Multilevel modeling course

10 nov 2010, the EMEA course (master-level) 'Linear Models for Continuous Variables' will start. The popular multilevel models for clustered data will be the main theme of this course. The course will be open for (EMEA-)students as well as PhD-students (UT). A basic understanding (bachelor-level) of linear regression models is required.

Blackboard-Webpage 197300140

Course description

Data in education (and psychology, medicine, public health, sociology etc) are often clustered or show a multilevel or hierarchical structure. For example in educational research, students are nested in classrooms, classrooms in schools , schools within schoolsystems, and so on. The homogeneity of measurements or results of students in the same cluster (class or school) leads to clustered data. Observations within each cluster are correlated, which violates the standard independence assumption. As a result, conclusions from standard statistical methods are invalid.

This course provides an introduction to the use of hierarchical or multilevel models that take into account dependencies between observations. The basic ideas and theory of hierarchical linear or multilevel models will bediscussed. Real data from studies in education, psychology and social sciences will be discussed.

Topics that will be covered include an introduction to multilevel analyses, ANOVA versus multilevel analyses, random intercept and slope models, two- and three-level models, hypothesis testing, model assessment, and longitudinal (repeated measures) data.

SPSS will be used to analyse clustered data.

	GW > Latest News OMD > Multilevel modeling course
<< GW Home Latest News OMD Members of the Department Dissertations Sitemap Search	Multilevel modeling course 10 nov 2010, the EMEA course (master-level) 'Linear Models for Continuous Variables' will start. The popular multilevel models for clustered data will be the main theme of this course. The course will be open for (EMEA-)students as well as PhD-students (UT). A basic understanding (bachelor-level) of linear regression models is required. Blackboard-Webpage 197300140 Course description Data in education (and psychology, medicine, public health, sociology etc) are often clustered or show a multilevel or hierarchical structure. For example in educational research, students are nested in classrooms, classrooms in schools , schools within schoolsystems, and so on. The homogeneity of measurements or results of students in the same cluster (class or school) leads to clustered data. Observations within each cluster are correlated, which violates the standard independence assumption. As a result, conclusions from standard statistical methods are invalid. This course provides an introduction to the use of hierarchical or multilevel models that take into account dependencies between observations. The basic ideas and theory of hierarchical linear or multilevel models will bediscussed. Real data from studies in education, psychology and social sciences will be discussed. Topics that will be covered include an introduction to multilevel analyses, ANOVA versus multilevel analyses, random intercept and slope models, two- and three-level models, hypothesis testing, model assessment, and longitudinal (repeated measures) data. SPSS will be used to analyse clustered data.