140.655.01
Analysis of Multilevel and Longitudinal Data

Location
East Baltimore
Term
1st Term
Department
Biostatistics
Credit(s)
4
Academic Year
2023 - 2024
Instruction Method
In-person
Class Time(s)
M, W, 10:30 - 11:50am
Lab Times
Wednesday, 9:00 - 10:20am (03)
Auditors Allowed
Yes, with instructor consent
Available to Undergraduate
No
Grading Restriction
Letter Grade or Pass/Fail
Course Instructor(s)
Scott Zeger
Contact Name
Scott Zeger
Frequency Schedule
Every Year
Prerequisite

140.621-624 or 140.651-654

Description
This course is an opportunity to master the key ideas about the analysis of multilevel data of which longitudinal data is an important special case. Students will be exposed to the statistical models used in analysis and will get some preliminary practice in conducting multilevel analyses.
Explores statistical models for drawing scientific inferences from multilevel and longitudinal public health data. Includes topics: multilevel causes in public health, longitudinal as a leading example of multilevel data, study design, exploring multilevel and longitudinal data; linear and generalized linear regression models for correlated data including marginal, random effects, and transition models; and handling missing data.  
Learning Objectives
Upon successfully completing this course, students will be able to:
  1. Prepare graphical or tabular displays of multilevel and longitudinal data that effectively communicate the patterns of scientific interest
  2. Implement and interpret a general linear model to make scientific inferences about the relationship between response and explanatory variables while accounting for the correlation among clustered responses.
  3. Implement and interpret marginal, random effects, or transitional generalized linear models to make scientific inferences when the clustered observations are binary, counts, or non-Gaussian continuous observations
Methods of Assessment
This course is evaluated as follows:
  • 30% Midterm
  • 40% Final Exam
  • 30% Project(s)
Instructor Consent
No consent required
Special Comments

The Implementation and Interpretation of Analysis of Longitudinal Data (Wednesday 9:00 - 10:20) is highly recommended for all students. The course faculty request that all cell phones be silenced during class time out of respect for both the faculty and students. The lecture notes will be posted as powerpoint and pdf files. The course faculty feel use of laptops during class time is fine for taking electronic notes.