Part 4 (I): Do you ought to show composition?

That is the fourth article in a series geared toward helping people to make a decision which style of chart to make use of in keeping with the message they are attempting to indicate to their particular audience.

The previous three articles focused on the next messages: Article 1, displaying the distribution of a single numerical variable; Article 2, showing the magnitude of a series of numbers; Article 3, comparing items.

The aim of this text is to point that are probably the most commonly used charts when showing Composition. Keep in mind that Composition pertains to a Whole that might be divided into individual Parts and the way each Part relates (absolutely or relatively) to that Whole. The evaluation might be Static (shows composition at a moment in time) or Dynamic (shows changes in composition over time).

Charts regularly used for displaying composition are as follows:

· Pie Charts

· Stacked Bar Charts

· Stacked Area Charts

· Waterfall Charts

· Mekko Charts

· Treemaps

In this text we are going to consider describing the next chart types: Pie Charts; Stacked Bar Charts; and Treemaps. In the next article, we are going to describe the remaining three.

Pie Charts (PCs) (Figure 1) are circular diagrams divided into wedged-like sectors used to display Parts of a Whole of mutually exclusive and never overlapping categories. The total circle represents the Whole while the wedges (slices, sectors, segments) represent the Parts. So, the total circle must represent the sum of all data and must consistently add as much as 100%. Numerical data included in a single slice must not be included in one other slice because, as previously indicated, sectors have to be mutually exclusive and overlapping is forbidden. Conceptually, they indicate a straightforward share of the Whole.

PCs encode numerical values through two visual markers: 1) the world of every sector; 2) the length of every sector across the perimeter of the circle. Unlike most other charts, the axis and scale of a pie chart are usually not linear.

It just isn’t easy for human beings to visually calculate areas or distances along the perimeter of a curve. That is the fundamental objection to this kind of chart and the origin of an infinite controversy: they’re quite simple to make, and audiences are accustomed to their use, but they’re very difficult to interpret in the event that they don’t include annotations and percentages that make clear the context.

Sometimes, the message delivered by PCs might be enhanced using the next alternatives: A1) Donut Charts; A2) Segment Separation.

A1: Donut Charts (Figure 2), conceptually such as pie charts, differ from them in that they’ve a blank space (like a hole) in the middle of the diagram where some type of additional information is displayed to boost the storytelling.

The blank space in the middle doesn’t allow to make a comparison of areas, so donut charts have just one visual marker: numerical values of each sector are only encoded via arc lengths along the perimeter of the circle.

A2: Segment Separation, the message might be enhanced by pulling out or separating one segment (or just a few) from the usual pie chart or the donut chart.

In fact, there have to be a well-founded reason to justify such a separation because, inevitably, the audience’s attention will likely be focused on that sector. As well as, there’s a visible distortion that makes it difficult to make direct comparisons with other sectors.

Finally, Pie Charts only show composition at a moment in time (Static Composition). More details about PCs might be present in my previous article.

Stacked Bar Charts (SBCs) (Figure 4) are rectangular bars that might be oriented vertically (horizontally). They’ve two axes: one axis shows categories, and the opposite axis shows numerical values with its corresponding scale. Each bar represents a principal category and it is split into rectangular sectors representing subcategories of a second categorical variable. The numerical value of every subcategory is shown by the peak (length) of those rectangular segments which might be stacked end to finish vertically (horizontally). The ultimate height (length) of every principal bar indicates the full amount of every category (except in 100% stacked bar charts).

There are two particular kinds of SBCs: 1) Easy Stacked Bars (Figure 4); 2) 100 Percent Stacked Bars (Figure 5).

Easy SBs place the absolute value of every subcategory over (after) the previous one whilst 100 Percent SBs place the percentage of every subcategory over (after) the previous one. Principal bars in Easy SBs habitually have different heights (lengths) whilst all of the principal bars have the identical height in 100 Percent SBs. You should use 100 Percent SBs when only relative differences matter while using Easy SBs when relative and absolute differences matter.

SBCs excel in showing composition changes over time (Dynamic Composition). For this kind of dynamic evaluation, it is crucial to make use of stacked bars oriented vertically with the variable related to time (days, months, years, temporal ranges) all the time on the horizontal axis (Figure 6).

Caution needs to be exercised with the variety of stacked sectors or when charting over long periods of time. It’s advisable to not stack greater than 4 or five sectors on each principal bar. The audience might also get confused when there are too many principal bars or greater than three sectors for very long periods of time. Given this example, our advice is to employ stacked area charts when you should display lots of temporal data and/or 4 or more sectors per principal bar.

More details might be present in my previous article.

This particular style of chart was invented by Ben Shneiderman, professor of Computer Science on the University of Maryland, when he was searching for “a compact visualization of directory tree structures” (#2).

In my very own words: “A Treemap is a rectangle-based visualization that lets you represent a hierarchically-ordered (tree-structured) set of information. The conceptual idea is to check quantities and show patterns of some hierarchical structure in a physically restricted space. For that purpose, rectangles of various sizes and colours are used to display the dataset from different perspectives. The goal just isn’t to point the precise numerical values but to ‘break’ the dataset into its constituent parts and quickly discover its larger and smaller components” (#3).

It was later found that they could possibly be an alternative choice to pie charts showing a A part of a Whole relationship. As the world of each rectangle is directly proportional to the numerical value it represents, they began for use to point relative proportions and differences between parts. The total rectangle area must represent the sum of all data. Treemaps only show composition at a moment in time (Static Composition).

Treemaps have two principal benefits against pie charts: 1) they will include ten or hundreds of Parts in a scheme of nested rectangles in a comparatively small space; 2) they code numerical values with areas, a greater visual attribute than arc lengths along the perimeter of the circle.

You should all the time indicate numerical values with proper annotations since the absence of a standard baseline seriously difficult the comparison between the rectangles that conform the parts.

More details might be present in my previous article.

Persistently, we now have to indicate Composition to our audience. This part to an entire evaluation just isn’t all the time easy to decode by our particular audience. Due to this fact, beforehand, we must analyze which methods we now have and what are their benefits and downsides related to our data and our message.

As previously indicated, six various kinds of charts might be used to indicate composition: Pie Charts; Stacked Bar Charts; Treemaps; Stacked Area Charts; Mekko Charts; Waterfall Charts. Here, we described three of them, particularly their characteristics, benefits, and a few precautions to be taken under consideration.

Stay tuned for the next article describing the remaining charts.

References

#1: https://serialmentor.com/dataviz/visualizing-proportions.html

#2 Ben Shneiderman (1992). “Tree visualization with tree-maps: 2-d space-filling approach”. ACM Transactions on Graphics. 11: 92–99. doi:10.1145/102377.115768.

#3 https://medium.com/towards-data-science/treemaps-why-and-how-cfb1e1c863e8

Should you find this text of interest, please read any of my 55 previous: https://medium.com/@dar.wtz