1.  Course Description and Objectives

The crucial step in data analysis is not in employing statistics but deciding which statistics to use. The research objective coupled with the data types determines the appropriate statistics that can be utilized. This seminar will introduce students to various types of statistical tools available that can help researchers to analyze data and build models. Emphasis will be placed on developing a working familiarity with some of the common statistical procedures and also on teaching a model building process with a focus on regression models. 

About regression, we will cover simple, multiple and nonlinear regression models with focuses on the model development stages and the model building skills. Students will learn how to:

1. specify a tentative model for application
2. estimate this tentative model from a dataset
3. perform extensive diagnostic checks to approximate the dataset
4. modify the model to approximate the dataset more closely
5. validate candidate models with new datasets
6. select a final model and perform estimates
7. use the final model for analysis, prediction and generalization

Applications with students’ own datasets will be used to illustrate the theories, fundamental concepts and skills. Students will obtain hands on experience in developing regression model and using statistical software.

2. Materials

1. Babbie, Earl 1998 The Practice of Social Research Belmont: Wadsworth Publishing
2. Fox, John. 1997 Applied Regression Analysis, Linear Models and Related Methods. Sage Publications
3. Sanford Weisberg 1985 Applied Linear Regression  John Wiley & Sons, Inc.
4. www.ResearchMethods.org/pol600.htm
5. APSA Style Guide for Political Science
6. Click here for more readings

3. Course Requirements

Each student will need to complete 7 problem sets and a final research paper.

Each student will be asked to bring his or her own research question, to select a dataset, and to build a regression model by following the model building stages and using the skills learned from this course.

4. Problem Sets

The emphasis of the problem sets is on understanding statistical concepts and obtaining familiarities with computing methods. Late homework will NOT be accepted.

5. Final Research Paper

Students are expected to bring one research question and a dataset to this course. Then, each student will apply the learned model building skills to his or her own research, and go through the model development stages to obtain a final model. A research paper based on this final model and the analyses need to be completed.  

There are five key components in the research paper:

• A critical review of the appropriate academic literature establishing the current state of debate, opinions on the issues, findings or short-comings.

• A clear and concise research question establishing the seminar participant's focus of the study and, if applicable how it differ from previous studies.

• The operationalization and measurement of the principal issues in the research question.

• The application of the appropriate statistical procedure and an interpretation of the results.

• A discussion of the implications and conclusions and how it pertains to the issue or debate.

6. Grades

The course grade will be determined using the following distribution:

                                                                                    Percentage

                                    Problems Sets                                 42%

                                    Seminar Participation                        18%

                                    Final Paper                                       40%

7. Software

We will use SPSS. But students are encouraged to use S+ or R.

8. Lectures

The lecture will be held in VKC 153 from 2:00pm to 4:50pm of every Wednesday. The instructor is Dr. Alex Liu whose office hour is 11:00am to noon of Wednesdays at VKC 342 with a telephone phone of 213-740-3297 and can be reached by email at alex@ResearchMethods.org  .

  BACK