πŸ’» Week 09 Lab

Revising Your Data Visualisation Strategy

Author

Dr Jon Cardoso-Silva

Published

19 March 2025

πŸ₯… Learning Goals
By the end of this lab, you should be able to: i) Identify common pitfalls in data visualisations, ii) Apply best practices to improve visual communication, iii) Critically evaluate visualisation choices, and iv) Create more effective visualisations for your Mini Project 2.
DS105W course icon

Last Updated: 19 March 2025, 22:30 GMT

πŸ“Time and Location: Friday, 21 March 2025. Check your timetable for the exact time and location of your class.

πŸ“‹ Preparation

To prepare for this lab, ensure you have:

πŸ›£οΈ Lab Roadmap (90 min)

Note to class teachers: I will leave you free to pick the two examples of visualisations to critique. You can pick examples from past DS105 projects (publicly available on GitHub pages) or from elsewhere online. Remember to share the links with your groups on Slack. After the 30 minutes of discussion, transition to individualised tech support for the remainder of the session.

🦸🏻 This lab is a TECH SUPPORT session with a focus on data visualisation!

The lab has two components:

  1. Group Discussion (20-30 minutes): A guided critique of visualisation examples
  2. Tech Support (60-70 minutes): Hands-on help with your Mini-Project 2 visualisations

Part 1: β€œWhat’s Wrong With This Plot?” Discussion

Your class teacher will lead a discussion analysing visualisations they find problematic and how to improve them.

When discussing the examples, consider:

  1. What issues do you notice with the visualisation?
  2. How does it fail to effectively communicate the data?
  3. What specific changes would improve it?
  4. How would you implement these improvements in Python?

Active Participation is Key!

This discussion will help you:

  • Develop a critical eye for visualisation quality
  • Understand common pitfalls to avoid
  • Learn concrete techniques to improve your Mini-Project 2 visualisations
  • Apply academic literature principles to practical examples
Remember the Grammar of Graphics Approach

Remember that in this course, we follow the grammar of graphics philosophy, which encourages us to think about visualisations as layers of graphical elements that map to our data. This approach has several advantages:

  1. Focus on data transformation first: The quality of your visualisation is directly proportional to how well you’ve prepared and structured your data.

  2. Declarative thinking: Rather than writing procedural code detailing how to draw graphics, you focus on declaring what elements should appear and how they map to your variables.

  3. Logical layering: Building visualisations as a series of logical layers helps create more thoughtful and effective graphics.

For your Mini-Project 2, consider these questions when planning your visualisations:

  • Have you transformed your data into the most appropriate format for visualisation?
  • Are you using the right geometric objects (points, lines, bars) to represent your data?
  • Do your aesthetic mappings (colour, size, shape) effectively convey your variables?
  • Have you added appropriate statistical transformations where needed?

Remember: Good visualisations start with good data preparation. Often, the most challenging aspect isn’t creating the plot but transforming your data into the right structure. Spending more time on data manipulation often leads to cleaner, more insightful visualisations. This is the β€œData for Data Science” course, after all!

Part 2: Tech Support for Mini-Project 2

For the remainder of the session, your class teacher will provide individualised support for your Mini-Project 2, with a special focus on:

  1. Visualisation Feedback: Get specific advice on your current visualisations
  2. Implementation Help
  3. Strategy Development: Planning your approach to visualising Reddit data
  4. Database Integration: Ensuring your visualisations connect properly with your database

🎯 ACTION POINTS:

  1. If you have existing visualisations, prepare to share them and receive feedback
  2. If you’re still collecting data, discuss your visualisation strategy
  3. Be prepared to implement changes to your visualisations based on feedback
  4. Take notes on the principles discussed to apply to your final report

Remember: The ✍️ Mini-Project 2 deadline is 26 March 2025, 8pm UK time. Use this session to refine your visualisations before the deadline!

πŸ“š Resources