Les,
> -----Original Message-----
> From: Leslie Carr [mailto:[log in to unmask]]
> Sent: 18 February 2008 16:59
> To: John Smith
> Cc: [log in to unmask]
> Subject: Re: RAE/REF research
>
>
> On 18 Feb 2008, at 16:29, John Smith wrote:
>
> > I understand why the new metrics need to 'prove themselves' by
> > showing they can consistently match the outcomes of the old
> accepted
> > system.
> Good. In other words, they correspond to what we would describe as
> our
> 'intuitions' about quality research or quality research groups.
I think I would say 'past prejudices' rather than intuitions (intuitions are very slippery and often unreliable things) :-) .
> > Yet it seems a waste of these new metrics if instead of being
> used
> > to explore new forms of assessment they are restricted to
> reproduce
> > the outcomes of the old measuring system.
> It would indeed be a shame if they were tied to reproducing the
> 2001
> or 2008 ideas of 'impact', 'significance', 'quality' or even
> 'rewardability'. But once they were accurately calibrated, we
> could
> extend them in any direction that we liked.
But are these the dimensions 'metrics' measure and does it matter?
[Incidentally we need a better name than 'metrics' since it just means some form of measurement {which begs another question :-) }, a fellow librarian suggests 'bibliometrics' but is this too close to 'book' for our computery friends?]
>
> > Imagine setting up an experiment to explore a new theory that
> > included in its design an implicit assumption that the old
> theory
> > was true.
>
> The RAE is not a scientific theory about how the world is
> constructed.
Actually it is a new theory of how to measure research (not all theories have to do with physics).
Regards,
John.
|