Hw5 is reading for grading!

[sepkamal]

Please see my homework 5 here: https://github.com/sepkamal/STAT545-hw-Kamal-Sepehr/blob/master/Hw05/HW05_factor_figure_management.md

Thanks

[menglinzhou]

Hi, @sepkamal, I'm glad to review your homework.

Factor Management

• You did a great job on dropping levels, because you not only use the function fct_drop(), but also compared it with 'filter()', and gave us a comment on how them works.
• The work of reordering shows that you really have a good understanding of this course. The diabetes research you mentioned about is really interesting.

File I/O

• You compared saveRDS()/readRDS() with read/write.csv(), and gave out a point of factor order which I never noticed before. Thank you!

Visualization design

• Fantastic job!! I really impressed with your plot of diabetes prevalence: beautiful color, awesome idea and nice code, you really show your ability and give me good instructions.

Clean up your repo

• You made a nice table of contents in top-level README and added links to individual pieces of work. It's very convenient to access to your work.

Overall, it's a check plus work!!

Menglin

[NSKrstic]

Peer Review:

Factor Management:

• Thorough investigation of what happens to factors after rows are filtered out (rows corresponding to the continent of Oceania). Subsequently performed dropping of factor levels after filtering, and verified that they were dropped.
• Successful reordering of factor levels based on "prevalence of diabetes" in different countries, by joining another dataset to gapminder. Original approach!

Writing/Reading of Data/Figures/Objects:

• Used both read.csv/write.csv and saveRDS/readRDS to examine what happens to the factor levels.
• The reason why you ended up with an "X" column with consecutive numerical values is that write.csv will also create a column of row numbers. I believe you can suppress reading that column using an argument in read.csv, but yes, it's kind of annoying. I recommend using read_csv/write_csv, but I've used a lot of read.csv/write.csv as well.
• Saved several figures to file using ggsave. Also demonstrated what happens if the plot object is not specified (saving of the most recent plot).

Visualization Design

• Well-designed figures, with good incorporation of colour.
• Covered what happens when using arrange to rearrange the order of the bars in you figure (there is no effect). Also noted that it was because of factor reordering that allows for rearranging bars in ggplot.

Additional Remarks:

• Repository is mainly clean, I would've liked to see you put your data and figures into separate folders for HW05.
• I'm glad you learned a lot from this assignment! I can tell that you did!

-Nikolas Krstic

[derekcho]

Hi @sepkamal, here are some comments about your hw05:

• The first part of the assignment was well done. You were able to successfully drop Oceania and its associated countries from gapminder. Alternatively, you can just use droplevels() to drop unused factor levels from both country and continent simultaneously
• Although your factor reordering looks correct from the code, you should include a table with the diabetes prevalence of each country so readers can verify the reorder worked
• File I/O was well done. Nice observations of differences between read/write_csv and read/saveRDS
• For visualization design, your plots look good and they incorporated several new elements
• Factor reordering and arrange for plots was done correctly, although it was a bit confusing because you did factor reordering first and it was done outside of this section. However, you were able to produce the correct plots and gave good explanations
• Well done overall! Your repo is clean and easy to navigate. Your report is organized and well presented

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