/materials

Primary LanguageHTMLMIT LicenseMIT

GESIS Fall Seminar in Computational Social Science 2022

Introduction to Computational Social Science with Python

  • Date: September 12-16, 2022
  • Time: 09:00-12:00 and 13:00-16:00 (including one 15 min break per session)

Lecturers

Dr. Milena Tsvetkova

Milena Tsvetkova is Assistant Professor of Computational Social Science at the Department of Methodology at the London School of Economics and Political Science. She completed her PhD in Sociology at Cornell University and postdoctoral training at the Oxford Internet Institute. In her research, she uses large-scale web-based experiments, network analysis of online data, and agent-based modeling to investigate fundamental social phenomena such as cooperation, social contagion, segregation, and inequality.

Dr. Patrick Gildersleve

Patrick Gildersleve is an LSE Fellow in Computational Social Science in the Department of Methodology at the London School of Economics and Political Science. Patrick graduated with a Masters in Physics from the University of Oxford, before completing his PhD at the Oxford Internet Institute in 2021. In his PhD research, he worked on studying the intersection of news media and Wikipedia. Patrick analysed how current events are recorded and accessed on the online collaborative encyclopaedia as well as its implications for theories of news values, newsworthiness, and collective attention dynamics. He has continued this work with an expanded research agenda around popularity and collective memory across platforms online.

Course Description

The course provides an introduction to the basic computational tools, skills, and methods used in Computational Social Science using Python. Python is the most popular programming language for data science, used widely in both academia and the industry. Students will learn to use common workflow and collaboration tools, design, write, and debug simple computer programs, and manage, summarize, and visualize data with common Python libraries. The course will employ interactive tutorials and hands-on exercises using real social data. Participants will work independently and in groups with guidance and support from the lecturers. The practical exercises are designed to demand more autonomy and initiative as the course progresses over the five days, culminating in an open-ended group project in the last afternoon session.

Course Prerequisites

This is an introductory course and no prior experience with programming is required. Basic understanding of statistics and some scripting experience (e.g., from building web pages or statistical analysis programs such as Stata) will be helpful but not needed.

Target Group

Participants will find the course useful if they:

  • Have no or limited technical and computational background
  • Have background in one of the social sciences (sociology, political science, psychology, etc.)
  • Would like to pursue research or professional career in computational social science or social data science (e.g., in academia, think tanks, government, NGOs, social media companies, tech startups)

Course and Learning Objectives

By the end of the course participants will:

  • Possess an understanding of the tools, methods, tasks, and goals of Computational Social Science
  • Design procedures and algorithms to solve data analysis tasks
  • Write simple programs in Python
  • Work confidently with pandas, matplotlib, seaborn, and other popular Python modules and libraries for data science
  • Use bash, Jupyter Notebook, and GitHub to write, run, collaborate on, and share programming code

Organisational Structure of the Course

The course will consist of two three-hour-long sessions. The morning session will use interactive instruction to introduce participants to the topic, demonstrate the new methods, and facilitate discussion. The afternoon session will make use of guided hands-on exercises with real-world data to practice the new material. Participants will work individually, in pairs, and in groups and the lecturers will be available throughout both sessions for consultation and support.

Software and Hardware Requirements

Participants require a laptop computer with Anaconda and git installed.

Recommended Literature to Look at in Advance

Day-to-Day Schedule and Literature

Day 1: Computation for Social Science

  • What is CSS?
  • Data, methods, and questions
  • Accountability, reproducibility, and ethics
  • Setting up your workflow
    • Installing Python with Anaconda
    • Introduction to Jupyter Notebooks
    • Introduction to Bash and GitHub
  • Introduction to programming with Python
    • Scalar data types, operators, and expressions
    • Variable assignment, printing, and comments
    • Non-scalar data types, indexing, and slicing
    • List and string methods

Recommended Literature:

Day 2: Writing Legible, Modular, and Optimized Code

Recommended Literature:

Day 3: Obtaining Data

  • Handling social data
    • Ethics of data access
    • Reading and writing common file types
    • More complex data types: time and dates, Unicode, etc.
  • Scraping web data
    • Inspecting webpages
    • Parsing static HTML with BeautifulSoup
    • Scraping dynamic pages with Selenium
  • JSON and working with APIs
    • The Anatomy of APIs
    • Authentication
    • Pagination

Recommended Literature:

Day 4: Analysing Rectangular Data

  • Introduction to pandas
    • Creating DataFrames
    • Accessing and filtering data
    • Computing summary statistics
    • Reading and writing data
  • Manipulating pandas DataFrames
    • Handling different data types
    • Combining data from different tables
    • Applying functions to DataFrames
    • Creating basic plots using pandas
  • Machine learning with scikit-learn
    • Machine learning (a very brief intro)
    • Scikit-learn
    • Training data vs test data
    • Random forests
    • Feature importance
    • Hyper-parameter tuning

Recommended Literature:

Day 5: Visualising Data and Analysing Non-Rectangular Data

Recommended Literature:

Additional Recommended Literature

  • Guttag, John V. (2016). Introduction to Computation and Programming Using Python: With Application to Understanding Data. MIT Press.
  • McLevey, John. (2021). Doing Computational Social Science: A Practical Introduction. Sage.