Professional Documents
Culture Documents
America's Leaning Ivory Tower: The Measurement of and Response To Concentration of Federal Funding For Academic Research Yonghong Wu
America's Leaning Ivory Tower: The Measurement of and Response To Concentration of Federal Funding For Academic Research Yonghong Wu
https://textbookfull.com/product/cracks-in-the-ivory-tower-the-
moral-mess-of-higher-education-1st-edition-jason-brennan-philip-
magness/
https://textbookfull.com/product/periphyton-functions-and-
application-in-environmental-remediation-yonghong-wu-auth/
https://textbookfull.com/product/educational-measurement-for-
applied-researchers-theory-into-practice-margaret-wu/
https://textbookfull.com/product/the-measurement-of-hate-crimes-
in-america-frank-s-pezzella/
America s Response To China A History Of Sino American
Relations Warren I. Cohen
https://textbookfull.com/product/america-s-response-to-china-a-
history-of-sino-american-relations-warren-i-cohen/
https://textbookfull.com/product/educational-measurement-for-
applied-researchers-theory-into-practice-1st-edition-margaret-wu/
https://textbookfull.com/product/scientific-integrity-the-rules-
of-academic-research-kees-schuyt/
https://textbookfull.com/product/sme-funding-the-role-of-shadow-
banking-and-alternative-funding-options-1st-edition-gianluca-
oricchio/
https://textbookfull.com/product/english-for-academic-research-
writing-exercises-adrian-wallwork/
SPRINGER BRIEFS IN POLITICAL SCIENCE
Yonghong Wu
America’s Leaning
Ivory Tower
The Measurement
of and Response to
Concentration of
Federal Funding for
Academic Research
123
SpringerBriefs in Political Science
More information about this series at http://www.springer.com/series/8871
Yonghong Wu
123
Yonghong Wu
Department of Public Administration
University of Illinois at Chicago
Chicago, IL, USA
This Springer imprint is published by the registered company Springer Nature Switzerland AG.
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
As a scholar in the field of public budgeting and finance, I have a genuine interest in
the allocation of government resources. Government allocation decisions affect the
operation of various public, private, and not-for-profit organizations and, ultimately,
the lives of people in this country. Because elected representatives make budgetary
decisions at all levels of government, government resource allocation is basically an
outcome of the political process. However, federal resource allocation does not end
when Congress passes appropriation bills and the president signs them.
Government agencies often need to further allocate appropriated funds to other
institutions and individuals. While public budgeting and finance scholars focus on
the politics of government budgeting, the post-appropriation allocation has not been
studied enough.
A few federal agencies such as the National Science Foundation and the
National Institute of Health receive billions of appropriated funds each year to
support academic research, and they allocate the funds to higher education insti-
tutions and researchers via a competitive selection process. With my bachelor’s
degree in applied physics and doctoral training in science and technology policy, I
understand the rationales of government funding of academic research and the need
for merit-based competition in funding scientific research. However, a competition-
based allocation mechanism has led to a substantial concentration of federal funding
in high-capacity states, while low-capacity states are largely left underfunded. The
uneven distribution of federal funding for academic research has been an important
public policy issue for decades.
Under the mandate of Congress, some federal agencies launched special pro-
grams to address funding concentration issue in the 1980s and 1990s. Although the
programs have shown some modest effects on research-capacity building through
investing in collaboration development and infrastructure improvement, insufficient
attention to the institutional environment limits the progress toward a more equi-
table distribution of federal research funding across jurisdictions. I feel the obli-
gation to thoroughly investigate the issue of uneven distribution of federal resources
and recommend ways to improve the effectiveness of government efforts in this
arena.
v
vi Preface
The book is the product of my long-term focus and research on this topic.
I sincerely hope that this work will have a positive impact on policy-making and
program implementation aimed at addressing this funding concentration in the
United States.
I am indebted to several people who assisted my research on this topic. Julia
Melkers was the co-author of my first paper on EPSCoR, which inspired my
decade-long interest in the topic. Eric W. Welch generously shared materials and
data that are helpful to this work. I also want to thank the external reviewers for
their thoughtful comments.
Finally, this book is dedicated to my wife, Karen, and my children, Jerry and
Cindy. This book would not have been possible without their love and support.
vii
viii Contents
ix
x List of Figures
xi
xii List of Tables
Like the epithetical ivory tower, academic institutions in the U.S. are a hierarchical
array of public and private universities, doctoral, master’s and baccalaureate univer-
sities/colleges, universities that belong to the Ivy League versus those that do not,
with varying academic reputations and funding sources. Diverse academic institu-
tions may be necessary to meet the various needs of people and society as a whole.
However, the geographic concentration of research-intensive higher education insti-
tutions creates potential issues of inequity. California and Massachusetts have more
top-tier, prestigious academic institutions than other states, creating an uneven dis-
tribution of educational benefits and the substantial concentration of federal funding
of academic research. This book describes the funding disparity aspect of higher
education in the U.S. and assesses the effectiveness of federal programs tackling this
issue.
The main goal of this chapter is to introduce readers to the broad context of
government funding of academic research. In the American science policy arena,
there have been continuous debates on peer-review versus equity-based approaches
to the allocation of federal research funding. The peer-review system has been the
primary mechanism for distributing federal government funding for research among
universities since shortly after World War II. Peer review ensures the production of
the best science by funding the most capable researchers in the country. As a result,
federal research funding has been concentrated in “high-capacity” states where many
of the most capable researchers reside, and a large number of “low-capacity” states
have received substantially less research funding from federal agencies.
In fiscal year 2016, all higher education institutions in the U.S. spent a total of
$67.7 billion in their conduct of research and development (R&D) in science and engi-
neering (S&E). Public institutions spent $44.2 billion, about 65.4% of total academic
R&D expenditures in that year. The money spent by higher education institutions
came from different sources. As the primary sponsor of academic research, the fed-
eral government provided $37.7 billion or about 55.7% of total R&D spending within
universities in 2016 ($23.1 billion of which went to public universities). State gov-
ernments contributed $3.7 billion, about 5.5% of total academic R&D expenditures
in 2016 ($3.4 billion to universities under their control). Other sources included the
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2020 1
Y. Wu, America’s Leaning Ivory Tower, SpringerBriefs in Political Science,
https://doi.org/10.1007/978-3-030-18704-0_1
2 1 The Funding of Academic Research in the U.S.
The term federalism is used to describe a system of government in which the power to
govern is constitutionally divided between a central (federal) governing authority and
constituent political units (like states). Scholars have developed different theoretical
models (dual, cooperative, and coercive federalism) that are dominant in different
times and applicable to different policy arenas. For instance, the cooperative fed-
eralism model posits that federal and state governments interact cooperatively and
collectively to solve common problems.
Federalism in funding academic research refers to the shared funding responsi-
bility for S&E research by both federal and state governments. Federalism in science
policy has been cooperative over time. Research funding comes from several sources,
including federal and state governments. The federal government never told state gov-
ernments to support or not support any particular type of research. Funding decisions
at the state level are independent of federal government, except for the cost-sharing
requirement of federal research grants. For instance, while the Bush administration
forbade the use of federal funds for research involving the destruction or creation of
embryos, some states could still step in to advance this vital research without federal
preemptions.
Fiscal federalism is based on the idea that a public service should be financed in
such a way that the benefits are confined to the jurisdiction financing the service.
This no-spillover arrangement is supposed to achieve and maintain efficient decision-
making at different levels of government. The application of this theory in scientific
research means that federal and state governments share responsibility in proportion
to the expected scope of benefit from specific research projects. Federal government
should support research projects that benefit the entire country (like basic research)
or meet national needs (like mission-driven research). On the other hand, it is the
states’ responsibility to support research projects with outputs benefiting primarily
their jurisdictions.
1.2 Trends of Fiscal Federalism in Funding Academic Research 3
45.00
40.00
35.00
30.00
25.00
20.00
15.00
10.00
5.00
0.00
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
2012
2014
2016
Fig. 1.1 Amounts of federally and state/local financed higher education R&D expenditures. Note
The black solid and dashed lines represent the amounts of federally and state/local financed R&D
expenditures for all higher education institutions (in billions constant dollars). The gray solid and
dashed lines represent the amounts of federally and state/local financed R&D expenditures for
public higher education institutions (in billions constant dollars)
In order to understand how fiscal federalism works, I examine the pattern and trends
of academic research funding over time in terms of annual shares of funding for
academic research by federal and state governments. The data source is the National
Science Foundation (NSF)’s Survey of Research and Development Expenditures at
Universities and Colleges. The survey has been collecting data since 1972 on sepa-
rately budgeted R&D expenditures within academic institutions by source of funds,
including federal government, state/local governments, businesses, higher-education
institutions, and other sources (NSF, 2011). The data are collected from universities
directly, using consistent, uniform definitions and collection techniques.
Figures 1.1 and 1.2 show the main pattern and time trends of federally and
state/local financed higher education R&D expenditures for all or public higher edu-
cation institutions, respectively. The actual amounts (converted to constant dollars)1
are presented in Fig. 1.1 and the shares as percentage of total expenditures are in
Fig. 1.2.
The two figures show that the federal government financed increasing amounts
of academic R&D during 1972–2016, whereas state government support was fairly
stable. The gap between federally financed R&D expenditures for all and public insti-
tutions indicates that federal agencies also provide substantial amounts of financial
1I use the GDP implicit price deflator with base year 2009 as of July 2017.
4 1 The Funding of Academic Research in the U.S.
80.0%
70.0%
60.0%
50.0%
40.0%
30.0%
20.0%
10.0%
0.0%
1972
1974
1976
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2010
2012
2014
2016
Fig. 1.2 Shares of federally and state/local financed higher education R&D expenditures. Note
The black solid and dashed lines represent the shares of federally and state/local financed R&D
expenditures for all higher education institutions (in %). The gray solid and dashed lines represent
the shares of federally and state/local financed R&D expenditures for public higher education
institutions (in %)
state share of academic R&D expenditures for all higher education institutions has
decreased almost continuously from slightly above 10% in the early 1970s to about
5.5% at the end of the period. Although state/local financed R&D expenditures are
mostly taken by public institutions, the state share of academic R&D expenditures
for public institutions has declined substantially, dropping from 14% in the early
1970s to 7.7% in 2016.
In addition to the overall pattern of shared funding by federal and state governments,
it is important to examine the distribution of federal funding of academic R&D. The
pursuit of efficiency in the conduct of academic research mandates the use of peer
review in allocating federal support in this arena. However, the dominance of gov-
ernment, particularly federal government, legitimatizes the concern for inequitable
distribution of federal resources across jurisdictions.
Figure 1.3 presents each state’s share of federal R&D support to academic institu-
tions in 2015. The federal R&D support refers to the federal obligations for academic
R&D in S&E fields. It covers all direct, indirect, incidental, or related costs resulting
from or necessary to the performance of R&D by private individuals and organi-
zations under grant, contract, or cooperative agreement, as well as demonstration
projects and research equipment (NSF, 2015). The data source is the NSF’s Survey
of Federal Science and Engineering Support to Universities, Colleges, and Nonprofit
Institutions. The survey includes all academic institutions that receive funding from
federal agencies that finance federal R&D obligations to the academic sector. The
data are collected from federal agencies directly.
The disparity in federal academic R&D funding is substantial. The top ten states
received an aggregation of over 60% of total federal R&D support to academic
institutions in 2015, while the bottom ten states received less than 2% of the total
federal support in that year. The disparity in per capita terms becomes less impressive
as the top ten states account for about 48% of the total U.S. population, and only
about 4.5% reside in the bottom ten states.2 However, the disparity remains quite
substantial when I compare individual states. The state of Maryland, with about
1.9% of the U.S. population, received 6.4% of federal R&D support; the state of
West Virginia, with about 0.9% of U.S. population, received only 0.2% of federal
R&D support.
The federal pursuit of efficiency in the conduct of scientific research coupled
with uneven distribution of research capacity plays a leading role in this funding
disparity. Competition for federal S&E research funding is primarily merit-based
via peer review. The concentration of academic research in a few states is a natural
result of agglomeration of prestigious research universities in those states. Aside from
getting more or less federal research dollars, education and training opportunities for
college students in S&E fields are affected, and spillover benefits from federally
funded research projects such as patents with potential commercial prospects and
the incubation of new industries and products become similarly concentrated. Federal
dominance coupled with the uneven distribution of federal funds raises legitimate
concerns about how federal academic R&D dollars should be distributed among
individual states.
The substantially uneven distribution of federal funding for academic research
has been an important public policy issue for decades. Under the mandates of
Congress, NSF launched the first Experimental Program to Stimulate Competitive
Research (EPSCoR) to support low-capacity jurisdictions in 1979, and several
other federal agencies established similar funding programs in the 1990s. The
number of EPSCoR-eligible jurisdictions increased from 5 in 1980 to 31 in 2012.
Decades of federal EPSCoR efforts have not delivered the expected results. As
the America COMPETES Reauthorization Act of 2010 states, “National Science
Foundation funding remains highly concentrated, with 27 states and two territories,
taken together, receiving only about 10% of all NSF research funding.” The book is
intended to provide a comprehensive assessment on the effectiveness of the EPSCoR
1.3 Substantial Disparity in Federal Funding of Academic S&E Research 7
index are used to describe the extent of concentration of academic research fund-
ing from several federal agencies. In particular, several group-based concentration
indices are introduced to succinctly summarize the level of jurisdictional concentra-
tion of federal obligation for academic R&D. The group-based concentration index
has the advantage to avoid false indication of policy effect. The chapter concludes
with a brief discussion of equity implications of the uneven distribution of federal
research funding by showing that academic R&D funding is closely tied to a state’s
educational opportunities and economic growth.
Chapter 3 describes federal government response to the uneven distribution of aca-
demic research funding. The chapter briefly reviews the history of federal EPSCoR
programs, particularly NSF’s EPSCoR and National Institutes of Health (NIH)’s
Institutional Development Award (IDeA), and describes evolving policy goals and
programmatic features of the programs and capacity-building activities in higher
education institutions in the eligible states. State-level EPSCoR coordination and
heterogeneity are discussed as well. I also develop an evaluative framework of
research capacity and competiveness as a conceptual guide to the subsequent multi-
level assessment of EPSCoR effects on research capacity at the individual level and
research competitiveness at the jurisdiction level. The framework encompasses tal-
ent, collaboration, support, and motivation as four key determinants of individual
research capacity, because the ability to conduct scientific research relies not only
on the scientific and collaborative abilities of the researchers but also on their access
to necessary facilities and equipment and encouraging institutional and work envi-
ronments.
The first part of Chap. 4 is an empirical test of the evaluative framework illus-
trated in Chap. 3. Using a recent data set of a sample of academic scientists, I develop
measures of talent, collaboration, support, and motivation, and examine how these
measures affect scientists’ research capacity as demonstrated by their grant-seeking
performance. The focus on scientists’ grant-seeking performance is closely related
to the primary goal of EPSCoR in the pursuit of a more equitable distribution of
federal research funding. After the evaluative framework is empirically validated, I
conduct an assessment of EPSCoR efforts in building scientists’ research capacity
in the eligible jurisdictions by comparing the mean values of the four key determi-
nants of individual research capacity between scientists in EPSCoR states and those
in other states. The analysis of variance results suggest that individual scientists in
EPSCoR states do not show significant weakness in research talent, collaboration,
and motivation, and they seem to perform equally well in grant-seeking as their
counterparts in non-EPSCoR states. But the results also reveal important frustra-
tions among scientists in EPSCoR states that EPSCoR initiatives might address and
mitigate.
Chapter 5 provides a comprehensive and updated assessment on the effectiveness
of the EPSCoR beyond the individual level. It begins with a descriptive analysis
of the mobility of the top 100 academic institutions in receipt of federal R&D
support from 1975 to 2015. This institution-level analysis reveals the dominance of
1.3 Substantial Disparity in Federal Funding of Academic S&E Research 9
non-EPSCoR institutions among the top competitors for federal funding of academic
R&D and a modest gain by academic institutions in EPSCoR states. It is followed
by a macro-level assessment showing that the two largest EPSCoR programs—NSF
EPSCoR and NIH IDeA—have been effective in reducing the concentration index
of the respective agency support to academic research, but the magnitude of the
effects is small. Two additional state-level assessments present quite modest effects
of NSF EPSCoR and NIH IDeA on a state’s shares of NSF and NIH obligations for
academic R&D, respectively. In consideration of the heterogeneity of state EPSCoR
programs, supplemental analysis is also performed on the share of NSF or NIH
funding of academic R&D for each state to identify the varying effects of EPSCoR
across the eligible states. These assessments are complementary to each other, and
collectively provide solid empirical evidence on the effects of EPSCoR on various
measures of research capacity and competiveness.
Chapter 6 focuses on the construction and renovation of research infrastructure in
higher education institutions. Research infrastructure is a critical pillar of academic
research capacity and has been a primary focus of EPSCoR since the early 2000s. I
first develop a measure of research density by comparing the R&D expenditures made
by academic institutions within a state with the size of its academic research facili-
ties. The analysis shows that EPSCoR states have a larger size of research facilities
relative to their academic R&D expenditures than non-EPSCoR states, indicating
that EPSCoR institutions have likely over-invested resources in physical research
infrastructure and do not utilize research facilities as efficiently as their counterparts
in non-EPSCoR states. The chapter also demonstrates that state governments have
been playing a more important role than federal government in funding of research
facilities. The empirical evidence furthermore shows that EPSCoR state governments
do not invest significantly more funds in research facilities than non-EPSCoR states.
I conclude in Chap. 7 with a synthesis of the analyses and a discussion of the impli-
cations for the future of EPSCoR and similar efforts to address the concentration of
federal funding for academic research. Although EPSCoR efforts have been effective
in building scientists’ research capacity, the limited effects at institutional, state and
national levels indicate the need for program improvement. Our empirical evidence
suggests that scientists are significantly dissatisfied with institutional environments
in the EPSCoR states, and this may limit the progress toward a more equitable distri-
bution of federal research funding at the institution or state level. I also find evidence
of redundancy and inefficiency in the construction and utilization of research facil-
ities in the EPSCoR states. The book therefore calls for a shift in EPSCoR strategy
from research collaboration and infrastructure to innovating and improving institu-
tional environments that help recruitment, retention, and motivation of S&E research
talents. The chapter also describes evolving political support for EPSCoR, and makes
additional recommendations to improve EPSCoR’s effectiveness.
10 1 The Funding of Academic Research in the U.S.
References
Bush, V. (1945). Science, the endless frontier: A report to the President. Washington, DC: U.S.
Government Printing Office.
Feller, I. (1997). Federal and state government roles in science and technology. Economic Devel-
opment Quarterly, 11(4), 283–295.
National Academies. (2013). The experimental program to stimulate competitive research. Wash-
ington, DC: The National Academies Press.
National Science Foundation. (2011). Academic research and development expenditures: Fiscal
year 2009 (NSF 11-313). Arlington, VA: National Science Foundation.
National Science Foundation. (2015). Federal science and engineering support to universities,
colleges, and nonprofit institutions: FY 2013 (NSF 15-327). Arlington, VA: National Science
Foundation.
Zuckerman, B. L., et al., (2015). Evaluation of the National Science Foundation’s Experimental
Program to Stimulate Competitive Research (EPSCoR): Final report. IDA Paper P-522. Science
and Technology Policy Institute.
Chapter 2
Geographical Concentration of Funding
of Academic Research
Building upon the context outlined in Chap. 1, this chapter illustrates the concentra-
tion of federal funding for academic research in the U.S. It also explores the under-
lying causes for and likely consequences of the uneven geographical distribution of
federal funding of R&D in the higher education sector.
This uneven pattern of distribution has persisted for a long time. Similar to Fig. 1.3,
Fig. 2.1 shows that the shares of federal R&D funding received by academic institu-
tions in the 50 states differed substantially in 1975. Together, the figures show that
federal support of academic R&D is concentrated in a few states, and a large number
of the 50 states have received minimal proportions. The top ten and the bottom ten
states largely overlap, even with 40 years between the data in the figures. California,
New York, Maryland, Pennsylvania, Massachusetts, Texas, Illinois, and Michigan
are among the top ten states in both 1975 and 2015. Washington and Ohio only appear
in the top-ten list of 1975, whereas North Carolina and Georgia make the list in 2015.
Although four states get in and out of the top-ten list over the period 1975–2015, the
change is not dramatic. For instance, Washington and Ohio dropped from 8th and
10th in 1975 to 13th and 11th in 2015. North Carolina and Georgia moved up from
12th and 19th in 1975 to 7th and 10th in 2015. In other words, the top winners of
federal academic R&D funding are virtually the same in 1975 and 2015.
Conversely, North Dakota, Nevada, Arkansas, Idaho, South Dakota, West Vir-
ginia, Wyoming, and Maine are in the bottom ten states in both 1975 and 2015.
Montana and Delaware were among the bottom ten states in 1975 (43rd and 44th)
and moved up to 39th and 40th in 2015. Alaska and Vermont, on the other hand,
were 35th and 38th in 1975, but dropped to 41st and 42nd in 2015. It seems that
the winners and losers in receiving federal academic R&D dollars in the U.S. have
stayed essentially the same for 40 years.
We can use one graph to show the disparity of federal funding of academic R&D
within a particular year, but the 45 figures for the years 1971–2015 are too complex to
reveal the pattern of change from year to year. To present a clearer picture of funding
over time, I develop some common descriptive statistics such as minimum, maxi-
mum, mean and standard deviation of state’s share of federal funding for academic
R&D in each year. Table 2.1 presents the statistics for the share of federal academic
R&D support by year from 1971 to 2015. Because NSF has been an important spon-
sor of academic research, I also present the statistics for the share of NSF academic
R&D support by year in Table 2.2.
2.1 Measurement of Concentration of Federal Funding … 13
Table 2.1 Descriptive statistics for state’s share of federal academic R&D support by year
Year Number of Mean Standard Minimum (%) Maximum (%)
states (%) deviation (%)
1971 50 1.97 2.92 0.09 14.69
1972 50 1.97 2.91 0.09 14.57
1973 50 1.97 2.94 0.08 14.92
1974 50 1.97 2.86 0.10 15.47
1975 50 1.97 2.84 0.11 14.98
1976 50 1.97 2.84 0.12 15.47
1977 50 1.97 2.78 0.10 14.85
1978 50 1.97 2.74 0.11 13.90
1979 50 1.98 2.78 0.10 14.27
1980 50 1.98 2.77 0.11 14.13
1981 50 1.98 2.81 0.11 13.65
1982 50 1.98 2.78 0.11 13.76
1983 50 1.97 2.85 0.08 14.37
1984 50 1.97 2.82 0.09 14.51
1985 50 1.97 2.81 0.08 14.47
1986 50 1.97 2.81 0.08 14.70
1987 50 1.97 2.78 0.07 14.54
1988 50 1.97 2.76 0.06 14.31
1989 50 1.97 2.74 0.08 14.69
1990 50 1.97 2.71 0.06 14.52
1991 50 1.97 2.66 0.07 14.32
1992 50 1.97 2.68 0.07 14.71
1993 50 1.97 2.60 0.09 13.87
1994 50 1.97 2.63 0.09 14.27
1995 50 1.96 2.60 0.08 14.28
1996 50 1.96 2.63 0.07 14.58
1997 50 1.96 2.60 0.10 14.37
1998 50 1.97 2.64 0.06 14.76
1999 50 1.96 2.61 0.06 14.42
2000 50 1.97 2.59 0.12 14.53
2001 50 1.97 2.53 0.12 13.88
2002 50 1.96 2.53 0.09 13.89
2003 50 1.97 2.53 0.10 13.98
2004 50 1.98 2.59 0.10 14.55
2005 50 1.98 2.56 0.09 14.38
(continued)
14 2 Geographical Concentration of Funding of Academic Research
Table 2.2 Descriptive statistics for state’s share of NSF academic R&D support by year
Year Number of Mean Standard Minimum (%) Maximum (%)
states (%) deviation (%)
1971 50 1.99 3.25 0.02 17.89
1972 50 1.99 3.49 0.01 19.05
1973 50 1.99 3.50 0.02 19.90
1974 50 1.99 3.53 0.03 20.25
1975 50 1.99 3.37 0.03 18.04
1976 50 1.99 3.58 0.03 20.84
1977 50 1.99 3.46 0.05 19.56
1978 50 1.99 3.49 0.04 19.68
1979 50 1.99 3.46 0.06 19.41
1980 50 1.99 3.46 0.04 19.50
1981 50 1.99 3.36 0.04 18.93
1982 50 1.99 3.43 0.04 18.81
1983 50 1.99 3.31 0.03 17.85
1984 50 1.99 3.23 0.04 16.82
1985 50 1.99 3.23 0.06 16.67
1986 50 1.99 3.19 0.04 16.94
1987 50 1.99 3.23 0.05 17.62
1988 50 1.99 3.12 0.07 16.47
1989 50 1.98 3.00 0.04 16.01
1990 50 1.98 2.92 0.09 15.27
1991 50 1.97 2.83 0.10 15.02
(continued)
2.1 Measurement of Concentration of Federal Funding … 15
Table 2.1 shows that the average values of state’s shares of federal academic
research funding are fairly stable over time, fluctuating within a quite narrow range
from 1.96 to 1.98%. Given the nearly constant mean values, we can focus on the
standard deviation as a reasonable measure of dispersion of the share measure of
federal funding of academic research. The trend of standard deviation is clearly
downward, even though the decrements are modest. It reached the maximum value of
2.94% in 1973, and almost continuously declined to 2.46% in 2009, and rose slightly
afterwards. As the overall disparity in state’s share of federal funding declined, the
concentration of federal support of academic research has been tapering over time,
at quite a slow pace.
Table 2.2 shows a similar decline of dispersion in state’s share of NSF funding
for academic research. The standard deviation was at its peak in 1976 (3.58%) and
16 2 Geographical Concentration of Funding of Academic Research
dropped to 2.43% in 2009, then rose briefly and declined again to 2.41% after 2012.
Similar to total federal support of academic R&D, the mean shares of NSF support are
fairly constant in this period with the exception of some minor sliding in 2009–2014.
The combination of mean and standard deviation is helpful to describe the dis-
tribution of an important variable. The standard deviation is a common measure
of overall deviations of individual observations from the mean. A smaller standard
deviation indicates that the individual observations converge to the mean. However,
it is not particularly useful in describing the degree of geographic concentration of
federal funding because the computation of standard deviation relies on the mean of
the data. In other words, the mean should be constant for standard deviation to be a
compatible measure of dispersion over time.
I introduce an alternate measure that is not dependent upon the mean value in
each year. Rather than calculating the sum of squared deviations from the mean (the
formula of standard deviation), I simply square each state’s share of federal academic
research funding and sum them up. In one year, 50 squared shares are aggregated to
get what I would call the concentration index in that year.
The concentration index has several desirable features. First, its range is from
1/50 (equal distribution of federal funding) to 1 (maximum concentration of federal
funding). The index takes the value 1/50 only when all 50 states get equal shares of
federal academic research funding. In the scenario of maximum concentration, all
federal funding goes to one state and the other 49 states receive nothing. Second,
when the index moves from the minimum (1/50) to the maximum (1), the degree
of concentration escalates. In other words, the index values close to 1 mean high
levels of concentration, whereas the values close to 1/50 represent low levels of
concentration.
The concentration index provides an opportunity to concisely summarize with
a single numeric value the degree of geographical concentration of federal support
of academic research. This index can be used as a thermostat for policy response,
indicating when government should take action as the level of concentration goes
beyond a certain benchmark. It also makes it easier to keep track of change of
concentration of federal funding from year to year. The index could be used in policy
assessment and program evolution because a significant decrement is likely a sign
of success for policy initiatives aimed at reducing geographic concentration.
I calculate the concentration index using the state shares of federal support of
academic R&D for years from 1971 to 2015.1 I also calculate the concentration
index for support provided by major federal sponsors of academic research such as
Department of Defense (DOD), Department of Energy (DOE), NIH and NSF. The
details are presented in Table 2.3.
1I use the total federal support of academic R&D in 50 states rather than in the entire U.S. as the
denominator in the calculation of a state’s share of federal support. The two share measures only
differ slightly. I make this choice to ensure that the concentration index is exactly 1 in the scenario
of maximum degree of concentration.
Another random document with
no related content on Scribd:
PLEASE READ THIS BEFORE YOU DISTRIBUTE OR USE THIS WORK
1.D. The copyright laws of the place where you are located also
govern what you can do with this work. Copyright laws in most
countries are in a constant state of change. If you are outside
the United States, check the laws of your country in addition to
the terms of this agreement before downloading, copying,
displaying, performing, distributing or creating derivative works
based on this work or any other Project Gutenberg™ work. The
Foundation makes no representations concerning the copyright
status of any work in any country other than the United States.
1.E.6. You may convert to and distribute this work in any binary,
compressed, marked up, nonproprietary or proprietary form,
including any word processing or hypertext form. However, if
you provide access to or distribute copies of a Project
Gutenberg™ work in a format other than “Plain Vanilla ASCII” or
other format used in the official version posted on the official
Project Gutenberg™ website (www.gutenberg.org), you must, at
no additional cost, fee or expense to the user, provide a copy, a
means of exporting a copy, or a means of obtaining a copy upon
request, of the work in its original “Plain Vanilla ASCII” or other
form. Any alternate format must include the full Project
Gutenberg™ License as specified in paragraph 1.E.1.
• You pay a royalty fee of 20% of the gross profits you derive from
the use of Project Gutenberg™ works calculated using the
method you already use to calculate your applicable taxes. The
fee is owed to the owner of the Project Gutenberg™ trademark,
but he has agreed to donate royalties under this paragraph to
the Project Gutenberg Literary Archive Foundation. Royalty
payments must be paid within 60 days following each date on
which you prepare (or are legally required to prepare) your
periodic tax returns. Royalty payments should be clearly marked
as such and sent to the Project Gutenberg Literary Archive
Foundation at the address specified in Section 4, “Information
about donations to the Project Gutenberg Literary Archive
Foundation.”
• You comply with all other terms of this agreement for free
distribution of Project Gutenberg™ works.
1.F.
Most people start at our website which has the main PG search
facility: www.gutenberg.org.