Polarización política y liderazgos en la red de Twitter de los parlamentarios catalanes

AutorRosa Borge Bravo - Marc Esteve Del Valle
CargoResident researcher at the Internet Interdisciplinary Institute, Associate Professor of Political Science (IN3-Universitat Oberta de Catalunya) - Researcher and PhD grant-holder, Doctoral Programme in Information and Knowledge Society, Internet Interdisciplinary Institute (IN3-Universitat Oberta de Catalunya)
Páginas43-60
IDP no 20 (June, 2015) I ISSN 1699-8154 Journal promoted by the Law and Political Science Department
www.uoc.edu/idp
Submission date: May 2015
Accepted date: May 2015
Published in: June 2015
Universitat Oberta de Catalunya
Abstract
This article seeks to test if Twitter is opening communication flows between parties and paving the way
for the appearance of network leaderships or if it is reducing them to representatives of the same party
or ideology and empowering party leaders. The study is based on a dataset spanning from 1 January
2013 to 31 March 2014 which covers all relationships (4,516), retweets (6,045) and mentions (19,507) of
Catalan parliamentarians. The data sustain that the political polarisation of Catalan parliamentarians’
Twitter flows is deeper in the relationship and retweet networks than it is in the mention and that even
if party leaders are still at the centre of communication flows, other parliamentarians are taking the
floor and becoming network leaders.
Keywords
parliamentarians, Twitter, polarisation, network leaderships, Catalonia
Topic
political science, communication, network analysis
ARTICLE
Political Polarisation
and Network Leaderships
in the Catalan Parliamentarians’
Twitter Network*
Rosa Borge Bravo
Resident researcher at the Internet Interdisciplinary Institute
Associate Professor of Political Science (IN3-Universitat Oberta de Catalunya)
Marc Esteve del Valle
Researcher and PhD grant-holder
Doctoral Programme in Information and Knowledge Society
Internet Interdisciplinary Institute (IN3-Universitat Oberta de Catalunya)
Rosa Borge Bravo, Marc Esteve del Valle
43
* The authors would like to thank the help given by Miguel Ángel Domingo (UOC) and Carles Llorach (UOC) in the data
mining process of the Catalan MPs’ Twitter interactions.
IDP no 20 (June, 2015) I ISSN 1699-8154 Journal promoted by the Law and Political Science Department
Eloi PuigEloi Puig
Jose R. Agustina
www.uoc.edu/idp
Universitat Oberta de Catalunya
44
Political Polarisation and Network Leaderships in the Catalan Parliamentarians’ Twitter Network
Eloi PuigEloi Puig
Rosa Borge Bravo, Marc Esteve del Valle
44
Polarización política y liderazgos de red en la red Twitter
de los parlamentarios catalanes
Resumen
Este artículo trata de probar si Twitter abre flujos de comunicación entre partidos distintos y allana el
camino para la aparición de liderazgos en la red o si los reduce a representantes del mismo partido o
ideología y empodera a los líderes del partido. El estudio se basa en un conjunto de datos que abarca
desde 1 de enero de 2013 al 31 de marzo de 2014 e incluye todas las relaciones (4516), retweets (6045) y
menciones (19.507) de los parlamentarios catalanes. Los datos sostienen que la polarización política de
los parlamentarios catalanes en los intercambios de Twitter es más profunda en las redes de relaciones y
las redes de retweets que en las redes de menciones y que, aunque los líderes del partido se encuentran
todavía en el centro de los flujos de comunicación, otros parlamentarios están tomando la palabra y se
están convirtiendo en líderes en la red.
Palabras clave
parlamentarios, Twitter, polarización, liderazgos de red, Cataluña
Tema
ciencias políticas, comunicación, análisis de redes
Introduction
MPs’ use of social media is opening a new online political
arena that has its own particularities and logics of power
distribution. In Catalonia, where 85% of parliamentarians
have a Twitter1 account, this new Internet-mediated political
realm is already a reality.
The MPs’ behaviour on Twitter has been studied from
multiple angles which go from the reasons behind the
parliamentarians’ adoption of Twitter (Lassen and Brown,
2010; Chi and Yang, 2010; Williams and Gulati, 2010) to
their use of this social networking site to communicate
with the electorate (Small, 2010; Margaretten and Gaber,
2014). More recently, some authors have studied the
phenomena of political polarisation (Gruzd, 2012) and
network leaderships (Gonzalez-Bailon and Wang, 2014)
in Twitter networks. However, as far as we know, no study
has examined political polarisation and leadership issues
on Twitter parliamentarian networks.
Therefore, this research attempts to analyse some of
the effects of Twitter on the directions (Bimber, 2014) of
parliamentarians’ online communication flows
2
. Specifically,
this article applies statistical analysis and social network
methods together in order to offer plausible and refutable
answers to the following questions: Is Twitter facilitating
Catalonian parliamentarians’ exposure to cross-ideological
opinions or is it confining them to like-minded clusters? Is
Twitter empowering official party leaders or paving the way
for the appearance of new network leaderships?
To do this, we have created a specific research engine to
gather all the Catalan parliamentarians’ relations -following/
follower- (4,516), retweets (6,045) and mentions (19,507)
1. An online social networking website and microblogging service that allows users to post and read text-based messages of up to 140
characters, known as tweets. It gives its users the opportunity to share short messages (tweets), respond to tweets (reply), mention other
users’ account (mention) or simply forward a tweet (retweet).
2. Manuel Castells defines these flows as “streams of information between nodes circulating through the channels of connection between
nodes” (Castells, 2009, p. 20).
IDP no 20 (June, 2015) I ISSN 1699-8154 Journal promoted by the Law and Political Science Department
Eloi PuigEloi Puig
Jose R. Agustina
www.uoc.edu/idp
Universitat Oberta de Catalunya
45
Political Polarisation and Network Leaderships in the Catalan Parliamentarians’ Twitter Network
Eloi PuigEloi Puig
Rosa Borge Bravo, Marc Esteve del Valle
45
from the 1st January 2013 – 31 March 2014 period. That data
covers an extraordinary and conflictive time in Catalonia in
which demands for independence and a referendum as well
as protests against austerity measures were widespread. We
have chosen the Catalan Parliament because the Catalan
party system is fragmented and ideologically polarised so it
provides a perfect context for examining cross-ideological
and inter-partisan interactions.
The paper is organised as follows: First, we will analyse
related work on political polarisation on Twitter and political
leaderships on Twitter networks. Continuing with details
of the design of the research, construction of variables
and stating the research hypotheses to be tested. Then we
will introduce the Catalan case. We will finally present the
findings of the article and discuss their implications.
2. Theoretical background
A growing body of research has begun to examine
the influence of social networking sites on political
communication from multiple angles, yet there is still a long
way to go in studying Internet-mediated social networks of
members of Parliament and the implications of these new
forms of communication between parliamentarians on party
leadership and relationships between parties. We will review
literature dealing with two aspects: a) Political polarisation
on Twitter; b) Twitter networks and political leaderships.
Political polarisation on Twitter
The phenomena of online political polarisation on Twitter
has been recently analysed in some papers amongst which
Yardi and Boyd’s (2010) study of peoples’ tweets on the
shooting of George Tiller revealed that “people are more
likely to reply to people who share the same view” (Yardi
and Boyd, 2010, p.13). In this same light, Conover et al’s.
(2011) study of more than 250,000 tweets from the six
weeks leading to the 2010 U.S. congressional midterm
elections proved that while “retweet networks exhibited
a highly modular structure segregating users into two
homogenous communities corresponding to the political
left and right, the mention network did not show this kind
of political segregation, resulting in users being exposed
to individuals and information they would not have been
likely to choose in advance” (Conover et al., 2011, p. 89).
Gruzd (2012) added a Canadian perspective to this field
of research and his network analysis of 5,918 tweets on
the 2011 Canadian Federal election concluded that “on the
one hand, there was a clustering effect around shared
political views among supporters of the same party in the
communication network of Twitter. On the other hand, there
was also evidence of cross-ideological connections” (Gruzd,
2012, p. 8). In this same line of research, Hsu and Park’s
(2012) study of the communication relationships between
members of the Korean National Assembly pointed to the
fact that intraparty connections were stronger than inter-
party connections on Twitter (Hsu and Park, 2012, p.5)
and Himelboim et al.’s social network and content analysis
of 10 controversial U.S. political topics on Twitter showed
that “political content was overall confined to like-minded
clusters of users” (Himelboim et al., 2013, p. 18). Also,
Conover at al., (2011) and Yoon and Park (2014) noticed
that while retweet and following networks reproduced
the known partisan split in the online world, the mention
network showed that ideologically opposed individuals
interacted with each other.
It seems therefore that political polarisation on social
media could be explained by the well-known sociological
phenomenon called homophily (McPherson et al., 2001),
by which interactions among similar people occur more
often than among dissimilar people, or, as Himelboim et
al. (2013) said, birds of a feather tweet together. However,
findings in the literature show that even if Twitter political
communications are reproducing the homophily concept
(Yardi and Bold, 2010; Conover et al., 2011; Himelboim et al.,
2012; Hsu and Park, 2012, Esteve and Borge, 2014), they are
also facilitating cross-party and cross-ideological discourse
(Conover et al., 2011; Gruzd, 2012).
To sum up, given the relation between the different types
of networks and the diverse political polarisation degree
reached by the parliamentarians in each one of these Twitter
layers, the following hypothesis is proposed:
H1: Communication flows of Catalan parliamentarians’
following-follower and retweet Twitter networks are
polarised by party affiliation and ideology while those of
the mention network reveal some cross-ideological and
cross-party connections.
Twitter networks and political leaderships
Recently, some authors have started to collaterally point out
the attributes of Twitter leaderships. Yoon and Park’s (2011)
study of Korean politicians on Twitter showed that while
IDP no 20 (June, 2015) I ISSN 1699-8154 Journal promoted by the Law and Political Science Department
Eloi PuigEloi Puig
Jose R. Agustina
www.uoc.edu/idp
Universitat Oberta de Catalunya
46
Political Polarisation and Network Leaderships in the Catalan Parliamentarians’ Twitter Network
Eloi PuigEloi Puig
Rosa Borge Bravo, Marc Esteve del Valle
46
the following-follower network was a social ritual network
(therefore with high reciprocated vertex pair ratio figures)
3
,
the mention network was a network of political support
characterised by the fact that some politicians received
the majority of mentions. Research carried by Hsu and
Park on communication relationships among members
of the Korean National Assembly also revealed that their
following-follower network followed a preferential power
law which is the same as saying that “several politicians in
the network where far more popular than the others” (Hsu
and Park, 2012). Lastly, González-Bailon and Wang (2013)
have studied the effects of brokers in bridging or bonding
(Putnam, 2000) social network clusters.
In sum, given the previous studies on the field which point
that Twitter networks follow a preferential attachment
power law skewed toward the relevant politicians or the
party leaders (Yoon and Park, 2011; Hsu and Park, 2012)
but that the relational power derived from the nature of
Twitter networks is paving the way for the appearance of
new sorts of network leaderships, not equivalent to official
party leaderships (González-Bailon and Wang 2013), the
following hypothesis is proposed:
H2: Communication flows in Catalan parliamentarians
Twitter network are empowering party leaders but also
paving the way for the appearance of network leaderships.
3. Data and methods
Our intention in this research is to test if communication
flows in the Twitter network of Catalan parliamentarians
are politically polarised or if, on the contrary, they facilitate
communication between parliamentarians of different
parties. We also wish to ascertain if communication flows of
Catalan parliamentarians’ Twitter networks are empowering
party leaders or paving the way for the appearance of
network leaderships.
In order to test our hypotheses empirically, first we manually
compiled a list containing usernames of members of the
Catalan Parliament on Twitter. There are 116 Catalan
Parliamentarians with Twitter accounts (85% of the total
number). Then we created a research engine to gather
all the relations, retweets and mentions of the Catalan
parliamentarians from 1 January 2013 to 31 March 2014,
which permitted us to quantitatively determine the
direction of Twitter communication flows among Catalan
representatives.
Once the directions of the communication flows were known,
we determined which parliamentarians were occupying a
centrality position in the three types of Twitter networks.
The centrality position was measured by collecting the
total number of parliamentarians following another
parliamentarian and the total number of retweets and
mentions a parliamentarian received. Then, regression
analyses were carried out (based on our own database)
to ascertain the individual attributes of the Catalan
parliamentarians triggering a centrality position in the
following–follower, retweet and mention Twitter networks.
The regressions were run on the 116 Catalan Parliament
members with Twitter accounts. Our independent variables
in the model include 7 dimensions
4
and are drawn from the
literature shown in previous sections:
1. Socio-demographic characteristics (age, gender, education);
2. Internet behaviour (having a blog or Facebook account)
and Twitter activity (total number of tweets; total number
of retweets sent; total number of retweets received; total
number of mentions sent; total number of mentions
received). We also introduced an estimation of the favourites
of the Catalan representatives based on their favourites per
day average from 1 January to 31 March 2014;
3. Electoral characteristics such as the party they belong to and
the number of legislatures they have attended Parliament
(up to three: 2006, 2010, 2012);
4. Parliamentary activity (number of legislative commissions in
which they participate and number of interventions in these
commissions and in the plenary sessions);
3. In a directed graph, the reciprocated vertex pair ratio is the number of adjacent vertices that are connected to the vertex with edges
in both directions divided by the number of adjacent vertices. For instance, in a parliamentarian Twitter network the vertices are the
parliamentarians holding a Twitter account.
4. See appendix for codification and description of the variables. The data was retrieved on January 2014. Most of the values of the
parliamentarians’ individual characteristics did not vary along the year.
IDP no 20 (June, 2015) I ISSN 1699-8154 Journal promoted by the Law and Political Science Department
Eloi PuigEloi Puig
Jose R. Agustina
www.uoc.edu/idp
Universitat Oberta de Catalunya
47
Political Polarisation and Network Leaderships in the Catalan Parliamentarians’ Twitter Network
Eloi PuigEloi Puig
Rosa Borge Bravo, Marc Esteve del Valle
47
5. Political position in the Parliament (role in the parliamentarian
group and Parliament and role in Parliament commissions),
at the party (party president) or as mayors when that is
the case;
6. Relational variables: In-Degree (Number of parliamentarian
followees a parliamentarian Twitter holds), Out-Degree
(Number of parliamentarians each parliamentarian is
following on Twitter);
7. Cleavages: Left (ERC, PSC, ICV-EUiA and CUP)- Right (CiU,
PP and C’s) and Catalan nationalists (CiU, ERC, ICV-EUiA
and CUP) and Non-Catalan nationalists (PSC, PP and C’s).
Lastly, we carried out a network analysis in order to find
the centrality position of Catalan parliamentarians’ in the
following-follower, retweet and mention networks but in a
different way that we previously did. This time, we measured
the centrality position of the Catalan parliamentarians
by using the total number of parliamentarians that
were following, retweeting or mentioning a particular
parliamentarian. By doing so, what we wanted to find was
not the centrality position of the Catalan parliamentarians
based on the quantity of retweets and mentions that
they received
5
(which has been previously done) but to
ascertain the distribution of the mentions and retweets
that a particular parliamentarian received. More precisely,
this network twist gives more weight to the distribution
of the mentions and retweets than to the quantity of
retweets and mentions that a parliamentarian received.
Consequently, the Catalan parliamentarians with a higher
centrality position will be those who receive more mentions
and retweets from the total number of deputies (and vice
versa).
4. The Catalan Parliament
and Twitter
The Catalan parliament is a good case study for several
reasons: First, the early adoption of networking sites by
the Catalan Parliament and its members. On 17 March
2009, the Catalan Parliament initiated the Parliament
2.0 project which consisted in ‘adapting the Parliament
to the new active role of users with social media’
(Benach, 2010). As a result, the Parliament launched a
YouTube channel and a Facebook and Twitter profile.
Furthermore, in October 2013, the Catalan Parliament
initiated the project Escó 136,
6
consisting in a web page
on which Catalan citizenry could leave comments and
suggestions to projects and laws proposed by the Catalan
Parliament7. Second, the ratio of Catalan parliamentarians
with Twitter profiles in 2014 (85%), was greater than
the number of parliamentarians with Twitter profiles in
the Spanish central parliament (52.6%)
8
, in the Spanish
senate (33.06%)
9
, in the German Bundestag (31.61%)
(Thamm and Bleier, 2013), in the UK House of Commons
(72.3%)10, but somewhat behind the number of politicians
with Twitter profiles in the US senate (100%) and in the
House of Representatives (90%)
11
.
It is also important to note the fragmentation and ideological
divisions in the Catalan party system that enables us to test
hypotheses related to interaction and communication intra
and between several parliamentarian groups in depth. That
is, specifically to test the polarisation degree among the
nodes of the network and their leadership role in terms of
brokerage and in-degree positions. In that regard, after the
25 November 2012 elections, the Catalan party system is
fragmented into a wide variety of fringe and medium-size
parties: CiU
12
(50 seats), ERC
13
(21 seats), PSC
14
(20 seats),
5. Evidently, in the case of the following-follower network the centrality position of the Catalan parliamentarians will be always measured by
the number of parliamentarians following a precise parliamentarian.
6. Seat 136 as the Parliament seats 135 members.
7. The comments and suggestions of the citizenry are transferred to the authorities in charge of elaborating Catalan legislation and they
will be annexed to the law initiative. See http://www.parlament.cat/web/participacio/esco-136
8. http://parlamento20.es/twitter-en-el-congreso-de-los-diputados
9. http://www.blogsdepolitica.com/el-senado-tambien-existe-presencia-y-actividad-en-twitter-de-los-senadores/
10. http://blogs.telegraph.co.uk/news/willheaven/100228088/how-long-before-every-single-mp-is-on-twitter
11. https://blog.twitter.com/2013/100-senators-and-57th-inauguration
12. CiU is a Catalan nationalist centre-right party.
13. ERC stands for republican left of Catalonia and it is a left wing and Catalan nationalist party.
14. PSC is the Socialist Party of Catalonia and it is federated with the Spanish Socialist Party.
IDP no 20 (June, 2015) I ISSN 1699-8154 Journal promoted by the Law and Political Science Department
Eloi PuigEloi Puig
Jose R. Agustina
www.uoc.edu/idp
Universitat Oberta de Catalunya
48
Political Polarisation and Network Leaderships in the Catalan Parliamentarians’ Twitter Network
Eloi PuigEloi Puig
Rosa Borge Bravo, Marc Esteve del Valle
48
1 2 3 4 5 6 7
1
1,5
2
2,5
3
3,5
4
4,5
5
Left-Right
Non Catalan Nationalist / Catalan-Nationalist
ERC
CIU
CUP
ICV
PSC
C’S
PP
PP
15
(19 seats), C’s
16
(9 seats), ICV-EUIA
17
(13 seats) and CUP
18
(3 seats). Indeed, the Catalan party system is distributed
along two main ideological cleavages (see Figure 1): left/
right-wing and Non Catalan nationalist/Catalan nationalist.
The figure below places Catalan parties on these two axes:
Several studies have already shown that social media
are contributing to equalise opportunities for political
communication between Catalan parties, since new, fringe
and medium-size parties as well as parties from varied
different political positions achieve greater online interaction
and participation than larger and more institutionalised
parties (Balcells and Cardenal, 2013; Esteve and Borge,
2013). We expect therefore that, in the particular political
context of the Catalan parliament, some of the network
leaderships that could arise from the Twitter interaction
between the parliamentarians will be members of the new
and medium-size parties.
5. - Findings
Political polarisation
Communication flows in Catalan parliamentarians’ Twitter
networks show both party and ideological polarisation.
However, the intensity of this political polarisation is
deeper in the following-follower and retweet networks
than in the mention network. More precisely, as can be
seen in Table 1 and Figure 2, members of the Catalan
15. PP is the Popular Party of Catalonia, which is a member of the Spanish Popular Party, a right-wing and Spanish nationalist party.
16. C’s stands for ‘Citizens’ and it is a relatively new centrist party that is against Catalan nationalism.
17. ICV-EUIA is a left-wing party self-defined as eco-socialist and an autonomous part of the Spanish United Left (IU).
18. CUPs stands for ‘Candidacies of Popular Unity’ and are an extreme left and Catalan nationalist coalition.
Figure 1: The position of Catalan parties on the political spectrum of
Catalonia according to Catalan respondents
Source: Second round of CEO surveys (20 June 2013); N=2000.
Table 1: Percentage of parliamentarians’ following relations over the total possible MPs’ following relations between
parliamentarian groups in the Catalan parliamentarians’ Twitter network (1 January 2013–31 March 2014)
Parliamentarian
Groups and their
cleavage positioning
Left (1)-Right (0)
Catalan-nationalist (1)/
Non Catalan nationalist
(0)
CiU
(35)
ERC
(21)
PSC
(21)
PP
(15)
ICV-
EUiA
(11)
C’S
(9)
CUP
(3)
CiU
0 1 71.93% 27.17% 16.53% 19.71% 12.03% 11.76% 3.92%
ERC
1 1 36.13% 90.48% 14.99% 13.33% 34.63% 5.29% 33.33%
PSC
1 0 28.43% 23.39% 84.29% 32.70% 44.16% 24.34% 14.29%
PP
0 0 27.84% 12.70% 18.41% 95.24% 7.88% 17.78% 4.44%
ICV-EUiA
1 0 22.73% 44.16% 41.13% 22.42% 95.45% 18.18% 18.18%
C'S
0 0 16.34% 8.99% 18.52% 50.43% 18.18% 100.00% 0.00%
CUP
1 1 33.33% 55.56% 30.16% 11.90% 24.24% 3.70% 100.00%
The MPs’ total possible following relations between or in the parliamentarians groups could reach 100% in each cell table, in the sense that any party could
follow the 100% of the parliamentarians of his/her group or of other group.
IDP no 20 (June, 2015) I ISSN 1699-8154 Journal promoted by the Law and Political Science Department
Eloi PuigEloi Puig
Jose R. Agustina
www.uoc.edu/idp
Universitat Oberta de Catalunya
49
Political Polarisation and Network Leaderships in the Catalan Parliamentarians’ Twitter Network
Eloi PuigEloi Puig
Rosa Borge Bravo, Marc Esteve del Valle
49
parliament Twitter network are more likely to follow
parliamentarians of their own political groups than
parliamentarians of other political groups. That is, the
percentage of MPs’ following relations over the total
possible parliamentarians’ following relations between the
parliamentarian groups in the Catalan parliamentarians
Twitter network show that MPs’ are more likely to follow
parliamentarians of their own groups than parliamentarians
of other political parties. In fact, the average percentage
of the total possible following relations reached by the
116 parliamentarians in 7 parliamentarian groups with
respect to the representatives of their political groups is
91.05% (average of the diagonal zone in Table 1) in the
following-follower network.
Also, this same homophily tendency can be observed
(see Table 2) in the data regarding the retweet network.
More precisely, the average percentage of the total
possible retweets reached by the 116 parliamentarians in
7 parliamentarian groups compared to representatives of
their political groups is 88.84% in the retweet network.
Moreover, the results in Figure 2, Table 1 and Table 2 also
reveal the influence exerted by cleavages in the Catalan
Figure 2 The following centrality of the Catalan parliamentarians’ Twitter Network (1 January 2013–31 March 2014)
*The nodes of the network are the 116 deputies with Twitter account
**The size of the nodes is equivalent to their in-degree centrality in the network
***The colour of the nodes indicates the political party they belong to: orange (CIU), yellow (ERC), red (PSC), blue (PP), green (ICV), brown (C’S)
and violet (CUP)
IDP no 20 (June, 2015) I ISSN 1699-8154 Journal promoted by the Law and Political Science Department
Eloi PuigEloi Puig
Jose R. Agustina
www.uoc.edu/idp
Universitat Oberta de Catalunya
50
Political Polarisation and Network Leaderships in the Catalan Parliamentarians’ Twitter Network
Eloi PuigEloi Puig
Rosa Borge Bravo, Marc Esteve del Valle
50
political system (left/right wing; Catalan nationalist/Non-
Catalan nationalist) on parliamentarians’ communications
flows. Specifically, at a party level, these results show:
1. CiU representatives follow (27.17%) and retweet (4.15%)
more the ERC parliamentarians than the others. Perhaps, the
government agreement reached between these two parties
and their common leaning towards Catalan independence
has some effect on the communication flows of CiU
parliamentarians. In any case, CiU is the party which follow
less its own representatives (71.93%).
2. The nationalist cleavage seems to be guiding the
communication flows of the ERC’s representatives for they
follow (CiU -36.13%- and CUP -33.33%-) and retweet (CiU
-5.0%-) Catalan nationalist parliamentarians more than the
others.
3. MP’s belonging to the PSC follow in a considerable proportion
the rest of the parties (more than 20%) and show a left wing
leaning orientation (ERC -4.37%- and ICV-EUiA - 6%-) on
the retweet network.
4. Parliamentarians belonging to the PP party follow right wing
(CiU -27.84%- and C’s -17.78%-) and Non-Catalan nationalist
(PSC -18.41%- and C’s) parliamentarians more than others,
but they retweet left wing Catalan nationalist parties (ERC
-1.19%- and CUP -0.34%-) almost on a par with non-Catalan
nationalist parties (PSC -0.34%- and C’s -1.36%-).
5. ICV-EUiA’s parliamentarians follow and retweet left wing
parties (ERC, PSC and CUP) more than the others.
6. Representatives in C’s follow Non-Catalan-nationalist parties
(PSC -18.52%- and PP -50.43%-) and retweet right wing
parties more (CiU -0.23%- and PP -0.79%-).
7. CUP MPs follow in higher proportion, first, Catalan nationalist
parties (CiU -33.33%- and ERC-55.56%-) and, second, left
wing parties (PSC -30.16%- and ICV-EUiA -24.24%-), yet they
retweet left wing parties more (ERC -5.33%-, PSC -3.33%-
and ICV-EUiA -13.33%-) than others.
Lastly, regarding the mention network of Catalan
parliamentarians on Twitter, Table 3 shows that political
polarisation is less deep in this network than in the following-
follower and retweet networks. In fact, the average percentage
of total mentions reached by the 116 parliamentarians
of the 7 parliamentarian groups with respect to the
representatives of their political groups is 66.12% (far lower
than 91.05% and 88.48% in the following-follower and
retweet networks). However, the results in Table 3 also point
out the influence exerted by cleavages in the Catalan party
system on mentions between the Catalan representatives:
1. The two parties more frequently mentioned by CiU
parliamentarians are those of a Catalan nationalist party
—ERC (10.72%)— and a Non-Catalan nationalist right wing
party —PP (8.91%)-.
Table 2: Percentage of parliamentarians retweet relations over the total MPs’ retweet relations between parliamentarian groups in
the Catalan parliamentarians’ Twitter network (1 January 2013–31 March 2014)
Parliamentarian
Groups and their
cleavage positioning
Left (1)-
Right (0)
Catalan-nationalist
(1)/ Non Catalan
nationalist (0)
CiU
(35)
ERC
(21)
PSC
(21)
PP
(15)
ICV-
EUiA
(11)
CS
(9)
CUP
(3) Tot al
CiU
0 1 93.83% 4.15% 0.67% 0.56% 0.67% 0.11% 0.00% 100%
ERC
1 1 5.00% 90.62% 0.52% 0.15% 3.56% 0.15% 0.00% 100%
PSC
1 0 2.87% 4.37% 84.52% 2.12% 5.24% 0.50% 0.37% 100%
PP
0 0 0.00% 1.19% 0.34% 96.76% 0.00% 1.36% 0.34% 100%
ICV-EUiA
1 0 0.50% 17.37% 3.11% 0.10% 78.31% 0.30% 0.30% 100%
C'S
0 0 0.29% 0.15% 0.00% 0.73% 0.15% 98.69% 0.00% 100%
CUP
1 1 1.33% 5.33% 3.33% 0.00% 13.33% 0.00% 76.67% 100%
IDP no 20 (June, 2015) I ISSN 1699-8154 Journal promoted by the Law and Political Science Department
Eloi PuigEloi Puig
Jose R. Agustina
www.uoc.edu/idp
Universitat Oberta de Catalunya
51
Political Polarisation and Network Leaderships in the Catalan Parliamentarians’ Twitter Network
Eloi PuigEloi Puig
Rosa Borge Bravo, Marc Esteve del Valle
51
2. ERC representatives mention more frequently left wing
parties, specifically the PSC (3.54%) and ICV (8.46%).
3. PSC parliamentarians focus their mentions on members of
a Catalan nationalist party —ERC (14.53%)— and a left party
wing —ICV (10.01%).
4. MPs in the PP concentrate their mentions on one Catalan-
nationalist and left wing party —ERC (12.19%)— and another
Catalan-nationalist but right wing party —CiU (6.65%).
5. Members of ICV-EUiA clearly focus their mention activity on
ERC (20.98%) and, in lower proportion, the PSC (4.92%) MPs.
6. The most frequently mentioned group by parliamentarians
in C’s is by large the Catalan-nationalist and left wing party
ERC (10.77%).
7. CUP members focus their mention activity on left wing
parliamentarians in ICV (32.21) and ERC (14.32%).
In short, the results in Figure 2, Table 1, Table 2 and
Table 3 show that communication flows between Catalan
parliamentarians’ Twitter network are polarised by party and
ideology. However, the degree of that political polarisation
is higher, first, in the following-follower and, second, in the
retweet network than it is in the mention network, similar
to what other studies (Connover et al., 2011; Yoon and Park,
2012) found. More precisely, Yoon and Park (2012) found that
following other politicians is a social ritual based on dyadic
reciprocity but the strategy of mentioning other politicians
is based on achieving more popularity, regardless of the
ideology of the mentioned politician. In fact, it is in the
mention network where we observe, in higher proportion,
that Catalan members of parliament not only mention
parliamentarians with their same ideological (left/right
wing or Catalan nationalist/Non-Catalan nationalist) stances
but also MPs holding politically opposite points of view.
Therefore we can accept our H1.
Network leaderships
As mentioned in previous sections of this article, the
network leaderships of the Catalan parliamentarians
following-follower, retweet and mention Twitter networks
will be analysed from a double perspective: On the one
hand, we will ascertain the individual characteristics of
the most followed, retweeted and mentioned Catalan
parliamentarians by carrying out regression analysis. On
the other, we will unfold the appearance of new network
leaderships by using network-analysis methods.
In order to understand the individual characteristics of
the most followed, retweeted and mentioned Catalan
parliamentarians, we carried out multiple regression
analyses. While the distribution of the in-degree variable
was normal, the distribution of the other two dependent
variables, mentions received and retweets received were
skewed towards the lowest limit, as it is usual among
Internet users (Hogan, 2008). That is, few parliamentarians
(as shown in Table 4) have a high centrality in these two
networks. Therefore, we transformed these two dependent
variables into two normal distributions (Mentions_LN10
Table 3: Percentage of the total possible MP’s mentions between the parliamentarians’ groups in the Catalan parliamentarians’
Twitter network (1 January 2013–31 March 2014)
Parliamentarian
Groups and their
cleavage positioning
Left (1)-
Right (0)
Catalan-Nationalist
(1)/ Non Catalan
Nationalist (0)
CiU
(35)
ERC
(21)
PSC
(21)
PP
(15)
ICV-
EUiA
(11)
CS
(9)
CUP
(3) Tot al
CiU
0 1 65.99% 10.72% 5.25% 8.91% 6.12% 2.57% 0.44% 100%
ERC
1 1 2.81% 76.21% 3.53% 4.44% 8.46% 3.70% 0.84% 100%
PSC
1 0 6.28% 14.53% 62.94% 2.66% 10.01% 3.00% 0.59% 100%
PP
0 0 6.65% 12.19% 1.85% 72.02% 3.74% 3.05% 0.51% 100%
ICV-EUiA
1 0 1.89% 20.98% 4.92% 3.52% 65.51% 1.89% 1.28% 100%
C'S
0 0 2.49% 10.77% 1.86% 2.56% 2.03% 80.18% 0.11% 100%
CUP
1 1 1.68% 14.32% 6.95% 3.89% 32.21% 0.95% 40.00% 100%
IDP no 20 (June, 2015) I ISSN 1699-8154 Journal promoted by the Law and Political Science Department
Eloi PuigEloi Puig
Jose R. Agustina
www.uoc.edu/idp
Universitat Oberta de Catalunya
52
Political Polarisation and Network Leaderships in the Catalan Parliamentarians’ Twitter Network
Eloi PuigEloi Puig
Rosa Borge Bravo, Marc Esteve del Valle
52
and Retweet_Rec_SQRT) by performing square root and
logarithmic transformations. Then we carried out three
multiple regression analyses to find the parliamentarian’s
characteristics triggering a centrality position in the in-
degree, retweet and mention networks
19
.
The results of the three regression analyses are statistically
significant with an adjusted R-square of 0.494 in the case
of the following network, 0.560 in the retweet network and
0.488 in the mention network. Specifically, the results shown
in Table 4 point to factors explaining the centrality of the
19. In our models the multicollinearity tests do not show the existence of any collinearity between the independent variables.
Table 4: Multiple Regressions on the centrality variables (In-degree; Retweets_recieved_SQRT and Mention_received_LN10) in the
Catalan parliamentarians’ Twitter network
Explanatory variables DV= In-Degree DV= Retweets_Received_SQRT DV= Mentions_Received_LN10
Gender
3.256 -1.102** 0.49
Age
.275** -.009 -.011**
Education level
-1.284 -.113 -.039
Blog
3.732 -.216 -.055
Facebook
-.484 -.292 .054
Total tweets
.004*** .000 -2.435E-005
Total mentions sent
-0.10 .010*** .000
Total mentions received
.000 .007***
Total retweets sent
-0.22 -.002 .001
Total retweets received
.017 .004***
Favourites per day
.471 -.022 -.017
Incumbency
2.172 .361 -.105
Participation in commissions
-.523 .027 .008
Intervention in commissions
-.017 -.001 .001
Intervention in the plenary
.039 -.006 .000
Parliamentarian position (0-1)
9.142** -.462 .214
Position at the legislative commissions (0-1)
.405 -.620 -.104
Political position (0-1)
-.325 -.889 .199*
In-degree
.043* .009**
Out-degree
.353*** .008 -.002
Right-left_0_1
-4.035 .499 -.095
Catalan nationalist/Non-Catalan nationalist (0-1)
.553 .384 -.001
Constant
6.003 3.352 1.891
Adjusted R
2
0.494 0.560 0.488
N
115 115 115
*p
IDP no 20 (June, 2015) I ISSN 1699-8154 Journal promoted by the Law and Political Science Department
Eloi PuigEloi Puig
Jose R. Agustina
www.uoc.edu/idp
Universitat Oberta de Catalunya
53
Political Polarisation and Network Leaderships in the Catalan Parliamentarians’ Twitter Network
Eloi PuigEloi Puig
Rosa Borge Bravo, Marc Esteve del Valle
53
Catalan parliamentarians in the following network, such as
their age (having more followers increases with age), out-
degree (number of parliamentarians someone is following),
total number of tweets and their parliamentarian position
(having a parliamentarian position increases the number of
followers). Moreover, as shown in Table 4, the characteristics
explaining the centrality position of Catalan parliamentarians
in the retweet network are: gender (being a woman diminishes
the likelihood of being central), mentions sent, in-degree
and the mentions received. Lastly, regarding the centrality
position in the mention network, Table 4 shows that age (the
number of mentions decreases with age), political position
(mentions increases if a parliamentarian holds a political
position), retweets received and in-degree are explanatory
factors of the dependent variable. In brief, analysis of the
factors explaining the parliamentarians’ centrality position
in the following, retweet and mention networks points to the
relevance of relational variables; parliamentarians holding a
centrality position in the three networks are those who are
the most followed and who at the same time concentrate
more retweets and mentions than others. However, holding
an official responsibility in the party or in Parliament
increases the number of following parliamentarians and
the number of mentions received, yet it has no influence
on the number of times a parliamentarian is retweeted.
The results observed in the regression analyses can also be
corroborated by the results shown in Table 5. According to
these results, parliamentarians holding a political position
in their parties are only receiving 26.22% of the retweets
(1,585 retweets) but taking the lead in the following and
mention networks. More precisely, as shown in Table 5,
parliamentarians holding a political position are the most
followed (average of In-degree= 41.86) and mentioned
(8,309) than all the others. Furthermore, among these
parliamentarians, it is important to note the dominant position
reached by the 6 party leaders in the in-degree (average
of 55.14), retweet (1,107) and mention (6,333) networks.
Nevertheless, for a deeper understanding of the centrality
position of the Catalan parliamentarians based not only
on the quantity of followers, retweets and mentions
that they have or receive but on the number of the
parliamentarians following, retweeting or mentioning a
particular deputy, we analysed the centrality position of
the Catalan parliamentarians in the following, retweet and
mention Twitter networks by using the total number of
parliamentarians following, retweeting and mentioning a
specific representative. In this way, we are able to better
detect the parliamentarians whose popularity is spread
between many parliamentarians and not only concentrated
in a few parliamentarians that are very active in retweeting,
mentioning and following others but only their own party
peers or leaders. The results obtained by doing that
network twist (shown in Table 6 and Table 7) are different
compared with those observed in the regression analysis
and in Table 5. More precisely, Table 6 shows that, of the
21 representatives
20
with a higher in-degree, 47.61 % hold
a political position (4 are party leaders) while the other
52.39 % do not. Also, regarding the mention network (see
Figure 3 and Table 6), among the 21 parliamentarians with
a higher mention centrality degree, 33.33 % hold a political
position (6 are party leaders) while the other 66.67 % do
not. But the most surprising figure comes when analysing
the retweet network (see Figure 4 and Table 6); of the 21
representatives with the highest retweet centrality degree,
only one holds a political position (a party leader).
In fact, the data showed below points to the appearance
of new network leaderships who do not need to hold
Table 5: In-degree (following), retweets and mentions in the parliamentarians’ Twitter network
Type of network All parliamentarians (115) Political position* (22) Only party leaders (6)
In-degree (following) (average) 38.93 41.86 55.14
Retweets (total) 6,045 1,585 1,107
Mentions (total) 19,507 8,309 6,333
*Political position: Parliamentarians holding a relevant political position (party leaders or mayors)
20. This number of representatives has been chosen as it approaches 20% of the total of the nodes of the network and we believe that it helps
to show the centrality position of some parliamentarians in the following, retweet and mention networks.
IDP no 20 (June, 2015) I ISSN 1699-8154 Journal promoted by the Law and Political Science Department
Eloi PuigEloi Puig
Jose R. Agustina
www.uoc.edu/idp
Universitat Oberta de Catalunya
54
Political Polarisation and Network Leaderships in the Catalan Parliamentarians’ Twitter Network
Eloi PuigEloi Puig
Rosa Borge Bravo, Marc Esteve del Valle
54
Table 6: Political position characteristics of the 21 Catalan parliamentarians with highest in-degree (following), retweet and
mentions network centrality in the Catalan parliamentarians’ Twitter network
Type of network Parliamentarians Political position* Party leaders (6)
In Degree (following) (average) 52.39 47.6 1 4
Retweets (%) 95.24 4.76 1
Mentions (%) 66.67 33.33 6
*Political position: the parliamentarians holding a relevant political position (party leaders or mayors)
Figure 3 The mention centrality of the Catalan parliamentarians’ Twitter network (1 January 2013–31 March 2014)
*The nodes of the network are the 116 deputies with Twitter account and those with labels are the 20% of parliamentarians with a higher mention
degree of whom 8 representatives out of 24 hold a relevant political position (@alber_rivera; @dolorscamats; @herrerajoan; @pere_navarro; @
higiniaroig; @aliciascamacho; @junqueras; @martarovira).
**The size of the nodes is equivalent to their mention centrality in the network
***The colour of the nodes is equivalent to the political party that they belong to: orange (CIU), yellow (ERC), red (PSC), blue (PP), green (ICV),
brown (C’S) and violet (CUP)
IDP no 20 (June, 2015) I ISSN 1699-8154 Journal promoted by the Law and Political Science Department
Eloi PuigEloi Puig
Jose R. Agustina
www.uoc.edu/idp
Universitat Oberta de Catalunya
55
Political Polarisation and Network Leaderships in the Catalan Parliamentarians’ Twitter Network
Eloi PuigEloi Puig
Rosa Borge Bravo, Marc Esteve del Valle
55
Figure 4 The retweet centrality position of the Catalan parliamentarians’ Twitter network (1 January 2013–31 March 2014)
*The nodes of the network are the 116 deputies with Twitter account and those with labels the 10% of parliamentarians with higher retweet
centrality of whom there is only one with a relevant political position, @martarovira (Secretary General of ERC).
**The size of the nodes is equivalent to their retweet centrality in the network
***The colour of the nodes is equivalent to the political party they belong to: orange (CIU), yellow (ERC), red (PSC), blue (PP), green (ICV),
brown (C’S) and violet (CUP)
relevant political positions or being a party leader to be
central in the Twitter networks of the Catalan parliament,
a fact that is also corroborated when we analyse the
parliamentarian position of the Catalan representatives.
In this vein, Table 7 shows that of the 21 representatives
with a higher in-degree, 42.85 % hold a parliamentarian
position (5 are party leaders) while the remaining 57.15 %
are not. Also, with respect to the mention network, of the
21 parliamentarians with the highest mention centrality
degree, 14.28 % hold a parliamentarian position (1 is a party
leader) while the remaining 85.72 % do not. But again, the
most surprising case is that of the retweet network, where
only 1 parliamentarian holds a parliamentarian position (and
he is also a party leader).
Table 7: Parliamentarian position characteristics of the 21 Catalan parliamentarians with highest in-degree (following), retweet and
mentions network centrality in the Catalan parliamentarians’ Twitter network
Type of network Parliamentarians Parliamentarian position* Party leaders (6)
In degree (following) (average) 57.15 42.85 5
Retweets (%) 95.24 4.76 1
Mentions (%) 85.72 14.28 1
* Parliamentarian position: the parliamentarians holding a relevant political position in the Catalan Parliament (Spokesperson or President of the
parliamentarian group, Secretary, Vice-President or President of the Parliament).
IDP no 20 (June, 2015) I ISSN 1699-8154 Journal promoted by the Law and Political Science Department
Eloi PuigEloi Puig
Jose R. Agustina
www.uoc.edu/idp
Universitat Oberta de Catalunya
56
Political Polarisation and Network Leaderships in the Catalan Parliamentarians’ Twitter Network
Eloi PuigEloi Puig
Rosa Borge Bravo, Marc Esteve del Valle
56
The data presented above show that official party leaders
with political or parliamentarian positions are dominant in
the following and mention networks but not in the retweet
relationships. Some authors have pointed out that the
former networks are subjected to partisan pressures and
attempts to reach more followers (Yoon and Park, 2012)
whereas the later could be more ruled by information
exchanges and personal interests. Consequently, the
data and the network analysis presented below seem
to clearly show that “communication flows of Catalan
parliamentarians’ Twitter Network are empowering party
leaders but also paving the way for the appearance of
Network Leaderships” (H2).
Conclusions
This article has shown that Twitter plays a double
game, it enclosures communication flows among party
representatives and empowers party leaders but it also
facilitates dialogue between MPs from different political
groups and paves the way for the appearance of network
leaderships.
From a political communication perspective, in line with
the results found by Yardi and Bold (2010), Conover et
al.,(2011), Himelboim et al., (2012) and Hsu and Park
(2012), our research points out some evidence for the
argument that communication flows among the Twitter
network of Catalan parliamentarians’ are polarised by
party affiliation and ideology. However, as Conover et al.,
(2011), Gruzd (2012) and Yoon and Park (2014) have already
corroborated, our results also confirm that the degree of
political polarisation is higher in the following-follower
and the retweet networks than in the mention network.
More precisely, the data sustain that, in the mention
network, Catalan parliamentarians not only communicate
with parliamentarians sharing their similar ideological
stances (left/right wing or Catalan nationalist/Non-Catalan
nationalist cleavages) but also with MPs holding politically
opposite points of view.
However, regarding the analysis of network leaderships, the
results related to the volume of the communication flows
between the parliamentarians (number of parliamentarians
following another parliamentarian, total number of
retweets and total number of mentions) and those linked
to the distribution of the communication flows (number
of parliamentarians following another parliamentarian,
total number of retweets a parliamentarian receives from
the other parliamentarians and total number of mentions
a parliamentarian receives from other parliamentarians)
reveal different dimensions of the Catalan parliamentarians’
Twitter network leaderships. The regression analyses have
pointed out that parliamentarians holding a centrality
position in the three networks are those who are the most
followed, hold relevant political positions and, at the same
time, concentrate more parliamentarian followers, retweets
and mentions than others. However, the network analysis
of the centrality position of the parliamentarians based
on the in-degree, mentions and retweets that they receive
from other nodes (representatives) in the three networks
has revealed that new network leaders (who do not hold a
political or a parliamentarian position) are concentrating the
majority of retweets and taking the lead in the following and
mention networks. This last result is more relevant since it
takes into account the number of different parliamentarians
retweeting and mentioning another parliamentarian and
not the gross volume of retweets and mentions that is
frequently affected by few parliamentarians giving heavy
support to their own peers.
Could our research be extended to other contexts? Our study
object is the directions of parliamentarians’ communication
flows on Twitter and we conceive parliamentarians’ online
communications as comparable entities across countries.
Consequently, our argument could then be applicable to
all kinds of political contexts. Moreover, the statistical and
network analyses used in the research are also prone to be
replicated in other studies.
However, a number of complementary issues remain to be
tested;
1. It would be necessary to explain exactly why the following,
retweet and mention networks are different in terms of
ideological polarisation and leadership. It seems that the
parliamentarians strategically decide to behave differently
in each of these layers of twitter-based communications.
2. It is reasonable to assume that other characteristics shaping
the parliamentarians’ communication flows play a role in
explaining their directions. For instance, the electoral
district of the parliamentarians, since relationships could
rise up among representatives of the same constituency
(Lassen and Brown, 2010), the core goals sought by parties
(Römmele, 2003) or the predominant issues on the Catalan
political agenda (Esteve and Borge, 2013).
IDP no 20 (June, 2015) I ISSN 1699-8154 Journal promoted by the Law and Political Science Department
Eloi PuigEloi Puig
Jose R. Agustina
www.uoc.edu/idp
Universitat Oberta de Catalunya
57
Political Polarisation and Network Leaderships in the Catalan Parliamentarians’ Twitter Network
Eloi PuigEloi Puig
Rosa Borge Bravo, Marc Esteve del Valle
57
3. It may be useful to identify not only the directions of the
communication flows between Catalan parliamentarians,
but also the content of the information transmitted through
Twitter interactions (Yoon and Park, 2012).
4. Finally, more studies extending the time span of our article and
carrying out comparative analysis between parliamentarians
of different and similar political systems could add powerful
insights to how MPs deal with social media.
References
BALCELLS, J.; CARDENAL, A. S. (2013). “Internet and electoral competition: The Case of Esquerra
Republicana de Catalunya”. Revista Española de Investigaciones Sociológicas, no. 141, pp. 3-28.
BENACH, E. (2010). Política 2.0. Barcelona: Bromera.
BIMBER, B. (2014). “Digital media in the Obama campaigns of 2008 and 2012: Adaptation to the
personalized political communication environment” [Special issue]. Journal of Information Technology
& Politics, vol. 11, no. 2, pp. 130–150. g/10.1080/19331681.2014.895691>
CARDENAL, A. S. (2011). “Why mobilize support online? The paradox of party behaviour online”. Party
Politics, vol. 19, no. 1, pp. 83–103. doi.org/10.1177/1354068810395059>
CASTELLS, M. (2009). Communication Power. New York, NY: Oxford University Press.
CHI, F.; YANG, N. (2010). “Twitter adoption in Congress.Review of Network Economics, vol. 10, no. 1,
pp. 1–49.
CONOVER, M.; RATKIEWICZ, J.; FRANCISCO, M.; GONÇALVES, B.; FLAMMINI, F. (2011). “Political polarization
on Twitter”. In Proceedings of the Fifth International AAAI Conference on Weblogs and Social Media,
AAAI, Barcelona.
ESTEVE, M.; BORGE, R. (2014). “Are social media changing party politics? Brokers among the members
of the Catalan parliament Twitter network”. In: J. BALCELLS, A. CERRILLO, M. PEGUERA, I: PEÑA-
LÓPEZ, M. J. PIFARRÉ, M. (coord). Internet, Law and Politics. A Decade of Transformations. Paper
presented at the 10th International Conference on Internet, Law and Politics, Universitat Oberta de
Catalunya, Barcelona, 3-4 July (pp. 745–766). Barcelona: Huygens. symposia/
idp2014/proceedings/>
ESTEVE, M., BORGE, R. (2013). “Abriendo brechas: centralización de las decisiones e interacción en línea
en CIU, ERC y el PSC” [Opening gaps: centralization of the decisions and online interaction in CIU, ERC
and PSC]. IDP. Revista de Internet, Derecho y Política, 17, pp. 45–62. Universitat Oberta de Catalunya.
ojs/index.php/idp/article/view/n17>
GRUZD, A. (2012). “Investigating a political polarization on Twitter: A Canadian perspective”. Paper
presented at the Internet, Politics, Policy 2012: Big Data, Big Challenges? Oxford, UK, September 20-21.
GONZALEZ-BAILON, S.; WANG, N. (2013). The bridges and brokers of protest campaigns using social
media. srn.com/sol3/papers.cfm?abstract_id=2268165>
GOLDBECK, J.; GRIMES, J. M.; ROGERS, A. (2010). “Twitter use by the U.S. Congress”. Journal of the
American Society for Information Science and Technology, vol. 61, no. 8, pp. 1612–1621.
HIMELBOIM, I.; Mc CREERY, S.; SMITH, M. (2013). “Birds of a feather tweet together: Integrating network
and content analysis to examine cross-ideology exposure on Twitter.” Journal of Computer-Mediated
Communication, vol. 18, no. 2, pp. 154-174. dx.doi.org/10.1111/jcc4.12001>
HOGAN, B. (2008). “Analysing social networks via the Internet.” In: R. FIELDING and G. BLANK
(eds). The handbook of online research methods. Thousand Oaks: Sage, pp. 141–161. dx.doi.
org/10.4135/9780857020055.n8>
HSU, H. L., PARK, H. W. (2012). “Mapping online social networks of Korean politicians”. Government
Information Quartely, no. 29, pp. 169–181. dx.doi.org/10.1016/j.giq.2011.09.009>
IDP no 20 (June, 2015) I ISSN 1699-8154 Journal promoted by the Law and Political Science Department
Eloi PuigEloi Puig
Jose R. Agustina
www.uoc.edu/idp
Universitat Oberta de Catalunya
58
Political Polarisation and Network Leaderships in the Catalan Parliamentarians’ Twitter Network
Eloi PuigEloi Puig
Rosa Borge Bravo, Marc Esteve del Valle
58
LASSEN, D.; BROWN, A. (2010). “Twitter: The electoral connection?” Social Science Computer, 000(00),
pp. 1–18.
McPHERSON, M.; SMITH-LOVIN, L.; COOK, J. M. (2001). “Birds of a feather: Homophily in social networks”.
Annual Review of Sociology, no. 27, pp. 415–444. org/10.1146/annurev.soc.27.1.415>
PADRÓ-SOLANET, A.; CARDENAL, A. S. (2008). “Partidos y política en Internet: Un análisis de los websites
de los partidos políticos catalanes” [Parties and policy in the Internet: An analysis of the Catalan
parties’ webpages]. IDP Revista de Internet, Derecho y Política, no. 6. www.uoc.edu/idp/6/dt/
esp/padro-solanet_cardenal.pdf>
PUTNAM, R. (2000). Bowling alone: The collapse and revival of American community. New York, NY:
Simon & Schuster. g/10.1145/358916.361990>
RÖMMELE, A. (2003). “Political parties, party communication and New Information and Communication
Technologies.Party Politics. Special Issue Party Politics on the Net, 9 (1), pp. 7–20. dx.doi.
org/10.1177/135406880391002>
SMITH, M., RAINIE, L., HIMELBOIM, I.; SHNEIDERMAN, B. (2014). “Mapping Twitter topics networks: From
polarized crowds to community clusters”.
s-uns.blogspot.com.es/2014/04/smith-et-al-2014-mapeando-comunidades.html>
THAMM, M., BLEIER, A. (2013). “When politicians tweet: A study on the members of the German Federal
Diet” In: ACM Web of Science, Paris. websci13.org/>
VERWEIJ, P. (2012). “Twitter links between politicians and journalists”. Journalism Practice, 6 (5-6), pp.
680–691. org/10.1080/17512786.2012.667272>
WILLIAMS, C. B.; GULATI, G. J. (2010). Communicating with constituents in 140 characters or less. Working
Papers (43). Retrieved from: http://opensiuc.lib.siu.edu/pn_wp/43
YARDI, S.; BOYD, D. (2010). “Dynamic debates: An analysis of group polarization over time on
Twitter”. Bulletin of Science, Technology and Society, vol. 30, no. 5, pp. 316–327. dx.doi.
org/10.1177/0270467610380011>
YOON, H. Y.; PARK, W. H. (2014). “Strategies affecting Twitter based networking pattern of South Korean
politicians: social network analysis and exponential random graph model”. Quality and Quantity, vol.
48, no. 1, pp. 409–423. dx.doi.org/10.1007/s11135-012-9777-1>
IDP no 20 (June, 2015) I ISSN 1699-8154 Journal promoted by the Law and Political Science Department
Eloi PuigEloi Puig
Jose R. Agustina
www.uoc.edu/idp
Universitat Oberta de Catalunya
59
Political Polarisation and Network Leaderships in the Catalan Parliamentarians’ Twitter Network
Eloi PuigEloi Puig
Rosa Borge Bravo, Marc Esteve del Valle
59
Appendix I
Description of variables and codification
Variable Codification
Gender 0= Male 1=Female
Age 28 to 66 years old
Level of education
1= Undergraduate
2= Bachelor
3= Master’s degree or PhD
Blog 1= yes
0= no
Facebook 1= yes
0= no
Total tweets From 0 to 3244.
Total mentions sent From 0 to 773.
Total mentions received From 0 to 1274.
Total retweets sent From 0 to 227.
Total retweets received From 0 to 361.
Favourites per day Average of the Parliamentarians’ favorites per day. Ranges from 0 to 26.07.
Incumbency
(2006, 2010 or 2012 Legislatures)
1= One Legislature
2= Two Legislatures
3= Three Legislatures
Number of commissions in which the parliamentarian participates From 0 to 16 commissions
Number of interventions in commissions From 0 to 445
Number of interventions in the Plenary From 0 to 262
Political position in Parliament and the parliamentarian group
1 - Spokesperson of the parliamentarian group
2 - President of the parliamentarian group
3- Parliament Secretary or Vice-President of the Parliament
4 - President of the Parliament
Recoded 0-1 (O= no position; 1= 1 to 4)
Political position in Parliament commissions
1 - Secretary
2 - Vice-President
3 - President
Recoded 0-1 (O= no position; 1= 1 to 3)
Political position in the political party
1 - Local mayor
2 - President of the party
Recoded 0-1 (O= no position; 1= 1 to 2)
In-degree Number of followers a parliamentarian’s Twitter holds. Ranges from 6 to 77.
Out-degree Number of parliamentarian Twitter accounts each parliamentarian is following. Ranges
from 0 to 115.
Right-left 1= Right
0= Left
Catalan nationalist and non-nationalist 1= Yes
0= No
IDP no 20 (June, 2015) I ISSN 1699-8154 Journal promoted by the Law and Political Science Department
Eloi PuigEloi Puig
Jose R. Agustina
www.uoc.edu/idp
Universitat Oberta de Catalunya
60
Political Polarisation and Network Leaderships in the Catalan Parliamentarians’ Twitter Network
Eloi PuigEloi Puig
Rosa Borge Bravo, Marc Esteve del Valle
60
Recommended citation
BORGE BRAVO, Rosa; ESTEVE DEL VALLE, Marc (2015). “Political Polarisation and Network Leaderships
in the Catalan Parliamentarians’ Twitter Network”. IDP. Revista de Internet, Derecho y Política. No. 20,
pp. 43-60. UOC [Accessed: dd/mm/yy]
/idp/article/view/n20-borge-delvalle/n20-borge-delvalle-pdf-en>
org/10.7238/idp.v0i20.2641>
The texts published in this journal, unless otherwise indicated, are subject to a Creative
Commons Attribution-NoDerivativeWorks 3.0 Spain licence. They may be copied, distributed
and broadcast provided that the author, the journal and the institution that publishes
them (IDP Revista de Internet, Derecho y Política; UOC)are cited. Derivative works are not
permitted. The full licence can be consulted onhttp://creativecommons.org/licenses/by-
nd/3.0/es/deed.en.
About the authors
Rosa Borge Bravo
rborge@uoc.edu
Resident researcher at the Internet Interdisciplinary Institute
Associate Professor of Political Science (Universitat de Catalunya)
http://in3.uoc.edu/opencms_portalin3/opencms/en/investigadors/list/borge_bravo_rosa
Internet Interdisciplinary Institute (IN3-UOC)
Av. Carl Fiedrich Gauss, 5
08860 Castelldefels
Marc Esteve del Valle
marcesteve2015@gmail.com
Researcher and PhD grant-holder
Doctoral Programme in Information and Knowledge Society
Internet Interdisciplinary Institute (IN3-Open University of Catalonia)
Internet Interdisciplinary Institute (IN3-UOC)
Av. Carl Fiedrich Gauss, 5
08860 Castelldefels

VLEX utiliza cookies de inicio de sesión para aportarte una mejor experiencia de navegación. Si haces click en 'Aceptar' o continúas navegando por esta web consideramos que aceptas nuestra política de cookies. ACEPTAR