Teams | Deliverables | Templates | Grades | Ideas | Examples

Project

The final project is a chance to show what you have learned in class and to create an example of both your coding and communications skills that you can show to employers and admission committees. The final deliverable is made up of two parts:

  1. An executive summary of your analysis. This is a professional report to be turned in as a PDF. This report clearly lays out the question(s) you are addressing, discusses the analysis you performed in the notebook, and draws a conclusion. This report should be no more than three pages long (not including figures and tables). This is not a technical document, but you should discuss any technical difficulties and any assumption you made in your analysis.

  2. Code and data. You will submit a Jupyter notebook and any other files needed to make the code work. This notebook should explain what you are doing (and have many useful comments) but does not contain your questions/conclusions. Think of this as the document you would give a coworker who would like to replicate your results and understand what you have done to get to those results.

Teams

The project can be completed in teams of up to three students. Students working in teams should not divide up the work so that one person does all the coding, one person does all the writing, etc. Each team member should contribute to all parts of the project.

Each team should choose a corresponding author. This person will submit documents through canvas. The other team members do not need to submit anything.

Deliverables

Proposal

The proposal is one-page document (PDF) and the accompanying Jupyter notebook and data files. Upload them as two files:

  1. The proposal document as a single PDF
  2. The code and data in a single zip file

The proposal document should include

  1. the questions your team plans to answer
  2. the data you will use to answer them
  3. one graph or table that demonstrates a pattern in your team’s data
  4. (be sure to list your team members and note which member is the corresponding author)

The Jupyter notebook and data files should load the data and produce the graph or table that is in the proposal document.

Note that your team does not need to have answers for your questions yet. The goal of the proposal is to come up with a set of questions and the appropriate data. The table or graph is to let me know that you all have figured out how to get the data into pandas and that your team has looked through the data.

Your questions and expected results will likely change a bit as you study the data more and learn new things. That is how research works. The proposal is to ensure that we agree that the project is off on the right foot.

Final project files

You must upload two files to canvas. Upload them as two files. Do not compress your project report.

  1. The project report as a PDF. Do not forget to include the names of your group members in the report.
  2. The code and data in a zip file
    • The code and data should be in one folder that you zip to create a single zip file.
    • Before you zip your code, make sure that there are no absolute file paths. If your code is in a folder on your computer named 'project' and the data file you want to read in is in a folder named 'data', your code should not have paths such as C:\Users\kimru\Dropbox\project\data\file.csv. All the paths should be relative to the folder your notebook is in, such as data\file.csv.

Templates

The proposal templates are MS Word documents. Students interested in academic research might want to try preparing the documents in LaTeX, which is the standard in technical writing. Overleaf is an easy-to-use online LaTeX editor.

Word templates

Proposal template // Project template

Latex templates

Project template

Grades

Projects will be graded on both the quality and readability of the code and the project report. The grading rubric is here. Some of the things I will be looking for include:

  1. Project idea: Is the question clearly stated? Are the data appropriate? Is the approach appropriate?

  2. Code: Does it run? Is it commented? Is it readable? Are complicated parts of the code explained?

  3. Project report: Is the report clearly written? Are the graphs labeled? Do they display graphical excellence?

Generating ideas

Many coding or statistics classes that require projects assign project ideas to students. Not here! Part of the data mentality that we are developing is being able to ask good questions. Like any skill, this takes some practice.

  • Read. Blogs, newspapers, magazines...almost anything can spark a question. Reading publications that are not afraid to show some data are even better: The Wall Street Journal; New York Times; Economist; FiveThirtyEight...

  • Think about other classes you have taken. What questions came up in labor class? Industrial organization? Sociology?

  • What interests you outside of economics? Sports, theater, food? What questions and data are out there?

  • Keep an idea journal. Mine is a text file on my computer. Anytime an idea strikes — no matter how crazy it might sound at the time — I write it down. Periodically I go over my journal to see what questions I might be able to make some progress on.

  • The course resources page is the beginning of a list of data sources and writing about data. Send me more ideas!

Examples

I write "Data Briefs" for the Center for Research on the Wisconsin Economy. The Briefs written by me are similar to the format of your projects. The Briefs written by others might be a bit longer than I require for this project but you might find them interesting.

Below are some projects from previous iterations of this course at UW – Madison. These projects represent a mix of topics and techniques, but you should not feel constrained by what you see here. Be creative: Explore datasets, read papers, and find something that interests you!

[This is a work in progress. I am waiting to hear from more students for permission to post their projects. I hope to post several more projects.]

  • Post-graduation earnings, Shuanger Chen, Fall 2018: "I study what factors affect average post-school earnings of a college or university in the US, and how they affect the after-school earnings." Report // Code

  • Crime in Chicago, Zhe Wang, Fall 2018: "My target of this report is to demonstrate the spread of the crime ratio in Chicago in 2016 and clarify the frequency of crime behaviors in different periods of time, to inform tourists when and where to go." Report // Code

  • Tax Cuts and Jobs Act of 2017, Sheridan Camarata, Spring 2020: "The objective of this research is to analyze the changes [to] existing federal tax revenue and aggregate financial data trends both pre- and post-TCJA." Report // Code

  • Economic Health during Pandemics, Charlie Mrkvicka, Catherine Peterson, and Emily Wurst, Spring 2020: "...[W]e examine and compare various economic health indicators, such as the unemployment rate and inflation rate, to better understand how they have been influenced during COVID-19 and other flu-like pandemics in American history: the Spanish Flu, H2N2, H3N2, and H1N1." Report // Code

  • Coronavirus Death Factors, Wally Estenson, and Michael Marotta, Spring 2020: "We wanted to determine what variables lead to higher deaths in counties across the [United States]." Report // Code

  • A Real Estate Valuation Model, Yu Shi Chua, Spring 2020: "The aim of this study is to find out what variables, if any, can significantly affect home value." Report // Code

  • Uneven effects of a reduction in import competition, Siying Li, Fall 2022: "I study the uneven effects of the reduction in import competition following the trade war. Which states seem to experience a larger reduction in import competition?" Report // Code