Guided-Project 7: Star Wars survey

Dear all

Please find below my 7th guided project analyzing the fivethirtyeight Star Wars survey .

I found Kwu’s comment on the “biased rating” due to default fillings very interesting and analyzed the difference between the biased and the corrected data.

I would be keen to get your feedback on my “Cross plot” visualizations Ratings vs View Percentages.
Thought hat they give in some cases better insights to the data, compared over the bar plots.

GuidedProj7_Star_Wars.ipynb (1.2 MB)

Click here to view the jupyter notebook file in a new tab

3 Likes

Hi @bender38! Thanks for sharing your project with the Community :slight_smile: I hope my feedback will still be valuable to you. Great that you’ve read that topic and addressed the problem of biased rankings. Your cross-plots look good and actually give a better overview of the differences between two groups.

Some suggestions from my side:

  • Make your section naming consistent, or in other words use correctly leveled sections, for example, the title is #, then major sections ##, and subsections of each section ###. Now, you have inconsistent leveling and it’s hard to follow
  • Check your code style and make it consistent. You can use JupyterLab Code Formatter together with your preferred code style, like black
  • Outline what questions you want to answer in the project and give more context to each section so that we understand better what you are trying to achieve
  • It’s better to remove plot grids and plots spines as they make plots difficult to read
  • You can also play with other colors for your plots to make them more attractive and natural. You can read this article, How to pick more beautiful colors for your data visualizations to start
  • Don’t forget to label all of your plots (for example, the plots [33] miss x labels. It’s also better to rotate “Male” and “Female” horizontally so that they are easier to read
  • Remove the commented code as it only distracts the readers
  • To be honest, I don’t know what plots [37] to [42] are for. They are very difficult to comprehend
  • Plots [43], [44], [46], [47] are difficult to read. The legends overlap with the plots, there are too many colors, the bars are too narrow, labels are vertical, and the plots are too small… You can read this article to make your plots more attractive with some easy techniques. If you want to learn more about data visualization, the book “Storytelling with Data” by Cole Knaflic is a very good start
  • Write the conclusions where you wrap up your findings

I hope my feedback was useful. Happy coding:)

1 Like