Você está na página 1de 18

How to Measure Innovation

How to Measure Innovation


Balkrishna C. Rao

Innovation will have an increasingly important place


in the future prosperity of the world. But can we
develop reliable models for quantifying innovations
that enahnce social capital? This piece describes
the work of one economist who has developed a
technically demanding but fascinating methodology
to help evaluate a nations quality and quantity
of innovation. Such measures are often technically
elaborate, but it is important to present high-quality
technical attempts at dealing with difficult but
important subjects.

umankind has perhaps been best defined by its creative abilities

since time immemorial. Thanks to human creativity, major


industrial revolutions have occurred, resulting in a modern
society with advanced amenities. The ushering in of globalization has
made innovation all the more important for member countries of the
Organization for Economic Cooperation and Development (OECD),
in view of explosive growth rates experienced by several countries of
the other world.
Three factors account for these rapid growth rates. First, countries
that comprise the East Asian tigers, i.e., Singapore, South Korea, Taiwan, and Hong Kong, are at the forefront of the technological race,

Balkrishna C. Rao is an assistant professor at the Indian Institute of TechnologyMadras in


Chennai, India.
Challenge, vol. 53, no. 1, January/February 2010, pp. 109125.
2010 M.E. Sharpe, Inc. All rights reserved.
ISSN 05775132 / 2010 $9.50 + 0.00.
DOI: 10.2753/05775132530105

Challenge/JanuaryFebruary 2010 109

Rao
possessing a knowledge-based workforce that performs high-technology tasks on a par with the developed world (Emerging Economies
2006). Second, a huge workforce has been unleashed by emerging
economies, which has led to the outsourcing of many routine tasks by
OECD member countries. According to Richard Freeman, an economist at Harvard University, the global workforce had effectively doubled in 2000 due to the influx of labor from China, India, the former
Soviet Union, and other emerging economies (Freeman 2005). These
nouveau entrants also possess a growing army of knowledge-based
workers who will play an important role in the future when emerging
markets cater to their domestic demands in addition to undertaking
larger proportions of knowledge-based tasks from OECD countries.
And third, advances in information and communications technologies
have made remote execution of numerous tasks possible.
All in all, a new global economy has arrived with a global village
of workers catering to the service and technological needs of OECD
countries. The corporate sector in OECD countries has benefited
immensely from outsourcing, even though globalization has had
a negative influence, to some extent, on their domestic workforce.
OECD member countries must embrace widespread innovations to
maintain a high perch in the economic pecking order.
Many now agree that the phenomenon of global warming has been
scientifically established by the Intergovernmental Panel on Climate
Change (IPCC). This assessment has created an impetus to make innovations in greening the global economy. Thus the advent of both
globalization and climate change during this century has made innovation still more important as an engine of growth for rich economies
of the developed world.
Hence the precise measurement of innovations is useful and perhaps even imperative for assessing the well-being of rich economies
in the twenty-first century (Rao 2007). In fact, accurate quantification
of innovations could also aid in trading this intangible commodity
through suitable derivative securities (Rao et al. 2008). The thrust to
quantify innovations has culminated recently with the U.S. Department of Commerce (2008) encouraging various segments of its society

110 Challenge/JanuaryFebruary 2010

How to Measure Innovation


to foster the measurement of innovation in the economy. Efforts to
quantify this intangible attribute are also being vigorously pursued
by the European Union, Britain, and the OECD.
An example of these country-based efforts is the European Innovation Scoreboard developed for the European Union, which uses
research and development spending, availability of venture capital,
and number of students pursuing science and engineering degrees
as some of its indicators to arrive at a suitable weighed innovation
performance score (Russo 2009). However, their Summary Innovation Index (SII) attributes equal weight to all the indicators used for
arriving at the score (Ministry of Economic Affairs 2006). In contrast
to an equally weighed index such as SII, individual innovation scores
have also been proxied in the European Union from any of the following sources: surveys among citizens and various industry sectors,
intensity of research and development within various sectors, and
data in publications (ibid.). Until now, research efforts by individual
groups have focused on the quantification of innovations using financial statements as in Caibano et al. (2000); using patent data
as a proxy to measure innovation performance as in Katila (2000),
Lanjouw and Schankerman (2004), and McAleer and Slottje (2005);
and new innovation indicators for traditional industries as in Marzal
and Esparza (2007), to include a few.
This paper proposes a framework to score individual innovations
using different approachesnamely, intuitive scorecards, regressionbased models, and artificial intelligence. It also enumerates a list of
variables to score individual innovations in a wide range of ways.
Such a framework is new. Unlike the innovation of equally weighted
measures followed by the European Union, this work ascribes unequal
weights to various input variables. The goal of this work is to develop
a robust model for scoring innovations in a wide variety of fields with
practicable input variables whose values can be readily obtained. Along
with innovations protected by legal covenants granting intellectual
rights such as patents, the models described in this work also score
innovations appearing in refereed publications and other sources.
Where possible, individual innovations have been scored according

Challenge/JanuaryFebruary 2010 111

Rao
to their viability for use in the real world. The framework proposed
below attempts to deal with the many uncertainites in scoring an
intangible attribute.

Targeting Proper Innovations


In this paper, I quantify only those innovations that have been vetted
thoroughly by an official body vested with the necessary powers for
the task. An appropriate candidate is the U.S. Patent and Trademark
Office (USPTO) or its counterpart affiliated with organizations such as
the OECD. However, any country can use the modeling methodology
presented as long as their creative ideas are put through a rigorous
examination by their governments.
Such a vetting process is designed to have the USPTO-like organizations grant a legal covenanttypically, a patentfor protecting
the intellectual rights of the owner. When the vetting process does
not result in an actual patent, the ideas might still merit the status
of a new innovation for assessment purposes, albeit without the accompanying intellectual rights. Innovations falling into the latter
category are documented in, among other sources, publications by
academia, national centers and laboratories, and refereed journals
of various professional bodies. Overall, both the patented and nonpatented innovations can be run through any of the models presented
in this work for computing what we call the innovation score (IS).
The individual innovation scores, in turn, will be totaled to determine the cumulative gross domestic innovation (GDI) number for
a developed country. Such a total score will thus be the sum of all
the creative efforts unleashed in a vibrant economy. Therefore, these
numbers will not be limited to scientific disciplines but will track all
creative ideas that are given legitimacy by official bodies. The idea
in this paper is to provide not only a score but also a benchmark
to which nations will aspire, thus encouraging more investment in
such research.
Ideally, the entire process of assigning scores to innovative ideas
should be put under the auspices of an official body established for

112 Challenge/JanuaryFebruary 2010

How to Measure Innovation


conferring legal rights to intellectual property. Thus, in addition to its
primary activity, the USPTO, if it becomes the primary body to score
innovations, is mandated to score individual innovations and tally up
the individual scores to obtain a number for GDI. The sources referred
to by the USPTO may include, as noted, research efforts carried out by
academia, prominent centers, and laboratories, and refereed articles
appearing in journals published by the Institute of Electrical and Electronics Engineers (IEEE), American Society of Mechanical Engineers
(ASME), the Nature Publishing Group, the American Association for
the Advancement of Science, and other professional bodies in the
developed world. The USPTO, and its counterparts in the developed
world, should be empowered to comb through these various sources.
In doing so, however, the USPTO must avoid a double count when
an idea is documented in more than one source. Furthermore the
USPTO is in an ideal position to gather other empirical data to aid in
the development of the models for scoring innovations.
In addition to scoring innovations for various commercial and noncommercial entities, the models proposed later in this work could be
used for setting a threshold value that would decide whether an idea
passes muster as a viable innovation. This will be made possible with
the evolution of the regression-based models, as more Intellectual
Property (IP) data points are gathered in the foreseeable future. This
will aid significantly in categorizing scores.

Variables Influencing the Innovation Score


The variables that influence the innovation score are below. Some
of these have been listed in Tables 1 and 2. The empirical values for
variables in Table 1 are either readily characterized or made available
from historical databases. Table 2 lists the remaining variables whose
values are categorized into different bins. All of the variables listed
in these two tables are numeric. The techniques presented below use
some or all of these variables to develop appropriate models for the
innovation score.
It should be noted that this section attempts to bring out the important

Challenge/JanuaryFebruary 2010 113

Rao
Table 1

Variables That Can Be Characterized Empirically


Type

Characteristic

Bin
(j)

No. (i)

Variable (Xi)

1
2

X1
X2

Numeric Impact within field


Numeric Societal benefit: Impact on end user

1
1

X3

Numeric Time taken for fruition

X4

Numeric Capital consumed

X5

Numeric Workforce participation

X6

X7

Numeric Employment and learning


opportunities for society
Numeric Number of respondents confirming a
creative culture

variables needed for computing the scores of innovations in widespread


application areas. It is not necessarily complete, and the USPTO together
with the research community in general can update and supplement
this list. The subsections to follow describe the various independent
variables appearing in the sequence indicated by Tables 1 and 2.
Relative Importance of the Innovation Within Its Field (X1)
As the title suggests, variable X1 assigns a number based on the significance of the innovation in a particular discipline. This value depends on
the ranking given to the innovation by reigning experts in the respective
field. Such a body of experts could come from within the USPTO or
from other professional bodies. This variable also gauges the originality of the innovation. We measure this variable, X1, by assessing the
disruptiveness of the innovation. It accords a higher number to a new
concept as opposed to incremental progress in the correct direction.
Societal Impact (X2)
This variable is a measure of an innovations impact on society. It
quantifies the usefulness of a creative good to the society at large. In

114 Challenge/JanuaryFebruary 2010

How to Measure Innovation


Table 2

Remaining Variables Going into the Innovation Score

No. (i)
8

10

Variable
(Xi)
X8

X9

X10

Type

Characteristic

Numeric Outsourcing innovative


work
Numeric Genesis of innovation, by
entity

Numeric Genesis of innovation, by


region

Bin (j) Classification (Cj)


1

Outsourced

Not outsourced

Academia

National labs

Private Sector

Development Level
High

2
3
11

12

13

14

X11

X12

X13

X14

Numeric Genesis of innovation, by


discipline

Numeric Type of legal


documentation

Numeric Award-winning innovation

Numeric Time to covenant


expiration

Medium
Low
Development Level
High

Medium

3
1

Low
Patents

2
3
1

Publications
Other sources
Prestigious award

2
3
1

An award
No award
Suitable

Not suitable

Challenge/JanuaryFebruary 2010 115

Rao
this regard, the pharmaceutical industry, for example, would have a
direct impact on the sick populace of a nation. Another example of a
commercial activity making a direct societal impact through its innovations is the mobile-phone industry. Among its myriad benefits,
mobile telephony has influenced economic growth in ways ranging
from aiding mundane fishermen in locating the best markets for their
catch to people doing sophisticated financial transactions with their
banks. In contrast, the benevolent effects of innovations in the communications and information technology areas will be felt indirectly
through better infrastructure, improved services, and other advantages. Thus, variable X2 will assign points based on the viability of an
innovation for practical use. This effort will equate X2 to the potential
number of end users. It should be predicted from statistical regression
techniques applied to historical data sets. Many commercial setups
routinely conduct reliable forecasts of potential end users, which is
critical for their business needs. Such estimates, where availablee.g.,
from the pharmaceutical and mobile-telephony sectorsshould be
utilized because of their accuracy. In their absence, requisite regression analyses should be carried out to procure the numbers for X2 in
various application areas. (Even though X2 is an independent input
variable for the intuitive-scorecard approach described in the next
section, it is the dependent variablethe unknown one we seekin
the regression-based approaches.
Time Required for Fruition (X3)
This variable is equated to the time it takes to realize an innovative
product. It consists of the actual time consumed in transforming a
creative idea to reality and the time to secure a legal covenant, such
as a patent, for the resulting creative product. The time to secure the
legal documentation is important because an innovative idea is scored
only at the successful conclusion of its documentation process. Thus,
for products involving large cycle times, such as pharmaceuticals, X3
will be large due to longer periods required for research and clinicaltrail activities, unforeseen disruptions to the discovery process, and

116 Challenge/JanuaryFebruary 2010

How to Measure Innovation


so on. In contrast, the mobile-phone industry will use short time periods for X3 because of the ever-decreasing cycle times in new-product
development. The objective is to reward efforts that require time. To
repeat, one of the purposes of the overall measure is to encourage
research to attain a high score.
Capital Consumed (X4)
This variable is equated to the other important resourcethe capital
consumed in realizing an individual innovation. Here again, the monetary resources represented by X4 could be spilt into two portions,
accounting for the actual innovation activity and the legal documentation process. In fact, monitoring X3 and X4 would aid in observing the
efficacy of the legal documentation process in terms of the resources
consumed. Alternatively, both variables X3 and X4 could be made to
account only for the resources consumed in realizing the actual innovation. This arrangement would exclude the component arising
from the legal documentation process. In any case, variables X3 and
X4 could be overseen to study the efficient allocation of resources for
realizing the successful fruition of an innovation.
Workforce Participation (X5)
This parameter is equated to the weighed sum of the number of team
workers who have contributed to the innovation. In other words, this
work classifies workers into the categories of advanced, postgraduate,
graduate, and other degrees. In getting a value for X5, the class of
workers with advanced degrees are given the highest weight, with the
other degrees receiving the least. Moreover, the terminology chosen
in this effort describes a Ph.D. as an advanced degree, an MS or an
MBA as a postgraduate degree, a BS as a graduate degree, and anything else as other degrees. Besides team effort, this variable would
characterize the level of skill going into a candidate innovation. In so
doing, it would encourage the development of advanced skill levels
for innovative activities in newer fields of study. Such quantification

Challenge/JanuaryFebruary 2010 117

Rao
of skill levels is also crucial to kindling the fire of knowledge among
the populace and creating an awareness of innovations in mundane
and specialized fields of study.
Employment and Learning Opportunities for Society (X6)
This variable should be set up to record the employment opportunities
as well as learning opportunities created by individual innovations.
The variable X6 is the sum of two separate components. The first
component would account for the workers who potentially would
be employed, with the second component tallying those individuals
who would acquire new skills or knowledge due to an innovation. It
should be noted that this variable is not the same as X2, outlined in
a previous section, which tracks the usefulness of the end product
of an innovative idea. The creation of this variable would facilitate
efforts to tackle unemployment and some of the negative effects of
globalization by using the innovative activities of a rich economy.
Organizational Approach (X7)
This variable measures the ability of a parent organization to foster
an environment that is conducive to creative activities. Numbers for
this parameter could be obtained from employee surveys until more
accurate methods are established. Therefore this variable should be
equated to the number of respondents of the total workforce affirming the creative culture of the concerned organization.
Outsourcing of Innovative Work (X8)
In this era of globalization, companies and other institutions based
in rich nations tend to outsource some of their innovative activities
to subsidiaries in emerging economies. Consequently any nation
encouraging more innovations in-house would benefit immensely
through the jobs and knowledge created and lodged within its boundaries. Accordingly, this variable assigns a higher number to indigenous

118 Challenge/JanuaryFebruary 2010

How to Measure Innovation


creative activities, thereby providing a fillip to efforts empowering
innovative work in the rich world.
Genesis of Innovation, by Entity (X9)
This variable assigns a number based on the organizational entity that
created a given innovation. An entity could be any of the following:
a. Universities and other seats of academia
b. State-sponsored bodies such as national laboratories and
centers
c. Private sector.
Hence, depending on the distribution of innovations among these
three segments, scoring can be aimed to maintain growth among these
important seats of innovation. The academia- and government-sponsored
centers usually engage in fundamental research as opposed to the private
sector, which harnesses the fruits of applied work. Ideally, this variable
could encourage the balance of fundamental and applied research.
Genesis of Innovation, by Region (X10)
As opposed to X9s genesis-by-entity accounting, this variable should
track the contributions of various regions within a developed economy. It could be used to rate innovations from regions with lower
economic growth clips as higher as opposed to those from vibrant
regions. Such scoring can boost efforts to harmonize the contributions
from all the regions of a rich economy. Although this variable comprises an input for scoring purposes, it could be monitored to gauge
the level of innovation across the length and breadth of a country and
subsequently motivate those regions or states that are lagging in their
innovative output. This can have a salutary effect of getting entire
nations on the innovation bandwagon.
Genesis of Innovation, by Discipline (X11)
This factor would assign a higher number to an innovation that falls
in a field of study that is experiencing relatively lower development

Challenge/JanuaryFebruary 2010 119

Rao
rates. Contrasted with X9 for geographic regions, this variable could
be utilized to monitor the level of innovation across various fields of
study, which could aid efforts to encourage creative output in both
scientific and nonscientific fields of study.
Type of Legal Documentation (X12)
The inherent rigors of the patent process merit a higher number for
ideas that are patented as opposed to being documented in publications. Accordingly this effort would support awarding numbers, in
descending order, to patents, refereed publications, and other sources.
As usual, publications referred to here include refereed journal articles
and reports and dissertations printed by academia and national centers
and laboratories. This factor, while tracking creative ideas possessing
intellectual rights, could be used in endeavors for detecting the double
count of innovations.
Awards for Innovations (X13)
Some of the innovations will receive prestigious awards for their
contributions to a particular field or for addressing societal issues.
Such innovations could be boosted further in their scores through
variable X13.
Safeguarding the Legal Covenant (X14)
This variable characterizes the tendency of the inventor, be it an organization or an individual user, to block the use of the innovation by
other suitable users for a considerable duration of time. The protection
of intellectual property rights is significant to commercial entities,
be it the pharmaceutical sector with longer gestation periods for their
drugs or the aggressive innovative environment prevalent in the mobiletelephony sector. Although individual inventors should be rewarded
by the free-market system for innovations having a positive societal
impact, they should also be reined in from their urge to hoard on their
discoveries. Thus X14 would grant a higher number to innovations
whose inventors agree to release their patents in due course of time.

120 Challenge/JanuaryFebruary 2010

How to Measure Innovation


This course of time would be longer for innovations in the pharmaceutical industry than for discoveries in the field of computer science.
Moreover, this variable applies to all other forms of legal covenants
granting intellectual rights, thereby not being limited to patents.

Methodology for Quantification


Three different modeling techniques are presented in this section for
evaluating the score of an individual innovation. These approaches are
the intuitive scorecard, logistic and ordinary regression, and neural
networks. This work leaves the choice of the proper technique, or combination of techniques, up to the end user, which can be a public organization such as the USPTO or a quasi-public or private-sector entity. The
results pertaining to the efficacy of these modeling techniques when
applied to actual data on innovations from the archives of developed
economies will be discussed in a future work. The modeling framework of this paper assigns a maximum score of 1,000 to an individual
innovation. This can facilitate assigning whole numbers as scores to
individual innovations. In contrast, a scale of 0 to 100 might eventually
lead to fractional scores that are cumbersome to interpret.

Expression for GDI


The innovation score for a single innovation is denoted by IS. The
individual innovation scores, given by IS, will in turn add up, as
noted earlier, to the aggregate GDI number. The subsections to follow
describe three different approaches for computing the IS of a given
innovation.

Intuitive-Scorecard Approach
This method has been used extensively in the financial services industry to quantify the credit risk of pools of clients (see, for example,
Rudd 2000). Its popularity stems from the simplicity of setting it up
and the ability to score applications according to a standard set of
points. As far as innovations are concerned, some or all of the inde-

Challenge/JanuaryFebruary 2010 121

Rao
pendent variables described above could be collected together. Despite
its subjective nature, the intuitive scorecard approach would require
official bodies to set the proper scores for the various categories. The
development of regression-based models, described in the next subsection, could be used by such official bodies to set the reference scores,
thereby standardizing the approach and adding rigor to it
A sample intuitive scorecard for the pharmaceutical sector is shown
by Table 3. It pertains to a new drug whose discovery by a major drug
company has been protected by intellectual rights granted through a
patent conferred recently. This scorecard is based on some of the variables from Table 1, whose values, as mentioned earlier, can be readily
characterized. The sixth column of Table 3 lists the intuitive weights
assigned to the various classification bins, with the last column presenting the corresponding scores for the pharmaceutical innovation. Individual scores in the last column correspond to the value of a variable
falling in the appropriate numeric range. These numbers add up to
an IS of 550, as shown by the last row of Table 3, for our new drug. As
touched upon earlier, our new drug can obtain a maximum score of
1,000. It should be noted that the numbers appearing in Table 3 are
not real. Even the numbers of bins, ranges of variables, and intuitive
weights displayed have been created to replicate the pharmaceutical
sector as closely as possible in accordance with PhRMA (2006). This
sample scorecard has been presented to illustrate the simplicity of
the approach.

Regression Approach
The use of statistical techniques (Montgomery and Runger 2003;
Neter et al. 1993) could overcome the subjectivity of scorecards. As
opposed to intuitive scoring, statistical techniques could be applied
to innovation data archived by professional bodies like the USPTO for
a more precise estimation of IS. In particular, a statistical technique
known as regression can be utilized to estimate the relation or trend
existing between a measure like IS and a data set of past innovations.
For example, a trend between certain kinds of innovations and the

122 Challenge/JanuaryFebruary 2010

How to Measure Innovation


Table 3

Sample Intuitive Scorecard

No. (i)

Variable
(Xi)

X1

X2

Characteristic
Impact within a
field

Impact on end user

Bin
(j)

X3

X4

Time taken for


fruition

Capital consumed

X5

Number of respondents confirming a


creative culture

Score
200

> 100

200

100

100

< 100

60

> 1 billion users

200

100

0.5 billion1
billion users
< 0.5 billion
users
15 years

510 years

100

1015 years

200

> $1 billion

200

$0.5 billion$1
billion
< $0.5 billion

100

3
5

Intuitive
Weight

3
3

Variable
Range

50
50

100

100

60

1030%

40

3050%

100

> 50%

200

Total Innovation Score (IS)

50

100
550

number of end users, which is one way we measure the impact on


society (X2 in Table 1), can be estimated.

Artificial Intelligence Approach


In recent years, artificial intelligence (AI) techniques have been
brought to bear on critical problems in science, engineering, and

Challenge/JanuaryFebruary 2010 123

Rao
finance. This is possible because of the huge computational power
unleashed by advancement in computer technology. An AI technique
that has found widespread applications in pattern recognition, signal
processing, and financial market analysis is neural networks. Neuralnetwork models mimic the massive processing capabilities of the
living brain wherein a three-dimensional network of neurons carries
out complex tasks (for example, Haykin 1999).
This is a far more complex way of modeling a score of a nations
innovation. But it will enable researchers to estimate the more complex relationship among varieies of innovations. It is still in the early
stages of development. Last but not least, the regression- and AI-based
techniques require empirical data sets for the independent and dependent variables. High-quality databases are critical for building an
accurate model whose predictions are reliable. Ideally, organizations
such as the USPTO and institutions concerned with other sources of
innovations must further develop these databases.
This effort has summarized some of the concepts underlying the quantification and measure of innovations. Besides identifying innovations
that are potential candidates for scoring, this work has also described
the independent variables needed for quantifying innovations. Clearly,
there is still much work to do. But we have a solid beginning.

For Further Reading


Caibano L., M.G. Ayuso, and Snchez M.P. 2000. Shortcomings in the Measurement of Innovation: Implications for Accounting Standard Setting. Journal of
Management and Governance 4, no. 4: 31942.
Emerging Economies Climbing Back. 2006. Economist, January 19: 6970.
Freeman, R. 2005. What Really Ails Europe (and America): The Doubling of the
Global Workforce. Globalist. Available at www.theglobalist.com/StoryId.
aspx?StoryId=4542.
Haykin, S. 1999. Neural Networks. Upper Saddle River, NJ: Prentice-Hall.
IPCC. 2001. Climate Change 2001: The Scientific Basis; Intergovernmental Panel on
Climate Change. Cambridge: Cambridge University Press.
Katila, P. 2000. Using Patent Data to Measure Innovation Performance. International Journal of Business Performance Management 2: 18093.
Lanjouw, Jean O., and M. Schankerman. 2004. Patent Quality and Research Productivity: Measuring Innovation with Multiple Indicators. Economic Journal
114: 44165.

124 Challenge/JanuaryFebruary 2010

How to Measure Innovation


Marzal, J.A., and Esparza, E.T. 2007. Innovation Assessment in Traditional Industries: A Proposal of Aesthetic Innovation Indicators. Scientometrics 72, no. 1:
3357.
McAleer, M., and Slottje, D. 2005. A New Measure of Innovation: The Patent Success
Ratio. Scientometrics 63, no. 3: 42129.
Ministry of Economic Affairs. 2006. Science, Technology and Innovation in the
Netherlands. The Hague.
Montgomery, D.C., and G.C. Runger. 2003. Applied Statistics and Probability for Engineers. New York: John Wiley.
Neter, J., W. Wasserman, and G.A. Whitmore. 1993. Applied Statistics. Boston: Allyn
and Bacon.
PhRMA (Pharmaceutical Research and Manufacturers of America). 2006. Pharmaceutical Industry Profile, 2006. Washington, DC.
Rao, B. 2007. Economic Recognition of Innovation. Paper presented at the Singapore Economic Review Conference (SERC), Singapore. Available at http://
worldbenefit.case.edu/research/paperseries/?p=128.
Rao B., A. Sundararajan., and S. Rao. 2008. Trading of Innovations to Combat
Globalization and Climate Change. Challenge 51, no. 5 (SeptemberOctober):
7084.
Rudd, O.P. 2000. Data Mining Cookbook: Modeling Data for Marketing, Risk and Customer Relationship Management. Hoboken, NJ: John Wiley & Sons.
Russo, G. 2009. NatureJobs. Nature 457, no. 12: 921.
U.S. Department of Commerce. 2008. Innovation Measurement: Tracking the State of
Innovation in the American Economy: A Report to the Secretary of Commerce by the
Advisory Committee on Measuring Innovation in the 21st Century Economy,
Washington, DC. Available at www.innovationmetrics.gov.

To order reprints, call 1-800-352-2210; outside the United States, call 717-632-3535.

Challenge/JanuaryFebruary 2010 125

Copyright of Challenge (05775132) is the property of M.E. Sharpe Inc. and its content may not be copied or
emailed to multiple sites or posted to a listserv without the copyright holder's express written permission.
However, users may print, download, or email articles for individual use.

Você também pode gostar