| layout | title | header | ||
|---|---|---|---|---|
single |
Digital Controversy Mapping |
|
This tutorial material consists of 21 modules of 15 to 45 minutes, for 2 full days in total. It covers some of the technical / practical aspects of controversy mapping with digital methods. As such, it is designed to complement teaching on the controversy mapping course.
Goal: learn how to harvest and explore data, formulate insights, and build relatable visualizations.
Data: we will mostly use Wikipedia data to keep things relatively simple, but the techniques generalize to other media platforms and datasets.
Tools: we will mostly use Tableau, Gephi, and Jupyter Notebooks. No experience required.
Each tutorial covers a part of this general process, from data harvesting (bottom) to building a controversy atlas (top). Once through all the tutorials, the routes on this map provide a set of options for completing the practical steps of a controversy mapping project.
The schedule breaks down in 4 half-days. Each one focuses on a different topic, and requires 4 hours of work, split into 4 sessions of 45 minutes with 15 minutes breaks. That schedule is, however, quite intensive. You may want to slow down or skip some parts. We indicate which parts are necessary and which ones are optional.
The activities must be done in order, as they build on each other and ramp up in complexity. Each one comprises instructions and a set of practical tasks.
Build annotated timelines with Tableau Software, ~4 hours.
1.1. Intro to Tableau software
Tutorial for Tableau and Google Slides for annotation.
30 min. Necessary.
1.2. Visualize a different dataset with Tableau
Advanced tutorial for Tableau, with exercises and active learning.
30 min. Necessary.
1.3. Build a simple dashboard
Tutorial for dashboards in Tableau.
30 min. Optional, but someone in your group should know how to do dashboards.
1.4. A timeline of words
Exercise and active learning.
30 min. Optional but recommended. You should at least check that you are capable of making a meaningful annotated visualization from a data file with Tableau and Google Slides.
1.5. Harvest a dataset
Just a short introduction to data harvesting.
15 min. Necessary.
1.6. Harvest data with a notebook
Tutorial for Google Colab (Jupyter notebooks).
30 min. Necessary and very important.
1.7. Activate your knowledge about Tableau
Exercise: activate what you have learned.
30 min. Optional but recommended. You should at least check that you are capable of making a meaningful annotated visualization from scratch (including harvesting).
Build annotated network maps with Gephi, ~4 hours.
1.8. Intro to Gephi & Visualize clusters
Tutorial for Gephi.
45 min. Necessary.
1.9. Visualize a bipartite network
Tutorial for Gephi (bipartite networks).
30 min. Necessary, but you can skim through it if you're comfortable with the tool.
1.10. Visualize a weighted network
Exercise: activate what you have learned about Gephi.
15 min. Optional.
1.11. From data to network with Table2Net
Tutorial for Table2Net and a knowledge activation exercise.
45 min. You need to know how to use Table2Net but the exercise is ultimately optional.
1.12. Activate your knowledge about Gephi
Knowledge activation exercise.
45 min. Optional but recommended. You should at least check that you understand how and why to visualize networks.
Write visual protocols for relational data, ~4 hours.
2.1. Follow the protocol: scrape a network with SeeAlsology
Introduction to protocols and SeeAlsology.
15 min. Necessary.
2.2. Write the protocol: scrape from one article with SeeAlsology
Activation exercise with a case of your choosing.
30 min. Necessary. You must check that you are autonomous when it comes to harvesting, mapping, visualizing, annotating and documenting a case. This is the most simple version possible.
2.3. Follow the protocol: co-reference network from a category
Excercise where you will use 2 new notebooks.
15 min. Strongly recommended, as you will try new notebooks.
2.4. Write the protocol: Article-editor network from a category
Exercise where you will use 1 new notebook.
30 Min. Recommended, as you will try a new notebook.
2.5. Do your own network from a category
Exercise activating everything you have learned so far.
1h30 (45 min + 45 min). Optional but recommended. You should at least check that you are autonomous when it comes to harvesting, mapping, visualizing, annotating and documenting a case, in a real-world situation.
Write visual protocols for other kinds of data, ~4 hours.
2.6. Follow the protocol: words, from manual curation to Tableau
Exercise where you will use 1 new notebook.
45 min. Recommended, as you will try a new notebook, but also because it ties back to what you have done with Tableau in the first tutorials.
2.7. Extend the protocol: natural language processing
Exercise where you will use 1 new notebook.
45 min. Necessary.
2.8. Write the protocol: Annotated Tableau dashboard of Scopus data
Exercise about a new kind of data (scientometric).
45 min. Optional but recommended, as scientometric data are often a key part of controversies.
2.9. Write the protocol: Annotated Scopus author-article network map
Exercise activating all the skills you have developed so far.
45 min. Optional, but check that you can develop and run your own research design on different kinds of data.
