Creation of ISS Spacewalks - Makeover Monday

-
Earlier this year, I created one of my first Makeover Mondays on Spacewalks. This post walks through my data preparation steps and considerations I made while creating the visualization.
Makeover Monday ISS Spacewalks

Preparing the data set

1. Examine the data set for completeness 
I sorted each column to find missing data. The Spacewalkers column had some "n/a" fields so I filled in these values manually.

2. Add extra information to data set
I wanted to add the nationalities of the Spacewalkers and countries of the spacesuits to the data set. 
For the nationalities, I used a vlookup to append the data based on the name. However, I did have to be careful about different name spellings used in the data set compared to Wikipedia. 

As for the spacesuits used in a spacewalk - I entered this manually based on the footnote symbols in Wikipedia. American spacesuits and Russian spacesuits were used depending on where the mission took place. 

3. Final check
I recreated the original visualization as a final check. Through this process, I discovered that some dates were mixed up in the data set. For example, Expedition 18 was listed as Dec 10, 2009 but it should be March 10, 2009 according to NASA. 

If you'd like to explore the cleaned data set, I've made it available on Google Sheets. The spacewalk detail and cleaned tabs are most relevant.

Creating the visualization

Design
My design was inspired by this visualization. I liked that they combined a summary bar chart but also displayed each spacewalk individually. However, I thought presenting the data in a circular format was hard to read - although it looks very pretty and fits the space theme!
Thus, I kept the stacked bar chart as a summary and added more information in the scatter chart below.

Colour choices 
The red and blue combination in the original visualization isn't the worst color combination and they both stand out well against black. However, they're very bright colours and it can be confusing because both Russian and American flags contain both colours. 
After many iterations, I eventually chose to use white and beige as they represent the colours of the American and Russian spacesuits. I also had to change the background colour to gray so that they could still be seen. 

Data displayed

I chose to exclude 2019 because the year wasn't finished yet and the data would be incomplete. Also, by doing that, I can have a year label at both the starting and ending year in the visualization (as I chose to only label every other year).

I used R to concatenate both of the spacewalker's names into a single label. Within my tooltip, I used the spacewalkers' names and mission. I didn't end up using the nationality of the spacewalker in my visualization as Data Studio currently has a limit of three dimensions for a scatter plot.

------

Overall, I'm happy with how my visualization turned out. I also appreciated the data preparation exercise I was able to get with this data set.

Comments

Post a Comment

Popular posts from this blog

Product Ideas: Getting more specific with locations

-
As a user, there are often times when you come up with product improvements based on the pain points you experience. I'll share two of my recent ideas centered around the theme of location.  Product : LinkedIn, Indeed, and other job search sites What : Add detailed job location (neighborhood or exact location) to job postings rather than simply a city/area. For remote jobs, allow for more detailed searches - is this remote worldwide or only remote within a certain country? Why : From a personal perspective, I live in a large city and the job location within the city matters a lot to me. A job might be listed in the same city that I live in, but it could take over an hour to commute to the location.  In general, for job seekers, this feature would allow them to better understand what the commute could look like. Employers would also benefit from a more relevant talent pool. Ie. job seekers can more readily self-select and employers will end up with applicants who are aware of t...
Read more

Tip: Create a radar chart in DS using built-in area charts

-
Image
For this week's Makeover Monday , I made a 4 sided radar chart in Data Studio (DS) with the built in area chart. Note: This method works well for 4 sided radar charts but may not work with more variables due to limitations within Data Studio 1. Data Preparation Date for X values An area chart in DS requires a date for the X axis and a metric for the Y axis. Since I wanted to put Spring/Fall on the Y Axis and Winter/Summer on the X axis, I had to convert the Spring/Fall percentages to dates. I did this by choosing Jan 1 2019 (an arbitrary date) to be "0" and added/subtracted by the percentage point to get the final date values. Use a top and bottom area chart My area chart is actually composed of two area charts - one for the top and bottom. The reason for this is because DS doesn't handle negative Y values well and the background colour gets affected. You'll get something like Figure 1 if you use a negative Y value and it would be a hassle to fix the ba...
Read more