Type of Credit: Elective
Credit(s)
Number of Students
Data in education, psychology, medicine, public health, sociology and other applied sciences are often clustered or show a multilevel or hierarchical structure. Data with this structure are usually dependent within each cluster and this dependency violates the assumption of independent observations which most standard statistical models and tests assumed. Hierarchical Linear Model is one of the modeling approaches which is developed to handle data with nested structure. This course provides an introduction to the use of hierarchical or multilevel models. Topics covered in this course include an overview of regression methods and models, introduction to multilevel models, random intercept and slope models, model buildings, hypothesis testing, and model assessment. In addition to the formulation and specification of hierarchical linear models, students will learn how to fit various HLMs using statistical packages in computer lab sections.
能力項目說明
The purpose of this course is to provide students with an introductory background in the basic principles and applications of hierarchical linear modeling (HLM) in psychology and educational research. The course will review both the conceptual issues and methodological issues in using hierarchical linear modeling. Computer lab sections are designed to introduce how various models can be fitted using statistical software.
| 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 Week |
課程主題 Topic |
課程內容與 指定閱讀 Content and Reading Assignment |
教學活動 與作業 Teaching Activities and Homework |
學習投入時間 Student workload expectation |
|
|
課堂 講授 In-class Hours |
課程 前後 Outside-of-class Hours |
||||
|
1 |
Overview of the course |
Overview syllabus Get to know students |
確認選課 課程內容介紹 課程進行方式說明 課程要求及評分標準 |
3 |
3 |
|
2 |
Regression |
Simple Regression Multiple Regression |
Coefficient Estimation Coefficient interpretation Variable Selection Model Evaluation Model Diagnosis |
3 |
6 |
|
3 |
Introduction to multilevel modeling |
Nested Data Structure Introduction to multilevel modeling Reading: Ch1+Ch2 |
Concept Data Structure
|
3 |
6 |
|
4 |
Related Models and Methods -1 |
Review related models and methods for analyzing nested data structure Reading: Ch3 |
Multiple regression mixed effects ANOVA |
3 |
6 |
|
5 |
Random Intercept |
Theory of Random Intercept Models Reading: Ch4 |
Model Specification Model Interpretation |
3 |
6 |
|
6 |
Computation Skills Lab 1 |
Data Setup / PROC MIXED SAS Introduction |
Analyzing Empirical Data Demo/Illustration |
3 |
6 |
|
7 |
Random Intercept and Slope Models |
Theory of Random Intercept and Slope Models Reading: Ch5 |
Model Specification Model Interpretation |
3 |
6 |
|
8 |
Computation Skills Lab 2 |
Analyzing Empirical Data
|
Analyzing Empirical Data Demo/Illustration |
- |
5 |
|
9 |
Midterm Week |
Review / Catch Up |
Review / Catch Up |
- |
6 |
|
10 |
Longitudinal data HLM |
Longitudinal Data: Multilevel Modeling Approach |
Sec01: Introduction Sec02: GLM-again |
3 |
6 |
|
11 |
Longitudinal data HLM |
Longitudinal Data: Multilevel Modeling Approach |
Sec03: WP Analysis Sec04: Model Comparison Sec05: Alternative CS |
3 |
6 |
|
12 |
Longitudinal data HLM |
Longitudinal Data: Multilevel Modeling Approach |
Sec06: RE of Time Sec07: MLMs to SEMs |
3 |
6 |
|
13 |
Longitudinal data HLM |
Longitudinal Data: Multilevel Modeling Approach |
Sec08: Estimation Sec09: WP Change over Time |
3 |
6 |
|
14 |
Holiday |
- |
- |
- |
5 |
|
15 |
Longitudinal data HLM |
Longitudinal Data: Multilevel Modeling Approach
|
Sec10: Handling Missing Sec11: Time Invariant Predictors |
3 |
6 |
|
16 |
Final Project |
Individual Presentation |
Final Report Presentation |
3 |
6 |
|
17 |
自主學習 |
課程內容回顧和統整 |
Final Course Review |
- |
5 |
|
18 |
自主學習 |
課程內容回顧和統整 |
Final Course Review |
- |
5 |
|
Students will be evaluated on the basis of homework, lab assignments, and participation
評分標準
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Snijders, T. & Bosker, R. (2012). Multilevel analysis: An introduction to basic and advanced multilevel modeling (2nd Edition). Thousand Oaks, CA: Sage. |
Moodle