A dual axis chart is a great way to easily illustrate the relationship between two different variables. They illustrate a lot of information with limited space and allow you to discover trends you may have otherwise missed if you’re switching between graphs. However, many data analysts hate them, but why? Dual axis charts, if used incorrectly, can be misleading. This post will walk you through when to use dual axis charts, and when to stick to a single y-axis.
Do Display Variables With Vastly Different Scales
To easily see two variables with very different scales, it’s much easier to see them within the same graph than to flip between two charts. It also uses up less real estate in Powerpoint presentations. In DataHero, simply drag on two numerical variables (plus another variable like date) and you’ll see a dual axis chart like the one below that displays the average price per square foot versus average rent per square foot:
We can now easily visualize these two variables within the same plane. The average rent per square foot is calculated on a monthly basis, whereas the average home price per square foot is calculated based on the overall sale price. This is why the scales have such a large discrepancy. By plotting these two variables on the same chart using a dual y-axis we can see that the average price per square foot dipped slightly in 2012 but increased again throughout the end of 2012 and into 2013. The average rent price per square foot, however, has increased steadily from January of 2011- August of 2013. Whereas owning is affected by changing market conditions, renting seems to move in only one direction.
Do Display Variables With Different Numerical Units
Similar to the previous idea, you can display two numerical values with different units with a dual axis chart as well. In the chart below we can see the relationship between number of homes for sale and average home price in California, for example. This compares a currency value with a rounded number value.
Don’t Overwhelm your Charts
A chart can only hold so much information before it becomes overwhelming. Stick to making one point with your chart, by displaying the relationship between two variables. In the example below, there are four lines in addition to the columns.
It becomes very visually confusing, switching between checking the legend and the chart over and over. You can’t even see the bottom line except for the triangles because the axes are so vastly different.
Don’t Forget Charting Best Practices
This one seems obvious but it’s easy to forget. Just because you have two axes doesn’t mean other charting best practices don’t apply. For example, a line chart isn’t the best way to represent categorical data. They’re best reserved for trends over time.
Lines imply a trend over time. Thus, the line in the chart above is confusing because it’s representing a trend with categorical data. Have more questions about the appropriate type of chart to use? Check out this post.
If you’re ready to give your own dual axis charting a try, sign up for a free DataHero account and import your own data.
Get the fastest, easiest way to understand your data today.Sign up for free