#http://www.tcd.ie/Political_Science/wordscores/software.html #[In STATA use control V to paste into the command line; control R to recall a previous command] #Open STATA. On the command line..... #Install wordscores net install http://www.tcd.ie/Political_Science/wordscores/wordscores #Change working directory to folder that contains texts cd "C:\Users\John\Desktop\wordscoresdata" #Convert texts to term frequency matrices wordfreq reagan1981.txt reagan1988.txt clinton1993.txt clinton2000.txt bush2001.txt bush2008.txt obama2009.txt obama2015.txt #Set reference document locations (can have multiple ref doc locations, but need at least two.) setref treagan1981 20.0 tobama2015 0 #Define dimension (this is completely arbitrary) wordscore conservatism #Calculate conservatism score of speeches textscore conservatism treagan1988 tclinton1993 tclinton2000 tbush2001 tbush2008 tobama2009 Results! 'Transformed Score is the estimated location of the speech on the conservation dimension where a higher value is more conservative' Textscore v23-Jun-2009 (c) 2009 Kenneth Benoit Dimension: CONSERVATISM Transformed Score Method: Laver-Benoit-Garry | Unique Trans- Trans- Total % Virgin | Raw Raw Scored formed formed Transformed Words Tot Text | Score SE Words Score SE [95% Conf. Interval] Scored Sc'd -----------------------+------------------------------------------------------------------------------------- treagan1988 | 7.5286 0.0753 744 5.7475 1.8792 1.9890 9.5060 3,953 82.0 tclinton1993 | 8.4344 0.0587 889 28.3691 1.4660 25.4371 31.3011 6,215 88.9 tclinton2000 | 7.4928 0.0527 947 4.8537 1.3169 2.2199 7.4875 7,593 84.0 tbush2001 | 8.0698 0.0779 742 19.2637 1.9445 15.3747 23.1528 3,742 86.2 tbush2008 | 7.2127 0.0684 817 -2.1409 1.7084 -5.5577 1.2759 4,642 81.5 tobama2009 | 6.8789 0.0585 1,522 -10.4760 1.4614 -13.3989 -7.5532 6,006 100.0