Party Advertising in Newspapers

Abstract

Journalists increasingly fear that their work may be influenced by advertisers’ interests. While studies have found that commercial advertisers exert pressure on news producers, equivalent research on political advertising is scarce. Using a rigorous methodological approach, we investigate whether party advertising led to three different forms of media bias in newspaper coverage of the Austrian general election in 2013. We combine data on party newspaper advertising (N≈1300), party press releases (N≈1900) and coverage in seven newspapers (N≈4200). There are no statistically significant effects of party advertising on media bias, though there is indicative evidence of differences between media genres. Our findings advance the study of advertiser pressure in the field of political journalism and political communication and clarify a heated public debate.

KEYWORDS:

• advertiser pressure
• media bias
• newspaper advertisements
• political advertising
• political campaigns

ACKNOWLEDGEMENTS

The authors would like to thank Laurenz Ennser-Jedenastik, Martin Haselmayer and Peter Maurer, as well as our anonymous reviewers, for their helpful comments and insights.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the authors.

Notes

1. The 2013 election was the first contested by two new parties that ended up gaining representation in parliament, NEOS (a liberal party) and Team Stronach (led by a business tycoon). Both seem to have shaken up the political advertising market. Team Stronach spent more than half of its considerable campaign budget on ads. Together they bought around half of the advertising shares in two media outlets (Die Presse and Der Standard).

2. The authors thank the Media FOCUS Research Ges.m.b.H. for providing the party advertising data.

3. We focus our analysis on newspapers, as they are the most important media for political advertising in Austria and still play an important role in the Austrian media system in general. Television news is still dominated by public broadcasting. However, regulations prohibit political ads on public broadcast networks, making substantive comparisons to commercial broadcasters impossible.

4. Inter-coder reliability scores (Krippendorff’s α) between seven coders are at 0.85 (speakers), 0.76 (evaluative statement), 0.78 (addressee) and 0.80 (issue) (N = 1123).

5. For each article, only claims in the title, the lead and the first paragraph were coded.

6. Here, inter-coder reliability scores (Krippendorff’s α) between six coders are at 0.99 (speakers) and 0.70 (issue) (N = 100).

7. NEOS have to be excluded from the agenda bias analysis since they only sent out about a dozen press releases during the election campaign.

8. Outlet-specific analyses confirm this finding: only for the Neue Kronenzeitung is there a statistically significant positive relationship between party ads and tonality bias (β = 0.003, p = 0.048); only Österreich shows a positive relationship between party ads and visibility bias (β = 0.001, p = 0.043); and for no medium is there a significant relationship between party ads and agenda bias. Of course, these outlet-specific tests are weak, as they are based on seven observations only.

Additional information

Funding

This research is supported by the Dissertation Completion Fellowship of the University of Vienna and conducted under the auspices of the Austrian National Election Study (AUTNES), a National Research Network (NFN) sponsored by the Austrian Science Fund (FWF) [S10908-G11].