This is the forth of five posts on History 100S: Text Analysis for Digital Humanists ans Social Scientists, a Spring 2017 course taught by Laura Nelson that exposed UC Berkeley students to cutting edge computational text analysis techniques. In this post we focus on a collaborative project by Arvind Iyengar, Jiayi Huang, Parth Singhal entitled An Exploratory Topical Analysis of Obama's Speeches Utilizing Text Analysis Techniques and Data Analysis and Visualizations.
The project asks the following two research questions:
- Was the political interest of the Obama administration, expressed in Obama's speeches, consistent with the actions of the legislative branch during his two consecutive terms in regards to the allocation of the national budget as decided by Congress?
- If any inconsistency does exist between the agenda of the executive adminstration and the legislative branch, are the inconsistencies associated with a difference in the political affiliations of the two branches of government?
Using topic modelling and bag of words methods, the authors address these questions through a text analysis of President Barack Obama's speeches made between 2009 - 2016. They then use statistical methods to compare the results with annual national budget data for the same years. The researchers find that "as the Republicans take control of Congress, the amount of discrepancy between Obama's speeches and the Congressional budget expenditure increases.
To learn more, read the project notebook here.
Posts in this series:
Text Analysis for Digital Humanists and Social Scientists, Part 1: Introduction
Text Analysis for Digital Humanists and Social Scientists, Part 3: The Evolution of Modern Hip Hop
Text Analysis for Digital Humanists and Social Scientists, Part 5: Text Encoding and Decoding
