Applied Scaling & Classification Techniques in Political Science (2019/20)

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 
Morning class: An introduction to textual analysis methods​ (first part; second part)
Reference text 1; Reference text 2
Lab Class: How to analyze texts with the Quanteda package. Read this file before our Lab 1!!! (scripts: a) 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) List President USA by party; d) sample of Japanese legislatives speeches (to open this file, please use the data compression tool WinRAR))
Assignment 1

Second Lecture
Morning class: Scaling texts (Unsupervised scaling algorithms: Wordfish) 
Reference text 1; Reference text 2; Reference text 3; Reference text 4
Lab Class: How to implement the Wordfish algorithm using the Quanteda package (packages to install in R before lab; Lab 2 script; Slide about Lab 2;  datasets: a) sample of Japanese legislatives speeches; b) UK party programs 1992 and 1997: to open this file, please use the data compressione tool WinRAR)
Assignment 2 (datasets for Assignment 1: a) the Irish party manifestoes; b) Japanese legislatives speeches. To open these files, please use the data compressione tool WinRAR)
​
Third Lecture
Morning class: Scaling texts (Supervised scaling algorithms: Wordscores) 
Reference text 1; Reference text 2; Reference text 3
Lab Class: How to implement the Wordscores algorithm using the Quanteda package (Lab 3 script)
Assignment 3

Fourth Lecture
Morning class: Scaling texts (some extensions); CMP; Social Media
References text 1; References text 2; References text 3; Reference text 4; Reference text 5; Reference text 6
Lab Class: packages to install in R before lab; a) Script about CA; b) Script about CMP; c) Script about Twitter
Assignment 4

Fifth Lecture
Morning class: Unsupervied classification methods (Structural Topic Models)
Reference text 1; References text 2
Lab Class: packages to install in R before lab; Slide about Lab 5; Scripts: a) clustering script; b) stm script
Assignment 5
Assignment 5 solution

Sixth Lecture
Morning class: Supervised classification methods (Dictionaries and Machine Learning Algorithms)
Reference text
Lab Class: packages to install in R before lab; Scripts: a) dictionaries (part 1); b) dictionaries (part 2)
Assignment 6
Assignment 6 solution (second part)

Seventh Lecture
Morning class:  Supervised classification methods (Machine Learning and Proportional Classification Algorithms)
Reference text 1; Reference text 2; Reference text 3
Lab Class: packages to install in R before lab; Script; dataset 1; dataset 2
Assignment 7 (dataset for the assignment: training-set; test-set)
​
Eight Lecture
​Morning class: Supervised classification methods (Cross-Validation)
Reference text 1; Reference text 2
​Lab Class: script 1; script 2; script 3; scritp 4
Assignment 8