POS 6208
Empirical Political Research
Spring, 2007
Dr. Kenneth Wald
Class: Tuesday, 8-10 periods  (3:00-6:00)
Room: Matherly 002
Telephone: 392-0262, x267
Office: 318 Anderson
Tu 10:30-11:30, 1:30-2:30,Weds 1:30-4:30 or by appointment

* Course Description and Purpose *
* Requirements *
* Books *
* Grades *
* Detailed Schedule with Assignments *

* Useful Links *
* Contact the Instructor *

Description & Purpose: This seminar is intended to equip students with the skills to conduct empirical research projects on social and political subjects. The skills emphasized include criticism/evaluation of research, problem and hypothesis formulation, concept development, measurement, data analysis and computer-based statistical analysis. The culmination of the course will be an original empirical research paper on some aspect of  social or political  behavior that interests you. About two-thirds of the semester will consist of collective meetings devoted to honing your knowledge of research concepts and methods. Along the way, you will prepare a number of assignments and short papers. Following the completion of that portion of the course, we will cease meeting as a group and you will devote your energies to the individual research effort. At the end of the term, we will reconvene for oral presentation of your research papers in a convention panel format.

Requirements: As in any graduate seminar, you are expected to attend each weekly meeting, read the assigned material thoroughly, participate actively in the discussions, and complete all assignments on time. Accordingly, two unexcused absences from the seminar will constitute grounds for dismissal and students will be penalized for handing in late assignments without advance authorization. During the last third of the semester, when we do not meet as a group, your weekly meetings with me count as seminar time.

Grading Criteria: Your final grade will be calculated in the following way:
class participation 10%
written assignments 15%
midterm examination 25%
final paper 50%

The standards for evaluating the final paper, the single largest component of the seminar grade, will vary from student to student in response to methodological background. I assume that all seminar participants have mastered the basics of research design (as conveyed in our "Conduct of Inquiry"), elementary statistics (STA 6126, POS 6737 or the equivalent), and some core questions in political behavior (as explored in POS 6207) or institutions. Those with skills and background beyond this level will be expected to produce more sophisticated research projects.

The midterm examination will be a take-home exercise in which you are allowed access to your books and notes. The goal of the examination is to determine your mastery of the key concepts and skills covered in the seminar to date.

Books: Readings will be taken from the following four paperback books and several articles that will be available on library reserve. The books are available only at Goerings Bookstore on campus.
Howard S. Becker, TRICKS OF THE TRADE, University of Chicago Press, 1998, 0-226-04124-7

John W. Creswell, RESEARCH DESIGN; QUALITATIVE, QUANTITATIVE, AND MIXED METHODS APPROACHES, 2nd ed., Sage Publications, 2003, 0-7619-2442-6

Philip H. Pollock III, AN SPSS COMPANION TO POLITICAL ANALYSIS, 2nd edition, CQ Press, 2005, 1-56802-996-9. (If you already have a manual for a statistical package, you do not need to purchase this.)

Students who are conducting quantitative analysis will require access to a statistical package. The datalabs maintained by CLAS have SPSS for Windows and STATA, the two most widely-utilized statistical packages in the social sciences. You may wish to purchase the SPSS Graduate Pack at the UF BookstoreHUB (for about $140). This entitles you to a permanent license for the program, which you can use in your later research here at the University and in your subsequent career. Employees of the university (including graduate assistants) may acquire a limited license for SPSS for $45; this license will expire annually, but can be renewed in subsequent yearswhile you are at UF.


1. Overview (January 9)

Review of POS 6207 (or equivalent source of ) research topics - be prepared to present your idea for a research topic to the seminar. Together, we will review the kinds of research topics proposed by seminar participants and discuss the designs' basic strengths, weaknesses, and possible pitfalls to avoid.

2. Doing Political Science Research (January 16)

Topic: Getting Started in Political Science Research

Assignment due in class: Write a 2-3 page essay (using an appropriate professional style) describing the process by which you arrived at your research agenda. Don't be afraid to offer personal history as it relates to your work.

Becker, chaps. 1-3
Creswell, chap. 1-2
Benjamin A. Most, 1990. Getting started on political research. PS: Political Science & Politics 23 (4, December), 592-6.
Kenneth D. Wald, "Homeland Interests in Hostland Politics: Politicized Ethnic Identity among Middle Eastern Heritage Groups in the United States," 2006.
Myron J. Aronoff, "Forty Years as a Political Ethnographer," Ab Imperio, 4/2006

3. Beginning Analysis (January 23)

Topic: Evaluating Research

Assignments due in class:
(1) Read David C. Nice, "Abortion Clinic Bombing as Political Violence," AJPS 32 (February 1988), 178-195. Then turn in a short written profile that describes the data and methodology used (theoretical framework, research design, sample, hypotheses, identification of independent and dependent variables, measures and conclusions). Indicate whether or not you would have recommended publication of this article and why or why not. (A full review normally evaluates the coverage of the literature, adequacy of the independent and dependent variables, the quality of the data, and, most importantly, whether the conclusions were justified by the analysis.)

(2) Turn in a brief summary of your research plans for the major paper. This should indicate the problem you wish to explore, the major theoretical traditions likely to be consulted, and the major data set(s) you will use for the empirical analysis. 

Creswell, chaps. 3-5

Roper Center
Selected Web Resources for Statistics in the Social Sciences - University of North Carolina Library
Internet Resources in Political Science - Poly Cy
Political Science and Public Affairs Reference & Web Resources - University of Washington Library
International Relations Data Sets - FSU
Social Science Data on the Net - UC-San Diego
Council of European Social Science Data Archives
Statistical Resources on the Web - University of Michigan

4. Conceptualization and Measurement (January 30)

Topic: Converting an Interest into a Researchable Topic.

Assignment due in class: Using  the dataset that you expect to use in your own analysis, present one or two simple bivariate hypotheses and the appropriate bivariate data analysis. If you do not have data, present the tables that you would produce without the cell entries.

Becker, chap. 4
Creswell, chaps. 7-8

5. Hypothesis Testing (February 6)

Topic: Developing Testable Hypotheses

Assignments due in class:

(1) Using the GSS data, construct indexes for religiosity, attitudes toward abortion, and sexual tolerance. Which, if any, of these indexes are valid? (Show your SPSS printouts and text that explains your interpretations.)

(2) Present a written list of the major hypotheses that you intend to test in your research paper.   You should present an argument for the face validity of your measurements, and tell me how you will determine one other form of validity or reliability for each of your measurements.

Becker, chap 5
Creswell, chaps. 6, 9-11
Coogan and Woshinsky, Science of Politics, chap. 3
Leege and Francis, Political Research, chap. 5

6. Advanced Analysis (February 13)

Topic: Using Computers for Quantitative Analysis

Assignment due in class:

(1) Using the GSS data, tell me 

(a) Are more Americans Democrats, Republicans, or Independents?
(b) Which value do Americans think is most important for children to learn? (Obedience, popularity, independent thinking, hard work, or helping others)
(c) Under which condition are Americans the most supportive of a woman's right to a legal abortion?
(d) Under which condition are Americans the least supportive of a woman's right to a legal abortion?

(2) Using the dataset that you expect to use in your analysis, present the appropriate univariate statistics (frequencies or
descriptives) for your principal independent and dependent variables.

You will be required to turn in printouts showing your work, as well as text that explains your interpretations.

Read: TBA

7. Finding Findings (February 20)

Topic: Making Sense of Empirical Data

Assignment due in class:

(1) Using variables in the GSS, formulate two hypotheses which might account for variations in Americans' attitudes on abortion. Present tables showing the association between the independent and dependent variables in your hypotheses. Write a short paragraph explaining whether the results tend to support your hypotheses.

(2) What statistic did Tufte use in calculating the "swing ratio"?   He finds that the swing ratio for Great Britain's House of Commons is higher than it is for the U.S. House of Representatives, and that the swing ratio for New Jersey is higher than that for New York.  What does that mean politically?  (Tufte's article is excerpted in Ethridge, Chapter 12.  The original article is Edward R. Tufte. 1973.  The Relationship between Seats and Votes in Two-Party Systems.  American Political Science Review  67:  540-554.)

Read: TBA

8. Data Analysis and Report Writing (February 27)

Topic: Presenting Findings

Assignment due in class:

Using data from the GSS, formulate three or more hypotheses explaining variations in Americans' attitudes toward abortion. Test these hypotheses in an appropriate multivariate analysis. Evaluate your hypotheses. (As usual, show me the printouts and write a short interpretation.)

Coogan and Woshinsky, Science of Politics, chap. 4
Leege and Francis, Political Research, chap. 10 - as needed

9. Last Things (March 6)

We will not meet as a class on this date. You will have two assignments due to me.

(1) Present a written preview of your intended analysis for the research paper. You will need to indicate which tests you plan to run and why. Also indicate what evidence will be necessary for you to confirm your hypotheses. Due in class on October 24th.

(2) The midterm exam will be handed out on Tuesday, March 6th and is due back to me one week later. I expect a hard copy delivered to me at my office or my mailbox in the department office.. Do not slide this under my door! 

10. No Class: Independent Work on Research Papers (March 20-April 10) Each student should meet with me weekly during this period to evaluate progress.

11. Research Panel (April 17-24)

Useful Links