Blog

BLOG

A focus on visualization: Line graphs

PUBLISHED ON 13 MAY, 2014

What story is your chart telling? We want to bring you back to one of the most popular blog post series focusing on visualization, and within this broad topic, specifically get into Line Graphs.

The original article was published more than five years ago. In this window of time, Sweetspot was revamped with a new interface and improved user experience, providing appealing and refreshed visuals for the chart types highlighted then.

While the concept of Line Charts and best practices explained by Holly McKendry haven’t changed, there are new flavors of the same visualization formats that are worth replacing. That’s why all Sweetspot images have been updated.

As any chart type, the line chart has its “pros” and “cons” as well. Savor the best practices for line charts with us and reap the benefits of meeting all the criteria.

If a picture is worth a thousand words, how much is a data visualization worth to your stakeholders? It depends on how well your visualization conveys critical information so insights can be taken from it!

To help you pick the best visualization to suit your insight delivery needs, we’d like to continue our visual series today with a closer look at the Line Graph.

The Line graph, unlike the bar graph, is better at representing data over time, and visually demonstrating changes that occur. Line graphs represent data along an interval scale, allowing readers to spot trends quickly. If you’re searching for patterns in your data, such as trends, fluctuations, cycles, rates of change, or simply how any two data sets relate to one another, the line graph is your best option.

When to use a line graph	When NOT to use a line graph
Line graphs are best used to represent: Distributions of data over time: trends Continuous data	Line graphs are generally not our best option for: Close comparisons of data Representing individual values

Types of line graphs:

Single series line graphs vs. Multiple series line graphs: Line graphs can include only one series or multiple series. Multiple series line graphs are valuable for allowing you to compare the performance of multiple variables over time.

Dual axis line graphs: are used when you wish to show multiple series with very different ranges on the same graph. By adding a second y-axis you allow yourself to measure along two distinct scales. However, dual scales can be confusing for readers, so ensure that the value a second axis will bring to your graph isn’t outweighed by interpretation problems. If your graph is complex to the point that you are unable to easily interpret the data, you should consider whether a dual axis graph is really the best option for you. Afterall, the goal is for your reader to take away meaningful insights- complex graphs can sometimes be counterproductive.

Important things to consider when creating your line graph:

• Frequency: Ask yourself how often is it necessary to measure your variable: daily, monthly, quarterly? If the x-axis is so full you cannot even read the dates, like in the graph below, you’ve definitely got the wrong frequency.

Take a step back and reconsider which frequency helps you to best identify patterns. The graph below gives readers a much clearer picture of what they need to know, by choosing a more-limited frequency for the scale:

• Quantity: As well as how often you measure, take into account how much you measure. Too many variables fill the chart, causing nothing but headaches and confusion.

Try to limit the amount of variables you measure to allow your chart to be read easily, or perhaps in this case, creating multiple graphs would be more effective than simply trying to ‘jam’ all these variables into one graph

Doesn’t that look better?

• Creativity: Just like bar graphs, you should limit your creativity when making line graphs. The key is to be creative, whilst maintaining or bettering the readability of your graph. You shouldn’t impede your reader from taking valuable insights just for the the sake of beautifying your graph.

We all know this pig is “cute”, but he is distracting potential viewers from taking any insights from this graph.

To add some creativity to your graph focus on the little details, such as color, or small graphics. Subtle details may be just what you need to dress up your graph, without taking attention away from what matters most.

 

 

One way to show your creative side could be replacing the value markers with a small symbol to represent your data. Not only does this add visual appeal to the graph, it also eliminates the need to display a legend. Just be sure your graphics aren’t overpowering, or so large that the reader can’t tell where the value should be marked.

 

If you prefer to keep traditional markers, you can still eliminate the legend by directly labeling the series lines with a symbol or title. Sometimes this helps to clarify which line is which, better than a color key.

 

• Dimension: 3D graphs have a way of creating unnecessary complications, although entertaining, readers must focus time and energy trying to decipher data that could be laid out much more clearly.

• Best form of line:

Picking a smooth line, like the graph on the left, makes the series line look as if it is flowing and peaks are evident but not sharp. If curves make interpretation difficult, stick to a traditional straight line, like the graph on the right. Straight series lines leave little room for interpretation error.

Did you know you can even make line graphs using a step line? Be careful however, the form of the line may trick readers into thinking it is a poorly made bar graph.

In conclusion:

We hope this series helps you to enhance your data visualizations, allowing you to pick the graph that most easily turns your data into actionable insights. After all, the right visualization speaks millions.

What else do you take into consideration when building the perfect line graph for your reporting or insight delivery needs?