Criticize and give feedback, please! -- Clean And Analyze Employee Exit Surveys

Hey y’all!

Here is my 6th guided project, please take a look! This one takes a while to finish, definitely takes way longer than expected. It was fun tho! And I learned a lot ;D

As an aspiring data scientist, I would like to use this project to help me gain more experience and knowledge in my data science journey. So I would really appreciate feedback and criticism, and I will definitely update this project based on that. Thanks!

Project last mission screen

Project link – github

update: the github link has a 7.2mb file, so it takes time to load. Couldn’t upload it here, because the upload limit is 4mb. Hence, I created a shorter version: This one has a reduced amount of graphs.

Let me know which one is better please, the original with more graphs? Or the new one (this one)? Because after looking at it, I kind of feel the original is too lengthy and some graphs are unnecessary. I’d really appreciate it if you guys let me know your preference. Anyway, here is the shorter file below.

6. [short version] Clean And Analyze Employee Exit Surveys --Dataquest guided Project.ipynb (3.8 MB)

[Shorter version] Project link – github

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


Hi @adrianzchmn please recheck your project it’s not opening in the Github, or you can kindly upload the notebook.


Hi @brayanopiyo18 the github link is working on my end, although it might take a while to load because the file is big.

As for uploading it here, since the original file is too big, I couldn’t upload the original. However, I’ve made some edits and uploaded the shorter version of the project (which I think is slightly better, because the original have too many unnecessary graphs). Please take a look. Let me know if this one works for you.

Hi @adrianzchmn,

Thank you for sharing your project. You’ve done an amazing job! I liked many things in your project: very profound and thorough analysis, perfectly structured, using bold and italic where necessary, very well-commented and clean code, pretty-printed tables, great visualizations, recaping intermediate results throughout the project. And those tables with the use of table_style() with graduated colors look really great!

Below are some suggestions from my side for your consideration. Hopefully, they will be useful.

  • Visualizations. It’s better to decrease the height of the bars in the plots [74], [76], and [78], then it will be easier to percept them. At the same time, for these plots [84], [88], [90], [94], and [99], just the opposite, it’s better to increase the bar height increasing also the figure size. Also, increasing the legend for these plots will be helpful, as well as for the plot [104]. As for the box plots ([27], [29], [31]), it’s better to add a title and remove the spines.
  • Probably it would be a good idea to put categorizing age in a separate section, just as you did for the institute service.
  • The code cells [32] and [33] could be removed, since after the last modifications nothing changed, and we can always refer to those previous results.
  • Throughout the project, you repeat too many times the questions made in the introduction, in their full form. While in general it’s a great idea to keep them in mind, you can refer to them just mentioning something like that: following our questions of interest that we made in the introduction, …
  • The code cell [21] - I would remove the commented-out code (well, for every piece of code there are alternatives, we always can choose).
  • After the code cell [47] it’s better to remove the intermediate results, since they are repeated after a couple of chapters, together with some new ones.
  • It can be a good idea to make the “Initial Analysis” chapter a section of the “Data Analysis” chapter, as they are very interconnected logically.
  • A small recap before the code cell [103] should be put after the section 11.6. Indeed, it recaps also that section.
  • In the beginning of the chapter 10, I noticed the word “screen”, inherited from the DQ mission screens. It’s better to substitute it with something else.
  • A best pactice is to use uniform style of quote marks throughout the project (or only single, or only double).
  • As for the conclusion, to make it more concise you can combine some categories. For example:
    -Tutor age 30s
    -Tutor age 40s
    -Tutor age 50s
    can be combined as “Tutor age 30s - 50s”. The same about some other categories.

All in all, great job your project! I wish you good luck with all your future projects, keep this high level!

Hi @Elena_Kosourova, thank you for the positive feedback, your kind words make me motivated to do more! And another thank you for taking the time to write clear and detailed and suggestions. This is awesome! I really appreciate it.

I will update this project based on all of your suggestions ;D
Will post it here once I’m done with it

1 Like

Hi @Elena_Kosourova here is the update. I followed all your suggestions, and it does look better now. Cheers! Thanks again!

github link
*I can’t upload the file here because it’s too big…

Anyway, if you can think of anything else for me to update, please let me know. I am going to keep making adjustments. Or else… I guess see you in the next project ;D

1 Like

Hi @adrianzchmn,

Your project is just perfect now! Especially the visualizations, they look very nice with the modified bar width. And your conclusion is more concise and more generalizing this time. Ah, and I like the cool picture that you inserted in the introduction! :grinning:

See you on the next project then. Happy learning!

1 Like