Homework 1 Feedback

[Gchism94]

Dear Harshit -- Below is the feedback for your assignment. Please review carefully, and stop by office hours if you have any questions about the feedback.

---

Feedback

Exercise 1: [10 / 10 points]

• Plot: [6 / 6 points]
• Narrative: [4 / 4 points]
• Feedback: I don't think you needed a different colored title. A larger font size is adequate. I would also like to see insight into what the data tells you and why a density plot was good for this.
• [x] Resolved

Exercise 2: [20 / 20 points]

a. [8 / 8 points]

• Histogram: [3 / 3]
• Boxplot: [3 / 3]
• Narrative: [2 / 2]
• Feedback: The color scheme is odd, I suggest using black lines. I would also like to see axes labels, and perhaps a cleaner theme. You also did not give me any insight into what each plot tells you about the data.
  • [x] Resolved

b. [5 / 5 points]

• Boxplot: [3 / 3]
• Narrative: [2 / 2]
• Feedback: I would like to see axes labels, and perhaps a cleaner theme. You also did not give me any insight into what each plot tells you about the data.
  • [x] Resolved

c. [2 / 2 points]

• Feedback: You correctly removed the legend, but there is no insight into why you did this.
  • [x] Resolved

d. [5 / 5 points]

• Time series plot: [3 / 3]
• Narrative: [2 / 2]
• Feedback: I would like to see axes labels, and perhaps a cleaner theme. You also did not give me any insight into what each plot tells you about the data. The lines could also be a bit larger in geom_line()
  • [x] Resolved

Exercise 3: [20 / 20 points]

a. [2 / 2]

• Feedback: You did not answer the question.
  • [x] Resolved

b. [2 / 2]

• Feedback: You did not answer the question.
  • [x] Resolved

c. [16 / 16] (2 points each)

• Feedback: Feedback: Good work, but I suggest using grid.arrange() from the gridExtra package to arrange each set of plots the way I did in the original question.
  • [x] Resolved

Exercise 4: [10 / 10 points]

• Plot: [6 / 6 points]
• Narrative: [4 / 4 points]
• Feedback: The boxplot approach is okay, but it would look better with a wider rendered plot. To do this, one step is to have the legend positioned horizontally on the top or bottom. You also did not give me any insight into what each plot tells you about the data.
• [x] Resolved Additional feedback: Watch out for cut-off titles in facets in the future. I will knock points off for this.

Exercise 5: [20 / 20 points]

Feedback: The plot looks good, but you did not give me any insight into what each plot tells you about the data.

• [x] Resolved

Overall: [20 / 20 points]

• [5 / 5] - Doesn't require reproduction: .md file is in repo, figures show up in the .md file on GitHub, no egregious errors requiring reproducing the entire analysis to be able to follow it.

• [3 / 3] - Code style: Line breaks after |> and +, spaces around operators, spaces after commas, lines not too long, etc.

• [3 / 3] - Code smell: Messy / unnecessarily complex code, difficult to follow, unaddressed warnings, etc. Feedback: Most of your annotations should be plain text outside of code chunks. (Next time I will ding you for this)

• [3 / 3] - Quality and quantity of commits: No uninformative/nonsense text in commit messages, entire assignment not committed in a single commit. Feedback: You only have a few commits and the messages are not very descriptive. How will you be able to revert to a past commit without losing basically all of your work? (Next time I will ding you for this)

• [3 / 3] - Reasonably sized figures. Feedback: See my feedback from question 4. (Next time I will ding you for this - see cut-off facet titles)

• [3 / 3] - Font size, organization: No crazy large text for narrative, questions answered in order and identified, easy to follow.

Late penalty

• [x] On time: No penalty
• [ ] Late, but same day (before midnight): -10 points
• [ ] Late, but next day: -20 points

[Gchism94]

Your submission earns 9.2/10, there are some unresolved issues:

1. You did not answer these parts of question 3 again: a. What does the following code do? Does it work? Does it make sense? Why/why not?

ggplot(county) +
  geom_point(aes(x = median_edu, y = median_hh_income)) +
  geom_boxplot(aes(x = smoking_ban, y = pop2017))

b. Which of the following two plots makes it easier to compare poverty levels (poverty) across people from different median education levels (median_edu)? What does this say about when to place a faceting variable across rows or columns?

ggplot(county %>% filter(!is.na(median_edu))) + 
  geom_point(aes(x = homeownership, y = poverty)) + 
  facet_grid(median_edu ~ .)

2. You additionally did not give any narrative on what your plot tells you about the data for exercise 4.

For next time: Your code chunks are rendering really oddly. It seems that you're trying to do all feedback and text inside of code chunks. The purpose of notebooks are to separate code chunks and text. I will not penalize you this time, but next time I will grade this much more strictly.

[Gchism94]

You've addressed the issues well. Make sure to pay close attention to my additional feedback for HW 02. I will take points off more strictly for these issues next time. You now have full-credit for HW 01.