- Dec 2017
generated by research stemming back over the previous 20 years.
There should be unique identifiers on various levels of participant inputs, so that things from REF2014 can be referenced unambiguously in REF2021.
REF also stipulates precise measurement periods for the impact of research, broadly amounting to impact occurring in the seven years between assessments, generated by research stemming back over the previous 20 years. Impact however can take many years to achieve (Hughes and Martin, 2012) and reflections on the REF 2014 have thus raised concerns over short-termism
I think the REF2021 should have a "check-in" section for assessing the "change in impact" of research reported in REF2014.