How the Media Shape Perceptions of Right-Wing Populist Leaders

Abstract

It is often argued that right-wing populist party leaders are dependent on the media for their public image, which in turn is key for their electoral success. This study tests this assumption by comparing the effects of the media coverage of 2 Dutch right-wing populist leaders with the effects of the coverage of leaders of established parties, in a real-life setting, by tracking campaign developments in the Dutch 2006 national election campaign. We combine panel survey data (n = 401) with repeated measurements of the party leaders' public images with a systematic content analysis of 17 media outlets (with a total of 1,001 stories), on the basis of the media consumption of individual respondents. Our results show significant effects of the content of media coverage on the public image of political leaders. However, only in 1 case (out of 10) is there a significant difference between right-wing populist party leaders and leaders of other parties in the strength of media effects. It thus seems that leaders of right-wing populist parties are just as dependent upon the media as leaders of other parties. The findings are discussed in the light of extant research on right-wing populist parties and media populism.

Keywords:

• populism
• media effects
• leadership
• right-wing populist parties
• electoral behavior

Notes

1. See http://gking.harvard.edu/amelia. Amelia produced five data sets in which missing values were imputed under different assumptions. For our analysis in Stata, we made use of Ken Scheve's MI program to automatically compute the quantities of interest and the standard errors. An analysis of the unimputed data matrix leads to very similar results and the same substantial conclusions.

2. We used the program MSPWIN 5.0 (Molenaar, Van Schuur, Sijtsma, & Mokken, 2002) to perform the scale analysis.

3. Legitimacy was measured with the following two items:

1.

To reach their goal some politicians are willing to ignore important democratic rules, while others will comply with these democratic rules under all circumstances. Below you see a number of current politicians. Could you tell me whether you think that they have always complied with the democratic principles and that they always will? (Responses could vary from 1 [doesn't care about democratic rules] to 7 [always complies with the democratic rules].)

2.

Sometimes people think a certain politician or party is dangerous. They are afraid that when that politician rises to power he or she will pose a threat to democracy. Others are of the opinion that this will not be the case. Below you will see a number of politicians. Imagine a situation in which this politician has risen to power—do you think that this politician would then pose a threat to democracy? (Responses could vary from 1 [If he/she rose to power he/she would pose a real threat to democracy] to 7 [If he/she rose to power he/she would definitely not pose a threat to democracy].)

Effectiveness was measured with two items as well:

1.	Some politicians have great influence on governmental policy; others do not have a lot of influence. Could you indicate for each of the following politicians whether you expect they will exert little or a lot of influence after the elections? (Responses could vary from 1 [will probably have little influence on policy] to 7 [will probably have a lot of influence on policy].)
2.	Some politicians you don't hear from, whereas other politicians exert a lot of influence on public debates. How important have the following politicians been in the public debate? (Responses could vary from 1 [does not shape the public debate at all] to 7 [shapes the public debate to a large extent].)

4. According to Mokken, the coefficient H (homogeneity of the items) has to be .30 or higher to be a scale. When H is higher than .50, it is a strong scale.

5. Because they are very often included as important explanatory variables, we additionally ran analyses in which we included measures of political interest and political knowledge. However, because we controlled for the lagged variable, the effects of these variables on the dependent variable were not significant, which is why we left them out of the analyses presented in this article.

6. To create the stacked data matrix, the generic independent variables had to be linearly transformed. As a result, their effects are positive.

7. We deliberately omitted Socialist Party (SP) leader Jan Marijnissen from the analysis, even though this party had great electoral success, the main reason being the populist nature of the party. We want to compare right-wing populist party leaders with leaders of established parties. Since the SP is often considered to be a left-wing populist party, we could have biased the results by including it in the group of established parties. We did, however, estimate our models with the SP included in the group of established parties, but this did not change the results substantively.

8. For more extensive information on the nature of the content analysis data, see Bos et al. (2010).

9. See http://academic.lexisnexis.nl/uva.

10. Because of a shortage of newspaper articles in which right-wing populist leaders were coded as one of the actors, we coded all of the articles in which Geert Wilders and Marco Pastors spoke about substantive matters. As a result, an extra 42 items were coded. These items were only used to estimate the positions of the right-wing populist leaders on the populism and authoritativeness dimensions.

11. See Bos et al. (2010) for a more detailed description of the intercoder reliability check and the results.

12. See Bos et al. (2010) for the formula.

13. The additional items referring to the ideological core of populism, anti-elitism, and heartland do not fit the scale.

14. In the original scale, we also included an item that measured whether the party leader referred to facts. However, due to low intercoder reliability results, we had to exclude the item from the scale.

15. Because the content analysis data are limited and provide only limited variance, we unfortunately cannot test interactions between the various aspects of the media coverage of right-wing populist leaders.

16. These variables are centered around their mean in order to use them in interaction terms.

17. The interactions are calculated not by using the original variables, but their deviations from the mean (Jaccard, Turrisi, & Wan, 1990).

18. We understand combining the two very different right-wing populist party leaders Geert Wilders and Marco Pastors into one dummy called “right-wing populist party leader” might raise questions with the reader. Tests did confirm, however, that including a dummy for each right-wing populist party leader individually did not change the results. We did not choose for this solution, because we believe that including two dummies and thereby doubling the number of interactions in our analysis would not lead to a substantive addition and would make the presentation of our results even more complex.

19. It can be argued that audience segmentation can have a polarizing effect; that is, media effects can differ between left-wing and right-wing parts of the electorate in such a way that the overall effect is insignificant. We have, however, tested for this and found that media effects are the same for audience members from all ideological backgrounds.