Multilevel Modeling: What it is, when you need it (and when you don't), and 4 important questions to ask every time you use it
Presenter: Amie Gordon, Postdoctoral Fellow, UC San Francisco
2121 Berkeley Way, Room 1104
Multilevel modeling (MLM) is everywhere these days. Reviewers are increasingly asking people to use this advanced approach to statistics and there are more and more online calculators devoted to helping people run MLM analyses. But MLM requires making a lot of choices, and without a clear understanding of what MLM is, it is easy to make mistakes. In this one hour whirlwind tour of MLM, I will introduce you to the topic, help you figure out when MLM is needed (and when it is not), and describe the 4 questions you should ask yourself every time you use it: (1) What is the structure of my data? (2) Are my effects fixed or random? (3) What type of centering should I use? (4) Which covariance matrices should I use?
Amie’s work explores the social, affective, and biological factors that shape interpersonal experiences. Her research methods frequently utilize dyadic and daily experience designs, which have led her to gain extensive experience with multilevel modeling. After realizing she also likes teaching statistics, Amie began to lead workshops on multilevel modeling for graduate students and faculty. She received her PhD in Social-Personality Psychology from UC Berkeley and is currently a Research Scientist at UCSF.