Share Description
Power and Sample Size for Longitudinal and Multilevel Study Designs is a five-week, entirely online course that covers innovative, research-based power and sample size methods, as well as software for multilevel and longitudinal studies. This course's power and sample size methods and software can be applied to any health-related or, more broadly, social science-related (e.g., educational research) application. All of the examples in the course videos are from real-world behavioural and social science studies that used multilevel and longitudinal designs. The course philosophy is to concentrate on the conceptual knowledge required to conduct power and sample size calculations. The course's goal is to teach and disseminate methods for selecting an appropriate sample size, culminating in the creation of a power/sample size analysis for a relevant research study in your professional context.
Learners will be able to :
- Use a framework and strategy for study planning;
- Write study objectives as testable hypotheses;
- Describe a longitudinal and multilevel study design;
- Write a statistical analysis plan;
- Plan a sampling design for subgroups, such as racial and ethnic groups;
- Demonstrate the feasibility of recruitment.
- Create a power and sample size analysis that corresponds to the planned statistical analysis.
Syllabus :
1. Introduction to Multilevel and Longitudinal Designs
- Course introduction and overview
- Review of basic statistical concepts
- Introducing longitudinal studies
- Studies with a single level of clustering
- Studies with multiple levels of clustering
- Multilevel and longitudinal studies
2. Foundations of Complex Multilevel and Longitudinal Designs
- Within and between independent sampling unit factors
- Understanding the hypothesis
- Power and type I error
- Choosing the test
- Correlation structure
3. Model Assumptions, Alignment, Missing Data, and Dropout
- Model assumptions
- Alignment of power and data analysis
- Predicting missing data and dropout
- Accounting for missing data and dropout
- Continuous, binary and Poisson outcomes
4. Inputs to Analysis, Recruitment Feasibility, and Multiple Aims
- Inputs for power analysis: Literature review
- Internal pilot studies
- Planned pilot studies
- Studying power via simulation
- Demonstrating recruitment feasibility
- Handling multiple aims
5. Ethics and Using Power and Sample Size Analysis to Get Funded
- Ethics of power and sample size
- Writing the sample size section for your grant
- Graphics for power and sample size
- Power for subgroup analysis
- Getting funded
