Lies, statistics, and Tableau, or my very short data visualization journey
My dive into the world of Tableau taught me two things: 1) this is a powerful tool for people who use statistics in their work, and 2) I have no idea what I’m doing with it.
My troubles began when I opened the sample workbook we were working with in class at the same time I had the NFL workbook open. At this point I’m not even sure how I managed to do this, since going back I don’t seem to be able to recreate what I did, but at the time I had open both workbooks and both data sets and it was just a complete mess. While I was figuring out what this disaster on my computer was, I wasn’t taking in the tips I needed to know in order to do things like change the color scheme. Now, looking back at it a couple weeks later, I still can’t figure out how to change the color scheme–Googling the answer tells me to do this on the color legend, but I can’t get the color legend to show up. Purple bar chart it is!
In Tableau’s defense, it’s clearly trying to help out the uninitiated novice at data visualization–this “Show me” tool, in addition to providing the best kind of chart for the data I’ve given it, also tells me what kinds of data I’d need if I wanted, say, a pie chart, and has helpfully color coded the different kinds of variables for me.
I think I’ve more or less manage the “click and drag” style of putting some measures in the “columns” box and some in the “rows” box, but how to actually put these charts together such that they would help me answer questions still seems like wizardry to me. I was so happy we were introduced to the public Tableau projects, because projects like this one gave me at least a glimpse of how I could conceivably use Tableau in research.
As they say at my old alma mater the University of Chicago, “That’s all very well and good in practice, but how does it work in theory?” I felt like I had a pretty good grasp on some of the issues involved in data visualization. Historically in my composition classes, I’ve had my students read a chapter out of Daniel J. Levitin’s Weaponized Lies: How to Think Critically in the Post-Truth Era, in which Levitin discusses the ways in which charted data can be used to mislead people. From the board room to the evening news to the floor of Congress, deceptive charts are everywhere, and we’re often inclined to believe them because data seem like “hard facts,” particularly when visualized. I still think I’m probably not bad at questioning charts or identifying misleading ones, but my humbling experience in Tableau has taught me that I’m a long way from being able to make them myself, or at least make them in an effective way. I might come back to this program after playing around more with AirTable. I’ve already used AirTable for another project, and I think it’s very plausible that at some point I will have academic data it might be useful to chart. That’s the time I’ll sit down with a cup of coffee and some YouTube tutorials and figure out just what the heck to do with Tableau.