Jessica Dixon

Chart Critique:

Here we have a bar graph from Pottermore, "global digital publisher of Harry Potter and J.K. Rowling’s Wizarding World," depicting instances of giggling by character(s) in the Harry Potter book series. This was actually shared on their Tumblr page and was originally published on pottermore.com as a larger graphic regarding the character Lavender Brown. Let's start with the positives.

I believe that Few would agree that using a bar chart was the best option. The overall goal here is to indicate to us which character (well, technically characters) giggles/giggle the most at a glance. This type of data corresponds to ranking relationships in business data and, according to the table provided in the reading, bar charts are a good choice for that when the scale includes a zero. Because of line length and 2-D positioning, we can take one glance at this and see that Lavender and Parvati do something the most.

That "do something the most" is where we begin to see the weaknesses of this chart. The people in charge of Visualizations at Pottermore appear to have fallen for the trap of arbitrary decision making that Few discusses. Rather than making only focused decisions to best represent the data, they went for flashy. I understand they want people to stop and look, but it gets a bit confusing. First of all, it is not clearly labelled what is being measured here. We have the term "Giggle-ometer" which does not exactly tell us all we need to know. An axis is labeled as "Instances of Giggling" but it, to me, almost appears as another bar in the chart particularly because each of the three bars is a different color. Few lets us know that this overuse of colors is a fairly common mistake. It potentially leads us to wonder why Lavender and Parvati's bar is gray. The graph would have been better served with a traditional x-axis which would not only tell us what is being represented but also give us the zero point which is implied but not outright given in this example. I also find it odd that despite not having a real x-axis but grid lines are included. We also have a couple of unneeded elements. Using the numbered x-axis would give us the tick marks to indicate the number of giggle instances allowing the numbers to be removed from the bars, leaving just the character labels. The more problematic extraneous element is the partially filled circle in each bar. I think what they are going for with that is the percentage of the total number of giggles that character or characters giggling comprises. But it doesn't work, as we now know, because our minds are not so great with size comparisons. At first glance, I actually thought it was indicative of smiling until I checked the numbers and percentages. That data probably would have worked better as a secondary graph.

Overall, the graph starts in the right direction by making correct decisions early on about the type of graph to use, but then it falls prey to its flashy, unnecessary elements.

Quiz:

I have neither given nor received aid while working on this assignment. I have completed the graded portion BEFORE looking at anyone else's work on this assignment. Signed, Jessica D. Dixon.