Visualizing Data Effectively
Welcome to Part Six: Visualizing Data Effectively!
In this sub-module you will learn about:
Choosing an appropriate chart type
Using color in data visualizations
Using type and text in data visualizations
Data Visualization
Creating effective visualizations is about more than just the data. Communicating your analysis means making careful decisions about the design of your charts and graphs to help you tell a story with the data.
For example, the bar chart that we created in Python in Section 2: Accessing and Analyzing Data shows us the information we need, but it isn't descriptive or nice to look at.
This chart shows the number of land animals (True, on the right) and the number of air animals (False, on the left) that were struck by aircraft in North Carolina. However, if you were looking at this for the first time, none of that would be clear.
We could redesign this visualization to make it better!
Selecting the appropriate chart type
When creating a visualization you must consider how you want to represent your data graphically. There are two main considerations:
What type of data do you have? (categorical, numerical, textual, more than one?)
What is the function of chart? What story are you trying to tell with this chart?
Here are a few common chart types:
Bar Charts
Function: comparison
how alike or different are these data?
Data types: categorical, numerical
Line Chart
Function: change/trend
how do these data change over time?
Data types: numerical
Choropleth Map
Function: comparison
how alike or different are these data?
Data types: spatial, numerical
Scatter Plot
Function: relationship
do these data change when those data change?
Data types: numerical
Make sure you consider the function of your chart before you create it. Check out The Data Visualisation Catalogue to help identify chart types that could be relevant for your data.
Using color in data visualizations
To complete this section...
Watch the video or read the text below to learn about how to use color in data visualizations.
Using color (written version)
Click on the arrow to the right to see the written instructions.
Color can be leveraged in creating data visualizations, but there are a number of things to consider when selecting colors:
Color has cultural associations (example: red and blue have political connotations in the US)
When visualizing quantitative data, use an ordered palette. An ordered palette can be a brightness ramp (e.g., light to dark shades of grey, blue, or red) or a hue ramp (e.g., cycling from light yellow to dark blue). In general, people interpret darker colors as representing “more.”
Hue is not naturally ordered. We don't see one color as being inherently bigger or smaller than the other, and so to use this to represent quantitative data is a mistake.
Consider accessibility: some color combinations aren't effective for viewers with colorblindness. Color combinations to avoid:
Green & Red
Green & Brown
Blue & Purple
Green & Blue
Light Green & Yellow
Blue & Grey
Green & Grey
Green & Black
Using type and text in data visualizations
To complete this section...
Watch the video or read the text below to learn about how to use use type and text in data visualizations.
Using type and text (written version)
Click on the arrow to the right to see the written instructions.
Choose fonts that are easy to read.
Centered and right aligned text are difficult to read, use left aligned text whenever possible.
Give text empty space around it (especially in boxes).
Keep type horizontal.
Be consistent with your font choices.
Make sure your visualization is clearly and consistently labeled (all axes, and a title).
More Resources
Identify good chart types for your data with The Data Visualisation Catalogue
Test out color combinations for accessibility with Coblis, a color blindness simulator
Find color palettes for data with Color Brewer