits on those pages.

I wonder if there is some selection bias taking place as well. According to the methodology page linked, they randomly contact people to be tracked, and convince them to sign up to have their online viewing habits tracked by giving them virus protection, "faster internet", sweepstakes entries, and "a chance to improve the internet".

I don't know about you, but that's not enough for me to let anyone track my internet surfing habits. I have good virus protection, a nice fast DSL connection, a healthy dose of skepticism for sweepstakes, and lots of privacy concerns. I think this is probably shared by most bloggers and readers, and I doubt too many of them would take up comScore's offer in exchange for thier surfing privacy.

My guess is prizes like those offered would appeal to the less educated, less tech savy, and less affluent - who are exactly the people who don't read blogs, or even know what they are. They are the people who would be willing to sign up with comScore.

There are probably other factors as well - I'm guessing, for example, that comScore's tracking software is probably Windows based, shutting out Mac and Linux users - who are a small percentage of overall internet users, but are frequently higher educated, more technical, and more wealthy that average.