Feedback about Visualizing Earnings Based On College Majors

I have done the guided project on Visualizing Earnings Based On College Majors and would like to receive your valuable feedback especially about my analysis.

Visualizing Earnings Based On College Majors.ipynb (923.7 KB)

1 Like

If I understand the problem correctly you are trying to answer these questions.

  • Do students in more popular majors make more money? Using scatter plots
  • How many majors are predominantly male? Predominantly female? Using histograms
  • Which category of majors have the most students? Using bar plots

It would make this a more fun read if you guided us better through your thought process with some text between the plots (and perhaps had fewer plots?) and explained your results better.

For example you say

Do students in more popular majors make more money? No, because by increasing the sample_size which shows more popular majors, median does not increase.

It is unclear to me how you reach this conclusion. Your thoughts about the data and the numbers are more valuable than the plots alone.

1 Like

Hi @jafarinasim, congrats on completing your project! I like that you explored the additional questions at the end and included box plots and the hex plots. I still have yet to do that on my project. :slight_smile: It looks like you know what you’re doing to create the plots.

I had some questions on parts of the analysis. For the predominantly male/female question, you said “According to ShareWomen there are 2 majors that are predominantly female which is 2/172 = 0.011 (1.1%) and there is one major with about 0% of Women which means predominantly male: 1/172 = 0.006 (0.6%).” I’m not sure how that conclusion was reached. In the histogram, the bars I boxed below all represent the number of majors that had 50% or more women. The tallest is showing nearly 30 majors with 70-80% women.

I also wasn’t sure about this conclusion when looking at the bar graphs of the top 10 and bottom 10 majors with regard to ShareWomen: “Comparing the percentages of women (ShareWomen) from the first ten rows and last ten rows shows that the last 10 rows have more ShareWoment than the first 10 rows and as the data is sorted by the earning it may show that the majors with higher ShareWoment receives higher salary.” Since the last 10 rows would represent the lower salaries, it would seem that the majors that are predominantly female receive lower salaries. The unemployment graphs also have a similar conclusion that I don’t think is supported in the graph and needs a second look.

If you update the project, please share it again so we can have a look!

1 Like

Thanks a lot for your feedback. I agree that I could have used fewer plots but since it was a guided project, I drew all the plots that were mentioned in the instruction.

About this question “Do students in more popular majors make more money?” I should say I did not figure out this question refers to which plot and which column shows more popular majors? I thought “Total” can be the best column to show popular majors but it was not mentioned in the project plots so I thought maybe the Sample_size also shows more popular majors. BTW I have updated my project and drew Median vs Total scatter plot as well but the result looks similar. The majors with a higher number of people do not earn more money as the scatter plot does not show a higher Median for a higher Total. Please let me know your idea about it.
Thanks a lot

1 Like

Thanks a lot for your feedback.
I may get the meaning of predominantly male or female wrong. I thought it means approximately all of them are male or female, so I looked only at this part of the plot:
About the analysis of the top and bottom 10 majors, I thought it is sorted in ascending order by salary, thanks for correcting me.
I have updated my project according to your comments and appreciate if you can review again.
Thanks a lot

1 Like

Thanks for sharing the update! Looks good.