Fandango 2015 Movie Ratings: Legit or Bias?

This project utilized a great combination of statistical analysis along with various graphical illustrations. The conclusion is interesting but not relevant to the ‘here and now’. Find out why.
Any feedback is welcome.

Fandango Movie Ratings.ipynb (782.9 KB)

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


Hi Bruce,

Your project is really something incredible! :heart_eyes: Everything is just perfect, dataviz (my favorites are box plots), covering and closing images, project structure, capturing storytelling (it’s read like a book!), project structure, clean code and excellent code commenting (all the necessary moments, and not too much). I liked your geniune interest and critical approach to the data: you did far much more than it was required in the instructions and, indeed, got some rather curious insights.

It is possible to come up with the right conclusion for this analysis but for the wrong reason.

This one is really cool, as well as the explanation thereafter, and in general the whole chapter Some Cautions to Note. Also, your finding that in 2016 Rotten Tomatoes and Flixster were bought by Fandango was a great discovery for me and it explains a lot of things. Amazing job your digging so deep in the data!

This time I don’t have much to suggest to you, probably only these few ideas:

  • The code cells [1], [2], and [9]. It’s better to separate somehow the intermediate outputs adding subheadings for each output and also empty lines in between.
  • File names mentioned in markdowns: you might consider enclosing them in backticks for better emphasizing.
  • Probably, it would be a good idea to define a function for creating KDE plots and another one - for box plots, since the codes for each plot type are similar.

Congratulations on having done such a cool job and good luck with your future projects! :star_struck: