2005 OST PSA report

Introduction

This is a report about indicators of the UK’s relative international research
performance in science, engineering, the social sciences and the humanities
and arts. It is the third report with these indicators and in this format. The
information content, analysis, commentary and overall structure have
developed in successive cycles. The main change this year is the inclusion of
data on research in the humanities and arts. This has extended the number of
sub-fields covered by some indicators.

The

Research Footprint

diagrams summarise the outcomes of analyses for six

leading indicators, comparing the research profile of the UK, the G8 and a
number of other leading research economies. The

thematic commentary,

following the Footprints, gives a broad overview of the UK’s performance in
terms of the selected indicators.

Background

The objective is to support a system for assessing outputs, outcomes and
impacts related to the Public Service Agreement (PSA) target to ‘improve the
relative international performance of the UK research base’. This target is
challenging. Many studies have indicated that successive advances in
research become increasingly expensive (the “sophistication” factor, Advisory
Board for the Research Councils (ABRC), ‘Strategy for the Science Base’,
1986) and the costs of improving relative performance rise in parallel.

The Atkinson Review of ‘Measurement of Government Output’ (2005) for the
Office of National Statistics affirmed that ‘the measurement of quality is central
to our concerns’. Sir Tony Atkinson recommended that all assessments of
output and productivity should account for this. This is what the OST sets out
to do, capturing information not only on what the research base produces for
the Science Budget investment but how its output is perceived internationally.

The UK is widely acknowledged to be an extremely effective research
performer. It is therefore difficult to improve significantly on this relative level of
achievement. Indeed, it will be difficult in some fields to maintain the UK’s
international status without, for example, additional investment that meets the
growing competition from technologically specialist research nations in the

Asia-Pacific economies. This year’s report particularly highlights the impact
that China is having as its research base expands.

Until 2002, the UK Office of Science and Technology (OST) employed a core
set of indicators that demonstrated the position of the UK and reflected
effectiveness in the use of research funding. This report describes an extended
basket of indicators established in 2003. Plurality in an indicator system is a
desirable feature, because over-dependency on any one indicator can be
misleading. A balanced set can take account of differences in the pattern of
performance between research disciplines, the interaction between inputs and
outputs and possible measures of efficiency and effectiveness, and year on
year fluctuations in any one indicator.

Assessing excellence is as important as measuring system average. The peak
of research excellence, however defined, includes those highly innovative
outcomes that are most likely to impact on economic performance. The
indicators in this report allow for disaggregation, to throw light on changing
patterns of selectivity and concentration within the UK science base. That work
is being developed in a parallel OST study to be published in 2006.

Other countries and communities – such as the EU, the NSF in the USA,
CWTS in the Netherlands and the OST in Paris - already publish reports about
national science and technology indicators on a regular basis. This report has
taken note of the good practice established elsewhere.

Data and Indicators

Every piece of research data should have three attributes: subject area, time
and location. Each attribute works at varying levels of detail and we need to
identify the best level for analysis. Data about research usually measure
something in one of three primary categories: input (usually financial), activity
(or proxies such as staff numbers) and outputs. Secondary indicators describe
the relationship between them. Sometimes, outputs can be followed through
into outcomes and impacts. The UK indicators include both primary and
secondary indicators and focus on impacts where possible. They are listed in
the table (below) on the “

Definition and description of indicators”.

Bibliometric data play a key part in these indicators. There are reasons for
being cautious about some uses of such data particularly with respect to social
science and to humanities’ and arts’ research (see

Background

sections after

the indicator pages). There is also great value and applicability. Crucially,
these data uniquely provide us with international comparisons of research
quality for most countries and by subject area.

The

Background

sections that come after the indicator data describe the main

data sources,

list the range of the OST

comparator group of countries

, the level

subject disaggregation

and the

time frames

used for comparisons. There is

also a discussion about the significance and interpretation of

bibliometric

indicators

and some cultural aspects of publication and citation behaviour in

different countries and disciplines.

International comparisons

are made across an OST comparator group of 25

countries. This includes the full G8 (UK, USA, Canada, France, Germany ,Italy,
Japan, Russia). In addition there is a combination of some larger and OECD
countries from different continents with research bases both similar and
contrasting in structure to the UK, and a spread of smaller nations with active
and rapidly growing research bases with specific strengths. These are
Belgium, Denmark, Finland, Netherlands, Poland, Spain, Sweden, Switzerland,
Australia, Brazil, China, India, Iran, Israel, Singapore, South Africa, South
Korea, and Taiwan. A separate line of analysis for a group of 15 EU countries
(EU15 = member countries in 2003) is also included where feasible and
appropriate.

Many of the graphs that illustrate performance indicators use short codes for
these countries, for clarity. These codes are linked to their countries in a table
in the

Background

sections.

Subject disaggregations

used in this report employ two systems of

categorisation. First, there are five main OECD categories (medical sciences,
natural sciences, engineering and technology, social sciences and humanities
[which includes the arts]). Second, subjects based on 69 UK Units of
Assessment (UoAs) are grouped by publication similarity into ten Super-UoAs
(Clinical, Pre-clinical/health, Biological sciences, Environmental sciences,
Mathematics, Physical sciences, Engineering, Social sciences, Business,
Humanities).

International R&D databases have historically focussed on science and
technology and are therefore have some deficits in social science and
humanities data. This does affect some analyses, and this is discussed further
in the

Background

sections.

The humanities and arts are included in the subject spread for the first time.
Not all the research indicators used in the natural sciences are well suited to
analysing research performance in these disciplines. Account can be taken of
feedback from readers and users of this report so as to ‘tune’ these
presentations for later editions.

Details of the countries and subjects are given in the relevant part of the

Background

section after the indicators.

Definition and description of indicators

Indicator number

Description of performance
indicator

Condition signalling
improvement

Level of disaggregation

Primary data sources

THEME 1

INPUTS including expenditure on
research

OECD MSTI 2005-1

1.01

GERD relative to GDP (R&D
intensity)

Increased proportion of R&D
specific spend

System

1.02

Publicly performed R&D
(PUBERD) as proportion of GDP

Increased proportion of R&D
specific spend

System

OECD MSTI 2005-1

THEME 2

OUTPUTS including people and
publications

2.01

Number and share of OECD PhD
awards

Increased count and increased
share by comparison with
competitors

System

OECD Education Database

OECD MSTI 2005-1

2.02

PhDs awarded per head of
population

Increased ratio

System

OECD Education Database

2.03

Number and share of world
publications

Increased count and increased
share by comparison with
competitors

System

ISI National Science Indicators 2004

THEME 3

OUTCOMES including research
recognition and citations; training
and research quality

3.01

Number and share of world
citations

Increased count and increased
share by comparison with
competitors

System

ISI National Science Indicators 2004

3.02

Number and share of world
citations in ten main research
areas

Increased national count and
share

SUoA

ISI National Science Indicators 2004

3.03

Rank of share of world citations
by nine main research fields -
frequency in top 3

More frequent presence in top
three among fields

System

ISI National Science Indicators 2004

Indicator number

Description of performance
indicator

Condition signalling
improvement

Level of disaggregation

Primary data sources

3.04

Share of citations relative to
share of publications

Increase in citation share
compared to source share within
field

SUoA, using NSI5 for
constant time frame

ISI National Science Indicators 2004

3.05

Number and proportion of uncited
publications

Decrease in ratio of uncited to
total sources

SUoA, using NSI5 for
constant time frame

ISI National Science Indicators 2004

3.06

Number and share of cited
publications

Increase in ratio of cited to total
sources

System

ISI National Science Indicators 2004

3.07

National share of papers in top
1% by citation count

Increase in share

System

ISI Essential Science Indicators

3.08

Citation impact (citations per
publication) relative to world
baselines

Increased impact compared to
world

System

ISI National Science Indicators 2004

3.09

Citation impact relative to world
baselines in ten main research
fields

Increased impact in main
research fields

SUoA

ISI National Science Indicators 2004

3.10

Variation and consistency of
research strength

Reduced ratio between variance
and average. Maximised ratio of
average/variance.

SUoA

ISI National Science Indicators 2004

3.11

Relationship between distribution
of research training across
subjects and research quality

Improved match in distribution;
improved research training power
(product of volume and quality)

Data related across
research fields (OECD
level)

OECD Education Database

THEME 4

PRODUCTIVITY – FINANCIAL
including outputs and outcomes
related to inputs

OECD Education Database

4.01

PhDs awarded relative to HERD

Increased ratio

System

OECD MSTI 2005-1

OECD Education Database

UK SET statistics and HESA data

4.02

PhDs awarded relative to HERD
in five main research areas

Increased ratio

OECD fields

OECD RDS 2004-2

Indicator number

Description of performance
indicator

Condition signalling
improvement

Level of disaggregation

Primary data sources

OECD Education Database

4.03

PhDs awarded relative to
PUBERD

Increased ratio

System

OECD MSTI 2005-1

ISI National Science Indicators 2004

4.04

Citations relative to GDP

Improved ratio of citations per
GDP compared to recent past

System

OECD MSTI 2005-1

ISI National Science Indicators 2004

4.05

Number and share of citations
relative to GERD

Increased citations per unit
expenditure

System

OECD MSTI 2005-1

ISI National Science Indicators 2004

4.06

Citations relative to PUBERD
(GOVERD + HERD)

Increased citations per unit
expenditure

System

OECD MSTI 2005-1

ISI National Science Indicators 2004

4.07

Citations relative to HERD

Increase in citations per unit
spend

System

OECD MSTI 2005-1

ISI National Science Indicators 2004

UK SET statistics and HESA data

4.08

Citations relative to HERD in five
main research areas

Increase in citations per unit
spend at the OECD macro-
category level

OECD fields

OECD RDS 2004-2

THEME 5

PRODUCTIVITY – LABOUR
including outputs and outcomes
related to non-financial measures

OECD Education Database

5.01

PhDs awarded per researcher

Increased ratio

System

OECD MSTI 2005-1

ISI National Science Indicators 2004

5.02

Publications per researcher

Increased relative output

System, NSI5 averaging

OECD MSTI 2005-1

ISI National Science Indicators 2004

5.03

Citations per researcher

Increase in citation ratio

System, rolling 5 year
averages

OECD MSTI 2005-1

THEME 6

PEOPLE

6.01

Researchers per thousand
population

Increased ratio

System

OECD MSTI 2005-1

Research Footprints

There are over 30 research indicators grouped under seven themes. This complex body of data provides an informative and comprehensive view of many aspects of
the comparative international performance of the research base, but it is not readily absorbed. We have used a Research Footprint® to illustrate key data.

Each country has a distinctive Research Footprint

of its international research competitiveness. Our figure uses six key indicators and provides a direct graphical

comparison of the performance of select comparator countries with the OST comparator group average. The shaded area is the ‘footprint’ of a stated country, which can
be compared directly with the dotted line that marks the average footprint for the group. Each axis measures a specific indicator, with the lowest level of performance
(low rank or zero activity) at the origin near the centre and the maximum value at the outer end of the axis. The area of the footprint has no statistical significance.

1.02 PUBERD

per

GDP

Theme: Inputs
Full title: Publicly performed R&D (PUBERD) as proportion of GDP
Description: Volume of publicly funded R&D relative to general economy

2.01

Share of OECD PhDs

Theme: People (research degree output)
Full title: Number and share of OECD PhD awards
Description: Highly skilled people: research degree output

2.03

Share of world publications

Theme: Outputs
Full title: Number and share of world publications
Description: Relative output volume

3.01

Share of world citations

Theme: Outcomes
Full title: Number and share of world citations
Description: Esteem measured by share of world citations

3.03

Lead citation share by research field

Theme: Outcomes
Full title: Frequency in top three for rank of world citation share by nine main
research fields
Description: Breadth of research strength measured by spread of dominance

6.02

Researchers per thousand workforce

Theme: People
Full title: Workforce research capacity
Description: Skilled R&D capacity within national workforce

3.01 Share of world citations

2.03 Share of world

publications

6.02 Researchers per

thousand workforce

1.02 PUBERD per GDP

2.01 Share of OECD PhDs

3.03 Lead citation share

by research field

Research Footprint

of comparative UK research performance

The Research Footprint

for the UK is compared in the next two pages, first

with other G8 nations and with the pattern for the EU15 as a whole and second
with a set of other leading research nations in the OST comparator group.

The display uses absolute values, not ranked position. Because of data and
analysis changes, the comparator average Research Footprint

– essentially

the reference point for others - covers a slightly different area to that of 2004
and relative performance of specific countries may therefore appear to expand
or retreat. The data coverage – for countries, years and fields – has improved
again since last year. This has identified a number of exceptional performers for
particular indicators, some of which appear to behave inconsistently and may
be amended in later reports. There are also some anomalous values (e.g.
those involving GDP for Russia).

The status of the USA is quite clear. It will continue to be a strong performer
across the board and contributes the maximum volume performance on most
indicators because of its sheer size, although its efficiency is certainly less
impressive and its effectiveness is being challenged. One example of an
exception is in terms of public expenditure on R&D as a proportion of GDP
(indicator 1.02). On that indicator there are smaller nations, such as Sweden,
Denmark and the Netherlands, that all spend relatively more. Other indicators
show that they all have relatively highly skilled populations and rising
performance in niche research areas.

The EU has no calculated value on indicator 3.03 [which would be a summation
of specific countries rather than an integrated figure] where it would score as
highly as the USA. It has more PhDs than the USA (indicator 1.01) and higher
average PUBERD per GDP (indicator 1.02) but relatively fewer researchers in
its total workforce (indicator 6.02), although the data are patchy in this area.

The UK has a substantial and growing share of world publications (indicator
2.03) and citations (indicator 3.01). Because it ranks second on average to the
USA, it also has a very good position on average ranking by major research
area (indicator 3.03). The UK’s share of OECD PhD awards (indicator 2.01)
matches its share of outputs but is less than its achievement on other
indicators. It has a relatively low (below average) concentration of researchers
within its workforce (indicator 6.02) by comparison with some of the smaller
countries with rising profiles.

Overall, the UK can be seen to have a good all round performance by
comparison with most of the G8 and all of the smaller countries, but the well
balanced performance of Germany – good business investment, strong PhD
output, above average proportion of research workers - is also notable. Each
report has confirmed that it is the major research competitor for the UK in
Europe. Japan’s strength is in its research work-force (indicator 6.02), as it is
for Sweden. While Switzerland has a strong average bibliometric performance
in many fields, in the Research Footprint

presentation its relatively small

research base and capacity becomes clear.

China has been given a Research Footprint

this year, replacing Belgium.

China has exceptional growth in its research base and is rapidly expanding its
volume of activity with rapid growth in GERD, growth in outputs in most fields of
science and technology and a very large R&D workforce with a relatively high
proportion of researchers. The Research Footprint

shows the extent to which

this latent strength has yet to be translated into quality outcomes. On these
criteria, China remains a ‘sleeping giant’ but the weight of its footfall can be
expected to change this picture significantly in the next few years.

Research Footprints

for UK, G8 countries (except Russia) and the EU15

USA

EUROPEAN UNION

CANADA

FRANCE

GERMANY

ITALY

JAPAN

3.01 Share of world citations

2.03 Share of World Sources

6.02 Researchers per

thousand workforce

1.02 PUBERD per GDP

2.01 Share of OECD PhDs

3.03 Lead citation share

by research field

3.01 Share of world citations

2.03 Share of World Sources

6.02 Researchers per

thousand workforce

1.02 PUBERD per GDP

2.01 Share of OECD PhDs

3.03 Lead citation share

by research field

3.01 Share of world citations

2.03 Share of World Sources

6.02 Researchers per

thousand workforce

1.02 PUBERD per GDP

2.01 Share of OECD PhDs

3.03 Lead citation share

by research field

3.01 Share of world citations

2.03 Share of World Sources

6.02 Researchers per

thousand workforce

1.02 PUBERD per GDP

2.01 Share of OECD PhDs

3.03 Lead citation share

by research field

3.01 Share of world citations

2.03 Share of World Sources

6.02 Researchers per

thousand workforce

1.02 PUBERD per GDP

2.01 Share of OECD PhDs

3.03 Lead citation share

by research field

3.01 Share of world citations

2.03 Share of World Sources

6.02 Researchers per

thousand workforce

1.02 PUBERD per GDP

2.01 Share of OECD PhDs

3.03 Lead citation share

by research field

3.01 Share of world citations

2.03 Share of World Sources

6.02 Researchers per

thousand workforce

1.02 PUBERD per GDP

2.01 Share of OECD PhDs

3.03 Lead citation share

by research field

3.01 Share of world citations

2.03 Share of World Sources

6.02 Researchers per

thousand workforce

1.02 PUBERD per GDP

2.01 Share of OECD PhDs

3.03 Lead citation share

by research field

Research Footprints

for other leading comparator nations

SWITZERLAND

SWEDEN

SPAIN

SOUTH KOREA

NETHERLANDS

DENMARK

CHINA

AUSTRALIA

3.01 Share of world citations

2.03 Share of World Sources

6.02 Researchers per

thousand workforce

1.02 PUBERD per GDP

2.01 Share of OECD PhDs

3.03 Lead citation share

by research field

3.01 Share of world citations

2.03 Share of World Sources

6.02 Researchers per

thousand workforce

1.02 PUBERD per GDP

2.01 Share of OECD PhDs

3.03 Lead citation share

by research field

3.01 Share of world citations

2.03 Share of World Sources

6.02 Researchers per

thousand workforce

1.02 PUBERD per GDP

2.01 Share of OECD PhDs

3.03 Lead citation share

by research field

3.01 Share of world citations

2.03 Share of World Sources

6.02 Researchers per

thousand workforce

1.02 PUBERD per GDP

2.01 Share of OECD PhDs

3.03 Lead citation share

by research field

3.01 Share of world citations

2.03 Share of World Sources

6.02 Researchers per

thousand workforce

1.02 PUBERD per GDP

2.01 Share of OECD PhDs

3.03 Lead citation share

by research field

3.01 Share of world citations

2.03 Share of World Sources

6.02 Researchers per

thousand workforce

1.02 PUBERD per GDP

2.01 Share of OECD PhDs

3.03 Lead citation share

by research field

3.01 Share of world citations

2.03 Share of World Sources

6.02 Researchers per

thousand workforce

1.02 PUBERD per GDP

2.01 Share of OECD PhDs

3.03 Lead citation share

by research field

3.01 Share of world citations

2.03 Share of World Sources

6.02 Researchers per

thousand workforce

1.02 PUBERD per GDP

2.01 Share of OECD PhDs

3.03 Lead citation share

by research field

Thematic commentary

Theme 1 - Inputs (including expenditure on research)

The UK is spending less on research as a proportion of its total economic
activity than its competitors and less than the EU15 average. It share of
comparator group R&D spend is about 4.5%, which is barely half its share of
group outputs (

Theme 2

and

Indicator 4.05

). Its ranked position on R&D spend

is also declining on current data.

GERD as a proportion of GDP indicates research intensity in the economy. UK
GERD was about 8% higher in 2003 compared with the average for the recent
past (

Indicator 1.01

) but it remains at only about 1.8% of GDP and has fallen in

regard to the OST comparator group average. Business expenditure in UK
R&D has fallen across the public sector (

Theme 7

). The UK is 6

among G8

nations and ranks only 17

among the twenty-one OST comparator group

nations for which data are available (average around 2.3% GDP and
progressively increasing). The most rapid increase in GERD is for China, which
now exceeds all but the USA and Japan.

GOVERD and HERD refer to expenditure on R&D performed in the
GOVernment and Higher Education sectors. The sum of these (PUBERD)
forms the bulk of the publicly funded research base. This has generally
increased in real terms for the UK over the last decade, but dropped in 2001-02
and the increase now is less than the average for the OST comparator group.
The UK is 7

among G8 nations (

Indicator 1.02

). UK HERD has increased in

real terms (see table in

Indicator 4.07

) but at a slower rate than the OST

comparator group.

Theme 2 - Outputs (including people and publications)

The UK’s share of global research people and publication outputs are both
around 9% The UK is ranked 3

globally in terms of people output and 2

terms of papers, despite a recent fall in volume. This contrasts with its rank on
inputs (

Theme 1

The UK’s share of PhD awards (now over 9%) is similar to that of similar
countries for which data are available but much less than Germany (

Indicator

2.01

). In relation to population size, UK output has been in line with a general

trend whereas Germany is a high outlier. Looking to the future, growth is higher
in South East Asia, and some smaller and research competitive European
nations also outrank the UK on this measure (

Indicator 2.02

The UK’s share of world journal article publications (slightly less than 9%) was
overtaken by Japan in 2003 but has now moved back to 2

behind the USA

(

Indicator 2.03

). A plateau in publication output over the last few years followed

a period of annual increases in many countries. The drop in UK share is likely
to be due to consolidation rather than contraction as it has not affected quality
(

Theme 3

Theme 3 - Outcomes (research recognition, citations; training and
research quality)

The UK is very efficient in terms of research outputs (as the contrast between

Theme 1 and 2

shows) and, generally, has fewer lower quality papers than

most of its competitors. The UK relative international research performance is
second behind the USA in terms of overall research recognition but it now
performs better than the USA in some specific areas though less well in others.
It has a good balance of strength in its performance, however, and has fewer
areas of relative weakness than some competitor nations. The rapid growth of
outputs from the China research base is likely to make the international picture
more dynamic over the next few years.

The UK’s total citation count is 2

highest to the USA and has risen to just over

12% of world citations (

Indicator 3.01

). It remains ahead of all other OST

comparator group competitors. Germany was closing but has not improved its
relative position in recent years. China is rapidly increasing its global share, as
are some smaller nations and this has effects on larger countries.

The UK is clearly ahead of competitors (albeit 2

to the USA) on citation count

in all of ten research fields including the new humanities field, but not in
mathematics, where it is 3

, and in physical sciences and in engineering, where

it is 4

(

Indicator 3.02

). Its share of global citations has increased in most fields

but not in environmental sciences. It is becoming clear that there are some
inconsistent annual oscillations and that longer term trends are more important.
Citation growth among smaller nations does ‘attack’ the global share of larger
countries but the trend among individual small countries is erratic.

A measure of consistency in research strength is found in the frequency with
which a country appears in the top tier of nations, ranked by citation count for
each of nine main fields (

Indicator 3.03,

see also

Indicator 3.10

). The USA is in

the top tier in all fields. The UK performs strongly on this indicator and is in the
top three in seven of nine fields. Germany is the only other nation that performs
well across many fields. It had improved to place in the top three in six fields
but has slipped to only five fields. No nation outside the G8 is placed in the top
three on this measure in any field.

Citation count usually increases with numbers of papers published (source
count). So, although numbers of citations (

Indicators 3.01, 3.02

) give one

measure of esteem, it is also important to look at share of global citations
compared to share of sources so as to account for this scaling.

Indicator 3.04

shows that the UK has a positive difference between share of citations less
share of outputs in eight of ten research areas, particularly biological, physical
and environmental sciences. Other countries have multiple positive balances
but these are smaller in scale than for the UK or concentrated in niche areas.

Not all papers are subsequently cited, so their impact on research remains
obscure. About 35% of UK papers remain uncited in any 5-year period but this
proportion and the UK’s share of the world's uncited papers has declined – so
its performance has improved - compared to the recent past (

Indicator 3.05

The decline is seen across all fields; it has the fewest uncited among G8
countries in Biological, Environmental and Social Sciences and Humanities.

The UK’s ratio of cited papers to total outputs is greater than for the OST
comparator group average and improving in all fields (

Indicator 3.06

The UK has the 2

greatest - and a growing - share (13.2%) of the world’s

most highly cited papers (

Indicator 3.07

). Its average research impact

(measured by the number of citations per paper for these data) is ranked 8

but

the volume of this high impact material is greater than for countries ahead of
the UK such as Switzerland, the Netherlands and Scandinavia. UK share has
again increased whereas that for the countries ahead of the UK on impact,
including the USA, has fallen.

Research impact is usually indexed by the average ratio of citations to sources.

Indicator 3.08

shows that, for the research base as a whole, the UK’s average

impact increased progressively in the last ten years and has stepped up
between the recent past (the five years to 2003) and 2004. This has moved the
UK even closer to the USA and further ahead of most other G8 countries but
Germany has also improved and is a close competitor. The UK now stands 5

in the world where it was 7

How does the pattern of research impact vary between disciplines? It is
reviewed at the level of the ten main research fields in

Indicator 3.09

. The data

show marked improvements for the UK in some areas where it is now ranked
ahead of the USA in pre-clinical & health sciences and in biological sciences.
Although ranked further back among the OST comparator group as a whole, its
impact has improved progressively in many fields over the last decade. The UK
has almost closed the gap on the USA in mathematics, and has a marked
improvement in physical sciences and a sustained upturn in engineering.

Is research strength broad based and diverse or does it peak only in specific
disciplines? Average research impact and variation in quality between fields
both contribute to the overall performance of the research base (see also

Indicator 3.03

). A more consistent performance provides more even capacity

and hence enables flexibility in response to opportunities.

Indicator 3.10

links

information on field specific impact and considers both average across fields
and variation between fields for leading research countries. The UK has both a
strong average and a relatively even performance. The UK position has been
consistent between 2000 and 2004 whereas the USA has dropped back and

the Netherlands has improved its average only at the cost of more concentrated
excellence.

Does people quality match publication quality? It is not possible directly to
measure the quality of highly skilled people trained in the research base, but we
can test the broad correlation between relative training volume and relative
research impact. There is a correlation between training and impact for the UK
(

Indicator 3.11

) whereas this is absent for most G8 countries. UK PhD awards

are concentrated in the natural sciences, which is where

Indicator 3.09

suggests that the UK has relative research strength. Thus the UK’s
advantageous productivity pattern (

Theme 1, Theme 2

) is matched by a very

positive quality pattern.

Theme 4 – Productivity - financial (outputs and outcomes related to
inputs)

The UK is highly productive across the research base by the measures applied
in this analysis, as has been indicated above. However, its position is
challenged in some fields and is not as uniformly strong as it has been in the
past.

Highly trained people are an important product of the research base.

Indicator

4.01

shows relative productivity as PhDs awarded in relation to general spend

in the HE sector (HERD) and by main research areas (

Indicator 4.02

). This is

discussed above in terms of output (

Indicator 2.01

Total UK PhD awards are similar to most other G8 nations, though less than
Germany and much less than the USA. Compared to the level of spend on
R&D performed in the HE sector (HERD,

Indicator 4.01

) the UK produces

relatively more PhDs per unit spend than most OST comparator group nations
and is ranked 5

, above the EU average. The USA is ranked 11

. Other

measures do indicate that UK quality is high, so the assets of the research base
appear to be used effectively although it is, of course, not necessarily the case
that higher people output per unit research spend is a quality measure.

At the level of five main research fields, the UK ranks 3

to Germany and the

USA in volume in the natural sciences but has moved ahead of Germany in
terms of PhDs per HERD, and is well ahead of OST group average.
Productivity has fallen back markedly in engineering. The correlation between
level of HERD and PhD awards is less clear in medical sciences, where the UK
is lower ranked and has productivity well below OST group average; Germany
remains well ahead. The UK is below the OST comparator group average in
the social sciences, although there are issues about data quality in this area,
but well ahead in the humanities (

Indicator 4.02

We can also index PhD output in terms of total PUBlic expenditure on R&D
(PUBERD,

Indicator 4.03

) and the UK then ranks 1

in the G8 ahead of

Germany and well ahead of the OST comparator group average against which
it is improving its position.

Research performance in relation to the general economy (GDP) is of less
direct significance. Although productivity is falling, the UK is 1

among G8

countries on citations per GDP (

Indicator 4.04

) but lies 7

overall behind

Switzerland and Scandinavian nations. Its performance is about 30% better
than the OST comparator group average whereas other G8 apart from Canada
are below that average.

The UK has dropped from 3

to 5

in the OST comparator group in terms of

citations per unit GERD but has maintained its performance over the recent
past and is well ahead of other G8 nations. As would be predicted from
examination of

Theme 1

and

Theme 2

, it share of citations per share of spend

is better than twice the OST comparator group average (

Indicator 4.05

UK PUBERD has fallen compared to the OST comparator group average over
the last ten years.

Indicator 4.06

shows that citations per unit spend have

improved over the period, while the USA and Canada have fallen on this
measure and the UK’s lead within the G8 has therefore increased. The UK is
now 3

to Switzerland and Denmark in the OST comparator group.

In terms of citations per unit spend as HERD (

Indicator 4.07

), the UK had led

the OST comparator group in the past and fell back slightly last year but is 2

to Switzerland. Its performance has remained well ahead of OST comparator

group average and improved on other G8 nations while the USA has dropped.
At the level of OECD fields (

Indicator 4.08

) the UK is now 2

to the Netherlands

in natural sciences. Denmark has moved further ahead in engineering, but the
USA has dropped behind the UK. In the social sciences the UK has improved
its performance relative to the OST comparator group average but the USA,
though declining, remains far ahead on this measure. In the humanities the UK
has a clear lead, although the significance of this is open to interpretation.

Theme 5 – Productivity - labour (outputs and outcomes related to other
measures)

The UK is highly productive in terms of labour productivity.

Skilled people are a key contribution to future resources for the knowledge
economy. The UK is 2

in the G8 but has gained further on Germany in terms

of PhDs awarded per researcher (

Indicator 5.01

). Only Switzerland and Spain

are more productive. The UK has maintained its level of output during a period
when the USA slipped back.

On relative productivity (papers published per researcher,

Indicator 5.02

) and

effectiveness (citations acquired per researcher,

Indicator 5.03

) the UK has a

strong lead position in G8 countries (excepting anomalous data for Italy) and
has been 3

overall behind Switzerland and the Netherlands. In the last year,

UK paper productivity has fallen slightly compared with the recent past
(previous five years) but citation acquisition has improved suggesting that
people have produced fewer but better papers.

Theme 6 - People

The UK has a weaker availability of highly skilled people with research training
– in relation to population and to the workforce as a whole - than its
competitors. However, there may be some shift in the skills balance among
research personnel.

We can index each country’s researchers and R&D personnel in relation to the
general population and the national workforce. As we move towards an
increasingly technology and knowledge based economy, these indicators may
take on added significance as a measure of national capacity to absorb and use
knowledge based opportunities. The Scandinavian countries are strong in this
regard.

In the UK, only 0.3% of the population as a whole would be classified as a
researcher (

Indicator 6.01

). This is less than the OST comparator group

average, where the UK is ranked 15

, and is lowest in the G8 where no nation

is much higher than 0.5%. The USA has reached 0.5% and is consistently
improving whereas there has been little change in the last decade for the UK.
Frequency of R&D personnel in the population is higher, typically 0.45%
(

Indicator 6.03

) but the UK is again below OST comparator group and EU

average and is the lowest except for Italy in the G8.

If comparison is made to the size of the national labour-force (

Indicator 6.02,

6.04

), the UK rank has slipped to 17

in the OST comparator group for

researchers (0.6% of labour-force) and 15

for R&D personnel (0.88% down

from 0.92% last year). In both cases, the UK is the lowest ranked except Italy
among G8 nations.

Change in the structure of research personnel in the labour-force is reflected in
a measure of researchers as a proportion of R&D personnel (

Indicator 6.05

This may be an indication of increasing professionalisation of research work.
The UK now ranks 8

in the OST comparator group, ahead of Denmark and

Sweden, and 2

to Japan in the G8. Its share of researchers as a proportion of

R&D personnel is increasing faster than average.

Theme 7 - Business expenditure

The UK has performed relatively well in the past in terms of business
investment in the HE research base as defined by OECD. Recent data suggest
that this investment is not being sustained.

Indicator summary pages

The body of this report is a page by page summary of the detailed quantitative analyses for each indicator. The layout for each page follows a similar pattern:

•

Report on indicator and the headline results.

•

Table of key results (actual values and ranked performance among comparators) for the latest year for which data are available and the average value for the
previous 5 years. The Table also shows current change in performance (ratio of activity between latest and recent, or difference in rank) and then ranks that
change among the OST comparator group and G8 competitors to reflect the relative as well as absolute shift in current UK performance.

•

Charts of data for UK and competitors (usually G8 plus select others) showing trends.

Additional explanatory notes are in the

Background

section at the end of this document.

Description of performance indicator

2 Citations in main research fields (SUoAs): national count and share of OST comparator group total

Condition signalling improvement

Increased national count and share

Biological Sciences

Average

1997 - 2001

Actual

2002

Ratio or

difference

UK - all OST

UK - G8

UK citation count - Actual

168,098.8

10,323

0.0614

OST average citation count

65,102.2

3,814

0.0648

UK citation count - Rank among OST group

UK citation count - Share of OST group

0.100

0.104

1.046

UK citation count - Share of world

0.123

0.133

1.083

Note that this table ranks both performance and change in performance

Ranked change in perf

Citation share among OST group for G8 nations (USA omitted) for SUoA Biological Sciences

0.02

0.04

0.06

0.08

0.1

0.12

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Year

s sha

re of

mpar

tor

oup

s w

ith

CANADA

FRANCE

GERMANY

ITALY

JAPAN

RUSSIA

Indicator Headline

In Biological Sciences, the UK citation
share overall is 2nd to the USA.

The UK's rank citation count remained
2nd highest to the USA for both 1997-
2001 and 2002 periods (difference = 0 -
remaining the same).

The change in UK citation share ratio of
1.046 between periods 1997-2001 and
2002 is second among G8 nations to
Germany which had a ratio of 1.092,
indicating that Germany's citation share is
growing at a slightly faster rate than the
UK.

The UK ranked position of 9th on change
in OST citation share is because other
countries, notably Poland and Spain in
this case, are increasing their citation
share at a faster rate than the UK - even
though their citation counts are smaller
than the UK. The average citation share
ratio of the 8 countries ahead of the UK is
1.331.

Between the periods 1997-2001 and 2002
the USA has shown a decrease in citation
share by a ratio of 0.978.

Although there were reductions in UK
citation share in both 1996 and 2001
there is a slight but progressive rise
across the period.

Graphs may omit USA &

EU 15 if inclusion would

distort vertical axis

Key results for UK (or OST group)

actual and ranked performance in

latest data year and average for

previous 5 years

Indicator identification and

description

One or more graphs illustrating

data scatter and performance

trend of UK and OST group or G8

Text may include commentary on additional issues arising from data or indicators

Change in UK performance

ranked among total OST group

and G8

Change (ratio or difference)

between last year and recent

Headline outcome of

analyses describing UK

status and recent

performance

Description of performance indicator

1.01 GERD relative to GDP (R&D intensity)

Condition signalling improvement

Increased proportion of R&D specific spend

Average

1998 - 2002

Actual

2003

Ratio or

difference

UK GERD as a percentage of GDP - Actual

1.85

1.84

0.99

OST average GERD as percentage of GDP

2.15

2.29

1.06

UK GERD per GDP - Rank

-1

UK GERD per GDP / OST average GERD per GDP

0.86

0.80

0.93

GERD as a percentage of GDP - G8 nations (Russia not included)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

RD as a percentage of GDP

USA

CANADA

FRANCE

GERMANY

ITALY

JAPAN

EUROPEAN
UNION

Data: OECD (MSTI)

Indicator Headline

The relative size of GERD (the most general
measure of the share of total GDP that is spent
on research and development, R&D) is an
indication of the relative research intensity of
the economy as a whole. GERD is an input
measure rather than a measure of research
performance. Nonetheless it is an important
contextual indicator alongside output measures
(Theme 2).

UK GERD makes up about 4% of the
comparator group total (this contrasts with UK
outputs which make up 9% of the comparator
group total outputs - see Indicator 2.03). In
absolute terms, adjusted for PPP, it is just over
10% of that of the USA and is smaller than that
of both France and Germany. In relative terms,
it is a lower percentage of GDP at 1.84% (as
measured in million 2000 $ - constant prices
and PPP, this compares with 1.89% when
measured in million current PPP $) than for the
OST group average and has fallen in recent
years by comparison to that average. It has
also fallen below the average for the EU15
nations.

The most rapid rate of increase in GERD is that
for China, for which no time series data were
available until this year. Although it has only
reached 1.3% of GDP, this is a relative
doubling over the decade and a fourfold
increase in absolute value. China's GERD
($80Bn PPP 2000) now exceeds all but the
USA and Japan and will soon approach 50% of
the EU15 total ($190Bn). We draw attention to
increasing China research activity on a number
of indicators.

cont./

Description of performance indicator

1.02 Publicly performed R&D (PUBERD) as a proportion of GDP

Condition signalling improvement

Increased proportion of R&D specific spend

Average

1998 - 2002

Actual

2003

Ratio or

difference

UK PUBERD as a percentage of GDP - Actual

0.60

0.57

0.96

OST comparator group average PUBERD as percentage of GDP

0.68

0.72

1.07

UK PUBERD per GDP - Rank

-3

UK PUBERD per GDP / OST comparator group av'ge PUBERD per GDP

0.88

0.79

0.90

PUBERD as a percentage of GDP - G8 nations (Russia not included)

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1.1

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

RD as a percentage of GDP

USA

CANADA

FRANCE

GERMANY

ITALY

JAPAN

EUROPEAN
UNION

Data: OECD (MSTI)

Indicator Headline

PUBERD is the sum of GOVERD (expenditure
on R&D performed in the Government sector)
and HERD (R&D in the higher education
sector). This thus accounts for the bulk of
public sector R&D expenditure. This is an
input measure, not an index of performance.

UK PUBERD is only just over 4% of the OST
comparator group total. In terms of PUBERD
relative to GDP, the UK has dropped three
places since 1998 and its PUBERD has fallen
slightly while the group average has increased.
It is therefore now at 80% of group average
whereas it had been ahead of the OST
comparator group average in 1995.

The UK ranks last in the G7 group on PUBERD
relative to GDP. It slipped behind the USA in
2001 and Italy last year. It is well below the
EU15 average.

The main trend over the period has for UK
PUBERD to be relatively static, as it has for
France and Germany. Elsewhere, there have
been significant increases for the USA,
Canada, Scandinavian nations, China, and
other Asian nations.

China is included in this data analysis for the
first time. It has clearly the fastest rate of
increase and its PUBERD has trebled in value
since 1995, increasing by 50% as a share of
GDP. It now stands at $30Bn (2000 PPP)
slightly les than half the total for the EU15
($65Bn). China is on track to overtake the UK
on PUBERD/GDP (this indicator) in the next
few years.

Description of performance indicator

2.01 Number and share of OECD PhD awards

Condition signalling improvement

Increased count and increased share by comparison with competitors

Average

1998 - 2001

Actual

2002

Ratio or

difference

UK PhDs count - Actual

12012

14232

1.18

OST comparator group average PhDs count

8948

9372

1.05

UK PhD count - Ranked performance

UK PhD count - Share OECD comparator nations (%)

8.64

9.76

1.13

Note that this table ranks both performance and change in performance

G8 nations: PhD share among OST comparator group (USA & Canada omitted; no data for Russia)

1999

2000

2001

2002

Year

ount

ry's percent

age share of

l OST

com

arat

or group PhD

UK
USA
FRANCE
GERMANY
ITALY
JAPAN

Data: OECD Education Database

UK - HESA data collection
procedure changed in
2001 (see headline note)

Indicator Headline

There are only 5 years' data on PhD awards in
the OECD Education database. There are also
some data gaps among the OST comparator
group countries. Data are present for G7
countries, although recent years are missing
for Canada. The data have proved stable and
the trends appear fairly level.

The UK's count and share have risen over the
period although this has not affected its ranked
position. Its share of PhD awards is similar to
its share of publication outputs (Indicator 2.03)
and much greater than its share of input
funding (Theme 1).

The USA is the major producer of PhD
students with a slightly declining 44,000 annual
out-turn, which is about 30% of the OST
comparator group total. This compares with
about 50% for the EU15.

Germany (16% down from 18%) ranks 2nd to
the USA, with the UK 3rd at just under 10% of
the group total. A number of countries do not
yet return OECD data on this indicator,
including China and India.

This is an important indicator because people
are a key output from the research base.
Absolute numbers indicate sustainable
capacity, but the share of the group total is
also valuable as a comparator with other
measures of input and output. Highly skilled
people reflect the capacity to make use of
knowledge, where other indicators indicate its
generation and impact. The numbers of people
available to the public and private sector
research base may also be critical to economic
innovation.

Description of performance indicator

2.02 PhDs awarded per head of population

Condition signalling improvement

Increased ratio

Average

1998 - 2001

Actual

2002

Ratio or

difference

UK PhDs awarded per 1000 population - Actual

0.20

0.24

0.04

OST comparator group average - PhDs awarded per 1000 population

0.20

0.21

0.01

UK PhDs awarded per 1000 head of population - Rank

UK PhDs awarded per 1000 population / OST comparator group average
PhDs awarded per 1000 population

1.01

1.12

0.10

PhDs output per capita, G8 nations (no data for Russia)

0.00

0.05

0.10

0.15

0.20

0.25

0.30

0.35

1998

1999

2000

2001

2002

Year

Ds awarded per 1000 popul

ati

UK
USA
CANADA
FRANCE
GERMANY
ITALY
JAPAN
EUROPEAN UNION

Data: OECD (MSTI), OECD Education Database

UK - HESA data collection
procedure changed in 2001
(see headline note)

Indicator Headline

The UK is ranked 2nd behind Germany within
the G8 in terms of PhD awards per head of
population (Germany produced 15% more
PhDs per head of population than the UK,
0.29 c.f. 0.24 PhDs per 1000 population), and
5th amongst all OST comparator group
nations for which there are data. The UK
remained close to the OST comparator group
and EU average throughout the period. Its
apparent improvement in the last two years is
attributable partly to data revision and partly to
cuts elsewhere.

Outside the G8, the OST comparator group
rankings are led by Sweden (average 0.39
PhDs per 1000 population), Switzerland
(0.38), and Finland (0.35). There are no data
for China or India. The EU average is 0.19
and the USA average is 0.15, so Europe is
producing more PhDs and many of these
come from leading research nations.

People are a key output from the research
base and form a key input of highly skilled
people to the workforce. The numbers of
PhDs awarded relative to population size as a
whole is a broad measure of the relative
training capacity and productivity of different
countries. Comparison between trainee
output and the size of the training population is
made in indicator 5.1.

cont./

Description of performance indicator

2.03 Number and share of world publications

Condition signalling improvement

Increased count and increased share by comparison with competitors

Average

1999 - 2003

Actual

2004

Ratio or

difference

UK - all OST

UK - G8

UK publication count - Actual

70348

69419

0.99

32219

33783

1.12

UK publication count - Ranked performance

8.41

7.90

0.94

UK publication count - % share of world

9.26

8.81

0.95

Note that this table ranks both performance and change in performance
Sum of comparator nations exceeds actual world total.
World total is corrected for international co-authorship, which creates duplication in OST total.

Ranked change in performance

UK publication count - % share of OST comparator
group

OST comparator group average publication count

Publication share among OST comparator group for G8 nations (USA omitted)

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

ercentage share of total

T com

arator

group publicat

ions

UK
CANADA
FRANCE
GERMANY
ITALY
JAPAN
RUSSIA

Indicator Headline

Publication share is a measure of research
activity relative to the rest of the world.
Volume is a scale measure of significance and
changes in share are an important indicator of
performance.

The UK is ranked 2nd behind the USA in
national share of global publications for 2004,
the most recent year. This means it has
regained its position relative to Japan.

The UK’s publication count has not increased
markedly, however, since it reached 70,000
outputs per year in 1999. There is some
periodicity in UK output associated with
assessment cycles. From 2001 there may be
some shift from greater volume to better quality
(see later indicators on citations).

Since 1998, the 5 year rolling average growth
for the OST comparator group has been
around 3%. In that period, the UK publication
total has been flat and therefore growth is
consistently less than the group as a whole,
but the USA's annual growth has also fallen
from 1% to 0.2%, and Japan from 5.4% to
2.3%. The UK ranked change in performance
(publication share 2004 compared with 1999-
2003 average) is 22nd among all the nations in
the OST comparator group, up from 25 last
year (France, Germany and Japan rank lower).

The UK had just under 8% of OST comparator
group publications, or 8.8% of global
publications, for 2004 [this difference is
because the sum of the OST comparator group
publications includes some duplication so the
individual national shares of the total are
uniformly depressed compared to true global
values (see methodology notes in
Background)].

Description of performance indicator

2.03 Number and share of world publications

Condition signalling improvement

Increased count and increased share by comparison with competitors

Indicator headline cont.

China has more than trebled its output from
13,500 publications in 1995 to 46,000
publications in 2004 and is ranked 2nd on
growth (Iran is ranked 1st and has increased
nearly ten-fold). Italy (at 10) and Canada (13)
are the only G8 counties ranked better than
20th on performance change. This indicates
that the smaller countries in the OST
comparator group are increasing their
publication share at a faster rate than the G8.

The consequence of the relative changes in
output, and the growth of new research
economies, is that the UK’s outputs will
progressively index less than in the past as a
share of world, although it may be greater
absolutely. It is inevitable that UK share will
continue to fall if growth elsewhere is
sustained, but this need not affect quality.

China’s total is now almost the same size as
France at just over 46,000. Together with
South Korea (19,000), Taiwan (13,000) and
Singapore (5,000), this Asia-Pacific group now
exceeds any one European country in terms of
publication volume. The opportunities for
international collaboration and the implications
for the balance of research endeavour are
likely to be very significant.

In 1998 the EU15 total outputs passed that of
the USA and now clearly exceeds it (2004: EU
= 37.9% of world, USA = 33.6%).

Publication share among fast-growing nations (Russia and China on second axis)

0.5

1.5

2.5

3.5

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

ercentage share of total

T com

arator

group publicat

ions

SPAIN

INDIA

SOUTH
KOREA

BRAZIL

TAIWAN

POLAND

SINGAPORE

IRAN

RUSSIA

CHINA

Description of performance indicator

3.01 Number and share of world citations

Condition signalling improvement

Increased count and increased share by comparison with competitors

Average

1999 - 2003

Actual

2004

Ratio or

difference

UK - all OST

UK - G8

UK citation count - Actual

555364

33426

0.06

OST comparator group average citation count

232286

13955

0.06

UK citation count - Ranked performance

UK citation count - % share OST comparator group

9.14

9.21

1.01

UK citation count - % share of world

11.52

12.23

1.06

Ranked change in performance

Citation share among OST comparator group for G8 nations (USA & EU15 omitted)

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

Percent

age share of

l OST com

arat

or group

cit

ions

UK
CANADA
FRANCE
GERMANY
ITALY
JAPAN
RUSSIA

Data: Thomson
ISI® National
Science Indicators

Indicator headline

Relative citations counts are the most general
measure of research excellence because
esteem is related to citation frequency. The
more times an article is cited by others, the
greater its influence is deemed to be.

The UK has improved on this indicator relative
to 2003. There is significant continuing growth
in China.

The UK is ranked 2nd to the USA in terms of
national share of global citations for 2004, the
most recent year. It has about 12% of global
citations (but 9% of OST comparator group for
reasons noted below).

UK rank has remained 2nd to the USA both as
an average for the recent period (1999-2003)
and for 2004 (difference in table = 0). Citation
counts are fewer in more recent years, but the
rate at which the count falls may vary between
nations.

UK citation share rose slightly between the
period 1999-2003 and 2004. The change (ratio
= 1.01) is better than that of other leading
research nations. Germany had been catching
up to the UK but has now plateaued while
France and Japan have fallen back. The
EU15’s combined share has declined slightly,
but the USA is also dropping – by about 0.5%
per year for several years.

One of the biggest increases is that for China,
which has accelerated to a more than four-fold
change over the decade (from 0.92% to
3.78%) and approaches the citation volume of
Italy (but the publication volume of France -
see 2.03). Ranked 18th in 1995 it now ranks
8th by volume.

cont./

Description of performance indicator

3.02 Number and share of world citations in ten main research fields

Condition signalling improvement

Increased national count and share

Clinical Sciences

Average

1999 - 2003

Actual

2004

Ratio or

difference

UK - all OST

UK - G8

UK citation count - Actual

415,732

22,616

0.05

OST comparator group average citation count

158,355

8,342

0.05

UK citation count - Rank among OST comparator group

UK citation count - % share of OST comparator group

10.05

10.43

1.04

UK citation count - % share of world

12.82

13.86

1.08

Note that this table ranks both performance and change in performance

Ranked change in performance

Citation share among OST comparator group for G8 nations (USA omitted) for SUoA Clinical

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

ount

ry's percent

age share of

l OST

com

arat

or group cit

ions wit

in SU

UK
CANADA
FRANCE
GERMANY
ITALY
JAPAN
RUSSIA

Indicator Headline

See 3.01 for background and 3.03 for
summary.

In Clinical Sciences, the UK citation share
overall is 2nd to the USA. Its share of world
citations has improved compared to the
previous year.

The UK's ranked citation count remained 2nd
highest to the USA for both 1999-2003 and
2004 periods.

UK citation count reduced at a ratio of 0.05
between the periods 1999-2003 and 2004.
This is very similar to the OST comparator
group average. (Citations are always fewer in
more recent years).

The change in UK citation share ratio between
periods 1999-2003 and 2003 (1.04) is second
only to Russia within the G8 nations.
Germany's citation share had been growing at
a slightly faster rate than the UK but this trend
seems to have levelled off while some other
nations have declined slightly.

The USA has shown little change in citation
share between the periods 1999-2003 and
2004 and seems to be stabilising after a long
period of reducing share.

The UK ranked position of 14th on change in
OST comparator group citation share is
because other countries, notably Poland, India,
Singapore and China in this instance, are
increasing their citation share at a faster rate
than the UK - although their citation counts are
smaller than the UK. China’s rapid growth
seems to be levelling off in this field.

Data: Thomson ISI® National Science Indicators

Description of performance indicator

3.02 Number and share of world citations in ten main research fields

Condition signalling improvement

Increased national count and share

Pre-Clinical and Health-Related Sciences

Average

1999 - 2003

Actual

2004

Ratio or

difference

UK - all OST

UK - G8

UK citation count - Actual

54,882

3,341

0.06

OST comparator group average citation count

18,441

1,051

0.06

UK citation count - Rank among OST comparator group

UK citation count - % share of OST comparator group

11.31

12.23

1.08

UK citation count - % share of world

14.66

16.60

1.13

Note that this table ranks both performance and change in performance

Ranked change in performance

Citation share among OST comparator group for G8 nations (USA omitted) for SUoA Pre-Clinical and

Health-Related Sciences

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

ount

ry's percent

age share of

l OST

com

arat

or group cit

ions wit

in SU

UK
CANADA
FRANCE
GERMANY
ITALY
JAPAN
RUSSIA

Data: Thomson ISI® National Science Indicators

Indicator He

adline

See 3.01 for background and 3.03 for
summary.

In the Pre-Clinical and Health-Related
Sciences, the UK citation share overall is 2nd
to the USA.

There is a considerable gap between UK
citation share 1995 - 2004 and the next closest
competitor (Germany). Consequently, the
UK's rank citation count remained 2nd highest
to the USA for both 1999-2003 and 2004
periods (difference in rank = 0).

UK citation count reduced at a rate similar to
the OST comparator group as a whole over the
period 1999-2004.

The improvement in UK citation share ratio of
1.08 between periods 1999-2003 and 2004
places it third among G8 nations behind
Russia and Japan, indicating that these
nations citation share are growing at a faster
rate than the UK.

The USA had shown a decrease in citation
share over the period since 1995 but recovered
slightly in 2004.

The UK ranked position of 12th on change in
OST comparator group citation share is
because other countries, notably China, India
and South Korea in this case, are increasing
their citation share at a faster rate than the UK -
even though their citation counts are less than
the UK. The average citation share growth
ratio of the countries ahead of the UK is +1.44.

Description of performance indicator

3.02 Number and share of world citations in ten main research fields

Condition signalling improvement

Increased national count and share

Biological Sciences

Average

1999 - 2003

Actual

2004

Ratio or

difference

UK - all OST

UK - G8

UK citation count - Actual

208,262

11,724

0.06

OST comparator group average citation count

78,877

4,496

0.06

UK citation count - Rank among OST comparator group

UK citation count - % share of OST comparator group

10.14

10.03

0.99

UK citation count - % share of world

14.36

14.58

1.02

Note that this table ranks both performance and change in performance

Ranked change in performance

Citation share among OST comparator group for G8 nations (USA omitted) for SUoA Biological Sciences

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

ount

ry's percent

age share of

l OST

com

arat

or group cit

ions wit

in SU

CANADA

FRANCE

GERMANY

ITALY

JAPAN

RUSSIA

Data: Thomson ISI® National Science Indicators

Indicator Headline

See 3.01 for background and 3.03 for
summary.

In Biological Sciences, the UK citation share
overall is 2nd to the USA. The UK's rank
citation count remained 2nd highest to the USA
for both 1999-2003 and 2004 periods
(difference in rank = 0). The overall picture for
the G8 has changed little this year.

The marginal change in UK citation share ratio
of 0.99 between periods 1999-2003 and 2004
is fourth among G8 nations behind the USA,
Canada and Russia. Germany, which is third
to the UK and USA in citation share, had a ratio
of 0.98, indicating that its citation share is
affected by similar factors. Canada has
recovered slightly from a period of decline.

The UK rank of 14th on change in OST
comparator group citation share is because
other countries, notably China (three-fold over
ten years), Singapore and Iran in this case, are
increasing their citation share at a faster rate
than the UK - though their citation counts are
smaller than the UK. China (1.6%) now
exceeds Belgium and Denmark and is of
similar scale to Sweden.

The USA has stabilised after a period of
decrease in citation share between 1995 and
2000 and has remained level at just under 40%
since 2001.

Although there are fluctuations in UK citation
share across the period, the trend is for a
consolidated performance at just over 10% of
OST group, but its share of world has actually
improved marginally over the ten year period.
This may reflect increasing UK collaboration
within the G8.

Description of performance indicator

3.02 Number and share of world citations in ten main research fields

Condition signalling improvement

Increased national count and share

Environmental Sciences

Average

1999 - 2003

Actual

2004

Ratio or

difference

UK - all OST

UK - G8

UK citation count - Actual

37,647

2,265

0.06

OST comparator group average citation count

13,554

920

0.07

UK citation count - Rank among OST comparator group

UK citation count - % share of OST comparator group

10.50

9.47

0.90

UK citation count - % share of world

13.40

13.10

0.98

Note that this table ranks both performance and change in performance

Ranked change in performance

Citation share among OST comparator group for G8 nations (USA omitted) for SUoA Environment

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

ount

ry's percent

age share of

l OST

com

arat

or group cit

ions wit

in SU

UK
CANADA
FRANCE
GERMANY
ITALY
JAPAN
RUSSIA

Data: Thomson ISI® National Science Indicators

Indicator Headline

See 3.01 for background and 3.03 for
summary.

In Environment the UK citation share overall is
2nd to the USA. The UK's rank citation count
remained 2nd highest to the USA for both 1999-
2003 and 2004 periods (difference = 0).

The change in UK citation share ratio of 0.90
between periods 1999-2003 and 2004 is 7th
among G8 nations and reflects a progressively
declining share since 2000. Germany’s share
also decreased where it had been growing and
seemed likely to overtake the UK. Canada is
now improving markedly.

The UK ranked position of 21st on change in
OST comparator group citation share is
because other countries, notably South Africa,
Poland and Iran in this case, are increasing
their citation share at a faster rate than the UK -
even though their citation counts are only a
tenth of the UK. The average citation share
ratio of the countries ahead of the UK is 1.1.

The USA has shown a marked decline in
citation share by over 2%, or a ratio of 0.95
between the periods 1999-2003 and 2004.
China citation share has been level at 2.5%
since 2001.

Description of performance indicator

3.02 Number and share of world citations in ten main research fields

Condition signalling improvement

Increased national count and share

Mathematics

Average

1999 - 2003

Actual

2004

Ratio or

difference

UK - all OST

UK - G8

UK citation count - Actual

5,743

339

0.06

OST comparator group average citation count

2,852

156

0.05

UK citation count - Rank among OST comparator group

UK citation count - % share of OST comparator group

7.63

8.36

1.10

UK citation count - % share of world

9.03

10.02

1.11

Note that this table ranks both performance and change in performance

Ranked change in performance

Citation share among OST comparator group for G8 nations (USA omitted) for SUoA Mathematics

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

ount

ry's percent

age share of

l OST

com

arat

or group cit

ions wit

in SU

UK
CANADA
FRANCE
GERMANY
ITALY
JAPAN
RUSSIA

Data: Thomson ISI® National Science Indicators

Indicator Headline

See 3.01 for background and 3.03 for
summary.

In Mathematics, the UK citation share overall
has been 3rd to the USA and (since 1998)
France. It has significantly improved its
position in 2004, with other G8 nations
generally dropping back in share. However, it
is now 3rd to the USA and China which has
moved up to 2nd - with 9.6% of OST group
citations - from 8th in 1995.

The change in UK citation share ratio of 1.10
between periods 1999-2003 and 2004 is 1st
among G8 nations Its position of 6th on
change in OST comparator group citation
share is a marked change from recent years (it
had been 15th last year). Only Belgium and
China were substantially better in improvement.

The USA had suffered a decreasing citation
share since 1995 but this pattern appears to
have stabilised between the periods 1999-2003
and 2004.

The UK's citation share has been more variable
over the period than for some other fields.
From the mid-1990s UK share has fluctuated
whilst France had increased steadily to
overtake the UK. This is an area which needs
further monitoring to verify that the UK's
changed position is real and sustained rather
than a statistical aberration.

Description of performance indicator

3.02 Number and share of world citations in ten main research fields

Condition signalling improvement

Increased national count and share

Physical Sciences

Average

1999 - 2003

Actual

2004

Ratio or

difference

UK - all OST

UK - G8

UK citation count - Actual

154,044

11,671

0.08

OST comparator group average citation count

77,813

5,788

0.07

UK citation count - Rank among OST comparator group

UK citation count - % share of OST comparator group

7.58

7.76

1.02

UK citation count - % share of world

10.65

12.09

1.13

Note that this table ranks both performance and change in performance

Ranked change in performance

Citation share among OST comparator group for G8 nations (USA omitted) for SUoA Physical Sciences

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

ount

ry's percent

age share of

l OST

com

arat

or group cit

ions wit

in SU

UK
CANADA
FRANCE
GERMANY
ITALY
JAPAN
RUSSIA

Data: Thomson ISI® National Science Indicators

Indicator Headline

See 3.01 for background and 3.03 for
summary.

In Physical Sciences, the UK citation share
overall is 4th to the USA, Germany and Japan.

The UK's rank citation count remained 4th
highest to the USA, Germany and Japan for
both 1999-2003 and 2004 periods (difference =
0). The UK appears to have lagged
significantly on these nations in the recent past,
but its relative impact has been well ahead of
Japan and moved ahead of Germany. There is
sustained evidence of slow but continuing
growth in UK citation share, and it is ranked
2nd on change among G8 nations. It has
moved well ahead of France (at 5th in the
group) and seems likely to overtake Japan.

China has moved up from 13th to 6th globally
on share in this area. moving ahead of Italy
with 4.8% of group citations. The UK ranked
position of 12th on change in OST comparator
group citation share is because other countries,
notably China, India and Poland in this case,
are increasing their citation share at a faster
rate than the UK.

USA share has dropped from over 35% to less
than 29%. The EU15, at 38%, has broadly
maintained its total share over the last ten
years.

Description of performance indicator

3.02 Number and share of world citations in ten main research fields

Condition signalling improvement

Increased national count and share

Engineering

Average

1999 - 2003

Actual

2004

Ratio or

difference

UK - all OST

UK - G8

UK citation count - Actual

58,774

3,082

0.05

OST comparator group average citation count

32,026

1,732

0.05

UK citation count - Rank among OST comparator group

UK citation count - % share of OST comparator group

6.90

6.85

0.99

UK citation count - % share of world

8.51

8.69

1.02

Note that this table ranks both performance and change in performance

Ranked change in performance

Citation share among OST comparator group for G8 nations (USA omitted) for SUoA Engineering

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

ount

ry's percent

age share of

l OST

com

arat

or group cit

ions wit

in SU

UK
CANADA
FRANCE
GERMANY
ITALY
JAPAN
RUSSIA

Data: Thomson ISI® National Science Indicators

Indicator Headline

See 3.01 for background and 3.03 for
summary.

In Engineering, the UK citation share overall is
4th to the USA, Japan and Germany.

The UK's rank of citation count remained 4th
highest to the USA, Japan and Germany for
both 1999-2003 and 2004 periods (difference =
0).

While the UK's ranked position is unchanged, a
trend of falling UK share of world citations over
the period from 1994 has been halted and
some uplift is emerging. A growing gap
between the UK and Germany has narrowed.
The flat UK citation share ratio is average
among G8 nations, and ahead of USA, Japan
and France.

China has moved up to 5th in OST group share
with 6.3% of citations, up from 11th in 1995,
and thus just behind the UK.

The UK ranked position of 17th on change in
OST comparator group citation share is
because other countries, notably Singapore,
China, Denmark and Finland in this case, are
increasing their citation share at a faster rate
than the UK - even though their citation counts
are smaller than the UK.

UK average research impact in Engineering
has improved and leading units are
outstanding. We have previously noted the
significant diversity in performance within the
UK research base.

Description of performance indicator

3.02 Number and share of world citations in ten main research fields

Condition signalling improvement

Increased national count and share

Social Sciences

Average

1999 - 2003

Actual

2004

Ratio or

difference

UK - all OST

UK - G8

UK citation count - Actual

9,703

600

0.06

OST comparator group average citation count

3,401

212

0.06

UK citation count - Rank among OST comparator group

UK citation count - % share of OST comparator group

11.08

10.87

0.98

UK citation count - % share of world

11.98

11.90

0.99

Note that this table ranks both performance and change in performance

Ranked change in performance

Citation share among OST comparator group for G8 nations (USA omitted) for SUoA Social Science

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

ount

ry's percent

age share of

l OST

com

arat

or group cit

ions wit

in SU

UK
CANADA
FRANCE
GERMANY
ITALY
JAPAN
RUSSIA

Data: Thomson ISI® National Science Indicators

Indicator Headline

See 3.01 for background and 3.03 for
summary.

Bibliometric data for the Social Sciences
cannot be compared directly with other
disciplines (see Background). Recent work
has shown that European comparisons must
be made with caution because Anglophone
journals are over-represented.

In the Social Sciences, the UK citation share
overall is 2nd to the USA. It remained 2nd
highest to the USA for both 1999-2003 and
2004 periods (difference = 0).

The change in UK citation share ratio between
periods 1999-2003 and 2004 is fifth among G8
nations and follows an earlier period of rapid
UK expansion. Evidence from research
assessment suggests that journal publication
is becoming increasingly important in this field.

The UK ranked position on change in OST
comparator group citation share is only 12th.
A number of other countries, notably Germany
but also Belgium and India in this case, are
increasing their citation share at a faster rate
than the UK - even though their citation counts
are smaller than the UK.

The USA has shown a decrease in citation
share of over 10% between 1995-2004. This is
partly accounted for by substantial shifts in the
database coverage to include a wider range of
European journals. This will make the data
more valuable in the future although it makes
current trends less clear.

China has less than 1% of citations in the OST
comparator group.

Description of performance indicator

3.02 Number and share of world citations in ten main research fields

Condition signalling improvement

Increased national count and share

Business

Average

1999 - 2003

Actual

2004

Ratio or

difference

UK - all OST

UK - G8

UK citation count - Actual

4,683

246

0.05

OST comparator group average citation count

1,625

0.05

UK citation count - Rank among OST comparator group

UK citation count - % share of OST comparator group

11.19

12.45

1.11

UK citation count - % share of world

13.13

15.25

1.16

Note that this table ranks both performance and change in performance

Ranked change in performance

Citation share among OST comparator group for G8 nations (USA omitted) for SUoA Business

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

ount

ry's percent

age share of

l OST

com

arat

or group cit

ions wit

in SU

UK
CANADA
FRANCE
GERMANY
ITALY
JAPAN
RUSSIA

Data: Thomson ISI® National Science Indicators

Indicator Headline

See 3.01 for background and 3.03 for
summary.

Bibliometric data for Business, like the Social
Sciences, should be treated with caution and
cannot be compared directly with other
disciplines.

In Business, the UK citation share overall is
2nd to the USA.

The UK's rank of citation count remained 2nd
highest to the USA for both 1999-2003 and
2004 periods.

The UK ranked position of 13th on change in
OST comparator group citation share is
because other countries, notably Singapore,
South Africa and Switzerland in this case, are
increasing their citation share at a faster rate
than the UK - even though their citation counts
are smaller than the UK.

USA citation share has fallen from 65% to just
over 51% in this area. EU15 share has risen
from less than 20% to 27% over the same
period, of which the UK has been a significant
part. China has more than doubled its share,
but only to 2.6%.

Description of performance indicator

3.02 Number and share of world citations in ten main research fields

Condition signalling improvement

Increased national count and share

Humanities

Average

1999 - 2003

Actual

2004

Ratio or

difference

UK - all OST

UK - G8

UK citation count - Actual

1,915

169

0.09

OST comparator group average citation count

533

0.08

UK citation count - Rank among OST comparator group

UK citation count - % share of OST comparator group

14.33

14.58

1.02

UK citation count - % share of world

15.19

15.34

1.01

Note that this table ranks both performance and change in performance

Ranked change in performance

Citation share among OST comparator group for G8 nations (USA omitted) for SUoA Humanities

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

Country's percentage

share of total O

comparator group citations w

ithin SUoA

UK
CANADA
FRANCE
GERMANY
ITALY
JAPAN
RUSSIA

Data: Thomson ISI® National Science Indicators

Indicator Headline

This is the first year in which any bibliometric
data for the Humanities and Arts have been
analysed in this report. It is necessary to treat
these data with great caution as other analyses
have confirmed the degree to which these
subjects continue to make use of other modes
of output, notably books, for their key
publications.

With that caveat in mind, and recognising that
citation counts are very low compared to the
numbers of researchers working in these
disciplines, we can nonetheless see that the
UK has a strong and improving position in the
database.

That said, it must be noted that non-
Anglophone journal coverage is inevitably
rather poor, and the literature is said to be
parochial where journal content and usage
would be more international for natural
sciences and technology.

We will continue to monitor the UK's position
and will work with other agencies, including the
newly established UK Arts & Humanities
Research Council, to explore the value and
relevance of these and other research
performance indicators in these fields. For the
present, we take the outcome reported here as
a very positive reflection of the health of the UK
research base in this area.

Description of performance indicator

3.03 Rank of share of world citations by nine research fields

Condition signalling improvement

More frequent presence in top three among fields

Average

1999 - 2003

Actual

2004

Ratio or

difference

UK frequency (out of 9) in top 3

6.6

1.061

Indicator Headline

The comparative level of national share of
world citations (3.02) is a good measure of
strength in a research field. This may be an
isolated peak of strength, however, or it may be
typical of general performance.

This indicator extends indicator 3.02 by looking
across the level of nine main research fields to
assess consistency of performance. The fields
roughly correspond to a University ‘Faculty’ or
group of Schools, and are defined by grouping
cognate Units of Assessment (i.e. those units
using similar literature). UoAs are subject
categories used in UK research assessment.

Nine main fields are used rather than ten
because we are as yet uncertain about the
value of national ranked position in regard to
the Humanities.

The measure is simply a count of the
occurrence of a country in the top 3 by rank
citation share. The UK ranks in the top 3 by
citation count for 7 of the nine main research
fields (indicator 3.02). This is a reflection of
strength in depth in comparative international
research performance. This has placed the UK
second to the USA, which wins in all nine
fields, in almost every year of the last ten. It
improved over Germany compared to 2003.

Only G8 countries appear in this ranking, since
smaller research nations are unable to
compete on volume. The performance profile
of Germany is notable, improving from placing
in 4 to placing in 6 fields last year but only 5 in
2004. Canada, by contrast, slipped back and
now ranks in 2-3 fields.

If the EU15 grouping were to be analysed in
the same way as the USA, it too would be
placed in the top 3 in 9 out of 9 categories.

Leading nations for ranked citation volume in nine research fields

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Fre

of oc

rre

top 3

ons

UK
USA
CANADA
FRANCE
GERMANY
JAPAN

Data: Thomson ISI® National Science Indicators

Description of performance indicator
Condition signalling improvement

Increase in citation share compared to share of papers within field

Country

Year

Clinical

Pre-Clinical

Biological

Sciences

Environment Mathematics

Physical

Sciences

Engineering

Social

Science

Business

Humanities

2004

0.38

1.18

2.02

1.43

0.94

1.27

0.27

-1.7

-0.88

0.87

USA

2004

5.39

7.07

8.59

4.63

6.34

8.09

5.36

5.53

11.63

6.73

CANADA

2004

0.37

1.08

-0.62

0.27

-0.34

0.44

-0.32

0.12

-0.71

-0.86

FRANCE

2004

-0.21

-0.73

0.13

0.37

0.45

-0.06

0.44

-0.22

-0.3

-3.98

GERMANY

2004

-0.4

-0.32

0.88

0.92

0.23

0.98

1.49

-1.25

-0.62

-2.46

ITALY

2004

-0.35

-0.49

-0.67

-0.56

-0.07

0.17

-0.35

0.18

-0.62

-0.43

JAPAN

2004

-2.03

-3.48

-1.34

-1.28

-1.89

-1.79

-0.72

-0.45

-0.85

-0.02

RUSSIA

2004

-0.35

-0.13

-1.04

-1.99

-1.35

-2.98

-1.43

-0.18

-0.03

-0.41

AUSTRALIA

2004

-0.26

-0.36

-0.69

0.14

0.23

0.09

-0.15

-0.45

-0.92

-0.4

BELGIUM

2004

0.13

0.26

-0.15

-0.07

0.25

-0.1

0.17

-0.1

-0.02

-0.15

BRAZIL

2004

-0.5

-1.16

-1.15

-0.46

-0.29

-0.51

-0.34

-0.35

-0.2

-0.15

DENMARK

2004

0.15

0.25

-0.06

0.26

0.14

0.26

0.03

-0.1

0.39

FINLAND

2004

0.1

0.16

-0.11

0.07

0.01

-0.02

0.01

0.16

-0.15

0.23

NETHERLANDS

2004

0.25

0.52

0.19

0.26

0.34

0.67

0.48

0.19

-0.27

0.98

POLAND

2004

-0.16

-0.28

-0.79

-0.44

-0.48

-0.63

-0.46

-0.06

-0.03

-0.09

SPAIN

2004

-0.38

-0.36

-1.18

-0.62

-0.36

-0.1

0.09

-0.07

-0.93

-0.92

SWEDEN

2004

-0.02

0.11

0.08

0.47

0.06

0.13

0.28

0.43

-0.12

-0.02

SWITZERLAND

2004

0.41

0.83

0.61

0.65

0.18

0.66

0.64

0.17

0.13

-0.33

CHINA

2004

-0.56

-1.03

-1.22

-1.55

-2.05

-3.34

-2.16

-0.52

-0.85

0.45

INDIA

2004

-0.52

-1.24

-1.93

-1.13

-0.83

-1.35

-0.97

-0.27

-1.37

-0.02

IRAN

2004

-0.08

-0.16

-0.14

-0.1

-0.23

-0.17

-0.15

-0.02

-0.06

0.01

ISRAEL

2004

-0.21

-0.06

0.14

-0.13

0.06

0.18

0.09

-0.41

0.07

SINGAPORE

2004

-0.1

-0.04

-0.03

-0.08

0.19

-0.25

-0.37

-0.15

-0.39

0.13

SOUTH AFRICA

2004

-0.09

-0.13

-0.47

-0.19

-0.07

-0.08

-0.12

-0.16

-0.25

-0.06

SOUTH KOREA

2004

-0.49

-0.85

-0.63

-0.46

-0.94

-0.93

-1.01

-0.23

-0.72

0.4

TAIWAN

2004

-0.48

-0.61

-0.45

-0.41

-0.49

-0.61

-1.05

-0.25

-1.41

0.05

3.04 Share of citations relative to share of publications

Percentage difference between share of citations and share of publications

within OST group of countries for most recent 5 years

The analysis assumes that each country 'uses' journals in a field in the same way. It is, in practice, evident that the general assumption is incorrect for the Social
Sciences and Humanities. Some countries' national publishing bases are less well covered by ISI and only more international work is included in the common
database. In these areas the USA also has an exceptional positive balance and comparative outcomes may thus be further flawed.

Bibliometric data for Social Sciences and Humanities should always be treated with caution and cannot be compared directly with other disciplines. In other
indicators, the analysis for these disciplines focuses essentially on the year to year trend. Only here is the comparison solely between countries.

Indicator Headline

The difference between a country's share of
world papers (sources) and its share of world
citations is like a balance in trading. There is an
investment in publications and there is an
acquisition of international recognition in
citations. Stronger research is reflected in a
greater citation share than source share.

The UK has a positive difference between its
share of world citations and its share of world
sources in eight of ten areas.
The analysis is carried out for a recent five-year
window at the level of Super-UoAs, with
Humanities introduced for the first time this year.

The USA has a strong positive balance in all
areas. Some other countries (Switzerland, the
Netherlands, Denmark, Sweden) also have
positive differences across multiple subject
areas but the absolute values of these
differences are generally less than that of the
UK. Germany has a marked positive balance in
Biological, Environmental and Physical Sciences
and Engineering. China has a growing output
but is generally in citation 'deficit'.

For smaller countries the trading balance is
mostly negative. This reflects the general
scaling relationship between output volume and
citation share, also seen at institutional level
within countries.

The balances in social sciences, business and
humanities are skewed towards the USA. The
balance of journal coverage in these areas is
not as representative as in the natural sciences.

The intention of this indicator is to identify more
effective research systems by comparing each
country’s share of OST group citations with its
share of outputs within a field. Citation impact
(cites per paper) is analysed in 3.09.

Description of performance indicator
Condition signalling improvement

Increase in citation share compared to share of papers within field

3.04 Share of citations relative to share of publications

2004 Share of Sources by Share of Citations

Pre-Clinical

USA

0.0001

0.001

0.01

0.1

0.0001

0.001

0.01

0.1

Share of Sources within OST comparator group

e of

Cit

ions

in O

T c

or grou

2004 Share of Sources by Share of Citations

Clinical

USA

0.0001

0.001

0.01

0.1

0.001

0.01

0.1

Share of Sources within OST comparator group

e of Ci

ons

thi

T c

mpa

tor

oup

2004 Share of Sources by Share of Citations

Biological Sciences

USA

0.0001

0.001

0.01

0.1

0.001

0.01

0.1

Share of Sources within OST comparator group

e of

Cit

ions

in O

T c

or grou

2004 Share of Sources by Share of Citations

Environment

USA

0.0001

0.001

0.01

0.1

0.001

0.01

0.1

Share of Sources within OST comparator group

e of Ci

ons

thi

T c

mpa

tor

oup

Data: Thomson ISI® National Science Indicators

Description of performance indicator
Condition signalling improvement

Increase in citation share compared to share of papers within field

3.04 Share of citations relative to share of publications

2004 Share of Sources by Share of Citations

Mathematics

USA

0.0001

0.001

0.01

0.1

0.0001

0.001

0.01

0.1

Share of Sources within OST comparator group

ons

T c

tor group

2004 Share of Sources by Share of Citations

Physical Sciences

USA

0.0001

0.001

0.01

0.1

0.0001

0.001

0.01

0.1

Share of Sources within OST comparator group

e of

Cit

ions

in O

T c

or grou

2004 Share of Sources by Share of Citations

Engineering

USA

0.0001

0.001

0.01

0.1

0.0001

0.001

0.01

0.1

Share of Sources within OST comparator group

of Ci

ons

thi

T c

mpa

tor

oup

2004 Share of Sources by Share of Citations

Social Science

USA

0.0001

0.001

0.01

0.1

0.0001

0.001

0.01

0.1

Share of Sources within OST comparator group

e of

Cit

ions

in O

T c

or grou

Data: Thomson ISI® National Science Indicators

Description of performance indicator

3.05 Number and share of uncited publications

Condition signalling improvement

Decrease in ratio of uncited to total sources

Part 1 - UK SUoA analysis

NSI5 2000

NSI5 2004

Ratio

UK uncited papers

130544

122771

0.94

UK published papers

338273

351230

1.04

UK uncited papers as a percentage of all sources

38.59

34.95

0.91

NSI5 data make use of papers in a five year period and citations to those papers within the same specified period.
Volume is therefore about five times the annual average in Indicator 2.03

G8 share of world uncited papers (USA & EU15 omitted)

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

of tota

l world unc

ite

CANADA

FRANCE

GERMANY

ITALY

JAPAN

RUSSIA

Data: Thomson ISI® National Science Indicators

Indicator Headline

The status of uncited papers is disputed.
Papers that receive no subsequent attention are
different in kind to papers that are even
infrequently cited. It is argued that negative
reports may be uncited yet can be important in
identifying blind alleys and other unfruitful areas.

Nonetheless, since uncited work has no
measurable influence, it may be inefficient to
invest undue resources in research that has no
apparent value to other researchers. It is
assumed that research management should
work towards a minimisation in the proportion of
papers that are uncited.

The UK has been successful in progressively
reducing both the proportion of its papers that
remain uncited in a five-year period (35% down
from 38.5%) and its share of the world's uncited
papers. Its total output increased but a smaller
number of these papers remained uncited at the
end of the period. It has improved its
performance in this regard relative to the EU
competitors and to Japan.

Note that in this indicator, a time factor needs to
be considered (see background notes). The
proportion of papers published in any one year
that remain uncited in later years will decline
with time. Thus a progressive year by year
decrease would be observed in the proportion of
any cohort that remains uncited. Fixed citing
windows are therefore used here for analyses of
uncited papers in different periods, to allow the
appropriate comparisons to be made. We
compare papers published in a 5 year period
with citations over the same years rather than
the paper's lifetime.

Description of performance indicator

3.05 Number and share of uncited publications

Condition signalling improvement

Decrease in ratio of uncited to total sources

Clinical

Pre-Clinical

& Health

Biological

Sciences

Environment Mathematics

Physical

Sciences

Engineering

Social

Science

Business

Humanities

Total UK

uncited

papers

2000

95220

13039

37691

11622

6251

43883

44802

9223

5511

9444

130544

2003

94139

11991

36651

11797

6237

43072

42360

9695

5632

9358

124904

2004

93175

11665

35761

11576

6261

41891

41119

9803

5753

9585

122771

Percentage change 2000 - 2004

2.19

11.78

5.40

0.40

-0.16

4.76

8.96

-5.92

-4.21

-1.47

6.33

UK - numbers of uncited papers in main subject areas (SUoAs) in overlapping 5 year windows

Indicator Headline cont.

The analysis is carried out both at a national
level for comparisons between countries and at a
more detailed SUoA level to develop our
understanding of the distribution of uncitedness.

Within specific SUoA fields: Pre-Clinical &
Health, Biological Sciences, Physical Sciences
and Engineering all saw a decrease in the
number of uncited papers between the 5 years
preceding 2000 and the 5 years preceding 2004.
This indicates an improvement in performance.

In the Social Sciences, Business and
Humanities' SUoAs there was an increase in the
number of uncited papers but this is
accompanied by increases in volume. The
proportion uncited decreased in all areas and the
UK now has the lowest proportion uncited in
Social Sciences among G8 countries.

Bibliometric data for the Social Sciences and
Humanities should be treated with caution and
cannot be compared directly with other
disciplines.

Description of performance indicator

3.05 Number and share of uncited publications

Condition signalling improvement

Decrease in ratio of uncited to total sources

Part 2 - Comparison between UK and other countries

Percentage of uncited papers by SUoA (NSI5 data)

100

Cli

nic

e-C

linic

olog

ical Sc

viro

ent

hem

ics

ical

ien

ces

gine

erin

cial

ienc

sine

ities

SUoA

ite

d pa

of tota

l s

our

UK 2000
UK 2004
OST Grp 2000
OST Grp 2004

Data: Thomson ISI® National Science Indicators

Indicator Headline

Across almost all subject areas, the UK
produces fewer papers that remain uncited
than does the OST comparator group on
average. The UK had more uncited papers
than average for the first time in Pre-clinical
and health sciences in 2003 and this increased
in 2004.

Changes in the UK's proportion of uncited
papers follow the OST comparator group trend
across most SUoA fields. For both the UK and
OST comparator group average there is
usually a fall between the two five-year periods
to 2000 and to 2004. The exception is the
Humanities, but citation rates are
fundamentally lower in this area. Here too,
however, the UK is more often cited than the
group as a whole or any other G8 country.

Description of performance indicator

3.05 Number and share of uncited publications

Condition signalling improvement

Decrease in ratio of uncited to total sources

G8 - Percentage of papers remaining uncited within period

Clinical Pre-Clinical

Biological

Sciences

Environment Mathematics

Physical

Sciences

Engineering

Social

Science

Business

Humanities

Total

uncited

papers

Five year period ending in 2000

51.8

55.0

40.3

41.3

56.6

46.1

58.3

60.2

79.4

130544

USA

48.4

48.0

40.1

43.7

55.3

45.2

58.2

58.0

57.5

79.4

466361

CANADA

49.6

53.4

42.3

40.9

58.5

45.6

58.5

59.7

60.8

82.3

64556

FRANCE

52.8

56.0

42.9

41.6

58.7

48.1

57.8

71.1

63.3

92.2

94260

GERMANY

51.4

52.2

43.1

43.5

59.9

45.0

56.8

72.4

64.0

86.3

122440

ITALY

53.0

54.5

46.1

48.8

58.9

46.8

59.6

66.5

64.3

87.0

56572

JAPAN

53.2

47.2

48.4

49.5

67.9

52.3

61.5

72.0

68.6

81.6

138398

RUSSIA

69.7

53.5

67.1

66.1

73.1

62.7

71.2

84.3

80.3

93.3

82266

Five year period ending in 2004

48.9

52.9

37.8

37.3

53.3

42.3

54.5

55.4

57.2

77.5

122771

USA

46.3

45.8

39.1

42.0

53.2

42.3

55.4

56.7

55.0

78.8

447314

CANADA

47.5

52.9

41.3

41.1

58.3

44.7

56.4

56.2

60.8

81.6

63910

FRANCE

50.7

53.9

40.7

38.3

54.7

44.5

53.6

64.3

60.8

92.3

89049

GERMANY

47.7

50.3

40.8

38.9

54.9

41.6

52.8

64.7

60.4

83.9

116494

ITALY

49.7

50.1

44.3

44.9

56.7

43.8

56.4

61.7

64.4

83.6

59796

JAPAN

48.6

43.7

46.5

46.7

64.0

48.6

57.9

67.8

67.7

79.1

133342

RUSSIA

68.5

50.4

63.8

66.0

68.9

59.1

65.2

82.8

63.6

91.5

72761

Difference between earlier and later periods (2000 value - 2004 value, +ve result = decrease in uncitedness)

3.0

2.1

2.5

4.0

3.4

3.8

4.8

2.9

1.9

7773

USA

2.1

2.3

1.0

1.7

2.1

2.9

1.4

2.5

0.6

19047

CANADA

2.1

0.5

1.0

-0.2

0.2

0.9

2.1

3.5

0.1

0.7

646

FRANCE

2.2

2.1

2.2

3.2

4.0

3.6

4.3

6.7

2.5

-0.1

5211

GERMANY

3.7

1.9

2.3

4.6

5.0

3.3

4.0

7.7

3.6

2.4

5946

ITALY

3.3

4.4

1.8

3.8

2.2

3.0

3.2

4.7

-0.1

3.4

-3224

JAPAN

4.6

3.5

1.9

2.8

3.9

3.7

3.6

4.2

0.9

2.5

5056

RUSSIA

1.2

3.1

3.2

0.1

4.2

3.6

6.0

1.5

16.8

1.8

9505

Description of performance indicator

3.06 Number and share of cited publications

Condition signalling improvement

Increase in ratio of cited to total sources

Part 1 - UK SUoA analysis

NSI5 2000 NSI5 2004

Ratio

UK cited papers

207729

228459

1.10

UK published papers

338273

351230

1.04

UK cited papers as a percentage of all sources

61.4

65.0

1.06

NSI5 data make use of papers in a five year period and citations to those papers within the same specified period.

G8 share of world cited papers (USA & EU15 omitted)

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

of tota

l world c

ite

d pa

CANADA

FRANCE

GERMANY

ITALY

JAPAN

RUSSIA

Data: Thomson ISI® National Science Indicators

Indicator Headline

Number and share of cited papers is the positive
counterpart of indicator 3.05, the number and
share of uncited papers.

Whereas uncited papers are a well defined
category, cited papers are a rather more diffuse
group including both low and high impact papers.
They are therefore more difficult to define in
terms of a performance indicator.

The UK's cited papers have increased recently at
a relatively faster rate than the total national
output (ratio 2004 vs. 2000).

The UK's cited papers as a proportion of all
outputs has consequently improved from 61 to
65%. The UK's share of world cited papers
increased to 2002 but has now fallen back
slightly.

The general G8 trend has been for an increase
in citedness between the 5 year periods to 2000
and to 2004, with improvements for most EU
countries. The immediate trend is for some
levelling in this pattern.

Citedness for China outputs has improved
substantially between the two five year periods
but remains significantly below the OST group
averages for the present.

Description of performance indicator

3.06 Number and share of cited publications

Condition signalling improvement

Increase in ratio of cited to total sources

Clinical

Pre-Clinical

Biological

Sciences

Environment

Mathematics

Physical

Sciences

Engineering Social Science

Business

Humanities

Total UK

uncited

papers

2000

88451

10669

55803

16534

4786

51299

32057

6098

3651

2452

207729

2003

96798

10527

58942

19218

5490

56411

34536

7651

4210

2689

226837

2004

97398

10379

58828

19449

5493

57152

34304

7889

4299

2788

228459

Percentage change 2000 - 2004

10.1

-2.7

5.4

17.6

14.8

11.4

7.0

29.4

17.7

13.7

10.0

UK - numbers of cited papers in main subject areas (SUoAs) in overlapping 5 year windows

Percentage of cited papers by SUoA (NSI5 data)

Cli

nic

e-C

linic

olog

ical Sc

viro

ent

hem

ics

ical

ien

ces

gine

erin

cial

ienc

sine

ities

SUoA

d paper

s as a per

cent

age of

l sour

ces

UK 2000
UK 2004
OST Grp 2000
OST Grp 2004

Data: Thomson ISI® National Science Indicators

Description of performance indicator

3.07 Number and share of publications in top 1% by citation count

Condition signalling improvement

Increase in share

Country

Papers

exceeding

threshold

Rank by

Papers

Papers as %

share OST

Group

Citations for

these papers

Rank by

Citations

Average

impact of

highly cited

papers

OST group

rank by

impact

G8 rank by

impact

USA

25,746

61.4%

2,208,042

85.76

5,528

13.2%

482,795

87.34

GERMANY

4,715

11.3%

384,599

81.57

JAPAN

2,921

7.0%

251,576

86.13

FRANCE

2,847

6.8%

240,515

84.48

CANADA

2,435

5.8%

214,811

88.22

ITALY

1,878

4.5%

156,234

83.19

SWITZERLAND

1,683

4.0%

148,561

88.27

NETHERLANDS

1,678

4.0%

141,250

84.18

AUSTRALIA

1,353

3.2%

113,567

83.94

CHINA

1,251

3.0%

66,017

52.77

SWEDEN

1,102

2.6%

101,435

92.05

SPAIN

1,100

2.6%

93,051

84.59

BELGIUM

775

1.8%

62,757

80.98

DENMARK

712

1.7%

59,708

83.86

ISRAEL

607

1.4%

56,291

92.74

RUSSIA

586

1.4%

44,730

76.33

SOUTH KOREA

560

1.3%

35,043

62.58

FINLAND

478

1.1%

46,109

96.46

POLAND

374

0.9%

29,984

80.17

INDIA

354

0.8%

26,291

74.27

BRAZIL

321

0.8%

22,805

71.04

TAIWAN

290

0.7%

17,652

79.35

SINGAPORE

169

0.4%

9,878

58.45

SOUTH AFRICA

129

0.3%

8,193

63.51

IRAN

0.0%

405

28.93

EUROPEAN UNION

15,972

1,267,368

0.00

Indicator Headline

Some publications have exceptional citation
rates compared to others in their field. The
national share of the most cited 1% of papers
is an indicator of interest. The data reported
here cover the five-year period 2000-2004.
The citation counts are not directly comparable
to last year because the point at which we
sampled the papers was later in the cycle
(November instead of January). Paper counts
and rank position are comparable, although
the total of papers in the set is greater as
Thomson journal coverage has increased.

The UK has increased its share from 12.9% to
13.2%, which is a further improvement over
each previous 5-year window. The UK is
second in volume of highly cited papers to the
USA. Its share (13.2%) is high relative to its
share of all papers (about 9%, see Indicator
2.03).

The UK has moved from 10th to 6th in the
OST comparator group in terms of average
impact of highly cited papers, and is 2nd to
Canada in the G8. The absolute volume of
other countries is, however, much lower.

On volume, the USA has fallen further back
from its 62.7% last time. The biggest
improvement among the G8 is for Germany, by
a further 0.5% on top of +0.4% last year.
China has also significantly improved its
position, up from 1% two years ago to 3% now.
It continues to have an atypical distribution of
peak activity, in relatively low citation areas,
thus resulting in a low average impact.

cont./

Description of performance indicator

3.08 Citation impact (citations/publication) relative to world baselines

Condition signalling improvement

Increased national impact

Average

1999 - 2003

Actual

2004

Ratio or

difference

UK - all OST

UK - G8

UK impact - Actual

7.91

0.48

0.06

OST comparator group average impact

6.23

0.36

0.06

UK impact - Rank

UK rebased impact - Actual

1.25

1.37

1.10

0.98

1.02

1.05

Note that this table ranks both performance and change in performance

Ranked performance change

OST comparator group average rebased
impact

Rebased average impact (world = 1.0) for G8 nations

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

Rebased I

pact

UK
USA
CANADA
FRANCE
GERMANY
ITALY
JAPAN
RUSSIA
EUROPEAN UNION

Data: Thomson ISI® National Science Indicators

Indicator Headline

The most frequently used index of research
performance is that of impact, measured as
citations per paper. This is widely accepted
internationally as an index of research quality.

The UK has improved its relative international
performance (Rebased Impact (RBI) of 1.37
[rebased against world = 1.0, see below] in
2004, 1.25 recent average) and its overall rank
(now 5th from 7th). It is 2nd to the USA (1.45
average recent impact) among G8 nations,
and is evidently closing the gap as it has done
throughout the last decade.

Germany has significantly improved its
position, from around world average in 1993 to
a 1.34 average RBI in 2003 and will continue
to challenge the UK and the USA.

The USA lies 2nd globally to Switzerland (1.74
average recent impact). Denmark lies 3rd
overall with the Netherlands 4th. Sweden is
6th but with a slightly higher recent average
impact than the UK.

China, which has had a major increase in
investment and output, has an average impact
of around 0.63 and lies 24th in the OST
comparator group. Its average RBI had
improved over the decade but has plateaued in
the last few years.

Citation rates vary between fields, and older
papers have more time to accumulate
citations. A common baseline is therefore
created to compare impact data. This baseline
is the world average figure for the stated field
and year. Setting actual performance figures
against a common reference point is called
‘rebasing’ (or normalisation) and the figure
usually quoted will be ReBased Impact (RBI).
[Data by field are shown in indicator 3.09.]

Description of performance indicator

3.09 Citation impact relative to world baselines in ten main research fields

Condition signalling improvement

Increased impact in main research fields

Clinical Sciences

Average

1999 - 2003

Actual

2004

Ratio or

difference

UK - all OST

UK - G8

UK impact - Actual

10.91

0.60

0.06

OST comparator group average impact

8.83

0.46

0.05

UK impact - Rank

UK rebased impact - Actual

1.18

1.30

1.10

0.96

1.00

1.04

Note that this table ranks both performance and change in performance

Ranked performance change

OST comparator group average rebased
impact

Clinical - Rebased Impact

0.8

1.0

1.3

1.5

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

Rebased I

pact

UK
USA
CANADA
FRANCE
GERMANY
ITALY
JAPAN
NETHERLANDS
EUROPEAN UNION

Data: Thomson ISI® National Science Indicators

Indicator Headline

For Clinical Sciences, the UK is ranked 4th in
terms of impact for 2004, the most recent year.
This is a significant uplift from last year and
places it 2nd among G8 nations to the USA.

The UK has moved ahead of other
competitors, has caught up with Canada as
predicted last year, but remains behind
Switzerland and Belgium within Europe.

UK rebased impact rose from 1.09 in 1995, to
1.18 for the period 1999-2003 and 1.3 for 2004
(world average = 1.0). The step up in 2002
seemed exceptional but is now seen to be part
of a sustained trend. This degree of
performance change (ratio = 1.10) is first
among G8 nations, except Russia where
performance is currently erratic. It is 8th
among the OST comparator group, up from
12th last year, where improvement is led by
South Africa and Poland. These
improvements among smaller nations are
rather variable from year to year.

USA impact has been more or less static.
China had risen above world average in 2003
but has slipped back to an RBI of 0.85 (0.81
over the recent past).

Description of performance indicator

3.09 Citation impact relative to world baselines in ten main research fields

Condition signalling improvement

Increased impact in main research fields

Pre-Clinical & Health-Related Sciences

Average

1999 - 2003

Actual

2004

Ratio or

difference

UK - all OST

UK - G8

UK impact - Actual

12.05

0.79

0.07

OST comparator group average impact

9.28

0.55

0.06

UK impact - Rank

UK rebased impact - Actual

1.30

1.54

1.19

1.01

1.07

1.06

Note that this table ranks both performance and change in performance

Ranked performance change

OST comparator group average rebased
impact

Pre-Clinical - Rebased Impact

0.6

0.8

1.0

1.2

1.4

1.6

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

Rebased I

pact

UK
USA
CANADA
FRANCE
GERMANY
ITALY
JAPAN
NETHERLANDS
EUROPEAN UNION

Data: Thomson ISI® National Science Indicators

Indicator Headline

In Pre-Clinical & Health-Related Sciences, the
UK is ranked 3rd in terms of impact for 2004,
the most recent year. Among G8 nations, it is
now ahead of both the USA and Canada. The
UK's performance has improved markedly
throughout the recent period (1999 to present).

UK rebased impact rose from 1.2 in 1996, to
1.3 on average for the period 1999-2003 and
was 1.54 in 2004 (world average = 1.0). The
degree of rebased performance change (ratio
= 1.19) is 6th among OST comparator group
nations, and 3rd to Japan and Russia within
the G8 group.

USA impact has been fairly flat through the
last ten years. The EU15 has shown a slow
but progressive average improvement. China
does not have a major research presence in
this area.

In 2003, a number of less research intensive
countries showed an exceptional improvement
in citation impact in this area. In Europe, the
Netherlands, Denmark and Switzerland are
leaders, but only the latter sustained this
improvement into 2004.

Japan's improvement is particularly marked
while France has performed rather less well in
this area than would be expected for its
research base generally.

Description of performance indicator

3.09 Citation impact relative to world baselines in ten main research fields

Condition signalling improvement

Increased impact in main research fields

Biological Sciences

Average

1999 - 2003

Actual

2004

Ratio or

difference

UK - all OST

UK - G8

UK impact - Actual

10.89

0.64

0.06

OST comparator group average impact

7.24

0.40

0.06

UK impact - Rank

UK rebased impact - Actual

1.28

1.40

1.09

0.84

0.88

1.04

Note that this table ranks both performance and change in performance

Ranked performance change

OST comparator group average rebased
impact

Biological Sciences - Rebased Impact

0.6

0.8

1.0

1.2

1.4

1.6

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

Rebased I

pact

UK
USA
CANADA
FRANCE
GERMANY
ITALY
JAPAN
NETHERLANDS
EUROPEAN UNION

Data: Thomson ISI® National Science Indicators

Indicator Headline

In Biological Sciences, the UK is ranked 3rd.
It has sustained the trend of the last ten years
and has caught up with the USA in terms of
performance. It remains behind Switzerland.

A possibly maverick result in 2004 (based on
a small number of papers) is that of
Singapore which has moved into 2nd place,
from an average of 9th in the recent past.
This unexpected result will need to be
reviewed next year.

There has been a relative international
improvement in UK rebased impact which
rose from 1.28 to 1.4 between the period
1999-2003 and 2004. UK average of 1.28 for
the recent period compares to 1.34 for the
USA and 1.44 for Switzerland. However,
neither of these countries has shown the
trend of improvement that the UK has
achieved. Only Germany's relative position
has improved as consistently over the period.

This is an area of strength in terms of volume
and performance for the UK. It is well ahead
of OST comparator group average and only
those nations mentioned above have
consistently competitive performances. It
makes a significant contribution to the overall
performance of the EU15 which has
otherwise slipped back slightly.

Description of performance indicator

3.09 Citation impact relative to world baselines in ten main research fields

Condition signalling improvement

Increased impact in main research fields

Environmental Sciences

Average

1999 - 2003

Actual

2004

Ratio or

difference

UK - all OST

UK - G8

UK impact - Actual

6.09

0.37

0.06

OST comparator group average impact

4.96

0.30

0.06

UK impact - Rank

-1

UK rebased impact - Actual

1.23

1.36

1.10

0.992

1.10

Note that this table ranks both performance and change in performance

Ranked performance change

OST comparator group average rebased
impact

Environment - Rebased Impact

0.6

0.8

1.0

1.2

1.4

1.6

1.8

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

Rebased I

pact

UK
USA
CANADA
FRANCE
GERMANY
ITALY
JAPAN
NETHERLANDS
EUROPEAN UNION

Data: Thomson ISI® National Science Indicators

Indicator Headline

Research performance in the Environmental
Sciences has been quite variable between
years. The UK has improved relative to the
USA and Germany, but its ranked impact in the
OST comparator group has slipped to 9th.

UK rebased impact has risen from below 1.1 in
the early 1990s to 1.23 for the period 1999-
2003 and 1.36 in 2004. Nonetheless, the
degree of rebased performance change (ratio =
1.10) is only 12th among OST comparator
group nations. This is because a number of
smaller nations (including South Africa, for
which we noted last year that the values
appear exceptional) have improved their
performance to an even greater degree. These
include the Netherlands and the Scandinavian
group.

While the UK has shown consistent
improvement in relative international
performance over the period, and has thus
overtaken the USA, this is a field in which a
number of countries have similar performance
and the degree of competition is therefore
marked. The Scandinavian countries have
better impact than the G8 lead group.
Germany's recent performance had taken it
well ahead but it has now been pegged back.
EU networks will, of course, allow the research
elsewhere in Europe to be accessed by other
nations in global environment networks.

Description of performance indicator

3.09 Citation impact relative to world baselines in ten main research fields

Condition signalling improvement

Increased impact in main research fields

Mathematics

Average

1999 - 2003

Actual

2004

Ratio or

difference

UK - all OST

UK - G8

UK impact - Actual

2.46

0.15

0.06

OST comparator group average impact

2.14

0.11

0.05

UK impact - Rank

UK rebased impact - Actual

1.17

1.38

1.18

1.02

1.00

0.98

Note that this table ranks both performance and change in performance

Ranked performance change

OST comparator group average rebased
impact

Mathematics - Rebased Impact

0.6

0.8

1.0

1.2

1.4

1.6

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

Rebased I

pact

UK
USA
CANADA
FRANCE
GERMANY
ITALY
JAPAN
NETHERLANDS
EUROPEAN UNION

Data: Thomson ISI® National Science Indicators

Indicator Headline

In Mathematics, the UK is ranked 4th in terms
of impact for 2004, the most recent year.
There has been a notable improvement in the
last two years and the UK's rank has changed
from 8th for the recent period (1999-2003).

UK rebased impact has fluctuated in this area.
It dropped to and moved around 1.2 for much
of the last ten years but has risen steeply
since 2002. The degree of rebased
performance change this year is 2nd among
OST comparator group nations, and 1st within
the G8 group. This suggests something of a
welcome renaissance amongst UK
mathematics.

Mathematics is a field in which smaller
countries can be competitive. Singapore had
improved to achieve a similar impact to the UK
but this year has fallen back. Australia has
also shown a sawtooth performance. Belgium,
Denmark and Switzerland have all been
ranked 1st in the recent past. China has
improved to around world average.

The USA was consistently around 4th among
the OST comparator group. Within the G8,
France improved its performance to 2001 but
then fell back while Germany has shown
significant recent improvement.

Description of performance indicator

3.09 Citation impact relative to world baselines in ten main research fields

Condition signalling improvement

Increased impact in main research fields

Physical Sciences

Average

1999 - 2003

Actual

2004

Ratio or

difference

UK - all OST

UK - G8

UK impact - Actual

7.76

0.60

0.08

OST comparator group average impact

6.22

0.43

0.07

UK impact - Rank

UK rebased impact - Actual

1.37

1.67

1.22

1.09

1.21

1.11

Note that this table ranks both performance and change in performance

Ranked performance change

OST comparator group average rebased
impact

Physical Sciences - Rebased Impact

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

Rebased I

pact

UK
USA
CANADA
FRANCE
GERMANY
ITALY
JAPAN
NETHERLANDS
EUROPEAN UNION

Data: Thomson ISI® National Science Indicators

Indicator Headline

In Physical Sciences, the UK is ranked 5th in
terms of impact for 2004. The UK's rank has
improved from 9th in the mid-1990s to 7th on
average for the recent period (1999-2003).

There has been a sustained rise in UK
rebased impact from 1.14 in 1994 to 1.37 in
the period 1999-2003 and an exceptional 1.67
in 2004, overtaking Canada (1.45) which fell
back steeply this year. The UK's recent
rebased performance change is 4th among
OST comparator group nations. Whether the
upturn in performance for the Netherlands is
real or artefactual will need examination next
year, as will the upward move for the UK,
Germany, France and Italy.

Physical Sciences are an important part of the
underpinning research for engineering and
technology development. The UK's
improvement in performance and its relative
citation impact should be seen against a
backdrop of a relatively low share of world
citations (Indicator 3.02), which may reflect a
limited capacity.

China has increased capacity but not
performance in this area and it remains at
around 0.7 RBI against the world average.
Denmark, Israel and Australia have been
strong performers in recent years.

Description of performance indicator

3.09 Citation impact relative to world baselines in ten main research fields

Condition signalling improvement

Increased impact in main research fields

Engineering

Average

1999 - 2003

Actual

2004

Ratio or

difference

UK - all OST

UK - G8

UK impact - Actual

3.78

0.21

0.06

OST comparator group average impact

3.62

0.19

0.05

UK impact - Rank

UK rebased impact - Actual

1.06

1.21

1.14

1.01

1.07

1.06

Note that this table ranks both performance and change in performance

Ranked performance change

OST comparator group average rebased
impact

Engineering - Rebased Impact

0.8

1.0

1.2

1.4

1.6

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

Rebased I

pact

UK
USA
CANADA
FRANCE
GERMANY
ITALY
JAPAN
NETHERLANDS
EUROPEAN UNION

Data: Thomson ISI® National Science Indicators

Indicator Headline

In Engineering, the UK is ranked 9th in terms
of impact for 2004, the most recent year. The
UK's rank has improved from 14th in the early
1990s and an average of 11th for the recent
period (1999-2003).

UK rebased impact rose from close to world
average in the mid-1990s to 1.21 for 2003
(world average = 1.0). The degree of rebased
performance change (ratio = 1.14) is 5th
among OST comparator group nations, and 1st
within the G8 group (excepting Russia).
There has been a relative international
improvement in the UK's performance, which
has been behind not only the USA among G8
nations but also its close European
competitors. The UK is now well ahead of
EU15 average and has clearly overtaken
France, with a sustained improvement over
three years. Its performance remains behind
some smaller European nations, however,
including the Scandinavian countries (Denmark
ranks 1st with RBI > 2), Belgium and the
Netherlands as well as Switzerland and
Germany.

China has improved its RBI from about 0.7 to
about 0.8 but remains at 20th in the OST
comparator group.

Description of performance indicator

3.09 Citation impact relative to world baselines in ten main research fields

Condition signalling improvement

Increased impact in main research fields

Social Sciences

Average

1999 - 2003

Actual

2004

Ratio or

difference

UK - all OST

UK - G8

UK impact - Actual

2.86

0.17

0.06

OST comparator group average impact

2.78

0.13

0.05

UK impact - Rank

UK rebased impact - Actual

0.92

0.94

1.02

0.88

0.74

0.84

Note that this table ranks both performance and change in performance

Ranked performance change

OST comparator group average rebased
impact

Social Science - Rebased Impact

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

Rebased I

pact

UK
USA
CANADA
FRANCE
GERMANY
ITALY
JAPAN
NETHERLANDS
EUROPEAN UNION

Data: Thomson ISI® National Science Indicators

Indicator Headline

As we have noted in Indicator 3.02, bibliometric
data for the Social Sciences should be treated
with caution and cannot be compared directly
with other disciplines. While the comparisons
here are like-for-like, journal coverage for non-
Anglophone European nations is probably in
deficit so international comparisons must be
interpreted with caution.

The UK is ranked 9th in terms of impact for
2003, the most recent year. Rank position is
more volatile in this field (e.g. see Italy) than in
the Natural Sciences but the UK's position as
well as performance has been consistent over
the decade.

UK rebased impact was below world average,
although it improved between the period 1999-
2003 and 2004. It is placed behind a number of
smaller nations which have impact above world
average. Coverage for these nations may be
selective, Less international non-Anglophone
journals and their contents may be absent from
the database. Thus, whereas the UK has a
diversity of journal articles in the database, for
some smaller countries only exceptional items
are covered and only the peak of performance
is analysed.

Overall, the UK's trend against this backdrop is
more important than its position. To that extent,
the improvement in its impact and rank are a
positive.

Description of performance indicator

3.09 Citation impact relative to world baselines in ten main research fields

Condition signalling improvement

Increased impact in main research fields

Business

Average

1999 - 2003

Actual

2004

Ratio or

difference

UK - all OST

UK - G8

UK impact - Actual

2.40

0.12

0.05

OST comparator group average impact

2.14

0.09

0.04

UK impact - Rank

UK rebased impact - Actual

1.00

1.16

0.85

0.93

1.09

Note that this table ranks both performance and change in performance

OST comparator group average rebased
impact

Ranked performance change

Business - Rebased Impact

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

Rebased I

pact

UK
USA
CANADA
FRANCE
GERMANY
ITALY
JAPAN
NETHERLANDS
EUROPEAN UNION

Data: Thomson ISI® National Science Indicators

Indicator Headline

The UK is ranked 8th in terms of impact for
2004. The UK's rank is similar to that for the
recent period (1999-2003). While the UK has
been affected by a volatile performance profile
with a peak in 1998 and a dip in 2000, it now
seems to be on an upward trajectory.

UK rebased impact rose from below world
average in 1994 to 1.1 in 1998 (world average
= 1.0). The degree of rebased performance
change (ratio = 1.01) is 7th among OST
comparator nations, and 2nd only to Russia
within the G8 comparator group.

There are similar fluctuations to those of the
UK in some other G8 countries (see chart) and
some very extreme variations in smaller
countries (possibly due to the effects of small
numbers of papers). Elsewhere, Belgium and
Switzerland have performed well on average
but otherwise only Sweden performs above
world average for any extended period.

The USA has had a consistently strong position
and dominates business and management
coverage. This reflects the USA's dominance
of the recorded and indexed literature in these
fields and its influence on journal coverage.

Description of performance indicator

3.09 Citation impact relative to world baselines in ten main research fields

Condition signalling improvement

Increased impact in main research fields

Humanities

Average

1999 - 2003

Actual

2004

Ratio or

difference

UK - all OST

UK - G8

UK impact - Actual

0.78

0.06

0.08

OST comparator group average impact

0.81

0.06

0.08

UK impact - Rank

UK rebased impact - Actual

1.17

1.15

0.99

1.15

1.00

Note that this table ranks both performance and change in performance

OST comparator group average rebased
impact

Ranked performance change

Indicator Headline

This is the first time that the Humanities and
Arts have been covered for bibliometric
analyses in this report. The data must be
interpreted with caution.

Paper counts are often rather low in the
national samples. We know that journal
outputs are not usually a main mode of
publication in these fields and that citation
behaviour is different to the sciences. We are
also aware that journal coverage is dominated
by Anglophone sources and that there may be
only limited input from non-European and non-
American nations, perhaps only the very peak
of their country's output.

In the context of those caveats the UK's
position in the chart accompanying this text
seems unexceptional. Its impact is above
world average, is ahead of many G8
competitors and compares well with the USA
which has much more diverse database
coverage.

No immediate conclusions should be drawn but
this indicator will be of widespread interest and
we will monitor its change for future reports.

Humanities - Rebased Impact

0.2

0.6

1.0

1.4

1.8

2.2

2.6

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

Rebased I

pact

UK
USA
CANADA
FRANCE
GERMANY
ITALY
JAPAN
NETHERLANDS
EUROPEAN UNION

Data: Thomson ISI® National Science Indicators

Description of performance indicator

3.10 Variety and consistency of research strength

Condition signalling improvement

Reduced ratio between variance and average. Maximised ratio of average/variance.

Country

5 years to

Average

1/Variance

Country

5 years to

Average

1/Variance

2000

1.14

40.30

2004

1.18

36.52

USA

2000

1.33

67.01

USA

2004

1.31

62.96

CAN

2000

1.05

32.52

CAN

2004

1.06

27.00

FRA

2000

0.86

11.03

FRA

2004

0.88

9.54

GER

2000

0.94

9.13

GER

2004

1.01

11.68

ITA

2000

0.87

19.57

ITA

2004

0.96

34.51

JAP

2000

0.87

38.85

JAP

2004

0.89

37.13

NED

2000

1.33

22.32

NED

2004

1.39

9.56

SUI

2000

1.12

2.97

SUI

2004

1.23

6.10

Rebased bibliometric impact across SUoAs

Analysis of rebased impact across SUoAs for select members of OST comparator group

Performance is maximised by combining high average impact with low variability

SUI

NED

JAP

ITA

GER

FRA

CAN

USA

SUI

NED

JAP

ITA

GER

FRA

CAN

USA

0.00

10.00

20.00

30.00

40.00

50.00

60.00

70.00

80.00

0.70

0.80

0.90

1.00

1.10

1.20

1.30

1.40

Average of rebased impact across SUoAs

ari

ance i

reb

ased i

pact across

2000

2004

Data: Thomson ISI® National Science Indicators

Indicator Headline

See below for Methodology.

It is argued that a balanced research economy
has both strength and evenness. That is, high
average quality across fields would be
enhanced by low variation between them.

This is a desirable combination because
research strength should not be unduly
concentrated. Investing in priorities is sensible
but research is long term; capacity takes years
to build. Not all needs and opportunities can
be foreseen, so the flexibility to shift into new
areas is worth paying for. It contributes to a
national capacity to appraise research
developments in other countries and to
respond to research opportunities.

We can visualise this characteristic of the
research base by looking simultaneously at
average impact and the reciprocal of variance
across fields. (We use the reciprocal because
we are interested in systems that minimise
variation; this is one of several alternative
measures of variability). We then look for
countries that maximise the net product.

For the UK, average impact is relatively high
compared to most G8 countries but less than
the USA. It is also less than the Netherlands
(NED) and Switzerland (SUI). Both the smaller
countries have greater variance between
disciplines, however, which suggests that their
research economy is relatively concentrated.

In the graph, the UK shows a positive shift in
average impact with a decrease in variance for
the two 5 year periods to 2000 and to 2004.
For the Netherlands impact also increases (as
it does for Switzerland), but variance increases
markedly (decrease in 1/variance), reducing
the net product.

Description of performance indicator

3.11 Relationship between distribution of research training across subjects and research quality

Condition signalling improvement

Improved match in distribution; improved research training power (product of volume and quality)

Medical

Sciences

Natural

Sciences

Engineering

and

Technology

Social

Sciences

Humanities

UK average rebased impact 1998 - 2002

1.18

1.32

1.08

1.07

1.16

UK average PhDs awarded 1998 - 2002

1825

4823

2004

2179

1582

0.94

0.96

0.99

0.91

0.97

1415

2356

1132

1858

937

OST comparator group average PhDs awarded
1998 - 2002

OST comparator group average rebased impact
1998 - 2002

Indicator Headline

What is the quality of the researchers we train?

People are a key output from the research base and a
major route for the transfer of knowledge and know-
how from producers to users. Highly trained people
with PhDs are a specialised part of that output. Direct
measurement of people quality is unfeasible, but we
can measure the association between the distribution
of training and research quality.

This year, Humanities & Arts data have been added to
the analytical coverage.

For the UK, there is a relative concentration of PhD
awards in Natural Sciences where average research
impact rebased against world benchmarks is also
highest. The data points are too few to calculate a
correlation but the general relationship between
concentration of training and performance appears
positive.

For the USA, the relationship between training volume
and research performance is positive only for sciences
and technology and a relative excess of PhDs is
awarded in the Social Sciences and, to a lesser extent,
in the Humanities and Arts.

There is no clear relationship between training
concentration and research quality for other G8
countries for which data are available and there is a
negative relationship for the OST comparator group
average. For Germany and Japan, too, there appear
to be relatively more PhDs awarded in fields where
rebased impact is relatively lower.

If the benefits of training are related to research
quality, then this positive relationship for the UK should
bring wider benefits to the research base. This is
therefore a marker of relative consistency in the
structure of the research base.

Indicator Headline cont.

We cannot assert that a country that manages its
training so as to ensure that highly qualified people
benefit from a rich research environment will
necessarily benefit at an economic level, but it seems
reasonable to suppose that where training is
associated with research quality this is at least unlikely
to be a disbenefit.

For this indicator, data are taken from the OECD
Education Database. These data are disaggregated
by fields that are most readily mapped to other OECD
fields rather than SUoAs.

Detailed exploration and consideration of the data may
reveal additional features. It is possible that a national
policy might actually advocate increased PhD output in
areas of relative weakness, to increase the pool of
trained people.

Description of performance indicator

3.11 Relationship between distribution of research training across subjects and research quality

Condition signalling improvement

Improved match in distribution; improved research training power (product of volume and quality)

UK 2002 - rebased impact vs. PhDs awarded by OECD field

Humanities

Social Sciences

Medical Sciences

Engineering &

Technology

Natural Sciences

1000

2000

3000

4000

5000

6000

7000

8000

0.8

0.9

1.1

1.2

1.3

1.4

1.5

1.6

Rebased Impact

PhDs awarded

USA 2002 - rebased impact vs. PhDs awarded by OECD field

Humanities

Social Sciences

Medical Sciences

Engineering &

Technology

Natural Sciences

2000

4000

6000

8000

10000

12000

14000

16000

18000

20000

0.8

0.9

1.1

1.2

1.3

1.4

1.5

1.6

Rebased Impact

ded

France 2002 - rebased impact vs. PhDs awarded by OECD field

Humanities

Social Sciences

Medical Sciences

Engineering &

Technology

Natural Sciences

1000

2000

3000

4000

5000

6000

7000

8000

0.8

0.9

1.1

1.2

1.3

1.4

1.5

1.6

Rebased Impact

PhDs awarded

Data: Thomson ISI® National Science Indicators,

OECD Education Database

Description of performance indicator

3.11 Relationship between distribution of research training across subjects and research quality

Condition signalling improvement

Improved match in distribution; improved research training power (product of volume and quality)

Germany 2002 - rebased impact vs. PhDs awarded by OECD field

Humanities

Social Sciences

Medical Sciences

Engineering &

Technology

Natural Sciences

1500

2500

3500

4500

5500

6500

7500

8500

9500

0.8

0.9

1.1

1.2

1.3

1.4

1.5

1.6

Rebased Impact

PhDs awarded

Italy 2002 - rebased impact vs. PhDs awarded by OECD field

Humanities

Social Sciences

Medical Sciences

Engineering &

Technology

Natural Sciences

1000

2000

3000

4000

5000

6000

7000

8000

0.8

0.9

1.1

1.2

1.3

1.4

1.5

1.6

Rebased Impact

ded

Japan 2002 - rebased impact vs. PhDs awarded by OECD field

Humanities

Social Sciences

Medical Sciences

Engineering &

Technology

Natural Sciences

1000

2000

3000

4000

5000

6000

7000

8000

0.6

0.7

0.8

0.9

1.1

1.2

1.3

1.4

1.5

1.6

Rebased Impact

PhDs awarded

OST comparator group average 2002 - rebased impact vs. PhDs

awarded by OECD field

Humanities

Social Sciences

Medical Sciences

Engineering &

Technology

Natural Sciences

1000

2000

3000

4000

5000

6000

7000

8000

0.8

0.9

1.1

1.2

1.3

1.4

1.5

1.6

Rebased Impact

PhDs awarded

Data: Thomson ISI® National Science Indicators,

OECD Education Database

Description of performance indicator

4.01 PhDs awarded relative to HERD

Condition signalling improvement

Increased ratio

Average

1998 - 2001

Actual

2002

Ratio or

difference

UK PhDs awarded per $M HERD - Actual

2.15

1.00

OST comparator group average - PhDs awarded per $M HERD

1.85

1.77

0.96

UK PhDs awarded per unit $M HERD - Rank

G8 nations and EU15: PhDs per HERD (no data for Russia)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

1998

1999

2000

2001

2002

Year

Ds awarded per HE

RD $M

2000 P

UK
USA
CANADA
FRANCE
GERMANY
ITALY
JAPAN
EUROPEAN UNION

Data: OECD (MSTI), OECD Education Database

UK - HESA data collection
procedure changed in 2001
(see headline note)

Indicator Headline

This indicator compares the output of people
with research degrees with the specific
research spend in HE (HERD).

The UK produces more PhDs relative to HERD
than the OST comparator group on average
(for the 17 countries for which data were
available). It is ranked 5th, behind Germany,
South Korea and Spain, with a similar output to
Switzerland and Finland. The EU15 is
significantly more productive than the USA,
which is ranked 11th.

The UK produces slightly in excess of 8% of
the world's PhDs, rather more than most of the
G8 nations (Indicator 2.01). The exceptions
are Germany and the USA who produce twice
and four times as many PhDs than the UK
respectively.

The relative volume of PhDs for each nation is
otherwise broadly similar to that for publication
output. Note that the general trend for
decreasing relative output is despite the data
being adjusted for PPP.

Highly skilled postgraduates are a key output
of the higher education sector. They transfer
knowledge to users and, more generally, they
transfer know-how and technological
advances. It should not be assumed that all
science and engineering postgraduates are
necessarily employed in science and
technology let alone research. More generally,
the gain to the economy is in having a trained
and technological workforce capable of
assessing and responding to technology
related opportunities and issues.

cont./

Description of performance indicator

4.02 PhDs awarded relative to HERD in five main research fields

Condition signalling improvement

Increased ratio

Medical Sciences

Average

1998 - 2001

Actual

2002

Ratio or

difference

UK PhDs awarded per HERD ($M 2000 PPP) - Actual

1.09

1.28

1.17

1.92

1.81

UK PhDs awarded per unit HERD ($M 2000 PPP) - Rank

OST comparator group average - PhDs awarded per unit HERD
($M 2000 PPP)

Indicator Headline

The indicator compares HERD expenditure
allocated to the Medical Sciences with the
numbers of PhDs awarded in this category.
Medical Sciences include both clinical and pre-
clinical areas, nursing and health.

The UK produces fewer PhDs per unit spend
than the OST comparator group average.
This is a much lower level of productivity per
unit spend than Germany but similar to that of
Finland, Australia and, amongst the G8
nations, Japan. The USA is substantially
below the group average.

UK PhD data are subject to a change in HESA
data collection from 2001 (see Background).
In this field, the UK continues its recent trend
of increasing PhD awards (rising from 2001 by
more than 10%). The UK remains the 4th
largest - 2,300 (behind Germany - 8,400,
Japan - 4,300 and USA - 3,800) by volume.

There is only a poor correlation across
countries between funding and PhD output,
implying that although both may scale with an
underlying size factor such as staff numbers,
there are other factors affecting training and
research volumes.

Medical Sciences - PhDs per HERD, G8 nations (no data for Canada, France, Italy or Russia)

0.0

1.0

2.0

3.0

4.0

5.0

6.0

1998

1999

2000

2001

2002

Year

Ds aw

ard

r HERD $M 2000 PPP

UK
USA
GERMANY
JAPAN
EUROPEAN UNION

Data: OECD (RDS), OECD Education Database, UK SET statistics and HESA data

Description of performance indicator

4.02 PhDs awarded relative to HERD in five main research fields

Condition signalling improvement

Increased ratio

Natural Sciences

Average

1998 - 2001

Actual

2002

Ratio or

difference

UK PhDs awarded per HERD ($M 2000 PPP) - Actual

2.49

2.59

1.04

1.83

1.76

UK PhDs awarded per unit HERD ($M 2000 PPP) - Rank

-1

OST comparator group average - PhDs awarded per unit HERD
($M 2000 PPP)

Natural Sciences - PhDs per HERD, G8 nations (no data for Canada, France, Italy or Russia)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

1998

1999

2000

2001

2002

Year

award

r H

2000 PPP

UK
USA
GERMANY
JAPAN
EUROPEAN UNION

Indicator Headline

The indicator compares HERD expenditure
allocated to the Natural Sciences with the
numbers of PhDs awarded in this category.
Here, the Natural Sciences include biological,
physical, environmental and agricultural fields.

The UK produces about 1.5 times as many
PhDs per unit spend as the OST comparator
group average. This is a similar level of
productivity per unit spend to Germany. The
USA is just below the group average and
Japan are just above the group average.

UK PhD data are subject to a change in HESA
data collection from 2001 (see Background).
In this field, the UK currently awards over
5,500 PhDs per year and is the 3rd largest
(behind USA - 11,000 and Germany - 7,000)
by volume.

There is a clear correlation across countries
between funding and PhD output, implying
that probably both scale with an underlying
size factor such as staff numbers, which affect
training and research volumes.

Data: OECD (RDS), OECD Education Database, UK SET statistics and HESA data

Description of performance indicator

4.02 PhDs awarded relative to HERD in five main research fields

Condition signalling improvement

Increased ratio

Engineering and Technology

Average

1998 - 2001

Actual

2002

Ratio or

difference

UK PhDs awarded per HERD ($M 2000 PPP) - Actual

1.99

1.69

0.85

1.56

1.35

UK PhDs awarded per unit HERD ($M 2000 PPP) - Rank

-3

OST comparator group average - PhDs awarded per unit HERD
($M 2000 PPP)

Indicator Headline

The indicator compares HERD expenditure
allocated to Engineering and Technology with
the numbers of PhDs awarded in this
category.

The UK has around one quarter more PhDs
per unit spend than the OST comparator
group average (for the 5-year period). This is
a similar level of productivity per unit spend to
Finland and Sweden and above Germany
(who are slightly below the OST comparator
group average in this measure). The USA is
similar to Germany, whilst Japan is
substantially below the OST comparator group
average.

The UK is close to the EU average, but its
productivity has dropped over the period.

In Engineering and Technology, the UK
awards about 2,000 PhDs per year (there has
been a drop of about 10% between 2001 and
2002). It is the 4th largest (to USA - 5,400,
Japan - 3,100 and Germany 2,400) by volume.
There are no available data for France, but its
research in this area has historically had good
impact.

UK PhD data are subject to a change in HESA
data collection from 2001 (see Background).

There is a correlation across countries
between funding and PhD output, implying
that probably both scale with an underlying
size factor such as staff numbers, which affect
training and research volumes although other
factors may also play a role.

Engineering and Technology - PhDs per HERD, G8 nations (no data for Canada, France, Italy or

Russia)

0.0

0.5

1.0

1.5

2.0

2.5

1998

1999

2000

2001

2002

Year

award

r H

2000 PPP

UK
USA
GERMANY
JAPAN
EUROPEAN UNION

Data: OECD (RDS), OECD Education Database, UK SET statistics and HESA data

Description of performance indicator

4.02 PhDs awarded relative to HERD in five main research fields

Condition signalling improvement

Increased ratio

Social Sciences

Average

1998 - 2001

Actual

2002

Ratio or

difference

UK PhDs awarded per HERD ($M 2000 PPP) - Actual

3.01

2.78

0.92

3.69

3.31

UK PhDs awarded per unit HERD ($M 2000 PPP) - Rank

OST comparator group average - PhDs awarded per unit HERD
($M 2000 PPP)

Indicator Headline

The indicator compares HERD expenditure
allocated to Social Sciences with the numbers
of PhDs awarded in this category. Social
Sciences includes business and management
fields.

The UK produces rather fewer PhDs per unit
spend than the OST comparator group
average. The range is much wider in this field
than the science-technology-based categories
and other analyses have thrown doubt on data
consistency for some countries.

The UK's average level of productivity per unit
spend is roughly half that of Germany, and
one third that of the USA. In this field, the UK
awards more than 2,000 PhDs per year and is
the 3rd largest (to the USA - 17,000 and
Germany - 4,000) by volume.

UK PhD data are subject to a change in HESA
data collection from 2001 (see Background).

There are no data available for Japan HERD
in Social Sciences.

Social Sciences - PhDs per HERD, G8 nations (no data for Canada, France, Italy or Russia)

0.0

2.0

4.0

6.0

8.0

10.0

12.0

1998

1999

2000

2001

2002

Year

award

r H

2000 PPP

UK
USA
GERMANY
EUROPEAN UNION

UK - HESA data collection
procedure changed in 2001
(see headline note)

Data: OECD (RDS), OECD Education Database, UK SET statistics and HESA data

Description of performance indicator

4.02 PhDs awarded relative to HERD in five main research fields

Condition signalling improvement

Increased ratio

Humanities

Average

1998 - 2001

Actual

2002

Ratio or

difference

UK PhDs awarded per HERD ($M 2000 PPP) - Actual

3.96

3.72

0.94

2.49

2.35

UK PhDs awarded per unit HERD ($M 2000 PPP) - Rank

OST comparator group average - PhDs awarded per unit HERD
($M 2000 PPP)

Indicator Headline

The indicator compares HERD expenditure
allocated to Humanities with the numbers of
PhDs awarded in this category. This is the
first time that Humanities data have been
analysed in this indicator. The data coverage
is sparse, reducing the feasibility of
interpretation

The UK produces a much larger number of
PhDs per unit spend than the OST comparator
group. This is a similar average level of
productivity per unit spend to Finland and
approximately twice that of Germany, which
itself is below the OST comparator group
average.

In this field, the UK awards close to 2,000
PhDs per year and is the 3rd largest (after the
USA - 6,000 and just behind Germany also
with 2,000) by volume. The UK recently
started producing more humanities PhDs than
France.

UK PhD data are subject to a change in HESA
data collection from 2001 (see Background).

There are no data available for the USA
HERD in Humanities.

Humanities - PhDs per HERD, G8 nations (no data for USA, Canada, France, Italy or Russia)

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

1998

1999

2000

2001

2002

Year

award

r H

2000 PPP

UK
GERMANY
EUROPEAN UNION

UK - HESA data collection
procedure changed in 2001
(see headline note)

Data: OECD (RDS), OECD Education Database, UK SET statistics and HESA data

Description of performance indicator

4.03 PhDs awarded relative to PUBERD

Condition signalling improvement

Increased ratio

Average

1998 - 2001

Actual

2002

Ratio or

difference

UK PhDs awarded per PUBERD (2000 $M PPP) - Actual

1.36

1.54

1.14

OST comparator group average - PhDs awarded per PUBERD
(2000 $M PPP)

1.17

1.15

0.99

UK PhDs awarded per PUBERD (2000 $M PPP) - Rank

UK PhDs awarded per PUBERD (2000 $M PPP) / OST comparator group
average - PhDs awarded per PUBERD (2000 $M PPP)

1.16

1.34

PhDs per PUBERD (2000 $M PPP), G8 nations (no data for Russia)

0.00

0.20

0.40

0.60

0.80

1.00

1.20

1.40

1.60

1.80

1998

1999

2000

2001

2002

Year

PhD

s awarded per PU

(

2000 $M PPP)

UK
USA
CANADA
FRANCE
GERMANY
ITALY
JAPAN
EUROPEAN UNION
OST GRP AVGE

Data: OECD (MSTI), OECD Education Database

UK - HESA data collection
procedure changed in 2001
(see headline note)

Indicator Headline

The indicator compares the output of people
gaining research degrees with the total public
research spend (PUBERD, the sum of
GOVERD and HERD).

The UK is ranked 1st within the G8 nations in
terms of PhD awards per PUBERD, and 4th
amongst the OST comparator group (behind
Poland, Switzerland and Spain). The UK
remains significantly above the OST
comparator group average throughout the
period 1998-2001.

Over the 5-year period only one other G8
nation has improved the PhD/PUBERD ratio -
Japan by 30%. The ratios for the USA and
France have continued to decline. A UK
decline in 2000 is obscured by the change in
data collection in 2001.

Outside the G8, the PhD/PUBERD ratio of
South Korea has levelled off after an increase
in the previous 4 years of more than 13%.
The EU15 average is fairly level.

UK PhD data are subject to a change in HESA
data collection from 2001 (see Background).

Description of performance indicator

4.04 Citations relative to GDP

Condition signalling improvement

Improved ratio of citations per GDP compared to recent past

Average

1998 - 2002

Actual

2003

Ratio or

difference

UK citations per GDP (2000 $M PPP) - Actual

0.49

0.12

0.24

OST comparator group average citations per GDP (2000 $M PPP)

0.35

0.09

0.26

UK citations per GDP - Rank

UK cites per GDP / OST comparator group average cites per GDP

1.38

1.27

0.92

Indicator Headline

This is the first of a series of indicators that
measure citation accumulation (hence research
esteem) in relation to the components of
investment.

This indicator measures the association
between total GDP and national citation count.

The UK is the leading country among the G8
group. The UK is well ahead of the USA which
is ranked 11th in the OST group overall (the UK
holds the 7th position). Several countries,
including the UK, Canada and the USA have
declined in performance relative to the OST
average. Germany's performance has recently
levelled, having improved slightly since 1995
throughout the time period.

UK relative international performance remains
well ahead of the average for the OST
comparator group, although it declined slightly
in 2001 and again in 2003 compared to the
OST group average.

The UK is ranked 7th overall for recognition
compared to general expenditure, behind
Israel, Switzerland, the Netherlands and the
Scandinavian countries.

The EU15 combined overall performance is
increasingly similar to that of the USA,
especially since the late 1990s.

Data for Russia are anomalous, because of the
rapid fall in the OECD figures for Russian GDP.

cont./

Citations relative to GDP rebased to OST comparator group average - G8 nations

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

tati

on /

GDP

rebased to OS

T com

arator group average

(=1.

USA

CANADA

FRANCE

GERMANY

ITALY

JAPAN

RUSSIA

EUROPEAN
UNION

Data: Thomson ISI® National Science Indicators, OECD (MSTI)

Description of performance indicator

4.05 Number and share of citations relative to GERD

Condition signalling improvement

Increased citations per unit expenditure

Average

1998 - 2002

Actual

2003

Ratio or

difference

UK GERD (2000 $M PPP)

27792

30203

1.09

UK citations / GERD (2000 $M PPP) - Actual

26.24

6.31

0.24

OST comparator group average citations / GERD (2000 $M PPP)

16.04

4.10

0.26

UK citations / GERD (2000 $M PPP) - Rank

2.09

2.08

0.99

UK share of OST comparator group citations/spend - Rank

UK share of OST comparator group citations / UK share of OST
comparator group GERD (2000 $M PPP)

OST comparator group citation share / OST comparator group GERD share for G8 nations

0.5

1.5

2.5

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

Rati

o ci

tati

on share /

RD share

UK
USA
CANADA
FRANCE
GERMANY
ITALY
JAPAN
RUSSIA
EUROPEAN UNION

Data: Thomson ISI® National Science Indicators, OECD (MSTI)

Indicator Headline

This indicator compares national citation
counts to Gross Expenditure on R&D (GERD)
including both public and private sector spend.
The private sector generally publishes far less
than the public sector and its expenditure is
therefore less likely to generate citations. As a
consequence, national economies that have a
balance towards the private sector will
generate fewer citations per unit GERD.

Within the OST comparator group the UK has
a strong performance in terms of relative share
of citations compared to relative expenditure.

The UK continues to lead among the G8
nations although its position is perhaps not as
strong as it has been previously.

UK GERD has increased in real terms over the
last ten years and again was greater in 2003
than the average for the recent past. The OST
comparator group average rate of increase is
dominated by USA figures: USA GERD
increased by over one third in the last decade.

Citations accumulate with time and are always
fewer in more recent years. The UK has
dipped very slightly in its relative international
performance on this indicator (an average
change in citations per GERD - ratio between
2003 and recent average is UK = 0.240, OST
comparator group average = 0.255).

Switzerland still leads in this indicator (32.2
cites/$M GERD average in recent 5 years) but
in the last 3 years Poland has risen through the
ranks from around 12th to 2nd or 3rd. The UK
is ranked 5th behind Switzerland, Poland, the
Netherlands and Denmark.

cont./

Description of performance indicator

4.06 Citations relative to PUBERD (GOVERD + HERD)

Condition signalling improvement

Increased citations per unit expenditure

Average

1998 - 2002

Actual

2003

Ratio or

difference

UK GOVERD + HERD (2000 $M PPP)

8930

9379

1.05

UK citations / GOVERD + HERD (2000 $M PPP) - Actual

81.08

20.31

0.25

48.05

12.12

0.25

UK citations per GOVERD + HERD (2000 $M PPP) - Rank

OST comparator group average citations / GOVERD + HERD (2000 $M
PPP)

Indicator Headline

It is useful to know about national research
performance in relation to levels of public
spending on the research base. While GERD
indicates total national expenditure on R&D,
and it is reasonable to anticipate some
interaction and gearing between public and
private sectors, private R&D investment is not
generally aimed at producing papers or people.

Indicator 1.02 shows that UK expenditure on
R&D performed in the public sector has risen in
real terms compared to the recent past. This
increase is less, however, over the last ten
years than for the OST group average.

The UK has maintained its strong relative
international performance in the G8 and the
wider OST comparator group.

Most countries have experienced a reduction in
the absolute number of citations/$M PUBERD
but the relative positions are unchanged.
Thus, the UK (81.1 citations/$M PUBERD on
average for recent years) is ranked 2nd
recently and 3rd in 2003 to Switzerland (129.3
cites/$M) and just behind Denmark. The USA
(55.2 cites/$M) is ranked 9th.

Of the G8, Germany (41.5 cites/$M) has
maintained its upward trend and Japan, though
not increasing as in previous years has
remained stable whilst other countries are
experiencing declines in this measure.

cont./

Citations per GOVERD + HERD rebased to OST comparator group average (= 1.0) - G8

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

2.0

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

tati

s p

er GOV

RD + HE

USA

CANADA

FRANCE

GERMANY

ITALY

JAPAN

RUSSIA

EUROPEAN
UNION

Data: Thomson ISI® National Science Indicators, OECD (MSTI)

Description of performance indicator

4.07 Citations relative to HERD

Condition signalling improvement

Increased citations per unit expenditure

Average

1998 - 2002

Actual

2003

Ratio or

difference

UK HERD (2000 $M PPP)

5786

6462

1.12

UK citations per HERD (2000 $M PPP) - Actual

128.85

29.48

0.23

OST comparator group average citations per HERD (2000 $M PPP)

80.07

18.87

0.24

UK citations per HERD (2000 $M PPP) - Rank

-2

Citations per unit HERD rebased to OST comparator group average (= 1.0) for G8 nations

0.0

0.5

1.0

1.5

2.0

2.5

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

tati

ons per HE

RD - (OS

com

arator group average =

UK
USA
CANADA
FRANCE
GERMANY
ITALY
JAPAN
RUSSIA
EUROPEAN UNION

Data: Thomson ISI® National Science Indicators, OECD (MSTI)

Indicator Headline

HERD is Expenditure on R&D performed in the
Higher Education sector. Here, citation counts
are compared with HERD.

UK HERD has increased markedly, over one
third in real terms over the last ten years and
by nearly a quarter between 1999 and 2002
alone. OST comparator group average has
increased by a similar amount over ten years.

Citations accumulate over time and are always
fewer in more recent years. UK actual citations
per HERD naturally decrease since expenditure
is greater and older papers have had more time
to accumulate citations. By comparison to the
average for the OST comparator group, the UK
has about 1.6 times as many citations per unit
expenditure.

The UK's position in 2003 is 2nd (29.5
cites/$M), it has apparently lost some ground to
Switzerland (36.0 cites/$M) and is now only
slightly ahead of Denmark (27.7 cites/$M). The
UK is ranked 4th with regard to average recent
performance in this indicator behind
Switzerland, Denmark and Russia. However,
the UK is still a clear leader amongst the G8
nations.

The EU15 has continued to improve its overall
performance relative to the USA. The ratio of
USA cites/$M HERD in 1995 to that of the
EU15 in the same year was 1.41 and for the
most recent year, 2003, this ratio had dropped
to 1.08. Average HERD for the most recent
five year period for the USA and the EU15 is
close to $40BN.

cont./

Description of performance indicator

4.08 Citations relative to HERD in five main research fields

Condition signalling improvement

Increase in cites per unit spend at the OECD field level

Medical Sciences

Average

1998 - 2002

Actual

2003

Ratio or

difference

UK HERD (2000 $M PPP)

1607

1821

1.13

UK citations per HERD (2000 $M PPP) - Actual

357.79

79.77

0.22

OST comparator group average citations per HERD (2000 $M PPP)

306.98

69.95

0.23

UK citations per HERD (2000 $M PPP) - Rank

-1

UK cites per HERD / OST comparator group average

1.17

1.14

Indicator Headline

For some countries, both HERD and citation
counts can be disaggregated by OECD field for
analysis in this indicator.

For the Medical Sciences, the UK acquires
more cites per unit HERD than the OST
comparator group average and is ranked 3rd
behind Denmark (on its own at over twice OST
average, but with a somewhat anomalous blip
in 2002) and Finland in recent years. Ranked
1st among G8 nations, UK performance is
broadly constant over the period. From 2001
Spain replaces Finland in second place, and in
2003, the Netherlands edges the UK into 4th
position.

The USA is ranked 7th, at 0.85 OST
comparator group average.

The data for Denmark show a general level of
performance that is exceptional and where the
profile departs substantially from that of other
countries. This is due to relatively low amounts
of HERD expenditure until a marked rise in
2002 which accounts for the drop in 2002.

The EU15 line should be regarded with caution.
The OECD data do not accurately reflect a true
sum and the index is therefore over-inflated.

HERD is Expenditure on R&D performed in the
Higher Education sector. Here, citation counts
are compared with HERD. It is not implied that
HE is the only agent acquiring citations, but in
many research economies it is the major
player.

Medical Sciences - Citations per unit HERD rebased to OST comparator group average (= 1.0) for

OECD nations

0.0

0.5

1.0

1.5

2.0

2.5

3.0

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

ions per H

(

ST com

arat

or group

average = 1.

UK
USA
GERMANY
JAPAN
AUSTRALIA
DENMARK
FINLAND
NETHERLANDS
POLAND
SPAIN
SWEDEN
EUROPEAN UNION

Data: Thomson ISI® National Science Indicators, OECD (RDS), UK SET statistics and HESA data

Description of performance indicator

4.08 Citations relative to HERD in five main research fields

Condition signalling improvement

Increase in cites per unit spend at the OECD field level

Natural Sciences

Average

1998 - 2002

Actual

2002

Ratio or

difference

UK HERD (2000 $M PPP)

1918

2080

1.08

UK citations per HERD (2000 $M PPP) - Actual

254.94

61.66

0.24

OST comparator group average citations per HERD (2000 $M PPP)

190.60

49.53

0.26

UK citations per HERD (2000 $M PPP) - Rank

UK cites per HERD / OST comparator group average

1.34

1.24

Natural Sciences - Citations per unit HERD rebased to OST comparator group average (= 1.0) for

OECD nations

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

ions per H

(

ST com

arat

or group

average = 1.

UK
USA
GERMANY
JAPAN
AUSTRALIA
DENMARK
FINLAND
NETHERLANDS
POLAND
SPAIN
SWEDEN

Indicator Headline

For some countries, both HERD and citation
counts can be disaggregated by OECD field for
analysis in this indicator.

For the Natural Sciences, the UK acquires
more cites per unit HERD than the OST
comparator group average and is ranked 1st
among G8 nations, and 2nd behind the
Netherlands overall. Recent UK performance
is steady after a period of decline.

The USA is placed bottom of the OST nations
in this indicator, and its performance continues
to decline.

Data: Thomson ISI® National Science Indicators, OECD (RDS), UK SET statistics and HESA data

100

Description of performance indicator

4.08 Citations relative to HERD in five main research fields

Condition signalling improvement

Increase in cites per unit spend at the OECD field level

Engineering and Technology

Average

1998 - 2002

Actual

2003

Ratio or

difference

UK HERD (2000 $M PPP)

1045

1202

1.15

UK citations per HERD (2000 $M PPP) - Actual

58.17

12.25

0.21

OST comparator group average citations per HERD (2000 $M PPP)

45.35

11.38

0.25

UK citations per HERD (2000 $M PPP) - Rank

UK cites per HERD / OST comparator group average

1.28

1.08

Indicator Headline

For some countries, both HERD and citation
counts can be disaggregated by OECD field for
analysis in this indicator.

For Engineering and Technology, the UK
acquires more cites per unit HERD than the
OST comparator group average and is ranked
1st among G8 nations, and 2nd behind
Denmark overall. Its recent performance
shows a steady decline, however.

The USA is ranked 3rd, with Germany in 4th
place, showing improvement over the period.

Engineering and Technology - Citations per unit HERD rebased to OST comparator group average (=

1.0) for OECD nations

0.4

0.9

1.4

1.9

2.4

2.9

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

ions per H

(

ST com

arat

or group

average = 1.

UK
USA
GERMANY
JAPAN
AUSTRALIA
DENMARK
FINLAND
NETHERLANDS
POLAND
SPAIN
SWEDEN

Data: Thomson ISI® National Science Indicators, OECD (RDS), UK SET statistics and HESA data

102

Description of performance indicator

4.08 Citations relative to HERD in five main research fields

Condition signalling improvement

Increase in cites per unit spend at the OECD field level

Social Sciences

Average

1998 - 2002

Actual

2003

Ratio or

difference

UK HERD (2000 $M PPP)

738

860

1.16

UK citations per HERD (2000 $M PPP) - Actual

119.93

22.47

0.19

OST comparator group average citations per HERD (2000 $M PPP)

63.05

11.89

0.19

UK citations per HERD (2000 $M PPP) - Rank

UK cites per HERD / OST comparator group average

1.90

1.89

Indicator Headline

For some countries, both HERD and citation
counts can be disaggregated by OECD field for
analysis in this indicator.

Bibliometric data for the Social Sciences
cannot be compared directly with other
disciplines. Recent work has shown that
European comparisons must be made with
caution.

For the Social Sciences, the UK acquires more
cites per unit HERD than the OST comparator
group average and is ranked 2nd behind the
USA overall. Its recent profile shows a steady
performance, while the USA is declining.

Germany is ranked 3rd, showing improvement
over the period.

The EU15 line should be regarded with caution.
The OECD data do not accurately reflect a true
sum and the index is therefore over-inflated.

Social Sciences - Citations per unit HERD rebased to OST comparator group average (= 1.0) for OECD

nations

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

ions per H

(

ST com

arat

or group

average = 1.

UK
USA
GERMANY
AUSTRALIA
DENMARK
FINLAND
POLAND
SPAIN
SWEDEN
EUROPEAN UNION

Data: Thomson ISI® National Science Indicators, OECD (RDS), UK SET statistics and HESA data

104

Description of performance indicator

4.08 Citations relative to HERD in five main research fields

Condition signalling improvement

Increase in cites per unit spend at the OECD field level

Humanities

Average

1998 - 2002

Actual

2003

Ratio or

difference

UK HERD (2000 $M PPP)

405

498

1.23

UK citations per HERD (2000 $M PPP) - Actual

5.24

1.14

0.22

OST comparator group average citations per HERD (2000 $M PPP)

1.09

0.27

0.25

UK citations per HERD (2000 $M PPP) - Rank

UK cites per HERD / OST comparator group average

4.82

4.18

Indicator Headline

This is the first time that data on the
Humanities have been analysed for this
indicator.

For some countries, both HERD and citation
counts can be disaggregated by OECD field for
analysis in this indicator.

For the Humanities, the UK acquires roughly
three times more cites per unit HERD than the
OST comparator group average and is ranked
1st. Its recent performance shows a modest
decline, but no other nation's performance is
close.

As is widely understood, citations in this field
are extremely low compared with other fields.

Humanities - Citations per unit HERD rebased to OST comparator group average (= 1.0) for OECD

nations

0.0

1.0

2.0

3.0

4.0

5.0

6.0

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

ions per H

(

ST com

arat

or group

average = 1.

UK
GERMANY
AUSTRALIA
EUROPEAN UNION

Data: Thomson ISI® National Science Indicators, OECD (RDS), UK SET statistics and HESA data

106

Description of performance indicator

5.01 PhDs awarded per researcher

Condition signalling improvement

Increased ratio

Average

1999 - 2001

Actual

2002

Ratio or

difference

UK PhDs awarded per researcher - Actual

0.07

0.08

1.11

OST comparator group average - PhDs awarded per researcher

0.06

0.99

UK PhDs awarded per researcher - Rank

UK PhDs awarded per researcher / OST comparator group average PhDs
awarded per researcher

1.21

1.36

1.12

PhDs per FTE researcher, G8 nations (no data for Russia)

0.00

0.02

0.04

0.06

0.08

0.10

0.12

1998

1999

2000

2001

2002

Year

Ds awarded per FTE

researcher

UK
USA
CANADA
FRANCE
GERMANY
ITALY
JAPAN
EUROPEAN UNION
OST AVERAGE

Data: OECD (MSTI), OECD Education Database

Indicator Headline

The labour productivity of the research base is
measured by the volume of outputs compared
to the numbers of highly skilled people
contributing to research.

This indicator is also one of sustainability. It
measures the relationship between the volume
of 'researchers' (see definition in Glossary)
already in the research base and the output of
highly skilled people gaining research
degrees.

The UK is ranked 2nd behind Germany within
the G8 in terms of PhD awards per
researcher, and 4th within the OST
comparator group (behind Switzerland and
Spain). The UK has remained well ahead of
the OST comparator group average
throughout the period.

The number of PhD awards per researcher in
the UK has declined in the most recent year.
Over the 4-year period it has shown apparent
growth of 15%, but this is largely due to data
adjustment. UK data for 2001 are affected by
a change in HESA data collection.

Both Japan and Belgium have seen increases
in this indicator of nearly 30% and more than
40% respectively. Japan trails the G8,
however, by a significant margin.

Both the EU average and the USA average
are falling over the period. This may have
implications for planning as it will affect the
availability of highly skilled people.

There are no data for China.

108

Description of performance indicator

5.02 Publications per researcher

Condition signalling improvement

Increased relative output

Average

1998 - 2002

Actual

2003

Ratio or

difference

UK papers per researcher - Actual

2.17

2.08

0.96

1.38

1.35

0.97

UK papers per researcher - Rank

-1

0.62

0.63

1.02

OST comparator group average papers per
researcher

OST comparator group average compared to
UK (UK=1)

Ratio of publications to FTE researchers for G8 nations

0.5

1.5

2.5

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

ublicat

ions per

esear

cher

USA

CANADA

FRANCE

GERMANY

ITALY

JAPAN

RUSSIA

EUROPEAN UNION

OST GRP AVGE

Data: Thomson ISI® National Science Indicators, OECD (MSTI)

Indicator Headline

Research culture is reflected in the numbers of
publications in international journals that are
produced by each researcher active within the
research base.

The UK's recent (2.17 papers/FTE) and current
(2.08 papers/FTE) performance is 3rd to
Switzerland (2.69, 2.61) and the Netherlands
(2.32, 2.22) as an average for the recent
period (1998-2002) and for 2003. [Italy - see
below].

The UK's ratio of papers per researcher has
fallen between the recent period and 2003.
For the OST comparator group, the average
number of papers per researcher also fell
slightly during this period.

UK output per researcher remains much higher
than the OST comparator group average. In
absolute terms the UK relative output has
improved to more than twice that of the USA.

The UK ranks first in the G8 for the average
recent period whereas several of the G8 are
below the OST comparator group average.
The performance of the EU15 has been fairly
stable over the 10-year period at around 1.4 to
1.5 papers per researcher whilst the USA
output has declined from 1.2 down to 0.95
papers/FTE.

China's productivity has doubled over the ten
years, but is only one-tenth that of the UK.

Italy's indexed papers/FTE researcher has
increased. This may be an artefact, however,
because of a diminishing researcher base.
That decline has stabilised, so a continuing
upward trend would indicate genuine
improvement.

110

Description of performance indicator

5.03 Citations per researcher

Condition signalling improvement

Increase in citation ratio

Average

1998 - 2002

Actual

2003

Ratio or

difference

UK citations per researcher - Actual

10.29

11.09

1.08

OST comparator group average citations per researcher

5.96

6.45

1.08

UK citations per researcher - Rank

-1

OST comparator group average compared to UK (UK=1)

0.56

1.00

Numbers of citations per researcher for G8 nations

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

tati

ons per FTE

researcher

UK
USA
CANADA
FRANCE
GERMANY
ITALY
JAPAN
RUSSIA
EUROPEAN UNION

Data: Thomson ISI® National Science Indicators, OECD (MSTI)

Indicator Headline

As with indicator 5.02, this is a simple and
readily understood index calculated at the level
of the research base as a whole.

Gross output per capita in international
journals is a good measure of volume of
research activity. It is also important to
consider the effectiveness of that output by
measuring the citations they generate: in other
words, we need a measure of quality as well
as quantity.

For the UK, there has been a clear and
substantial trend of improvement in citations
per researcher throughout the 10-year time
period, indicating a relative improvement in
research quality as measured by peer esteem.

The UK's relative international position has
improved in line with that of Switzerland and
the Netherlands. Along with Denmark, these
nations have an average of more than 10
citations per researcher per year over the 10-
year period.

China's rapidly increasing output (see 2.03) is
generated by a huge researcher population.
Its impact still lags (see 3.01) and its
effectiveness by this indicator is less than 0.5
citations per FTE researcher at present.

Italy has nominally overtaken the UK this year,
but reservations about the apparent Italian
performance (see Indicator 5.02) apply here
also.

cont./

112

Description of performance indicator

6.01 Researchers per thousand population

Condition signalling improvement

Increased ratio

Average

1998 - 2002

Actual

2003

Ratio or

difference

UK researchers as a percentage of total population - Actual

0.28

0.30

1.07

0.31

0.34

1.11

UK researchers as a percentage of total population - Rank

0.91

0.88

OST comparator group average - researchers as a percentage of total
population

UK researchers as a percentage of total population / OST comparator
group average researchers as a percentage of total population

Researchers compared with total population, G8 nations

0.1

0.2

0.3

0.4

0.5

0.6

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

Researchers as a percentage of total

popul

ati

UK
USA
CANADA
FRANCE
GERMANY
ITALY
JAPAN
RUSSIA

Data: OECD (MSTI)

Indicator Headline

The numbers of researchers in the population
as a whole provides a measure of the resource
capacity of each country in regard to research
innovation, and an indication of the
technological orientation of the country.

Researchers (see Glossary definition)
constitute less than 0.5% of the population in
those countries in the OST comparator group
for which data are available (22 countries).

For the UK, researchers as a percentage of
the total population has again risen slightly in
2003 compared to the average for the recent
past. This is part of a gradual longer term
trend and is similar to but slightly lower than
the average rise for the OST comparator group
as a whole.

The UK proportion of researchers in the
population is much lower than for most of the
G8 nations and below that of the OST
comparator group average. The UK rank
amongst the OST comparator group in this
indicator has remained steady at around 14.

Many research economies elsewhere in
Europe have increased their relative
researcher population by a greater percentage
than the UK and other G8 nations.

China has a much smaller relative researcher
population but this has risen from 0.04% in
1995 to 0.07% in 2003. The total is about half
that of the USA.

114

Description of performance indicator

6.02 Researchers per thousand workforce

Condition signalling improvement

Increased ratio

Average

1998 - 2002

Actual

2003

Ratio or

difference

UK researchers per 1000 workforce - Actual

5.61

5.92

1.05

OST comparator group average - researchers per 1000 workforce

6.21

6.85

1.10

UK researchers per 1000 workforce - Rank

-2

0.90

0.86

UK researchers per 1000 workforce / OST comparator group average
researchers per 1000 workforce

Researchers per 1000 workforce, G8 nations

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

Researchers per 1000 workforce

UK
USA
CANADA
FRANCE
GERMANY
ITALY
JAPAN
RUSSIA
EUROPEAN UNION

Data: OECD (MSTI)

Indicator Headline

The numbers of researchers in the workforce
provides a measure of the resource capacity
of each country in regard to research
innovation.

The UK continues to perform poorly compared
to the other G8 nations in this indicator. The
USA proportion of researchers in the
workforce continues to rise (9.1 per 1000
workforce, recent average) and is well ahead
of the UK (5.6 per 1000 workforce). Finland
remains the leader for this indicator with 13.4
researchers per 1000 workforce. Japan and
Sweden are both around the same recent
average proportion as the USA (9.7 per 1000
workforce in both countries).

The UK's volume of researchers as a
proportion of the workforce has increased from
5.1 per 1000 workforce in 1995 to 5.9 per 1000
workforce in 2003. The ratio between the
latest year and the recent past is in line with
this trend.

The UK has been consistently ranked around
15th over the longer period. Since 2001,
however, both South Korea and Taiwan have
moved ahead of the UK. Recent change for
the UK is similar to but less than the OST
comparator group average. The UK remains
more than 10% below average for the OST
comparator group of 22 countries for which
data were available.

Researchers (see Glossary definition) make
up a specific subcategory of R&D personnel,
and provide a particular measure of research
capacity. Both these terms are somewhat
broadly defined and there may be
inconsistency between national figures.

116

Description of performance indicator

6.03 R&D personnel per hundred population

Condition signalling improvement

Increased ratio

Average

1998 - 2002

Actual

2003

Ratio or

difference

UK R&D personnel as a percentage of total population

0.46

0.45

0.98

0.50

0.53

1.07

UK R&D personnel as a percentage of total population - Rank

-1

0.93

0.85

OST comparator group average - R&D personnel as a percentage of total
population

UK R&D personnel as a percentage of total population / OST comparator
group average R&D personnel as a percentage of total population

R&D personnel as a percentage of total population G8 nations (no data for USA)

0.20

0.30

0.40

0.50

0.60

0.70

0.80

0.90

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

Personnel as a percent

age of

l populat

ion

UK
CANADA
FRANCE
GERMANY
ITALY
JAPAN
RUSSIA
EUROPEAN UNION

Data: OECD (MSTI)

Indicator Headline

The numbers of R&D personnel in the
population as a whole provides a measure of
the highly-qualified human knowledge base of
each country in regard to research and
development, and an indication of the
technological orientation of the country.

The UK proportion of R&D personnel has
remained at around 0.45%. Amongst the G8
nations (the USA is not included in this
indicator), the UK has a lower proportion of
R&D personnel in the population than every
member, except Italy. Within the OST
comparator group as a whole the UK is ranked
15th, a decline of 3 positions since 1995. The
UK has also dropped below EU15 average.

R&D personnel account for just over 0.5% of
the workforce, on average, in the 21 OST
comparator group countries for which data are
available (there is no data available for the
USA).

The OST comparator group average has
improved steadily from 0.45% in 1995 to
0.53% in 2003.

R&D personnel (see Glossary definition)
provide a general measure of research
capacity, of which researchers (see Glossary
definition) are a particular set. Both these
terms are somewhat broadly defined and there
may be inconsistency between national
figures.

118

Description of performance indicator

6.04 R&D personnel per hundred workforce

Condition signalling improvement

Increased ratio

Average

1998 - 2002

Actual

2003

Ratio or

difference

UK R&D personnel as a percentage of workforce - Actual

0.92

0.88

0.96

1.01

1.07

1.06

UK R&D personnel as a percentage of workforce - Rank

-1

0.91

0.83

UK R&D personnel as a percentage of workforce / OST comparator group
average R&D personnel as a percentage of workforce

OST comparator group average - R&D personnel as a percentage of
workforce

R&D personnel as a percentage of workforce, G8 nations (no data for USA)

0.4

0.6

0.8

1.0

1.2

1.4

1.6

1.8

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

Personnel as a percen

ge of

workf

rce

UK
CANADA
FRANCE
GERMANY
ITALY
JAPAN
RUSSIA
EUROPEAN UNION

Data: OECD (MSTI)

Indicator Headline

The numbers of R&D personnel in the
workforce provides a measure of the highly-
qualified human knowledge base of each
country in regard to research development.

The UK has less than 1% of its workforce
classified as R&D personnel in OECD data
and its relative concentration has declined
while the average has improved. The UK
percentage has fluctuated around 0.9% in the
last 10 years but has not changed significantly.
The pattern has generally been slightly better
for other G8 nations, except Russia.

The UK is below the OST comparator group
average, which increased steadily over the last
decade. The OST comparator group average
(over 1% of workforce) is for a group of 21
countries, excluding the USA, for which data
are available. Within this group the UK has
been ranked 14th in most years. This index
has changed little for the UK whilst the OST
comparator group average and the EU15
average have increased, reducing the UK
'share' of the total. Meanwhile smaller
countries including Singapore, the
Scandinavian countries and Belgium have
moved ahead of the average.

It might be expected that R&D personnel
should increase as a proportion of workforce in
a knowledge-based economy, but issues of
classification may affect data reporting.

R&D personnel (see Glossary definition)
provide a general measure of research
capacity, of which researchers (see Glossary
definition) are a specific part. Both these
terms are somewhat broadly defined and there
may be inconsistency between national
figures.

120

Description of performance indicator

6.05 Researchers as a proportion of R&D Personnel

Condition signalling improvement

Increased ratio

Average

1998 - 2002

Actual

2003

Ratio or

difference

UK researchers as a proportion of R&D Personnel - Actual

0.61

0.67

1.10

0.58

0.61

1.04

UK researchers as a proportion of R&D Personnel - Rank

1.05

1.10

OST comparator group average - researchers as a proportion of R&D
Personnel

UK researchers as a proportion of R&D Personnel / OST comparator
group average researchers as a proportion of R&D Personnel

Researchers compared with R&D personnel, G8 nations (no data for USA)

0.3

0.4

0.5

0.6

0.7

0.8

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

Researchers as a proporti

on of R&D P

ersonnel

UK
CANADA
FRANCE
GERMANY
ITALY
JAPAN
RUSSIA
EUROPEAN UNION

Data: OECD (MSTI)

Indicator Headline

Although numbers of both researchers and
R&D personnel have appeared somewhat
static for the UK as a proportion of workforce,
the data in this indicator suggest that a shift
has taken place. This shift was in line with
other G8 nations but may have accelerated.

What these data show is that researchers are
increasing as a proportion of R&D personnel.
There is no overall decrease in R&D
personnel; for the UK these numbers
increased slightly although they have now
plateaued. This outcome therefore seems to
reflect the growing professionalisation of the
research workforce in a knowledge-based
economy, as traditional technical roles become
more complex and sophisticated.

The UK researchers/R&D personnel ratio has
improved in the last few years both in absolute
and relative terms. There was an increase last
year and an increase of 10% between the
recent and current year figures. In addition,
the UK has improved its ranking from 12th to
8th in the most recent year.

The researcher proportion is highest for
Singapore and South Korea, both countries
with more than three quarters of their
researchers as a proportion of R&D personnel.
China also performs strongly in this regard and
ranks 3rd. Italy, the Netherlands and
Switzerland have the lowest researcher
proportions, all with less than half. There are
no USA data.

Researchers (see Glossary definition) make up
a specific subcategory of R&D personnel, and
provide a particular measure of research
capacity. Both these terms are somewhat
broadly defined and there may be
inconsistency between national figures.

122

Description of performance indicator

7.01 Business R&D investment in publicly performed R&D (BE-PUBERD as a proportion of PUBERD

Condition signalling improvement

Increased ratio

Average

1998 - 2002

Actual 2003

Ratio or

difference

UK - percentage of BE-PUBERD within total PUBERD - Actual

9.16

6.61

0.72

6.82

6.78

0.99

UK - percentage of BE-PUBERD within total PUBERD - Rank

-3

1.34

0.98

OST comparator group average percentage of BE-PUBERD within total
PUBERD

UK percentage of BE-PUBERD within total PUBERD / OST comparator
group average percentage of BE-PUBERD within total PUBERD

BE-PUBERD as a percentage of PUBERD - G8 nations (no data for Russia)

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

-P

UBE

RD as p

ercen

tag

e o

f P

USA

CANADA

FRANCE

GERMANY

ITALY

JAPAN

EUROPEAN
UNION

Data: OECD (RDS & MSTI)

Indicator Headline

BE-PUBERD is R&D performed in the publicly
funded sector (Government plus Higher
Education) that is funded by the Business
Enterprise sector. Business expenditure via
investment in other sectors may reflect
confidence in the research and relevance to
business objectives.

The UK had ranked ahead of other G8 nations
but has dropped behind Germany and Canada.
It is now close to OST comparator group and
EU15 average, but has fallen to 8th among
OST comparator group nations in terms of
business investment in publicly funded R&D.
This puts it behind Poland (which bounced
back from a 'data' drop last year), the
Netherlands, Belgium and Finland.

The biggest increases on the recent 5 year
average are Italy (28%), Switzerland and
Denmark (17%) and the Netherlands, Finland
and Germany (9%) compared with a stable
OST comparator group average.

The long term trend for the EU15 seems to be
an increase on this indicator. By contrast, the
USA lags on Europe and is declining.

China ranks 19th in the OST comparator
group.

Some 2003 data are produced by forecasting.
Forecasted data are 'sense checked' by
comparing forecasts in previous reports with
outcomes.

124

Description of performance indicator

7.02 Business R&D investment in GOVERD (BE-GOVERD as a percentage of GOVERD)

Condition signalling improvement

Increased ratio

Average

1998 - 2002

Actual

2003

Ratio or

difference

UK percentage of BE-GOVERD within total GOVERD - Actual

13.58

8.95

0.66

7.88

7.99

1.01

UK percentage of BE-GOVERD within total GOVERD - Rank

-2

1.72

1.12

UK percentage of BE-GOVERD within total GOVERD / OST comparator
group average percentage of BE-GOVERD within total GOVERD

OST comparator group average percentage of BE-GOVERD within total
GOVERD

BE-GOVERD as a percentage of GOVERD - G8 nations, not USA (no data for Russia)

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

-GOV

D as a percentage of GOV

UK
CANADA
FRANCE
GERMANY
ITALY
JAPAN
EUROPEAN UNION

Data: OECD (RDS & MSTI)

Indicator Headline

BE-GOVERD is R&D performed in the
Government sector that is funded by the
Business Enterprise sector.

For the UK, R&D expenditure from business
sources as a proportion of GOVERD remains
volatile, having dipped to a low-point in the mid-
1990s, peaked in 1999 and is now dipping
again. Nonetheless, UK BE-GOVERD is still
above the OST comparator group average.

The UK remains ahead in the G8 group and
ranks 4th to the Netherlands (which is of a
similar scale in terms of BE-GOVERD), Finland
and Poland overall, declining to 6th in 2003.
There have been substantial improvements in
both Finland and Poland over the period.

There are no data for the USA or China.

Some 2003 data are produced by forecasting.
Forecasted data are 'sense checked' by
comparing forecasts in previous reports with
outcomes.

126

Description of performance indicator

7.03 Business R&D investment in PNPERD (BE-PNPERD as a percentage of PNPERD)

Condition signalling improvement

Increased ratio

Average

1998 - 2002

Actual

2003

Ratio or

difference

UK percentage of BE-PNPERD within total PNPERD - Actual

12.84

10.50

0.82

18.07

18.02

1.00

UK percentage of BE-PNPERD within total PNPERD - Rank

0.71

0.58

OST comparator group average percentage of BE-PNPERD within total
PNPERD

UK percentage of BE-PNPERD within total PNPERD / OST comparator
group average percentage of BE-PNPERD within total PNPERD

BE-PNPERD as a percentage of PNPERD - G8 nations (no data for Germany, Italy or Russia)

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

-P

RD as a percentage of P

USA

CANADA

FRANCE

JAPAN

EUROPEAN
UNION

Data: OECD (RDS)

Indicator Headline

BE-PNPERD is R&D performed in the Private-
Non-Profit sector that is funded by the
Business Enterprise sector.

This indicator has seen substantial change
since last year due to revision to UK PNPERD
data. Total PNPERD has seen a substantial
increase, particularly in recent years, leading to
an apparent relative decrease in the proportion
funded by the Business Enterprise sector.

UK business source expenditure as a
proportion of PNPERD dropped by 50%
between 1998 and 2002, beginning to show
growth again only in 2003.

Among G8 nations, the UK is 4th with Canada
and France well behind Japan. Japan is
exceptional on this indicator. Among the OST
comparator group, the UK's rank position has
remained steady at 10.

In terms of absolute scale of BE-PNPERD,
apart from Japan and the US, only France and
South Korea (another strong performer in this
indicator) approach the UK level.

The OST comparator group average has
remained almost static, but there are marked
variations between countries and greater
fluctuations within some such as Sweden.
Spain, in particular, has markedly improved its
position.

Some 2003 data are produced by forecasting.
Forecasted data are 'sense checked' by
comparing forecasts in previous reports with
outcomes.

128

Description of performance indicator

7.04 Business R&D investment in HERD (BE-HERD as a percentage of HERD)

Condition signalling improvement

Increased ratio

Average

1998 - 2002

Actual 2003

Ratio or

difference

UK - percentage of BE-HERD within total HERD - Actual

6.65

5.56

0.84

6.82

6.83

1.00

UK - percentage of BE-HERD within total HERD - Rank

-3

0.98

0.81

UK percentage of BE-HERD within total HERD / OST comparator group
average percentage of BE-HERD within total HERD

OST comparator group average percentage of BE-HERD within total
HERD

Business R&D investment in HERD - G8 nations (no data for Russia or Italy)

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

-HE

RD as p

ercen

tag

e o

f HE

USA

CANADA

FRANCE

GERMANY

JAPAN

EUROPEAN
UNION

Data: OECD (RDS & MSTI)

Indicator Headline

BE-HERD is R&D performed in the Higher
Education sector that is funded by the
Business Enterprise sector. There is a general
but rather weak correlation between volume
HERD and the level of BE-HERD (see graph).
In other words, BEHERD as a share of HERD
is fairly consistent across a range of countries.

For the UK, the overall increase in relative
business R&D funding within the HE sector in
the mid-1990s plateaued, dipped at the turn of
the millennium, and is now below the EU15
average. There has been a very similar dip for
the USA which is now well below the EU.

BE-HERD has fallen back in South Korea,
where it was high, but is strong and continues
to be maintained in Germany. Both Canada
and Germany are ahead of the UK in absolute
and relative terms. Their data cannot,
unfortunately, be disaggregated at the more
detailed level of Indicator 7.05. Belgium is a
srong performer in this indicator, and
Switzerland, Finland and the Netherlands all
show strong recent growth.

The OST comparator group average fluctuates
from year to year. It generally fell from the
early 1990s but picked up in 2001 because of
simultaneous increases in a number of
European countries. The UK is consequently
now below group average and its rank position
has fallen from 7th to 10th.

Some 2003 data are produced by forecasting.
Forecasted data are 'sense checked' by
comparing forecasts in previous reports with
outcomes.

130

Description of performance indicator

7.05 Business R&D investment in HERD (BE-HERD as %ge of HERD) in five main research fields

Condition signalling improvement

Increased ratio

Medical Sciences

Average

1998 - 2002

Actual

2003

Ratio or

difference

UK - Percentage of BE-HERD within total HERD - Actual

6.51

5.47

0.84

5.85

6.00

1.03

UK - Percentage of BE-HERD within total HERD - Rank

-2

1.11

0.91

OST comparator group average percentage of BE-HERD within total
HERD

UK - Percentage of BE-HERD within total HERD / OST comparator group
average percentage of BE-HERD within total HERD

Indicator Headline

BE-HERD is R&D performed in the Higher
Education sector that is funded by the
Business Enterprise sector.

Data for OECD field Medical Sciences BE-
HERD are available only for 8 countries.
These do not include USA, Canada, France
and Germany. This explains the somewhat
unusual mix of countries shown in the attached
graph.

UK Medical Sciences BE-HERD is second
highest to Japan in absolute terms and third
highest to Poland and Spain as a proportion of
HERD among OST comparator group
countries.

The UK's steady decline reflects a fall-off in BE-
HERD since 2000 in the context of (sometimes
rapidly) rising HERD. Both BE-HERD and
HERD decreased in 2003.

Denmark has been very volatile and Finland
and Poland have seen significant fluctuations
although on a smaller volume than the UK.
Increases have not always been sustained.

Some 2003 data are produced by forecasting.
Forecasted data are 'sense checked' by
comparing forecasts in previous reports with
outcomes.

The EU15 figure is based on a limited data set.

Medical Sciences - BE-HERD as a percentage of HERD

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

-HE

RD a

of HE

UK
JAPAN
AUSTRALIA
DENMARK
FINLAND
POLAND
SPAIN
SWEDEN
EUROPEAN UNION

Data: OECD (RDS & MSTI), UK SET statistics and HESA

132

Description of performance indicator

7.05 Business R&D investment in HERD (BE-HERD as %ge of HERD) in five main research fields

Condition signalling improvement

Increased ratio

Natural Sciences

Average

1998 - 2002

Actual

2003

Ratio or

difference

UK - Percentage of BE-HERD within total HERD - Actual

6.70

5.25

0.78

4.30

4.27

0.99

UK - Percentage of BE-HERD within total HERD - Rank

-2

1.56

1.23

OST comparator group average percentage of BE-HERD within total
HERD

UK - Percentage of BE-HERD within total HERD / OST comparator group
average percentage of BE-HERD within total HERD

Natural Sciences - BE-HERD as a percentage of HERD

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

-HE

RD a

of HE

UK
JAPAN
AUSTRALIA
DENMARK
FINLAND
POLAND
SPAIN
SWEDEN
EUROPEAN UNION

Indicator Headline

BE-HERD is R&D performed in the Higher
Education sector that is funded by the Business
Enterprise sector.

Data are available only for 8 countries at the
OECD field level. These do not include USA,
Canada, France and Germany. This explains
the somewhat unusual mix of countries shown
in the attached graph.

UK Natural Sciences BE-HERD is still highest
in absolute terms (by 1/3rd over Spain) but is
now 3rd as a proportion of HERD among the
group of comparator countries. This element of
UK business funding declined in 2001 after
substantial growth in the last decade.

UK research impact is also strong in this area
and research volume is high in Life Sciences.
Relatively high levels of business investment
may reflect confidence in UK research.

Spain, Finland and - to a lesser extent -
Australia have seen significant relative
increases.

The average OST comparator group figure for
this indicator is fairly stable.

Some 2003 data are produced by forecasting.
Forecasted data are 'sense checked' by
comparing forecasts in previous reports with
outcomes.

Data: OECD (RDS & MSTI), UK SET statistics and HESA data

134

Description of performance indicator

7.05 Business R&D investment in HERD (BE-HERD as %ge of HERD) in five main research fields

Condition signalling improvement

Increased ratio

Engineering and Technology

Average

1998 - 2002

Actual

2003

Ratio or

difference

UK - Percentage of BE-HERD within total HERD - Actual

10.85

9.57

0.88

9.34

8.98

0.96

UK - Percentage of BE-HERD within total HERD - Rank

-1

1.16

1.06

OST comparator group average percentage of BE-HERD within total
HERD

UK - Percentage of BE-HERD within total HERD / OST comparator group
average percentage of BE-HERD within total HERD

Indicator Headline

BE-HERD is R&D performed in the Higher
Education sector that is funded by the
Business Enterprise sector.

Data are available only for 8 countries at the
OECD field level. These do not include USA,
Canada, France and Germany. This explains
the somewhat unusual mix of countries shown
in the attached graph.

UK Engineering and Technology BE-HERD is
highest in absolute terms (just ahead of Japan)
among OST comparator group countries but
lower as a proportion of HERD than that of
Finland. It is of a similar proportion to Spain
and Australia. This element of business
funding has fallen only marginally in absolute
terms and as a proportion of HERD in the UK
in the last decade.

Denmark has seen a significant increase
moving ahead of the UK in 2003, while
Sweden has fallen back somewhat on this
indicator.

UK research impact had been less strong in
this area but has recently improved (see
Indicator 3.09). The pattern of business
investment could reflect wider issues, but the
high absolute level should be noted as an
expression of confidence. Annual variations
may reflect other economic issues faced by the
engineering industry in the UK.

Some 2003 data are produced by forecasting.
Forecasted data are 'sense checked' by
comparing forecasts in previous reports with
outcomes.

Engineering and Technology - BE-HERD as a percentage of HERD

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

-HE

RD a

of HE

UK
JAPAN
AUSTRALIA
DENMARK
FINLAND
POLAND
SPAIN
SWEDEN
EUROPEAN UNION

Data: OECD (RDS & MSTI), UK SET statistics and HESA data

136

Description of performance indicator

7.05 Business R&D investment in HERD (BE-HERD as %ge of HERD) in five main research fields

Condition signalling improvement

Increased ratio

Social Sciences

Average

1998 - 2002

Actual

2003

Ratio or

difference

UK - Percentage of BE-HERD within total HERD - Actual

4.43

3.44

0.78

2.82

1.00

UK - percentage of BE-HERD within total HERD - Rank

-1

1.57

1.22

UK percentage of BE-HERD within total HERD / OST comparator group
average percentage of BE-HERD within total HERD

OST comparator group average percentage of BE-HERD within total
HERD

Indicator Headline

BE-HERD is R&D performed in the Higher
Education sector that is funded by the
Business Enterprise sector.

Data are available only for 7 countries at the
OECD field level. These do not include USA,
Canada, France, Germany or Japan. This
explains the somewhat unusual mix of
countries shown in the attached graph.

UK Social Sciences BE-HERD as a proportion
of HERD continues to be the highest in
absolute terms among comparator countries,
marginally ahead of Spain.

The UK is third highest to Spain and Finland as
a proportion of HERD. Finland has improved
significantly over the period, and overtook the
UK in 2001. The Poland line drops because
BE-HERD declines to almost zero while HERD
steadily rises in that country.

This element of business funding has grown
slowly in absolute terms, in line with total
HERD in the UK in the last decade. This
pattern is typical of the rest of the group.
Overall across the OST comparator group
there is a steady rise in business expenditure
in these HE research areas in line with the
general level of public investment.

Some 2003 data are produced by forecasting.
Forecasted data are 'sense checked' by
comparing forecasts in previous reports with
outcomes

Social Sciences - BE-HERD as a percentage of HERD

-1

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

-HE

RD a

of HE

AUSTRALIA

DENMARK
FINLAND

POLAND

SPAIN

SWEDEN
EUROPEAN UNION

Data: OECD (RDS & MSTI), UK SET statistics and HESA data

138

Description of performance indicator

7.05 Business R&D investment in HERD (BE-HERD as %ge of HERD) in five main research fields

Condition signalling improvement

Increased ratio

Humanities

Average

1998 - 2002

Actual

2003

Ratio or

difference

UK - Percentage of BE-HERD within total HERD - Actual

1.30

1.07

0.82

1.76

2.00

1.14

UK - percentage of BE-HERD within total HERD - Rank

0.74

0.54

OST comparator group average percentage of BE-HERD within total
HERD

UK percentage of BE-HERD within total HERD / OST comparator group
average percentage of BE-HERD within total HERD

Indicator Headline

This is the first time that data for this indicator
have been analysed for Humanities and Arts.

BE-HERD is R&D performed in the Higher
Education sector that is funded by the
Business Enterprise sector.

Data are available only for 7 countries at the
OECD field level for Humanities & Arts. These
do not include USA, Canada, France,
Germany or Japan. Data for Sweden only
cover some years. This explains the
somewhat unusual mix of countries shown in
the attached graph.

UK BE-HERD is about half that of Spain but
much more in absolute terms than other
countries. As a proportion of HERD, the UK is
broadly in the main group of countries where
data are available. The level of investment
has been stable over the period.

The index for Spain is extremely high and this
is due to high BE-HERD not low HERD. The
Poland figures are slightly erratic.

Excepting Spain, the UK appears to attract a
relatively good level of BE-HERD into a well
supported research base in these disciplines.

Some 2003 data are produced by forecasting.
Forecasted data are 'sense checked' by
comparing forecasts in previous reports with
outcomes.

Humanities - BE-HERD as a percentage of HERD

1995

1996

1997

1998

1999

2000

2001

2002

2003

Year

-HE

RD a

of HE

UK
AUSTRALIA
DENMARK
FINLAND
POLAND
SPAIN
SWEDEN
EUROPEAN UNION

Data: OECD (RDS & MSTI), UK SET statistics and HESA data

140

143

Data and sources

With the exception of the UK, the main data sources used for OST indicators
are:

•

Finance and people – OECD

•

Publications – Thomson Scientific

For the UK, OST has drawn our attention to the most recent data available from
SourceOECD and ONS.

The OECD is the main provider of internationally comparable data on research
and development. Its two products on the measurement of science and
technology, Main Science and Technology Indicators (MSTI, 2005-1, first
edition) and Research and Development Statistics (RDS, 2004-2, second
edition) provide the basis for much of the data used in these analyses. The
latest editions of each of these products were used to generate the indicators
listed in the table below. In addition, OECD provides the only reliable
international comparisons of educational data via its online Education
Database.

The OECD provides comments on a number of the data points in RDS and
MSTI, explaining their derivation or discussing their accuracy. These
comments have not been reproduced here but are available to the interested
reader when referring to the original data. Sources can be found at:

http://www1.oecd.org/dsti/sti/stat-ana/index.htm

There are some points of difference between MSTI and RDS. MSTI has been
the preferred database for most of the analyses in this report as it provides data
on a greater range of countries. RDS, however, provides data disaggregated at
the level of fields of science and also covers PNPERD.

Data are presented for the years 1995 to 2003/4. Gaps of one or two years in
the time series have been filled by interpolation, whilst missing values at either
end of the time series were created by extrapolation where appropriate.

Rolling 5 year averages have been created for researchers, population and
labour-force data in order to enable like-for-like comparisons with ISI data. The
average is produced from the value for the year in question, and the four years
which precede it.

In this report, OECD data are usually available for 21 countries. Coverage for
the 17 OECD nations is broadly complete, but data for some countries are

missing from some tables. This may be because there were no data available,
or that there were so many missing data points in the data available that no
meaningful attempt to interpolate and extrapolate could be made.

Where necessary and feasible, OECD data has been supplemented by data
sourced from EUROSTAT, the UN, the Higher Education Statistics Agency
(HESA), and the statistics’ portals of individual national governments.

Financial data is given in units of Million constant US$ at 2000 prices and
corrected for Purchasing Power Parity (PPP). This is a change from last year,
where PPP-1995 was used.

The interpretation of OECD science and technology data is governed by the
Frascati Manual, which has become the internationally recognised methodology
for collecting and using R&D statistics. Some basic definitions from the
Frascati manual appear below; detail is in the Glossary.

The OECD Education Database provides internationally comparable data on
key aspects of education systems. It makes use of data collected by UNESCO,
OECD and EUROSTAT. The interpretation of OECD education data is
governed by the OECD publication ‘Data Collection on Education Systems:
Definitions, Explanations, and Instructions’.

Indicator Basic

source

BE-GOVERD

OECD Research and Development Statistics

BE-HERD

OECD Research and Development Statistics

BE-HERD by field of science

OECD Research and Development Statistics

BE-PNPERD

OECD Research and Development Statistics

GDP

OECD Main Science and Technology Indicators

GERD

OECD Main Science and Technology Indicators

GOVERD

OECD Main Science and Technology Indicators

HERD

OECD Main Science and Technology Indicators

HERD by field of science

OECD Research and Development Statistics

National populations

OECD Main Science and Technology Indicators

PhD graduates

OECD Education Database

PhD graduates by field of science

OECD Education Database

PNPERD

OECD Research and Development Statistics

R&D personnel

OECD Main Science and Technology Indicators

Researchers

OECD Main Science and Technology Indicators

Labour (work) force

OECD Main Science and Technology Indicators

144

There has been a change in UK postgraduate data because of changes in data
collection by HESA, the UK Higher Education Statistics Agency. This affects
the status of awards made from dormant registrations and increases the
numbers of PhD awards from 2001 onwards by about 4.5% compared to
previous data. More information is available in an article published by HESA at
the time, see:

http://www.hesa.ac.uk/holisdocs/pubinfo/student/changes.htm

All publication and citation data are provided by Thomson Scientific®. The
NSICCOD2004 was the specific database from which figures were taken for
these analyses. Two main methods are used in analysing these data:

•

NSI1

: analyses based on data from the most recent (or any specific)

calendar year use the ISI NSI1 data frame, looking at the numbers of
articles published and the citations they have accumulated to date;

•

NSI5

: analyses based on a select period are most effective if a five-

year window is taken, using the ISI NSI5 data frame. This takes the
publications for a stated five-year period (e.g. NSI5 for 2004 is the 5-
year window 2000-2004) and the citations to those articles in the same
five-year period.

Frascati manual data definitions (