進階結構方程式模型專題
Special Topic in Advanced Structural Equation Modeling
| 節 | 週二 |
|---|---|
5 13:20–14:10 | 進階結構方程式模型專題 HA217 3 節連堂 |
6 14:20–15:10 | |
7 15:30–16:20 |
* 根據陽明交大上課時間表所列
課程內容介紹, Course Contents: Structural equation models are a class of statistical techniques that incorporate regression analysis, path analysis, confirmatory factor analysis, and full scale models incorporating both measurement and structural components. These techniques are useful for both experimental and non-experimental data; for cross-sectional datasets; for multiple-group comparisons; and for longitudinal datasets. This course will cover the advanced topics of structural equation models including: mediation and moderation in structural equation models measurement invariance testing, interaction effects between latent factors, robust estimation and multilevel structural equation models, latent growth models. 課程目標, Course Goals: There are four major goals of this course: to understand the concepts related to Structural Equation Modeling; to be able to specify your own models and analyze the data using one of the SEM programs; to be able to interpret the statistical findings to lay persons. We will be using the SEM software program including R, MPLUS (and/or AMOS) to perform the statistical analyses
You are encouraged to work with other students on the assignments. You will work with (maximum) one partners (i.e., 2 students per group) on the final presentation. Ph.D. student need to do his/her own final presentation. You may analyze your own data or data which have been collected by other individual (as long as that individual has not analyzed the data addressing the same research questions you are attempting to answer). The final presentation (15-20 minutes) should include the following four sections: introduction, method, results, and discussion. You should apply the advanced SEM techniques you learn from this course to your final project. Your group should schedule a meeting with me to talk about your final presentation at least two weeks before the date of the final presentation.
You are encouraged to work with other students on the assignments. You will work with (maximum) one partners (i.e., 2 students per group) on the final presentation. Ph.D. student need to do his/her own final presentation. You may analyze your own data or data which have been collected by other individual (as long as that individual has not analyzed the data addressing the same research questions you are attempting to answer). The final presentation (15-20 minutes) should include the following four sections: introduction, method, results, and discussion. You should apply the advanced SEM techniques you learn from this course to your final project. Your group should schedule a meeting with me to talk about your final presentation at least two weeks before the date of the final presentation.
Grades will be based on the following: a) Participation (10%, come to class, look very happy, and talk about statistical models; go to facebook (交通大學教育所-Adv. SEM應援團I) and share what you thought) b) In classroom presentation (10%) c) Assignments (40%) d) Final in-class presentation (30%) e) Final Oral Exam (10%)
| 週次 | 主題 |
|---|---|
| 第 1 週 | Brief review of SEM (matrix algebra, matrix representation of the structural equation models, mean structure and variance-covariance structure) |
| 第 2 週 | Brief review of SEM (matrix algebra, matrix representation of the structural equation models, mean structure and variance-covariance structure) |
| 第 3 週 | Mediation models in SEM (mediation effect, cross-lag model, Panel model) |
| 第 4 週 | Mediation models in SEM (mediation effect, cross-lag model, Panel model) |
| 第 5 週 | Moderation models in SEM (Multi-group analysis) & Measurement Invariance in SEM |
| 第 6 週 | Moderation models in SEM (Multi-group analysis) & Measurement Invariance in SEM |
| 第 7 週 | Introduction of multilevel structural equation models for cross-sectional data |
| 第 8 週 | Introduction of multilevel structural equation models for cross-sectional data |
| 第 9 週 | Exploratory factor analysis in SEM |
| 第 10 週 | Longitudinal data analysis in SEM (latent growth models) & Multilevel SEM for longitudinal data and latent growth models |
| 第 11 週 | Longitudinal data analysis in SEM (latent growth models) & Multilevel SEM for longitudinal data and latent growth models |
| 第 12 週 | Mixture model and Growth Mixture model |
| 第 13 週 | Mixture model and Growth Mixture model |
| 第 14 週 | Interaction effects between latent factors |
| 第 15 週 | Interaction effects between latent factors |
| 第 16 週 | Advanced topics (Structural equation models with categorical observed variables, Monte Carlo study and power analysis) |
| 第 17 週 | Advanced topics (Structural equation models with categorical observed variables, Monte Carlo study and power analysis) |
教科書: 1. Class handouts will be distributed in each class. 2. Kline, R. B. (2016). Principles and Practice of Structural Equation Modeling (4th Ed.). New York, NY: Guilford. [Required] 3. Hancock, G. R., & Mueller, R. O. (2006). Structural equation modeling: A second course. Greenwich, CT: Information Age Publishing. [Required] 參考書, Reference: 1. Brown, T. A. (2006). Confirmatory Factor Analysis for Applied Research. New York, NY: Guilford. [Optional] 2. Bollen, K. A. (1989). Structural equations with latent variables. New York, NY: Wiley. [Optional] 3. Kaplan, D. (2009). Structural Equation Modeling: Foundations and Extensions (2nd Ed.). Thousand Oaks, CA: Sage. [Optional] 4. Raykov, T., & Marcoulides, G. A. (2006). A First Course in Structural Equation Modeling (2 edition). Mahwah, NJ: Psychology Press. [Optional, E-book]
- 地點
- 教師: HA202B 助教:電資大樓7樓E04室
- 時間
- 教師:若需當面討論,請事先以Email或電話預約時間。 助教:粘美玟(Office hour: 2H, 5EF)
- 聯絡方式
- 教師:分機:31843; Email: jiunyuwu@mail.nctu.edu.tw 助教:分機:58078; Email:anke801211@gmail.com