STA9750 matata668 MiniProject #01

[matata668]

Hi @michaelweylandt!

I've uploaded my work for MiniProject #01 - check it out!

https://matata668.github.io/STA9750-2024-FALL/mp01.html

[michaelweylandt]

Thanks @matata668 !

Peer Feedback:

• @Cdiallo77
• @nikitagtm
• @whaleron
• @SethKauf

It is now time for the peer feedback round for Mini-Project #01. Please review @matata668's submission for this mini-project and provide peer feedback.

Using the rubric at https://michael-weylandt.com/STA9750/miniprojects/mini01.html#rubric, please grade their submission out of a total of 50 points.

For each of the five categories, please give them a separate score and provide a total (sum) score across the entire assignment. Feel free to assign extra credit if you feel it is warranted (following the rubric).

If you give a score of less than 5 for any category, please provide a suggestion for improvement. (You can also give suggestions for any element they did well - more feedback is always great!)

As you go through this peer feedback exercise, think about what you particularly like about this submission and how you can incorporate that approach in your future work. If something is particularly insightful or creative, give some kudos!

Evaluators: This should take you around 15 minutes per peer feedback. You are not required to engage in substantial back-and-forth with @matata668, but you are of course welcome to initiate a discussion.

@matata668: please engage fully with your peers. They are here to help you!

Submission URL should be: https://matata668.github.io/STA9750-2024-FALL/mp01.html

Feel free to link to other repos, the course documentation, or other useful examples.

Thanks! @michaelweylandt

CC: @charles-ramirez

[whaleron]

Hey @matata668 ,

Great job on this! Here’s the breakdown: Written Communication: 8 Project Structure: 9 Formatting & Display: 9 Code Quality: 10 Data Preparation: 10 Total: 46 I loved the depth of your analysis, and your additional questions are interesting! Just a couple of minor suggestions to make it even better: For data display, maybe avoid showing the tables every time the code runs to keep things a bit cleaner. You could also add some headings or highlight the questions for better readability. Overall, fantastic work! 😊

[SethKauf]

Hi @matata668 ,

Good job on your project!

My breakdown for grading:

Written Communication: 8 The questions are all set up before each question, while the conclusions and analysis are answered in nice bold font. A couple times, the flow of the project was a bit off. I think limiting the DT table selection in the beginning and adding it again at some point throughout the project might have made it all pop better. The Task headers were inconsistent throughout most of the project.

Project Skeleton: 10 I ran the code and it all worked as intended and re-produced the analysis you provided.

Formatting & Display: 9 The only tables were the DT tables from the beginning and on some console printouts to the Quarto file. For the printouts, if a number was shown, it was usually just as an unbroken integer. Since you put the number in a better format in the analysis/written communication part as an actual formatted number, I don't think this was too bad, but it might be something to keep in mind going forward.

Code Quality: 7 The code functioned as intended which is good, but there were virtually no comments throughout the code about which block did what. There were a few areas with repetitive code ie when dealing with the 2019 to 2020 date ranges.

Data Preparation: 10 N/A

Overall: 44

I think there are some rough-around-the-edges spots to improve on, but overall really nice work. I really liked the analysis and felt like you were able to dig into the data and find the answer to some interesting questions. Mostly agree with the previous comment about data table displays and I would strongly encourage using comments to convey what you want your code to do, e.g: when you produced the following code block:

mta_nyc_covid_analysis <- USAGE_Drop |>
  filter(Agency == "MTA New York City Transit", 
         year(month) == 2020, 
         month(month) %in% c(2, 3)) |>
  group_by(month) |>
  summarize(total_passenger_trips = sum(Passenger_Trips, na.rm = TRUE)) |>
  arrange(month) |>
  mutate(percentage_change = (total_passenger_trips - lag(total_passenger_trips)) / lag(total_passenger_trips) * 100) |>
  filter(!is.na(percentage_change)) |>
  ungroup()
print(mta_nyc_covid_analysis)

I thought a comment at the top explaining both what you were calculating and how would have been perfect:

# Takes data from all MTA NYC Transit Modes for Feb-2020 and Mar-2020 and
# returns the monthly percentage difference using the dplyr lag function
mta_nyc_covid_analysis <- USAGE_Drop |>
  filter(Agency == "MTA New York City Transit", 
         year(month) == 2020, 
         month(month) %in% c(2, 3)) |>
  group_by(month) |>
  summarize(total_passenger_trips = sum(Passenger_Trips, na.rm = TRUE)) |>
  arrange(month) |>
  mutate(percentage_change = (total_passenger_trips - lag(total_passenger_trips)) / lag(total_passenger_trips) * 100) |>
  filter(!is.na(percentage_change)) |>
  ungroup()
print(mta_nyc_covid_analysis)

But again, overall very nice work!

[matata668]

Thanks @matata668 !

Peer Feedback:

• @Cdiallo77
• @nikitagtm
• @whaleron
• @SethKauf

It is now time for the peer feedback round for Mini-Project #1. Please review @matata668's submission for this mini-project and provide peer feedback.

Using the rubric at https://michael-weylandt.com/STA9750/miniprojects/mini01.html#rubric, please grade their submission out of a total of 50 points.

For each of the five categories, please give them a separate score and provide a total (sum) score across the entire assignment. Feel free to assign extra credit if you feel it is warranted (following the rubric).

If you give a score of less than 5 for any category, please provide a suggestion for improvement. (You can also give suggestions for any element they did well - more feedback is always great!)

As you go through this peer feedback exercise, think about what you particularly like about this submission and how you can incorporate that approach in your future work. If something is particularly insightful or creative, give some kudos!

Evaluators: This should take you around 15 minutes per peer feedback. You are not required to engage in substantial back-and-forth with @matata668, but you are of course welcome to initiate a discussion.

@matata668: please engage fully with your peers. They are here to help you!

Submission URL should be: https://matata668.github.io/STA9750-2024-FALL/mp01.html

Feel free to link to other repos, the course documentation, or other useful examples.

Thanks! @michaelweylandt

CC: @charles-ramirez

Hey @matata668 ,

Great job on this! Here’s the breakdown: Written Communication: 8 Project Structure: 9 Formatting & Display: 9 Code Quality: 10 Data Preparation: 10 Total: 46 I loved the depth of your analysis, and your additional questions are interesting! Just a couple of minor suggestions to make it even better: For data display, maybe avoid showing the tables every time the code runs to keep things a bit cleaner. You could also add some headings or highlight the questions for better readability. Overall, fantastic work! 😊

Hey@whaleron,

Thank you so much for the feedback. I am actually going to try and remove the tables and find ways to highlight the questions. Thank you for the idea

[matata668]

Hi @matata668 ,

Good job on your project!

My breakdown for grading:

Written Communication: 8 The questions are all set up before each question, while the conclusions and analysis are answered in nice bold font. A couple times, the flow of the project was a bit off. I think limiting the DT table selection in the beginning and adding it again at some point throughout the project might have made it all pop better. The Task headers were inconsistent throughout most of the project.

Project Skeleton: 10 I ran the code and it all worked as intended and re-produced the analysis you provided.

Formatting & Display: 9 The only tables were the DT tables from the beginning and on some console printouts to the Quarto file. For the printouts, if a number was shown, it was usually just as an unbroken integer. Since you put the number in a better format in the analysis/written communication part as an actual formatted number, I don't think this was too bad, but it might be something to keep in mind going forward.

Code Quality: 7 The code functioned as intended which is good, but there were virtually no comments throughout the code about which block did what. There were a few areas with repetitive code ie when dealing with the 2019 to 2020 date ranges.

Data Preparation: 10 N/A

Overall: 44

I think there are some rough-around-the-edges spots to improve on, but overall really nice work. I really liked the analysis and felt like you were able to dig into the data and find the answer to some interesting questions. Mostly agree with the previous comment about data table displays and I would strongly encourage using comments to convey what you want your code to do, e.g: when you produced the following code block:

mta_nyc_covid_analysis <- USAGE_Drop |>
  filter(Agency == "MTA New York City Transit", 
         year(month) == 2020, 
         month(month) %in% c(2, 3)) |>
  group_by(month) |>
  summarize(total_passenger_trips = sum(Passenger_Trips, na.rm = TRUE)) |>
  arrange(month) |>
  mutate(percentage_change = (total_passenger_trips - lag(total_passenger_trips)) / lag(total_passenger_trips) * 100) |>
  filter(!is.na(percentage_change)) |>
  ungroup()
print(mta_nyc_covid_analysis)

I thought a comment at the top explaining both what you were calculating and how would have been perfect:

# Takes data from all MTA NYC Transit Modes for Feb-2020 and Mar-2020 and
# returns the monthly percentage difference using the dplyr lag function
mta_nyc_covid_analysis <- USAGE_Drop |>
  filter(Agency == "MTA New York City Transit", 
         year(month) == 2020, 
         month(month) %in% c(2, 3)) |>
  group_by(month) |>
  summarize(total_passenger_trips = sum(Passenger_Trips, na.rm = TRUE)) |>
  arrange(month) |>
  mutate(percentage_change = (total_passenger_trips - lag(total_passenger_trips)) / lag(total_passenger_trips) * 100) |>
  filter(!is.na(percentage_change)) |>
  ungroup()
print(mta_nyc_covid_analysis)

But again, overall very nice work!

Hey @SethKauf ,

Thank you so much for your feed back. I actually agree with the inconsistency. As, I have tried multiple times to edit it and it looked fine but then when uploaded to github it would look a little wonky. I am going to work on the issues brought up. Thank you so much! Appreciate it :)

[nikitagtm]

Congratulations matata668 for completing your MINI#01.

Things I Like:

1. The report is well-written and flows naturally, with clear explanations of key steps and findings.
2. Code quality is excellent, with proper comments, clean structure, and no errors.
3. Tables are well-formatted and easy to interpret, contributing to a polished presentation.

Suggestions:

1. It would help to include better labeling for tables and figures to guide the reader through the analysis.
2. While the data preparation is solid, adding more innovative insights or utilizing additional data columns could deepen the analysis.
3. Instead of displaying the entire tibble output like this, It would be more efficient to directly show the result as a statement.

   A tibble: 1 × 2
   Agency                      Total_VRM
   <chr>                           <dbl>

Rating: Written Communication: 9/10 Project Skeleton: 10/10 Formatting & Display: 9/10 Code Quality: 10/10 Data Preparation: 10/10 Total: 48/50

[matata668]

Thank you so much for your feedbacks.

[Cdiallo77]

Hi @matata668

I enjoyed reading your work and it helped me understand a lot more this assignment and to find other ways of coding.

• Written communication: 9.75/10 - I like the intro and the details in your findings.You just used the word "Conclusion" a lot.
• Project Skeleton 10/10 - N/A
• Formatting: 10/10 - I Tasks are separated, and easy to read.
• Code quality 10/10 - very nice. I enjoyed reading it, it helped me too.
• Data prep: 10/10

Good job, keep up with the good work. Cdiallo77