Organic, fine chemistry - Science, Technology and Innovation

drawerbeamerΒιοτεχνολογία

6 Δεκ 2012 (πριν από 4 χρόνια και 8 μήνες)

140 εμφανίσεις


Measuring the Impact of
Science in the Knowledge
-
Driven Economy


Martin Meyer

Steunpunt O&O Statistieken

K.U. Leuven,
Belgium

SPRU

University of Sussex, Brighton, UK

martin.meyer@knowledgeflows.net

www.knowledgeflows.net
,
www.techtransfer.org.uk



Outline


Different Ways to track the impact of
science:


a focus on science
-
technology linkage


comparison of indicators


How to measure use beyond the
economic?


Outlook

Different ways of linking science
and technology


Patent citation analysis:


Linking references in patents to scientifc
publication data


Tracking university
-
owned and other
academic patents


variety of aproaches


To
illustrate…

A Typical
US Patent

SCI

Patent Citations


Popular indicator but

problematic
interpretation:


not necessarily direct, causal link between science
and technology


’some form’ of, at times, reciprocal relationships
between s & t


Interpretating Patent Citation
Links


Explorative analysis of a small set of patents in nanotechnology to see to
what extent their science citations can be related to science
-
related
knowledge inputs into the inventions (
Research Policy,
29 (2000) 409
-
434
)



Inventor interviews indicated that


inventions tend to be based on
general experience

in research and teaching
rather than specific science inputs,


inventions in individual cases were stimulated by (often
uncited
) articles in
downstream specialist or trade journals,


research papers cited in patents as
general background

information,


science references used to attack, restrict, or modify claims,


inventors indicated that some front
-
page citations were
not known

to them prior
to the examination process.


Conclusion:


misleading to assume that
p
atent citation links would necessarily indicate
knowledge transfer from science to technology, especially with respect to the
cited paper and citing patent.


Interpretating Patent Citation
Links (2)


Tijssen et al.

examined the extent his interpretation can be applied
to the general context.


study was based on a sample of 50 patents from the set of Dutch
USPTO patents.


survey of the inventors on scientific contributions to their patents.


The authors suggest that many domestic non
-
patent citations are
researcher/inventor given and therefore likely to refer to ‘direct links’
but ‘not necessarily causal relationships’ between science and
technology.


NB

Authors inquired about link of patent to ’science’, not about a
knowledge
-
input
relationship between patents and cited scientific
papers



If interest on personal or direct knowledge input, other indicators may be
more appropriate

University
-
owned and other
academic patents


Seem to be more appropriate in this
context


Yet severe limitations if one looks at
university
-
owned patents only!!!

Motivation: Modes of TT

University
Ownership


of
IPR

No

University
Ownership of IPR

University

Transfer
Office
Involved

University Transfer
Office
Not

Involved


Mediated

Mode’


Direct
Mode’


Intermediary

Mode’

Only here university

listed as assignee

Prof.
Pertti Törmälä,


Tampere Univ Tech

Methodology

’Collaborative

patents’

Surveying
inventors

’Industrial uni
-

related patents’

’Academic

patents’

US patents

Personnel registry

Matching inventor with
researcher names

Researcher/
inventor

name pairs

Cleaning lists
(homonyms)

University
-
related

patents

Interviewing
Inventors

Case Studies

Technology Profiles

Regional Distributions

Inventor concentrations

Specializations

Maps

Sectoral Distributions

Relevance of research

Public Funding

Utilization

Licensing

Exemplifications

Opinions about IP regulations

Analyses based on
data generated in this
project:

13000 inventor names in 6,500 patents

~24,000 names in personnel registries
of 12 Finnish universities

----------------------------------------------------

Bibliometric matching

(on family and first name/initial basis)


Identified ~
1200

hits

----------------------------------------------------

Personal follow up of hits (email,
telephone)

----------------------------------------------------

655

hits proved correct


530

patents
related to universities


610
individualized

questionnaires sent
out to 270 inventors

----------------------------------------------------

(inventors were asked to respond
patent
-
by
-
patent)

----------------------------------------------------

Response rate: 39.8%


32 case studies

(not discussed here)

Results of the linkage:


530

university
-
related patents with
285

university
-
researchers as
inventors could be identified


For comparison:


less than 40

patents were assigned to universities, their TLOs or public
funding agencies


About
100

patents contained citation links to Finnish authored SCI papers


University
-
related patents account for
about 8%

of the total national
inventive output


conservative measure:


as only two years of personnel registry data was available for analysis and
US patent data was used (time lag, long examination times, especially for
life science related applications)


US Patents (IY)

1997

2000

Personnel registry data:

1986

2001

Researcher patents:

A comparison

36
120
494
242
0
100
200
300
400
500
600
Finland
Flanders
University-owned
Not university-owned
7%
33%
93%
67%
0%
20%
40%
60%
80%
100%
Finland
Flanders
University-owned
Not university-owned
Issues to be explored:


Differences in Linkage Patterns:


Can they be explained?


National importance of researcher
inventions


Where are research
-
based inventions
utilized?

Differences in Science
-
Technology
Linkage Patterns


Telecoms (ICT) largest field of academic patents


Yet
Chief Technological Contributions

of University
-
based Inventors
in
Pharmaceuticals,

Biotechnology, and Organic chemistry


SEM
ORG
ANA
CPP
OPT
EDE
TEL
MCT
MAP
PHA
BIO
MED
MAC
MET
SUC
INT
AVT
AGR
EPT
0%
2%
4%
6%
8%
10%
12%
14%
16%
18%
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
Specialization*
Patent Share**
Niche
Contributions
Background
Contributions
Background
Contributions
Core
Contributions
Core
Contributions
Marginal
Contributions
SEM
ORG
ANA
CPP
OPT
EDE
TEL
MCT
MAP
PHA
BIO
MED
MAC
MET
SUC
INT
AVT
AGR
EPT
0%
2%
4%
6%
8%
10%
12%
14%
16%
18%
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
Specialization*
Patent Share**
Niche
Contributions
Background
Contributions
Background
Contributions
Background
Contributions
Background
Contributions
Core
Contributions
Core
Contributions
Marginal
Contributions
* compares share of field amongst academic patents to overall patenting activity

** here the share of fields amongst university
-
related patents

ICT
ANA
PHA
BIO
ORG
MED
MAP
EDE
Other
0%
5%
10%
15%
20%
25%
30%
35%
40%
0%
5%
10%
15%
20%
25%
30%
35%
40%
Academic Patents
Patent Citations
Science link of tech fields:

Share of tech. areas of acad. pat. v their share of science citations

How to Explain Differences in Linkage:

An Inventor Survey


(Purely) Academic patents:

All

the inventors were working
in a university at the time of the invention (
36% of the
university
-
related patents
), mostly in:


Analysis, measurement, control (19% of the academic patents)


Pharmaceuticals/cosmetics (15%)


Organic, fine chemistry (13%)


Collaborative patents:

One or more inventors were
working in a university while the others were employed in
industry at the time of the invention (
25%
), mostly in:


Pharmaceuticals/cosmetics (21%)


Macromolecular chemistry/polymers (10%)


Organic, fine chemistry (10%)


‘Industrial’ patents:

The inventors were working in the
university but were employed by industry at the time of the
invention (
39%
), mostly in:


Telecommunications (28%)


Materials processing (14%)


Surfaces, coating (9%)


Biotechnology

or
organic chemistry

feature prominently among
purely academic

and
collaborative patents
,


telecommunications

patents dominate the third category of

industrial’ patents
.






Different types of university
-
related patenting seem to coincide with
different technological areas and may point to different transfer
patterns.


Areas such as telecommunications and materials processing seem to
be more characterized by
person
-
embodied

transfer and exchange of
knowledge.

National Impact????


Overall
importance of
tech fields with
researcher
activity still low


Patent Share
here reflect the
importance of
tech fields at
the country
level (not from a
researcher
perspective)


Focus on large firms
as patent owners


Few start
-
ups


Potential gaps in
biotech/med.
engineering


Many ‘unassigned’
patents (~16%)


Distributed unevenly
across all universities


Up to 46% of patents
remained unassigned
in one instance


Foreign assignations

Where are

Researcher

Inventions

utilized?


Where are inventions
exploited?



Purely
Academic

Collaborative

Industrial

1

Respondent
’s research directly relevant but not
funded

25%

21%

44%


Respondent’s research directly relevant and funded

64%

69%

47%


Respondent’s research NOT directly relevant

11%

10 %

9%

2

Funding for patent invention

51%

28%

35%


No funding for patent inventi
on

49%

72%

65%

3

Funding for further utilization

25%

13%

18%


No funding for further utilization

75%

87%

82%

4

Licensed

41%

22%

38%


Not licensed

59%

78%

62%


0%
10%
20%
30%
40%
50%
60%
70%
80%
Start-up
Estd. SME
Large firm
University
Publ. Res. Ctr.
Other/no util.
Academic
Collaborative
Industrial
Where are the start
-
ups, research

spin
-
outs ???

Even low rate among

purely academic ventures

... which tended to receive funding

for IP protection but not for other purposes

Anecdotal evidence
raises further questions
about effectiveness of
support mechanisms...


Pointers to
use beyond
the
economic?

Non
-
Economic Impacts of
Science?


Not all academic inventions identified were
used in a commercial context


Substantial Share of Patents are used in
another context or not at all


Case studies point to a number of
motivational issues

Inventor Motivations


Scientists may be
motivated also by
non
-
commercial
reasons:


increasing chances
of getting research
funding,


attracting students,


keeping control
over the idea and
its exploitation, or


‘keeping research
ticking over’



Another study
corroborates results

#

Characteristics

Type*

Employment
status of the
inventor

Mode of
Utilization

Ultimate purpose
of, or motiva
tion
for invention

Defining characteristic

1

Inventor/entreprene
urs in start
-
up
companies

Recently self
-
employed

Use in newly
established
company

Entrepreneurial
intent, technical
interest

Considerable technical
knowledge and relative
lack of business s
kills

2

Proprietor
-
inventors
in existing SMEs

Self
-
employed for
long period of time

Use in their
(already
existing)
company

Use in firm,

Technical motivation

Inventors with business
experience

3

Patent transfers
and Licensing

Mixed

Use in
organization of

new owner /
licensee

Technical motivation

Failure to find
appropriate partners**

4

Academic inventors

Employed by
university

Mixed

Technology and
research based

Institutional environment
provides support and
security but is major
hurdle too

5

Invention
too novel
for intended market

Mixed

No apparent
application

Technology
-
based,
also commercial
intent

Failure to commercialize
or abort efforts at an
appropriate stage

6

Non
-
financial
-
success

Mixed

No utilization
in economic
context

Making invention
public
ly accessible,
Increasing public
wealth,
Technical interest

Not explicitly financial
motivation

7

Secret inventions

Mixed

No permissible
use

Technology
-
based,
also commercial
intent

Failure to commercialize


Conclusions


S T Indicators:


More science
-
technology links through academic patent approach than
others


need for using several linkage indicators (esp. exploring
differences)


Limitations as to what can be inferred from patent data alone
necessitates combination with survey tools to explore innovation
process further


TT:


Considerable level of inventive activity


Different types and intensities of links depending on tech areas


Highly concentrated, connected mostly to large firms


University start
-
ups/spin
-
off companies not necessarily the major
vehicle of tech transfer


Non
-
economic impacts:


Pointed to a number of patented inventions which are used in other
commercial contexts, or in response to increased commercialisation of
science


Outlook


Extending data coverage


Extending research to
cover EPO and domestic data


Extending the
time dimension

of the study


Zooming in on
relevant technology

areas


Comparison with data from
other countries


Exploring further approaches:


Inventor
-
author analysis:


Co
-
active knowledge producers: Does economically oriented activity
of academics affect their scientific impact?


Exploring non
-
commercial utilisation of patents


What are academics’ motivation for patenting?