After five days of long days at work, lots of head scratching, complaining, and ‘eureka’ moments, Dashboard Week has finally come to an end! I have to admit that my capacity to write long and interesting posts has been strongly diminished by the amount of brain power I’ve had to dedicate to finding, cleaning, and vizing data these past few days, so I’m going to keep this post quite short. I’m first going to go into the last theme of the week – historical climate data – and then go into my general feelings about what this week felt like and what future DSers should expect from it. Let’s go!
Climate Change: Does it really need any more proof?
After seeing how much we struggled Monday through Thursday to obtain data and transform it into something we could use in Tableau – a step we usually only go to around 5 or 6 pm – Andy promised us “no data prep on Friday”. And boy did he deliver! We were given access to a massive historical data set of temperature data dating back all the way to 1827! Of course, only a few countries actually collected information back there, but as we move into the 1900s, we start getting really interesting year-on-year information for pretty much every country. I must admit that we were all both relieved and excited to see we wouldn’t have to do anything in Alteryx that day. It was the little motivational push we needed to get through the final day of this truly tough week.
The only problem with having so little time to work on a dashboard is that if you really only have one shot at an idea. What I mean by that is that by the time you actually have a decently working dashboard and realize whether or not you are conveying an interesting idea or story, you really don’t have time to go back and change anything if you realize that, well, your viz isn’t all that great. This is how I felt on Friday. Even though climate data is one of my personal favorites, the idea I had didn’t turn out so great once I put my viz together. My objective was to separate the yearly data into different global economy cycles (Industrial Revolution, Machine Age, Information Age, etc.) to see how the change in the organization of economies and shift from manufacturing to service-based economies, or the late kick-off of industrialization in some countries, impacts the rate of increase of temperatures in these nations. In some countries, like the United States, the viz turned out quite cool. For the majority of them, however, the dearth of data, or the small incremental changes in temperature, didn’t really look all that great in a viz. Still, I believe I managed to show that global warming is a real and dangerous phenomenon that the world should really pay more attention to. Here’s what the evolution of the global average minimum and maximum temperatures look like (the black dotted line shows the overall average temperature).
Here’s what the evolution of the global average minimum and maximum temperatures look like (the black dotted line shows the overall average temperature).
Here’s a screenshot of what the evolution of the temperatures in the US looks like. As you can see, in this case, the viz works quite well, as we can clearly note the dramatic impact of the industrial revolution at the end of the 1800s, and how the temperatures stabilized more in the late 1900s as the economy shifted away from heavy manufacturing and more towards services that are generally “lighter” on the environment (I do mean “lighter”, not better, or energy-intensive).
All in all had I had more time, I would probably have opted for a different dashboard altogether had I noticed the upward temperature trends did not stick out as much with this graph. Neither one of my best or my worst, but one I will definitely revisit in the next few weeks! Thank you for the data again, Andy! (And EXASOL).
So How Does It Feel To Go Through Dashboard Week?
This question has one simple answer: amazing.
Even though I may have sounded a little flustered at times in my blog posts this week, Dashboard Week has been hands down the best experience I’ve had so far at the Data School. It’s amazing to see what my teammates and I were able to accomplish after receiving roughly two months of training from the School’s many coaches. Before starting this program, I had only vaguely read about APIs, had never opened Alteryx, had only produced a Tableau viz for my application, and was generally not a very skilled keyboard warrior. Now, I can instantly recognize HTML and JSON and identify which child I’ll need to parse my data, I can create complicated charts full of table calculations and LODs in a matter of hours if not minutes, or find raw data hidden in the root code of pretty much any website. I’m not telling you this to boast, not at all. I’m just trying to convince you to apply to the Data School. Seriously, I get paid to do nothing but sit around with like-minded people, learn, and build interesting data stories about wine and basketball. What more does one need?
Dashboard Week was a perfect opportunity to synthesize all the knowledge we had accumulated over the past two months, to identify which skills we still need to perfect and which topics we still aren’t as familiar with, and to produce some high-quality dashboards. It was tough and it was long, but at the end of the day, you know that with the help of the people around you that you are always going to find a solution and get the data you need to build the dashboard you were tasked with building. For any future DSers reading this, don’t be scared: we all make it. And we all come out far more skilled and ready to meet any objectives our clients’ will have for us one day.
That’s all I’ve got, thanks for reading, and stay tuned for more blog posts coming soon!