5 most common chart types and when to use them
Do you frequently find yourself in a situation when you have a chunk of data but don’t know how to present it? Do you have a rough time choosing the right chart from the dozen of options that spreadsheet programs offer you? There are bars, columns, lines, pies, areas, maps, and more. Which one is best to use? How can you fit your needs into the visualization and make the chart communicate your desired message?
First, ask yourself these questions:
- Who is my audience?
- How should my audience use the information conveyed by this chart?
- What is the main message/story of the visualization?
- What do I want to show?
The answer to the last question defines the chart type you want to use to tell your story. Here we will discuss the 5 most common chart types and when to use them.
1. Column chart
Use a column chart when you want to show the comparison between a few items. When using a column chart, make sure your Y axis starts from 0, so the scale is correct and your visualization is not misleading. Also, to make it easier for your audience to compare item values, sort the chart items either in an increasing or decreasing manner, depending again on what you want to show.
2. Bar chart
A bar chart is similar to a column chart, but you can use a bar chart when you want to show the comparison between many items, which might not easily fit in an horizontal column chart because of width limitations.
3. Pie chart
When you want to show the composition of a whole in percentages, you should use a pie chart. Always make sure that the sum of the parts equals 100%. While using a pie chart and labeling it with percentage values, always indicate the absolute value of the sum. It’s recommended that a pie chart not to be too fractionated. Use 7 or fewer items in your chart or it can be difficult for your audience to see differences between items and compare them to each other.
4. Line chart
A line chart is used when you want to show change over time. You can have multiple lines on a single chart to show change in each category. Make sure the lines are easily differentiable from each other and correctly labeled to make it easier for your audience to see change over time and the comparison between categories at a given time period.
5. Scatter chart
When you want to show the relationship between two variables, for example, GDP per capita and the software piracy rate, use a scatter plot, which shows the correlation between these two variables. Be careful and always remember that correlation does not imply causation!
Sometimes a simple table is the best way to communicate your data. That’s when:
- Your aim is for your audience to be able to find a specific value quickly;
- Your dataset is too big to be visualized in a single chart and you need to show all of the data;
- The dataset has data in various measurement units and you need show all of the data.
What to keep in mind when creating a data visualization:
No 3D charts
3 dimensions change the surface area of the chart items. This can make it difficult to visually perceive differences correctly and compare values to each other. Some people might think that 3D charts are beautiful, not mainstream, and exciting, but it removes the most important thing from the visualization: its functionality. So don’t use 3D charts.
Make it as simple as possible
When working with a large dataset for days, weeks, or even months, we tend to think that everything in the dataset is important and we want to show them in as detailed form as possible. The truth is that in most cases, our audience does not share the same excitement about that particular dataset. Their attention span is limited so we need to catch their eyes and reach their brains with our visualizations in a short period of time. Being simple really helps. Complicated charts scare people away. Losing our audience’s attention is the last thing we want to achieve.
Test your visualizations
Here at JumpStart, every visualization we create is run by our whole team to make sure that everything is clear and understandable. After that, we test our visualizations with a group of our supporters, informally called JumpStart’s external review board, and we get their feedback before a visualization is released to the wider public. You can do the same! Before putting your visualization in a report, presentation, or a news story, show it to your colleagues, friends, and family members to see if they can read it. If not, then your visualization may not do its job well and you can improve it with revisions.
Remember, a data visualization is supposed to make it easier for a reader to understand the data, see the trend, and get the message/story you are trying to communicate. Don’t make it vice-versa! We will talk about poor visualizations in upcoming blog posts. Stay tuned!