Skip to content

Data Visualisation June 2020

Jump to bottom
Gabriel Bodard edited this page Aug 25, 2020 · 42 revisions

Data visualisation for Ancient and Modern History, Languages and Literature

Data visualisation

Tutors: Jonathan Blaney, Gabriel Bodard, Christopher Ohge, Naomi Wells

  • This online workshop will use a mix of real-time and asynchronous teaching to introduce participants to a range of text and data visualisation tools and methods. We will work hands-on with online resources such as Voyant Tools, Tableau Public, and Raw Graphs, and discuss some of the issues and implications of visualisation in academic work and media more broadly. This event is targeted to research students and early career researchers, but all are welcome; booking is essential. Please commit to attending all three sessions. The workshop will involve sessions at 2-3pm on Wednesday June 10, Thursday June 11, and Thursday June 18, with individual and group work to be carried out in between.

Thank you for registering for the Data visualisation for Ancient and Modern History, Languages and Literature workshop jointly run by the IMLR, IHR, IES and ICS at the University of London. This training will be offered over three separate short meetings at 2pm (UK time), on Wednesday and Thursday next week and the following Thursday. It is essential that you commit to attending all three sessions, and to doing at least a couple of hours of preparation and practice between the sessions; there will also be group work and discussion.

All of the exercises involving Voyant Tools, Tableau Public and RawGraphs will involve the online versions of these tools. If you want a slightly more stable version of the tool (and to be prepared in case websites are down or similar) you may optionally install the desktop version of Voyant (https://github.com/sgsinclair/VoyantServer) and/or a 14-day free trial of Tableau (https://www.tableau.com/en-gb/products/desktop/download).

Discussion and theory

Readings

Example visualisations

  1. History of the Atlantic slave trade from Slate.

  2. W.E.B. Du Bois Paris Exposition

  3. Mona Chalabi museum demographics

  4. Ages at Death in Roman Cyrenaica

  5. ClimateSpiral (by Robert Rohde)

  6. Pepsi Rebrand (2008) (see whole Twitter thread)

  7. US Airforce Cross Domain Escalation (see Twitter thread)

  8. Election leaflets from LibDems and Conservatives.

  9. Niger maize prices at market

Discussion questions

  1. What cognitive and/or affective effect does this visualization have on you as a viewer/reader?
  2. How do you think that effect differs from reading the same data in other forms (e.g. as text or in a spreadsheet)?
  3. Can you imagine other ways this data could be visualised that would be more/less effective?
  4. What do you imagine is the intent of the creator of the visualisation by presenting the data in this way?

Practical session 1: Voyant Tools

Tutorials

Exercise

  1. In one tab, go to https://voyant-tools.org.
  2. In another tab, open the folder of Melville texts on the github repo. When downloading, make sure to click on the Raw button before downloading. Visualise in Voyant by clicking Upload. (Navigate to the folder in which you downloaded the texts; in the folder, hold down the command or Windows button to select all three texts.)
  3. We will divide into four groups:
    1. Group 1: Early Melville. Download melville_typee.txt, melville_omoo.txt, and melville_mardi.txt. (Jessica, Joey, Sara)
    2. Group 2: Mid-career Melville. Download melville_moby-dick.txt, melville_pierre.txt, and melville_confidence-man.txt. (Ellis, Sarah, Colleen)
    3. Group 3: Melville the poet. Download melville_battle-pieces.txt, melville_late-poems.txt. (Jordon, Kelly, Emma)
    4. Group 4: Melville's last work. Download billy-budd.txt (Carlo, Simona)
  4. Each group should write up some initial impressions of what kinds of themes and ideas might be showing in these groups of texts.
    1. Did you edit the stop words list? (Did you notice any stop words in the results?)
    2. What word trends did you find compelling?
    3. List and take note of three prominent word linkages (hint: use the Links and/or Terms Berry feature).
    4. What collocates distinguished themselves in the text?
    5. Did keywords-in-context searches of high-frequency terms change your mind about themes and ideas?
  5. Export your results (or one of your individual visualisations) to a URL and post the link to the Workshop Forum.
  6. Have a look now at this visualisation of all of Melville's texts.
    1. Do the results surprise any of you, given your attention to a selection of texts before?

Practical session 2: Tableau Public and RAWGraphs

Tutorials

Exercise

  1. You will be assigned a dataset as a group, and be instructed to work with either Tableau or RAWGraphs
  2. After watching/reading relevant tutorials and familiarising yourself with the dataset in CSV form, open the dataset on Tableau/RAWgraphs
  3. Although it will be useful to discuss initial reflections on the dataset with your group, you may find it easiest to try creating visualisations with your dataset individually as you familiarise yourself with the Tableau/RAWGraphs tools
  4. Discuss and share with your group practice visualisations or questions about the dataset
  5. As a group, aim to agree on 2-4 visualisations to present to the whole group at next week's sessions
  6. Particularly, in Tableau, there are many different ways to export or share your visualisations (e.g. you may want to experiment with creating a public dashboard) but for the purposes of this exercise, it will be easiest to share image/vector files with the tutors to be added to the GitHub repository

Groups and datasets

Additional datasets to explore (all incomplete or inconsistent for other reasons, but try to look at a couple of them and see what you can make of them)

Discussion questions for Exercise 1 and 2

  1. What were the advantages/limitations of these tools in relation to the types of visualisation you wanted to create?
  2. Did visualising your data using these tools make you think differently or reveal things that weren’t evident to you in CSV/text form?
  3. What challenges did you encounter with the datasets themselves and did you have to make any changes to the CSV/text files to produce more effective visualisations?
  4. How effective were these tools for visualising complex or ambiguous data? Was there anything you felt the visualisations you created obscured?
  5. Did others in your group tend to produce similar or very different visualisations, and why do you think this is?
  6. How could you reconceive of your visualisation as capta?
  7. Is your visualisation representing/communicating only the results of the analysis by Voyant/Tableau, or communicating your own research and interpretation?

Further resources

Clone this wiki locally