Project 2

Presenting Your Data (Quickly!)

Project 2 is an in-class presentation which will occur during class on Tentatively Tu-Th, March 17-19th, in class (Tuesday and Thursday both). Each group will make and present a short data analysis using one of the data sets you listed in Section 2 of Project 1 (or a new one, I’m not keeping track). Your presentation will introduce and detail a new ggplot geom_ or aesthetic mapping that we haven’t yet covered in class. You will be limited to a total of six slides (plus a cover slide), presented in no more than eight minutes, including questions/discussion.

Some specifics

Explore your data

You will present exactly two visuals based on a dataset of your group’s choice. Your task is to find one interesting relationship in the data that you wish to share. The relationship should be the sort that inspires further investigation of the data, which you will do in one of your slide. For instance, your Rats data from Project 1 might have shown that summer months had higher numbers of rat sightings than did winter months, but this was only true in two of the five boroughs (I’m making this up, it may or may not be true in the data, and you are NOT to use the Rats data from Project 1). Then, for further exploration, you might look at weekend vs weekdays, as an example.

You will present your analysis in the form of two visuals made with ggplot. No memo, no text. You get two visuals to communicate it.

Communicate your data with a new geom_ or aesthetic

Each group will be assigned a ggplot geom_ or aesthetic. The list is at the bottom of this page. Your plots must show both your interesting relationship in the data, and one must do so with this assigned geometry or aesthetic. This may mean adjusting your analysis to be compatible with the geometry or aesthetic.

Put it all together

Your presentation will have six slides.

  1. A cover slide with your group members names on it.

  2. A slide briefly introducing your dataset, showing a snippet of the relevant data.

    • In your presentation, you will highlight the relevant variables and the overarching topic you are exploring. For example “We are using the US Census ACS 5-year estimates to examine the relationship between income and commuting time. In this data, each observation is a census tract, and we have data on median tract income and average commuting time as well as marital status.” (where your slide shows a few rows of the data with the relevant values, plus any other variables you think are helpful.)
  3. A slide showing, through visualization, a summary of the “interesting relationship” in the data you’re exploring. For example, if you’re interested in unemployment during COVID, you’d first show a visualization of how unemployment changed 2019-2020-2021 across counties or states or something.

    • In your presentation, you will discuss the “interesting relationship” you found.
    • Either this visualization or your previous visualization must use your assigned geometry or aesthetic
  4. A slide visualizing some feature of the “interesting relationship” in the data – drill down by some other variable and explore the relationship further from the first.

    • In your presentation, you will discuss the detailed feature of the “interesting relationship” you found, and why you think it is happening.
    • Either this visualization or your previous visualization must use your assigned geometry or aesthetic
  5. A slide showing the code from your visual that used your assigned geometry, function, or aesthetic. The point is that others can learn about the aesthetic or geometry and see a use case of how to use it

  • In your presentation, walk through the code necessary to make the plot. Focus on the new geometry/aesthetic.
  1. Next steps on your data analysis and exploration.
  • As a class, we’ll discuss some ideas and ask questions here.

Submit it on D2L

Submit a single powerpoint file on D2L that contains your cover slide + your 5 content slides by the night before presentation (Tentatively Monday, March 16th, 11:59PM). I will assemble everyone’s submissions into one big powerpoint that we can move through without delay. You can submit as a Powerpoint or as a PDF (Powerpoint is preferred as I will load them all into one giant powerpoint).

This will be short

There are five content slides and three of you in the group. Each person can take 1-2 slides. Do not plan on speaking for more than 1 minute or so per person, we have a lot of class to get through. Just as a good visual boils the message down to it’s most pure form, so too should your presentation.

Grading

Out of 40 points:

  • Data analysis (10 points): Did you find an interesting relationship? Does it give important context or lead one to want to explore the data further?

  • Visual (10 points): Is the visual clean, well-formatted, and free from errors. Does it adhere to our standards for good visualization? Does it clearly communicate it’s message with only basic discussion and description?

  • Application of new geometry/aesthetic/method (10 points): Did you successfully show your assigned geometry or aesthetic? Was the code clear enough for others to create something similar with minimal work?

  • Presentation (10 points): Was your group organized and concise? Did you clearly cover all the discussion necessary in the allotted time?

Remember, this is a 8-minute presentation that we’ll run back-to-back-to-back over all the groups, so efficiency is key. The plot has to hit fast and hard!

Assigned Geometries / aesthetics / methods

If you have any questions, please email me jkirk@msu.edu.

Many examples (from which I created this list) can be found here. You can (and should!) use outside resources to get an understanding of what your assigned geometry/aesthetic/method looks like. For some of these geometries/aesthetics/methods, I’ve added a link to a package or example that should help.

Not every dataset will work with your geometry. You may have to use one of your other datasets to find data that will work. If your group is assigned a method, just use that method in whatever visualization best illustrates it. Email me if you have questions.

Group Name Last Name geometry, aesthetic, or method name Additional Info
Broom Campbell shape
Broom Gibson shape
Broom Melnyk shape
Cli Casas Ponce cut_number https://ggplot2.tidyverse.org/reference/cut_interval.html
Cli Scherkenbach cut_number https://ggplot2.tidyverse.org/reference/cut_interval.html
Cli Staniak cut_number https://ggplot2.tidyverse.org/reference/cut_interval.html
Dplyr Luebs Stacked area chart with geom_area See https://r-statistics.co/Top50-Ggplot2-Visualizations-MasterList-R-Code.html
Dplyr Surian Stacked area chart with geom_area See https://r-statistics.co/Top50-Ggplot2-Visualizations-MasterList-R-Code.html
Dplyr Weiermiller Stacked area chart with geom_area See https://r-statistics.co/Top50-Ggplot2-Visualizations-MasterList-R-Code.html
Dtplyr Bjork ggMarginal
Dtplyr Warfield ggMarginal
Dtplyr Wilson ggMarginal
Function Fusion Capodivacca alpha Find an application where alpha is useful
Function Fusion Mitchell alpha Find an application where alpha is useful
Function Fusion Spring alpha Find an application where alpha is useful
Httr Deacy cut_width https://ggplot2.tidyverse.org/reference/cut_interval.html
Httr Iulianelli cut_width https://ggplot2.tidyverse.org/reference/cut_interval.html
Httr Riley cut_width https://ggplot2.tidyverse.org/reference/cut_interval.html
Keynes to Success Broyles Area chart with geom_area See https://r-statistics.co/Top50-Ggplot2-Visualizations-MasterList-R-Code.html
Keynes to Success Danner Area chart with geom_area See https://r-statistics.co/Top50-Ggplot2-Visualizations-MasterList-R-Code.html
Keynes to Success Spadoni Area chart with geom_area See https://r-statistics.co/Top50-Ggplot2-Visualizations-MasterList-R-Code.html
R Squad Lute geom_rain https://github.com/njudd/ggrain
R Squad Tehzib geom_rain https://github.com/njudd/ggrain
The Spy Kids Kleinerman geom_bin2d
The Spy Kids Webb geom_bin2d
The Spy Kids Ziedins geom_bin2d
Three’s Company Abdalla shape
Three’s Company Mohamed shape
Three’s Company Zhou shape
Trifecta Dennis treemapify/ggplotify See https://r-statistics.co/Top50-Ggplot2-Visualizations-MasterList-R-Code.html
Trifecta Evans treemapify/ggplotify See https://r-statistics.co/Top50-Ggplot2-Visualizations-MasterList-R-Code.html
Trifecta Nulsen treemapify/ggplotify See https://r-statistics.co/Top50-Ggplot2-Visualizations-MasterList-R-Code.html
conflicted Benchimol geom_xsidedensity (or ysidedensity) https://github.com/jtlandis/ggside
conflicted Karaka geom_xsidedensity (or ysidedensity) https://github.com/jtlandis/ggside
conflicted Perttunen geom_xsidedensity (or ysidedensity) https://github.com/jtlandis/ggside
forcats Calley cut_width https://ggplot2.tidyverse.org/reference/cut_interval.html
forcats Miller cut_width https://ggplot2.tidyverse.org/reference/cut_interval.html
forcats Wolson cut_width https://ggplot2.tidyverse.org/reference/cut_interval.html
googledrive Kitchen shape
googledrive Raczkowski shape
googledrive Varshney shape
googlesheets4 Al Zaid geom_errorbar Hint: do this with regression results
googlesheets4 Gupta geom_errorbar Hint: do this with regression results
googlesheets4 Weber geom_errorbar Hint: do this with regression results
haven Allen geom_errorbar Hint: do this with regression results
haven Halstead geom_errorbar Hint: do this with regression results
haven Warren geom_errorbar Hint: do this with regression results
jsonlite Berenfeld transition_time https://www.datanovia.com/en/blog/gganimate-how-to-create-plots-with-beautiful-animation-in-r/
jsonlite Curtin transition_time https://www.datanovia.com/en/blog/gganimate-how-to-create-plots-with-beautiful-animation-in-r/
jsonlite Paszkiewicz transition_time https://www.datanovia.com/en/blog/gganimate-how-to-create-plots-with-beautiful-animation-in-r/
lubridate Pandit geom_tufteboxplot See https://r-statistics.co/Top50-Ggplot2-Visualizations-MasterList-R-Code.html
lubridate Valine geom_tufteboxplot See https://r-statistics.co/Top50-Ggplot2-Visualizations-MasterList-R-Code.html
lubridate Yoder geom_tufteboxplot See https://r-statistics.co/Top50-Ggplot2-Visualizations-MasterList-R-Code.html
modelr Dannenberg geom_count
modelr Jain geom_count
modelr Shah geom_count
pillar Dominguez shape
pillar Koczara shape
pillar Wesley shape
purrr Hao ggMarginal
purrr McGuirl ggMarginal
purrr Shenoy ggMarginal
readr McCarthy geom_violin
readr Paterson geom_violin
readr Serna geom_violin
readxl Cheerla geom_xsidedensity (or ysidedensity) https://github.com/jtlandis/ggside
readxl Holden geom_xsidedensity (or ysidedensity) https://github.com/jtlandis/ggside
readxl Tran geom_xsidedensity (or ysidedensity) https://github.com/jtlandis/ggside
reprex Labrador geom_dumbbell See https://r-statistics.co/Top50-Ggplot2-Visualizations-MasterList-R-Code.html
reprex McElroy geom_dumbbell See https://r-statistics.co/Top50-Ggplot2-Visualizations-MasterList-R-Code.html