Examination of Witnesses (Questions 420
- 428)
WEDNESDAY 5 MAY 2004
PROFESSOR SIR
KEITH O'NIONS,
MR RAMA
THIRUNAMACHANDRAN AND
PROFESSOR JOHN
WOOD
Q420 Paul Farrelly: Would you agree
that it would have been helpful to have somebody responsible for
competition matters from the DTI coming to the Committee?
Professor Sir Keith O'Nions: Given
the nature of the conversations we have had this morning and some
of the areas where the three people in front of you have been
less than helpful, either because we do not have the experience
or the responsibility, it must be right. There must be others
that you could have usefully spoken to.
Q421 Dr Harris: Turning to the issue
of public interest and public policy, which is more OST than DTI,
you will be relieved to hear, do you think it is reasonable that
public money should be spent in, for example, but not only, biomedical
research where a question is asked, the result is negative, therefore,
in a sense, not of positive interest and that is never published?
Professor Sir Keith O'Nions: When
you say that it is never
Q422 Dr Harris: It is the problem
of publication bias basically, that negative results, publicly
funded or otherwise, are not published, so that is a huge problem.
If you have 20 studies you could get one with a 5% chance of showing
a positive result but if that is the only one published it looks
like evidence when, in fact, it is not.
Professor Sir Keith O'Nions: As
a general assertion, I do not think I can agree with you. The
whole nature of experimental science and research is that negative
results are published and they are extremely important because
of that. If something does not work, that is a very useful piece
of knowledge which often is published because it tells somebody
else "don't bother to go and do this" or if perhaps
you believe it should have worked you will have another go at
it. As a general assertion I do not think I can agree with you.
Q423 Dr Harris: I am quite surprised,
I thought it was generally agreed that there was a problem, not
that every negative result was not published but that many were
not. I am talking about funded money through the Research Councils
or otherwise being used for a project where one does not insist
on publication and, therefore, the researchers when they get a
negative result do not have to write it up, they can move on to
something that hopefully will be positive and high impact and
more RAE based. I am just wondering whether there should be a
system whereby all results from publicly funded research grants
are published and the Government should facilitate there being
a journal, for example, of negative results where this could be
done. Clearly peer review needs to take place but I would be surprised
if peer review said "This is not positive, get out of our
journal". Has that been thought about and discussed? This
is topical in the science community, certainly the bioscience
community.
Professor Wood: I have read an
awful lot of papers which have got nothing in them because nothing
happened. I am sure others have as well.
Q424 Chairman: You have published
a few!
Professor Wood: If I have I will
not admit it.
Q425 Dr Harris: You just have to
read the abstract.
Professor Wood: I would say that
the Research Councils do demand publication in as much as that
all research after it is finished is assessed and one of the key
criteria for whether the work has been done properly is how much
dissemination there has been of those data. That does feed back
in quite strongly. There are some areas where more speculative
research is undertaken where the initial proposition is almost
essentially totally flawed. I can think of one which I was involved
in which took eight years to sort things out, but we managed to
publish an awful lot of papers on the way to finding that it was
totally flawed.
Q426 Dr Harris: This is tied into
open access, is it not? Because of what you have just said, and
you were joking about it, it is not a gripping read and, therefore,
you are not going to sell many journals that are full of things
that have not worked, whereas open access would be a way of ensuring
that could be done, although there is still a cost to pay for.
Professor Wood: I think you are
probably being a bit too black and white here because some negative
results can be extremely useful and journals do accept that. Personally
I can think of
Professor Sir Keith O'Nions: Just
take cold fusion.
Professor Wood: I did not publish
that.
Professor Sir Keith O'Nions: Nor
did I. Many, many articles on negative results were published.
Can I just shift that a little bit. A great deal of research in
universities is done by people working for their PhDs and when
people write up their PhDs they are usually fairly fulsome and
do report a great deal of things that may not have gone right,
negative results and so on. I think there is probably a higher
proportion of that sort of result in that domain perhaps than
from people who are further on in their careers. That material
is available, it is not published but it is all available from
universities and so on. I think quite a lot of things that are
perhaps not as exciting do actually end up being written up and
are accessible.
Q427 Dr Iddon: I do not think we
could finish this session without referring to the university
end of the public libraries. They are under great pressure because
of increasing usage and, indeed, the increasing amount of published
work, particularly in the research area. Certainly, in my view,
the money awarded to university libraries in particular has not
kept pace with those developments. This morning we have been discussing
the transition period which is obviously going to increase costs
on everybody who wants to buy in publications, whether they be
hard copy or on the internet. Do you agree with me that university
libraries are going to be grossly under-funded if we do not offer
them some special assistance during this transition period?
Mr Thirunamachandran: I have already
given the figures regarding the funding which goes into university
libraries. University libraries will always claim, and with some
justification, that prices of certain journals and certain books
have gone up well ahead of the retail price index so that means
they are struggling to keep pace in real terms. What we have done
is to try to ensure that institutions collaborate more in terms
of what they can do in provision. There is a small fund of just
under five million which we provide called the Research Library
Access Fund which is specifically aimed at those institutions
which are prepared to collaborate and provide access to neighbouring
and other institutions to improve access and to be more cost-effective
in that way.
Professor Wood: From our perspective,
this intermediate area is a problem for us and we are looking
at it and populating various scenarios. In the end I think open
access with institutional repositories is the way to go and then
we will start to see savings, but in the meantime it is a problem.
Q428 Dr Iddon: Will the move to "author
pays" help or hinder this situation?
Professor Wood: Again, it depends
on the cost. This goes back to copyright and all the other things
that have been mentioned, the cost of archiving. It could go either
way.
Chairman: Can I say to the three of you,
thank you very much for coming along. I hope the discussion has
engendered some enthusiasm to get back there with some alacrity
to get a move on to get some decisions and hopefully our report
will play a part in that as well. Thank you very much for taking
the time.
|