suminwei2772 / STAT545-547-hw-Wei-Lisa

0 stars 0 forks source link

Homework 03 is ready for grading #3

Open suminwei2772 opened 7 years ago

suminwei2772 commented 7 years ago
  1. README.md for hw03

  2. gapminder_dplyr_ggplot2_exploration.md

hannahdxz commented 6 years ago

Hi Lisa! Thank you for inviting me on your GitHub. Here's my peer review for Homework 03. Hopefully you find these comments helpful!

Your README file in your HW repository was easy to navigate with the links for each homework session. The README file for HW03 was also very easy to navigate with links. The HW03 README was very informative, and concise.

Great exploration of the Gapminder data. I liked how you organized your analysis with section headers, which made your code very clear and easy to follow. You demonstrated a good knowledge of coding syntax. Your use of dplyr() and ggplot2() is quite diverse, and you used kable() functions to display your data. For each table, there is a relevant figure for visualization which was very helpful and informative to help understand the table. You explained both the code and the plots thoroughly so it is easy to understand how you got those plots and what does each plot means.

You tackled 5 tasks, which exceeded the requirements in number of tasks! I think you did a really great job exploring the data in the section Task 1: Compute trimmed mean (and untrimmed mean) of life expectancy, You had two trimmed means and you plot them on the same plot with the untrimmed mean, that ggplot was very impressive and informative!

One suggestion might be in the Task 3: Look at spread of gdp per capita for each continent, you plotted the distribution of the continents with the greatest and lowest spreads. I would suggest to use box plot in this section, side by side box plot would give a very clear comparison between these two continents.

Overall, thank you for your great work and I hope you enjoy this class!

estennw commented 6 years ago

Hi Lisa,

This is a great piece of work, and you've made some beautiful plots. You exceed the number of tasks required, while delivering high performance on each task nevertheless.

Code Your code is really nice, and well structured. I want to advice you to add comments to your code, however it's so easy to read that it's really not necessary to add any comments. You use dplyr really efficient, and you don't use any more code than needed. Perfect!

Tables It's always nice to see some well crafted kable tables. The tables look nice, and your use of input variables like align and col.names gives the tables a finished look. Tips:

gapminder %>% 
  filter(lifeExp < 59.47444)  %>% 
    group_by(continent,year) %>% 
      summarize(distinct_countries=n_distinct(country)) %>% 
        spread(key=year, value=distinct_countries) %>%
          knitr::kable(format="markdown",align=c(rep('c',times=7)), padding=2) 

which results in this

continent 1952 1957 1962 1967 1972 1977 1982 1987 1992 1997 2002 2007
Africa 52 52 51 50 50 49 43 39 38 39 41 40
Americas 19 15 13 10 8 7 5 2 1 1 1 NA
Asia 29 26 25 23 19 14 11 8 7 6 3 1
Europe 5 3 1 1 1 NA NA NA NA NA NA NA

Of course, it's not very pretty with the NA's but it's worth considering. This can also be done with the other tall table.

Figures Amazing, really nice looking, and so sparse that each figure is easily interpretable. I was especially impressed by the trimmed means plot. That plot tells you a lot about whether outliers are located above or below the median.

Impressive work, it was a nice read!

pgonzaleze commented 6 years ago

Hi @suminwei2772 , here are some comments about your work: At least three tasks: Yes
A table and figure for each task: Yes
Comments on tables/figures: Yes Reflections on Process: Yes Bonus (side-by-side layout, new table packages): Yes (side-by-side)

Your marks will be distributed later, Regards,

Pedro G.