#SWDchallenge: explain the change
When you get right down to it, almost every time we graph data with a time component to it, we’re showing our audiences a visual representation of change. Are our annual sales increasing? Is our net promoter score trending in the right direction? Has our overall crop yield been declining faster year-over-year? Did our promotional campaign lead to a sales increase?
Visualizing change across time, then, seems like the simplest task possible. Pick two points in time; collect the values for those two points; and voila—you’re ready to visualize change.
This bar chart shows a change over time...
…as does this line graph (technically, a slopegraph, but you get it):
When you add more data points in, you can get a better sense of change over time, for sure. You can see trends, and relationships.
You can make analytic assessments based on these aggregated measures too, if you have enough data points.
The thing is, all of these visualizations are representing the aggregate values of many, many observations. While it’s easy to visualize a summary of change, it can be more of a challenge to explain the change...or, to put it another way, to be explicit about what the components of that change were.
If I tell you that my department had 20 employees at the end of Q2 2020, and 20 employees at the end of Q3 2020, you might think that it was an uneventful few months; this visualization that only shows headcount at the end of every quarter supports this idea.
In fact, you might think that the four-person drop in headcount from Q1 to Q2 was the big story here, and that we had stabilized our staffing numbers from Q2 to Q3.
In reality, however, those four people, who quit right at the end of Q2, were just the tip of the iceberg. In the first week of Q3, 15 more people from the department quit en masse. Only a heroic effort on the part of recruitment, senior management, and our other department heads managed to bring our headcount back up from 5 to 20—with a combination of new hires, transfers, and reorganization—in a mere 3 months.
That’s a much more interesting story, and potentially a critical one to tell, for the health of our organization. We would need to take another approach than just using a simple bar chart, then, in order to convey a fuller, and more accurate, picture of the situation to our audience.
This month, your challenge is to create a chart that tells the REAL story behind a change over time...something that allows the viewer to see the components of the change, and not just the summary numbers. (Or, maybe you prefer to take the opposite approach—visualize something where the aggregated data would make you believe that the situation is calm and uninteresting, while the underlying components of that change couldn't be more volatile.)
Share your creation in the SWD Community by November 30th at 4PM PST. If there is any specific feedback or input that you would find helpful, include that detail in your commentary.
A few chart types that might appeal to you this month include:
The waterfall - we have had a challenge dedicated to the waterfall, but that was several years ago, and you might feel that it's high time to revisit it. This chart type lends itself particularly well to showing the components of change.
The stacked bar or the stacked area - these can be extremely attention-grabbing chart types, but making sure they are informative and clear to your audience will be the tricky part.
The connected scatterplot - if your data’s journey from A to B takes a meandering path through the entire alphabet on the way to its destination, this might be your visual of choice.
But don’t feel obliged to limit yourself to just these...nor should you feel like “chart choice” is the only way to approach this challenge. Maybe you feel that animation or interaction are the key things to consider here! As is always the case for our challenges, you are only limited by your own creativity and imagination.
