Multilevel Analysis of Multiple Indicators Cluster Survey MICS2, Sudan

Multilevel models are models specifically geared toward the statistical analysis of data that have a hierarchical or clustered structured. Such data arise routinely in various field, for instance in education research, where pupils are nested within schools, family studies with children nested within families, and medical research with patients nested within physicians or hospitals. Clustered data may also arise as a result of specific research design. For instance, in large scale survey research the data collection is usually, for economic reasons, organized in some sort of multistage sampling design that results in clustered or stratified design. Another example is longitudinal designs; on way of viewing longitudinal data is as a series of repeated measurements nested within individual subjects.