|
|
Applied Scaling & Classification Techniques in Political Science using text data
Syllabus
Course aims and objectives
Students will learn how to employ some widely discussed methods advanced in the literature to analyze political texts and to extract from them useful information for texting their own theories.
First Lecture
24/11/21 Theory: An introduction to text analytics
Reference texts: (1; 2)
24/11/21 Lab class: An introduction to the Quanteda package (a) packages to install; b) Lab 1 slides; scripts: Lab 1 script; datasets: a) Boston tweets sample; b) Inaugural US Presidential speeches sample (to open this file, please use the data compression tool WinRAR); c) sample of Japanese legislatives speeches
First assignment (due: 30 November 2021)
Second Lecture
1/12/21 Theory: From words to positions: unsupervised scaling models
Reference texts (1; 2; 3)
2/12/21 Lab class: How to implement the Wordfish algorithm (scripts: a) packages to install; b) Lab 2 script (part I: Wordifsh); c) Lab 2 Script (part II: playing with Twitter); d) Lab 2 script (EXTRA: estimating bootstrap confidence intervals in Wordfish); e) Lab 2 script (EXTRA: estimating Wordshoal); dataset: a) UK party programs 1992 and 1997 (to open this file, please use the data compression tool WinRAR)
Second assignment (due: 7 December 2021) (dataset for Assignment 2. To open this file, please use the data compression tool WinRAR)
Third Lecture
8/12/21, 10:00-12:00 Theory: From words to positions: Supervised scaling models ; script: how Wordscores works
Reference texts (1; 2; 3, 4)
9/12/21 Lab class: How to implement the Wordscores algorithm (a) Lab 3 slides; b) Lab 3 script (part I: Wordscores); c) Lab 3 script (part II: geography and Twitter); d) dataset for the second part of the Lab)
Third assignment (due: 14 December 2021)
Fourth Lecture
15/12/21 Theory: From words to issues: unsupervised classification models
Reference text (1):
16/12/21 Lab class: How to implement a Topic Model (scripts: a) packages to install; b) Lab 4 script (topic model); c) Lab 4 script (EXTRA: estimating a cluster analysis); datasets: Guardian 2016; Guardian 2013).
Fourth assignment (due: 21 December 2021)
Fifth Lecture
1/5/22 Theory: (Part 1): From words to issues: structural topic models; (Part 2): Dictionary models
Reference texts (1, 2):
1/6/22 Lab class: How to implement a Structural Topic Model and dictionary models (scripts: a) packages to install; b) Lab 5 script: STM; c) Lab 5 dictionaries; d) Lab 5 dictionaries and Twitter; datasets: a) NyT; b) data for topical content analysis)
Fifth Assignment (due: 11 January 2022)
Sixth Lecture
1/12/22 Theory: From words to issues: semi-supervised classification models
Reference texts (1, 2):
1/13/22 Lab class: How to implement a semi-supervised classification model (scripts: a) packages to install; b) Lab 6 script: Newsmap; c) Lab 6 script: KeyATM)
Sixth Assignment (due: 18 January 2022)
Seventh Lecture
1/19/22 Theory: From words to issues: supervised classification models (part 1; part 2)
Reference text (1):
1/20/22 Lab class: How to implement supervised classification models (scripts: a) packages to install; b) Lab 7 script; datasets: a) disasters training-set; b) disasters test-set; c) Nationality)
Seventh Assignment (due: 25 January 2022) (datasets for Assignment 7: a) UK training set; b) UK test set) - SOLUTION with 3 class-labels
Eighth Lecture
1/26/22 Theory: How to validate the results you get from machine learning algorithms (part 1); The importance of the training-set (part 2)
Reference text (1, 2, 3):
1/27/22 Lab class: How to apply k-fold cross validation (scripts: a) package to install; b) Lab 8 script (part A); c) Lab 8 script (part B); d) Lab 8 script (part C); e) secondo training-set for the lab; f) inter-coder reliability script
Eigth Assignment (due: 1 February 2022) (datasets for Assignment 8: training-set movie reviews)
Students will learn how to employ some widely discussed methods advanced in the literature to analyze political texts and to extract from them useful information for texting their own theories.
First Lecture
24/11/21 Theory: An introduction to text analytics
Reference texts: (1; 2)
- Grimmer, Justin, and Stewart, Brandon M. 2013. Text as Data: The Promise and Pitfalls of Automatic Content Analysis Methods for Political Texts. Political Analysis, 21(3): 267-297
- Benoit, Kenneth (2020). Text as data: An overview. In Luigi Curini and Robert Franzese (eds.), SAGE Handbook of Research Methods is Political Science & International Relations, London, Sage, chapter 26
24/11/21 Lab class: An introduction to the Quanteda package (a) packages to install; b) Lab 1 slides; scripts: Lab 1 script; datasets: a) Boston tweets sample; b) Inaugural US Presidential speeches sample (to open this file, please use the data compression tool WinRAR); c) sample of Japanese legislatives speeches
First assignment (due: 30 November 2021)
Second Lecture
1/12/21 Theory: From words to positions: unsupervised scaling models
Reference texts (1; 2; 3)
- Proksch, Sven-Oliver, and Slapin, Jonathan B. 2008. A Scaling Model for Estimating Time-Series Party Positions from Texts. American Journal of Political Science, 52(3): 705-722.
- Proksch, Sven-Oliver, and Slapin, Jonathan B. 2009. How to Avoid Pitfalls in Statistical Analysis of Political Texts: The Case of Germany. German Politics, 18(3): 323-344.
- Curini, Luigi, Hino, Airo, and Atsushi Osaki. 2020. Intensity of government–opposition divide as measured through legislative speeches and what we can learn from it. Analyses of Japanese parliamentary debates, 1953–2013, Government and Opposition, 55(2), 184-201
2/12/21 Lab class: How to implement the Wordfish algorithm (scripts: a) packages to install; b) Lab 2 script (part I: Wordifsh); c) Lab 2 Script (part II: playing with Twitter); d) Lab 2 script (EXTRA: estimating bootstrap confidence intervals in Wordfish); e) Lab 2 script (EXTRA: estimating Wordshoal); dataset: a) UK party programs 1992 and 1997 (to open this file, please use the data compression tool WinRAR)
Second assignment (due: 7 December 2021) (dataset for Assignment 2. To open this file, please use the data compression tool WinRAR)
Third Lecture
8/12/21, 10:00-12:00 Theory: From words to positions: Supervised scaling models ; script: how Wordscores works
Reference texts (1; 2; 3, 4)
- Laver, Michael, Kenneth Benoit, John Garry. 2003. Extracting Policy Positions from political texts using words as data. American Political Science Review, 97(02), 311-331
- Egerod, Benjamin C.K., and Robert Klemmensen (2020). Scaling Political Positions from text. Assumptions, Methods and Pitfalls. In Luigi Curini and Robert Franzese (eds.), SAGE Handbook of Research Methods is Political Science & International Relations, London, Sage, chapter 27
- Martin, Lanny W., and Georg Vanberg. 2008. A robust transformation procedure for interpreting political text. Political Analysis, 16: 93-100
- Kruspe, Anna et al. (2021). Changes in Twitter geolocations: Insights and suggestions for future usage. arXiv:2108.12251v1
9/12/21 Lab class: How to implement the Wordscores algorithm (a) Lab 3 slides; b) Lab 3 script (part I: Wordscores); c) Lab 3 script (part II: geography and Twitter); d) dataset for the second part of the Lab)
Third assignment (due: 14 December 2021)
Fourth Lecture
15/12/21 Theory: From words to issues: unsupervised classification models
Reference text (1):
- Robert, Margaret E., Brandon M. Stewart, Dustin Tingley, Christopher Luca, Jetson Leder-Luis, Shana Kushner Gadarian, Bethany Albertson, David G. Rand. 2014. Structural Topic Models for Open-Ended Survey Response. American Journal of Political Science, 58(4), 1064-1082
16/12/21 Lab class: How to implement a Topic Model (scripts: a) packages to install; b) Lab 4 script (topic model); c) Lab 4 script (EXTRA: estimating a cluster analysis); datasets: Guardian 2016; Guardian 2013).
Fourth assignment (due: 21 December 2021)
Fifth Lecture
1/5/22 Theory: (Part 1): From words to issues: structural topic models; (Part 2): Dictionary models
Reference texts (1, 2):
- Roberts, Margaret E., Brandon M. Stewart, Dustin Tingley. 2014. STM: R Package for Structural Topic Models. Journal of Statistical Software, https://cran.r-project.org/web/packages/stm/vignettes/stmVignette.pdf
- Grimmer, Justin, and Stewart, Brandon M. 2013. Text as Data: The Promise and Pitfalls of Automatic Content Analysis Methods for Political Texts. Political Analysis, 21(3): 267-297
1/6/22 Lab class: How to implement a Structural Topic Model and dictionary models (scripts: a) packages to install; b) Lab 5 script: STM; c) Lab 5 dictionaries; d) Lab 5 dictionaries and Twitter; datasets: a) NyT; b) data for topical content analysis)
Fifth Assignment (due: 11 January 2022)
Sixth Lecture
1/12/22 Theory: From words to issues: semi-supervised classification models
Reference texts (1, 2):
- Kohei Watanabe and Yuan Zhou (2020) Theory-Driven Analysis of Large Corpora: Semisupervised Topic Classification of the UN Speeches. Social Science Computer Review, DOI: 10.1177/0894439320907027
- Shusei Eshima, Kosuke Imai, and Tomoya Sasaki (2020). Keyword Assisted Topic Models, arXiv:2004.05964v1
1/13/22 Lab class: How to implement a semi-supervised classification model (scripts: a) packages to install; b) Lab 6 script: Newsmap; c) Lab 6 script: KeyATM)
Sixth Assignment (due: 18 January 2022)
Seventh Lecture
1/19/22 Theory: From words to issues: supervised classification models (part 1; part 2)
Reference text (1):
- Olivella, Santiago, and Shoub Kelsey (2020). Machine Learning in Political Science: Supervised Learning Models. In Luigi Curini and Robert Franzese (eds.), SAGE Handbook of Research Methods is Political Science & International Relations, London, Sage, chapter 56
1/20/22 Lab class: How to implement supervised classification models (scripts: a) packages to install; b) Lab 7 script; datasets: a) disasters training-set; b) disasters test-set; c) Nationality)
Seventh Assignment (due: 25 January 2022) (datasets for Assignment 7: a) UK training set; b) UK test set) - SOLUTION with 3 class-labels
Eighth Lecture
1/26/22 Theory: How to validate the results you get from machine learning algorithms (part 1); The importance of the training-set (part 2)
Reference text (1, 2, 3):
- Grimmer, Justin, and Stewart, Brandon M. 2013. Text as Data: The Promise and Pitfalls of Automatic Content Analysis Methods for Political Texts. Political Analysis, 21(3): 267-297
- Cranmer, Skyler J. and Desmarais, Bruce A. (2017) What Can We Learn from Predictive Modeling?, Political Analysis, 25: 145-166
- Curini, Luigi, and Robert Fahey. 2020. Sentiment Analysis. In: Luigi Curini and Robert Franzese (eds.), Sage Handbook of Research Methods in Political Science and International Relations, London: Sage, chapter 29
1/27/22 Lab class: How to apply k-fold cross validation (scripts: a) package to install; b) Lab 8 script (part A); c) Lab 8 script (part B); d) Lab 8 script (part C); e) secondo training-set for the lab; f) inter-coder reliability script
Eigth Assignment (due: 1 February 2022) (datasets for Assignment 8: training-set movie reviews)