Use pie charts to compare the sizes of categories to the entire dataset. To create a pie chart, you must have a categorical variable that divides your data into groups. These graphs consist of a circle (i.e., the pie) with slices representing subgroups. The size of each slice is proportional to the relative size of each category out of the whole.

These charts can use proportions or summary statistics to determine the sizes of the slices. For example, you can create a pie chart that shows the proportion of each sales type (electronics, software, accessories, etc.). Or create one that displays total sales by those categories.

Pie charts shine when you need to assess the relative sizes of categories to the entire dataset.

At a minimum, pie charts require one categorical variable. To learn about other graphs, read my Guide to Data Types and How to Graph Them.

## Example Pie Chart

A company wants to determine the proportion of employees in each job category.

Pie charts typically contain the following elements:

- Circle (“pie”) representing all observations.
- Circle segment (“pie slice”) for each category.
- Optionally, slices can have labels indicating their relative (percentage) or absolute size (count or summary statistic).

For the company, the largest category is manufacturing, followed by R&D. The smallest group is janitorial.

## Interpreting Pie Charts

Pie charts provide a broad overview of the categories you’re studying. By comparing and contrasting the size of the slices, you can evaluate the relative magnitude of each group. In the chart below, four colors (white, silver, black, and grey) comprise nearly three-quarters of all new car colors in 2012.

When assessing more than one pie chart, compare the sizes of the categories between charts. Understanding how the slices for the same groups change between pie charts can help you recognize the relationships in your data. The graph below displays total sales by category for two locations. The East location has relatively more Laptop sales and fewer Desktop sales than the West location. The two locations are approximately equal for Mobile and Software sales.

## Limitations of Pie Charts and Comparing Them to Bar Charts

Pie charts effectively illustrate the different sizes for parts of the whole. However, these graphs have shortcomings that can limit their usage. To use a pie chart, consider the following:

- Use when your primary goal is to compare the parts to the whole.
- The category totals must add up to the overall total.
- Pie charts are best for simple data arrangements.

When these three points are true, pie charts are a compelling choice. People viewing them will understand the data easily. However, if these three points don’t apply to your data, consider a different graph. Pie charts require categorical data. Consequently, bar charts are an excellent alternative because they also use categorical data *and* have greater formatting flexibility.

Let’s explore a couple of these limitations in more detail.

**Related post**: Guide to Bar Charts

### Parts Must Add Up to the Whole

Pie charts can display summary statistics for the categories, but the parts must sum up to the whole. For example, they can display total sales by group because the group sales add up to the total sales.

However, if you want to graph the average sales price per transaction by category, you can’t use a pie chart. Average sales by category do NOT add up to the average for all sales. In other words, average category sales are not parts of a whole and, therefore, inappropriate for pie charts.

Fortunately, category averages are just fine for bar charts because those graphs do not assume the variable function (i.e., average) represents parts of the whole.

Bar charts provide greater flexibility for choosing metrics to compare across groups because they don’t need to sum to the total value.

### Too Many Categories and Too Few Formatting Options

Even when you have appropriate data where you are comparing parts to a whole, you might need to use a bar chart instead.

Maybe your data have many groups, and the slices become too small to identify. Perhaps you have two categorical variables and want to explore the relationship between them. Or your categories have subcategories.

Pie charts provide relatively few formatting options to handle more complex data arrangements. However, bar charts allow you to stack, cluster, and otherwise organize the bars in ways that can handle more complex data and many categories.

In short, pie charts are a fantastic option for comparing parts to a whole when you have a simple data arrangement with a reasonable number of categories. Unfortunately, various data issues can either be invalid or create pie charts that are hard to interpret. In these cases, consider using a bar chart, which is more flexible on all those fronts.

## Comments and Questions