The different channels of university-industry knowledge transfer: Empirical evidence from Biomedical Engineering

1
The different channels of university-industry
knowledge transfer Empirical evidence from
Biomedical Engineering

• Reg Brennenraedts
• Dialogic Innovatie InteractieBrennenraedts_at_dial
  ogic.nl
• Bart Verspagen
• Technische Universiteit EindhovenB.Verspagen_at_tm.t
  ue.nl
• Rudi BekkersTechnische Universiteit
  EindhovenR.N.A.Bekkers_at_tm.tue.nl
• DIME, Workshop on Technology transfer from
  universities A critical appraisal of patents,
  spin-offs and human mobility September
  29-30/2006, Lausanne

2
Talk outline

• Research questions
• Theoretical framework
• Methodology
• Analysis of the data
• Conclusions
• Discussion

3
Research questions
Research questions

• How do industry-science relations take place at
  the faculty of Biomedical Engineering at the
  Eindhoven University?
• What is the relative frequency of the different
  forms of ISRs?
• What is the perceived importance of the
  different forms of ISRs?
• Which factors influence the pattern of ISRs?

4
Context
Theoretical framework

• Science increasingly more important for economic
  growth
• European paradox
• Europe excels in scientific research
• However
• Commercial/technological performance in high tech
  sectors is decreasing
• Large differences in ISRs occur between countries
  and universities
• Purpose of ISRs ? Knowledge transfer

5
Forms of ISRs
Theoretical framework

• Publications
• Participation in conferences, professional
  networks and boards
• Mobility of people
• Other informal contacts
• Cooperation in RD
• Sharing of facilities
• Cooperation in education
• Contract research and advisement
• Intellectual property rights
• Spin-offs and entrepreneurship

6
What causes different footprints of ISRs?
Theoretical framework

• Between sector variation
• Knowledge base (in casu BME)
• Nelson and Winter (entrepreneurial / routinized)
• Schumpeter (Mark I widening / Mark II deepening)
• Pavitt (supplier dominated, production intensive,
  science)
• Within sector variation
• Reputation of a researcher
• Exact type of research one conducts
• Applied vs. not-applied
• Multidisciplinary vs. monodisciplinary
• Social network of an individual
• Weak ties (acquaintances)
• Strong ties (friends)
• National or Regional System of Innovation
• Policy of faculty or university (regarding ISRs)

7
Faculty of Biomedical Engineering (BME)
Theoretical framework

• 200 employees, 400 students
• Founded in 1997
• Cooperation between
• Eindhoven University (TU/e)
• Maastricht University (UM)
• Teaching hospital Maastricht (azM)
• Focus on
• BMTE (BioMechanics and Tissue Engineering)
• MBEMI (Molecular BioEngineering Molecular
  Imaging)
• BIOMIM (BIOMedical Imaging and Modeling)
• Knowledge base in Physics, Chemistry,
  Mathematics, Electronics, Medicine Biology

8
Back to our research question
Methodology

• Relative frequency / perceived importance of the
  different forms of ISRs?
• Which factors influence the pattern of ISRs?
• Dependent variables frequency/perceived
  importance of ISRs
• Independent variables Properties of researchers

9
Obtaining data (i)
Methodology

• Focus on knowledge producers, not RD managers
• Publication and citation scores
• Web of Science database
• Questionnaire
• Population are all the researchers employed at
  BME (n138)
• Response gt62 (n85)

10
Obtaining data (ii)
Methodology

• Questionnaire contains questions regarding
• Background of the researcher
• Position at university?
• Other position?
• In the past employed in industry?
• Monodisciplinary or multidisciplinary research?
• Applied or fundamental or experimental research?
• Patents?
• Frequency/perceived importance of forms of ISRs
• 21 different forms of ISRs

11
Methodology
12
Relative frequency and perceived importance of
ISRs
Analysis of the data

• Highly correlated
• Correlation coefficient 0.95
• Rank correlation 0.92
• People do what they find important
• No possibility (or need) to distinguish between
  these variables
• Further analysis is conducted using the sum of
  these scores

13
Perceived importance and frequency of ISRs (top-5)
Analysis of the data

• 1. Conferences and workshops
• 2. Refereed scientific journals or books
• 3. Joint RD projects with the industry
• 4. Networks based on friendship
• 5. Presentation of research at the industry

14
General patterns in ISRs (using Factor Analysis)
Analysis of the data

• Factors
• 1. Entrepreneur
• 2. Dense cooperation
• 3. Formal network
• 4. Science
• 5. Informal network

15
Towards a taxonomy (i)(using cluster analysis)
Analysis of the data

• Clustering the respondents by their factor
  scores
• Cluster I (n24) high factor scores on informal
  networking
• Cluster II (n18) high factor scores on science
  
• Cluster III (n14) high factor scores on
  science, formal network, informal network
• Note Factors entrepreneur and dense cooperation
  are not preffered by a specific group

16
Towards a taxonomy (ii)(using multinomial logit
regression)
Analysis of the data

• Cluster II (opposed to Cluster I) is typified by
  researchers who usually
• Do not have another appointment
• Do not have worked in a firm
• Do not have any patents
• Do have conducted mainly fundamental
• Do have many publications.
• Cluster III (opposed to Cluster I) is typified by
  researchers who usually
• Do have another appointment (relative risk ration
  gt60)

17
The taxonomy (i)
Conclusions

• Some ISRs are appreciated by a broad set of
  respondents
• Spin-offs, patents, contract research, et cetera
  (factor Entrepreneur)
• Exchange of personnel, sharing facilities, joint
  RD, et cetera (factor Dense cooperation)
• Some ISRs are preferred by a specific type of
  respondent.

18
The taxonomy (ii)
Conclusions

• Cluster I
• prefers friendships, presentation at the
  industry, et cetera (factor informal networking)
• Cluster II
• prefers refereed publications, conferences,
  supervision of a Ph.D. student (factor science)
• contains Traditional academics
• Cluster III
• prefers many different channels (factors
  informal networking, formal networking, science)
• contains academics with more then 1 appointment

19
Policy implications
Conclusions

• Much variation found in transferring knowledge
• Policy should be aimed at a multitude of
  channels and a wide range of channels.
• Academics with a strong reputation prefer to use
  the traditional (rather passive) channels
• Although an interesting match for the industry,
  could possibly be hard to motivate to use the
  more active channels of knowledge transfer.

20
Suggestions for further research
Conclusions

• Research at another sector (faculty) ? verify
  within sector variations
• Research at broad scope of faculties ? find
  between sector variations
• Research at the industry ? do they have the
  same opinion?

21
Discussion
Discussion
22

• N24

23

• N24