Contents
UKRN Academy
Find out more about our PhD students and their projectsOrganisation and membership
The UKRN Academy connects doctoral students at UKRN member institutions who are working on topics related to reproducibility. It was initially formed of students working on projects directly funded by UKRN institutions as part of their commitment to the UKRN, but is open to all doctoral students in the UK working on research reliability, transparency, metascience or other meta-research / research-on-research topics. The UKRN contact for the Academy is Tom Stafford, institutional lead for the University of Sheffield.
PhD student profiles
Profiles are ordered alphabetically by first name.
Aoife O'Mahony
Cardiff University
Creating a comparative database of registered reports
My project involves creating a comprehensive database of stage 2 Registered Reports and comparing their characteristics with regular research articles. This will clarify how Registered Reports differ from standard research articles and to what extent they are associated with indicators of greater rigour and transparency. These insights can be used to inform journal policies.
Supervisors: Chris Chambers, Candice Morey
Ben Meghreblian
Cardiff University
Metascience and tool development for Registered Reports
The project aims to design, build, and evaluate tools to encourage adoption of the Registered Reports publishing format by journals and authors – this includes:
1) community tools for channelling advocacy, and collating feedback from authors and reviewers; 2) a study design template, aiding authors’ submissions, with elements linked in a standardised, structured format; 3) a policy builder for journal editors
Supervisors: Chris Chambers, Loukia Tzavella
Christopher Osborne
University of Oxford
How can access to research data be enhanced?
Research findings should be verifiable. At a minimum, this means that the data used to produce them should be accessible. However, evidence suggests that data are often inaccessible, limiting the extent to which findings are verifiable. To address this, I will operate within a behaviour change framework to explore how data can be made more accessible, and so findings more verifiable.
Supervisor: Laura Fortunato
Emma Wilson
University of Edinburgh
Developing a meta-research framework for preclinical models of monogenic ASD
I aim to develop an online living evidence summary of research related to rodent models of ASD/ID, using machine learning and natural language processing to automatically classify and extract data. I will then use this resource to investigate the robustness and reproducibility of research in this area, and recommend actions for improvement.
Supervisors: Emily Sena, Peter Kind
Katie Drax
University of Bristol
Investigating tools to improve the reliability & validity of biomedical research
Studying how tools can help researchers increase the quality of their research. Projects to date include a systematic review of reporting quality in UK Biobank studies, a crowdsourced data analysis, and a qualitative study examining the experiences of stakeholders involved in funder-Registered Report journal partnerships.
Supervisor: Marcus Munafo
Mark Gibson
University of Bristol
Exploring factors that influence the replicability of results generated by large population-based cohort studies
My aim is to investigate the current levels of transparency and replicability in epidemiological studies which use large population-based cohorts (with a focus on Mendelian randomisation) and outline the sources of analytical flexibility and poor practice which reduce replicability.
Supervisors: Marcus Munafo, Rebecca Richmond
Qianying Wang
University of Edinburgh
Biomedical natural language processing for preclinical evidence synthesis
My research focuses on developing tools to automatically identify risk of bias and PICO phrases in preclinical publications, which helps to speed the process of systematic review, provide information to guide research improvement, and support translation from preclinical to clinical research. The tasks involving biomedical text classification, question answering, named entity recognition and other natural language processing techniques.
Supervisor: Malcolm Macleod
Robbie Clark
University of Bristol
An exploration of the role and utility of the philosophy of science in psychology research
My aim is to understand how different philosophies of science and epistemologies shape research conduct and data synthesis in psychology research. My ultimate goal is to develop a coherent framework which reconciles the Popperian ideals of falsification with the human disposition towards theories of confirmation and inference to the best explanation.
Supervisors: Marcus Munafo, James Ladyman
Zuzanna Zagrodzka
University of Sheffield
Evaluating the perceptions of bias, replication and transparency in evolution and climate change science
My research focuses on exploring the perception of bias, replication and transparency in evolution and climate change science. Looking at the scientific landscape of researchers, publishers, technologies, repositories, funders, institutions – their policies, their visions, their products, their incentives – to calculate the impact of changing open research policies and areas for improvement.
Supervisors: Andrew Beckerman, Anna Krystalli