Lectures – Overview

Week 1: What Is Computational Social Science (CSS)?

This lecture answers the following questions:

  • What does the term CSS mean?
  • Which principles underly a new CSS?
  • Which methods does CSS encompass?

Week 2: New Possibilities

This lecture answers the following questions:

  • What are digital trace data?
  • What are their advantages for our research?
  • Which examples of smart research exist?

Week 3: Data Acquisition I

This lecture answers the following questions:

  • How do we acquire digital trace data via web scraping?
  • What are considerations in terms of law and ethics?

Week 4: Data Acquisition II

This lecture answers the following questions:

  • How do we digitize text?
  • How do we digitize speech?

Week 5: Text as Data I

This lecture answers the following questions:

  • What do we mean by “text as data”?
  • What do we mean by “bag of words”?
  • Which methods exist more generally?
  • … and what do they measure?

Week 6: Text as Data II

This lecture answers the following questions:

  • What is “Machine Learning”?
  • What are use-cases for unsupervised and supervised ML?

Week 7: Text as Data III

This lecture answers the following questions:

  • How can we go beyond the “bag of words”?
  • What does context tell us about words?
  • … and how is it relevant for social scientists?

Week 8: Text as Data VI

This lecture answers the following questions:

  • What’s the latest and greatest in Natural Language Processing (NLP)?
  • What do these new methods bring to the table?
  • Which requirements do they have?

Week 9: Geo Data I

This lecture answers the following questions:

  • Why shall we take spatial data into account?
  • What does spatial data tell us about humans?
  • What are the particularities of spatial data?
  • How to measure distance?
  • Moran’s I

Week 10: Geo Data II

This lecture answers the following questions:

  • How do we perform statistical inference on human behavior using spatial data?
  • Which ways exist to get around spatial autocorrelation?

Week 11: Simulation I

This lecture answers the following questions:

  • Why shall we simulate human behavior in an agent-based fashion?
  • What is its relationship with sociological theory?
  • How can we use it for mechanism-based explanations?

Week 12: Simulation II

This lecture answers the following questions:

  • How can we get to reasonable assumptions for ABMs?
  • Where do these data come from?
  • How can we compare the model outcomes to the real world?

Week 13: Paper Presentations

No slides for this lecture except for some housekeeping things. Students are supposed to present their progress.