Empirical Research Paper Project

“How can I know what I think until I see what I say?” — E. M. Forster on Writing

“No economist has achieved scientific success as a result of a statistically significant coefficient. Massed observations, clever common sense, elegant theorems, new policies, sagacious economic reasoning, historical perspective, relevant accounting, these have all led to scientific success. Statistical significance has not.” — McCloskey and Ziliak, “The Cult of Statistical Significance” (1996: 112)

Overview

Your task is to write a research paper on any topic in political economy of your choosing so long as it has an empirical component. The purpose of this paper is threefold: (1) to demonstrate to me that you have mastered the material of this course, (2) to develop your writing, communication, and data analysis skills, and (3) to help you develop, strengthen, and/or modify your own views by grappling with them in writing. As a reminder, this paper constitutes 25% of your final course grade.

I will spend a significant portion of one class discussing more about the writing process, and provide a detailed guide to help you choose a topic, craft arguments, and write a good paper.

I am NOT looking for a survey of existing research, a series of block quotes showing what some economist said about X, a regurgitation of my lectures, a list of pros and cons with a last minute conclusion, a book review, and so on. I also do NOT want you to compress everything about econometrics you learned from this class into a single paper. You should only use those insights that are relevant to your topic and your argument (it may only be one or two things!).

I am looking for a paper that attempts to answer a specific research question of interest to economists by using data analysis. You should be able to summarize your paper in one or two sentences – the specific research question your paper addresses, your method for answering it, and your results. It should be a reasonably original paper (it is difficult with limited knowledge, time, and data to offer something truly original), and it should be your own take on the topic. I expect you to, at the very least, have one multivariate regression used to reach your conclusions. I do not expect or require you to find statistically significant results.

Please note that, while you are not required to, I highly recommend discussing your ideas with me over email or in person. I will also read any drafts you would like to submit to me early, and provide you with helpful comments, subject to my own availability. I will stop accepting drafts to provide comments by November 30.

Length, References, & Mechanics

I hesitate to give formal length requirements, because most students will write the bare minimum, and also because different papers have different optimal lengths. What truly matters is that your paper is long enough to say what you need to say, to say it well, and to say it briefly. Ceterus paribus, if papers \(A\) and \(B\) say the same thing, but paper \(B\) says it in half the length (without sacrificing key arguments), paper \(B\) is a better paper. You will also find that simply by including the necessary components of an empirical paper (plots, tables, etc.), your page count will by necessity increase on its own.

Since you of course still want to know what a good length will be, my rule of thumb is that your paper should be about \(\mathbf{10\pm 5}\) pages (size 12 font, double spaced, 1" margins). This will depend on the topic you have chosen to written on, your data, and your own personal writing style.

I would also like you to use scholarly references, that is, articles from economics journals and cite them properly. I do not have a minimum requirement of the amount of references, but I expect you to have at least 2-5 scholarly references, depending on your paper topic and thesis. I am not particularly picky about exactly how you format your citations or bibliography, just please be consistent, and do not use footnotes or endnotes (only because they annoy me). I suggest the APA author-year-page in-text citation format that is fairly standard in economics journals, i.e.: “The division of labor is limited by the extent of the market,” (Smith 1776: 27). Look at my slides or my handouts for a suggested bibliography style. If you use .bib files, the default formatting is fine.

Sources for Inspiration

Here is a list of a few places you might consider to dig up some information on your topic, or to help you find a topic. Just be sure to read and cite the actual sources that these secondary sources cite.

Data Sources

While it is one thing to find a topic to write on, it is an altogether different animal to find data to use to test empirical research questions. You will find out quickly that the constraint to writing an empirical paper is not the set of topics or questions to write on (though that is often a challenge itself!), but the data available to use.

Depending on the topic, you can also collect your own data, and many times you will want to create a custom dataset by simply combining data from different sources.

General Databases of Datasets

Good R Packages for Getting Data in R FormatSome of these come from Nick Huntington-Klein’s excellent list.

Below are packages written by and for R users that link up with the API of key data sets for easy use in R. Each link goes to the documentation and description of each package.

Don’t forget to installinstall.packages("name_of_package")

first and then load it with library().

Here is a list of good data sources depending on the types of topics you might be interested in writing on:

Most government agencies and non-profit organizations have publicly available data on the issue you are concerned with, and will be your first place to look.

Grading Rubric and Deadlines

While it may be possible for many papers in your college career, is not a paper you can write the last minute and do well on. To ensure that you do not get too far behind, I have split the assignment into stages that are due at different intervals over the semester. Note that your topic can and may change depending on what you are able to find and work with. The hardest part is finding data that allows you to test a research question. It is primarily for this reason that writing an empirical paper on what you want is very difficult.

Assignment Points Due Date Description
Abstract 5 Sun Oct 11 Short summary of your ideas
Literature Review 10 Sun Nov 1 1-3 paragraphs on 2-3 scholarly sources
Data Description 10 Sun Nov 15 Description of data sources, and some summary statistics
Presentation 5 Thurs Nov 19 Short presentation of your project so far
Final Paper Due 70 Tues Nov 22 Email to me paper, data, and code

All assignments are due by emails to me.

The remaining 70% for the final product are broken down as follows:

Category Points
Persuasiveness 10
Clarity 10
Econometric Validity 20
Economic Soundness 20
Organization 5
References 5
TOTAL 70

Emailing Your Final Version to Me

When you send your final email (by Tuesday November 22), it should contain the following files:

  1. Your final paper as a .pdf. It should include an abstract and bibliography and all tables and figures contained within it.
  2. The (commented!) code used for your data analysis (i.e. loading data, making tables, making plots, running regressions). These can be either .R files: one or multiple (one-per-task) are equally fine OR a .Rmd file. I want to know how you reached the results you got! Reproducibility is the goal!
  3. Your data used, in whatever original format you found it (e.g. .csv, .xlsx, .dta)

Again, you are not obligated to use R Markdown to write your paper. Microsoft Word is fine.