Data Visualizations
Tips for Evaluating Data Visualizations
1. Identify the intended purpose of the visualization
Ask yourself: What is the purpose of the visualization? What is the claim?
2. Consider other factors that may shape the data and the visualization.
Ask yourself: What wasn't measured that could impact how the data is represented?
3. Consider the creator
Ask yourself: How may the creator be biased? Can you find more about who did the primary research? Do they have authority and experience in this area? How big was the sample size and how inclusive was it?
4. Reflect and Interpret
Ask yourself: what is the takeaway from the visualization? After evaluating the visualization, did your understanding of its claim change at all?
5. Infer further
What other information can you infer based on your interpretation of the visualization?
- The Data Visualization CatalogueAn overview of different types of data visualization, including descriptions, examples and tools for generating.
- Visualizing Your DataAre you creating your own data visualizations? Check out our guide on visualizing your data and get in touch if you need help!
Evaluating Data and Statistics
It's just as important to evaluate the underlying data and statistics being represented in the visualization. Statistics can be easily manipulated or shared out of context in all types of media.
- Analyzing StatisticsQuestions to consider when evaluating statistics.
- Becoming Data Literate in 3 Simple StepsProvides guiding questions for evaluating data; from the Data Journalist Handbook I.
Examples of Misleading Data Visualizations
Truncated Axis
Take note of where the Y axis starts: 1.4 million instead of 0.
Corrected Axis
Changing the Y axis baseline to zero has a big impact on the visualization. In the previous version, there looked to be more significant differences in the number of visits, but with the adjusted Y axis, we can see that is not the case.
Confusing Pie Chart
Each slice of the pie is the same size but the labels showing different values add up to 193%. Each person was asked about their opinion of each candidate, so a pie chart is not the best graph for representing this data.
Better Bar Chart
Instead of a pie chart, this bar chart is a better was to represent the data of each person's response about each candidate. The baseline starts at zero and the different colors help distinguish the different candidates.