Course content

Problems and advantages with dependent observations.
Summary measures (Area under curve (AUC), coefficient of slope, min/max value).
Covariance and correlation.
Adjustment for baseline measurement.
Variance components.
Linear mixed effect models.
Use of relevant software (Stata)

Learning outcome

After completing the course, the student should be able to:
* Understand the nature of dependency in clustered and repeated measurements, and how this dependency alters the approach to statistical analysis and modeling; as well as the consequences of not taking this information into account.

* Identify clusters and potential dependency by inspecting the design and/or viewing the resulting data set.

* Understand the principles of experimental design in which experimental factors vary both between and within clusters.

* Perform simple, descriptive analyses such as obtaining sample covariance and correlation, and corresponding graphical plots such as scatter plots, to illuminate key features of data with clustered and/or repeated observations.

* Perform and interpret variance component estimation, and adjustment for baseline value in randomized trials using analysis-of-covariance.

* Perform and interpret linear mixed regression models with random intercept, with random intercept and random slopes for covariates; using appropriate software.

* Understand the special nature of observations made along the time axis, including the possibility of autoregressive residuals.

Learning methods and activities

Lectures and guided excercises.

Course information will be published at http://folk.ntnu.no/eiriksko/KLMED8008/KLMED8008v16.html
before the course starts.

Recommended previous knowledge

KLMED8004 and KLMED8005, or equivalent.

Required previous knowledge

Master or equivalent degree in medicine, health related subjects, or social sciences. Medical students at The Student Research Programme. Candidates with a lower degree may be assessed individually.

Course materials

Textbook: Rabe-Hesketh & Skrondal (2012). Multilevel and Longitudinal Modeling Using Stata; Volume 1: Continuous Responses. Stata Press

Credit reductions