Through the course you will be acquainted with the fundamental theories and application of multilevel models. The focus will be on using these methods for applied research in veterinary epidemiology. You will also gain practical competency in statistical software for analyzing data.
Introduction to clustered data; linear mixed models, using software (MLwiN, Stata, R); Generalized linear mixed models; GLMM in MLWiN, Stata, R; Multilevel model diagnostics; Diagnostics in MLWiN.
Random slopes and contextual effects; Alternative approaches for discrete data; Repeated measures for discrete and continuous outcomes.
- Monday: Introduction, clustered data and linear mixed models
- Tuesday: Generalized linear mixed models and multilevel model diagnostics
- Wednesday: Visualisation of models, random slopes and contextual effects
- Thursday: Repeated measures, advanced models for discrete data
- Friday: Course wrap up, assignments.
After the course, an assignment will be handed in by each student, based upon analysing own data.
Through the course the participants will be expected to reach the following competence:
- Be familiar with the basic theories for analysing random effects and hierarchical data in veterinary epidemiology
- Be able to read and understand scientific papers using these models
- Be aware of different software options for analysing such data
- Be able to establish, fit and evaluate models for continuous and categorical data
- Be able to present results from the analyses into a format fit for a scientific publication
- Be able to use advanced epidemiological methods in interpreting research data, to better utilise observational data within an intervention context
Continuous interaction with the course teachers to provide supervision and ensure each student's progress. The students are required to actively take part in practical exercises. The final credits for the course will be given after evaluation of the post-campus assignment.
The course consists of theoretical lectures mixed with exercises. The course will provide opportunity to work on own data under skilled supervision. A review of last days topics is given every morning.
Lectures 25 hours
Seminars and tutored exercises 25 hours
Independent work 100 hours (pre-course reading and assignment 40 hours, course exercises 15 hours, POST-course assignment 45 hours)
Total 150 hours
The student needs to be familiar with the basics of veterinary epidemiology, and basic statistics including simple linear regression techniques.
Admission for NOVA courses is handled by the course organiser/ the NOVA member institution organising the course. Please see the links in the margin for more information.
Costs and other information
Please find information on costs, accommodation and other practical information here.