MPU: Quality vs Quantity

Alexander Berezin berezin at MCMAIL.CIS.MCMASTER.CA
Thu Jan 25 20:14:46 EST 1996


QUALITY versus QUANTITY

No matter how stupid and pointless 'paper counting'
(and citation counting) may appear, it at least has
one strong advantage: it is fully objective and 
undisputable. Any-one can open Science Citation Index
and count your papers (by Subject Index) and citations
and get the same number. 

But now we want to talk about QUALITY. Fine. And here 
we jump into the realm of a pretty subjective stuff,
especially in view that much of the breakthrough
ideas may not be easily understood by the
contemporaries.

It is almost certainly that NO fully satisafctory
solution can exist about which will all people will
feel perfectly happy and fair. The only practical 
(reasonably good) solutions are those which make some 
account of all factors involve (paper/citations quantity
AND expert's assessment of the quality of the work of
the researcher). In practical terms this again means some 
kind of a sliding scale, because NO selectivity model can 
account for this superpositioning of several factors.
(the cutting boundary is bound to be arbitrary).

(see more below)

On 25 Jan 1996, Gregory R. Harriman wrote:

> 
>      While we both appear to agree that things could and should be done to
> improve the quality of published papers, we are on diametrically opposite
> ends of the spectrum when it comes to how to accomplish that.  You appear
> to take a big-brother approach and want to arbitrarily dictate how many
> papers a person should be allowed to publish.  This is dogmatic and
> unfair.

BEREZIN:
You extremize what I am saying. I am not suggesting that inverted
U-curve should be anything like dictatorial measure, it is simply
should be a waring sign to (whatever mechanism) is used for 
funding decisions.


(HARRIMAN):

>  Some scientists might stuggle to publish one paper a year.  Many,
> perhaps the majority, may find that 2-4 papers a year is reasonable in
> order to maintain quality.  A few talented ones might be able to publish
> 10 or more excellent papers a year.  

BEREZIN:
Precisely. But the trouble with the present (selectivity)
method that the guy with one paper per year is (almost)
certainly won't get a penny. The sliding scale (with basic
grant minimum) will eliminate this anomaly. 

(HARRIMAN):

> It is "wrong" IMHO for anyone ("the
> science police") to dictate how many papers a scientist can or should
> publish.  Let him/her strive to see how far they can go and let their
> scientific peer group decide if they are successful.

BEREZIN:
This is, in fact, about the same as I am saying in the sliding 
scale advocacy. If you step away from the 'selectivity' approach 
towards 'fund researchers, not proposals' this is what you
are proposing (peer assessment is more robust mechanism than
paper counting). 

(HARRIMAN):
>      In terms of how to implement a plan to promote quality over
> quantity...  Again, take the approach suggested by Dr. Tivol.  
> If academic promotions committees looked not at the quantity, but 
> the quality, of papers in a CV and promoted the researcher 
> accordingly, this would significantly decrease the motivation and 
> reward for publishing many, often marginal, papers.

BEREZIN:

How exactly you propose to change premises on which promotion 
committees operate ?

There are numerous cases when people with many papers were 
not promoted while people with much fewer papers (and 
citations) were promoted, but not because their papers 
are 'better', but because some other factors (often their
'political' weight in the institution) were affecting the
decision. 

In any case, I would not make a very strong link between
publish-perish (MPU) problem and promotions. A lot of MPU
and publish-perish stuff goes from those who are already 
at the Full Professor rank, so them it is not the promotion 
but grantsmanship which drives their MPU machinery. If you
want to offer a solution, you have to propose a robust
mechanism which will rank quality of researchers 
separately from their quantity and then blend the both. 

(in case, you are indeed unhappy about how the things are done 
in the present system, although I admit, I did not get this 
impression from your postings - I still not quite see what 
makes YOU wanting radical changes which apperently funds
you adequately). 	  

> 
> Greg Harriman
> 
> 



More information about the Bioforum mailing list