Symplectic Elements REF2029 Steering Group : 11 March 2025

Key discussion points from the 11 March 2025 Elements REF2029 Steering Group meeting.  The slides are available here.

As most interested parties will already be aware, for REF2029 there’s no longer a ‘REF1 Researcher’ submission element. Instead, institutions will capture FTE on the Unit of Assessment (UoA) in Elements in line with HESA returns and stated averaging calculations, allowing for automated calculations of the required research outputs per UoA. However, while researchers and outputs are formally decoupled, there’s still a strong need to track staff and their UoA assignments in Elements to support reporting and management.

Previously, we considered adding a dedicated UoA field in the Elements User Data Model, but since staff can now be assigned to multiple UoAs (assigned per contract, rather than per individual), that approach is no longer practical.

CKU Research Outputs and the Uncertainty of Decoupling

A big question still hanging in the air is how staff will be linked to research outputs in REF2029. While full decoupling is a possibility, recent REF discussions have hinted at the return of maximum output limits per staff member. One of the key concerns is how to demonstrate attribution to the submitting Unit if researchers aren’t directly linked.  Another issue is the inclusion of outputs from non-academic staff (e.g. technicians). 

Given the uncertainty, the group discussed different ways institutions might populate UoAs with research outputs in Elements. Some of the suggested approaches included:

• Bulk uploading outputs directly

• Automatically pulling in outputs based on staff assignments, with an option to manually add more

• Marking outputs with eligibility statuses (e.g. ‘eligible,’ ‘submitted,’ ‘not submitted’)

Even if full decoupling happens, institutions will likely still need to report on the representativeness of submitted outputs - and this means Elements will need to keep some kind of researcher-output linking for reporting, even if it’s not officially required for submission.

The OA Monitor already supports compliance with REF2029 OA policy. However, because of differences between REF2021 and REF2029, institutions will need to set up at least two separate OA compliance schemes - possibly four, depending on embargo period management.

One tricky issue (so far) is that REF2029 has reclassified a key exception. The ‘publication actively disallows OA deposit’ exception, previously an Access Exception (REF2021), is now a Deposit Exception (REF2029). This shift could create some complexity in mapping to the correct value in the submission, so we’re waiting for clarity from the REF team before making system changes. We are also still to add the two new REF2029 technical exceptions to Elements.

Elements currently supports user-centric review & approval workflows through the Assessment Module. Based on feedback, we’re now expanding this to introduce ‘Object-centric’ review & approval workflows. This will allow for:

• Reviewing and scoring individual outputs

• Manual & automatic nomination of individual outputs for review

• Flexible review structures, including multi-reviewer workflows

• Bulk import of existing reviews to make it easier to integrate ongoing evaluations

This functionality will be delivered as a new module requiring an additional license. The first phase will focus on peer review of publications, with future expansions to impact activities and other assessment types. The exact licensing and cost structure for this module is still under review, and further details will be shared as soon as they are finalised.

We also reviewed expected development timelines, with the understanding that we’re working within the REF team’s shifting schedule:

• Object Review / Review Hub: First phase tentatively planned for v7.1 (June 2025), but exact features are still TBC.

• REF-Specific Development: CKU policy guidance expected in spring/summer 2025, with the earliest realistic release for REF-specific functionality in v7.3 (November 2025).

• Awaiting Confirmation: No word yet on when data requirements, validation rules, or the submission XML schema will be published.

Given these uncertainties, REF development will be iterative, adapting as new information emerges.

With a large group of colleagues signed up (the REF is clearly a popular subject!), we discussed whether to split into a smaller working group that would meet more frequently. The larger steering group could shift to quarterly meetings, while the smaller group would focus on the finer details of implementation.

We anticipate that the next meeting will take place following the release of CKU policy guidance.

• REF2029 Steering Group _ Mar 2025.pdf