Finally, I am able to share my latest guided project : Visualizing Earnings Based On College Majors
It took me so long to finish this. When I first saw this, it was only 6 slides, and I’m like It will be done very soon! But it wasn’t the case. It was a bit difficult for me to understand the concepts, especially reading the plots.
So if anyone of you have some time to go through my project, please let me know if my plot reading skills are any good. I’m not very sure if I had interpreted the plots correctly or not. So if you can please guide me on this, I would be very grateful.
Looking forward to your feedback.
Last mission screen of the Guided Project
P4_Guided Project_Visualizing Earnings Based On College Majors.ipynb (787.0 KB)
Click here to view the jupyter notebook file in a new tab
Hello @jithins123, thanks for sharing your work. Also thanks for always reviewing peers project in the community, I usually find your reviews very informative.
You’ve done a very good work in your project, everything starting from the title is nice and informative. The presentation style too is easy to understand and follow along.
Nice descriptions and visualizations @jithins123! Good Work!
This is really a great project from goal to conclusions you’ve jotted everything and explained very clear. One thing I personally noticed after reading your project end to end is, it is easy to understand for Non tech people as well which is really outstanding!
Awesome work & Cheers!
Thanks a lot Victor @info.victoromondi for your appreciation. Glad to know that you have liked my reviews. And thanks a lot for taking time to review my guided project.
Thank you Ryan @masterryan.prof thanks for going through my project.
Thanks a lot @prasadkalyan05 for going through the project end to end. Glad you liked the presentation.
Would be nice to know if you guys have found any point where I could have done better.
Your project looks perfect, as all your previous ones: well-structured, very profound analysis, clean code, curious insights and conclusions. I totally agree with your reasoning that sample size for many majors here is rather low and this can significantly affect the results. Also I liked your observation that the majors ranking is practically based on median salaries in the descenidng order. When I was doing the same project, I thought that there was a more complex reasoning behind, like combination of median, unemployment rate, total number etc. Now I see that it was more straightforward
It was a great idea to analyze full-time and part-time employment calculated in %, instead of using the numbers themselves. In this case you immediately distinguished some clear patterns in the data. I hadn’t arrived there before, now I know that it was helpful to do so. Also I liked that you added titles to the plots: since those created automatically in pandas include only column names that are not always self-explanatory in our dataset, it’s better to write them directly.
Here are few comments from my side what can be improved.
- The markdown cell after the code cell . It isn’t the highest unemployment rate, but the most frequent one.
- For bar plots and box plots sections it’s better to add separate subheadings.
- For box plots it would be good to mention also their quantiles and min-max values.
- The bar plot in the code cell . I would add more observations about mostly female and mostly male majors, and those least popular for both genders.
- About popularity vs. median. I agree, from our data it seems to be no correlation between them. Rather I would say that popularity influences the dispersion of earnings: less popular majors show larger dispersion.
Well, nothing else to add. Good job, great project as usual. See you on the next projects then.
Thanks a lot @Elena_Kosourova for this line by line review! You have gone really deep into this project and thanks a lot for pointing out a few things for the improvement. I have updated my notebook with your suggestions. See you soon on the next project. I’m sure you will be able to finish that one faster.