Node xl book chapter oct 3

1
Think Link
Chapter XX. NodeXL Scholarship in Korea
Kyujin Jung, Ph.D.
Tennessee State University
Weiai Xu, Ph.D.
Northeastern University
Han Woo Park, Ph.D.
YeungNam University
In Korea, NodeXL has been not only widely
utilized for linking theory to practice but also
applied for investigating diverse features derived
from social media. To our surprise, NodeXL
manuals in Korean were proudly included in the
most viewed slides of 2013 according to
Slideshare.net. Furthermore, NodeXL Korea user
group has been formed in 2014 in order to
promote the NodeXL in terms of open tool in
social media network analysis. Regarding the
details of the NodeXL’s popularity in Korea,
please refer to the following document available at
the website:
http://www.slideshare.net/hanpark/note-about-
formation-of-node-xl-korea-users-group-
hanwoopark. This chapter aims to examine
thoroughly how NodeXL has been used for
analyzing political phenomena, public policy, and
cultural innovation cases under Korean socio-
economic contexts. In the following section, we
provide a review of the literature using NodeXL
on social media in the field of political
communication, public policy, and cultural
innovation. Then, this chapter concludes with
discussion of implications and directions for future
research.
X.1 NodeXL in Political Science and Public
Policy
In the Korean field of political science and public
policy, NodeXL has played a pivotal role in
examining communication patterns among
politicians and communities [1][2][3][4], political
participation [5], governments’ social media use
[6], and homeland security policy [7] on social
media such as Twitter and Facebook. Particularly,
previous researches focused on both introducing
NodeXL to the Korean academia and proposing a
methodological strategy to collect social media
network data. For instance, Kim and Park [5]
introduced NodeXL as a tool based on the
application programming interface (API) to
readers which helps researchers to collect data
from Twitter users. Using NodeXL, they not only
collected the number of Twitter followers and
followings, the list of Twitter followers’ and
followings’ ID, and the number of Tweets each
user published but also visualized the follower-
based and following-based matrices. As a result,
they found that resource-deficient politicians are
more likely to secure Twitter’s potentials than
other politicians.
Furthermore, NodeXL has provided
critical data-visualization functions, which allow
researcher to draw a range from entire networks to
ego-networks representations and to map data
attributes to visible properties e.g., nodes’ size,
color, shape, and transparency. In this vein, Choi
et al. [6] used NodeXL to detect and represent
innovation processes in social network data
collected from Twitter, highlighting that “a
particularly important aspect of NodeXL is that
social science researchers can improve the
efficiency of data visualization analysis by
examining data retrieved from a specific
geography and timeframe (p. 46)”. Through their
analysis using NodeXL, the results showed
evidence that innovation processes are mainly
emerged from overlapped members who

2
participated in multitudinous innovation
communities. Using NodeXL, Cho and Park [9]
also visualized ego-networks within Twitter
activities of the Korean Ministry for Food,
Agriculture, Forestry, and Fisheries to
demonstrate the relationships its followers and
followings. From a result of its ego-network
analysis, this research found that the Ministry’s
efforts to directly communicate with the public
under a specific event while Twitter has the
potential to facilitate risk communication.
With advanced network simulation
techniques, NodeXL is suitable to elaborate
politicians’ networking behavior [2][8][9][4] and a
particular event such as the 2013 North Korea’s
nuclear test [7]. In order to provide a better
understanding of the dynamics of the discussion
network, one hand, Choi et al. [9] collected the
longitudinal network data derived from Tweets
with the former president’s name and three types
of the messages (i.e., followings, mentions, and
retweets by using NodeXL applications. Using
sample from November 1, 2011, to April 20, 2012
included 26,150 Twitter users and 892,034
relationships, the results showed that the
discussion about President Myung-Bak Lee was
dominated by Twitter users who already had
considerable influence both online and offline. On
the other hand, Yoon and Park [2] utilized
NodeXL to collect data including Korea’s national
assemblymen and the most influential political
figures. Then, they tried to link the data to
exponential random graph models in order to
predict Twitter-based networking patterns of
actors. Particularly, it can be considered as a
critical step to fostering NodeXL by testing
hypothesized structures in social media network
data as advanced applications of NodeXL [10]. As
shown in Figure X.1, on top of that, Jung and Park
[7] proposed a systemically-designed research
using NodeXL. Following the case of the 2013
North Korea’s nuclear test, they collected social
media network data from Twitter for four weeks.
Again, they initially collected data for respectively
two weeks before and after the nuclear test. Based
on the longitudinal data, they showed the change
of Twitter-based networks’ attributes such as size
of network, density, reciprocity rate, and page
rank index. By collecting the data with key words
in English and Korean through NodeXL, they
found evidence that the impact of social contexts
matters in the evolution of each international and
Korean networks.
Figure X.1 During the 2013 North Korea’s nuclear test, changes in the average geodesic network distance and density

3
Regarding to natural disaster such as
tropical typhoons and floods occurred in Korea,
scholars in the field of emergency management
started using NodeXL to investigate patterns of
risk communication on social media. Since
citizens use social media as the source of disaster
information for helping them evacuate from
disaster-impacted sites, NodeXL can be used not
only to collect real-time social media network data
but also to contribute to decision-making
procedures of principal agencies such as national,
provincial, and local emergency operations centers.
According to the Seoul research report [8], for
instance, NodeXL seems to fill the gap of a
current risk management system operated by the
Seoul Metropolitan Government. The report
recommends NodeXL for public administrators to
identify emergent needs of victims and affected
communities, which facilitate stakeholders to
make a timely decision during disaster responses.
As a result of the social media network analysis
using NodeXL, the report presented policy
implication that local governments within the City
of Seoul are recommended to periodically monitor
social media and then provide customized disaster
information that citizens need after a disaster.
Recent frequent occurrences of
overwhelming events raise urgent need for
studying effective risk communication on social
media. In this point of view, NodeXL is a key to
examine the patterns of risk communication
embedded in social media networks. Song et al.
[11] utilized NodeXL to identify the differences
patterns between Korean and international
contexts by comparing Korean and international
networks based on the social amplification of risk
framework. The results in Figure X.2, from
NodeXL analysis focusing on interpersonal risk
communication in the context of the Sewol ferry
disaster, show that the Korean risk communication
network was more fragmented, and its clustering
was more sparsely knitted based on the impact of
issues and the physical proximity of the disaster.
Figure X.2 After the Sewol ferry disaster, Korean risk communication networks on Facebook

4
X.2 NodeXL in Cultural Innovation
NodeXL provides a systematic look over the
fabrics of our digital culture. The tool addresses a
fundamental shift of power in cultural innovations.
The contemporary culture is increasingly mediated
through participatory internet platforms in which
networked cultural consumers encounter, evaluate
and recreate cultural symbols. What drive cultural
innovations are not single cultural institutions,
governments or policies, but a networked online
community of active cultural participants.
NodeXL can be used to visualize and measure the
structure of the community, thus providing
compelling insights into two essential qualities of
digital culture: virality and meme.
Virality ensures the wide exposure of
cultural offerings to the global audience. Strategic
promotion of national cultures becomes possible
through viral videos. The successful cases include
Korean artist Psy’s horse-dancing Gangnam Style,
which has become synonymous with Kpop. Meme,
referring to a host of user-generated culture
inspired by the original viral symbols, expands the
longevity of viral culture [12]. Following the
release of Gangnam Style, various genres of remix
and mash-ups bubbled up to create a viable
cultural ecosystem. The two qualities call for a
paradigm shift from studying established cultural
enforcers to studying networked individuals in
cultural systems. In light of the explosive growth
of viral digital culture, in which Korea has played
a dominant role, a new research strain has been
developed to explore the interplay in the Kpop
mediated through digital platforms. Such
knowledge is critical in understanding how nation-
states and corporates strategically enhance their
brand images [13]. The research can be integrated
to build a hybrid webometric model of cultural
innovations, built on webometrics enabled by
NodeXL. Figure X.3 shows a visualization of the
webometric model used in Xu, Park, and Park’s
[13] work.
Figure X.3 The webometric model for examining YouTube-based cultural innovations through NodeXL

5
The hybrid model uses NodeXL to
examine three layers of ties in YouTube-based
networks of cultural innovations. The first layer is
based on reply-to-comment behavior among users,
meaning that A is connected to B when A replies
to B’s comments on a video [14]. Such directional
ties reveal exchanges of ideas within the online
community formed on collective cultural
evaluation. Using the case of Gangnam Style, Xu,
Park and Park [13] showed that YouTube
commentators, while being the minority of their
peer, nevertheless engaged in continuous and
reciprocal exchanges of cultural critique.
Interestingly, center users in such discussion
networks—those who are most active in replying
others’ comments and most frequently targeted by
other users, voiced critical opinions towards the
cultural offering. In a way, these influential users
are not necessarily evangelists but critical
connoisseurs. Their opinions possibly resonate
with the rest of the community. Among the active
commentators, their intensively of participation
varied by gender: male commentators, in the case
of Gangnam Style, were more active than females,
and those residing in countries culturally similar to
Korea were more likely to express favorable
attitudes. Figure X.4 displays the network graph
described in the above study. Taking a
longitudinal look [15], the influential commenters,
majority of them are amateur users, retained their
central dominance over time. But the structure of
the entire discussion network changed: the
network became smaller, less cohesive and more
compartmentalized, reflecting the waning of
audience interest and engagement, and possibly,
the increasing divide among users over cultural
Figure X.4 YouTube network based on users’ reply-to-comment activities in the diffusion of Gangnam style

6
tastes and opinions.
The second layer of ties reflects shared
video taste/interest among users. The ties are
based on co-subscription to the same YouTube
channel, that is, A and B are connected when both
subscribe to the same channel [14]. This
symmetric network tends to be denser than reply-
to-comment networks because subscription
requires minimal cognitive and time input than
commenting. By tabulating the most shared
channels, Xu, Park and Park [13] showed that
commentators in the case of Gangnam Style
shared topical interest in Kpop and peculiar
content. The shared video interest disintegrated
over time [15] — as time went by, commentators
shared less video interest, reflected in decreasing
density in the network of co-subscription.
The third layer of ties is based on user-
generated content, in comments and user-
generated videos. Semantic networks feature
topical links based on semantic co-occurrences
within the same text [16]. Semantic analysis—a
key feature in NodeXL—has been used to reveal
salient and thus influential concepts in user
discussions. The study of Gangnam Style on
YouTube showed that discussions by
commentators focused on the cultural origin of the
video and related the content to the broader
national and cultural image of a foreign country
[13]. NodeXL also features connections between
cultural objects in meme. This object-object
network underlies topical similarities between two
videos and the mutual attention they are able to
drive. In this network, videos A and B are tied
when both are commented on by the same user
[14]. By examining the network structure and
unique positions, the analyses can reveal
influential cultural objects in the entire meme
ecosystem. A study of meme inspired by
Gangnam Style showed that memetic creations fall
in the categories of remix, reproducing
Figure X.5 A network of meme, with links indicating mutual audience attention attracted by two user-generated videos.

7
background music, parody, physically imitating
the horse-dancing, making verbal comments
through response videos and etc. [17]. Figure X.5
visualizes a network of memes identified in Xu et
al., (in press). The genre reflects varying degree of
user participation, and accordingly, result in
different levels of audience attention: for example,
a few remix and physical imitation videos
attracted a large number of commentators, but
reaction videos (verbal commenting) drew most
cross-commenting, likely due to controversial
nature of opinion expression. Based on central
positions in the meme network, the content
provided by traditional mass media gave the viral
cultural symbol wide publicity. But the dominant
role of traditional media was later shared by
amateur users.
Taking into account the three types of ties,
the networked perspective towards cultural
innovations on YouTube reveals the structure of
the social system in diffusion of innovations.
NodeXL is critical in visualizing the relationship
patterns in the social system, whether it is the
relationship formed on social interaction, shared
interest or topical similarities. Discussing salient
actors in the network reveals the important roles of
various diffusion actors such as innovators and
early adopters. Examining the changing landscape
of the social system also reveals various diffusion
stages such as evaluation, trial, etc. Overall, the
insights add important pieces to the Rogers’ [18]
classical diffusion framework.
X.3 Directions for Future Research using
NodeXL
While much of the current NodeXL research in
Korea has focused on capturing the nature of
social media network, its users may overlooked a
qualitative perspective that facilitate researchers to
fill the gap of quantitative results derived from
NodeXL’s applications. For instance, testing
patterns of interactions among actors can explain
structural relationships such as reciprocal and
transitive ties, but it seldom clarifies potential
factors to build the patterns. The qualitative
approach to NodeXL’s applications helps
researchers systemically design their data
collection procedures and analysis methods. Prior
to using NodeXL, contacting and interviewing key
stakeholders engaged in certain policy issues
sheds light on the motivation of their interactions
with other actors on social media network. To
undoubtedly answer a research question derived
from theoretical considerations, NodeXL presents
the key to data collection and analysis methods,
but future research should consider its applications
as a critical lens to analyze and evaluate social
phenomena and previous research by
incorporating the qualitative perspective.
Cultural innovations, on the other hand, is
an interesting case for testing classical social
science theories such as diffusion and opinion
leadership. In the theoretical testing, NodeXL
provides a structural view of the entire
communication system. The networked
perspective, in future studies, can be connected
with individual-level analyses of important actors.
For example, in the diffusion of viral cultural
symbols, patterns of individual network positions
can be compared to the user’s self-reported data.
In addition, NodeXL and the network perspective
it represents provide new territories for testing old
theories. For example, there is a recent call for
combining social network analysis with agenda-
setting [20]. In addition, findings from the
aforementioned YouTube studies of cultural
innovations can be interpreted in conjunction with
findings of global cultural innovations on Twitter
[3][19]. NodeXL can be used to map connections
between attributes and objects in cultural
innovations, as researchers previously did, and
then such connections can be compared and
contrasted with individual perception of the
salience of certain cultural objects.

8
In terms of data validity and reliability
issues, studying political theory and public policy
embedded in social media network recommends
us to narrow down a case of research topics with
social events such as the 2012 president election
and the 2013 North Korea’s nuclear test. That is,
because, the network collected and visualized by
NodeXL might be part of an entire network on
Facebook or Twitter. In future research, social
media network data tends to be more complicated
and/or overlapped with not-relevant stakeholders,
and thus it should be intentionally designed for
data collection and methods such as keywords and
time points used by a researcher. Furthermore,
sharing data collection and cleaning procedures
with others through NodeXL Graph Gallery can
be an opportunity to increase the validity and
reliability of social media network data.
Finally, the combination of network
perspective and analyses of individual perception
requires us to integrate network data and survey
data. This represents a new frontier for future
methodological development of NodeXL studies.
Another possible methodological innovation, in
the context of cultural innovations, is to compare
YouTube-based networks with networks on their
social media platforms. YouTube arguably can be
viewed as an entertainment media platform, more
for content consumption and less for social
interactions, whereas, on Facebook, social needs
become a more salient need. It is worthy
discussing how culture diffuses through networks
underlying different types of relationship ties.

9
References
[1] Choi, S., Park, J. Y., & Park, H. W. (2012).
Using social media data to explore
communication processes within South
Korean online innovation communities.
Scientometrics, 90, 43-56.
[2] Yoon, H. Y. & Park, H. W. (2014). Strategies
affecting Twitter-based networking pattern
of South Korean politicians: social
network analysis and exponential random
graph model. Quality & Quantity, 48, 409-
423.
[3] Choi, S. C., Meza, X. V., & Park, H. W.
(2014). South Korean Culture Goes Latin
America. International Journal of
Contents, 10(1), 36-42.
[4] Cho, I., Choi, S. C., & Park, H. W. (2015).
Speech Acts in Televised Presidential
Debates and Facebook Messages: The
Case of the 2012 South Korean
Presidential Election. Journal of the
Korean Data Analysis Society 17 (3),
1185-1201.
[5] Kim, M. & Park, H. W. (2012). Measuring
Twitter-based political participation and
deliberation in the South Korea context by
using social network and Triple Helix
indicators. Scientometrics, 90, 121-140.
[6] Cho, S. E. & Park, H. W. (2012). Government
organizations’ innovative use of the
Internet: The case of the Twitter activity of
South Korea’s Ministry for Food,
Agriculture, Forestry, and Fisheries.
Scientometrics, 90, 9-23.
[7] Jung, K. & Park, H. W. (2014). Citizens'
Social Media Use and Homeland Security
Information Policy: Some Evidences from
Twitter Users during the 2013 North Korea
Nuclear Test. Government Information
Quarterly, 31, 563-573.
[8] Jung, K. (2014). Social Media Use for
Building Safe Seoul: Focusing on Civic
Engagement in Emergency Management.
Seoul Institute (Seoul, Korea), Seoul
Research Report 2013-PR-54.
[9] Choi, M., Sang, Y., & Park, H. W. (2014).
Exploring political discussions by Korean
twitter users. Aslib Journal of Information
Management 66 (6), 582 - 602
[10] Jung, K., Park, S. J., Wu, W., & Park, H. W.
(2015). A Webometric Approach to Policy
Analysis and Management using
Exponential Random Graph Models.
Quality & Quantity, 49 (2), 581-598.
[11] Song, M., Jung, K., Park, J. Y., & Park, H. W.
(2015). Different Structure of Risk
Communication Networks during the
Sewol Ferry Disaster: Comparative
Approach between Korea and International
Networks on Twitter and Facebook. The
Global Information Technology
Management Association (GITMA) 2015
conference proceedings, 82-104.
[12] Shifman, L. (2012). An anatomy of a
YouTube meme. New Media & Society,
14(2), 187-203.
[13] Xu, W. W., Park, J. Y., & Park, H. W.
(2015a). The networked cultural diffusion
of Korean wave. Online Information
Review, 39(1), 43-60.
[14] Hansen, D, Shneiderman, B, & Smith, M.A.
(2011). Analyzing social media networks
with NodeXL. Burlington, MA: Morgan
Kaufmann.
[15] Xu, W. W., Park, J. Y., & Park, H. W.
(2015b). Longitudinal Dynamics of the
Cultural Diffusion of Kpop on YouTube.
Telematics and Informatics, 39(1), 43-60.
[16] Chung, C. J., & Park, H. W. (2010). Textual
analysis of a political message: The
inaugural addresses of two Korean
presidents. Social Science Information,
49(2), 215-239.
[17] Xu, W. W., Park, J. Y., & Park, Kim, J.Y., &
Park, H. W. (Forthcoming). Networked
cultural diffusion and creation on YouTube:
An analysis of YouTube memes. Journal
of Broadcasting & Electronic Media.
[18] Rogers, E. M. (2003). Elements of diffusion.
Diffusion of innovations, 5, 1-38.
[19] Meza, X. V., & Park, H. W. (2014).
Globalization of cultural products: a

10
webometric analysis of Kpop in Spanish-
speaking countries. Quality & Quantity, 1-
16.
[20] Guo, L. (2012). The application of social
network analysis in agenda setting research:
A methodological exploration. Journal of
Broadcasting & Electronic Media, 56(4),
616-631.

Node xl book chapter oct 3

More Related Content

What's hot (20)

Viewers also liked (11)

Similar to Node xl book chapter oct 3 (20)

More from Han Woo PARK (20)

Recently uploaded (20)

Node xl book chapter oct 3