A Guide to Bar Charts

In previous sections of this guide on descriptive statistics, you learned everything from constructing bar charts to when they should be used. In this section, you’ll find an overview of all you’ve learned from previous chapters, including all the different types of bar charts. In addition, we’ll also summarize the best practices for constructing and displaying bar charts. 

Bar Charts

If you’ve ever kept up with the news, you’ve likely stumbled across a bar chart. The ubiquity of bar charts is in part due to their simplicity. While they can be very basic, they’re powerful tools for conveying information about any data set.

Bar charts are often confused with histograms because of the fact they both visualize data using bars. The main difference between a histogram and a bar chart has to do with the kind of data that they display. While histograms display two quantitative variables, bar charts display one or more categorical variables against a quantitative variable.

In the table below, we’ve summarized the main differences between histograms and bar charts by comparing their characteristics.


CharacteristicHistogramBar Chart
Vertical AxisQuantitative variableQuantitative variable
Horizontal AxisQuantitative variableCategorical variable
BarsNo space between themSpaces between them
Order of BarsOrder mattersOrder doesn’t matter


In general, there are three types of bar charts, which include:

  • Clustered Bar Chart
  • Stacked Bar Chart
  • 100% Column Bar Chart

All of these bar charts are used for different purposes and each can be constructed either vertically or horizontally.




Clustered bar charts are what you typically think of when bar charts are mentioned. They typically look like the image below.


Bar chart


Looking at this image, we can see that there are categories on the horizontal axis and a quantitative variable on the vertical axis. Sometimes, to better display the data, people construct bar charts horizontally. This looks like the following image.


Horizontal Bar Chart


As you can see, the data is now displayed horizontally instead of vertically. This means that the categorical variable is now displayed on the vertical axis while the quantitative variable is now represented by the horizontal axis.These clustered bar charts are used when you want to make a simple comparison between the total amounts of categorical variables.



Stacked bar charts are bar charts that are used when you want to compare the totals and proportions of each categorical variable. They look like the image below.


Stacked bar chart 2


In the image above, you can see that there are two qualitative variables, gender and city, with gender having two categories, male and female. In the stacked bar chart, the totals are represented by the bar as a whole. These bars are, in fact, the same as in our previous example. The difference lies in the division of these totals between the two categories.Meaning, you can compare both the total for each city’s population as well as the division of gender within each total population count. Stacked bar charts can also be presented horizontally, which is shown in the image below.


Horizontal Stacked Bar Chart


Here, just like the last example for bar charts, the quantitative variable is now on the horizontal axis while the categorical variables of city and gender are now on the vertical axis. 

100% Column

While it may sound strange, the 100% column bar chart is also an easy bar chart to comprehend. This bar chart is used when you want to compare the proportion of each categorical variable to the total amount. This would look like the image below.


100% stacked column 2


Notice from the image above that the 100% column chart is no different from the clustered and stacked bar chart in terms of the placement of the variables. The only difference is that now, instead of displaying the quantitative variable as an integer, the total population is presented in percentages.This is because the chart reports the proportion of each category of gender compared to the total population size. Meaning, from the chart we can see that males represent about 45% of the total population in all three cities. Regardless of the fact that each city has a different total population, they all have about the same distribution of males and females. This graph can also be displayed horizontally.


Horizontal 100% stacked column


This time like in clustered and stacked bar charts, the qualitative data is displayed on the vertical axis and the quantitative variable is displayed on the horizontal axis. 

Combination Charts

What you may already know is that visualizations can be combined in order to better present information. These types of charts are generally known as combination charts. Bar charts are typically only combined with two other types of charts: line and area charts.

Using fictitious average ages, we can also give information using a combination chart. Using a line chart to convey this information would look like the following.


Combination Line and Bar Chart


Using an area chart, we could display the same information in the following chart.


Combination Area and Bar Chart


A line chart would be better to use in this case as area charts are typically used more for indicating a change in volume across categories or time. However, for illustrative purposes, we’ve included it here.Notice that the second quantitative variable has been placed on what is known as a “secondary axis,” which basically means that the line and area chart follow the second vertical axis, marked on the right side as “Average Age,” and not the first one.


Best Practices

In the table below, you’ll find a summary of the best time to use each type of chart we’ve reviewed in this section.


Type of Bar ChartWhen to Use It
ClusteredUsed when wanting to compare totals between different qualitative variables or changes over time
StackedUsed for comparing totals and the different categories within one or more qualitative variables or for changes over time
100% ColumnUsed when wanting to compare the proportions of different categories within one or more qualitative variables or for changes over time


In general, when constructing any type of chart, there are a couple of general rules you should stick to in order to make sure the information you want to display is as understandable as possible. These best practices can be found in the table below.


Uses of Bar ChartsBest Practices
Constructing Bar ChartsWhen constructing a bar chart, you can make it easier for readers by:

  • Not crowding the bars to closely together
  • Making the units understandable to the audience
  • Make sure you are displaying a quantitative variable and one or more qualitative variables
Displaying Bar ChartsWhen choosing how to display your bar chart, make sure you:

  • Always start the vertical axis, where the quantitative variable is, at zero
  • The exception to the previous rule is when there are no marked differences between the bars – in which case you can start the axis at a higher value to make these differences more apparent
  • Be cautious when using effects such as shadows or 3-D effects


Did you like the article?

1 Star2 Stars3 Stars4 Stars5 Stars (1 votes, average: 5.00 out of 5)


Located in Prague and studying to become a Statistician, I enjoy reading, writing, and exploring new places.

Did you like
this resource?


Download it in pdf format by simply entering your e-mail!

{{ downloadEmailSaved }}

Your email is not valid

Leave a Reply

Notify of