Multilevel Modeling for Design and Analysis - Online Course

ABSTRACT

The three challenges of statistics are:

• • Generalizing from sample to population.
  • Generalizing from treatment to control group.
  • Generalizing from observed measurements to the underlying constructs of interest.

Traditional statistical methods need to be updated when we move beyond simple models of random sampling, constant effects, and accurate measurements. In this seminar, we consider general challenges of design and analysis in a world of nonrandom samples, varying treatment effects, and noisy data.

A key tool here is multilevel modeling, which is designed for structured data such as voters within states, students within schools, and grouped or clustered designs in surveys and experiments. It’s also useful in settings with more than one source of uncertainty—for example, causal inference with varying treatment effects.

This Distinguished Speaker Series seminar will consist of three hours of lecture, held live* via the free video-conferencing software Zoom. Each hour will include a mix of examples, methods, and discussion, including responding to your questions.

The first hour sets up a conceptual framework for thinking about inference and generalization in the context of multiple sources of variation. The second hour covers multilevel modeling as it applies to sampling and causal inference. The third hour gets into open questions in multilevel modeling and generalization. We will discuss many examples, mostly in social science but some in biology and medicine.

The course is not hands-on, but it will include analyses using R and Stan.

*The video recording of the seminar will be made available to registrants within 24 hours and will be accessible for four weeks thereafter.  That means that you can watch all of the class content and discussion even if you cannot participate synchronously. 

Closed captioning is available for all live and recorded sessions. Live captions can be translated to a variety of languages including Spanish, Korean, and Italian. For more information, click here.