IPH 431: Statistics for Humanities Scholars (3 units)
A survey of statistical ideas and principles. The course will expose students to tools and techniques useful for quantitative research in the humanities, many of which will be addressed more extensively in other courses: tools for text-processing and information extraction, natural language processing techniques, clustering & classification, and graphics. The course will consider how to use qualitative data and media as input for modeling and will address the use of statistics and data visualization in academic and public discourse. By the end of the course students should be able to evaluate statistical arguments and visualizations in the humanities with appropriate appreciation and skepticism.
Details. Core topics include sampling, experimentation, chance phenomena, distributions, exploration of data, measures of central tendency and variability, and methods of statistical testing and inference. In the early weeks, students will develop some facility in the use of Excel; thereafter, students will learn how to use Python or R for statistical analyses.
IPH 430: Data Manipulation for the Humanities (1 unit)
The course will present basic data modeling concepts and will focus on their application to data clean-up and organization (text markup, Excel, and SQL). Aiming to give humanities students the tools they will need to assemble and manage large data sets relevant to their research, the course will teach fundamental skills in programming relevant to data management (using Python); it will also teach database design and querying (SQL).
Details. The course will cover a number of “basics”: the difference between word processing files, plain text files, and structured XML; best practices for version control and software “hygiene”; methods for cleaning up data; regular expressions (and similar tools built into most word processors). It will proceed to data modeling: lists (Excel, Python); identifiers/keys and values (Excel, Python, SQL); tables/relations (SQL and/or data frames); joins (problem in Excel, solution in SQL, or data frames); hierarchies (problem in SQL/databases, solution in XML); and network graph structures (nodes and edges in CSV). It will entail basic scripting in Python, concentrating on using scripts to get data from the web, and the mastery of string handling.
IPH 432:Programming for Text Analysis (3 units)
This course will cover the core data-scientific concepts required for analyzing large corpora of texts and will introduce basic programming together with text-analysis techniques relevant to the humanities. (There will be very slight overlap with the programming instruction in the statistics and data-management courses.)
Details. Students will learn to calculate basic corpus-statistics, and will develop facility with such techniques as tokenization, chunking, extraction of thematically significant words, stylometrics and authorship attribution. Later in the course, more advanced topics from natural language processing such as stemming, lemmatization, named-entity recognition, part-of-speech tagging will be introduced along with a survey of text-classification terminology.