Statistical Science


UCL/RSS Symposium on Teaching Statistics in Higher Education 2019

The Department of Statistical Science, together with the Royal Statistical Society, hosted a Symposium on Teaching Statistics in Higher Education on the 3rd April, 2019. The programme included talks by distinguished speakers in Statistics Education along with panel discussions on the talks and the future of teaching statistics.

Summary of the day

Dr Margaret MacDougall, from the University of Edinburgh, made a presentation on recent survey-based research on the statistical learning needs of undergraduate medical students. The evidence draws on experiences of medical graduates in order to identify key competencies in statistics and probability to ensure preparedness for good clinical practice. An update was also provided on the project 'Statistics in Medicine: A risky business?' which was set up with the aim of improving understanding and retention of key statistical concepts within the context of clinical practice, and enhancing informed clinical decision making based on statistically valid interpretations of research findings available in the medical literature.

Dr Charalampos Chanialidis, from the University of Glasgow, discussed the benefits and challenges of having an online MSc programme. In particular, presentation focused on a recent MSc programme in Data Analytics launched by the School of Mathematics and Statistics within the University of Glasgow. The presenter touched upon various practical aspects of designing and teaching online courses such as assessment, marketing, engagement with students, and delivery of the course content. One of the video demonstrations included Lightboard technology which enables the lecturer to face the camera while writing on a transparent surface with fluorescent markers.

Dr Elke Thönnes, from the University of Warwick, highlighted the need for undergraduate programs in statistical science to provide students with the appropriate skills to succeed in an increasingly complex data-centric world. The presentation discussed authentic learning and assessment followed by an account of the personal experiences in implementing some of these approaches in undergraduate modules in applied statistics at Warwick.

Dr Paul Northrop, from UCL, described one of the research-based learning initiatives launched by the Department of Statistical Science. This learning exercise requires groups of six students to read an allocated research paper, conduct an interview with an author of the paper, and subsequently submit a short report for formative assessment that communicates key themes of that paper to a non-specialist audience. The presenter shared his personal experience of running this assessment over the past 5 years, and discussed difficulties and challenges faced by students.


Dr Margaret MacDougall, The University of Edinburgh

Progression towards open access environments in the teaching of statistics to non-specialists in medicine and allied health sciences and promoting curriculum reform through a practitioner-focused evidence base


Firstly, I shall provide an update on the project 'Statistics in Medicine: A risky business?' that was previously funded by the Mathematics, Statistics and Operational Research Network of the Higher Education Academy and which, through further funding, has expanded to deliver open access statistics resources for non-specialist learners in statistics from Medicine and allied health sciences. Secondly, I will report on recent research on the statistical learning needs of undergraduate medical students that I led through funding from the University of Edinburgh Principal's Teaching Award Scheme. This survey-based research relies on a rich evidence base gleaned from existing medical graduates concerning which competencies in statistics and probability are required for clinical practise. It is therefore intended to support curriculum reform through introduction of learning opportunities in statistics and probability into the undergraduate medical curriculum. This session ought to inspire educators who are faced with the challenges of promoting statistical learning within non-specialist disciplines.


Dr Margaret MacDougall is the statistician at the University of Edinburgh responsible for providing advice on research design and statistical analysis to undergraduate medical students. She has conceived and managed eleven funded research projects in medical education and statistics and supervised and managed research staff and students through various spin-off projects. She has served as an invited speaker, conference organizer and chair at a range of national and international events relating to learning and teaching in higher education. Margaret uses her research as a basis for curriculum innovation and enhancing student learning at undergraduate and postgraduate levels and has published widely in the fields of education, medicine, philosophy and statistical methodology.

Dr Charis (Charalampos) Chanialidis, The University of Glasgow

Online distance learning in Data Analytics: Benefits, challenges, and suggestions


The School of Mathematics and Statistics, within the University of Glasgow, launched an MSc programme in Data Analytics two years ago. This is part-time and targeted at those who are already in employment where there is strong interest in qualifications which are viewed as improving subsequent employment prospects and an enormous demand for data analytics expertise.
This particular programme has considerable innovation in its content, management, delivery, and tutorial support. For example, a film studio has been created with a novel `light board’ to allow lecturers to be face-to-camera while writing. As a result, the work required by the staff members most closely involved has been prodigious. Any new programme/course requires substantial development effort but an online one doubly so.
In this talk, I will point out some of the benefits and challenges of having an online MSc programme in a Statistics school/department and provide suggestions for teaching online courses.


Dr Charalampos Chanialidis is currently a Lecturer at the University of Glasgow in the School of Mathematics and Statistics and the programme director of the online MSc in Data Analytics. Before that, he was a Post Doctoral Research Associate at the University of Glasgow in the School of Mathematics and Statistics and the Urban Big Data Centre, working with Marian Scott and Adrian Bowman.
Prior that, Charis was a Post Doctoral Research Fellow at the University College Dublin in the School of Mathematical Sciences, working with Nial Friel and a PhD student of Statistics at the University of Glasgow in the School of Mathematics and Statistics. His supervisors were Ludger Evers and Tereza Neocleous and the title of his thesis is Bayesian mixture models for count data.
Charis obtained his MSc degree in Statistics and Operational Research at the University of Athens. His dissertation's title is inference and variable selection for quantile regression models (under the supervision of Loukia Meligkotsidou). He obtained his BSc degree from the Department of Mathematics at the same University.

Dr Elke Thönnes, The University of Warwick

Authentic learning and assessment in applied statistics


Statistics graduates are expected to be able to problem-solve and “think with data” (Hardin, 2015), work collaboratively in a team and communicate effectively complex statistical methods both to specialist and lay audiences (ASA, 2014; QAA, 2015). Clearly these are all desirable qualities, but how do we equip our students with such competencies?
Authenticity is often raised as an essential concept in competency-focussed teaching and learning. Definitions of authentic learning and assessment vary across the literature but often refer to an alignment between learning activities and real-life contexts. In this talk I will give a brief overview of approaches to authentic learning and assessment and then describe my own experiences in implementing some of these approaches in a module in applied statistics.
ASA - American Statistical Association (2014), Curriculum Guidelines for Undergraduate Programs in Statistical Science.
Horton, Nicholas J., and Johanna S. Hardin. (2015) Teaching the Next Generation of Statistics Students to “Think with data”: Special issue on Statistics and the Undergraduate Curriculum. The American Statistician 69, no. 4.  259-265.
QAA – Quality Assurance Agency for Higher Education (2015): Subject benchmark Mathematics, Statistics and Operational Research.


Dr Elke Thönnes is a teaching-focussed Associate Professor in the Department of Statistics, University of Warwick.  She is a Senior Fellow of the Higher Education Academy and an Alumni Fellow of WIHEA, the Warwick International Higher Education Academy.
At Faculty level she acts as one of three Student Engagement Coordinators who support student representation and engagement in the enhancement of the student experience within the Faculty. Within the Department of Statistics she is currently the Director of Undergraduate Studies with particular responsibility for the Mathematics and Statistics and the MORSE degrees. She has taught a variety of modules in Mathematics and Statistics ranging from first year undergraduate to Master’s level, but more recently has focussed on the development of Applied Statistics modules with an explicit competency and employability focus.

Dr Paul Northrop, University College London

Connecting Students with Statistical Research


UCL is committed to engaging its students in the research conducted by its staff. Its Connected Curriculum aims to ensure that all students are able to participate in research-based education, exposing them to recent research and enabling a dialogue between students and researchers. We describe one of the research-based learning exercises initiated by the Department of Statistical Science at UCL. Students are put into groups of six people and assigned a research paper to read. Their task is to prepare, and submit for formative assessment, a short report that communicates the key themes of the paper to a non-specialist audience. To help them, they conduct a one-hour interview with an author of the paper. We reflect on the experience of running this assessment over the past 5 years. Feedback from students and staff has mostly been very positive. We also consider ways in which the assessment could be adapted.


Dr Paul Northrop is an Associate Professor at University College London, who joined the Department of Statistical Science in 2005.
Dr Northrop was awarded a PhD in Statistics from University College London in 1996 for a thesis on the spatial-temporal modelling of rainfall processes and extended this work as a NERC postdoctoral research fellow at University College London and Imperial College. Prior to joining UCL, he was a Departmental Lecturer in the Department of Statistics at the University of Oxford from 1999 to 2005.
Dr Northrop’s research interests lie mainly in developing and applying statistical methods for the environmental sciences, with application areas including off-shore engineering, climate science and hydrology. He has taught a variety of Statistics modules.  Currently, he teaches an introductory Probability and Statistics course to first year undergraduates and course based on one of his research interests, the modelling of extreme values, aimed at MSc students and final-year undergraduates.

Organising Committee

Dr Niloufar Abourashchi, Dr Alex Donov, Dr Simon Harden, Dr Thomas Honnor, Dr Elinor Jones, Dr Matina Rassias