Special Topics in Clinical Psychology (497-0-1)

Topic

Multi-Level Modeling

Instructors

Daniel K Mroczek
312/503-7718
633 N. St. Clair, 19th Floor
Office Hours: Monday 12-2 pm

Meeting Info

Swift Hall 210: Tues 12:30PM - 3:20PM

Overview of class

This course will give students a basic grounding in the class of statistical techniques known as multilevel modeling (MLM). MLM deals with "nested" data structures. This is a common situation, such as when measurement occasions are nested within people, when children are nested within classrooms, or workers nested within companies. MLM goes by other names as well, such as hierarchical linear modeling (HLM), mixed effects models, random coefficient modeling, and random effects models. These techniques have been applied to many different research questions within the biomedical, behavioral, social sciences, and even engineering. We will focus on MLM as it is used in behavioral and social science research, with examples from education, lifespan development, and emotion research. Specifically, we will discuss in detail the three main types of multilevel models that are used in the social sciences: 1) the clustered-observations or organizational models (e.g., children nested within classrooms), 2) growth-curve or traejctory models (e.g., estimation of change with measurements nested within people over months or years), and 3) within-person variation models (daily or hourly measurements nested within people over days or weeks). Students will also learn how to use SAS Proc Mixed for conducting MLM analyses, and we will touch on using R for MLM as well. Students are assumed to have taken at least two graduate statistics courses and have a solid understanding of regression analysis.

Class Attributes

Attendance at 1st class mandatory

Enrollment Requirements

Enrollment Requirements: Pre-requisite: Student must be part of the Psychology PhD program to enroll.