How Second-Order Is the Regional Level? An Analysis of Tweets in Simultaneous Campaigns

AuthorSteven Lannoo,Evelien D’heer,Nicolas Bouteca
Published date01 December 2017
Date01 December 2017
DOIhttp://doi.org/10.1177/0032321717697346
Subject MatterArticles
https://doi.org/10.1177/0032321717697346
Political Studies
2017, Vol. 65(4) 1021 –1039
© The Author(s) 2017
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/0032321717697346
journals.sagepub.com/home/psx
How Second-Order Is the
Regional Level? An Analysis
of Tweets in Simultaneous
Campaigns
Nicolas Bouteca, Evelien D’heer and
Steven Lannoo
Abstract
This article puts the second-order theory for regional elections to the test. Not by analysing
voting behaviour but with the use of campaign data. The assumption that regional campaigns
are overshadowed by national issues was verified by analysing the campaign tweets of Flemish
politicians who ran for the regional or national parliament in the simultaneous elections of 2014.
No proof was found for a hierarchy of electoral levels but politicians clearly mix up both levels in
their tweets when elections coincide. The extent to which candidates mix up governmental levels
can be explained by the incumbency past of the candidates, their regionalist ideology, and the
political experience of the candidates.
Keywords
second-order elections, Twitter, campaign, elections, Belgium
Accepted: 30 December 2016
Regional elections have often been perceived as inferior elections (Schakel and Jeffery,
2013: 324) which voters primarily interpret through the prism of politics at the national
level (Erk and Swenden, 2010). According to some authors, this subordination is prob-
lematic. By putting the sub-states in the shade of the national level, there is no room for
regional peculiarities. Moreover, it would mean a profound weakening of democracy. By
looking to regional elections through a national lens, regional and national issues are
mixed up. As a result, it is difficult to hold politicians accountable for the policy decisions
which they make (Anderson, 2006: 459; Fabre and Maddens, 2009).
Department of Political Science, Ghent University, Ghent, Belgium
Corresponding author:
Nicolas Bouteca, Department of Political Science, Ghent University, Universiteitstraat 8, Ghent 9000,
Belgium.
Email: Nicolas.Bouteca@Ugent.be
697346PSX0010.1177/0032321717697346Political StudiesBouteca et al.
research-article2017
Article
1022 Political Studies 65 (4)
The popular idea that the regional elections are less important than national elections
stems from the second-order election (SOE) model, which Karlheinz Reif and Hermann
Schmitt (1980) originally designed for the European elections. According to their analy-
sis, regional elections do not have their own dynamics and should be seen as a function of
politics at the centre. This means that governmental levels are mixed up within federal
states and that the national level is the first-order political arena. But to what extent is this
true? According to some scholars of regional electoral politics, the SOE-model has been
copied too wholesale to elections within federal states (Erk and Swenden, 2010: 194;
Schakel and Dandoy, 2013b: 8; Schakel and Jeffery, 2013: 326).
In this article, we will examine this rank order of elections in federal states and this
will be done in an innovative way. The SOE-model will not be tested by looking at vot-
ing behaviour (Schakel and Dandoy, 2013b; Schakel and Jeffery, 2013) or by analysing
how media report on elections (De Vreese et al., 2007), as is usually the case. Instead,
we move the attention to the third factor that is important for the understanding of this
phenomenon: the parties and politicians. The way that they communicate during election
campaigns also determines how media and voters think about the rank order of elections
(Reif and Schmitt, 1980: 14). For example, if politicians frame regional elections as a
national test, the media and voters will be inclined to do so too. Thus, the analysis of
how politicians campaign may be crucial in our understanding of the election process
in multilevel states.
Up to now, campaigns have only rarely been analysed from a second-order perspec-
tive. There is some pioneering research to the second-order character of campaigns for
the elections to the European Parliament (De Vreese, 2003, 2009; De Vreese et al., 2007;
Gagatek, 2010), but the applicability of the SOE theory for regional elections has hardly
been tested with campaign data. Scholars who refer to this question (e.g. De Winter et al.,
2006; Versmessen, 1995) often state that regional campaigns are dominated by national
politics without systematically analysing this ‘campaign dominance’. Moreover, the
scarce systematic campaign studies which exist on this subject are usually descriptive or
make use of subjective perceptions on campaign dominance instead of analysing the
‘objective’ campaign behaviour of parties or politicians (Tenscher, 2014: 39).
In this article, the second-orderness of the regional level is tested by a content analysis
of the tweets that Flemish political candidates sent during the campaign for the 2014
regional and federal elections in Belgium. Because the elections were held on the same day,
they provide a unique opportunity to test the rank order of elections (Dandoy, 2013: 966).1
There are basically three reasons why analysing tweets is interesting for the study of
‘campaign dominance’. First of all, Twitter has become an important campaign instru-
ment (Kruikemeier, 2014: 131), especially because it is increasingly used as a source for
journalists. Hence, what politicians say on Twitter does not only impact on their followers
but also leads to extra attention in media that reach a more diverse public, such as televi-
sion and newspapers (Broersma and Graham, 2012; Strömbäck and Van Aelst, 2013).
Second, tweets are more interesting for this kind of content-driven analysis of the SOE-
model than other sources which can be used for content analysis. Twitter is a more
dynamic campaign instrument than manifestos or leaflets. The latter are often designed
before the actual campaign which makes them less convenient to study the real campaign.
Campaign speeches do not have this problem, but they reach fewer people and are far less
available than tweets. Moreover, by looking at tweets instead of the media, we circum-
vent the mediating role of journalists and are able to concentrate on the behaviour of poli-
ticians as such (Broersma and Graham, 2012: 408). Finally, there are several practical

To continue reading

Request your trial

VLEX uses login cookies to provide you with a better browsing experience. If you click on 'Accept' or continue browsing this site we consider that you accept our cookie policy. ACCEPT