General information

Homework

The homework assignments are almost entirely empirical, with each assignment focusing on a specific identification strategy, research question, and health-related dataset. I will provide all answers in R; however, you are free to use any software you’d like. Just keep in mind that I will be best able to answer your questions if you also use R. I can also answer most any question if you choose to use Stata. I can point you in the right direction with Python, but I likely can’t provide you with any sample code or specific syntax advice.

Please submit all of your homework answers as a link to a GitHub repository through Canvas. If you keep the repository private, please be sure to share the repository with me using my ian.mccarthy@emory.edu email address. I strongly encourage you to use RMarkdown, which will help you to easily create PDFs and is much better for writing data-oriented documents. If you are not familiar with R Markdown, take a look at the following resources:

• R Markdown: The Definitive Guide
• Using R Markdown for Class Reports

If you’re new to R Markdown, then there will be some growing pains. Please be patient with yourself and stick to it. One of the goals of the course is to develop good workflow habits. This means doing work in a way that minimizes mistakes and maximizes reproducibility. R Markdown exists exactly for these reasons. It allows you to keep your data analysis and the description of that analysis together in a single document or group of documents. It’s really easy to introduce mistakes when copying and pasting empirical results into a text document, and it’s even easier to waste hours (days/weeks?) re-pasting results into your text document and re-writing the results accordingly. Doing everything in R Markdown avoids all of these issues. In the long run, it’s much more efficient.

Note that, for your submission, I do not need to see all of your code. You can include snippets of code files to highlight the most important parts of your work, but please don’t include all of your code or large printouts of data in the final PDF submission.

Each homework assignment is worth 30 points toward your final grade (12% each) and consists of 10 individual questions. Each question is worth 3 points, graded as follows:

• 1 point for an attempted answer as of the initial due date
• 1 point for an attempted answer as of the revision date
• 1 point for a correct final answer as of the final due date

Due to the staged structure of each assignment, your GitHub repository should include three answer keys. To make it easy to navigate everyone’s repositories, we need everyone to adopt the same directory structure and naming convention. Please be sure to name your solution documents as follows: lastname-firstinitial-hwk(num)-(submission num). For example, if I’m finalizing the initial submission for homework 1, then the file should be named mccarthy-i-hwk1-1.pdf. A complete directory structure is provided below. Please adopt this structure as it will also help you to keep your analysis and data organized. Finally, please note that you should not include the “data” folder in git or github. The data files are too big and should never be tracked in that way. Instead, please be sure to add the data folder to your gitignore file.

homework1
|   README.md
|   mccarthy-i-hwk1-1.pdf
|   mccarthy-i-hwk1-2.pdf
|   mccarthy-i-hwk1-3.pdf
|
|---data (not tracked in git or pushed to github)
|   |   input (symbolic links)
|   |   output (analytic data sets)
|
|---data-code
|   |   _build-data.R
|   |   1_data1.R
|   |   2_data2.R
|
|---results
|   |   table1.tex
|   |   figure1.png
|   |   hwk1_workspace.R
|

Project

We do not have a final exam in this class. Instead, we have a final project that asks you to submit a 5-page policy brief. Please see the Final Project for more details.

Participation

Participation is graded based on your engagement in small-group homework discussions. At the end of each assignment review day (the final due date for each assignment), each member of each small group will be asked to evaluate their peers on a scale of 1-3, with 3 indicated high level of engagement and effort with the group. Please see the Participation for more details on the peer evaluation process.

Your median score across peers in your group will be your participation grade in that assignment. At the end of the semester, your peer grades will be totaled. Your final participation grade is then based on your total number of points out of 10. Note that there are 15 possible points, so there is some margin for error.