Statistical Modeling Final Paper/Project

Instructions for the first draft ...

This paper/project gives you an opportunity to apply the statistics that you are learning in class in an interesting real-world setting. You can either use one of the two data sets offered in class to conduct your analyses, or you can choose your own data set. Most students should use one of the two data sets offered in class. But if you are participating in an honors program (in any department), then you can use a data set that you plan on using for your honor's thesis.

Early in the semester (see the syllabus for the exact date), our class will be visited by two representatives from the Emory library. One of the representatives (Dr. Rob O'Reilly) will talk with you about how to obtain a data set for your project from collections of many interesting data sets freely available to the Emory community. The other representative (Chris Palazzolo) will talk with you about print resources available from Emory library that relate to your project. These print resources will be important to you even if you are using one of the two data sets offered in class for your project. It is very important that you attend this session.

For this first draft, you are to come up with a topic that you plan to write about. You also need to identify a data set that you will use to conduct your statistical analyses. If you are using one of the two data sets offered in class by your instructor, then you will already have the data in the proper format (SAS). If you are using a different data set, then you will need to obtain the data set from the library and convert it into the SAS format. Dr. Rob O'Reilly (from the library) will tell you how to do this when he talks to the class in September. He can also help you do this.

The final paper/project will be submitted as various drafts. The final paper should be anywhere from 10 to 20 pages. Quality is what matters, not length. The final version of the paper/project is due on the last day of normal class. The due dates for the previous drafts are in the syllabus.

Your first draft should be in the range of 3 to 5 pages, and it should include an introduction, a discussion of the problem under investigation, a modest literature review that is relevant to the problem (try to obtain at least five initial print references), a discussion of the data to be used in the project, and a clear description of the unit of analysis. Most of the references should be normal print sources, such as journal articles and/or books. Use Internet sources (i.e., url's) sparingly.

In terms of format, I have some initial suggestions. I will add to these suggestions later. Always use a title page. Always staple pages in the upper left corner. Do not use plastic covers for your papers. The second page should contain a one-paragraph abstract. I know that your first draft will have little to summarize in an abstract, but make a brief summary of what you do have. You will add to this in future drafts. Place the word "Abstract" centered at the top of the abstract page. The third page will begin with your introduction that will include a brief statement of your research topic. Shoot for between one and three pages. Following the introduction will be a brief review of the relevant literature. Following this will be a discussion of your data set. Use topic headers for each section of your paper. Thus, you will have sections with the headers such as "A Brief Review of the Literature," "The Data," and so on. Do not use a header for the introduction. Double space with 12 pt. Times New Roman font. Use normal one inch margins. Number all pages beginning with the page that begins the introduction.

Spend time reading Gary King's really good advice on writing your first empirical (and potentially publishable) paper. My advice generally parallel's his, although I do not want you to collaborate with this project/paper. This project/paper should reflect only your own work. Although it is perfectly OK (and advisable) to interact with others during the semester, sharing discoveries and insights. If you choose to replicate someone else's findings (as suggested by Professor King), then that is fine. But that is probably more appropriate if you are doing this project with the intent of using it as part of an honor's thesis. Replication is enormously helpful for students to fine-tune their skills. But the best scholars eventually come up with their own ideas (from scratch, so to speak). So you gain a lot when you try to do this early on. Remember that all scholarship (replication or otherwise) is based on prior research that has been conducted by others, so there is a lot of similarity between replication and other more fully original work. You learn either way!

Hand your draft in on time. This is important.

This entire site is Copyright © 1997-2008 by Courtney Brown. All Rights Reserved.
DISCLAIMER
URL: www.courtneybrown.com