Final Project
Introduction
For the final project, you and your group mates (groups of 3-4 people) get to choose one of the following two options: 1) Technical Presentation or 2) Data Analysis.
Each group will be provided with a private repo that all members as well as course instructional staff will have access to. Final projects will be “submitted” by pushing the project requirements to the group repo by the deadline.
Presentations will be submitted on Canvas.
Written, visual, and presented content will be graded on their technical merits as well as the effectiveness of their communication.
Option 1: Technical Presentation
Groups who choose the technical presentation route will make slides for a presentation that effectively communicates/teaches an advanced statistical topic1 and/or an R package2.
Deliverable: R Markdown + Slides
Slides will be required for the presentation and they must be generated from either an RMarkdown document or a quarto document. Chapters 4, 7, and 8 of the R Markdown: The Definitive Guide discussion options for generating slides/presentations from R Markdown documents. (Presentations from quarto have similar documentation here.) Students should commit both the .Rmd (or .qmd) document and the rendered slides to their GitHub repo.
This presentation must teach the details of the R package, the statistical topic, or both at a level appropriate for students in this course. (i.e. You can assume your audience knows how to program in R, know about the tidyverse, know linear regression, etc.) And, you must demonstrate how to use the package and/or carry out the statistical analysis in R.
Deliverable: Presentation
Students must also present their slides in a presentation that is 10-15min long. This presentation will be pre-recorded and submitted on Canvas. For this option, all students must participate in the presentation.
Deliverable: General Communication
This will be a communication targeted to the people who you think should know about this package/statistical analysis. Here, you can assume your audience knows about R/data analysis in general, but you want to distill your presentation down to the most important aspect someone would want/need to know if they were going to use what you’ve chosen to present on.
Option 2: Data Analysis
Groups who choose the data analysis route will carry out a full data science project. This will include question formation, finding the data, doing background research, wrangling the data, doing EDA, analyzing the data, and answering your question of interest.
You can think of this as a mini case report in the fact that the process is the same, but we would not expect the data wrangling to be quite as extensive as what was done in the case studies. That said, we want to see demonstration of the skills you’ve learned in the class, so we will be looking for some data wrangling in your case study. If you have a single dataset that requires no wrangling, consider if additional datasets could be incorporated to answer your question(s) of interest more deeply.
You are strongly encouraged to think of your topic/question before looking for datasets. More interesting case studies start with the topic/question. Boring case studies look for the dataset first.
Deliverable: Report (.Rmd + HTML)
Your analysis will be submitted as an .Rmd document and rendered to HTML (both of which should be pushed to GitHub).
This will likely not be quite as long as a case study in this course, but will likely have the same sections.
Deliverable: Presentation
Students must present their case study in a presentation that is 3-5min long. What you use to visually support this presentation (slides, or something else) is up to you but should follow the effective communication aspects discussed in class. This presentation will pre-recorded and submitted on Canvas. For this option, at least one group member must present the project (in other words, not everyone has to “speak” but everyone in the group is responsible for the contents).
Deliverable: General Communication
This will be a communication targeted to the general public (non-technical, non-data scientists) conveying the most important finding(s) from your project.
Option 3: CS02 + Additional Data
Students can choose to carry out CS02 for their final project; however, students will have to find an additional dataset on a related topic (pollution, climate change, etc.) and incorporate that into the ir final report. See CS02 documentation for details on report and general communication deliverables.
Deliverable: Presentation
Students must also present their project in a presentation that is 3-5min long. This presentation will be pre-recorded and submitted on Canvas. For this option, at least one group member must present the project (in other words, not everyone has to “speak” but everyone in the group is responsible for the contents).
NOTE: Prior to 12/11 The above paragraph was incorrect. As a result, if a CS02 + additional data group submits a longer presentation 10-15min presentation, there will NOT be a deduction given the error I made.
Group Feedback
There will be a form to submit upon submission of the final project to provide feedback about working with your group mates. As with the case studies, this is meant to motivate not scare. Most groups work out really really well and everyone contributes to the best of their ability. However, if and when that doesn’t happen, I want to be sure I’m aware of the circumstances and follow up as necessary.
Footnotes
Advanced statistical topics are any topics beyond what would be covered in an intro stats course or any topics covered in depth in this course.↩︎
R Packages that can be chosen in this course are any R package that is not covered in detail in this course. (i.e.
tidyversepackages,tidymodelsandbroomwould not be options). If a package was used but only briefly mentioned in class, you can choose that. If you’re unsure, ask!↩︎