« Some Perspective on the Twitter Paradigm | Main | Politweets »
The comments to this entry are closed.
Sun | Mon | Tue | Wed | Thu | Fri | Sat |
---|---|---|---|---|---|---|
1 | 2 | 3 | 4 | 5 | ||
6 | 7 | 8 | 9 | 10 | 11 | 12 |
13 | 14 | 15 | 16 | 17 | 18 | 19 |
20 | 21 | 22 | 23 | 24 | 25 | 26 |
27 | 28 | 29 | 30 | 31 |
Doesn't this show that the BuzzMetrics show no predicative power when it comes to elections?
I wonder if it makes sense to divide buzz into two types, at least when it comes to blogs -- blogs that are first-hand, describing the thoughts and views of the person about their first-hand experiences. A mathematician writing about his/her research, a high schooler on their boyfriend, and so on -- and blogs that are more journalistic, which tend to write and talk about things in the news, or on second-hand experiences in general. The mathematician talking about popular math topics, for example.
Would "authentic" buzz then be better measured from the first group? (and hopefully have more predicative power).
Posted by: Deniz | January 08, 2008 at 03:22 AM
well this search is a bit misleading, because blog chatter around "clinton" includes Bill (possibly even chelsea, though more remote).
using their full names gives a clearer picture, and you can see that obama actually started to draw even with (and even pass) hillary at the beginning of december, well before iowa caucuses and the january spike (just ran a search).
Posted by: avin | January 11, 2008 at 02:12 PM