In our Data Visualization 101 series, we cover each chart type to help you sharpen your data visualization skills.
For a general data refresher, start here.
Bar charts are a highly versatile way to visually communicate data. Decidedly straightforward, they can convey the message behind the numbers with impact and meaningful clarity, making complex data easy to understand at a glance.
Let’s look at what makes bar charts so great and how to best use them.
What It Is
The bar chart is a chart with rectangular columns proportional in length to the values they represent. Simply put, longer bars equal bigger numbers. On one axis these bars compare categories, while on the other they represent a discrete value.
A small slice of bar chart goodness from our foolproof guide to grilling meats.
Where It Came From
Some have credited 14th century Frenchman Nicole Oresme with the invention of the bar chart. He was certainly onto something when he vastly predated Sir Isaac Newton’s ideas on physics and proposed plotting the velocity of objects over time. His proto-bar charts lacked one thing, though: data.
Like the pie charts we previously covered, the bar charts we know today owe their success to data visualization pioneer William Playfair. He first included them in his 1786 work, “The Commercial and Political Atlas.”
Playfair updated the detailed work in 1801, noting that he was onto something with his newly invented statistical graphics:
“That I have succeeded in proposing and putting into practice a new and useful mode of stating accounts, has been generally acknowledged, that it remains only for me to request that those who do not, at the first sight, understand the manner of inspecting the Charts, will read with attention the few lines of directions facing the first Chart, after which they will find all the difficulty entirely vanish, and as much information may be obtained in five minutes as would require whole days to imprint on the memory, in a lasting manner, by a table of figures.”
In the graphic, Scotland’s imports and exports with 17 countries in 1781 are represented.
Playfair’s calligraphy-filled “Chart Shewing the Value of the Quarter of Wheat in Shillings & in Days Wages of a Good Mechanic from 1565 to 1821.”
When to Use It
Bar charts are used to showcase discrete data—that’s data that is based on counts and can only be certain values. They are most effectively to:
- Show change over time (e.g., the net monthly earnings of Tesla Motors in a year)
- Compare values of different categories (e.g., a year’s fishing yields for different species of fish)
- Compare parts of a whole (e.g., the percent distribution of Netflix rentals across genres)
The standard bar chart can be used with bars aligned either vertically or horizontally. Vertical orientations are best used for chronological data or when negative values are involved.
Horizontal orientations are best used when charting different categories, especially with long labels.
Stacked bar charts allow you to showcase and compare part-to-whole relationships.
For instances where the percentage distribution of categories is the key message (not total value), a 100% stacked chart is ideal. (Remember when we said bar charts are sometimes better than pie charts? This is one of those instances.)
Best Practices for Designing Bar Charts
1. Start the Y-Axis at 0.Starting at a value above zero truncates the bars and doesn’t accurately reflect the full value.
2. Space bars appropriately.The space between bars should be roughly ½ bar width.
3. Use consistent colors.
Use one color for bar charts. You may use an accent color to highlight a significant data point.
This post originally appeared on Visage.
Originally published on Jan 7, 2015 12:16 PM