高等數量方法

This course comprises two main sections: meta-analysis and hierarchical linear models. The meta-analysis section presents the concepts and practices of meta-analysis, with particular emphasis on Hunter and Schmidt’s approach. The hierarchical linear models section is application-oriented, covering the basic concepts and practices of HLM using Mplus software. In addition, the course introduces the use of Monte Carlo simulation for the experience sampling method (ESM).

Statistical Methods and data Analysis, Multivariate Analysis

Description of Course Details: 1. Homework and Assignments: Individual Assignment Doctoral students will complete a total of eight individual assignments, which account for 40% of the final grade, whereas master’s students will complete a total of three individual assignments, which account for 30% of the final grade. Late submissions will result in a deduction of the assignment grade. Students are encouraged to discuss ideas with classmates; however, all assignments must be completed and written independently. Plagiarism will result in a score of zero for the assignment. PhD IA Discussion Each PhD student must respond to every question and deliver a five-minute presentation. The presentation must be submitted on time via the E3 platform.  PhD 1: Artifacts issues (@3/11 1pm) 1. What are artifacts issues? 2. Please introduce the fixed-effect model and the random-effects model.  PhD 2: Meta-analysis article review (@3/20 10am) Students are required to complete the following tasks: 1. Read Bloom et al. (2025)’s meta-analysis article. 2. Prepare a 5-7 slide PowerPoint presentation with in-depth methodological analysis. 3. Record a video presentation explaining the study and your critical evaluation. 4. Submit the slides and the video.  PhD 3: Hunter-Schmidt model (@3/25 1pm) 1. Please explain what Bare Bones meta-analysis is.  PhD 4: Research questions and literature search strategies (@4/8 1pm) 1. Each PhD student must select a meta-analysis article on an OBHR topic (SSCI top-tier) and critically analyze its research question and literature search strategy. Specifically, explain what the research question addresses and why it is important, describe the article’s search strategy and process, and justify why these specific strategies were adopted. 2. Each PhD student is required to deliver a 10-minute presentation.  PhD 5: Relative weights analysis (@4/15 1pm) 1. Deliver a 10-minute presentation on the basic concepts of RWA from Jak (2000).  PhD 6: Multi-level analysis concepts (@4/22 1pm) 1. Deliver a 10-minute presentation on the basic concepts of multi-level construct from Chen et al. (2004).  PhD 7: Multi-level analysis methods (@5/6 1pm) 1. Deliver a 10-minute presentation on the multi-level analysis from Heck and Thomas (2020) (CH3: Multilevel Regression Models).  PhD 8: Multi-level analysis methods (@5/20 1pm) 1. Deliver a 10-minute presentation on the multi-level analysis from Heck and Thomas (2020) (CH7: Multilevel Structural Equation Models). Master’s IA Discussion  IA 1 (@3/20 1pm) Students are required to complete the following tasks: 1. Watch one doctoral student’s recorded presentation (PhD2). 2. Submit a 2-slide PowerPoint reflection and feedback on the presentation (e.g., key takeaways, strengths, and questions).  IA2 (@4/15 1pm) 1. Why we need to do a meta-analysis? Please provide three reasons. 2. Submit a 2-4 slide PowerPoint.  IA3 (@5/13 1pm) 1. Each student must select an SSCI top-tier article on an OBHR topic that employs multilevel analysis and prepare a 3-5 slide PowerPoint presentation. The presentation should explain (1) why the focal topic requires a multilevel construct and (2) what analytical methods were used to test the hypotheses. Each student is required to deliver a 5-minute presentation. Project Project (50%) Doctoral students will complete an individual project, which accounts for 30% of the final grade, whereas master’s students will complete a group project (up to four students per group), which accounts for 50% of the final grade. Students may choose to conduct either a meta-analysis project or a multilevel project. The term paper should report analytic results (e.g., reliability analysis, CFA, and HLM, as appropriate to the selected project type). A late project will result in a deduction of your project scores (10% each day). Please upload your assignment to E3 platform by the deadline. No plagiarism! You will share your assignment score with your classmates who have identical assignment answers to you. Impression & Participation (20%) The starting point of the class participation score is zero. In-class exercises or activities will be used as records of your participation. Attendance Policy Exams, group discussions, roll calls, or in-class activities will be used as your attendance records. Each student is exempt from 3 hours of absence with no questions asked. You will be assessed 1 point against your final grade for each absent hour. Sick leaves without diagnosis statements or medical receipts will be treated as an unexcused absence, depending on the circumstances. You will fail this class if you miss more than 12 class hours. 1. Exams and Quizzes: No. 2. Evaluation and Grading Policy: Individual Assignment; Project Report; Impression & Participation. Pedagogy and other supplementary information (websites, TAs, handouts and/or databases): Course handouts will be provided.

• Artifact Corrections in Meta-Analysis
• Hierarchical Linear Models

Schmidt, F. L. & Hunter, J. E. (2015). Methods of meta-analysis: Correcting error and bias in research findings. Sage. Kline, R. B. (2011). Principles and practice of structural equation modeling. New York: Guilford Press.

地點

教授研究室

時間

周五 13:00-15:00

聯絡方式

chenyy@nycu.edu.tw