Presidential and Vice Presidential Debate Stats and Script This page contains a number of resources useful for performing basic textual analyses of the presidential debates. A self-contained Python script parses each debate, and outputs a number of files listing: summary statistics of the speakers' responses (e.g., average response length in words), the most frequently used words and phrases, a compressed timeline showing when phrases were repeated, a list of the longest words used by each candidate, and each speaker's responses independent of the other speakers' speech. Statistics for all televised debates are available. Sample findings A word of caution with these statistics: These statistics look for exact word matches. This means they ignore semantically similar content that may have been expressed using different words. Word stemming is being used to help normalize the stats (ie, "lasts, lasted, and lasting" all reduce to "last"), but stemming is not foolproof. In the cases where stemming is used, all variations encountered are listed. First Presidential Debate Here are a few of the findings from the first debate, in no particular order: Kerry's average response length was 215 words. Bush's was 155 Bush said the phrase "hard work" 11 times Bush said "Saddam Hussein" 9 more times than he said "Bin Laden" (14 vs. 5 occurrences) Bush said "free Iraq" 12 times. Kerry did not say that exact phrase at all Bush said "mixed messages" 5 times The longest words Bush uttered were "transformational" and "counterterrorism". Kerry's were "extraordinarily" and "authoritatively". Vice Presidential Debate Edwards' average response length was 181 words, Cheney's was 165 The top two-word phrases used by Edwards: vice president (48 times, #1 two-word phrase) John Kerry (33 times) American people (26 times) Health care (23) For Cheney, the top two-word phrases (filtering out some common phrases, such as "we've got"): I think (23 times, the #1 two-word phrase) United States (11 times) Saddam Hussein (9 times) John Kerry (9 times) "Bin Laden" was said 2 times by Cheney, 8 times by Edwards Second Debate In this debate, Bush did not use the phrase "mixed messages" at all Bush upped his response length to 181 words, compared to his 155-word average response length in the first debate The second most frequently used 2-word phrase by Kerry in the second debate was "health care", which he said 19 times. (The number one two-word phrase used by Kerry was "I'm going") Third Debate Stats for the final debate are posted, but I have not analyzed them. Downloads You can download the Python script to generate the stats on your own, or you can download the files it generated directly. The script is completely self-contained, and even includes the debate text. debate_stats.py (The script) debate_stats.zip (All the generated stats, in one zip file; includes the script) The files outputted are tab-delimited data files that can be read into other software, like Matlab or Excel. Currently, the comment format is conducive to Matlab, though you can erase the "% " lines to more easily import them into something like Excel. For repeated phrases, there are now HTML files that format the output into tables more conducive to browsing. Text files remain for importing the data into third-party software. To give you an example of the files' contents, here are the most common phrases, composed of two words, uttered by Kerry and Bush in the first debate: 09-30-04_Bush_repeated_phrases_2.html 09-30-04_Kerry_repeated_phrases_2.html Here are the individual files: 09-30-04_summary_statistics.txt 10-05-04_summary_statistics.txt 10-08-04_summary_statistics.txt 10-13-04_summary_statistics.txt 09-30-04_Bush_longest_words.txt 09-30-04_Bush_repeated_phrases_1.html 09-30-04_Bush_repeated_phrases_1.txt 09-30-04_Bush_repeated_phrases_2.html 09-30-04_Bush_repeated_phrases_2.txt 09-30-04_Bush_repeated_phrases_3.html 09-30-04_Bush_repeated_phrases_3.txt 09-30-04_Bush_repeated_phrases_4.html 09-30-04_Bush_repeated_phrases_4.txt 09-30-04_Bush_repeated_phrases_5.html 09-30-04_Bush_repeated_phrases_5.txt 09-30-04_Bush_speaker_only_transcript.txt 09-30-04_Bush_word_lengths.txt 09-30-04_Bush_word_stats.txt 09-30-04_Kerry_longest_words.txt 09-30-04_Kerry_repeated_phrases_1.html 09-30-04_Kerry_repeated_phrases_1.txt 09-30-04_Kerry_repeated_phrases_2.html 09-30-04_Kerry_repeated_phrases_2.txt 09-30-04_Kerry_repeated_phrases_3.html 09-30-04_Kerry_repeated_phrases_3.txt 09-30-04_Kerry_repeated_phrases_4.html 09-30-04_Kerry_repeated_phrases_4.txt 09-30-04_Kerry_repeated_phrases_5.html 09-30-04_Kerry_repeated_phrases_5.txt 09-30-04_Kerry_speaker_only_transcript.txt 09-30-04_Kerry_word_lengths.txt 09-30-04_Kerry_word_stats.txt 09-30-04_Lehrer_longest_words.txt 09-30-04_Lehrer_repeated_phrases_1.html 09-30-04_Lehrer_repeated_phrases_1.txt 09-30-04_Lehrer_repeated_phrases_2.html 09-30-04_Lehrer_repeated_phrases_2.txt 09-30-04_Lehrer_repeated_phrases_3.html 09-30-04_Lehrer_repeated_phrases_3.txt 09-30-04_Lehrer_repeated_phrases_4.html 09-30-04_Lehrer_repeated_phrases_4.txt 09-30-04_Lehrer_repeated_phrases_5.html 09-30-04_Lehrer_repeated_phrases_5.txt 09-30-04_Lehrer_speaker_only_transcript.txt 09-30-04_Lehrer_word_lengths.txt 09-30-04_Lehrer_word_stats.txt 10-05-04_Cheney_longest_words.txt 10-05-04_Cheney_repeated_phrases_1.html 10-05-04_Cheney_repeated_phrases_1.txt 10-05-04_Cheney_repeated_phrases_2.html 10-05-04_Cheney_repeated_phrases_2.txt 10-05-04_Cheney_repeated_phrases_3.html 10-05-04_Cheney_repeated_phrases_3.txt 10-05-04_Cheney_repeated_phrases_4.html 10-05-04_Cheney_repeated_phrases_4.txt 10-05-04_Cheney_repeated_phrases_5.html 10-05-04_Cheney_repeated_phrases_5.txt 10-05-04_Cheney_speaker_only_transcript.txt 10-05-04_Cheney_word_lengths.txt 10-05-04_Cheney_word_stats.txt 10-05-04_Edwards_longest_words.txt 10-05-04_Edwards_repeated_phrases_1.html 10-05-04_Edwards_repeated_phrases_1.txt 10-05-04_Edwards_repeated_phrases_2.html 10-05-04_Edwards_repeated_phrases_2.txt 10-05-04_Edwards_repeated_phrases_3.html 10-05-04_Edwards_repeated_phrases_3.txt 10-05-04_Edwards_repeated_phrases_4.html 10-05-04_Edwards_repeated_phrases_4.txt 10-05-04_Edwards_repeated_phrases_5.html 10-05-04_Edwards_repeated_phrases_5.txt 10-05-04_Edwards_speaker_only_transcript.txt 10-05-04_Edwards_word_lengths.txt 10-05-04_Edwards_word_stats.txt 10-05-04_Ifill_longest_words.txt 10-05-04_Ifill_repeated_phrases_1.html 10-05-04_Ifill_repeated_phrases_1.txt 10-05-04_Ifill_repeated_phrases_2.html 10-05-04_Ifill_repeated_phrases_2.txt 10-05-04_Ifill_repeated_phrases_3.html 10-05-04_Ifill_repeated_phrases_3.txt 10-05-04_Ifill_repeated_phrases_4.html 10-05-04_Ifill_repeated_phrases_4.txt 10-05-04_Ifill_repeated_phrases_5.html 10-05-04_Ifill_repeated_phrases_5.txt 10-05-04_Ifill_speaker_only_transcript.txt 10-05-04_Ifill_word_lengths.txt 10-05-04_Ifill_word_stats.txt 10-08-04_Bush_longest_words.txt 10-08-04_Bush_repeated_phrases_1.html 10-08-04_Bush_repeated_phrases_1.txt 10-08-04_Bush_repeated_phrases_2.html 10-08-04_Bush_repeated_phrases_2.txt 10-08-04_Bush_repeated_phrases_3.html 10-08-04_Bush_repeated_phrases_3.txt 10-08-04_Bush_repeated_phrases_4.html 10-08-04_Bush_repeated_phrases_4.txt 10-08-04_Bush_repeated_phrases_5.html 10-08-04_Bush_repeated_phrases_5.txt 10-08-04_Bush_speaker_only_transcript.txt 10-08-04_Bush_word_lengths.txt 10-08-04_Bush_word_stats.txt 10-08-04_Gibson_longest_words.txt 10-08-04_Gibson_repeated_phrases_1.html 10-08-04_Gibson_repeated_phrases_1.txt 10-08-04_Gibson_repeated_phrases_2.html 10-08-04_Gibson_repeated_phrases_2.txt 10-08-04_Gibson_repeated_phrases_3.html 10-08-04_Gibson_repeated_phrases_3.txt 10-08-04_Gibson_repeated_phrases_4.html 10-08-04_Gibson_repeated_phrases_4.txt 10-08-04_Gibson_repeated_phrases_5.html 10-08-04_Gibson_repeated_phrases_5.txt 10-08-04_Gibson_speaker_only_transcript.txt 10-08-04_Gibson_word_lengths.txt 10-08-04_Gibson_word_stats.txt 10-08-04_Kerry_longest_words.txt 10-08-04_Kerry_repeated_phrases_1.html 10-08-04_Kerry_repeated_phrases_1.txt 10-08-04_Kerry_repeated_phrases_2.html 10-08-04_Kerry_repeated_phrases_2.txt 10-08-04_Kerry_repeated_phrases_3.html 10-08-04_Kerry_repeated_phrases_3.txt 10-08-04_Kerry_repeated_phrases_4.html 10-08-04_Kerry_repeated_phrases_4.txt 10-08-04_Kerry_repeated_phrases_5.html 10-08-04_Kerry_repeated_phrases_5.txt 10-08-04_Kerry_speaker_only_transcript.txt 10-08-04_Kerry_word_lengths.txt 10-08-04_Kerry_word_stats.txt 10-13-04_Bush_longest_words.txt 10-13-04_Bush_repeated_phrases_1.html 10-13-04_Bush_repeated_phrases_1.txt 10-13-04_Bush_repeated_phrases_2.html 10-13-04_Bush_repeated_phrases_2.txt 10-13-04_Bush_repeated_phrases_3.html 10-13-04_Bush_repeated_phrases_3.txt 10-13-04_Bush_repeated_phrases_4.html 10-13-04_Bush_repeated_phrases_4.txt 10-13-04_Bush_repeated_phrases_5.html 10-13-04_Bush_repeated_phrases_5.txt 10-13-04_Bush_speaker_only_transcript.txt 10-13-04_Bush_word_lengths.txt 10-13-04_Bush_word_stats.txt 10-13-04_Kerry_longest_words.txt 10-13-04_Kerry_repeated_phrases_1.html 10-13-04_Kerry_repeated_phrases_1.txt 10-13-04_Kerry_repeated_phrases_2.html 10-13-04_Kerry_repeated_phrases_2.txt 10-13-04_Kerry_repeated_phrases_3.html 10-13-04_Kerry_repeated_phrases_3.txt 10-13-04_Kerry_repeated_phrases_4.html 10-13-04_Kerry_repeated_phrases_4.txt 10-13-04_Kerry_repeated_phrases_5.html 10-13-04_Kerry_repeated_phrases_5.txt 10-13-04_Kerry_speaker_only_transcript.txt 10-13-04_Kerry_word_lengths.txt 10-13-04_Kerry_word_stats.txt 10-13-04_Schieffer_longest_words.txt 10-13-04_Schieffer_repeated_phrases_1.html 10-13-04_Schieffer_repeated_phrases_1.txt 10-13-04_Schieffer_repeated_phrases_2.html 10-13-04_Schieffer_repeated_phrases_2.txt 10-13-04_Schieffer_repeated_phrases_3.html 10-13-04_Schieffer_repeated_phrases_3.txt 10-13-04_Schieffer_repeated_phrases_4.html 10-13-04_Schieffer_repeated_phrases_4.txt 10-13-04_Schieffer_repeated_phrases_5.html 10-13-04_Schieffer_repeated_phrases_5.txt 10-13-04_Schieffer_speaker_only_transcript.txt 10-13-04_Schieffer_word_lengths.txt 10-13-04_Schieffer_word_stats.txt Have fun, Mike (mterry@cc.gatech.edu