Statistical analysis is brought to bear in the social sciences under a wide variety of conditions, using a wide variety of data, to test hypotheses about a wide variety of relationships. Therefore, in order to read the current literature thoughtfully, and to choose and conduct the appropriate data analysis technique for a project, a political scientist must be exposed to the wide range of statistical approaches available.

Our goal for this course is to provide you with this sort of exposure. We will present some of the most commonly seen, as well as some of the most cutting edge, statistical analysis techniques that go beyond multiple regression analysis. These techniques include ARIMA modeling, vector autoregression, cointegration models, multinomial and bivariate logit, ordered probit, event history analysis, factor analysis, pooled cross sectional analysis, bootstrapping, Monte Carlo simulation, graphical presentation and analysis, among others.

While it would be impossible to master each of these in the course of a semester, we believe that by the end of the course, for each technique discussed a student should be able to:

• -identify the goal of the technique,
• -understand the sort of data used in it and the relationships being modeled,
• -understand the differences between this and related techniques for modeling the specified relationships with this sort of data,
• -conduct and interpret simple analysis using the technique,
• -know where to go to learn more about the technique and ones related to it,
• -interpret and critique published results of the analysis.

Further, each student will demonstrate mastery of at least one of the techniques discussed by using it in a piece of original research, as discussed below.

We will spend much time in the computer lab, and we will use a variety of computer programs (e.g., RATS, GAUSS, LIMDEP, SPSS). This is necessary because no single program can, nor will ever be able to, do all the types of analysis we may need. While instruction in these programs is part of the course, students must have a mastery of at least one general statistical package (e.g., SAS, SPSS, NCSS, etc.) before entering the course.

The basic prerequisite of the course is at least two progressive semesters of statistics, yielding a thorough understanding of (at least) basic statistical inference and ordinary least squares regression analysis.

READINGS:

Much of the reading for the course will be from the following Sage QASS monographs (the "little green books"), which are available at the bookstore:

NOTE: The final two books in this list (Jacoby and Mooney) are due to be published in March, which should be in time for our discussion of them in the course. However, if they do not come in or if you want to read them sooner, I have the manuscripts which you can xerox at any time (this might actually be the cheaper way to go!).

There will also be a variety of articles and book chapters assigned weekly. See the course outline for details.

Students are expected to make at least their initial run through the readings before attending class.

ASSIGNMENTS:

Grades for the course will be based on two components: problem sets and a final paper.

Problem sets:

Problem sets will be assigned for each section of the course. These will consist of conducting and/or interpreting data analysis, and will be graded on a 0-1-2 scale, with a premium being placed on timely and thorough completion. Assignments will be due at class time on the specified date, and NO LATE PROBLEM SETS WILL BE ACCEPTED.

Most of these problems sets will require computer analysis. It is important, however, that you never turn in raw computer output. Your job is to conduct the analyses appropriately, interpret the results, and then present them in a useful way for the reader. Take as models for interpretation and presentation the sort of approach you see taken in journal articles using these techniques.

Final paper:

Each student will identify a significant research question in his/her field of study, gather data which represents the variables involved, and conduct analysis to address that question using one of the techniques discussed in the course. In order for us to provide guidance in the preparation of this paper, you will be required to turn in various brief intermediate papers throughout the semester. Always submit 2 copies of each assignment.

The schedule of these assignments
Due January 14	the general topic of the paper
Due January 28	the research question and hypothesis, and an extended substantive bibliography
Due February 11	a description of the data to be used, and how and when it is to be secure
Due March 18	a description of the data analysis technique to be used, how it is to be applied in this case, and a discussion of why it is the appropriate approach.
Due April 1	the data analysis results done to address the question
Due April 25	two copies of the completed paper

We expect each student to be in regular consultation with us as this paper develops.

The final paper is expected to be a significant piece of original research, along the lines of a journal article or research note. We anticipate that more than a couple of these papers will eventually find their way into the public domain, whether as national conference presentations and/or as published articles.

Final course grade weighting:

• All problem sets: 50%
• Final paper: 50%

STUDENTS WITH SPECIAL NEEDS

If you are a person with a disability and anticipate needing any type of accommodation in order to participate in this class, please advise us and make appropriate arrangements with Disability Services (293-6700).

SOCIAL JUSTICE STATEMENT:

West Virginia University is committed to social justice. We concur with that commitment and expect to maintain a positive learning environment based upon open communication and mutual respect, and non-discrimination. Our University does not discriminate on the basis of race, sex, age, disability, veteran status, religion, sexual orientation, color or national origin. Any suggestions as to how to further such a positive and open environment in this class will be appreciated and given serious consideration.