Corrective or Confirmative Actions? Political Online Participation as a Consequence of Presumed Media Influences in Election Campaigns

ABSTRACT

In the investigation of political participation, one potential influencing factor has so far hardly been considered: it is to a large extent unknown whether perceived influences of (online) media result in increased political (online) participation. To prove this, a survey among the population in the German federal state North Rhine-Westphalia was conducted in the context of an election (n = 485). The results show that perceived strong online influences led to an increase in online communication activities in the election campaign. The findings indicate that the online participation can be hardly interpreted as corrective, but rather as confirmative actions.

KEYWORDS:

• Confirmative actions
• corrective actions
• influence of presumed media influence
• online media
• political participation

Notes

1. Consider two individuals that assess media influences on themselves and on others on 5-point scales. Person A estimates the influence on himself with the value “1” and the influence on others with the value “2.” Person B estimates the influence on herself with the value “4” and the influence on others with the value “5.” Thus, in both cases the difference in perception is 1, although person A perceives weak media influences while person B perceives strong influences.

2. The answers to the two latter items were recoded before they were included in the index: the values 1 and 2 became 0; the values 3, 4, and 5 became 1. As Cronbach’s alpha was very low, the analyses described in the next chapter were also conducted for the individual items. Because the results were very similar to those of the index, they will not be presented in detail.

3. All variables for perceived influences and their evaluation were standardized in order to facilitate the interpretation of the results (Hayes, Glynn, & Huge, 2012).

4. Statistics show that all tolerance values are above .38 and variance inflation factors (VIF) are below 2.7. This suggests that there are no problems of multicollinearity. However, presumed influences of television and newspapers (r = .713; p < .001), television and Internet (r = .372; p < .001), and newspapers and Internet (r = .342; p < .001) correlate. Thus, it was tested in separate regression analyses whether the perceived influence of the Internet still predicts communication activities when perceived influences of television and newspapers are not included in the models. This is the case.

5. The relevant statistical values indicate a good model quality (cf. Hu & Bentler, 1999): χ2 = 2.22 (df = 4, p = .70); comparative fit index (CFI) = 1.00; root mean square error of approximation (RMSEA) = .00.

6. In addition to that, the agenda-setting dimension (β = .22; p < .001) and the candidates’ image dimension (β = .21; p < .001) have a direct influence on the voting decision dimension.

7. All indirect effects were bootstrapped (5,000 bootstrap samples), as recommended by Hayes, Preacher, and Myers (2011).

8. The statistical values for those path analyses also indicate good model qualities. “General communication” model: χ2 = 2.11 (df = 4, p = .72); CFI = 1.00; RMSEA = .00; “contradictions to differing opinions” model: χ2 = 1.40 (df = 4, p = .84); CFI = 1.00; RMSEA = .00.

9. In online versions of newspapers these barriers do not exist because articles can be commented on easily. However, from the German wording in the questionnaire it becomes quite clear that “newspaper” refers to the off-line version only. Nevertheless, it cannot be ruled out that some respondents had the online version in mind when answering the question.

10. Cognitive pretests conducted in the context of this study revealed that many respondents label the influences of different online media (e.g., Facebook, Twitter, YouTube, news Web sites) generally as “Internet influences” (for cognitive pretests, see Collins, 2003).

Additional information

Funding

This work was supported by the German Research Association, research group “Political Communication in the Online World,” subproject 3, grant number 1381.

Notes on contributors

Uli Bernhard

Uli Bernhard is a research associate at the Department of Communication and Media Studies, University of Düsseldorf, Germany. His current research areas include political communication, presumed media influences and its consequences, and content analyses of radio programs.

Marco Dohle

Marco Dohle is a research associate at the Department of Communication and Media Studies, University of Düsseldorf, Germany. His research focuses on political communication, presumed media influences and its consequences, and the selection and effects of media entertainment.