Data Visualization

Page Banner
Why Visualize?

Through data visualizations such as graphs, maps, etc. We can identify important patterns which allows us to better understand our data to perform further analysis. At the same time, a good data visualization will also allow us to easily communicate information about our data.

Understanding Visualization

Sometimes you may encounter a visualization that you do not understand. You might just not be familiar with that type of visualization, often because different disciplines have their "go-to" types of visualizations that work well with their data.
Or perhaps you haven't dealt with charts and graphs a lot before. You may need to work on developing your data visualization literacy here or here.
Check out these links to help you make sense of what you find:

Understanding Visualizations by IBM

  • ​​​​​Quick general overview of common visualizations

The Data Visualization Catalogue
  • Explains many visualizations with breakdowns that clearly show you how to read them
7 Data Visualization Types You Should be Using More (and How to Start) by Evan Sinar
  • Explains how to understand and use some less familiar visualization types
Top 5 things to look for (or visualisation in a hurry) by SeeingData
  • Tips for finding the key information to look for in a visualization. Check out this whole section on data visualization literacy for more tips on understanding visualizations
Data Visualization for Human Perception by Stephen Few
  • Explains how to create a good data visualization and examines the effective of different types of data visualization
How to read and interpret network graphs by Nodus Labs
  • Explains how to understand network graphs
Dot Plots: A Useful Alternative to Bar Charts by Naomi Robbins
  • Bar charts/graphs come up often, learning about some alternative formats, not as commonly mentioned, might be a useful addition to your data visualization toolkit
Why not to use two axes, and what to use instead by Lisa Charlotte Rost
  • Dual axes graphs can be problematic. Read this article to make sense of the problems and alternatives to a dual axes graph
Best Practices of Visualization
  • Don’t manipulate your visualization to better fit your argument.
It is easy to manipulate an axis to exaggerate small differences or de-emphasize differences in a data visualization. Create data visualizations that are true to your data to allow better understanding.
  • Choose a data visualization that fits the data and your purpose
Always choose a data visualization that fits the data and your purpose for visualization. For example, when comparing data, one may choose to use a bar graph. When trying to identify trends, use a scatter plot. When trying view the change over time, use a line graph.
  • Don’t overcrowd a data visualization
Simpler data visualizations are easier to understand. If a lot of information needs to be displayed, try to break it up into multiple graphs.
  • Choose a data visualization that doesn’t hinder understanding
Fancy data visualizations may look nice to the eye, but they are often hard to understand. They will likely distort the data to defeat the purpose of data visualization. Take a 3D pie chart as an example. Due to perspective, it will be much more difficult to the proportions of each slice.
Galleries and Examples

Tableau Public Gallery
Google Charts Gallery
R Graph Gallery
Stata Graph Examples