Employee exit survey cleaning and analysis

Hello everyone! I would like to share with you a guided project from “Data cleaning and analysis” mission. Any feedback would be appreciated.

I have several questions regarding this project:

  1. I always check my pd series after cleaning and transforming them. Is it ok to put these checks into the final version of the notebook (like I did this time)? Is it ok to write “Check” as a comment before them or is it too obvious comment?

  2. I want to learn dashboards since I heard it is a very useful skill in the industry. Do you think dataset from this project is ok to try for my first dashboard? Or it needs to be something with more time series data and sophisticated graphs rather than just a series of bar plots?

Thank you in advance for your answers and feedback!

https://app.dataquest.io/m/348/guided-project%3A-clean-and-analyze-employee-exit-surveys/11/next-steps

Employee exit survey.ipynb (560.5 KB)

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

I get the 503 error on nbviewer. I’m not sure if it’s just my problem. Could you confirm @alekseikondratenko7?

It works for me @artur.sannikov96. I also got dimilar problems yesterday from Safari. However, it works well from Chrome.

Now it works! Thanks for sharing your project:)

What I liked:

  • Good overall organization
  • Clearly defined objections
  • Summary of conclusions at the beginning
  • Code commenting

What can be improved:

  • You have some typos
  • In the function stage() that classifies workers in 4 experience groups the I would give the argument a more descriptive name like “age”
  • At least in the “Data Analysis” section I would briefly reintroduce what questions you are trying to answer to
  • Remove the legend from the “Dissatisfaction by working experience” plot and label the X axis. It should be as clear as possible
  • The same goes for the “Dissatisfaction by working experience and gender” plot: missing X label
  • In the “Data Analysis” section you still do some cleaning (like Age). Maybe it’s better to move it to the appropriate section?
  • The “Dissatisfaction by age” plot: legend and X label problems
  • You may also expand the conclusions section if you’d like to not repeat the conclusion summary

Regarding your questions:

  1. Yes, it’s okay:)
  2. Dashboards are very different and may be used with whatever data set including this one. Just make sure that the dashboard makes sense.

Happy coding and good luck with your future projects:)

Thank you for your comments, @artur.sannikov96! Yes, I agree with you that the cleaning of the “age” column would better fit in the “Data cleaning” section for the consistency purposes.

1 Like