I recently left a position where I was responsible for reporting frequently on the state of renewable energy projects in operation. There is a lot of data to parse in technical fields and it is easy for important information to be lost amid a barrage of terms and numbers. One of the most important skills I brought to the job was the ability to distill hundreds of data points and variables into one image I could use as a central point of reference without the need for further slides.
Before proceeding, I have to give a nod to the guru of displaying quantitative data, Edward Tufte. In fact, my interest in being able to put multiple data points into a concise image began when I attended a workshop of his in 2003. In his book, “The Visual Display of Quantitative Information,” Tufte says, “Graphical displays should:
show the data
induce the viewer to think about the substance rather than about methodology, graphic design, the technology of graphic production, or something else
avoid distorting with the data have to say
present many numbers in a small space
make large data sets coherent
encourage the eye to compare different pieces of data
reveal the data at several levels of detail, from a broad overview to the fine structure
serve a reasonably clear purpose: description, expiration, tabulation, or decoration
be closely integrated with the statistical and verbal descriptions of the data set.”
That is a lot to try to accomplish in one graphic, and it can seem difficult to do. Fortunately, graphing tools in programs such as Microsoft Excel and others are evolving as our need for visualizing data grows.
The easiest lever to pull is of course color. It has long been a favorite to show percentages, magnitude of the data point, changes over time, and different types of data. (think solar output versus wind, hydropower, etc.) A bit more complex is using multiple vertical axis. In the example below, in addition to using various colors in the monthly columns, I measured one variable on the left y-axis and linked it to the magnitude of the monthly columns and used the right y-axis to correspond with the red and blue lines. (The graph is a cross section of a larger graph and I have removed all labels for proprietary reasons.) These options allowed me to convey at least six data series of a single project at once. (I contemplated filling the gap between the two horizontal lines to emphasize the magnitude between them, but it would have obstructed the data conveyed in the changing colors of the columns.)
Sometimes data can be represented even if there is no formal way to label it. Last I checked, Excel does not account for ordinal directions, but it can be assumed. The figure below, Napoleon’s March from Paris to Moscow and the subsequent retreat, is one of the most popular examples of excellent graphical representation of data. The width of the brown line (the advance) goes from west to east. The black line (the retreat) goes from east to west. The width of the line in each direction indicates the number of troops. Along the route there are instances of troops splitting away and tracking their movement with the cardinal directions based on a common understanding of North, East, South, and West orientation.
Lastly, I am becoming a big fan of spark lines, a visual tool gaining in popularity. Using columns (a.k.a. sparks, or whiskers) you can represent a complex data set. Below is a spark line showing an American football team’s season. Sparks above the x-axis represent victories and those below represent losses. This simple graph can be enhanced with the magnitude of the spark indicating the magnitude of the win or loss, its color can signal if the game was a divisional, conference, or non-conference game, and an outline can represent whether or not the game was played at home. You can also use a pattern within the column to indicate something like whether or not the quarterback through any touchdown passes. All of these elements can be combined into one graphic summarizing a variety of data types and points.
[New England Patriots Winn-Loss record for the 2018 season.]
These are just some of the options at a person’s disposal to create visuals conveying large amounts of data. Workplaces are becoming more data centric, and as we turn to metrics to capture market segment, understand trends, and maybe even come up with some predictive moves, the ability to take in as much - if not more - with the eyes as the ears is becoming more and more essential.
Put more simply, a picture is worth a thousand words.
-•-