The eu’s Legitimacy in the Eye of the Beholders

4.3 – The Mass Media: A Positive or Negative Influence on Legitimacy?

The relationship between the mass media and the EU’s legitimacy is the final subject of this chapter. Since newspapers and television reach a vast audience, they are an almost indispensible means for politicians to get their message across. As already mentioned the mass media’s functioning is criticised. Let us see how they cause their own deficits, but also how one can overcome them.

There are several ways in which the mass media is a flawed mediator and may contribute to the legitimacy deficit. Bursens and Baetens distinguish three ‘media deficits’.⁶⁷ First, there is a knowledge deficit. The EU is largely unknown to the public, which leads to an inability to evaluate or a tendency to favour negative evaluations of its legitimacy.⁶⁸ Secondly, the motivational deficit, which means that if the EU is not presented as relevant, people will not be motivated to act in anyway – whether voting or seeking information. Thirdly, there is the comprehension deficit. This is a specific form of knowledge, namely whether people actually understand the European political system. The three deficits are closely related. To a certain extent the comprehension is part of the knowledge deficit and the motivation deficit a result of those two.

Fortunately, Bursens and Baetens develop several ways in which these media deficits can be overcome. First, there should be more EU-coverage – quantity – and on a more regular basis – regularity. Secondly, episodic and thematic coverage should be a balanced. Thirdly and fourthly, coverage should present the EU policies and the EU institutions in more detail. Finally, there should be more nationalised and personified coverage (2004: 12-14; 27). The first two methods are more focussed on news-coverage than editorials, although the quantity of contributions can be influenced.

The EU’s novel system of governance with its complex decision-making procedures, institutional design, terminologies and its distance from the people are hard to ‘sell’ (De Vreese 2003: 65-67). ‘Mass media coverage remains useless if people don’t read, hear, remember or understand the information offered by newspapers’ (Bursens and Baetens 2004: 13). According to many scholars, the EU will catch more attention if becomes more comprehensible and is presented attractively (c.f. Ibid., 14). The question is how to formulate such a message (Perse 2001: 132).⁶⁹ The two final methods might offer a solution for the third and fourth requirement. First, give the EU ‘a face’ – personification – and secondly, explain the consequences for the national situation – nationalisation.

In conclusion: The mass media is clearly not a perfect mediator. It has shortcomings – possible biased coverage and media deficits – but also offers opportunities – the emerging European public discourse. Further, there are ways to overcome the inherent media deficits. The question is whether contributors to the public discourse use these methods in their contributions and whether they actually work.

Chapter 5 – Methodology: From Strategy to Hypotheses

The methodological issues concerning this research are addressed in this chapter. First, the research strategy is introduced. Then the methods of inquiry are presented including a description of the actual research process. This is followed by a discussion of its reliability, validity and limitations and how these issues were addressed. Finally, the concepts from the theoretical framework are operationalised and three hypotheses formulated.

5.1 – The Research Strategy

The strategy chosen is a comparative case study. This is a multiple-case study, which ‘investigates a particular contemporary phenomenon in depth and within its real life context’ by comparing the findings of several cases (Yin 2009: 18-19). The focus is the (implicit) conceptualisation of legitimacy in discourse when assessing the EU’s legitimacy and the consequences for the evaluation thereof. The locus is the public discourse within three countries: the Netherlands, the United Kingdom and France. The three main reasons for selecting⁷⁰ these three countries were, first of all, that they are all Member States of the EU. Secondly, the three countries (would) have voted no in the referendum on the ‘Constitution for Europe’, which indicates a perceived legitimacy deficit. Thirdly, they represent different political-administrative cultures, which could lead to differences in perspectives. This is Siedentop’s analysis (2000), in which he compares the United Kingdom, France and Germany. Germany is replaced by the Netherlands, because in this way the perspective of a relatively small country is included. This is a defensible replacement, because the German and Dutch attitude toward the EU and administrative culture are comparable (Ibid.).

5.2 – The Methods of Inquiry

The main source of evidence into the EU’s legitimacy deficit are opinionated articles published during two periods of two year in two national newspapers per case.⁷¹ This is a proven source for the analysis of public discourse (e.g. Bursens and Helsen 2004; Liebert and Trentz 2008; Peters et al. 2005). Articles from the national newspapers (mass media) are a type of document. This source has the following strengths and weaknesses (Yin 2009: 102):

Strength	Weaknesses
Stable – can be reviewed repeatedly	Retrievability – can be difficult to find
Unobtrusive – not created as a result of the case study	Biased selectivity – if collection is incomplete
Exact – contains exact names, references, and details of an event	Reporting bias – reflects (unknown) bias of author
Broad Coverage – long span of time, many events, many settings	Access – may be deliberately withheld

Figure 5.1: Strengths and Weakness of Documentation
‘Public discourse, however, represents only a specific segment of public communication. Research on public discourse focuses not on information, or ‘news’, but on public commentary, interpretation, and debate’ (Peters et al. 2005: 141; Emphasis in original). The aim of the research is to analyse public discourse therefore only opinionated articles were selected.

The articles were retrieved from two periods of two years. Period one runs from the first of January 2000 till the first of January 2002 and period two from 30^th of May 2004 till the same date two years later: 2006. A major ‘EU event’ took place in the middle of each period. In period one the Euro was introduced (January 1, 2001) and in period two France (May 29, 2005) and the Netherlands (June 1, 2005) voted no to the Constitution for Europe.

For each case two qualitative newspapers⁷² were analysed during both periods. For the Netherlands the newspapers were the NRC Handelsblad and De Volkskrant. The Independent and The Guardian were selected for the United Kingdom. And finally for France Le Figaro and Libération were analysed. The newspapers were selected on the basis of Oosterbaan and Wansink (2008).⁷³

The newspapers were accessed via the databank: Academic Nexus Lexus.⁷⁴ A keyword search was performed in this databank during both periods in all six newspapers.⁷⁵ Access was thus not an issue. Further, a pilot search was conducted for both periods in the Dutch newspapers, which verified that retrievability was not a problem either.

The first method of inquiry is a quantitative content analysis. The articles were subject to a close-reading analysis. During this process observations on the legitimacy of the EU were extracted from the articles. These articles and observations were coded, which we will return to in more detail in the operationalisation. This quantitative data was entered into the statistical analysis computer program SPSS version 17 in order to accurately calculate relevant statistics.

A possible method for determining significance is the Pearson Chi-squire. This task is a relatively simple with the help of SPSS. Almost always the result was: p is less than 0,05 even less than 0,005, and thus significant. This though feels like an empty ‘victory’ for it stays unclear, which facet is related to which evaluation? And whether this relationship is strong? Instead of any of the possible significance tests a different approach was taken. The distribution of judgements within the facets is compared to the distribution of judgements within the overall discourse.

For example, the distribution of observations in the Dutch discourse with regards to the evaluation of the EU’s legitimacy is 70% negative and 30% positive – This is not the case, but for argument’s sake let us say it is. We then turn to the two dimensions. We find that of the observations using a universalist concept of legitimacy 80% is negative and only 20% is positive, whilst of the observations conceptualising legitimacy using a nationalist concept 60% is negative and 40% is positive. Then we can conclude that both conceptualisations lead to a negative perception, but observations using a universalist dimension are more negative than expected, whilst those using a nationalist dimension are less negative than expected.

The analysis might show an author’s bias, but more importantly it might not be completely reliable, because of the ‘transformation’ of qualitative data into quantitative data. A metaphor might help explain the use and possible shortcomings of a quantitative analysis. The quantitative analysis of the articles might be compared to coding 500 portrait-pictures. One analyses the hair colour, skin colour and eye colour of each picture. After analysis one ends up with 1500 observations in ‘groups’ of three. This quantitative approach was the only way to effectively deal with such a large number of observations, but how does one know the data is reliable?

Maybe the combination black, yellow, black is very common and this might lead one to suspect that the population contains a high percentage of people of Asian descent, but during the data analysis it became clear though that the vast majority was, for example, Japanese. Further, people might have dyad their hair – whose hair is originally blue? – in which case the percentage of people of Japanese descent might be even higher than the statistics indicate. The quantitative data was too abstract and thus dismissed these conclusions. In order to prevent such a misrepresentation and check whether the quantitative results are reliable, they are placed back into context.

This brings us to the second method of inquiry: the qualitative content or narrative analysis. After presenting the quantitative data, some flesh will be put back onto the bones, so to say, by placing the data in its original context. During the close-reading and SPSS codification process notes were made – mental, digital and in ink – on the qualitative differences between the discourses, which form the basis of this analysis.

The third method of inquiry is the secondary analysis, which Robson describes as: “[the] analysis of data already collected in some other than the present study” (2002: 552). Other researchers have written on the subject in journals, books, and other forms of academic work (Babbie 2001; Robson 2002). This method will add to the reliability of the empirical findings by comparing them to other researchers’ results.

Finally, the researcher has participated in one congress – ‘Festival der Bestuurskunde 2009: Democratie tussen Wal en Schip’ organised by ‘Vereniging voor Bestuurskunde’ on February 12, 2009⁷⁶ –, one master-class on ‘The role of the media’ by Tarik Ramadan at the Erasmus University Rotterdam on March 27 and April 3, 2009, and one public gathering on the European elections – ‘Kieswijzer Live’ at the Arminius Church, Rotterdam on May 27, 2009. During these events several unstructured, but very insightful conversations took place with stakeholders and scholars with regards to legitimacy, the EU and the role/crisis of the media. These findings are only used for purposes of triangulation.