Bar Graphs vs. Line Graphs - a comparison

Data visualization tools are powerful for presenting complex information in a visually appealing and easily understandable way. However, when the presenter doesn’t choose the correct data visualization tool, they do not add clarity. Choosing the wrong chart type or graphic can distort the message and lead to confusion. It is important to present the data accurately and objectively, without distorting or skewing the information to fit a particular narrative. In this article, I will discuss the proper use of Bar Graphs and Line Graphs.

Bar Graphs	Line Graphs
Show value at a specific point in time, or total value.	Shows changes over time
Good for comparing multiple values at the same point in time.	Good for comparing multiple data points over the same duration of time.
Harder to show trends	Good to find trends

The Bar Graph, also referred to as Bar Chart or Column Graph, is a visual representation of data that uses rectangular bars of varying lengths to display and compare different values or groups of values. Each bar on the graph represents a specific category or group, with the height or length of the bar indicating the magnitude or value of the data being represented. One axis of a bar graph represents the categories or groups being compared, while the other represents the scale or measurement being used. The bars on a bar graph can be arranged horizontally or vertically depending on the preference of the presenter. Sometimes a third value, such as time is added, and the graph can be animated to better show the new dimension.

This type of graph is ideal for comparing data across different categories at a specific point.

Bar graph of Best-Selling Music Artists 1969 - 2019

The Bar Graphs used in the video above looks beautiful and seems to effectively shows the record sales from 1969 - 2019, showing when new artist enter and exit the chart with the addition of the timestamp in the bottom right corner.

“Trend spotting” is not possible with 1 bar graph, however, because this is a video, the viewer is able to compare multiple Bar Graphs overtime to gain more insight into the data.

For example, if you want to know when Michael Jackson first enters the chart you have the ability to find the timestamp, and view the 2 frames where MJ isn’t there, followed by the one where he is, basically comparing 2 bar graphs, and guessing from the time in the bottom right. He shoots up the charts in Q4 of 1979, Off the Wall was released in August ‘79.

While trends are harder to spot using a Bar Graph, total magnitude is usually easy. The more significant the difference, the easier it is to recognize using a Bar Graph. When the magnitude of the difference is less significant, as in the bottom 3 usually, it is hard to decipher without additional information or captions.

Ultimately though, without using the video format, this data cannot be displayed as a Bar Graph. A Bar Graph does not provide the most important thing, values over time.

What about a Line Graph?

A Line Graph would allow you to see trends over time and it would also show time, so you can mark specific dates that effected record sales. A Line Graph would show the exact moment, and the direct effect, via the slope of the line, to illustrate the trend following that moment.

Examples: The adoption of cassette tapes, cds, MJ allegations, and Napster and streaming.

One final issue with this video, is that the raw data set is not available, and the data conversion method is a secret. According to the video description, the data set is

yearly certified record sales. Numbers are worldwide and adjusted to twelve months trailing average.

and not total so as a viewer I am forced to make assumptions. In this case, I assume that the data resets yearly, this may explain the over-exaggerated jumps and falls in popularity.

Overall, it’s a cool visual, but I’m still confused on the numbers especially when some of them descend as a year goes on. How can I sell a less total in Q4 than Q2? Let’s take a look at some other music sales data over time, but this time let’s see how a Line Graph handles it.

The Line Graph, also referred to as Line Chart or Line Plot, is a visual representation of data that uses lines to connect individual data points. Line Graphs are used to compare multiple quantitative values over a specific period of time. One axis of a line graph, usually the x, represents the time period being covered, while the other, the y, usually represents the values being measured. The lines created by connecting the data points over time show data trends.

This Line Graph above is a great way to visualize LP sales for the 20 year period between 1993 and 2013. As stated in the table this is a line graph because it is showing changes over time and trends. At first glance, it quickly shows an overall upward trend since ‘93, dips due to the rise of digital after ‘00 and resurgence in ‘08 since. If Cds and Digital streams were closer in total, a line could be added to represent those formats and pinpoint the date between ‘00 and ‘10 when Digital took over, and the downward trend of cd sales. A bar graph would not be able to convey the same data points, or type of information.

Screen shot from the Rise and Fall of music formats are line chart

There are many different examples of Line Graphs. Above are just two examples, a simple, The LP is Back, and an advanced line graph, The Rise and Fall of music formats. The Rise and Fall is an example of an Area Line Graph, taken from an interesting article about all sales of recorded music over a 50 year span. The Area Line Graph uses the total area to show the total value and the area, or specific lines and colors, represents the specific value.

The animation in the article is nice but unnecessary because the real power comes from this view, the final graph.

So what do I use?

BOTH! Bar Graphs and Line Graphs have the potential to wield great power when used correctly. However, when the designer confuses the role of the visualization, making it more decorative and less informative, the data can get obstructed or skewed, this can leave the end user confused and frustrated. 

For a quick refresher, refer to this table.

Bar Graphs	Line Graphs
Show value at a specific point in time, or total value.	Shows changes over time
Good for comparing multiple values at the same point in time.	Good for comparing multiple data points over the same duration of time.
Harder to show trends	Good to find trends

Editing assistance from Jaime Radakovich . Thanks J.