The goal of this project is to create a research paper of the form published in standard economics journals. Instruction in this class is intended to give the student

experience in the forms of analysis done by practicing economists in business, government, and consulting. These economists routinely publish research papers on the

applied topics that they are working on. Thus you should see your audience as people who read economics journals and the editors who decide on which papers to publish

and which to turn down. In short, you should assume that your audience is experienced in reading economics papers and is sophisticated about their analysis.

All research papers are exercises in persuasion, in which the author tries to convince the reader that he or she has discovered something new and that it is worth the

reader?s time to understand the importance of the discovery. Thus a successful paper is one where the reader knows exactly what the author thinks he or she has

discovered and why it is important. You should keep this in mind at all times as you write this paper.

In order to keep your focus on what is important, it will be useful to think of the paper as the test of an hypothesis. What you will prove in the paper is the truth

of the hypothesis. Within the first page or two, you should clearly state the hypothesis and why it is important to know whether it is true or false. You then lay out

the logic of the test and then do an econometric test of your hypothesis to confirm the validity of your idea. This is the form that essentially all economics research

papers follow and yours should follow it, too.

All economics papers start with an introduction or problem setting. This should give the background of your issue and try to convince the reader that it is important

to know whether your idea is true or false. The introduction usually ends with a paragraph telling the reader how the argument will be made in the remainder of the

paper. This section is generally between 2 and 3 pages long.

The second section of most papers is a literature review. This gives the reader an idea of where your analysis fits with what others have done. Depending on what the

hypothesis is that you will be testing, the literature review can be lengthy or only an article or two. Even if you think that you are the first person ever to attack

your subject, you MUST include a discussion of the most closely related literature so that your reader knows where your result fits in the existing literature. The

length of this section is highly variable?perhaps between 1 and 5 pages.

Once the reader knows what you will prove and where it fits in the existing literature, you need to explain how you will prove the hypothesis with the data that you

have. This is the most important part of the paper since it lays out for the reader how you want your results to be interpreted. This section is typically called a

theory section, or a model specification section. This is where the author succeeds or fails to convince the reader that something new has been discovered, so this is

often the longest and most carefully constructed part of the paper. It may also be the first part of the paper that you will want to write. The theory section often

leads up to an algebraic representation of the functional form that you will estimate, but this is not always necessary. A rough guideline is that this section should

be three to five pages long.

Some papers have data sections that are separate from the theory section, and some papers combine the two. In almost all cases, the theory (a prediction of what the

effect is of one economic force on another) is built on a more general level than the measurement. The job of the data section is to convince the reader that the data

you have selected reasonably represent the economic forces that are part of your theory. For example, you might have a theory that higher incomes will lead to higher

levels of traffic safety. A successfully written data section convinces your reader that the data series you have selected to represent income is reasonable and that

you have reasonably measured ?safety.? It is a good idea to give enough documentation in the paper so that an interested reader can replicate the test that you have

done.

The heart of your paper must be several multiple regressions using data from the series linked to the class web page. You are welcome to use other data series as well,

so long as at least one of the series is from a link on the class web page. There should be a main regression showing your preferred model and then several variations

on the theme to show the robustness of your result or, equivalently, the sensitivity of your result to minor modifications of the functional form. The regression must

be reported in a standard form, the easiest of which is a table with the dependent variable on the top of each column, independent variable names on each row, and

coefficient value and standard error as entries. You will wish to show which coefficients are significant. See Woodbury?s style manual for examples of a standard form.

Please do not include unformatted output in your paper. Reformat the output in standard form..

The results section is often short, but with a longer discussion section in which you repeat the interpretation that should be given to the result you have shown in

your regression table. By the end of the results section, your reader should be able to see how you have proved the hypothesis that you laid out in the introduction.

Including tables, this section of the paper is rarely more than 4 pages

All research papers end with a conclusion section that summarizes the findings of the paper, that suggests extensions to the research, and that give the reader a sense

of where the author thinks that the results are most vulnerable. One or two pages is plenty.

After completing the paper, you should write an abstract. This is a 100 word summary of what is in the paper which goes on the title page directly below the paper

title and the name of the author. An efficient reader will generally read the abstract to get an idea of what the paper is about and then turn to the conclusion to get

a sense of how the paper is argued and what to watch out for in terms of strengths and vulnerabilities of the analysis. After the Abstract and Conclusion, most readers

will turn to the introduction and read through from beginning to end.

The final section of the paper is a reference section which shows work that you have referenced throughout the paper. You should plan on referencing everything that

you read as you prepare the paper. The references should be in any standard form that you like. Woodbury gives some examples. Please make sure that you reference at

least some academic literature on the subject as well as journalistic treatments and data sources. A paper that does not reference academic literature is far less

credible than one that does, and part of your job is to make a credible case for the believability of what you have discovered in your research. Please make sure that

you give credit to ideas and passages that you have borrowed from others. All professors make a routine plagiarism check on term papers that have submitted. You do not

want to be found using someone else?s ideas and language without attribution. You can not get into trouble if you attribute what others have done to the original

authors. You can get into deep trouble if you plagiarize.