When to Use Bar Charts Instead of Pie Charts
Unless you’re a data scientist or you don’t have another data visualization job, choosing the right way to present information can be overwhelming. Software like Excel can insert any type of data into just about any type of chart, but which one is right for the story you’re trying to tell?
Bar charts and pie charts are two common ways to represent data , and while they can sometimes be used interchangeably, they are not appropriate for every type of dataset. Here’s how to choose between a bar chart and a pie chart.
When to Use a Histogram
Histograms have two axes: categories (months of the year) on one and numerical value (number of visitors to the national park) on the other. So, you are comparing the total number of visitors in January with the number of visitors counted in other months, and perhaps observe which months have the most visitor traffic. Histograms can be oriented horizontally or vertically.
Bar charts can also be used to combine datasets or show comparisons of sub-categories, such as the number or percentage of votes received by each candidate for office, by age group of voters.
When to Use a Pie Chart
On the other hand, pie charts show categories as a fraction of a whole . You can think of a circle as a sum – 100% – and the width of each slice is determined by how much of the sum it represents compared to other slices. Using a similar example, a pie chart could show the percentage of votes received by each candidate out of the total votes cast by all voters.
Pie charts are only really useful when the individual data points you include make up a whole, and that whole is itself part of the story. If you’re trying to show how a part relates to a whole, such as what percentage of your budget is spent on rent, a bar chart might not make it as clear.
When is it better to use a bar chart rather than a pie chart
There are several scenarios where, despite the fact that the data represents parts of a meaningful whole, a pie chart would be a poor choice:
- Too many categories . If the pie slices are too narrow, it is difficult to determine their relative size. In addition, the chart becomes crowded and the labels and annotations are hard to read.
- The fragment sizes are too similar . If the parts of a circle are of close but not equal proportions, the relative size can be difficult to interpret. At first glance, and without clear labels, pie charts may not accurately convey the exact proportion that a slice represents.
- You have multiple datasets . Pie charts can only represent one set of data, which, again, makes up the whole. For example, they cannot compare salaries for certain job categories, separated by gender.
While pie charts are especially good for showing percentages of a whole, bar charts can also be useful for showing percentages rather than absolute numbers, especially when the above limitations exist or if the categories aren’t mutually exclusive—for example, a survey that allows users to choose multiple answers or categories for one question.
In general, use a bar chart when you’re comparing category to category, and a pie chart when you’re comparing part to whole. But a bar chart can still be better for visualizing part-to-whole if you have many parts (more than five) or if the proportions of the pie slices are difficult to determine (for example, multiples of 1/3 or 1). /four). There is no turning back: pie charts are useless for comparing categories if the whole has no value. Conclusion: Bar charts are simply more versatile than pie charts.