Social media are often attributed the potential to strengthen democratic processes by bridging the gap between politicians and citizens (e.g., Coleman & Blumler, 2009). Whether this can be realized depends on how politicians use social media in their daily routines. First studies indicate that politicians are better evaluated by citizens when they use social media interactively instead of simply broadcasting information (e.g., Lee & Shin, 2012). However, previous research has paid little attention to explaining why politicians use social media in which ways. Furthermore, research has overlooked the importance of the demand-side, i.e. what users expect from politicians.
Our considerations are backed by the reciprocal effects approach, which describes the influence of (expected) media coverage on the protagonists of this coverage (Kepplinger, 2017), and by the norm of reciprocity, which states that people repay what other people have done to them (Fehr & Gächter, 2000). While the first approach implies that politicians have potential outcomes of their social media activities in mind, the second implies that politicians benefit when they fulfill the expectations of other social media users. However, most social media users do not express their expectations explicitly. Consequently, politicians can only assume what social media users expect from them. As politicians state that they are aware of what citizens expect from them (Lüders et al., 2014), their social media activities should be influenced by their perceptions about the expectations of social media users (H1).
Moreover, according to the equalization-normalization debate, the politicians’ party determines how politicians use social media. Although the impact of party size varies between different social media channels (e.g., Quinlan et al., 2017), politicians from smaller parties tend to be more interactive social media users (e.g., Kalsnes, 2016). Thus, politicians from smaller parties should use social media more frequently for interactive activities than politicians from larger parties (H2).
A possible explanation for this difference is that politicians of different parties act based on different user expectations. As supporters of smaller parties use social media more frequently in an interactive way (Koiranen et al., 2017), politicians from smaller parties (compared to politicians from larger parties) should more strongly assume that social media users expect interactive activities from them (H3).
Finally, we ask whether these effects vary between political levels (RQ1) and between social media channels, specifically between Facebook and Twitter (RQ2).
In spring/summer 2016, surveys were conducted among German national (n=118; response rate=18.6%) and local parliamentarians (n=859; response rate=24.5%). Both samples were approximately representative of these groups.
Controlled linear regression analyses largely confirmed H1: politicians use social media more frequently interactive, the stronger they perceive that social media users expect this behavior from them. Further analyses largely confirmed H2 and H3 for local politicians, but not for national politicians. Moreover, politicians obey perceived user expectations to a greater extent on Facebook than on Twitter.
The results underline the importance of the (perceived) demand side for explaining politicians’ social media communication and demonstrate how the equalization-normalization debate (and others) can benefit from the inclusion of perceptual processes.
