Political Science 310 (formerly, C10)

Fall, 2001

Elementary Statistics for Political Research

Mr. Janda

Go directly to
Course Outline by topics

This course in statistics is designed to integrate research methods with the substance of political and social research. Statistical techniques and applications have been chosen for their relevance to political science, a field that many people do not regard as quantitative. This approach should make the study of statistics more meaningful and valuable to those who are uncertain about the role of statistical analysis in the softer social sciences.

In addition to its focus on political and social research, this course differs from most other statistics courses in using the SPSS computer program as an integral part of statistical training. Computers are important and commonplace tools of political research, and it is vital for students, as well as professors, to know how to use standard computer programs for statistical analysis. Each student will be expected to record a small set of data for use in statistical exercises with the computer and to perform other analyses on data called from computer storage.

Until this year, I taught statistics using SPSS on central computers ("hardin" and "seldon") that operate under the Unix operating system, which students accessed through microcomputers and remote terminals. SPSS also runs on microcomputers, but the program costs $250 when purchased under a Northwestern site license and then costs an additional $125 per year renewal fee. SPSS also requires a lot of computer power. I was reluctant to ask students to incur the cost until I was severely criticized last year in a student's CTEC comment:

Perhaps I was wrong and the student right. In any event, I managed to get SPSS to lower its cost of the program so that students can purchase the base version of SPSS, which lists for $599 without an annual renewal fee, for only $90. I will teach this version of the program, expecting that students will either purchase it individually, share a purchased copy, or use it in one of the campus computer laboratories, including the one in Scott Hall, Room 401. Additionally, I will also keep available on my class website the syntax commands for running SPSS on a central computer.

REQUIRED
PURCHASE

SPSS Career Starter Program 10.0 (Chicago: SPSS Inc., 1999). Available for $90 only at Norris Bookstore

Norris has both PC and Mac versions of SPSS at this price--which includes a 400 page SPSS Base 10.0 Applications Guide. That Guide, my lecture notes, and class handout will serve as at the supporting text for the class. For students who want a genuine textbook, I can recommend Alan Brynam and Duncan Cramer, Quantitative Data Analysis with SPSS Release 10 for Windows: A Guide for Social Scientists (Philadelphia: Taylor and Francis, 2001: ISBN 0-415-24400-5), which sells for $30 new over amazon.com and $12.95 used.

LECTURE NOTES

It's difficult to take accurate notes in a statistics class while also following the lecture. So I will make my notes available on my personal web site, janda.org/c10, for you to print out.

TEACHING
METHODS

The course will be conducted in lectures four days a week--on Monday, Tuesday, Thursday, and Friday--at Noon in Swift Hall 107, where we have video equipment to display output. Optional discussion sections (I will hold them at my option and you can attend them at your option) will be held on Wednesdays. There is just too much material to be learned in one quarter to cover it well in only three lectures per week. Although you are asked to attend lectures an extra hour a week without increased course credit, remember that I am teaching the extra lecture without increased pay. I would not impose the burden on us if it were not needed to cover statistics properly in one quarter.

EVALUATION OF
PERFORMANCE

Your performance in this course will be evaluated through a 1/3 term exam (worth 15% of the final grade), a 2/3 term exam (worth 25%), a final examination (worth 35%), and a short paper (worth 25%). Your scores will be cumulated to produce a total point score on which your grade will be based.

TEACHING
ASSISTANTS

I will be assisted by an advanced graduate student, Gabriela Nava-Campos, who will hold weekly office hours to help with your computer runs and answer questions about the lectures. She will also grade your research papers.

RESEARCH
PAPER

The paper is intended to demonstrate your competence in applying statistics to a specific research problem of your own formulation. You will be expected to devise a concise and non-trivial hypothesis that can be tested with data available for computer analysis. You must state briefly the source of your research problem or question, formulate it as a hypothesis for testing with available data (more on this later), execute the appropriate test, and draw a conclusion about the truth or falsity of the hypothesis--within a maximum of 5 double-spaced typewritten pages of text (not including tables or figures). Evaluation of this exercise will be based mainly on clarity of presentation and statistical craftsmanship rather than on the substantive or theoretical importance of the problem.

The expected form and style of the paper can be likened to the "research notes" that are occasionally published in political science journals. We will be reading several examples of such articles. A one-page progress report stating the hypothesis and data set is due on November 19. Your 5-page paper will be due on December 3--the same day that we begin reviewing for the final examination.

YOU SHOULD PURCHASE A LOOSELEAF NOTEBOOK to contain your lecture notes and the numerous handouts that you will receive in class. Unless you have some way to organize that material for effective retrieval later, you will be left with a confusing pile of papers.

READINGS AND
ASSIGNMENTS

Reading assignments from the required texts (and from material on the web) are specified on subsequent pages in the syllabus. Be certain to read the accompanying "comments" on the assignments, for they will direct your attention to what you are expected to learn.