Final Project Guidelines

Overview

The goal of this project is to give you an opportunity to integrate the course material into your own academic, professional or personal interests. This can mean many different things. Here is a non-exhaustive list of potential forms the project may taken

Apply the course material to a research project you are already working on.
Read a paper which utilizes tools from the course and if possible reproduce the results (see Papers at the bottom).
Use the method's from the course to challenge a claim in the news or on social media.
Analyze some data you've found online and discuss the implications (see Data Sources at the bottom).

You will turn in a project proposal and a final write-up at the end of the semester. Given the very open ended nature of this project, it's difficult to come up with a concrete set of guidelines. If you feel you can develop a project which helps develop your understanding of the material but does not fit these guidelines, please contact me early in the semester.

Project proposal (30%) - due 10/16

The project proposal is a short (1) page note you will turn in roughly half way through the semester. These we be reviewed by myself and your peers.

(10pts) Is there a clear question you want to answer? The proposal should be as specific as possible. Try to avoid vague statements such as ``I want to better understand method X". Instead, set a concrete goal such as "I want to understand why method X and method Y give different conclusions on this example" or "I want to understand what happens if you control for some variable in this paper".
(10pts) Is the relevance of the techniques in this course clear? It's okay if you want to build on things we've done, for example you might be interested in time-series which can be re-framed as regression problems, but you shouldn't be working on something totally unrelated.
(5pts) Is the proposal clearly written and self-contained? Don't use jargon someone in the class might not know. Try to write in a way that gets someone excited about your project!
(5pts) Is it clear how the project will align with your own goals? It can apply to your personal, academic or professional interests. Be creative!

Final write up (60%)

The final project will include a write up (I except 3-5 pages) will be turned in at the end of the semester. This should include an introduction (largely taken from the proposal) explaining the background. A brief Methods section outlining the analysis you've preformed. A Results section explaining what you have or haven't learned. A discussion section where you describe what went right, what went wrong, what you might do differently, future direction etc.

Grading points:

(50pts) Does the write up include the following:
- introduction (5pts) - explains the question you are asking and why it is interesting to you
- methods (20pts) - explains how you are approach the problem, or, if you are summarizing the results of an existing paper/textbook chapYer, what the author's approach and how you are interpreting it using what we've learned in class. You should use the material we've covered in class correctly!
- results (20pts) - what have you concluded? Has your analysis revealed something new about the relationship between variables in the world? Have you found an error in a published paper or preprint?
- and discussion (5pts) - what are some future directions? have you answer the question you set out to answer? why or why not?
(10pts) Is the project clearly written? I should be able to reproduce your results with very little effort. This means documenting code and including all relevant data sets. This is all good practice for submitting scientific papers or writing reports. You should also include a bibliography. I don't care what format as long as I can find the parameter you cite.

You don't need to successfully solve the problem you set out to solve. In fact, I anticipate some of you wont. Research is hard and most things we try are false starts. What is important is that you learn something from the experience. Maybe you realize the question you set out to answer isn't actually the right question to ask, or maybe an analysis you thought was fairly straightforwards turns out to be vastly more complicated. Then explain why!

Python notebook (10%) - OPTIONAL

You should also include a colab notebook which can reproduce all of the simulations and data analysis. If you decide not to include a coding portion to your final project, the write up will count for an additional 20% spread out equally over each rubric item. If you do analyze any data or run any simulations, you should include a python notebook though and I expect most project will involve this.

(5pts) Does the notebook run? If there is any data that needs to be loaded to get the notebook running, give clear on instructions on how to obtain it.
(5pts) Is it well documented? Anyone else in the class should be able to understand what each line does (you don't need to document functions we have already used repeatedly in class).

Papers

Preprint servers, such as psyarxiv, medrxiv and biorxiv are a great place to find new papers. it's nice if the data is included in the paper, but if not you can still explore the methods they use by generating simulated data. Here are few papers that might make good projects:

Data Sources