The term paper in this course is intended to take you through the process of constructing your own empirical project. Your paper will use data to answer some interesting economic question. The grade you receive on your paper will be a function of the rigor and clarity of your analysis. In order to get a good grade:
Q: When is deadline for having my topic approved?
A: Friday, November 4.
Q: When is the first draft due?
A: Friday, November 25.
Q: When is the final paper due?
A: Monday, December 5, at 10:30 am.
Q: What if I miss a deadline?
A: Getting your topic approved on time will be treated as an assignment, and
graded on a pass/fail basis. Submitting a complete first draft on time will be
treated as an assignment, and graded on a pass/fail
basis. Your final paper loses one letter grade for each day it is late. If it is
only a few hours late, you will lose a portion of a letter grade.
Q: What should my topic be?
A: I take a very broad view of what is an interesting economic question: contemporary economists investigate issues like crime, education, families, and health. Look for a question that is interesting to you, so that you will be motivated to work hard. Remember that I
do this for a living, so anything that is interesting to you is likely to be interesting to me.
In order to have a topic approved, you must convince me that it is feasible. This means that:
Q: How can I get started on picking a topic?
A: The best way to get a topic is to look for an existing research article that
investigates a subject you are interested in, and uses data and methods that are within
your ability and time frame. Then you can get (some of) the data used in that
article, replicate (some of) the results, and add your own extensions. For 90% of students,
I strongly recommmend using an article whose analysis is based on state-level data. The reason
for this recommendation is that most of the work in empirical research is not in running the
regressions, it's in getting and organizing the data. Individual-level data is often very
complex in structure, with thousands of observations. State-level data comes in a manageable
size, and can be linked across multiple sources, but is large enough to be used with more
advanced methods like instrumental variables or fixed effects.
Q: What do “replicate” and “extend” mean?
A: “Replicate” means that you try to obtain the same data the original authors
used, and perform the same analysis of that data to see if you get the same results.
This is harder than it sounds, as authors often leave out important details or even make
programming mistakes. Usually you will fall short of a full replication.
“Extend” means that you add your own spin on the analysis.
For example, you could try alternative specifications of the model, or you could see if the
results are sensitive to outliers.
Q: I have an idea for a paper that doesn't fit your recommendations. Can I still get it approved?
A: Maybe. Before I approve any topic, I need to be convinced that it is feasible
and will satisfy my requirements. If you follow my recommendations, approval will be easy: find a paper, and find the data, and you should be fine. If you decide to go your own way, you'll need to give me a more fully-developed plan before I give approval.
Q: Can I use my paper for this class as the paper for another class?
A: Yes, if you notify me and the other instructor approves. However, your paper for
this course will be judged in terms of the expectations of this course. If you give me a brilliant
game theory paper with no data, I will give you an F.
Q: How long should the paper be?
A: There is no minimum or maximum length. A typical paper will be 10-15 pages.
Q: What should the paper look like?
A: I will provide a template file in MS Word format.
Q: How do I turn in the paper?
A: Please submit your paper using WebCT. Submit the
paper (including all tables and figures) in a single Word or PDF file.
You must also submit a ZIP or RAR compressed file containing all
data and computer code used to generate the results in your paper.
Q: What are the penalties for academic dishonesty?
A: Anything from a warning to expulsion, depending on the circumstances.
Q: How do I avoid those penalties?
A: Section 4.1.2 of the University's Code of Academic Integrity and Student Conduct(at http://www.sfu.ca/policies/gazette/student/s10-01.html) clearly identifies and prohibits several forms of academic dishonesty, including plagiarism and submitting purchased essays as your own work. The Code applies to this term paper and has the last word on what is and is not acceptable. A few class-specific guidelines: