Oct 3, 2008 | Parsing the Presidential Debates

I’ve written a simple Python script that enables you to parse the words from each of the Presidential and Vice Presidential debates.

Download the transcript for the Biden - Palin debate

Download the transcript for the Obama - McCain debate

Here’s the code. Some features:

• You can ignore short words or particular words, like determiners.
• The code automatically removes the all caps speaker identifier.
• It also removes the moderator’s sentence.

Some interesting points for the Biden - Palin debate:

people (46)
voted (42)
energy (32)
change (25)
maverick (12)

The Obama - McCain debate:
right (31)
spending (30)
people (27)
nuclear (27)

Update: On a related note, in Tuesday, October 14th’s Science Times, an article mentions James W. Pennebaker, a psychologist at U of T, who has done research in precisely the field of word analysis and social behavior as a result, and maintains a blog as well!