Type of Credit: Required
Credit(s)
Number of Students
This is the second full course in quantitative methods for the graduate program of political science at NCCU. The course introduces students to regression models for the analysis of quantitative data, and provides a basis of knowledge for more advanced statistical methods. It will also have a substantial programming/computation focus. The course assumes basic math literacy, including familiarity with probability theory, properties of estimators, rudimentary calculus, and linear algebra, as well as mastery of the basic statistics taught in the previous course.
能力項目說明
The bulk of the course will focus on regression models for continuous response variables, and will include discussions of the mathematical bases for such models, their estimation and interpretation, model assumptions and techniques for addressing violations of those assumptions, model diagnostics, and topics related to model specification and functional forms. The course will also address a range of other topics, including maximum likelihood, generalized linear models (logit, probit, etc.), and Bayesian inferences.
| 教學週次Course Week | 彈性補充教學週次Flexible Supplemental Instruction Week | 彈性補充教學類別Flexible Supplemental Instruction Type |
|---|---|---|
|
週次 |
課程主題 |
課程內容與指定閱讀 |
教學活動與作業 |
|---|---|---|---|
|
1 |
Course Introduction |
Alley, Joshua. 2021. “An Open Collection of Political Science Research with OLS Models and Cross-Sectional Data.” Political Methodologist blog, September 8, 2021. Roberts, Margaret E. 2018. “What is Political Methodology?” PS: Political Science & Politics 51:597-601. |
Methodology and methods in political science The advantages and disadvantages of using quantitative methods |
|
2 |
Correlation and Simple Linear Regression |
Agresti, ch 9 |
Least squares prediction Linear regression model Inferences for the slope and correlation Model assumptions and violations |
|
3 |
Multiple linear regression (I) |
Agresti, ch 10, 11 |
Multiple regression model R square Interaction effects Comparing models Standardized regression coefficients |
|
4 |
Multiple linear regression (II) |
Agresti, ch 12 and 13 |
Multiple linear regression with categorical explanatory variables Dummy variables ANOVA Categorical predictors Statement of Research Question due by end of class |
|
5 |
Multiple linear regression (III) |
Agresti, ch 14 |
Model selection Diagnostics Multicollinearity Nonlinear relationships |
|
6 |
GLM |
Long and Freese, ch 4 and 9.1-9.3 |
GLM: a generalization of linear regression Maximum likelihood Description of Hypothesis due by end of class |
|
7 |
Models for binary outcomes (I) |
Long and Freese, ch 5 |
Logistic regression for binary variables |
|
8 |
Models for binary outcomes (II) |
Long and Freese, ch 6 |
Tests and interpretation of logistic regression |
|
9 |
Ordinal logit models |
Long and Freese, ch 7 |
Models with ordinal dependent variables Introduction to Data Source due by end of class |
|
10 |
Multinomial logit model |
Long and Freese, ch 8 |
Models with nominal dependent variables |
|
11 |
Midterm exam |
Midterm exam |
Midterm exam |
|
12 |
Bayesian analysis (I) |
McElreath, ch 1-2 |
Why Bayes? How Bayes? |
|
13 |
Bayesian analysis (II) |
McElreath, cha 3 |
Sampling for Bayesian Analysis Proposal of Research Methods due by end of class |
|
14 |
Bayesian analysis (III) |
McElreath, cha 4 |
Linear Models I: Preliminaries |
|
15 |
Bayesian analysis (IV) |
McElreath, ch 5 |
Linear Models III: Multiple Predictors |
|
16 |
Bayesian analysis (V) |
McElreath, ch 6 |
Assessing & Improving Model Performance |
|
17 |
Term paper oral report |
Please circulate your paper by the designated deadline |
Please read other students’ papers and be prepared to offer comments in class |
|
18 |
Term paper oral report |
Please circulate your paper by the designated deadline |
Please read other students’ papers and be prepared to offer comments in class Final Paper due by end of class |
This course is organized around lectures, readings, and labs. You are expected to attend lectures as well as lab sections. To pass the class, ALL assignments must be completed.
Attendance and Participation (15%)
Homework (40%)
Data Analysis Project and Presentation (25%)
Midterm (20%)
Attendance and Participation (15%): Your preparation, presence, and participation are crucial. Please complete the required readings, be on time for each class, bring all relevant readings, and contribute energetically to the class and lab activities. Your class participation grade will be assessed based on lecture attendance and your contributions to lab activities. Teaching assistants may distribute additional section syllabi that detail specific lab expectations and requirements. Please note that unexcused absences in lectures or in the lab will count heavily against your grade. An absence will be excused only with documentation of medical necessity or with prior approval from your Teaching Assistant.
Homework (40%): Students will complete five problem sets designed to test their comprehension of the material covered in class (see the schedule for the exact due dates). You may consult with your classmates (see the study group policy below). However, each student must write up and turn in their own work/assignment. Assignments deemed too similar to another student’s assignment will receive a score of 0. Working (struggling) on the homework is the only sure way to master the material. All homework assignments are due at the beginning of the class. I am willing to receive them via email or in person.
Data Analysis Project and Presentation (25%): Students will utilize techniques of quantitative analysis included in this course to explore a research question in any field of social sciences. Students will learn how to access the many survey data sets maintained by the Inter-University Consortium for Political and Social Research (ICPSR). The paper should include a clear description of the following parts: research question, literature review, theory, hypotheses, concepts and variables, data, methods, results, conclusion and implications. Papers must be written in a scholarly style with footnotes (or endnotes) and references in the author-date system. For the formats, please refer to the guidelines of the Journal of Electoral Studies. In order to keep students on track, final project assignments will be assigned throughout the semester (please see the details in the “Schedule” Section). These assignments will be graded and counted as a part of your final project score. Students will present their work in the last two weeks of the semester.
Midterm (20%): The exam is open book, open-note. A calculator is necessary, hopefully, one with which you are familiar (with functions no more than + - x ÷ ⎷ xy log exp M). Laptop computers are not permitted during the test. Mark your calendar now because it is very unlikely that I create make-up tests or re-schedule tests for any one person.
Policy on using Generative AI tools: It is totally fine for students to consult with Generative AI tools for learning purposes. However, for completing their assignments and the data analysis project, all the input should be the students' own original work. Violations of this policy will be considered academic misconduct.
指定書目
Agresti. 2018. Statistical Methods for the Social Sciences (5th Edition).
Long. and Freese. 2014. Regression Models for Categorical Dependent Variables Using Stata. 3rd Edition, Stata Press, College Station
McElreath. 2016. Statistical Rethinking: A Bayesian Course in R and Stan. CRC Press.
參考書目
Weisberg, Sanford. 2013. Applied Linear Regression, 4th Ed. New York: Wiley. (ALR’s Wiley page.)
Fox, John. 2015. Applied Regression Analysis and Generalized Linear Models, Third Edition. Thousand Oaks, CA: Sage Publications.
Faraway, Julian J. 2016. Extending the Linear Model with R: Generalized Linear, Mixed Effects and Nonparametric Regression, 2nd Ed. London: Chapman & Hall.
Fox, John, and Sanford Weisberg. 2019. An R and S-Plus Companion to Applied Regression, Third Edition. Thousand Oaks, CA: Sage Publications.
Gelman, Andrew, Jennifer Hill, and Aki Vehtari. 2020. Regression and Other Stories. New York: Cambridge University Press.