Day 1:
On the first day, the MCAR, MAR, and MNAR mechanisms of missing data are discussed. You will learn how to explore patterns of missing values, discover basic solutions for handling missing data, and be introduced to the Multiple Imputation procedure.
Day 2:
The second day focuses on building a well-structured imputation model. A demonstration will be given on how to apply Multiple Imputation using software such as R and SPSS. The use of Multiple Imputation for multilevel data will also be addressed.
Day 3:
On the final day, we delve deeper into practical choices when applying Multiple Imputation, such as determining the number of imputed datasets. Specific applications will be covered, including Multiple Imputation for survival data and for questionnaire data. You will also learn how to analyse data following Multiple Imputation, using various pooling methods for commonly used statistical techniques.
In addition to lectures, participants will practise extensively with epidemiological and clinical case studies involving missing data using R (Studio) and SPSS software. There will also be hands-on exercises with an AI chatbot that simulates a clinical researcher facing a realistic missing data problem from clinical research practice.
