What a great turnout for the primary yesterday. I think Arriana Huffington said it best with, “A Great Night for Hillary, a Crappy Night for Pollsters“. And if that’s not enough info on polling, here’s another link from Zogby.

While we didn’t pick Hillary as a winner against Obama, our numbers were much closer than what pollsters predicted. We accurately picked the order of republican candidates with McCain coming in first, Romney second, and Huckabee third. On the democratic side, we had predicted a very tight race between Clinton and Obama with Obama winning and Edwards coming in a distant third. The race was indeed tight between Clinton and Obama but the blog data still showed Obama with the advantage. I will try to dig into the data a bit more and post on the collective wisdom of the blogging crowd but at first glance it appears that showing emotion was a positive for Mrs. Clinton.

The proof points are definitely real and in this case closer than the polling results indicated. Social media measurement and analytics rule the day.

Stay tuned for Super Tuesday!

Tags: Analyzing Consumer Generated Media, Blog Analysis, Mass Media/Culture by Kevin Yordy
No Comments »

In the days leading up to the Iowa caucus, Collective Intellect set up tracking to see how accurately our technology could transfer blog sentiment and activity monitoring into a tangible projection ahead of the caucus. We set up search agents for each of the major candidates which ensured that their name was mentioned near “Iowa”. We combined blog activity with our sentiment algorithm, and we were able to use the resulting “score” to accurately predict the actual caucus results for the major candidates:

Then we went back and looked to see if the results would have changed if we would have only used traditional news sources, and left out blogs. We found that the GOP projection would have been the same ranking, but if we had used only traditional news activity for the Democratic side, it would have projected Clinton as the winner:

We have repeated this technique for New Hampshire, creating new search agents where the candidate must be mentioned near “New Hampshire” instead of “Iowa.” The following is our projection for the New Hampshire primary:

There are no major surprises in this case, and it is more in tune with the traditional news sources this time, except that we see John Edwards coming in a more distant third in New Hampshire. However, future projections for Super Tuesday and beyond are likely to reveal trends and surprises that a simple reading of traditional news media might miss.

Note: Ron Paul has been left out of these projections simply because in the past, the massive amounts of posts, re-posts and links from his online supporters have skewed our data far more than any other candidate’s, and it has been demonstrated recently that this online support is not transferring much to the polls.

Tags: Analyzing Consumer Generated Media, Mass Media/Culture by Kevin Yordy
1 Comment »