Visualize categorical values through bar charts
A brief start about Bar charts
Bar charts are one of the most common visual tools we use. They are best known for comparing data categories. Even if they seem so usual for us, we might still have some handy tips to improve their usage.
In this article, we are going to take a look at its origin, apply a bar chart to a use case and see a
few tips to improve the way we use bar charts.
A rollback in history
According to " Quantitative Graphics in Statistics: A Brief History.", [1] Nicole Oresme, a French philosopher published one of the first known bar charts around the 13th century in a publication called " The Latitude of Forms ". In fact, it can be considered as a proto bar chart and an old one!
Oresme aimed to categorize the velocity of moving objects in time.
Applying Bar Charts
In this section, we will define some key terms and get a detailed explanation of a bar Chart.
Notions
First things first, let's see some notions!
Axis is a line used to place values representing our context. For example, we placed week numbers on our X-axis.
Bar is the representation of our measure. It has a fixed width and a varying length. The length is determined by the value in our Y-axis dependant on our X-axis. The X-axis represents categorical data items and the Y-axis represents our scale or measure.
Categorical data also referred to as qualitative data represents a fixed number of values without being ordered. For instance, weeks are categorical
In " The Big Book of Dashboards: Visualizing Your Data Using Real-World Business Scenarios" [2], it is told that bars are widely used in data visualization because they are often the most effective way to compare categories.
Let's take a look at Figure 3. We can observe that we have weeks on our X-axis and Integer numbers on our Y-axis. We should keep in mind that our X-axis is not in its commonplace due to the fact that this is a vertical bar chart. Our Y-axis represents hours, more specifically the downtime duration of Product Line 1. So if we take week 3, we can interpret it as " In week 3, product line got 5 hours of duration.
This was our elementary interpretation. Now let's take a look at some tips to get a deeper understanding of the bar chart.
Some tips:
In this section, we will get some tips when using a bar chart.
Keep the width, fix... fix!
Even if most of the visualization tools do it by default, it is always better to check that the bars have the same width. In fact, our measure is the length. In other words, when the length of our bar changes, it indicates that our content varies.
Order the bars
As stated in " The Big Book of Dashboards: Visualizing Your Data Using Real-World Business Scenarios " [3] Bar charts are mostly used to spot the biggest and smallest items.
To apply this, we can improve the visibility of our message by ordering our bars. Let's consider that our message is to show the most important downtime duration. In the example below, we have the answer of it almost immediately alongside the context ( the week number and total duration ).
Y =! 0
For those unfamiliar with the subsection title, it actually means Y different from 0. As seen in Figure 5, our Y-axis starts from 2. It can lead to misinterpretation! Indeed why does our downtime duration start from 2 hours? That is why we should consider starting our Y-axis from 0.
Histograms: Bar charts?
Before concluding, we should remind ourselves that Histograms look pretty the same as bar charts.
Yet they have different purposes that we can discuss in an upcoming article.
Coming to an end
We can conclude that bar charts are pretty handy when we compare important changes among categorical data.
Concerning small changes Line charts might be useful in this case.
What about other types of bar charts such as stacked bar charts or column bars ( vertical bar charts)?
Stay tuned with us for those ones!
References
[1] Beniger, J.R., and Robyn, D.L. (1978). Quantitative Graphics in Statistics: A Brief History. The American Statistician, 32(1), pp.1–11.
[2] Wexler, S., Shaffer, J. and Cotgreave, A. (2017). The Big Book of Dashboards: Visualizing Your Data Using Real-World Business Scenarios. Hoboken, Nj: John Wiley & Sons.
[3] Wexler, S., Shaffer, J. and Cotgreave, A. (2017). The Big Book of Dashboards: Visualizing Your Data Using Real-World Business Scenarios. Hoboken, Nj: John Wiley & Sons.
Comentarios