Knowledge Exchange Framework (KEF)

The Knowledge Exchange Framework (KEF) is Research England’s national exercise for assessing and benchmarking how English higher education providers perform knowledge exchange (KE) activity — the two-way sharing of knowledge, expertise and other assets between universities and non-academic partners for economic and societal benefit. It is administered by Research England, part of UK Research and Innovation (UKRI), and is one of three linked national assessment exercises alongside the Research Excellence Framework (REF), which assesses research quality, and the Teaching Excellence Framework (TEF), which assesses teaching. KEF is distinct from both: it does not score research outputs or teaching provision, it scores the breadth and intensity of a provider’s engagement with business, the public sector, civil society and local communities.

Who is in scope

KEF’s population is defined by Higher Education Innovation Funding (HEIF) eligibility: providers that receive a Research England HEIF allocation in the relevant academic year are included in the exercise. This ties KEF conceptually to knowledge exchange (KE) more broadly — HEIF is the funding stream that supports KE activity, while KEF is the accountability and benchmarking mechanism that reports on it.

The seven KEF perspectives

Each participating provider’s KEF dashboard reports performance across seven broad perspectives, shown as a polar area (radar-style) chart rather than a single composite score:

1. Research partnerships
2. Working with business
3. Working with the public and third sector
4. Continuing professional development (CPD) and graduate start-ups
5. Local growth and regeneration
6. IP and commercialisation
7. Public and community engagement

Underlying each perspective is a set of quantitative metrics (largely drawn from the Higher Education Business and Community Interaction, HE-BCI, survey and other HESA data) and, for some perspectives, narrative statements submitted by the provider.

Clusters and quintiles, not a single ranked league table

Because a specialist art college, a large research-intensive university and a regional teaching-led institution engage in fundamentally different kinds of KE activity, KEF does not compare all providers against one another directly. Instead, Research England groups providers into peer clusters of institutions with broadly similar characteristics — scale of research activity, subject mix, and institutional type. Within each cluster, a provider’s performance on each perspective is shown as a quintile (one of five performance bands) relative to its cluster peers, not as an absolute rank. Research England is explicit that clusters themselves are not ranked against each other, and KEF is designed as a benchmarking and self-improvement tool rather than a competitive league table in the way REF results are sometimes treated.

Relationship to REF, TEF and HEIF funding

KEF is often described as the third leg of a national assessment triad alongside REF (research) and TEF (teaching), but the funding mechanics differ meaningfully. REF outcomes directly inform the selective allocation of quality-related (QR) research funding. KEF does not work the same way: HEIF allocations are calculated through a separate formula based primarily on a provider’s KE-related income over a rolling multi-year period (drawn from HE-BCI and related HESA data), not from KEF quintile scores. KEF participation is a condition tied to receiving HEIF funding, and KEF’s evidence base has been used to inform policy thinking about how KE activity might be funded in future, but as of the most recent cycles KEF results do not themselves determine the HEIF funding formula.

Cadence

KEF has run in successive annual or near-annual cycles since its first full exercise (KEF1, 2021), with providers and cluster boundaries periodically refreshed using updated REF and HE-BCI data. Research England publishes cycle-specific guidance (circular letters) ahead of each round covering data-return requirements, timelines and any methodology changes.

Worked examples

• A research-intensive university’s engineering faculty compiles data on IP licensing income, contract research, spin-outs and CPD short courses as part of its institution-wide KEF return, and receives a quintile position on the "IP and commercialisation" and "working with business" perspectives relative to its assigned peer cluster.
• A smaller, teaching-led provider that receives a HEIF allocation submits KEF data on local growth and regeneration and public and community engagement, allowing it to benchmark against similarly-sized, similarly-focused peers rather than against a research-intensive comparator with a very different KE profile.

What KEF is not

• It is not a research-quality assessment — that is REF’s role, using peer review of outputs, impact case studies and environment statements.
• It is not a teaching-quality assessment — that is TEF’s role.
• It is not, on its own, the mechanism that calculates HEIF funding allocations, even though HEIF-recipient status determines who is in scope for KEF.
• Its quintile bandings are not a single sector-wide ranking; a quintile is only meaningful relative to a provider’s assigned peer cluster.

References

• Research England / UKRI, Knowledge Exchange Framework (kef.ac.uk) — About the KEF, perspectives and clustering methodology.
• UKRI, "Knowledge exchange framework" overview and KEF5 circular letter (ukri.org).