Social Science 534
Quantitative Analysis in
Social Science
Thursday
1504
INSTRUCTOR: Dr. WU Xiaogang
OFFICE: 3377
PHONE: 23587827
EMAIL: sowu@ust.hk
OFFICE HOURS: Wednesday
INSTRUCTIONAL ASSISTANT: Mr. Stephen WAN Hoi Ming
OFFICE: 3001
PHONE: 23587817
EMAIL: sohmsw@ust.hk
OFFICE HOURS: TBA
OVERVIEW
This course surveys the
statistical models commonly used in empirical social research, building on the
topics covered in SOSC 509. We will start with a basic review of multiple
linear regression models, and then consider models and methods that should be
used when the assumptions of the classical linear regression model (CLM) are
violated. We will cover methods for categorical data analysis, including
techniques for binary, ordered and unordered polytomous
(dependent) variables, frequency counts, and censored and truncated variables;
methods for longitudinal data, including panels and event histories. Additional
topics may also be introduced to those who seek further training in
quantitative analysis.
PREREQUISITE
SOSC509 or its equivalent is
required.
COURSE WEBSITE AND DISCUSSION BOARD
I have established a WebCT account for this course (http://webct.ust.hk). You may use your ITSC
username and password to log in and find the course syllabus, homework
assignments, data sets, and links to other learning resources. You may also
post your questions and exchange ideas with me, IA, or your fellow classmates
on the discussion board. Lecture notes will be distributed in class.
● Required Textbooks
Jeffrey
M. Wooldridge. 2003. Introductory
Econometrics: A Modern Approach (2nd edition)
Daniel A. Powers and Yu Xie 2000. Statistical
Methods for Categorical Data Analysis.
● Recommended textbook
Sophia Rave-Hesketh and Brian Everitt 2000. A
Handbook of Statistical Analyses Using STATA (3rd edition) Chapman
& Hall/CRC.
The three
books have been ordered via the HKUST bookshop. The recommended textbook is
indeed a second choice. Students who wish to master in categorical analysis
using STATA should manage to purchase the following book:
Scott J. Long and Jeremy Freese
2001. Regression Models for Categorical Dependent
Variable Using Stata.
● Supplementary
You may be asked to read a few
research articles as exemplary applications of the models covered in class.
They will be distributed in electronic format.
REQUIREMENTS
You will be assessed through 4
assignments (40% of final grade), a short critique of a quantitative paper
(10%), a take-home midterm exam (15% of final grade), and a term paper (35%).
Stephen, the Instructional Assistant, will help you use STATA to solve homework
problems in lab sessions.
COMPUTING
You will be doing substantial
amount of data analysis with the aid of a software package called STATA 8.0,
which many of you have got familiar with via SOSC509. You can either purchase STATA
under “STATA Grad Plan” program, or use it in Social Science Computing lab. You
can get it by a discounted price.
You are assumed to have taken
SOSC509 and known the basics for using STATA. The tutorial sessions are focused
on how to use STATA to estimate and interpret specific models taught in
lectures. If you have no previous experience of using STATA, you should spend
some extra time on the software. Your IA will be available for help.
TOPICS
AND TENTATIVE SCHEDULE
Class 1
[4 February] Linear Regression I
Powers
and Xie 2
Wooldridge
2-4
Rave-Hesketh and Brian Everitt 3
Class 2
[5 February] Linear Regression II (Makeup for 18 February)
Wooldridge
6-7
Rave-Hesketh and Brian Everitt 4
[11
February] Chinese Lunar New Year [No Class]
[18
February] [No Class, Makeup on 5
February]
Term Paper Proposal Due on 22 February (Tuesday)
Class 3
[25 February] Linear Regression
Wooldridge 8-9
Rave-Hesketh and Brian Everitt 4
Class 4
[4 March] Binary Logit and Probit Models
Powers
and Xie 3
Wooldridge
17.1
Rave-Hesketh and Brian Everitt 6 &
7
Homework 1 Due
Class 5 [11 March] Ordered
Logit, Multinomial Logit, and Conditional Logit Models I
Powers and Xie 6 & 7
Rave-Hesketh and Everitt 6 & 7
Class 6
[18 March] Ordered Logit, Multinomial Logit, and Conditional Logit Models II
Powers
and Xie 6 & 7
Wooldridge
17.3
Rave-Hesketh and Everitt 6 & 7
Homework 2 Due
[25
March] Good Friday [No Class]
[1
April] TAKE-HOME MIDTERM EXAM
Class 7 [8 April] Poisson
Regression, Log-linear Models for Contingency Tables
Powers
and Xie 4
Class 8
[15 April] Censored and Truncated Regression Models
Wooldridge
17.2-17.5
Class 9 [22 April] Event
History Analysis I
Powers
and Xie 5
Rave-Hesketh and Everitt 12
Homework 3 Due
Class
10 [29 April] Event History Analysis II
Powers
and Xie 5
Rave-Hesketh and Everitt 12
Class 11 [6 May] Models for
Panel and Other Types of Clustered Data
Wooldridge
13-14
Rave-Hesketh and Everitt 12
Homework 4 Due
Class
12 [13 May] Paper Presentation
No
Paper Critique Due.
TERM
PAPER DUE AT
Supplementary
1. King, Gary 1986. “How to Not Lie with Statistics:
Avoiding Common Mistakes in Quantitative Political Science.”
American Journal of Political Science. 30:
666-87.
2. Xie, Yu and Emily Hannum 1996 “Regional Variation
in Earnings Inequality in Reform-era
3. Meng, Xin and Paul Miller 1995 “Occupational Segregation and Its
Impact on Gender Wage Discrimination in
4. Cheng, Yuan and Jianzhong
Dai 1995. “Intergenerational Mobility in Modern
5. Kung, James Kai-sing. 2002. “Off-farm Labor
Markets and the Emergence of Land Rental Markets in Rural Markets.” Journal of Comparative Economics
30-395-414 [Tobit model]
6. Raymo, James M. and Yu Xie.
2000. “Income of the Urban Elderly in Post-reform
7. Walder, Andrew G., Bobai Li, and
Donald Treiman 2000. “Politics and Life Chance in a State Socialist Regime:
Dual Career Path into the Urban Chinese Elite, 1949-1996” American Sociological Review 65:191-201. [Piecewise exponential
model]
8. Wu, Xiaogang, and Donald
Treiman. 2004 “The Household Registration System and Social Stratification in