Measurement

Cross-Cutting

Instructional Course

(600785) IC 8: Longitudinal Data Analysis and Practical Workshop using R: Part II Advanced Topics

Presenter(s)

• 

  Allan Kozlowski, PhD, BSc (PT)

  Director of Outcomes Research
  Mary Free Bed Rehabilitation Hospital
  Grand Rapids, Michigan
• 

  Keith Lohse, PhD

  Assistant Professor
  University of Utah
  Salt Lake City, Utah

Rehabilitation researchers and clinicians often deal with outcomes that evolve over time and are not suited to binary categorization of pre- post-assessments, as is common in other areas of medicine and healthcare. This course will build on our previously offered introduction to modeling longitudinal outcomes in a lecture and practical workshop format. Advanced topics include methods to assess data and model structure visually, statistically, and conceptually; and to fit more complex models and outcome types. While not a prerequisite, course participants will benefit from having a working knowledge of the introductory course content and of the R environment.

Learning Objectives:

• Describe the structure of curvilinear and non-linear (e.g., negative exponential) models.
• Examine relationships of data to model fit by examining data patterns over time.
• Discuss the use of fixed and time-varying covariates in curvilinear models.
• Fit models continuous-curvilinear, non-continuous (e.g., dichotomous), and non-linear (e.g., negative exponential) outcomes in the R statistical environment.