Social Media Activities of Political Communication Practitioners: The Impact of Strategic Orientation and In-Group Orientation

ABSTRACT

The question of why political communication practitioners use social media for strategic political communication activities has rarely been investigated. By using well-established theoretical approaches of communication research, such as the influence of presumed influence approach, this study sought to determine the extent to which the subjective perceptions of German political communication practitioners explain their professional social media activities. The results of a survey (N = 1,067) indicate that the more political communication practitioners perceived that other political communication practitioners used and were influenced by Facebook and Twitter, the more often they used social media themselves. In contrast, the presumed reach of Facebook and Twitter among politicians, journalists, and citizens, as well as the presumed influence of both media on these groups, were not related to the practitioners’ social media activities. These findings suggest that the practitioners’ social media activities are driven more by an in-group orientation toward their colleagues and less by a strategic orientation toward external stakeholders.

Acknowledgments

The authors would like to thank Georg Milde, René Seidenglanz, and Kristina Sinemus (Quadriga University of Applied Science Berlin, Germany) for the cooperation in data collection. Moreover, we thank the two anonymous reviewers for useful comments on earlier versions of this article.

Funding

This publication was created in the context of the Research Unit “Political Communication in the Online World” (grant number 1381), subproject 3, which is funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation).

Notes

¹ On the one hand, single-item measures do not allow researchers to estimate the measures’ internal-consistency reliability (e.g., Loo, 2002), which is problematic when studying individuals’ perceptions and behavior. If the respondents do not understand the provided items in the same way, the measurements could be incorrect. On the other hand, such measures provide an easy and efficient tool, which is important when surveying difficult-to-study groups such as political communication practitioners. Because the use of single-items measures is a proven and tested procedure in the research on presumed influences (e.g., Cohen et al., 2008; Lim & Golan, 2011), they are also used in this article.

² In addition, the perceived influence on oneself was measured in the same way.

³ The presumed reach and presumed influence are related to one another. Correlation analyses between all related independent variables (for example, the presumed reach of Facebook among journalists and the presumed influence of Facebook on journalists) indicate a noticeable but not overly strong relationship on average (r = .35, p = .000). Because presumed reach and presumed influence are separate theoretical constructs and this article aims to detect the relationship between these constructs on the one hand and the social media use of the communication practitioners on the other, both constructs are treated separately.

⁴ Surveying independent and dependent variables in the same questionnaire poses the risk of common source bias (Jakobsen & Jensen, 2015). Ideally, perceptions (independent variables) would have been measured in the survey, yet behavior (dependent variables) would have been measured via the actual social media activities of the political communication practitioners. This was not possible because the respondents were requested anonymously. However, to reduce common source bias, the independent and dependent variables were measured at widely separated positions in the questionnaire (as suggested by Podsakoff, MacKenzie, & Podsakoff, 2012, p. 549).

⁵ In the additional regression analyses, almost all relationships between the political communication practitioners’ social media activities, the presumed reach of Facebook and Twitter among their colleagues, and the presumed influence of Facebook and Twitter on their colleagues are positive. However, they often failed the statistical significance level of p < .05, probably because of the reduced sample size. Nevertheless, the described relationship was stronger in practitioners who worked for companies compared to those who worked for state institutions or for associations, unions, parties, or civil society organizations.

⁶ Surprisingly, the presumed reach among journalists on Facebook and the presumed reach among politicians on Twitter are slightly negatively related to political communication practitioners’ social media usage. However, bivariate correlations indicate that these results should not be over-interpreted. First, these bivariate correlations show a positive relationship between the political communication practitioners’ presumed reach of Facebook among journalists, the respondents’ calling other peoples’ attention to political issues via Facebook (r = .09, p < .01), and their gathering political information via Facebook (r = .15, p < .001). Second, the bivariate relationships show that there is no significant relationship between the presumed reach of Twitter among politicians and the respondents’ calling other people’s attention to political issues via Twitter (r = -.05, ns) or gathering political information via Twitter (r = .00, ns).

Additional information

Funding

This publication was created in the context of the Research Unit “Political Communication in the Online World” (grant number 1381), subproject 3, which is funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation).