November 3, 2020

Do metrics matter?

Long story short: In Australia we have a metricised assessment of research quality that universities are compelled to complete. QUT recently put in a submission that basically said they think the metrics ERA tracks are not a very good measure of research quality. QUT also contrasts the peer assessments in ERA (which have been largely stable over time) with the more metricised assessments (which have shown improvements).

Indeed, the stability of peer review assessments may be read as a commentary on the assumptions about the fitness of metrical approaches as proxies for underlying quality, particularly over time. The divergence has now widened to the point that the fitness of the metrics as proxies for quality requires empirical validation.”

Whoa. Once more for those in the back:

the fitness of the metrics as proxies for quality requires empirical validation”


Previous post
Design is about asking meta-rational questions In poorly-understood domains, science requires a meta-rational approach to induction: in this situation, what method will give a meaningful answer?
Next post
Jack Reacher loves milsurp “Reacher had very few civilian clothes, and some of them weren’t really civilian. His off-duty pants were Marine Corps khakis about thirty years