Die Wirkung ?guter“ Argumente

Starting from a discursive perspective on public communication, we ask whether the deliberative quality of journalistic articles on policy topics influences the amount of user participation on news websites. The theory of deliberative democracy often serves as a source of normatively desirable discursive requirements. Here, however, we present empirically verifiable assumptions that we combine with explanations about media content from the spiral of silence theory and explanations of user behavior from the civic voluntarism model and the construct of self-efficacy. The spiral of silence theory assumes that people do not express their opinions in public when they feel their opinions belong to the minority, because they fear to be socially isolated. The media play a crucial role in this process, as their depiction of public discourses influences the perception of public opinion. Furthermore, the theory assumes that mass media form a media tenor, presenting similar representations of opinion distributions. In online contexts, we argue, we can no longer expect to observe a single media tenor, but we must define more individually perceived, selection-based media tenors. Thus, users are only to be expected to fear isolation in cases where a majority opinion is presented in online contents they selected which stand against their opinions. Based on the civic voluntarism model and previous literature, we understand writing comments as a form of participation that is most likely to be carried out by persons who are interested in politics, have a high level of self-efficacy and feel stimulated to participate by the immediate environment, i.?e., the journalistic article and the comment section.We therefore assume that the more representatives of the opposition, civil society and ordinary citizens are presented in the media, the more users will participate in the comment section (H1). Furthermore, we also expect articles containing conflicting and counterarguing utterances to lead to more participants than articles without this kind of utterances (H2). Regarding the rationality of the discourse, we suppose that the presence of justifications of arguments also increases users’ participation (H3). Articles often report on current debates that aim to lead to a decision. Considering self-efficacy assumptions and the relatability of a debate for the users, we assume that participation rates will be higher when decisions are discussed that lie in the future (H4a), whereas decisions that have already been taken are expected to reduce participation rates (H4b). Furthermore, we also expect articles with a reference to German domestic politics to attract higher participation rates (H5).In order to test our hypotheses, we conducted a content analysis of 400 online newspaper articles. We selected two national (WELT Online and ZEIT Online) and two regional (Rheinische Post Online and Tagesspiegel Online) online newspaper outlets, one of each belonging to the left-wing liberal and one to the conservative journalism spectrum. From the politics news feeds of the four online newspaper outlets between January 1, and March 31, 2016, we selected a systematic random sample of articles with enabled comment sections. The deliberative criteria were measured either on the level of a value-based utterance made by a representative of the government, the opposition, civil society or an ordinary citizen or on the level of the news article. Utterance-based criteria were then aggregated to the news article level. Furthermore, we measured the number of users that participated in the discussion in the comment section. We then estimated a zero-inflated negative binomial regression model in order to predict the number of participating users.Results show that the majority of deliberative criteria serve to increase the amount of user participation. The more opposition politicians are cited in a news article, the more users participate in the comment section which hints to the conclusion that discursive equality triggers participation. Regarding the reciprocity of discourses, conflicting and counterarguing utterances also increase the number of users who participate in a discussion. Another explanation of a higher number of commentators is the presence of justified arguments, i.?e., when the discourse can be classified as rational. While future decisions did not show any effect on user participation, past decisions in fact reduced the amount of participation. Furthermore, if the topic was related to domestic German politics, participation rates also increased, which represents relatability for people’s lives.All in all, we conclude that discourse-related criteria should be integrated more often into the analysis of the conditions of user participation, as they can add to explaining when people feel motivated to write comments. They are connected to considerations about opinion climates both as depicted by the journalistic article and as present in the comment section itself as discussed in the spiral of silence theory. The more opinions journalists present in an article and the better the underlying discourse, the more we can expect users to participate in the discussion. We have also seen that articles on decisions that have already been taken are not as frequently commented on, and that a relation to domestic topics leads to a more active participation. Taken together, we argue that user comments can be a meaningful addition to deliberation research, because they can serve as a measure of the effects of deliberative content, e.?g., in the news.This study focused on the contents of the journalistic articles, while the contents of user comments were not integrated into the analysis. In future studies, this gap should be closed with content analyses of user comments. Furthermore, the assumptions we made about users’ motivations and behavior also need to be confirmed in experimental designs and surveys.