I’ve seen super cool moving graphs on the internet before but never knew how people created them. This week, I discovered the powers of Flourish, a wonderful tool that facilitates the process of data visualization. As someone who’s only ever used RStudio and ggplot2, this was a whole new world for me. Using data from the New Zealand Department of Internal Affairs, I was able to produce this neat graph.
Lines & bar charts & plots, oh my!
I chose to visualize this data with an animated line chart also known as a “bump” chart. Despite the CSV file being pretty “tidy” to begin with, I had to transform some of the rows into columns to truly achieve the clarity I wanted with this dataset. What’s great about this visualization is that the bright colors are not only aesthetically pleasing but they also quickly convey the changes in the rank of each name over time! When one line crosses another, it’s pretty easy to see which name is more popular in a given year.
The interactive aspect also allows the viewer to easily switch between two different y-axes: one displaying the number of births registered to the name in a given year and another displaying the rank. The viewer can also hover over or click individual lines to isolate certain names as well as click along different points of the x-axis to go forward and backward in time.
The power of Digital Arts and Humanities
Of course, this visualization is not without its downfalls. In an effort to make the graph appear less cluttered, I decided to only display the top 3 names for females and males. As a result, I had to omit a lot of data from the original dataset. Additionally, I couldn’t figure out how to make the y-axis label change with the different views of the chart. Therefore, when viewed out of context, it might be a little vague as to what the y-axis represents at first.
As guest lecturer Lin mentioned on Thursday, digital humanists have the power to use existing data to highlight compelling stories. However, with great power comes great responsibility. It is in our hands to communicate data in a way that is not only engaging but also accurate and representative. Therein lies the debilitating dilemma: how can we craft simple and effective visualizations without leaving out parts of the story?
6 thoughts on “Visualizing Data with Flourish”
Leave a Reply
This site uses Akismet to reduce spam. Learn how your comment data is processed.
I really like the way you chose to display the data! I didn’t know that was called a bump chart, but I think it does a really good job showing the data across time, which was an interesting aspect. This was something I was interesting in creating, but was only able to get a simple, non-animated line graph. Your explanation of rearranging the data makes a lot more sense to me, as well as choosing to omit some data for clarity purposes. Similarly, I think you raise a really interesting question at the end. The data you present is clear, but you did have to omit some data which might not give the whole picture.
This is a really neat way to present the data! I ended up just using a regular bar chart on Rawgraphs.io but I wish I’d seen this when doing the assignment. For the question you pose at the end, I think it’s important to factor in that using an aesthetic representation for data should not take priority over telling the story. I would rather use a more simple display using most, if not all the data that people can understand rather than a more alluring display in which I would have to eliminate potentially important data points. However ideally there should be a balance and effort should be put towards each end of data visualization. Great work!
I legit smiled and thought “WWHHOOAA” as the lines flowed through the graph! This is a very neat way of displaying the data! I think the animation is particularly effective because it allows the viewer to comprehend all of the trends over the years, and allows the viewer to see tie breakers if two lines appear to overlap.
I love how you utilized the animation feature of flourish! I also ran into the same thing where I had to make some rows into columns to present the popularity of names over time. I was struggling with how to present all of the names without it looking cluttered- just presenting the top 3 names for each is a good way to solve that problem. Very cool!
I had some trouble along the way with rawgraphs so it’s cool to see someone who used Flourish have more success. I think the question you raise at the end is very thoughtful and makes the reader take a beat to think. On my end, I’m not sure we’re really able to unless it’s a small, controlled data set, which is why I think your adaptation + visualization of the data we were given works so well. Your chosen names and chart style create a few trends that I picked up on, and unless the reader’s goal is to learn absolutely everything there is to know about New Zealand’s early 2000s top baby names, this chart with 6 names works perfectly well at communicating the message.
I love the way you chose to display your data! The line chart race (or whatever it’s called) is a really cool way to visualize the most popular names over time. I also must commend you for your ability to make this feel more like a blog post than an academic assignment. I think that it is difficult for some to insert their own personality and quirks into their writing but you did it really well! Great job!