Course Outline
Required Texts
Grade Requirements
Course Calendar
PHONE (312) 413-7274 OFFICE HOURS By Appointment or Luck
CLASS TIME Mon./Wed.:  3:00-4:15 CLASS MEETS IN BSB 4133
Course Overview
 Data Analysis for Political Science II (POLS 501) is the second course in modern methods of data analysis for the social and behavioral sciences. In terms of the level of difficulty of the material covered, this is an intermediate course, in the sense that the course material does not span advanced modeling techniques. Prior to enrolling in this course however, you should have successfully completed either Data Analysis for Political Science I (POLS 401) or a similar graduate-level course. Feel free to use my input in deciding whether your training in applied statistics qualifies you for POLS 501. If you fail to meet with me during the Add/Drop period, but subsequently show up in my office sobbing because you feel inadequately trained to complete this course, I can do no more for you than to give you some tissues and bid you adieu.

In this course, I have two goals: (a) to provide you with a firm grounding in the rudiments of regression analysis and maximum likelihood estimation for both continuous and categorical data and; (b) to train you in judicious use of modern statistical software (STATA 8) for applied data analysis. STATA 8 is available in all public labs on campus, and also available for purchase through ACCC (see their website for details). Purchasing STATA is not mandatory. A word of caution to those who do not like to teach themselves: While you will often receive handouts with relevant commands for statistical routines, I expect you to familiarize yourself with the software using STATA’s online help facility, the manuals (you may borrow the same from me to photocopy relevant sections), the STATALIST archive (you must sign-up; this is mandatory!), and STATA’s online help library will more than meet your needs if you tap these resources. Do not expect to be spoon-fed.

It is of utmost importance to your sanity and, of course, your education that you inform me as soon as you feel you are sinking in the web of material we cover. Each module builds upon the preceding one and hence missed class sessions and/or incomprehension of any material covered in a preceding class will only confuse you, and at an increasing rate. Work hard to build a strong foundation and if you do, you will come to see that data analysis can be fun, challenging and financially rewarding. Of course, where would we be if it were not exasperating as well! Consequently, patience, enthusiasm, an open and inquisitive mind, and a willingness to work hard, and a sense of humor will stand you in good stead. If you do not have an appetite for rigor, either develop one by January 10, 2005 or then drop this course.

If you have any conditions or challenges that may make it difficult for you to meet the requirements of this course or that may leadyou to require extra time on assignments, let me know so that we can make the necessary arrangements.

Texts and Other Requirements
  1. Gujarati, Damodar N. 2003. Basic Econometrics. New York: Mc-Graw Hill.
  2. Aldrich, John H., and Forrest D. Nelson. 1984. Linear Probability, Logit and Probit Models. Sage University Paper series on Quantitative Applications in the Social Sciences., 07-045. Newbury Park, CA: Sage Publications.
  3. Berry, William D. and Stanley Feldman. 1985. Multiple Regression in Practice. Sage University Paper series on Quantitative Applications in the Social Sciences., 07-050. Newbury Park, CA: Sage Publications.
  4. Berry, William D. 1993. Understanding Regression Assumptions. Sage University Paper series on Quantitative Applications in the Social Sciences., 07-092. Newbury Park, CA: Sage Publications.

 These texts are available via online sellers (for e.g., Amazon.Com). You may even find used copies.

The following readings are required for this course. While some are available from JSTOR and other online databases, a few are obtainable from the journals in the UIC library. This list may grow as we proceed.

Achen, Christopher H. 1977. "Measuring Representation: Perils of the Correlation Coefficient." American Journal
           of Political Science

Achen, Christopher H. 1990. "What Does 'Explained Variance' Explain? A Reply." Political Analysis 2:173-84.

Bollen, Kenneth A. and Robert W. Jackman. 1985. "Regression Diagnostics: An Expository Treatment of Outliers and Influential Cases." Sociological Methods and Research 13(4):510-42.

Chatterjee, Samprit and Frederick Wiseman. 1983. "Use of Regression Diagnostics in Political Science." American Journal of Political Science 27(3):601-13.

Cohen, Jacob. 1994. "The Earth is Round (p < .05)." American Psychologist 49(12):997-1003.

Downs, George W. and David M. Rocke. 1979. "Interpreting Heteroscedasticity." American Journal of Political

Geddes, Barbara. 1990. "How the Cases You Choose Affect the Answers You Get: Selection Bias in Comparative Politics." Political Analysis :131-50.

Hagle, Timothy M. and Glenn E. Mitchell. 1992. "Goodness-of-Fit Measures for Probit and Logit." American Journal of Political Science 36(August):762-84.

King, Gary. 1986. "How Not to Lie With Statistics: Avoiding Common Mistakes in Quantitative Political Science." American Journal of Political Science 30(3):666-87.

King, Gary. 1991. "Truth is Stranger than Predication, More Questionable than Causal Inference." American Journal of Political Science 35(4):1047-53.

Lewis-Beck, Michael and Andrew Skalaban. 1990. "The R-squared: Some Straight Talk." Political Analysis 2:153-200.

Lewis-Beck, Michael. 1986. "Comparative Economic Voting: Britain, France, Germany, Italy." American Journal of Political Science 30(2):315-46.

Long, J. Scott and Pravin K. Trivedi. 1992. "Some Specification Tests for the Linear Regression Model." Sociological Methods and Research 21:161-208.

Nagler, Jonathan. 1994. "Scobit: An Alternative Estimator to Logit and Probit." American Journal of Political Science 38(1):230-55.

Petersen, Trond. 1985. "A Comment on Presenting Results from Logit and Probit Models." American Sociological Review 50(1):130-31.

Stimson, James A. 1981. "Interpreting Polynomial Regression." In Linear Models in Social Research,  Peter Marsden. Los Angeles: Sage Publications.

Stimson, James A. 1985. "Regression in Space and Time: A Statistical Essay." American Journal of Political Science 29(4):914-47.

Stolzenberg, Ross M. 1980. "The Measurement and Decomposition of Causal Effects in Nonlinear and Nonadditive Models." In Sociological Methodology, Karl F. Schuessler. San Francisco: Jossey Bass.

Stolzenberg, Ross M. and Daniel A. Relles. 1990. "Theory Testing in a World of Constrained Research Design." Sociological Methods and Research 18(4):395-415.

·         You will need a calculator capable of performing basic mathematical and statistical operations.

·         You must also hold an active email account.

·         Bookmark the URL for the course:

There are also a growing number of online statistics texts that explain the innards of statistics in an easy-to-understand manner, using everyday language and examples. A few of the better online statistics texts are linked below. You may be surprised at their usefulness. 
Internet Glossary of Statistical Terms
Statistics at Square One
Introductory Statistics: Concepts, Models, and Applications
The Data Analysis Briefbook
Statistics: The Study of Stability in Variation
Introduction to Probability
Virtual Laboratories in Probability and Statistics
HyperStat Online
The Statistics Homepage
The Statlets at Duke
The Research Randomizer

STATA: Basic through Advanced Commands with Examples

J Scott Long's Helpful Utilities

Grade Requirements
Your course grade is a function of your ability to work hard and to persevere. Three elements determine your course grade – a series of short assignments (40%), and two non-cumulative exams (30% each).

Grading Scale in Effect: A (94-100); B (84-93); C (74-83);  D (64-73); E (0-63)

Note: I assume you will not miss class without giving me prior notice (in person, via email, or by phone) of both the cause and date(s) of your absence(s). I administer make-up exams only for excused absences. Late assignments earn no points and there are no make-up assignments or extra credit.   

Course Calendar
Because we set our own pace, the Course Calendar lists neither examination dates nor session dates beyond the first day of classes. You will receive sufficient advance notice about examination dates, homework assignments, and reading instructions for upcoming class meetings.
Course Introduction

The Classical Linear Regression Model (CLRM)


Extending the CLRM 

Dummy Variable Regression



Model Specification and Diagnostic Testing

Discrete Choice ...
The Linear Probability Model

Censoring and Truncation 

Ordered Logit/Probit 

Multilevel Models  

Data Sets


Heteroscedasticity Multicollinearity Autocorrelation
Dummy Extending CLRM Model Specification LPM/Logit/Probit
NES 2000 Tobit (1)  Tobit (2) Multilevel Models Ordered Logit/Probit