PGCAP Project: Using Students as Editors and Co-Creators to Improve the Accuracy of Captions in SED

My action research project was concerned with improving the provision of captioning for recorded media in the School of English and Drama. It addressed the challenges of meeting new requirements in accordance with the Public Sector Bodies (Websites and Mobile Applications) (No. 2) Accessibility Regulations 2018.

To be compliant with these regulations, public sector bodies (including HE institutions) should be accurately captioning all recorded media on their websites (including lectures on Virtual Learning Environments). To be compliant captions need to be 99% accurate. However, despite legal requirements and the clear benefits that captions bring to diverse groups (including, but not limited to, disabled students) research shows that many HE institutions have struggled accurately captioning recorded media.

Producing captions is resource-intensive and many HE institutions rely on the use of automated captioning software as the first part of the process. However, there are issues with the quality of automated captioning. During the pandemic I worked as a lecturer and as SED Disability and Accessibility officer, and through my own experience and discussion with colleagues I identified several obstacles to captioning recorded lectures:

1) accuracy: usually only 80-90% accurate, and software tended to miss key technical vocabulary

2) workload: editing ten minutes of lecture material averaged an hour, an unsustainable addition to lecturers’ workload

3) technical challenges with using caption editing tools:  These limitations of auto captioning tools have been recognised and debated in key HE forums, These discussions clarify that auto captioning content will not make it fully compliant, and that a level of human editing and intervention will be required.

Sector-wide discussions reveals uncertainty as to where responsibility for resourcing editing lies: with academics, schools or institutions?  At a local level, I was concerned with how this lacuna in provision means many students do not have full access to recorded media at QMUL. 

One exciting solution I encountered in research is to incentivise students to support with the editing process. This has been the inspiration for my PGCAP action research proposal, which investigates how we can improve accessibility of online media by using appropriately-skilled and knowledgeable students as co-creators to edit and improve captions produced by automated software for modules in SED. Students will be incentivized to volunteer through the creation of a special ‘SED Accessibility Champion’ HEAR award and the opportunity to be sponsored for a QMUL SEED award. Recruitment materials for volunteers would also emphasise the academic benefits and opportunity of engaging more closely with learning materials. An important potential benefit is making participating students feel empowered in creating an inclusive learning environment. This also aligns with strategic priorities in the ‘Education and Student Experience’ section of Strategy 2030, including developing opportunities for students to be co-creators. I have been working with Jonathan Boffey, digital learning technologist in SED, to begin testing workflows and we are excited to expand and evaluate this project in 2022-23.

My experience of the PGCAP, and ADP7219 in particular, have been really valuable in helping me to materialize my idea into action this year, and provided me with exciting new frameworks to take the project forward.

About PGCAP projects

PGCAP is the Postgraduate Certificate Academic Practice, delivered by the Queen Mary Academy to staff teaching Queen Mary students.

The final module of PGCAP is Action (practitioner) Research. As part of this module, participants develop an Action Research Project Proposal. This is an excerpt from one such project proposal.