Analyzing Text with Computers
This course will teach students how to analyze very large volumes of text, assisted by computers. This is computer-assisted in that it does not mean push-button results where an algorithm is meant to replace the researchers' analytical insights. Rather it is meant to augment researchers' analytical insights and skills through machine-reading of text collections too large for human reading and coding, e.g. large collections of interview transcripts, collections of doctrine and regulations, and collections of speeches/press releases. Example applications include comparing interviews with PTSD patients vs. interviews with people w/ PTSD symptoms but not a diagnosis as part of a health study, or identifying attitudes and opinions in a large collection of public vs. private documents from the Communist Party of China. Ultimately, students will be able to make empirically grounded claims about unstructured textual data on the scale of millions of words.