Missing Data in Mental Health
Date: 19th March 2018
Venue: Computer Room C, Main Building. Institute of Psychiatry, Psychology & Neurosciences (IoPPN), King's College London. View Map.
The target audience of this course is mental health researchers needing to analyse incomplete data. Participants should be familiar with running Stata from the command line (i.e. not using menus) at least to the level of fitting a multiple regression model to complete data. No prior knowledge of missing data methods is assumed.
Participants will need to bring their own laptop computer with Stata version 13 or higher. Those with KCL computer accounts should be able to access Stata version 1.2 by running Stata.exe from the network drive.
Cost and Booking
- External £100
- KCL Staff £75
- KCL Student £50
- Other Student £75
- King's Health Partners £75
That is, 50% discount to King's College London PhD students, 25% discount to other students and staff at King's College London and King's Health Partners.
Booking for this course has now closed.
Professor Sabine Landau (Academic Lead)
Professor Andrew Pickles (Academic Lead)
Sabine is a Professor of Biostatistics and Lead of the Causal Modelling Group. She joined the Institute of Psychiatry as Lecturer in 1997 after early work in agricultural research (Rothamsted Research Harpenden) and obtaining a PhD in Applied Statistics from the University of Nottingham.
Her interest is in developing and applying statistical methods for behavioural research. Past methodological work has been on a variety of topics, including Cluster Analysis and the Analysis of Spatial Cell Patterns. Her current focus is on Causal Inference, in particular the development of methods and their application to learn more from clinical trials and to fully exploit information provided by observational data sources.
Causal modelling has an important role to play in advancing behavioural research methodology and can lead to new substantive insights. For example relevant techniques can be used to address questions about treatment mechanisms in trials. (How and for whom do treatments work?). A causal mediation analysis of a parenting programme led to a better understanding of the parenting components changed by the programme and translated into an improvement in child conduct. Another route of enquiry is the specification of conditions under which psychological therapies work. Sabine is involved in projects that develop methods for assessing treatment effect modification by post-randomisation variables (therapeutic alliance, compliance, contamination) to address such questions.
Sabine has extensive experience and expertise in Biostatistics and Trials. She currently serves as the senior trial statistician for a number of trials, in particular for the evaluation of complex interventions (psychological therapies). Most of her trials work is NIHR funded and affiliated to the King’s Clinical Trials Unit. She also supports the NIHR by serving on their funding panels. She is currently the statistics editor for the Journal of Child Psychiatry and Psychology.
Sabine is also active in education, supervising MSc and PhD students, co-leading the Research Methods and Statistics module for the MSc in Forensic Mental Health, running a number of advanced statistics courses for IoPPN staff and postgraduate students (including multilevel modelling, missing data) as well as leading a summer school in Causal Modelling and Evaluation.
Sabine is the departmental representative of the Diversity & Inclusion Self Assessment Team (D&I SAT) and a member of several IoPPN committees. She also acts as the department’s postgraduate research (PGR) lead.
See Sabine's research portal here.
From an early background in natural sciences, urban planning and geography Andrew joined the MRC Child Psychiatry Unit as statistician in 1986, moved to the Dept of Biostatistics in Manchester in 1998, returning to the IoPPN in 2010.
His principal interests have been in multivariate longitudinal data, with most applications being in what is now called life-course analysis. These studies have involved genomic, physiological, psychological and social measures and processes and have spanned neonates to geriatrics. A good part of this work has focussed on autism, with recent work highlighting intervention studies and longer term outcomes.
Equally rewarding have been the impact on statistical modelling and software (e.g. structural equation modelling in Stata) and the clinical impact on patient care (e.g. NICE guidelines).
See Andrew's research profile here.
Your place will not be confirmed until payment has been made. Failure to cancel without sufficient notice will forfeit your course fee and access to future courses. If you would like to pay by internal transfer, please contact email@example.com