Project Review: US Medical Project
Review: I found this project easy and I am curious to know about my code review
Time Estimate: The project took me approximately 5 hours to complete.
Code Repository: You can find my code repository here: US Medical Project Repository
You’re just doing EDA for this project; you haven’t made any predictions nor have you conducted hypothesis testing or anything of that nature (so, maybe remove those references to that language in the notebook). As such, the notebook should have a brief intro (which you have), show the results from your analysis/output of the code (summary statistics), and a conclusion that wraps up your findings and where you can list shortcomings or possible next steps (as you learn more python).
If you look at the range of the charges column, you would see that there are a few outliers that pull the mean. So, perhaps mean isn’t the best stat and median might be more appropriate here.
It would be helpful to see the code output from the functions for the descriptive stats. You’re telling a story about what’s in the data first, then you could do some significance testing for potential correlations between variables (ex: two tailed t- test for a binary categorical variable (women v. men) and a quantitative variable (charges), stuff like that.) at a later point.
Good work. Just maybe review exploratory data analysis to fix a few things.