I had to fight a bit with Flourish, but I’m very happy with my end result. I chose a stacked bar chart and I changed the years column to be a text column so I can choose it as a row filter. This way, I created a the year selection option so the viewer doesn’t get overwhelmed by all the data all at once, as Lin suggested in her talk, and instead can view the popularity of names by year. I realized pretty early on that I would have to limit some of the data because either the names didn’t fit on the chart, or it was just too much information all at once to fit into the visual, which the system could not categorize all at once. Also because of this, I had to manually choose the colors of the rows as custom overrides for every single name. I chose to do boy names blue, and girl names pink, which felt a bit outdated and reductive, but it clearly communicates the point I think. I couldn’t make the system understand gender as another category and this was the easiest one to do manually so I’m happy with that decision. I also put axis titles so a viewer who doesn’t know what the data set is can also understand what is going on. But the chart was generated already with the two genders separated, so that was really convenient. But I added the counts on the end of the rows myself for clarity of numbers as it wasn’t very obvious where they exactly stood on the x axis. I think this graph style works with the data set because although it doesn’t let the viewer compare the years side by side, it gives them all of the information available on the data set in a clear and full manner. No data has been eliminated for clarity, but rather has been separated into digestible year options.
Your visualization is super impressive! It captures every dimension of the data and tells a rich story about how baby names changed over time. Also, the interactive nature of your graph is very engaging. Other than Lin’s warning against making the graph overwhelming, did Lin’s discussion of design principles make you change your mind at all during the process? If so, how exactly?
I love how you used a stacked bar chart to visualize the data. It’s really structured, and I can see how the trends changed over the years without getting overwhelmed by all the numbers. I also like how you made the year selection options rather than making the whole series of them. This is a new way of viewing the data! Nice work.