Social Networks 64 (2021) 148–157
Available online 30 September 2020
0378-8733/© 2020 Published by Elsevier B.V.
Hubs and Authorities in the Koch Brothers Network
Patrick Doreian a,b,
*, Andrej Mrvar a
a
University of Ljubljana, Slovenia
b
University of Pittsburgh, United States
A R T I C L E I N F O
Keywords:
Directed networks
hubs and authorities
Koch Brothers network
Roles in networks
A B S T R A C T
The empirical focus is centered on the large social movement network created by the Koch Brothers to further
their aims of transforming the US. The network was obtained by using the VOSON web crawler given a starting
set of known allies of the Koch Brothers. This produced a large directed network with links between units. We
propose using the idea of hubs and authorities as another way of considering the roles played by the units within
this network. These roles may be more complex than has been realized. We have included an analysis of the core
interests for the members of this network.
This manuscript is constructed as follows. Section 1 introduces the
study of large network of allies assembled to promote the libertarian
ideas of the KBs to have the US government run to serve their primary
interests plus other interests relevant for their allies. Section 2 provides a
description of this network and the way it was assembled as a potent
force in American politics. A discussion of the ideas of hubs and authorities
(Kleinberg, 1999) are presented in Section 3 along with arguments
as to why we think these ideas are more useful, at least for our
purposes here, than traditional ideas regarding centrality, long a staple
approach for network analysists. We describe the main authorities and
the roles they play in spreading the most important information through
the KB network, especially their most important allies in Section 5. We
note that the notion of roles, as studied within the social network
tradition, is usually associated with blockmodeling. However, we think
that the idea of hubs and authorities can be very helpful for understanding
roles, albeit in a very different fashion.
1. Introduction
As has been noted before (see, for example, Doreian and Mrvar,
2020), the inspiration for a far-reaching project to construct the structure
of the Koch Brothers (KBs) network of allies found its inspiration in two
books. One is Democracy in Chains (MacLean, 2017) containing a
detailed history of the historical intellectual foundations of a substantive
approach dovetailing perfectly with the libertarian views of the KBs. The
history has a long pedigree going back to a mere two decades after the
founding of the US. The views of John Calhoun, seventh vice president
(1825-1832) of the US who defended slavery and promoted state’s
rights, remain alive, at least in the minds of the KBs, especially state’s
rights, and the leaders of their allied units. An economist, James
Buchanan, following this tradition, developed a theoretical economic
and political approach called ‘public choice theory’. Its core concern
centered on how incentives affected governmental action. For this work,
devoid of any genuine empirical data, he received a Nobel Economics
Prize. It was, instead, an openly political statement. Charles Koch found
Buchanan’s arguments fully consistent and most useful for achieving his
“unrealized dream of liberty, of a capitalism all but free of governmental
interference” (MacLean, 2017: xxiv). Buchanan spent many years
working at George Mason University (GMU) which has received huge
amounts of money from the KBs (especially during 2005-17). Its Foundation
received close to $86 million. Following closely behind, its
Institute for Humane Studies received almost $35 million and the Mercatus
Center receiving a seemingly paltry amount $9 million from the
KBs Greenpeace (2014). All of these contributions had a huge impact in
promoting and extolling libertarian ideas across multiple venues.
2. Studies of the Koch Brothers network
Dark Money (Mayer, 2017) lays out in great detail the ways in which
many billionaires, including the KBs, drove the rise of the radical right in
* Corresponding author at: University of Ljubljana, Slovenia.
E-mail address: pitpat@pitt.edu (P. Doreian).
Contents lists available at ScienceDirect
Social Networks
journal homepage: www.elsevier.com/locate/socnet
https://doi.org/10.1016/j.socnet.2020.07.010
Social Networks 64 (2021) 148–157
149
the US and its takeover of the Republican party, especially with the
current representation of the GOP in Congress (see Krugman, 2019).
This is not a new phenomenon. For four decades, the KBs have funded
the development of a powerful political movement (Mayer, 2017, xviii)
aimed at achieving their goal to operate freely without having to deal
with intrusive governmental constraints. See also Skocpol and
Hertel-Fernandez (2016) who studied this network.
Our large KB network was constructed as follows. Both Democracy in
Chains and Dark Money provided lists of units known to be among the
KB’s active allies. We formed a combined list of these organizations. For
each, its URL was located. All of these URLs were combined to form an
initial list of organizational URLs. The program VOSON, developed by
Ackland (see 2018) is designed to, among other things, identify links
between websites in a directed network. The initial list of URLs was
submitted1
to VOSON to identify web links involving the units identified
in Democracy in Chains and Dark Money. There were 176 such URLs2
submitted to VOSON. The result of the search was a network with 17,
212 units. Later we had to ‘clean’ this network by removing information
streams with widely varying content to focus only on the units in the
network of KB allies. The result was a network of 1081 units known as
KB allies. All of the data analyses that follow were completed using Pajek
(Batagelj and Mrvar, 1998). A fuller description of our methodology is
provided in Doreian and Mrvar (2020).
The websites for every identified unit were visited repeatedly to
extract textual material regarding their identity, ambitions, the organizations
they support and some of their many actions taken proudly to
promote the libertarian and conservative interests. These were recorded
in six large multiple background Word files used to categorize the
identified units in terms of their core interests and to extract keywords
identifying them. For the first task, the following categories were
created: i) Koch Industries (KI) units; ii) KB allies or likely KB allies; iii)
units likely to be useful for the KBs but not known as KB allies; iv) units
opposed to KB or were likely to be KB enemies – many of which were
committed environmental protection groups; v) other units having interests
apparently inconsistent with KB interests; vi) national US news
media units; vii) local or regional US news media units; viii) news media
units in other countries; ix) other non-US units (a reminder that the
reach of the KBs is a long international one); x) seemingly neutral units
but not media; xi) potentially irrelevant units (that may have survived
our efforts to winnow and remove them) and xii) US governmental units.
The largest three categories are: KB allies (having 466 units); Koch opponents
(163 of them) and 63 Koch Industry (KI) units. This last group is
a primary, but not the only, source of the wealth supporting the activities
of the KB network of allies.
Visiting the KI websites was remarkably frustrating but not at all
surprising given the secretive nature of the KBs regarding many of their
activities. This was especially the case regarding financial information.
Fig. 1 shows a network image of Koch Industries. While being very
secretive internally on their websites, they do have many links between
their websites and other units to other units to which they are linked.
These were obtained by using the VOSON web crawler. Of course, these
are ‘indirect’ links and can only serve as proxies for the real connections
involving KI units. But given their complete secretiveness, this is the best
data we can have.
It is not surprising that KI has the most links. Included in this figure
Fig. 1. The Network of units belong to Koch Industries directly or through purchasing or licensing agreements.
1
The VOSON analysis was performed by Rob Ackland to generate a Pajek
network file which he shared with us. We appreciate greatly his generosity in
doing this.
2
This is slightly less than the number of URLs in the above assembled list as
some identified units were defunct at the time of our search. Also, there were
some whose URLs could not be located. More consequentially, perhaps, some
required visitors to log in before getting access. However, after reading much of
the bile in so many of the websites of KB allies, we opted to not log into them.
P. Doreian and A. Mrvar
Social Networks 64 (2021) 148–157
150
are units providing funding to other members within the Koch Industries
umbrella. Other production units have been brought into this fold
through mergers, acquisitions and licensing agreements. Examining the
KI websites provides no mention of their polluting actions and their
creation of unsafe workplaces. Leonard (2019a) noted that David Koch
recognized very early that dealing with environmental cleanup issues
greatly threatened the profits of KI. Leonard (2019b: 187-190) goes
much further in documenting the repeated accumulation of fines by
Koch Industries for environmental pollution. Mayer (2017: 338) reported
that in 2012, KI was the top producer of toxic waste. She (2017:
155), quoted a representative of The Center for Public Integrity (a unit in
the set of KB enemies identified in our network), that the KI’s “pattern of
pollution was striking not just for its egregiousness but also for its
willfulness.” Layzer (2012) documented the opposition of conservatives,
in recent times, to environmental regulation of all sorts.
While issues regarding climate change may be viewed as less central
overall, they are among, if not the, core motivating issues for the KBs.
This has been noted extensively by Dunlap et al. (2016), Elsasser and
Dunlap (2013), Carroll et al. (2018), Brulle et al. (2016), Supran and
Oreskes (2017) and Farrall (2016). Another line of thought is relevant
featuring Oreskes and Conway (2010) in their book, Merchants of Deceit.
See also Begley (2007) and Bardon (2019). Among other things, deceit
involves obstructing the truth, especially about science. One prominent
target of the climate change deniers is the Intergovernmental Panel on
Climate Change (2001). See also National Research Council (2001),
McCright and Dunlap (2003) and Brownlee (2005).
3. Hubs and Authorities
Given the immense profusion of software designed for analyzing
networks in recent decades, network analysts have a huge variety of
tools to employ in their research designed to help promote our understanding
of both the structure of networks and the processes generating
them. While being fully aware of the prominence of using centrality
measures (Freeman 1979), we wonder if they have been overused,
especially when researchers compute all possible centrality measures
before inductively choosing those best serving their narrative interests.
See also Hummon et al. (1990) for a general view of this early literature.
They pointed to a separation of substantive issues and experimental
pursuits that inspired the creation of centrality measures. In essence,
creating and computing centrality measures became an end in its itself.
Here, we focus instead on the computer algorithm, HITS, developed by
Kleinberg (1999), one designed for ranking Web pages linked in a
directed network – exactly what the VOSON search was used for.
It seems that the terms hubs and authorities, while defined precisely
by Kleinberg, now have multiple meanings. This is not unreasonable.
Soldano et al. (2017) present a method for identifying hub-authority
cores in directed networks in which they search for dense cores in the
form of having vertices that are highly connected. This is an important
problem, one fully worthy of further attention. The network we study
here is directed, and as they put it, the vertices have attributes in the
form of identified units. But our concern here is not to identify densely
connected communities of units as we want to identify units as either
hubs or authorities in the sense of Kleinberg (1999).
Hubs are vertices pointing to other vertices thought to be important
in the operation of directed networks. Authorities are these vertices. Put
differently, good hubs point to many authorities while a good authority
are vertices pointed to by good hubs. While this appears to be rather
circular in its execution, the algorithm leads to useful measures of the
extent to which vertices are hubs and authorities. More importantly,
they point to the roles played by units in any network that, most likely, is
obscured when focusing solely on measures of centrality.
As noted above, roles have been seen as being identified within the
blockmodeling tradition. They have been identified, primarily, through
the links that vertices have to other vertices. In this sense, this is strictly
structural. A second sense, but integrally connected to the first sense, is
that this is also culturally defined, especially in well-defined contexts.
One clear example is school systems with administrators, teachers and
students who have clearly identified roles. We have absolutely no
quarrel with this line of thinking. But in the context of the KB network
considered here, it seems abundantly clear that there are units within KB
network of allies that produce documents and statements designed to
persuade allies in joining them while condemning opponents. They
clearly have roles role. Identifying them as authorities is critically
important. But so too, is identifying hubs whose role, in this network, is
to direct others to the important authorities. These roles are complementary
and need to be examined separately. We acknowledge that this
is expanding the notion of ‘roles’ but that this is a useful extension.
In Kleinberg’s (1999) approach, each vertex, i, has two non-negative
weights assigned to them. Let xi be the authority weight for vertex i and
let yi be the hub weight for i. HITS starts with each i having arbitrarily
assigned nonzero values. Let A denote the adjacency matrix of the
directed network and AT
its transpose. The two sets of weights are
updated by successive iterations. Each xk is updated by the operation
(AT
A) x(k-1)
. Similarly, yk is updated by (AAT
) y(k-1)
where k denotes
successive iterations.
This introduces another consideration as whether the approach we
have adopted is preferable to computing the eigenvalues of AT
A or AAT
.
We use an iterative algorithm (updating) for computing hubs and authorities
since it is much faster than computing these eigenvectors. It
works for large networks as well. But it is known that iterative algorithm
gives accurate results already with low number of iterations. In case of
this Koch Brothers network, only 17 iterations are needed for 6 valid
decimal places of Hubs and Authorities scores.
Fig. 2 shows a box plot of the authority scores for units in the network
identified by the VOSON web crawler. Clearly, this distribution is highly
skewed. Without surprise, some authorities are far more important than
others – this is to be expected as the money flows supporting this
network vary greatly. The units deemed more important for promoting
‘the messages’ receive more resources. Exactly the same holds for the
Fig. 2. A box plot for the authority scores for the KB allies.
P. Doreian and A. Mrvar
Social Networks 64 (2021) 148–157
151
box plot of hub scores. Our primary attention is focused on the most
important authorities while also considering the important hubs in this
directed network. While we acknowledge that identifying communities
as densely connected subgraphs is an important problem, this is not our
focus here as it features roles, as modified by our above description,
rather than communities of authorities that do not identify roles.
4. Hubs and authorities in the Koch Brothers network
We focus our narrative attention in this section on some of the top ten
authorities as identified by the HITS algorithm as implemented in Pajek
(Batagelj and Mrvar, 1998). However, for reasons of space we consider
only a few them here.
The top ten authorities are the Cato Institute (a libertarian think tank
headquartered in Washington, D.C) and funded heavily by the Koch
brothers. Indeed, Charles Koch turned his private foundation into the
Cato Institute. It supports total economic freedom for the likes of the
Koch brothers; the Reason Foundation which has also received funds
from by the KBs. It promotes outsourcing public services to private
corporations within a program advocating shrinkage of the national
government (MacLean, 2017: 143); the Heritage Foundation, founded
by Joseph Coors, stressing, among other things, solutions based solely on
the principles of free enterprise, limited government, individual
freedom, traditional American values and a strong national defense.
Some of its funding comes from Koch Family Foundations; the Foundation
for Economic Freedom (FEE) seeking to ‘spare’ private enterprises
from ‘governmental intrusion’; the Independent Institute
(following the generic von Mises approach3
for protecting private enterprise
and property rights); the Heartland Institute, founded in 1973, is
an organization promoting climate change denial, along with obstructing
all efforts for protecting the environment. It was founded in 1984
and is a part of the second wave of conservative mobilization (Skocpol
and Hertel-Fernandez, 2016); the Mercatus Center (also a heavily funded
unit by the KBs at George Mason University); the Foundation for
Individual Rights in Education (FIRE) devoted to expunging the alleged
liberal influence in academia.
Fig. 3 shows the links to the Cato Institute as an authority. The
structure is quite simple. Every unit shown on the right is a known ally of
the KBs. All point to the Cato Institute as an authority. However, there
are three reciprocated links, shown in solid black lines, which could be
viewed as unusual. But all of them are programs within the Cato Institute.
There is another link to Human Progress from the Cato institute.
However, it is another one if its projects. We have noted that many of the
larger units in the KB network are umbrella outfits containing multiple
units with coordinated agendas supporting the broader libertarian
agenda of the KBs. Thinking of them as separate ‘independent’ units is
nonsensical. They are joined in an extensive network of KB allies.
The second authority is the Mercatus Center shown in Fig. 4. As
Fig. 3. The Cato Institute as an authority.
3
From https://mises.org/profile/ludwig-von-mises, a clear statement of his
approach is made evident. He claimed that “the expansion of free markets, the
division of labor, and private capital investment is the only possible path to the
prosperity and flourishing of the human race.” It is not surprising that this
perspective fits well to interests of the KBs and many conservatives.
P. Doreian and A. Mrvar
Social Networks 64 (2021) 148–157
152
noted above, it is a core unit at the George Mason University where two
programs have been, in essence, taken over by the Koch Brothers
through their massive financial investments in them. One is the Antonin
Scalia Law School with the other being its Economics Department. They
each contain many programs devoted to promoting the libertarian
agenda of their funders. We note that most of the units listed in the right
do belong the KB network – but not all. Most of the 24 citations from the
Mercatus Center, something thought to be unusual for authorities, are to
units fully involved in the KB network of allies. But others are not. We
think that the apparent divide between hubs and authorities may be far
more complex than has been realized. We speculate that academics in
the two units funded heavily by the KBs are alert to publications in the
general literature and will seek to rebut them in economic and legal
outlets sympathetic to their views.
The third authority we display is the Reason Foundation as shown in
Fig. 5. Again, many hubs point to it. As we understand them, hubs and
authorities are thought to play very different roles in directed networks. If
so, this viewpoint may need to be expanded to reflect a more complex
view of reality given that the Reason Foundation, also funded very
heavily by the Koch brothers, is also a very prominent hub. This suggests
this unit is more structurally important for promoting the KB interests as it
plays two roles. It provides useful ‘knowledge’ for many KB allies while
pointing to other authorities having value in this network.
Even so, it is more important to consider authorities in conjunction,
Fig. 4. The Mercatus Center as an authority.
P. Doreian and A. Mrvar
Social Networks 64 (2021) 148–157
153
given their huge potential for coordinated influence on other actors in
this network rather than view them as separate entities. Fig. 6 displays
part of the KB network for which the top ten authorities are placed in a
small central circle. Without surprise, there are links between them
given their shared interests and coordinated actions. There are many
links between these authorities and their allies. While this figure is not
that large, it is merely one part of a much larger network. Even so, it
shows the reach and coordination of a network seeking to control the US
economy and political discourse to serve the interests of the KBs.
Without surprise, the network structure of the hubs in the KB
network is very similar to those of the authorities. There is little value in
reporting them here even though its distribution is slightly less skewed.
However, as with authorities, it is important to consider them in
conjunction. Fig. 7 provides the relevant image. The units with the top
hubs scores are: The Reason Foundation (described above); its TV unit
devoted to catering to followers of its vision with core messages being
displayed in visual formats while attacking enemies; Wichita Liberty (a
known KB ally); ARRA News Service (a very far-right news venue
devoted to promoting far-right views and attacking many perceived
opponents); the State Policy Network4
(a non-profit outfit., which is an
umbrella organization supporting the conservative and liberation
agendas at the state level Mayer (2017: 272). It promotes laws constructed
in conjunction with the American Legislative Exchange Coun­
cil5
(McLean (2017: 210)) to promote copy-cat laws obstructing
democratic laws at the state level. The John William Pope Foundation
(funding many right-wing causes in North Carolina); Forbes Magazine
discussing free-market issues; The Globe (a Canadian news outlet); News
Alert (promoting, among other things, libertarianism, property rights,
the struggle against the ever-expanding 14th Amendment, governmental
corruption, the struggle against socialism and MAGA; and the
Free Republic (another very far-right venue). In the main, these strong
hubs point to authorities promoting the KB agenda. We acknowledge the
arguments of Soldano et al. (2017) regarding the study of combined hub
communities and authority communities. But these are separate issues.
At some point, we hope that they will be combined.
5. The core interests of the KB allies
In a separate study, we constructed a list of important keywords
describing their interests in detail. The list of keywords in alphabetical
order is: against political correctness, American heritage and traditional
values, anti-affirmative action, racism, anti-feminist and hostility to
women, anti-immigrant, anti-labor action and opposing minimum wage
laws, arbitration enforcement, attacking Social Security and the social
safety net, attacking liberal bias in academia, blocking Obamacare and
subsequent healthcare reform, blocking environmental regulation,
Christian faith, climate change denial, conservative, criminal justice
reform, economic growth, evangelical, free markets, grass roots, gun
rights, heritage, Islamophobia, jerrymandering and redistricting, less
governmental regulation, libertarian, liberty, limited government,
mainstream media bias, monitoring and restricting voting, obstructing
consumer protection, opposing abortion, opposing campaign finance
Fig. 5. The Reason Foundation as an authority.
4
This unit is a prominent authority ranking fourteenth among all authorities.
5
This unit is also a prominent hub in the eighteenth place among hubs.
P. Doreian and A. Mrvar
Social Networks 64 (2021) 148–157
154
reform, opposing liberal/progressive/Democratic agendas, opposing net
neutrality, private property rights, privatization, pro-family and traditional
family roles, promoting conservative justices to US courts, public
choice theory, right to life, rule of law, strong defense, taxes: avoiding,
lowering or eliminating them, Tea Party, white supremacy and white
nationalism, and women’s health.
The result is a two-mode network where KB allies form one set of
units and the keywords form the other. There is a total of 544 units of
which 497 are allied units of the KBs and 47 are keywords. It is
straightforward to cluster these interests in Pajek using the extent to
which all pairs of keywords are shared by the KB allies. The resulting
dendrogram is shown in Fig. 8.
The clusters can be interpreted both at fine scale level and by looking
at larger clusters. We focus on the latter. The top cluster of interests has
the keywords: monitoring and restricting voting; rule of law; Tea Party;
public choice theory; grass roots; economic growth; promoting conservative
justices to US courts, right to life; white supremacy and white
nationalism; and opposing campaign finance reform. All are well known
right-wing ideas.
The second cluster has heritage, against political correctness, antifeminist
and hostility to women, jerrymandering and redistricting,
obstructing consumer protection, opposing net neutrality, criminal justice
reform, evangelical, women’s health, and arbitration enforcement.
The third cluster contains attacking liberal bias in academia, private
property rights, mainstream media bias, gun rights, American heritage
and traditional values, pro-family and traditional family roles, antiimmigrants,
Islamophobia, strong defense, opposing abortion and
Fig. 6. Part of the KB network of allies with the top ten authorities placed in a small central circle.
P. Doreian and A. Mrvar
Social Networks 64 (2021) 148–157
155
Christian faith6
. Again, they form a coherent set of interests.
The fourth cluster contains: liberty; libertarian; anti-labor action;
opposing minimum wage laws less governmental regulation; blocking
Obamacare and subsequent healthcare reform efforts; attacking Social
Security and the social safety net, taxes - avoiding, lowering or eliminating
them. They form another coherent cluster of keywords. The final
cluster contains: free markets; limited government; privatization; conservative;
climate change denial; blocking environmental regulation;
freedom; opposing liberal/progressive/Democratic agendas.
6. Summary, limitations and future work
We have provided an analysis of the directed KB network of allies
using the concepts of hubs and authorities. In addition to computing
authority and hub scores for all units in the KB network, we looked
closely at the most important authorities and hubs to show their operational
roles in spreading ideologically based information and coordinating
actions of allies promoting an aggressive agenda regarding their
preferred form of limited government in the US and elsewhere.
One limitation of this study is that our data are based on one cross
sectional wave. However, a second web crawl has been completed using
the same set of starting organizations. A second potential limitation was
addressed by examining the broad concepts holding the KB network
together. Both the two-mode network and its two projections (Everett
and Borgatti, 2013 and 2020) featuring the keywords and the KB allies
will be examined further.
We have shown that using hubs and authorities is a useful approach
for capturing the operation of roles in a different fashion. Our hope is
Fig. 7. Part of the KB network of allies with the top ten hubs placed in a small central circle.
6
Other organizations are primarily Roman Catholics in their orientation and
declare their support for some KB interests.
P. Doreian and A. Mrvar
Social Networks 64 (2021) 148–157
156
that others will find this approach for their studies of directed networks
and the roles of units withing them.
References
Ackland, Robert, 2018. Virtual Observatory for the Study of Online Networks (VOSON)
Software for Collecting and Analysing Online Networks (2004-2018). https://resea
rchers.anu.edu.au/publications/140076.
Bardon, Adrian, 2019. The Truth About Denial: Bias and Self-Deception in Science,
Politics, and Religion. Oxford University Press, Oxford.
Batagelj, Vladimir, Mrvar, Andrej, 1998. Pajek - A Program for Large Network Analysis.
Connections 21 (2), 47–57.
Begley, Sharon, 2007. August 13. The truth about denial, vol. 150. Newsweek,
pp. 20–29.
Brownlee, Jamie., 2005. Ruling Canada: Corporate Cohesion and Democracy. Fernwood,
Halifax.
Brulle, Robert J., Aronczyk, Melissa, Carmichael, Jason, 2016. Corporate promotion and
climate change: an analysis of key variables affecting advertising spending by major
oil companies, 1986-2015. https://doi.org/10.1007/s10584-019-02582-8).
Carroll, William, Graham, Nicholas, Lang, Michael K., Yunker, Zoe, 2018. The Corporate
Elite and the Architecture of Climate Change Denial: A Network Analysis of Carbon
Capital’s Reach into Civil Society. Canadian Review of Sociology 55 (3), 425–450.
Doreian, P., Mrvar, A., 2020. The Koch Brothers and their climate change denial social
movement. In: Tindall, D., Stoddart, M.C.J., Dunlap, R.E. (Eds.), Handbook of AntiEnvironmentalism.
Edward Elgar Publishing.
Dunlap, Riley E., McCright, Aaron M., Yorosh, Jerrod H., 2016. The Political Divide on
Climate Change: Partisan Polarization Widens in the U.S. Environment 58 (5), 4–23.
Everett, Martin G., Borgatti, Stephen P., 2013. The dual-projection approach for twomode
networks. Social Networks 35, 178–203.
Everett, Martin G., Borgatti, Stephen P., 2020. Partitioning multi-mode networks.
Chapter 9 in.
Elsasser, Shaun W., Dunlap, Riley E., 2013. Leading Voices in the Denier Choir:
Conservative Columnist’s Dismissal of Global Warming and Denigration of Climate
Science. American Behavioral Scientist 57 (6), 754–776.
Farrall, Justin, 2016. Corporate funding and ideological polarization about climate
change. Proceedings of the National Academy of Science, U.S.A. 113 (1), 92–97.
Freeman, Linton C., 1979. Centrality in social networks: Conceptual clarification". Social
Networks 1 (3), 215–239.
Greenpeace, 2014. KOCH POLLUTION ON CAMPUS: Academic Freedom Under Assault
from Charles Koch’s $50 million Campaign to Infiltrate Higher Education.
Hummon, Norman P., Doreian, P., Freeman, Linton C., 1990. Analyzing the Structure of
the Centrality-Productivity Literature. Knowledge 11 (4), 460–481. https://www.
greenpeace.org/usa/global-warming/climate-deniers/koch-pollution-on-campus/.
Kleinberg, Jon, 1999. Authoritative sources in a hyperlinked environment. Journal of the
Association for Computer Machinery 46 (5), 604–632.
Intergovernmental Panel on Climate Change, 2001. IPCC Third Assessment Report:
Contributions of IPCC Working Groups. IPCC, Geneva, Switzerland.
Krugman, Paul, 2019. The Party that Ruined the Planet: Republican climate change
denial is even scarier than Trumpism. New York Times. Dec. 12, 2019.
Layzer Judith, A., 2012. Open for Business: Conservatives’ Opposition to Environmental
Regulation. MIT Press, Cambridge, MA. USA.
Fig. 8. The dendrogram for the clustering of the core interests of the KB allies.
P. Doreian and A. Mrvar
Social Networks 64 (2021) 148–157
157
Leonard, Christopher, 2019a. David Koch was the Ultimate Climate Change Denier. New
York Times. Aug 13, 2019.
Leonard, Christopher, 2019b. Kochland: The Secret History of Koch Industries and
Corporate Power in America. Simon & Schuster, New York.
MacLean, Nancy, 2017. Democracy in Chains: The Deep History of the Radical Right’s
Stealth Plan for America. Penguin Random House, New York.
Mayer, Jane, 2017. Dark Money: The Hidden History of the Billionaires Behind the
Radical Right. Anchor Books, New York.
McCright, Aaron M., Dunlap, Riley E., 2003. Defeating Kyoto: The Conservative
Movement’s Impact on U.S. Climate Change Policy. Social Problems 50 (3),
348–373.
National Research Council, 2001. Climate Change Science. National Academy Press,
Washington, DC.
Oreskes, Naomi, Conway, Erik M., 2010. Merchants of Doubt: How a Handful of scientists
Obscured the Truth on Issues from Tobacco Smoke to Global Warming. Bloomsbury,
New York.
Supran, Geoffrey, Oreskes, Naomi, 2017. Assessing ExxonMobil’s climate change
communications (1977-2014). Environmental Research Letters 12, 1–18.
Skocpol, Theda, Hertel-Fernandez, Alexander, 2016. The Koch Network and the
Rightward Shift in U.S. Politics. In: Presented at the Annual Meeting of the Midwest
Political Science Association. Chicago, IL, 2016.
Soldano, Henry, Santini, Guillaume, Bouthinon, Lazega, Emmanuel, 2017. HubAuthority
Scores and Attributed Directed Network Mining. IEEE International
Conference on Tool with Artificial Intelligence.
P. Doreian and A. Mrvar