Exploratory Deliverable - Summer 2021 Final Project

Due: Your presentation

Logistics

  1. As a team, please sign up for your team’s presentation slot here on July 27th, 2021. Your presentation and discussion with us should last no longer than 15 minutes. Each group should only have one sign-up, and please let us know if your group absolutely cannot make it to any of the available time slots.

  2. Prepare a Zoom presentation for Ellie and your project TA. Your presentation should contain the information presented in the final project roadmap. More specifically:

    • In the For each of your research questions, provide at least one piece of analysis section: Be sure to also justify, in your presentation, why you chose that piece of analysis – how does using this piece of analysis help with answering your posed research questions?
    • In the Provide an interpretation for each analysis part: Be sure to include all the statistics that come into play (e.g., what is the p-value? what is the prediction accuracy and how does this differ from a baseline model that does random guessing/most likely scenario guessing?)

  3. Feel free to reach out to your mentor TA anytime along the way if you need any support with your final project. You can even ask to go over your presentation with your mentor TA, if you'd like!

    Your mentor TA will reach out to you to give your group feedback on your data deliverable stage (sometime by July 21). Your group should get started on your exploratory deliverable as soon as you can without waiting for our feedback from the previous stage, as your work should have little to no dependence on the data deliverable feedback.

Rubric

Score range Reasoning
90-100 (1) For each of the research question, there is a defined goal (does not necessarily have to be well defined), (2) for each defined goal, there is at least one well-motivated analysis method - meaning the chosen method has to answer the research question, (3) for each analysis, there is a robust interpretation of the results that is precise and appropriately-hedged - coefficients and p-values should be interpreted correctly, and models should be compared against a random guessing model, and (4) there should be a good discussion of existing limitations and future directions - what the team plans the concrete hypothesis/prediction task will be.
80-89 Only three out of four criteria were fully met.
70-79 Only two out of four criteria were fully met.
60-69 Only one out of four criteria were fully met.
below 60 None of the above criteria were fully met.