leila tahmooresnejad_impact of public funding on the development of nanotechnology a comparison of...

Impact of Public Funding on the

Development of Nanotechnology

A Comparison of Quebec,

Canada and the US

Leila Tahmooresnejad – Polytechnique Montréal

Catherine Beaudry – Polytechnique Montréal

Andrea Schiffauerova – Concordia University

1st International Conference of Ne3LS Network

November 2012

Outline of the presentation

Motivation

Theoretical Framework

Data and Methodology

Network

Hypotheses

Econometric models

Regression results

Conclusion

1 November 2012 2 Leila Tahmooresnejad - Catherine Beaudry

Andrea Schiffauerova

Motivation

Public funding for research facilitates the production of knowledge and is a key element for innovation in high technologies

Facilitate the diffusion of knowledge

Develop new technologies

Universities and their affiliated centers play a vital role in National innovation systems (Hall et al.,

2003; Link & Scott, 2004; Zucker, Darby & Armstrong, 2002)

1 November 2012 3

Leila Tahmooresnejad - Catherine Beaudry Andrea Schiffauerova

Nanotechnology (I/II)

Emergence of nanotechnology over recent years

was the starting point for many changes in a vast

number of industries.

High competitive advantage for companies (Canton, 1999)

Creation of new companies (Porter et al., 2007)

Nano-enabled products with optimal features (Armstrong,

2008; Vokhidov and Dobrovol’skii, 2010)

Potential markets (Knol, 2004; Roco, 2007; Malanowski and

Zweck, 2007)

Nano-related jobs (Freeman and Shukla, 2008)



Nanotechnology (II/II)

Nanotechnology requires considerable investment

Most of countries are following the US in initiating nanotechnology programs and increasing the allocated funds (Sargent, 2008)

Canada lags behind in the race of nanotechnology

1 November 2012 5


Theoretical Framework (I/II)

Positive correlation between federal research

funding and scientific outputs (Adams and Griliches,

1998; Payne and Siow, 2003; Blume-Kogut et al. 2009).

More government research funding results

more papers (Payne and Siow , 2003)

More government research funding results

more patents with a lower rate (Payne and Siow,

2003)



Theoretical Framework (II/II)

High quality research should obtain more

citations (Raan et al., 2003)

Citations are 'proxy' (Cronin, 2005)

Papers and Patents of researchers, who

received funding, may receive more citations

e.g. Patents of researchers, who received NSF

funding, received more citations compared with

those of other researchers in Nanoscale Science

and Engineering (Huang et al., 2005).

1 November 2012 7


Objectives

Measure the impact of grants and contracts on the outputs of academic researchers

Papers ( quantity and quality)

Patents ( quantity and quality)

Measure the impact of scientific and technological networks ( co-publication and co-invention networks)

Compare these impacts in Quebec, Canada and the US



Data and Methodology

Data (I/II)

Scopus

Extraction of nanotechnology scientific papers by using specific keywords in the title, abstract and keywords

Selection the articles where there is at least one Canadian author

United States Patent and Trademark Office (USPTO)

Extraction of nanotechnology scientific patents by using specific keywords in the title, abstract and keywords

Selection the patents where there is at least one Canadian inventor



Data (II/II) Systèmes d’information de la recherche universitaire (SIRU) for

Quebec Amounts of grants and contracts received by researchers in Quebec

Database of three granting councils (CIHR(Canadian Institute for

Health Research), NSERC (Natural Sciences and Engineering Research

Council), SSHRC (Social Sciences and Humanities Research Council of

Canada)) Amount of grants received by Canadian researchers

Nanobank Papers of the researchers in the US

Patents of the researchers in the US

Amount of grants( NIH (National Institutes of Health) and NSF(National

Science Foundation) received by researchers in the US



Methodology

Matching databases

Creating a unique identifier for each individual

researcher

Data cleaning

Creating co-publication and co-invention

networks

Calculating network characteristics and the

position of researchers

1 November 2012 12


Network (I/III) A, B and C have published an

article or are the inventors of a patent

A, B and E have published an article or are the inventors of a patent

C and D have published an article or are the inventors of a patent

Degree of a node Number of links that are directly

connected

A, B and C have 3 connections

E has 2 connections

D has 1 connection

C

A B

D

E



Network (II/III)

Centrality degree

indicates the number of actors that are connected to a

specific actor

Geodesic distance

Distance (shortest path) between two nodes

Betweenness centrality of a node

is defined as the proportion of all geodesic distances

between two nodes that includes this node.

It makes the node more powerful since it can control the

knowledge flow between the other pair of actors



Network (III/III)

Clustering coefficient

if two nodes are connected to the specific third

node, they may also be connected to each other.

It is computed as the fraction of pairs of neighbors

of an actor that are directly connected each other.

C

A B

D



Hypotheses (I/II)

Hypothesis 1a: Nanotechnology scientists/ academic inventors who receive more public funding contribute to more publications/patents compared with scientists/ academic inventors who receive less or no public funding.

Hypothesis 1b: Nanotechnology scientists/ academic inventors who receive more public funding contribute to higher quality publications/patents compared with scientists/ academic inventors who receive less or no public funding.

1 November 2012 16


Hypotheses (II/II)

Hypothesis 2a: A better network position of

scientists/ academic inventors has a positive effect

on the number of papers/patents to which a

scientist/ academic inventor contributes.

Hypothesis 2b: A better network position of

scientists/ academic inventors has a positive effect

on the quality of papers/patents to which a

scientist/ academic inventor contributes.

1 November 2012 17


Econometric Models (I/II) nbArtit / nbPatit

nbCitit

nbClaimit

é

ë

êêê

ù

û

úúú

=a + bS1TotSubvMoy3it-l + bS2 TotSubvMoy3it-l[ ]2

+bC1TotContMoy3it-l + bC2 TotContMoy3it-l[ ]2+ bP1nbPat3it-1 + bP2nbPat3it-1

2

+gbBtwCentXit-2 +gc1CliqnessXit-2 +gc2 CliqnessXit-2[ ]2

+gbp[BtwCentXit-2 ´nbPat3it-1]+gbc[BtwCentXit-2 ´CliqnessXit-2 ]

+dt dtt

å +n i +eit

The amount of average grants / contracts that are received in 3 years preceding the patent application / paper publication with one year lag

The betweenness centrality of academic –inventors /scientists in the co –invention/ co –publication network over 3 years preceding the patent application/ paper publication with 2 years lag

The cliquishness centrality of academic –inventors / scientists in the co –invention / co –publication network over 3 years preceding the patent application /paper publication with 2 years lag

TotSubvMoy3it-l TotContMoy3it-l

BtwCentXit-2

CliqnessXit-2



Endogeneity Problem

The explanatory variables are linked together since one can explain the other. The number of papers/patents is explained by the total

grants/contracts received

Two–Stage Residual Inclusion (2RSI) and Two –Stage –Least –Squares (2SLS)

Instrumental variables Age :the number of years since the beginning of the career

of researcher in nanotechnology

Chair :value 0 if a researcher has no chair, 1 if he has an industrial chair, 2 for being a chair from two councils of the Canadian federal granting, 3 for a scientist who is a Canada Research chair

1 November 2012 19


Econometric Models (II/II)

ln totSubvMoyXit-1( ) = a1 +x Ait-ii=2

4

å + lA1Ageit-1 + lA2Age2

it-1 + lChChairi + lanbArtMoy3

+Variables1stStage+ n1i +e1it( )

nbArtit / nbPatit

nbCitit

nbClaimit

é

ë

êêê

ù

û

úúú

= a2 + bG1 ln TotSubvMoy3it-1( ) + bG2 ln TotSubvMoy3it-1( )éë ùû2

+ n1i +e1it[ ]

+bC1 ln TotContMoy3it-1( ) + bC2 ln TotContMoy3it-1( )éë ùû2

+gbBtwCentXit-2 +gc1CliqnessXit-2 +gc2 CliqnessXit-2[ ]2

+gbp[BtwCentXit-2 ´ nbPat3it-1]+gbc[BtwCentXit-2 ´CliqnessXit-2 ]

+ dtdtt

å +n2i +e2it



Regression results



Comparison

Quebec

Rest of Canada

The US



The Impact of Public Funding

on Papers

Quebec (contracts and grants)

The number of papers

Positive impact of grants after threshold (right graph)

Negative impact of contracts (left graph)

Positive impact of network characteristics

Positive impact of having patents

1 November 2012 24


The number of citation

Negative impact of grants before threshold (right

graph)

Positive impact of contracts (left graph)



1 November 2012 25

Quebec (only grants)

The number of papers (left) and citations (right)

Positive impact of grants until reach the

threshold



The Impact of Public Funding on the number of papers (left graph), and on the number of citations (right graph) 1 November 2012 26

Rest of Canada

The number of papers (left) and the number

of citations (right)

Positive impact of grants until reach the threshold



1 November 2012 27

The US The number of papers (left) and the number of

citations (right)

Positive impact of grants on the number of papers


Positive impact of network characteristics (only

citations)

1 November 2012 28

The Impact of Public Funding

on Patents


The number of patents

Positive impacts of contracts after pass the

threshold

No effect of grants


cliquishness)

1 November 2012 30

Quebec (contracts and grants) The number of citation

Positive impact of contracts after a certain threshold (

left graph)


(right graph)


1 November 2012 31


The number of claims

Positive impact of contracts after pass the threshold (left

graph)

Positive impact of grants after pass the threshold (right

graph)

Positive impact of network characteristics (cliquishness)

1 November 2012 32

Rest of Canada


No effect of grants


cliquishness)



Rest of Canada

The number of citations (left) and the

number of claims (right)



cliquishness has effect on citation)

1 November 2012 34

The US


Positive linear impact of grants


cliquishness)



The US

The number of citation (left) and the

number of claims (right)

Positive impact of grants


cliquishness)

1 November 2012 36

Conclusion

Conclusion (I/III)

Scientists work in bigger teams, but inventors

are in smaller groups

Scientific network is more interconnected

compared with technological networks which

are fragmented

Having central positions in scientific networks

has more positive impact on the papers

compared with technological networks



Conclusion (II/III) Positive impact of grants on scientific productions and their

quality but there is a threshold for this impact in Canada

Positive impact of grants on scientific productions and their quality in the US, the threshold only for the citation

No impact of grants in Canada on the number of patents , but positive impact of grants in the US on the number of patents

Positive impact of grants on quality of patents, but there is the threshold in Canada

Positive impact of grants on quality of patents with no threshold in the US



Conclusion (III/III)

Negative impact of contracts on the number of papers

Positive impact of contracts on the quality of papers

Positive impact of contracts on the number of patents after passing the threshold

Positive impact of contracts on the quality of patents considering the threshold

Contracts are more crucial for patents, but we could not measure this impact for the rest of Canada and the US

1 November 2012


40

Thank you



leila tahmooresnejad_impact of public funding on the development of nanotechnology a comparison of...

Documents

patents of researchers

us patents

quality patents quantity

emergence of nanotechnology

impact of grants

impact of public funding

inventors of apatent

usingspecific keywords