MEMORANDUM 1
Submission from Evidence Ltd
SUMMARY
The UK's exceptional relative international
performance in science and technology, reflected in the data we
present in our annual report to the Office of Science & Innovation,
should make it an ideal research partner for many other countries.
Other data confirm that international collaboration is expanding
very rapidly, as are the domestic research bases of countries
such as China, India and Iran. That means that not only are opportunities
for international partnerships increasing but also the diversity
of likely outcomes will increase: there will be new approaches
to research as well as a greater global knowledge pool. The UK
has a good share of international collaboration, but it is not
as strong as might be anticipated, it is not expanding as rapidly
as some countries, and it is less consistent in the biomedical
areas where the UK has a position of world leadership on research
quality. The UK might benefit from a more assertive approach to
collaboration and from promoting the mobility of its own researchers
as well as encouraging visits from elsewhere.
BACKGROUND
1. We note that the Science and Technology
Committee has decided to hold an inquiry into the international
policies and activities of the Research Councils. We are submitting
this note and attachments to the Committee so that it is aware
of emerging evidence relevant to that inquiry and on which we
are currently working for UK Government offices but which is not
yet in the public domain.
2. Evidence Ltd works for the Office of
Science & Innovation (OSI) and individual Research Councils,
and for the Higher Education Funding Councils (HEFCs) and HE Institutions
in analysing their research activity and performance in order
to support better management of their research portfolios. For
OSI, we have produced the annual Public Service Agreement (PSA)
target indicators on the "relative international performance
of the UK research base". We also publish the UK Higher Education
Research Yearbook, which this year included a Foreword from the
Chief Executive of HEFCE.
3. Evidence has particular experience in
using bibliographic and bibliometric data (data on publications
and their citations) to analyse research activity.
4. We are currently carrying out work for
the Foreign & Commonwealth Office (FCO) Science and Innovation
network, to evaluate changing patterns of UK collaboration with
China and India. We have just completed work for OSI on "international
patterns of research collaboration" but this remains unpublished
at this time.
RELATIVE INTERNATIONAL
COLLABORATION
5. In current work for OSI and the FCO S&I
network on international collaboration, which should be published
later in 2007, we find that the UK has a well developed network
with other leading economies but it is not expanding its collaborative
links as quickly as some competitors. This is noticeable in considering
links with China and India.
6. Research collaboration can be measured
by analysing inputs (money), activity (the numbers of projects
and the numbers of people involved) and outputs (publications).
7. It is difficult to assign funding to
international activity because many collaborative links draw not
only on dedicated funds but on separate funding held by one or
both partners. This is particularly true of bottom-up collaboration,
driven by researchers' desire to be involved in high-quality or
particularly innovative research, whereas the dedicated funding
that emerges from top-down collaboration initiatives tends to
be driven by policy priorities. Thus, on balance, analysing dedicated
funding may concentrate more on the research platform than the
research peaks.
8. On project and people numbers, it is
difficult to validate data and acquire useful international comparisons.
We simply do not know the substantive value of a "visit"
of a researcher to or from a laboratory in another country. We
do not have the information on the pattern of movement for other
countries.
9. Publication data are therefore the key
evidence of international collaboration, because researchers do
not freely give away co-authorships so where a paper carries addresses
from more than one country we know the data is likely to be valid.
Databases provide ready international comparisons, and data can
quickly be aggregated by field. Furthermore, the associated citation
data allows us to attach quality measures.
10. The main data source is Thomson Scientific®,
which covers around 8,700 journals of international standing.
About 35% of 700,000 catalogued research articles published by
UK-based researchers over the last ten years have a co-author
from another country. The rate of international collaboration
has increased progressively, but the UK has a slower rate of increased
collaboration than some of its competitors.
11. The UK has a record of close collaboration
with the USA. About 30% of its international links are with that
country.
12. The UK also collaborates closely with
France and Germany. Its links with Germany have expanded more
rapidly than those with the USA (1.5 vs 1.4 for 2001-05 compared
with 1996-00, see table below). The relative increase in collaboration
for France and Germany with the UK is greater than it is for them
with the USA (France 1.39 vs 1.31, Germany 1.50 vs 1.37). The
leading European nations have therefore increased their relative
interaction, which is likely to be a consequence of EC cohesiveness
measures.
RECENT INCREASE IN COLLABORATION (RATIO 96-00/01-05)
MEASURED BY CO-AUTHORSHIP OF RESEARCH PUBLICATIONS
| UK | USA
| FRANCE | GERMANY
| CHINA | INDIA |
| UK | — | 1.40
| 1.39 | 1.50 | 1.94
| 1.65 |
| USA | 1.40 | —
| 1.31 | 1.37 | 2.23
| 1.54 |
| FRANCE | 1.39 | 1.31
| — | 1.40 | 2.05
| 1.48 |
| GERMANY | 1.50 | 1.37
| 1.40 | — | 1.96
| 1.81 |
COLLABORATION OUTSIDE
THE G7
13. The UK has good and growing links with both China
and India and, as a research-excellent partner, should be well
positioned to build on these. However, Germany is increasing its
general rate of collaboration with both countries more rapidly
than the UK (table above).
14. For the UK, in many science and technology fields
the rate at which collaboration is increasing is slower than that
for other major research economies. The problem is therefore pervasive.
15. The UK has particular research strengths in the Biological
sciences.[1] China and
India have thus far focussed much of their expansion in the Physical
and technological sciences. (for example, China published 132,000
papers in Physical sciences in 2001-2005 compared with 28,000
in Biological sciences; the UK published similar numbers in the
two areas, as did the USA). It is essential that the UK places
itself in a good position to capitalise on its existing competency
so as to take advantage of the opportunities for collaboration
that will appear as China and India move into the bio-medical
area.
THE BENEFITS
OF COLLABORATION
16. International research collaboration provides access
to larger pools of innovative ideas, emerging knowledge and highly
trained people. The work it produces also has "value added".
17. In a study carried out for Sir Gareth Roberts' study
of "International Partnerships of Research Excellence",
Evidence demonstrated that both UK and US research partners gain
added value from their collaboration compared with their domestic
outputs.
18. We have now extended that analysis and created an
Impact ProfileTM to illustrate the overall gain made by the UK
through its international links over the last ten years. The unpublished
note on this is attached. A related analysis on five-year windows
will be included in the current work for OSI.
19. It is quite clear that significant value accrues
to international collaborative work, but it is important to disaggregate
the reasons for this. Perhaps the most important factor is opportunity
cost: researchers only spend time and resources to create links
when they look promising at the outset. The exception to this
is when national agencies put up money for such links, when the
researchers' goal changes and becomes the acquisition of a share
of the inputs rather than the outcomes.
RESEARCHER MOBILITY
20. China offers not only "more" research but
also "different" research. There is a growing research
base in Iran which will offer further challenges to western research
paradigms. The UK can acquire knowledge of what China and Iran
are doing, albeit belatedly, by reading what they publish. It
can only find out how and why if it actively collaborates and,
particularly, if UK researchers travel to and work in China and
Iran as well as those researchers travelling to the UK.
21. In 2004, in a report published by the Higher Education
Policy Institute (HEPI), we used bibliographic data to track the
career pathways of samples of individual highly-cited researchers
from a number of countries (highly cited researchers are those
who have an exceptional number of publications that are recorded
as being in the most-cited 1% of world publications).[2]
22. Rather than indicating problems of "brain drain",
which had been repeatedly claimed, we showed that the typical
pathway for the highest performing UK researchers was to spend
some time overseas—particularly in the USA—before
returning to the UK to pursue a successful career. By contrast,
the few USA researchers who left north America tended not to return.
23. We argued that international research mobility not
only contributed to individual career development but also conferred
significant national benefit. Evidence has noted that some smaller
countries with profiles of exceptional research excellence are
associated with relatively high levels of researcher mobility.
A notable example is Switzerland where most elite researchers
have spent time abroad and where national research performance
is world-leading in some fields, which is unusual for a relatively
small research economy.
24. We suggest to the Committee that it is essential
that more UK researchers should be encouraged to spend time overseas
at an early stage in their career. UK research mobility is better
than some countries but is not exceptional in European terms,
and the UK is therefore losing the opportunity to gain from making
contacts with and learning lessons from others.
25. Contacts made early in a researcher's career also
lead to collaborative links which are then maintained for life.
This enables the UK to tap into innovative research across a global
network and to gain value form investments made in other economies.
THE RESEARCH
COUNCILS AND
OSI
26. We have not sought to comment on the current international
initiatives run by the UK Research Councils but to provide evidence
to contextualise the significance of the Committee's inquiry.
27. For the same reason, we have not provided full details
from our report on international collaboration to OSI. OSI will
wish to convey its own views when it has had an opportunity for
due reflection and interpretation. However, with OSI's agreement,
we could supply more detailed information if this is useful to
the Committee.
OVERVIEW
28. We have been analysing UK research data for over
ten years. Research is an increasingly global activity and both
knowledge and knowledgeable people are sourced globally. If the
UK is to maintain the competitive edge that we have demonstrated
in our annual reports for OSI, then it must do so by ensuring
that it is intimately linked to the most research-active economies
elsewhere. We suspect that this will not be done by open-ended
agreements on the development of cultural links but through much
more specific compacts on clearly identified research objectives
supported by tangible resources and with clear and substantive
benefits to each party. Allied to this must be programmes that
motivate UK researchers to work in those partner countries and
thereby to further UK understanding of those research cultures
and enhance our ability to interact. Relying on links maintained
primarily by visits of others to the UK, where their objective
is to learn about UK research culture but not to convey their
own, will no longer suffice.
April 2007
1
http://www.dti.gov.uk/science/science-funding/budget/uk-research-base/page29207.html Back
2
http://www.hepi.ac.uk/pubdetail.asp?ID=181&DOC=Reports Back
| 
