Ofqual and data-sharing

The 2020 summer GCSE and A level result period was challenging for everyone, particularly for young people awaiting the outcome that would decide their immediate future. Whatever one thinks of the methods employed to determine grades, of the policy changes that occurred or of the final outcome, everyone would agree that the process was less than optimal.

No doubt there will be many post-mortems of the period. Academics will have a fruitful time teasing apart the events and learning what went wrong and why. Trust in the system that awards GCSEs and A levels has clearly been damaged. There is a perception that the system does not work properly and that there is secrecy over the data and processes that drive the awarding of exam grades. Rebuilding trust in the system that determines young people’s futures is essential and urgent.

I was therefore pleased when Amanda Spielman, Chief Inspector of Schools, and Dame Glenys Stacey, Interim Chief Regulator, asked me to chair an expert group looking into how data on schools and young people’s performance could be better and more widely shared. Our aim is to improve research that enhances the quality of our qualifications system.

We must give the public confidence that the right lessons are learned from summer 2020. We must have high quality evaluation of the judgements made in awarding grades, carried out from a variety of perspectives. We must make sure that the evaluation of school leavers’ achievements properly informs university and college entrance policy. And, finally, we need stronger relationships to be built between the regulator and the statistical, policy and education research communities.

This is an ambitious agenda. It should not just be about summer 2020, but about creating better data sharing and a better way of ensuring the right research questions are asked. In this way, teachers can be better informed about the outcomes of their teaching, and help to guide school leavers to the right course, university, college, apprenticeship or job.

Data is at the heart of the issue. We need to make sure that the right data is made available to the right researcher in the right format. Modern data mining is very powerful: linking data-sets allows more powerful and wider analyses. Many of the data-sets we would like to link are “owned” by different groups and may use different definitions for variables, so there are technical challenges. There must be safeguards in place as well. We need to protect the identity of individual students, and there must be caution whenever analyses identify small groups, individual schools, geographical areas, or personal characteristics.

To begin to address these issues, Ofqual has put together a group of external experts to act as an advisory board to a data-sharing project which involves Ofqual, Ofsted, UCAS and ONS.

The data-sharing advisory board members are:

Iain Bell, Deputy National Statistician at the Office for National Statistics (ONS)

Nick Hillman, Director of the Higher Education Policy Institute (HEPI)

John Jerrim, Professor of Social Statistics at UCL Institute of Education

Laura McInerney, Chief Executive and co-founder of Teacher Tapp

Sandra McNally, Professor of Economics at the University of Surrey

Natalie Perera, Chief Executive of the Education Policy Institute (EPI)

Dave Thomson, Chief Statistician at FFT Education Datalab

Professor Julius Weinberg, Chair

This advisory board that I chair will look at the research questions that could be asked and the methods used, the data that could be shared and the principles to determine access to the data. We’ll need to look at promotion of the use of data to researchers and how to make sure the findings are made broadly available to practitioners and policy makers. This is how we can have impact, and improve the future of young people.

The group held its first meeting, which was lively, challenging and achieved its most important purpose. Ofqual attendees said external participants raised issues they had not thought of and helped them to reorder some of the priorities. That is the reason for having such a group: there would be little point if it did not change or influence things. It was good having a range of voices from statistical research to policy wonks interrogating the assumptions that had been made in setting the project up.

We recognise that there are many people whom we would like to have on the group. There is a tension, however, especially when trying to do things at pace and hope to report in May, between having a small group and a larger more representative one (particularly when trying to have a lively participative meeting as a virtual conference call). We will devise ways that people can contribute to our thinking. We may bring in others to add their voice in areas we feel could otherwise be neglected, or bring others in to transient working groups we set up around particular issues.

Determining how data might be shared more effectively sounds fairly dry. Yet I think this project could effect some fundamental change. It is primarily about trust, openness and accountability. Trusting others to use and interrogate data to determine how well our exam (and education) system is performing. Allowing independent researchers to hold the system to account. Being confident and open. Ultimately openness and accountability will, I hope, rebuild the trust that young people, their parents, teachers and others should have in the exam system.

If you would like to know more about this project then please do contact us at: [email protected]