Skip to content

Latest commit

 

History

History
102 lines (63 loc) · 21.8 KB

alternative_curriculum.md

File metadata and controls

102 lines (63 loc) · 21.8 KB
Raw

ALTERNATIVE CURRICULA

In addition to the main videos and course materials developed by SICSS Co-Founders Chris Bail and Matt Salganik, organizers of SICSS partner sites have produced a variety of alternative curricula that serve the needs of different audiences. For example, some of these materials present code in alternative languages such as Python, and others are directed at non-American audiences. Click on the links below to explore these materials in detail.

Istanbul, Turkey (Kadir Has University)

Matti Nelimarkka and Akin Unver, taught in English using R

Pre-SICSS: Tasks

  • Theory and practise rarely meet in the real world. Therefore instead of writing tutorials which are long, fancy and full with nice codes, we are going to suggest three introductory guides and two abstract case studies, and provide guidance every week till SICSS via zoom and Slack. The tasks cover data collection, cleaning and reporting. We are leaving the analysis to SICSS weeks. Tasks

Day 1: Introduction and Ethics

  • What is research ethics and why it matters? A global perspective Slides

Day 2: Automated Data Collection

  • Digital trace data: conceptualization, opportunities and problems Slides

Day 3: Text Analysis

Day 4: Machine Learning

License

All slides are released under a CC-BY license, and all code is released under an MIT license.

Bamberg, Germany (University of Bamberg)

Julian Hohner, Thomas Saalfeld, and Carsten Schwemmer, taught in English using R

Day 1: Introduction and Ethics

Day 2: Collecting Digital Trace Data

Day 3: Social Network Analysis

Day 4: Automated Text Analysis

  • Basics of quantitative text analysis Materials

License

All slides are released under a CC-BY license, and all code is released under an MIT license.

Chicago, Illinois (Northwestern University)

Kat Albrecht, Natalie Gallagher, and Tina Law, taught in English using R

  • Teaching Computational Social Science: Lessons and Strategies Video, Slides
  • Modeling and Experiments Slides

Oxford, England (Oxford University)

Ridhi Kashyap, Nicolo Cavalli, and Taylor Brown, taught in English using R

Day 2: Collecting Digital Trace Data

  • What is digital trace data? Pros and cons of digital trace data. Research designs involving digital data. Video, Slides
  • Tools and techniques for working with digital trace data. Challenges of ethics and access with digital traces. Video, Slides, Tutorial

Day 3: Computational Text Analysis

  • History of text analysis, and sources of data in the modern age Video, Slides
  • Three methods to large-scale text analysis--how they work and what they can('t) do. Slides

Day 5: Non-probability Surveys and Machine Learning

  • Probabilistic linkage; Mixing census and big surveys; Multilevel Regression and Post-stratification; ML and Bayesian approaches. Video, Material
  • General form of learning problems, conceptual differences between inference and prediction, supervised and unsupervised prediction, linear model selection and regularization, tree-based methods and SVMs, followed by two examples. Video, Materials
  • A Kaggle style prediction (classification) competition using healthcare data. Video

Day 6: Experiments

  • Basic concepts on experiments; Literature on experiments (scale and complexity; classifiers and treatment; bots and platforms); computational approaches to experiments. Video

License

All slides are released under a CC-BY license, and all code is released under an MIT license.

Research Triangle Park, NC (RTI International)

Antje Kirchner, Craig Hill, Alan Blatecky, Helen Jang, and Jacqueline Olich, taught in English using R

Day 5: Machine Learning and Synthetic Populations

  • A Brief Introduction to Machine Learning Slides
  • Synthetic Populations Workshop Slides

License

All slides are released under a CC-BY license, and all code is released under an MIT license.