Missing data? The TARMOS framework for accessible, reproducible analysis
Abstract
Although there is increasing guidance on how to handle missing data, practice is changing slowly and misapprehensions abound. Importantly, the lack of transparency around methodological decisions is threatening the validity and reproducibility of modern research.
I will describe how the recently published ‚Treatment And Reporting of Missing data in Observational Studies‘ (TARMOS) framework seeks to address these concerns, arguing that multiple imputation (which can be used in an ever increasing range of settings) provides a natural tool to implement this framework. The need to assist analysts with implementing multiple imputation has motivated recent work developing ‘mi.doc’ – an expert system in the statistical software R, which I will briefly demonstrate.
The ideas will be illustrated with an analysis of data from the UK Avon Longitudinal Study of Parents And Children (ALSPAC), exploring whether there is a causal link between smoking at 14 years and educational attainment at 16 years.
References:
Carpenter, J. R., Bartlett, J. W., Morris, T. P., Wood, A. M.,Quartagno, M. and Kenward, M. G. (2023) Multiple Imputation and its Application (second edition). Wiley.
Lee, K. J., Tilling, K. M., Cornish, R. P. Little, R. J. A., Bell, M. L., Goetghebeur, E., Hogan, J. W. and Carpenter J. R. on behalf of the STRATOS initiative. Framework for the treatment and reporting of missing data in observational studies: The Treatment And Reporting of Missing data in Observational Studies framework. Journal of Clinical Epidemiology (2021), vol 134, pages 79–88.
Vortragender
James Carpenter is professor of medical statistics at the London School of Hygiene & Tropical Medicine and MRC Investigator in Trials Methodology at the MRC Clinical Trials Unit at UCL, London. His principal research interests are coping with missing data in clinical trials and complex hierarchical models (especially using multiple imputation methodology), estimands and sensitivity analysis, meta-analysis and novel trial designs.
Termin
Datum: Mittwoch, 20 September
Beginn: 10:00 – 11:00 Uhr
Ort: AULA