Linear mixed modeling of treatment/intervention studies
Ph.d.-kurs
- Studiepoeng
- 2
- Undervisningssemester
- Haust, Vår
- Emnekode
- CDP991
- Talet på semester
- 1
- Ressursar
- Timeplan
Emnebeskrivelse
Læringsutbyte
On completion of the course the student should have the following learning outcomes defined in terms of knowledge, skills and general competence:
Knowledge:
Knowledge about main analytical approaches to treatment/intervention studies with repeated measurements including random effects models, covariance structure models
Knowledge about the application of linear mixed models in current research
Knowledge about data assumptions, and diagnostic criteria for evaluating of model assumptions
Knowledge about ways to express effect size in linear mixed models
Knowledge about tools for visualization of treatment effects
Skills:
Skills to manage data in suitable form for longitudinal analysis with standard software
Produce an analysis plan that correspond to relevant research questions in linear mixed models
Skills to perform analysis of treatment effects and/or change using standard software in studies with repeated measures
Skills to interpret output from analysis of linear mixed models
Skills to visualize individual differences, and overall treatment effects
General competence:
Ability to use linear mixed models in treatment/intervention research
Ability to interpret research findings in published research using linear mixed models
Undervisingsperiode
Studiepoeng
2 ects
All together 18 hours with teaching
Undervisingsstad
Undervisingsspråk
Påmelding og -fristar
Krav til forkunnskapar
Obligatoriske arbeidskrav
Vurderingsform
Kven kan delta
Primarily doctoral candidates. Others may apply. Level of study: PhD
ECTS will be only be given to candidates with admission to a doctoral education at the University of Bergen.
Utfyllande kursomtale
The course content deals with modern approaches for modeling treatment effects in multi-wave treatment/intervention studies. The course highlights key dilemmas and data analytic decisions from a user perspective.
The course objective is to provide a learning situation where participants gain experience with data management, key analytic decisions, and proper interpretations of results from longitudinal treatment/intervention studies. Topic include:
- Data management and data quality assurance in treatment/intervention studies
- Longitudinal data analysis
- Mixed-linear models of treatment/intervention effects
- Multilevel model of change in treatment/intervention studies
- Sample size, power and effect size
- Missing data assumptions, intention-to-treat analysis
- The complete workflow of longitudinal data analysis: Case studies
TEACHING METHODS
The 3-day course will be structured as a series of lectures (12-15 hs), lab sessions (6hs) and self-study (24hs) with relevant exercises. Participant are expected to bring their own laptop with SPSS installed.
The workshops include analysis of didactic data sets, but participants are encouraged to bring in their own datasets.