Politician versus party? : personalization in political news coverage : a comparison during election campaign periods and routine periods in Germany and the United Kingdom

(1)

Politician versus Party?

Personalization in political news coverage – A comparison during

election campaign periods and routine periods in Germany and

the United Kingdom

Master Thesis

by

Jasmin Tokpinar

Graduate School of Communication – Track Political Communication

Supervisor: Dr. Magdalena Wojcieszak

Student-ID: 10973451

Email: j.tokpinar@gmail.com Word count: 5057

(2)

I. Introduction

The field of political communication has changed significantly in the last two decades. Starting with the fast developing mediatization (Strömbeck, 2008) of politics and other societal institutions, the increasing professionalization together with the rise of political marketing has changed the way in which politics and politicians’ interacting with the media and the public. Blumler and Kavanagh (1999) have labelled these developments as “The Third Age of Communication”. Within this context, the personalization of politics and especially personalized political news coverage got in the focus of political communication research. In the framework of election campaigns personalization in news coverage generally refers to the shift of focus from parties and programs towards leaders and candidates and their personal lives (van Aelst, Shaefer, Stanyer, 2012). In parliamentary regime types however, this development is evaluated critically. Considering that the media representation of politics concentrates on individuals and therefore neglects the institutional and systematic structural representation of society. Further, keeping in mind that the vast majority of literature in political communication is analysing election campaigns, claiming generalization seems problematic. What about personalization in news coverage when no important elections take place? Do political news still concentrate on individuals? Only if researching this additional aspect, the dynamics of the election term on personalization can be determined. Therefore, the main research question is raised as:

RQ: To what extent is personalization driven by the election campaign period?

Answering this core question bares academic and also societal relevance. First, comparing personalization within an election period with personalization patterns during routine periods of political news coverage could enable a more comprehensive approach. Therefore, it could give more space to draw generalizations in personalization studies and in terms of how media covers politics. Second, findings that classify personalization rather to campaign periods can

(3)

help relativizing and turning down concerns of a systematically insufficient media representation.

II. Theoretical Background

The Conceptualization of Personalization

Personalization studies have become a popular topic within political communication research. This may be due to the fact that personalization is multi-dimensional phenomenon that refers to communicative interactions between politics, the media and the public, concepts that are core aspects of political communication. Therefore, personalization refers to the campaign profession, the voting behaviour and the political media coverage. In this study however, the personalization in media coverage is the main focus. Almost all studies that are dealing with personalization, are analysing media coverage during election campaigns. Although the analysed countries and cases are showing similarities and even overlap at times, authors come to very different conclusions concerning the existence and degree of personalized news coverage. One important reason for this, may be the differences in how different authors conceptualize and define personalization. In general, personalization is considered as a concept that is indicated by the increasing visibility and concentration on politicians as individuals in media coverage rather than on parties, issues or organizations (Gerth, Rademacher, Pühringer, Dahinden, Siegert, 2009; Holtz-Bacha, Langer, Merkle, 2014; Krisie, 2010; van Aelst et al, 2012; Zeh and Hopmann, 2013). Therefore, generally extant studies focus on the frequency and degree of concentration on politicians in media coverage of campaigns in comparison to the appearance of parties and issues. Krisie (2010) distinguishes between a “general personalization” which refers to a predominant coverage of individual politicians over parties and organizations in general, and a “particular personalization” which isolates the focus on the coverage of the top 10 and eventually on the top two most frequently mentioned personalities (Krisie, 2010). While Gerth et al.(2009) define personalization as the

(4)

amount of news selection oriented to a small number of prominent political actors, Zeh and Hopmann (2013) analyse personalization by considering news stories that are focussing only on the prime minister candidates.

As noticeable, personalization seems to be strongly related to an increasing of the importance of leaders and / or top candidates. In the literature this development is also termed as “presidentialization” (van Aelst et. al, 2012: 206) of political news coverage as consequence of a personalized reporting. In an in-depth literature review van Aelst et. al (2012) have developed an operationalization of the term, whereby they emphasize that personalization of news coverage consists of two forms. The first dimension is labelled as “Individualization” which concerns the focus on individual politicians as central actors in the political arena. The second dimension of personalization is labelled “privatization” (van Aelst et. al 2012:210) which concerns the focus on personal qualities and characteristics.

This study is defining personalization in the sense of the presidentialization trend, that refers to the first dimension labelled as “Individualization”. Therefore, following this conceptualization, in this framework personalization is defined as a concentrated visibility on leaders and top-candidates over their parties. The reason to focus only on this first dimension lies against the background that this study aims to compare differences in leader visibility between two different time frames. Studying whether personalization exist in both time periods is a scientifically logical first step, before analysing the kind and specifics of this personalization. Latter refers to the second dimension of “privatization”

Previous Findings

Due to the different concepts and definition frameworks discussed above, findings as to the extent and the nature of personalization differ between studies. Krisie (2010) for example

(5)

analysed personalization patterns in campaign coverage across six western European countries (United Kingdom, Germany, Netherlands, France, Austria, Switzerland) and concludes that there is no significant degree of personalization. However, Holtz-Bacha et al. (2014) who also compared campaign coverage in Germany and the United Kingdom not only show an increase in personalised news but also similar patterns between the two countries. This is important in the case of Germany, considering that previous studies that looked at campaign coverage could not confirm a noticeable trend of personalization in the sense of a leader and candidate dominance over parties (Kaase, 1994; Wilke and Reinamann, 2001). Single-case studies in the United Kingdom also find an increase in leader visibility over time (see Foley 2000; Mughan, 2000; Langer, 2011). In general research shows that personalization in election campaign coverage has increased in the last fifty years (Dalton, McAllister, Wattenberg, 2000). The introduction of televised political debates (Reinemann and Wilke, 2007), and especially the upcoming of TV duels between the top candidates in Germany and more recently also in the United Kingdom are a good example. Here the increasing dominance of top-candidates and leaders over their parties in media representation can be easily observed. Furthermore, the research review shows differences in the degree and form of personalization between types of media outlets (see Zeh and Hopmman, 2013) and also emphasizes specific context issues of political news. The study of Gerth et al. (2009) analysing a referendum campaign in Switzerland seems interesting to mention at this point. Also here the findings are showing a high concentration on the top-ten actors and a dominance of governing politicians over the opposition in Swiss campaign coverage, both trends indicating personalization.

The Comparison of Campaign and Routine Periods

Although research in the field of personalized campaign coverage appears highly comprehensive, covering campaign periods in single country (Foley, 2000; McAllister, 1996; Schulz and Zeh, 2005), in a cross-national perspective (Balmas and Sheafer,2014;

(6)

Holtz-Bacha et. al, 2014, Krisie, 2010; Zeh and Hopmann, 2009), longitudinally (Balmas & Sheafer, 2014; Krisie, 2010) and as based on specific political issues (Gerth et. al, 2009), all studies are limited by focusing on campaign periods only. A significant research gap appears when it comes to comparative analyses that compares the extent of personalized coverage during the election period to non-election or rather routine periods of politics. Campaign periods are exceptional in the process of politics and therefore they must be compared to non-campaign periods (Gulati, Just, Crigler, 2004). Deacon, Golden and Billig (2001) also conclude: “Analytical focus on the campaign alone may not reveal all we need to know about the potency of the media in defining the civic culture of politics.” (2001:677). During election campaigns, journalists are facing different working conditions. Further politicians are much more active and eager in order to get media attention, whereby at the same time voters are much more attentive to political news (van Aelst & de Swert, 2009). These factors are presumed to increase a rather personalized campaign coverage in the news. Therefore, the first two hypotheses are stated as:

H1: The degree of visibility of the top-candidates/ leaders will be higher in campaign periods than in routine periods

H2: In routine periods the coverage will be more balanced between leaders and their party

The Study of van Aelst and de Swert (2009) is important to mention in this context, because it is the only study that does include the analysis of political news within the routine-period. However, considering only one country (Flandern, Belgium) this study lacks an additional comparative perspective. Furthermore, besides the fact that this study is not focusing on personalization, considering Belgium as a single-case also appears problematic. Due the specific institutional constellation in the country, distinguishing between campaign periods

(7)

and routine periods is more difficult than in other political set-ups. From a scientific point of view, a cross-country comparison would deliver more significant insights, especially in countries where the time between election campaigns and no-elections periods is longer than Belgium.

The Comparison of Germany and the United Kingdom

Analysing Germany and the United Kingdom will put much more relevance to the findings, since both countries are one of the most important industries in Europe. Further in both countries the time frame between elections and the time frame where no elections take part are much longer and therefore easier to identify. Also this would give further relevance to possible findings assigned to the specific time frames. The comparative study of Holtz-Bacha et al. (2014) between Germany and the UK gives important more up-to date insights to the field, however as other research, their research is limited to a campaign period.

This thesis aims to address the limitations of this two previous studies in the field, by conducting a cross-country comparison over two different time periods. Comparing Germany and the United Kingdom seems highly feasible because both countries share structural and institutional similarities that are important for the dynamics of personalization. (Holtz-Bacha et al.,2014). Although earlier studies conclude different degrees of personalization in election coverage in these countries (see, Wilke and Reinemann 2001; Mughan 2000), the primarily position of leaders and top-candidates has been constantly high over time. Further both countries are parliamentary regime types with strong party systems. Still they differ in political structure, with Germany being a consensus democracy and the UK following a majoritarian principle of representation. The media systems in both countries are characterized by a high newspaper readership and strong public broadcast service, but do still differ from each other significantly. Therefore, the media landscape in the UK is much more fragmented, highly competitive and openly partisan. In Germany on the other site, the

(8)

newspaper market is less fragmented, given there is only one tabloid paper and no mid-market newspaper at all. Partisanship is also fewer and not openly showed in German journalistic culture (for an overview: Hallin and Mancini, 2004). Besides these theoretical and structural considerations, it is also important to catch up on the findings of Holtz-Bacha et al. (2014) that compared the personalized election coverage in Germany (2009) and the United Kingdom (2010). Overall they found, that in both countries the visibility of the leaders and top-candidates is rather higher than the parties’. Whereby this in-balance between candidates and their parties is much bigger in the UK than in Germany. In Germany also the candidates of the other smaller parties where represented, while in the United Kingdom media mainly concentrated on the first three parties. This showed that in Germany there was overall a higher party representation in the news than in the United Kingdom (Holtz-Bacha et al.,2014). In the framework of these findings, the third and fourth hypotheses for the analysis of the last general elections in Germany (2013) and the United Kingdom (2015) are raised as:

H3a: British leaders are more visible in comparison to their parties, than German leaders – H3b: British leaders are more visible in general than German leaders in the news coverage

H4: In general, parties are more visible in the German news coverage than in the UK

IV. Methodology

This thesis aims to elaborate the impact of the election campaign period on “media personalization” in Germany and the United Kingdom. As discussed above, such an impact can be analysed the best when it is also compared to the dynamics of personalization during routine times of political news coverage. Thereby the focus lies on the increasing visibility of top-candidates and leaders over their parties.

(9)

Research Units

For this, quantitative content analysis of two German and two British broadsheet newspapers were conducted. In order to ensure a representative sample, the newspapers were selected upon terms of availability, circulation rate and political standing. For Germany the liberal-left newspaper Süddeutsche Zeitung with a daily circulation of 367.9241 (STATISTA, 2016) and the conservative-right newspaper Die Welt with a daily circulation 187.8662 (STATISTA, 2016) were chosen. For the UK the centre-left newspaper The Guardian with a daily circulation of 166.0393 (Press Gazette, 2016) and the conservative-liberal Financial Times with a daily circulation of 209.2644 (Press Gazette, 2016) were chosen. The articles of the

Süddeutsche Zeitung were collected from its online archive; all other articles were collected

through the LexisNexis database.

Time Periods

The sample is covering two time frames labelled here as “campaign period” and “routine period”. The campaign period covers the last 7 days prior to the general elections in Germany 2013 and the general elections 2015 in the United Kingdom. Therefore, for Germany the week from the 15th of September 2013 until the 22nd of October 2013, for the UK the week from the 30th of April 2015 until the 7th of May 2015 were analysed. All articles that referred to the general elections and included at least one of the listed top-candidate and or party (see also appendix) in the headline were included. That resulted in a sample size for the campaign period of N = 166 articles (Germany N=90; UK N= 76), balanced represented by all outlets. To balance the distribution of the articles on the outlets, systematic sampling for The

1 as of December 2015 2 as of December 2015 3 as of December 2015 4 as of December 2015

(10)

Guardian and the Süddeutsche Zeitung was applied. Therefore every 8th article of The

Guardian and every second article of the Süddeutsche Zeitung were analysed.

The routine-period was determined by applying constructed week sampling following the conceptualization of Riffe, Aust and Lacy (1993). Therefore, one constructed week represents six months (Sundays excluded). Including one month as a puffer, for Germany this resulted in a time frame starting from the 21st of October 2013 until the 22nd of March 2014. The weekdays were sampled systematically as: October 21, 2013 (third Monday), November 26,2013 (fourth Tuesday), December 4, 2013 (first Wednesday), January 9, 2014 (second Thursday), February 21, 2014 (third Friday) and March 22, 2014 (fourth Saturday). For the United Kingdom the routine period covers the 8th of June until the 7th of November 2015. The constructed week is represented as: June 8, 2015 (first Monday), July 14, 2015 (second Tuesday), August 19, 2015 (third Wednesday), September 24, 2015 (fourth Thursday), October 30, 2015 (fifth Friday) and November 7, 2015 (first Saturday). Here, all articles in the political news section that are domestic news and refer to the listed party-leaders and or parties in the headline were included. That resulted in a sample size for the routine period of N = 92 (Germany N= 47; UK N= 45), which is expectably smaller than the sample of the campaign period. All suitable articles were coded, except for The Guardian, where again only every 8th article was analysed.

Design and observed variables

Two codebooks for each period were created. For the campaign period, the first six German parties and their top-candidates and the first four British parties and their top-candidates are representing the observed variables. The articles were analysed by coding the number of mentions of the top-candidates and their parties in the headline, the main body of the article and the number of times the candidates got quoted. The reason to create two codebooks for each period lies against the background that the variable names and therefore the concerned

(11)

politicians changed after the elections in both countries. In the UK, due to the election outcomes the leaders of the Labour and Liberal Democrats resigned. Therefore, for the second codebook these leaders were replaced by the new party leaders. Although not a direct consequence from the election outcomes, also the politician names in the German case had changed. Therefore, for the content analysis of the German outlets during the routine period these changes were considered and adjusted. The reason for this is simply, that in Germany the party leaders and the top-candidates can be the same person (as in case of Angela Merkel for example) but all other parties this two positions are divided among different party representatives. However, the changing names, do not influence the analysis negatively, since the research focus remained the same. Therefore, the literature in personalization puts the stress on party leaders and or top-candidates, whereby in most countries these are the same individuals. For academic consistency however, it is very important to adjust the analysis to the specifics of the German party system and also to the consequences of the electoral outcomes in the United Kingdom. The importance lies in the fact to observe the individuals that media is focusing on as representatives, and these individuals change after an election period. However, for both periods and countries, the variables remained the same. That way both time periods and both countries could be analysed and compared to each other in one dataset. To illustrate this: Variable 1 for example as “Leader_Candidate_CDU_Headline” for the mentions of Angela Merkel in the headlines, and “CDU_Headline” for the mentions of her party in the headlines (For all considered politicians and parties: see appendices). In order to test and ensure the quality of the codebooks, two coders participated in the data analysis. The second coder, analysed a sup-sample of 10 articles for each period: the inter-coder reliability testing resulted in a score of =.84. The analysis was conducted by using the the data analysis software SPSS. In order to answer the hypotheses and research question all variables had to be comparable in both time periods and in both countries. For this, the variables were

(12)

grouped in three categories: “Country”, “Time period” and “Presence” in each analysis section (headline, article, quote) and analysed with frequency analysis.

Table 1. Number of coded articles by time period and country

Country Campaign Routine Total

N % N % N %

Germany 90 54,2 47 51,1 137 53,1

UK 76 45,8 45 48,9 121 46,9

Total 166 100,0 92 100,0 258 100,0

V. Results

Since the research focus covers two comparative aspects, the results will be presented accordingly. First the findings will be considered in the light of the comparison between the campaign period and the routine period for Germany and the United Kingdom. Putting the findings in an overview set up for each time period first, will help understanding them more clearly when compared more detailed between the two countries.

As table 2 and table 3 are showing, the visibility of candidates and leaders does differ significantly between the campaign and the routine period in Germany and the UK. In Germany the top-candidates are more than six times more present in the headlines during the campaign, than in routine times. Further they were mentioned 130 times in all articles analysed during the campaign period in comparison to only 44 times in the routine period. However, putted in perspective, this is still a high presence also in routine times considering the sum of mentions represents 93,6 % of the total number of articles. The high difference in the German candidate visibility during the campaign and routine also shows itself by the

(13)

categories “quotes”. Therefore, the politicians were quoted almost three times more during the campaign than during the routine political news coverage.

Table 2. Presence of German candidates and leaders in each section in campaign vs. routine periods

Section Candidate/Leader mention sum (DE)

Campaign Period Routine Period

N % N % Headline 52 57,8 8 17,0 Article 130 144,4 44 93,6 Quotes 31 66,0 12 25,5 N= 90 N= 47

Note: The totals are smaller than the sums of the categories because more than one candidate or leader could be quoted and mentioned in the same headline and or article.

Similar findings can be observed in the case of the visibility of UK party leaders. However, when it comes to headline presence, the candidates were mentioned slightly more often during the routine period (62,2%) (Table 3) as they were mentioned during the campaign period (57,9%) (Table 3). This is the main difference when the two countries are compared here. Although the candidate’s headline presence in both countries shows almost the same degree with 57,8 % in Germany (Table 2) and 57,9% in the UK (Table 3) during the election time, British leaders are noticeably much more present in the headlines during routine times than in Germany. The higher leader visibility in campaign times, also can be seen in the article presence and number of quotes. Compared to the routine period, in both sections the party leaders were present almost four times more often during the election time (see Table 3).

(14)

As illustrated the findings are indicating a significantly higher visibility of the top-candidates and leaders during the campaign period in comparison to the routine period. Except minor differences, this is the case for both countries. Therefore, the first hypothesis H1 can be confirmed as: the degree of visibility of top-candidates and leaders is higher in campaign periods than in routine periods.

Table 3. Presence of British candidates and leaders in each section in campaign vs. routine periods

Section Candidate/Leader mention sum (UK)

Campaign Period Routine Period

N % N % Headline 44 57,9 28 62,2 Article 162 213,2 44 97,8 Quotes 24 53,3 8 17,7 N= 76 N= 47

Note: The totals are smaller than the sums of the categories because more than one candidate or leader could be quoted and mentioned in the same headline and or article.

Table 4 is presenting the overall ratios of candidate/leader mentions in relation to their parties’ mentions in the headline and articles. Looking at the results concerning the United Kingdom the leader / party ratio in headline presence is 0,7 in both time periods. Also in the article presence the ratio degree with 0,7 in the campaign period and 0,6 in the routine do not differ significantly. That means, the coverage of party leaders and their parties seems to be rather balanced but not different during election or non-election times. In terms of personalization, leaders do not dominate but are basically almost present as much as their parties. Since the overall ratio degrees are smaller than 1 however, party presence is still overweighing the news coverage in general. In contrary to that, in Germany there is a

(15)

noticeable difference between the campaign and the routine period, especially considering the headline presence. However, the ratio here of 0,1 (Table 4) indicates a high domination of parties over leaders in routine times.

The second hypothesis H2 is stating that, in routine periods the coverage will be more balanced between parties and their leaders. Considering the findings here however, this can not be confirmed. All in, party visibility is overweighting the leaders in both countries. The UK coverage is rather balanced in routine times, however at the same level as in campaign times (see Table 4). In Germany the coverage is not balanced but more party dominated in routine times. In contrary to the hypothesis the coverage gets more balanced in campaign times instead of routine times.

Table 4. Comparison of Politician vs. Party presence in total

Section Candidate/Leader vs. Party sum

Germany UK

Leader/Party Leader / Party

Ratio  Ratio 

Campaign Routine Campaign Routine

Headline 0,4 0,1 0,7 0,7

Article 0,4 0,3 0,7 0,6

Note: 1= equal presence; 1 > Party presence is higher; 1< Candidate/Leader presence is higher

The analysis shows that the leader and party presence distribution is not differing significantly between the two time periods in both countries. That means findings from the campaign period can be generalized also for the routine period when it comes to personalization tendencies. Therefore, when comparing the two countries directly in the following sections, the findings referring to the campaign period will be presented.

(16)

As discussed above, Table 4 shows that British leaders are more visibly in comparison to their parties than German leaders. This confirms the third hypothesis H3A stated as:

“British leaders are more visible in comparison to their parties, than German leaders”.

Besides the general ratio calculation, this can also be observed when the mentions of leaders and candidates alone are compared to each other in both countries. Considering the two main leader presence, Figure 1 shows, that David Cameron (Conservative Party) and his opponent Ed Miliband (Labour Party) were mentioned in over 70% of the articles. In comparison to that Angela Merkel (CDU) was mentioned in over 50% of the articles and her main opponent Peer Steinbrück (SPD) in 27,8%. Also when looking at the category “other leaders”, the leaders of the smaller parties in the UK were slightly mentioned more often in the articles than the leaders of the other parties in Germany. When it comes to headline presence however, leaders of the smaller parties are more visible in Germany than in the UK. Probably, out of the reason that the headlines in the UK are dominated by the leaders of the two first parties. Findings show, that there is also a higher general visibility of leaders in the UK in comparison to the visibility of leaders in Germany. Therefore, the sub-category of hypothesis H3B can also be confirmed.

Figure 1. Mentions of Candidates/ Leaders in UK and Germany in Comparison

52,2 27,8 64,5 72,4 73,7 67,1 21,1 11,1 25,6 26,3 27,6 17,1 0 10 20 30 40 50 60 70 80 90 100

A. Merkel P. Steinbrück Other top-candidates

DE

D.Cameron E. Miliband Other leaders UK Pe rc e n tag e o f ar ti cl e s Article Headline

(17)

The fourth and last hypothesis H4 is stating that, parties in the German news coverage are generally more visible than parties in the British news coverage. The ratio overview in Table 4 already indicated that parties are more visible than their leaders/candidates in both countries and both time periods. However, in the German case the ratio scales for both time periods and and both article sections are noticeably smaller than in the UK. Which shows an over presence of parties over candidates. Considering it in total numbers, Figure 2 shows that parties in Germany were mentioned two times more often in the headlines than parties in the UK during the campaign period. Overall parties were mentioned 343 times in Germany in comparison to 221 in the UK during campaign periods. Therefore, the findings also confirm the fourth hypothesis H4.

Figure 2. Mentions of Parties in the UK and Germany in Comparison

Note: The totals are smaller than the sums of the categories because more than one party could be mentioned in the same headline and or article.

343

221

122

63

Parties combined DE Parties combined UK

Mentions of Parties in Total

Article Headline

(18)

VI. Conclusion and Discussion

The aim of this thesis was to compare personalization in news coverage during campaign and routine periods of politics between Germany and the United Kingdom. For both countries the analysis could not detect an explicit personalization in the news coverage, in the sense of a higher visibility of leaders and candidates in comparison to their parties. Nevertheless, the first hypothesis H1 could be confirmed, thus leaders and candidates are significantly more visibly during campaign periods than routine periods. In general, parties are more visible than their leaders in both countries, regardless of the time period. However, it could be shown that personalization tendencies are much higher in the UK than in Germany. In the United Kingdom, the visibility of politicians is noticeably higher than in Germany. This also led to the confirmation of the third Hypothesis H3. In Germany on the other hand, a clear dominance of party presence was found. The fourth hypothesis could also be confirmed, thus in general parties in Germany are significantly more visibly than in the United Kingdom. The most interesting finding is probably the low impact of the campaign period when it comes to personalization. It was found that the distribution of leader vs. party visibility in both countries, does not differ significantly in both time periods. Therefore, the coverage is not influenced by the campaign period, in the sense to give more presence to leaders and candidates over their parties. Due to the missing difference between the time periods, the second H2 hypothesis could not be confirmed. Coming back to the research question, to what extent is personalization driven by the campaign period? Answering the research question appears difficult, considering that this study could not find personalization per se. The biggest impact of the campaign period on the news coverage in both countries is, that there are significantly more articles covering politics and politicians are much more present in the news than in routine periods. However, this fact should rather be assigned to the special issue-context of elections than to personalization tendencies. Facing the results this study concludes, that personalization is not directly driven by the campaign period. Therefore this

(19)

study also does not confirm with previous studies that found personalization patterns in Germany and the UK (Holtz-Bacha et. al, 2012; Mughan, 2000). However, this study is limited in certain terms. Therefore, sample period is probably in general too short. All other studies that are dealing with a similar topic, are covering a analysis period of at least three weeks before the elections. Also for generalizing over a routine period, further studies should extent the constructed week sampling covering at least one year. The limited scope of a master thesis however did not allow a longer sample period. Nevertheless, this study pioneered in conducting a two-country comparison in two different time frames. Therefore further research in a larger set up and with extended resources should definitely be pursuit.

(20)

VII. References

Balmas, M. and Sheafer, T. (2014). Charismatic Leaders and Mediated Personalization in the International Arena. Communication Research, Vol. 41(7) 991–1015.

Blumler, J. G. and Kavanagh, D. (1999). The third age of political communication: Influences and features. Political Communication 16 (3): 209–30.

Brettschneider, F. and Gabriel, O. (2002). The nonpersonalization of voting behaviour in Germany. In A. King (Ed.), Leaders' personalities and the outcomes of democratic elections. Oxford: Oxford University Press.

Dalton, R., McAllister, I., and Wattenberg, R. M. (2000). The consequences of partisan

dealingnment. In R. Dalton, and R. M. Wattenberg (Eds.), Parties without partisans. Political change in advanced industrial democracies. Oxford University Press, pp. 37-63.

Deacon, D., Golding, P., and Billig, M. (2001). Press and broadcasting: ‘real issues’ and real coverage. Parliamentary Affairs, 54(4), 666-678.

Foley, M. (2000). The British Presidency. Manchester: Manchester University Press.

Garzia, D. (2011). The personalization of politics in Western democracies: Causes and consequences on leader follower relationships. The Leadership Quarterly, 22(4), 697-709. Garzia, D. (2012). The Rise of Party/Leader Identification in Western Europe. Political Research Quarterly 66(3) 533– 544.

Gerth, M., Rademacher, P., Pühringer, K., Dahinden, U.and Siegert, G. (2009). Challenges to Political Campaigns in the Media: Commercialization, Framing, and Personalization. Studies in Communication Sciences, Vol. 9(1):149-170.

Gulati, G. J., Just M. R., and Crigler, A. N. (2004). News coverage of political campaigns. In L. L. Kaid (Ed.), Handbook of political communication research, pp.237-256.

(21)

Hallin, D. C. and Mancini, P (2004). Comparing Media Systems. Three models of media and politics. ”The forces and limits of homogenization”

Holtz-Bacha, C., Langer, A-I. and Merkle, S. (2014). The personalization of politics in comparative perspective: Campaign coverage in Germany and the United Kingdom. European Journal of Communication , Vol. 29(2) 153–170.

Kaase, M. (1994). Is there personalization in politics? Candidates and voting behavior in Germany. International Political Science Review, 15, 211–230.

Kriesi, H. (2010). Personalization of national election campaigns. Party Politics, Vol. 18(6) 825–844.

Langer, AI. (2007). A historical exploration of the personalisation of politics in the print media: The British Prime Ministers (1945–1999). Parliamentary Affairs 60(3): 371–387. Langer, AI. (2011). The Personalisation of Politics in the UK: Mediated Leadership from Attlee to Cameron. Manchester: Manchester University Press.

Lees, C. (2014). Party Politics and Bundestag Party System after the 2013 Federal Election, German Politics and Society, Issue 111 Vol. 32, No. 41–53

McAllister, I. (1996). Leaders. In L. LeDuc, R. Niemi, & P. Norris (Eds.), Comparing democracies. Elections and voting in global perspective. Thousand Oaks, CA: Sage Publications.

Merkle, S. (2015). Personalisierung und genderspezifische Bertichterstattung im Bundestagswahlkampf 2013 – ‚Ausnahmefall’ Angela Merkel oder typisch Frau. In: Holtz-Bacha, C. Die Massenmedien im Wahlkampf. Die Bundestagswahl 2013. pp.217-247

Mughan, A. (2000). Media and the Presidentialization of Parliamentary Elections. New York: St.Martin’s Press.

(22)

Reinemann, C. and Wilke, J. (2007). It’s the Debates, Stupid! How the Introduction of Televised Debates Changed the Portrayal of Chancellor Candidates in the German Press, 1949- 2005. The Harvard International Journal of Press/Politics, 12, 92–111.

Riffe, D. , Aust, C.F. and Lacy, S.R. (1993). The effectiveness of random consecutive day and constructed week sampling in newspaper content analysis. Joumalism Quarterly, Vol. 70, No. 1.pp. 133-139.

Schulz, W. and Zeh, R. (2005). The changing election coverage of German television. A content analysis: 1990–2002. Communications 30: 385–407.

Strömbäck, J. (2008): Four phases of Mediatization: An analysis of the mediatisation of politics. International Journal of Press/Politics 13(3): 228-246.

Swanson, D. and Mancini, P. (1996). Politics, Media and Modern Democracy: An International Study of Innovations in Electoral Campaigning and their Consequences

Van Aelst, P., and De Swert, K. (2009). Politics in the news: Do campaigns matter? A comparison of political news during election periods and routine periods in Flanders (Belgium). Communications, 34(2), 149–168.

Van Aelst, P., Sheafer, T. and Stanyer. J (2012). The personalization of mediated political communication: A review of concepts, operationalizations and key findings. Journalism, Vol. 13(2) 203-220.

Wagner, A. and Weßels, B. (2012). Parties and their Leaders. Does it matter how they match? The German General Elections 2009 in comparison. Electoral Studies 31: 72–82

Wattenberg, M. (1991). The rise of candidate-centered politics. Cambridge: Harvard University Press.

(23)

coverage– a study of the German press since 1949. European Journal of Communication 16(3): 291–314.

Zeh, R. and Hopmann, N-D (2013). Indicating mediatization? Two decades of election campaign television coverage. European Journal of Communication, Vol. 28(3) 225-240.

Sources

Press Gazette (2016). “National press ABCs”.

URL: http://www.pressgazette.co.uk/national-press-abcs-december-2015-cut-price-daily-star-grows-sales-and-hurts-daily-mirror (retrieved: 20 January, 2016)

Statista.com (2016). Verkaufte Auflage der überregionalen Tageszeitungen in Deutschland im 4.Quartal 2015. “Circulation of the national daily newspapers in Germany in the fourth

quarter of 2015”. URL: http://de.statista.com/statistik/daten/studie/73448/umfrage/auflage-der-ueberregionalen-tageszeitungen/ (retrieved: 20 January, 2016)

(24)

Appendix

Leaders / Candidates (campaign) # “Angela Merkel (CDU)” “Peer Steinbrück (SPD)” “Rainer Brüderle (FDP)“ “Katrin Göring-Eckardt and Jürgen Trittin (Die Grünen)” “Sarah Wagenknecht and Gregor Gysi (Die Linke)” “Gerda Hasselfeldt (CSU)” “David Cameron (Conservative)” “Ed Miliband (Labour)” “Nicola Sturgeon (SNP)” “Nick Clegg (LibDem)”

Leaders / Candidates (routine) # “Angela Merkel” (CDU) “Siegmar Gabriel” (SPD)” “Patrick Lindner (FDP)” “Cem Özdemir and Simone Peter (Die Grünen)” “ Katja Kipping and Bernd Riexinger (Linke)” “Horst Seehofer (CSU)” “David Camerion (Conservative)” “Jeremy Corbyn (Labour)” “Nicola Sturgeon (SNP)” “Tim Farron (LibDem)”

Parties# “Christlich Demokratische Union (CDU)” “Sozialdemokratische Partei Deutschlands(SPD)” “Freie Demokraten (FDP)” “Die Grünen” “Die Linke” “Christlich Soziale Union (CSU)” “Conservative” “Labour” “Scottish National Party (SNP)” “ Liberal Democrats (LIBDEM)”

Codebook Items – Visibility of Politicians and Parties during campaign period

1. A number identifying the article (1….)

2. The day the article was published (e.g. 25/11/15)

3. The title of the outlet (SZ=1, Die Welt= 2, The Guardian=3, Financial Times=4) 4. The title/ headline of the article

5. Number of mentions of the top- candidates / leaders in the headline (a number ranging from 0 to infinity) (Only mentions referring to the full name or last name are counting)

(25)

5b. ( ) Peer Steinbrück 5c. ( ) Rainer Brüderle

5d. ( ) Katrin Göhring-Eckardt / Jürgen Trittin 5e. ( ) Gregor Gysi / Sarah Wagenknecht 5f. ( ) Gerda Hasselfeldt

5g. ( ) David Cameron 5h. ( ) Ed Miliband 5i. ( ) Nicola Sturgeon 5j. ( ) Nick Clegg

6. Number of mentions of the top-candidates / leaders in the main body of the article (a number ranging from 0 to infinity)

6a. ( ) Angela Merkel 6b. ( ) Peer Steinbrück 6c. ( ) Rainer Brüderle

6d. ( ) Katrin Göhring-Eckardt / Jürgen Trittin 6e. ( ) Gregor Gysi / Sarah Wagenknecht 6f. ( ) Gerda Hasselfeldt

6g. ( ) David Cameron 6h. ( ) Ed Miliband 6i. ( ) Nicola Sturgeon 6j. ( ) Nick Clegg

7. Number of direct quotes attributed to the top-candidates/ leaders (a number ranging from 0 to infinity)

7a. ( ) Angela Merkel 7b. ( ) Peer Steinbrück 7c. ( ) Rainer Brüderle

7d. ( ) Katrin Göhring-Eckardt / Jürgen Trittin 7e. ( ) Gregor Gysi / Sarah Wagenknecht 7f. ( ) Gerda Hasselfelt

(26)

7h. ( ) Ed Miliband 7i. ( ) Nicola Sturgeon 7j. ( ) Nick Clegg

8. Number of mentions of the parties in the headline (a number ranging from 0 to infinity) (References to party ideologies also count. E.g.: Christian-Democrats = CDU or Liberals= FDP, References to parties through colours in German case count! E.g: “Schwarz” for CDU/CSU, “Rot” for SPD, “Gelb” for FDP, “Grün” for Die Grünen, “ Violette” for die Linke)

8a. ( ) CDU 8b. ( ) SPD 8c. ( ) FDP

8d. ( ) Die Grünen 8e. ( ) Die Linke 8f. ( ) CSU

8g. ( ) Conservative 8h. ( ) Labour 8i. ( ) SNP 8j. ( ) Lib Dem

9. Number of mentions of the parties in the main body of the article (a number ranging from 0 to infinity)

9a. ( ) CDU 9b. ( ) SPD 9c. ( ) FDP

(27)

Codebook Items – Visibility of Politicians and Parties during routine period

1. A number identifying the article (1….)

2. The day the article was published (e.g. 25/11/15)

3. The title of the outlet (SZ=1, Die Welt= 2, The Guardian=3, Financial Times=4) 4. The title/ headline of the article

5. Number of mentions of leaders in the headline (a number ranging from 0 to infinity) (Only mentions referring to the full name or last name are counting)

5a. ( ) Angela Merkel 5b. ( ) Siegmar Gabriel 5c. ( ) Patrick Lindner

5d. ( ) Cem Özdemir / Simone Peter 5e. ( ) Katja Kipping / Bernd Reixinger 5f. ( ) Horst Seehofer

5g. ( ) David Cameron 5h. ( ) Jeremy Corbyn 5i. ( ) Nicola Sturgeon 5j. ( ) Tim Farron

6. Number of mentions of leaders in the main body of the article (a number ranging from 0 to infinity)

6a. ( ) Angela Merkel 6b. ( ) Siegmar Gabriel 6c. ( ) Patrick Lindner

(28)

7. Number of direct quotes attributed to leaders (a number ranging from 0 to infinity) 7a. ( ) Angela Merkel

7b. ( ) Siegmar Gabriel 7c. ( ) Patrick Lindner

8. Number of mentions of the parties in the headline (a number ranging from 0 to infinity) (References to party ideologies also count. E.g.: Christian-Democrats = CDU or Liberals= FDP, References to parties through colours in German case count! E.g: “Schwarz” for CDU/CSU, “Rot” for SPD, “Gelb” for FDP, “Grün” for Die Grünen, “ Violette” for die Linke)

8a. ( ) CDU 8b. ( ) SPD 8c. ( ) FDP

9. Number of mentions of the parties in the main body of the article (a number ranging from 0 to infinity)

9a. ( ) CDU 9b. ( ) SPD

(29)

9c. ( ) FDP