berezin at MCMAIL.CIS.MCMASTER.CA
Wed Dec 27 20:03:46 EST 1995
On 27 Dec 1995, Keith Robison wrote:
> Alexander Berezin (berezin at MCMAIL.CIS.MCMASTER.CA) wrote:
> : BEREZIN:
> : Dear Kathy, Bill and others:
> : We generally can't do too much to change people. All we
> : can't do is to change (improve) the SYSTEM to the point that
> : the (partcicular type) of unethical behavior will not be
> : of a great ('Darwinian') advantage.
> : In scientific research so far the grantsmanship, secrecy, etc
> : all encourage this (unethical) type of behaviour as the
> : potential rewards of it still outweight the risks.
> : To change that, we need to change the reward system.
> : For example (no matter how crude this scheme looks), let's
> : take the folowing count:
> : Professor A ("fat cat"):
> : Finding $ 500,000 per year
> : Group produces 10 papers per year
> : Yield: 1 paper for $ 50,000
> : Professor B ("small guy")
> : Funding: $ 50,000
> : 3 papers per year
> : Yield: 1 paper per $ 17,000 (3 times better)
> : Present reward system ignores this (the 3 times better
> : efficiency) and by all categories (promotions, awards,
> : fame, etc) Prof A will be counted as superior, while
> : small fish will likely be spit upon.
> : Should we take YIELD (cost of paper) as a prime parameter,
> : the reward system will be immensely better and more
> : efficient. (I can foresee all the screams of "fat cats"
> : on such a proposals: of course, it is possible to improve
> : it beyond simple paper counting, but even the above would be
> : a step forward).
> Now you are descending into self-parody! Under such a system, the
> drive to publish lots of weak papers is immensely increased.
No, not at all. You are forgetting that I am not saying
that simple paper counting in the end. Paper assessment
(including, but not limited, to citations in Sci.Cit.Index)
can be included as well. Impact assessment (relative
to dollars spend) will NOT encourage many short/weak papers.
Rostum Roy of PennState wrote several papers how all this
can be made a viable scheme, much better than present APR
yes/no scheme. It is the PRESENT system which encourages
many thin/weak papers (the system can't really demarcate
them), what is called LPI (least publishabe unit).
> than some abstract perceived correlation between paper production and
> funding, you've made a concrete linkage!
irrelevant, in view of the above comment.
> Keith Robison
> Harvard University
> Department of Cellular and Developmental Biology
> Department of Genetics / HHMI
> robison at mito.harvard.edu
More information about the Bioforum