Analysis of the State of the Union Addresses by the American presidents determines which words are characteristic for Democratic and which for Republican presidents. It turns out that "yes" is a word characteristic for one of the parties, while "don't" is frequent in the speeches of the other party.
3. ABOUT ME
• Ph.D. in Linguistics
• Book: Linguistic Profiles: from Form to
Meaning via Statistics
• Blog Reading the Data Leaves
https://newyearaddresses.com, analyzing
the language of New Year Addresses by
Soviet and Russian Leaders and how it
correlates with historic events
5. DATA COLLECTION
AND PROCESSING
• State of the Union Addresses collected by Gerhard Peters
and John T. Woolley:
http://www.presidency.ucsb.edu/sou.php
• Beautiful Soup:
http://www.pythonforbeginners.com/python-on-the-
web/web-scraping-with-beautifulsoup/
• Pyhton SnowballStemmer:
https://pypi.python.org/pypi/snowballstemmer
• MySQL for data aggregation
• R for statistical analysis and visualization
6. 100 MOST FREQUENT
WORDS
1790 1791 1792 1793
act 810.7013 2196.8366 1446.4802 1539.2509
action 0.0000 0.0000 0.0000 0.0000
administration 810.7013 439.3673 0.0000 0.0000
America 0.0000 0.0000 0.0000 0.0000
…
ipm – items per million
Total words in the address of 1790: 2467
The word act is attested: 2 times
actipm = (2/2467)* 1000 000 = 810.7013
14. DEMOCRATIC WORDS VS.
REPUBLICAN WORDS
• Formula:
• Threshold: appears in at least 20 documents after 1933
http://www.livescience.com/34241-democratic-republican-
parties-switch-platforms.html
• 100 most republican and 100 most democratic words
Schmid, Hans-Jörg (2003), "Do women and men really live in different cultures?
Evidence from the BNC." In: Andrew Wilson, Paul Rayson and Tony McEnery, eds.,
Corpus linguistics by the Lune: a Festschrift for Geoffrey Leech, Peter Lang, Frankfurt,
185-221.
15. WORDS CHARACTERISTIC TO THE
DEMOCRATIC AND THE REPUBLICAN
PARTY
WHICH IS WHICH?
"DON'T" PARTY "YES" PARTY
20. SUMMARY
• Attestations of words in the State of the Union addresses
reflect historical changes in United States
• Two most important dimensions in the State of the Union
addresses are the historical era (Dimension 1) and the
economic situation (Dimension 2).
• Quantitative analysis can help find political preferences in
the text