KEF2: Return of the KEF

What's going on with the Knowledge Exchange Framework?

James Coe is Associate Editor for research and innovation at Wonkhe, and a partner at Counterculture

This week Research England has set out its final decisions on the second version of the KEF – the sequentially titled KEF2.

An explainer sets out how design, metrics, methodology, timescales, and narratives, will be tweaked for this latest version.

As a reminder, the KEF is the sector-wide exercise to capture the quality of knowledge exchange taking place. In the broadest sense, this is how institutions take the research from inside their walls to the outside world to make social change. Seven activities are measured – these are; research partnerships; working with business; working with the public and third sector; skills, enterprise, and entrepreneurship; local growth and regeneration; intellectual property and commercialisation; and public and community engagement. There is no KEF submission as such but scoring is based on existing data and short narratives by universities. In turn, universities are then grouped into clusters to provide comparisons to other universities.

There is much in KEF2 that remains identical to the OG KEF. It is refreshing to see consistency  in an exercise which is still in its infancy. However, some changes include:

  • changes to the metrics which underpin the seven areas. These are relatively technical but it’s interesting to see the inclusion of trade journals within the co-authorship metrics under research partnerships. Equally, under IP and commercialisation there is an additional metric on average external investment for spin-outs that survive more than three years,
  • The purpose of KEF is to measure the strength of knowledge exchange and it’s therefore unsurprising to see significant attention paid to how KEF results are presented and labelled. Most of what appeared in KEF survived but in KEF2 results will be presented in quintiles rather than in deciles. In effect, quintile five will represent very high engagement and quintile one will represent very low engagement.
  • There are some changes to the methodology section which are incredibly technical. Following feedback from the KEF process there has been a move away from employing a scaling step in calculations to control for outliers toward a three year average, which will control for variable results over a longer time period.

The KEF has previously been characterised as the third leg of the higher education excellence stool along with the TEF and REF. Currently, it lacks some of the profile and attention paid to the TEF and REF by the sector more broadly. This could be in part because of funding mechanisms, its recency, or because of the relative complexity of metrics.

There may also be an issue around how incentives are built into the higher education system. As has been argued elsewhere on this site it still feels too broad and too metrics driven to capture the essence of knowledge exchange in universities. This review document already hints toward discussions for KEF3 (the KEF and the last crusade?), and perhaps this could be the opportunity for a much wider debate on not just whether metrics, methodology, and design are right, but whether the KEF in its current form is capable of capturing the true social value of knowledge exchange.

One response to “KEF2: Return of the KEF

  1. I agree it is far too metrics driven, and the social impact, and value are not often apparent. The overlap with REF impact is understated, and, taking a leaf out of REF impact could help KEF. Also understated are sufficient details on how the environment created the KEF outcomes, especially around social capital, and dare I say it,, social mobility.

Leave a Reply