UCL, Bloomsbury & East London Doctoral Training Partnership - Quantitative Social Science Pathway


The Quantitative Social Science pathway is a collaboration between the Department of Medical Statistics at the London School of Hygiene and Tropical Medicine (LSHTM), the Department of Quantitative Social Science and the Centre for Longitudinal Studies, both at the UCL Institute of Education (UCL IOE).

We seek applicants for postgraduate training in the application of quantitative methods to substantive issues in the health and social sciences and/or in the development and evaluation of statistical methods.

+3 studentships are available for PhD study at the School. Applicants will have a Master's degree in medical or social statistics, or equivalent qualifications or experience.

Type of applicant

Applicants should be interested in developing and applying quantitative methods in health and/or social science, with a background either in medical statistics or quantitative social science including (but not only) economics, geography, sociology, social policy and psychology or other quantitative backgrounds such as maths or statistics.

Why study in the Department of Medical Statistics at LSHTM?

The Department of Medical Statistics is an internationally recognised centre of excellence for research into statistical methods for health and social science data, with an outstanding reputation for masters and research level teaching. Topics of expertise include missing data, longitudinal data, causal inference, and structural equation modelling, with ESRC and MRC funded research ongoing in these areas. Staff make the major contribution to the Royal Statistical Society accredited MSc in Medical Statistics, which includes a broad quantitative methods core appropriate for both social and medical data. (Please note that this MSc is not included in our ESRC-funded training pathways). 

Advanced Training

As students progress, we anticipate they will need additional training to address methodological issues that arise and to understand how related methodologies complement each other. To meet this need, students will be able to draw on extensive advanced training expertise in both LSHTM and UCL IOE to assemble a tailored programme of advanced study. Courses available include statistical analysis with missing data, statistical methods in epidemiology, longitudinal modelling, structural equation models, bootstrapping, and causal inference. In addition, all students will be expected to present their research as part of the regular Quantitative Methods PhD seminar series, organised jointly between the LSHTM and the UCL IOE.

Available topics for this pathway

We invite applications for an ESRC funded PhD project on:

The PhD would be based in the Department of Medical Statistics at LSHTM. This is a preliminary application round and the successful applicant from this stage will be taken forward to the main ESRC application stage.  Applicants will be contacted by email and informal interviews are expected to take place between 2nd and 4th January. Find out more about the project.


Example of other topics for PhD research

Examples of possible topics for PhD research include the following.  In addition, we encourage students to come with their own proposals.

  • The design of efficient, practical, longitudinal studies
  • Cluster randomised trials in education
  • The use of multiple imputation for missing data, in conjunction with sampling weights and other predictive estimands
  • Appropriate imputation strategies for missing data in complex social surveys, such as the British Household Panel Survey
  • Sensitivity analyses for non-response in social science
  • Statistical methods for understanding the effect of social networks

Examples of recent projects that have been funded via this pathway are:

  • Correcting for exposure measurement error in non-linear associations using the Bayesian family of methods, Christen Gray, Department of Medical Statistics
  • Developing strategies for handling missing data in time-to-event analyses: Incorporating variable selection, variable transformation and time-varying effects, Orlagh Carroll, Department of Medical Statistics

We welcome any topic from excellent candidates but are also especially interested in students who would like to research either of the following topics using advanced quantitative approaches.