HW04 is ready for grading

[sepkamal]

Please see my homework 4 here: https://github.com/sepkamal/STAT545-hw-Kamal-Sepehr/blob/master/Hw04/hw04-tidaydata.md

[navysealtf9k]

Hi @sepkamal

I found your analysis to be very well organized. You provided good documentation of the steps you were taking which made it easy to follow your train of thought.

I really enjoyed the graph you plotted. I liked your choice of colour schemes. This made it very easy to spot trends. Good job on adding a trend line to make this clear. One thing I suggest is perhaps having the code for your ggplot calls in separate lines. I was told by one of the TA's that this makes it more readable and hence easier for them to grade.

Your code structure was good. You made extensive use of the dplyr and ggplot packages. I can easily tell you are quite comfortable in doing so. Nice Job loading Knitr and then calling on Kable to produce your table. I had some difficulty doing so myself and hope to learn from what you have done!

You did a good job going over the different join functions. it was a pretty novel idea to use data on Diabetes for the different countries. You made extensive use of the different commands which was interesting. In addition you also filtered your data to verify certain functionalities of the different join commands. This too was a pretty neat add on to your analysis

One suggestion I have is perhaps mention your reflections on the different tasks in the README file. That would make it easier for the reader of your work to understand your pain points for this activity.

Overall I think you did a good job and I look forward to having an opportunity to review your work again.

Regards

[derekcho]

Hi @sepkamal, here are some comments about your hw04:

• Your first task was well done! You were able to successfully use the spread function to obtain a table where you have the year on the y-axis and countries on the x-axis with the life expectancies of each country in the cells. In addition, you were able to produce the correct scatterplot that utilizes wide form data. My only suggestion would be to manually set the axis values for the x and y axes so that they are the same. In your scatterplot, it looks like Canada and Iran have the same life expectancies until you read the axes.
• For the second task, it might have been better to work with a smaller subset of your data. It is hard to tell what is removed with you make a join if you can’t see the entire dataset. In addition, you should try to show these tables with kable() to make it look nicer. Aside from those issues, you did a good job of explaining the effects of each type of join and provided appropriate examples as well. Nice work!
• Overall, your report was well organized, but try not to show super long tables when a smaller table will do. Your coding style was generally neat and efficient and your repo is clean and easy to navigate.

Your grade will be emailed to you at a later date.