Missing data in multilevel models

Missing data is an unavoidable feaure of most datasets. Many datasets also exhibit multilevel structure with either data collected from clusters of indivudals or with several observations collected on individuals over time. Imputation of missing values helps to extract maximum value from this highly structured data, but more research is needed on the best way to impute whilst respecting the multilevel structure of the existing data.

Members

Dr Alice Richardson

Associate Professor Cate De'ste

Nidhi Menon

Updated:  20 September 2017/Responsible Officer:  Director/Page Contact:  Webmaster