Employee Exit Surveys Guided Project

Share
,
#1

Hi everyone,

I’m currently going back and redoing the guided projects including the next steps and would love your feedback on my (hopefully) improved work.

https://app.dataquest.io/jupyter/notebooks/notebook/Redone.ipynb#

Guided Project Clean And Analyze Employee Exit Surveys.ipynb (418.3 KB)

Thanks in advance,
Ivelina

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

10 Likes
#2

I like your project so much, I have a lot to learn, thank you for sharing.

3 Likes
#3

Thanks for the kind words, @jemartinezm1!

Good luck on your learning journey!
Ivelina

2 Likes
#4

good job I liked the way you present your plots. you can check mine too here

2 Likes
#5

Hi Ivelina,

Very good and interesting project, great job! I liked a lot your idea to investigate also other factors (position, employment status) and especially their combinations, like gender+age. Below are some of my suggestions, hope they will be useful.

  • It’s better to re-launch all the cells when the project is already finished, for the cells to be in order and starting from 1.
  • The code cell [226] - probably it should be explained somewhere in a markdown cell why we decided to drop exactly those columns.
  • You can add the titles to your plots to improve the visual perception.
  • Some of the code cells (for example, [262], [269], [271], [277], [278], [279], [282], [283], [286]) can be written in one line.
  • I was curious about the last line of your conclusion and checked it. But there’re only 3 of them! :grinning: These 100% values are always a little bit suspicious in terms of possible undersampling.

Happy coding with your future projects!

3 Likes
#6

Hi Elena,

Thanks for the detailed and constructive feedback! Very valid points, not just for this project but for any project in general! :slight_smile:

Happy coding to you too!

Ivelina

2 Likes
#7

Hi Ivelina,

It is my first time that i’m giving feedback,
Overall is very good presentation, but i have to agree with Elena in some code cells there are missing markdown explanations, very good work!

2 Likes
#8

Hi Ivelina,

Your notebook was very helpful to me. Thank you. I tried to summarize it below.

  • Import required modules (pandas, numpy, matplotlib.pyplot)
  • Read the data sets
  • Review the data sets (info, head)
  • Check missing values (“Not Stated” => NaN)
  • Drop irrelevant columns
  • Rename columns (same names for mutual columns)
  • Drop irrelevant rows (keep only employees resigned)
  • Create “institute_service” column for DETE data (it keeps data of years worked)
  • Find employees resigned bacause of dissatisfaction
  • Combine the data sets
  • Categorize data in “institute_service” (years => categories)
  • Analyze the data
    • Dissatisfied by service category
    • Dissatisfied by age
    • Dissatisfied by institute
    • Dissatisfied by gender
    • Dissatisfied by employment status
    • Dissatisfied by positions
    • Dissatisfied by age and gender
  • Conclude

By the way, this is my first comment in DQ. I hope there is nothing wrong with the comment. :blush:

2 Likes
#9

Very good summary, @cenkbursali! :slight_smile:

Happy learning,
Ivelina