This thesis examines intergroup biases that exist both in the reporting of Christian-Muslim conflict in Indonesia, and in the perceptions of media reports of Christian-Muslim conflict in Indonesia. Most existing research on media bias in Indonesia has taken a descriptive approach, utilising either sociological or journalistic perspectives. This thesis will employ an experimental, quantitative approach, couched within the language and theoretical framework of social identity theory.
The primary focus of Study 1 is to examine whether the newspapers showed bias in the extent to which they were prepared to identify the religion of those who engaged in violence. It was predicted that newspapers would be more inclined to specifically identify the religion of those committing atrocities when the atrocities were committed by outgroup members than when they were committed by ingroup members. When atrocities were committed by ingroup members, it was expected that newspapers would use more ambiguous language in association with the protagonists so as to disguise the role their religious group played in the conflict. Two hundred and thirty five articles reporting Christian-Muslim conflict in Ambon were coded according to the group allegiance of the newspaper and the language they used to describe the perpetrators of violence. Evidence supporting the predictions was found for the Muslim newspaper. The Christian newspaper, in contrast, was shown to use ambiguous language throughout.
The remainder of the studies do not focus on bias within the media per se, but on the biases that exist in how people perceive the media. Research on the hostile media bias shows that, when exposed to standardized news footage, both sides of a conflict tend to see the footage as being biased against their group. Two studies were conducted in which participants read a standardized article on the conflict and rated the extent to which they perceived the article to be biased against their group (Studies 2 and 3). In addition to testing the cross-cultural generalizability of the hostile media bias, a primary aim of these studies was to examine whether the group allegiance of the media has an effect on perceptions of bias. Participants were either told that the newspaper article was published in a Christian newspaper, published in a Muslim newspaper, or they were given no information about the group allegiance of the newspaper. Study 2 (N= 97) was conducted using participants who were residents of Ambon. Study 3 (N= 238 Jakarta residents) was designed to overcome some of the methodological problems within Study 2. Results showed evidence for a hostile media bias among both Muslims and Christians. Furthermore, Study 3 revealed that both religious groups expected the newspapers to favour their own religious group. This, in turn, resulted in participants perceiving that the articles were biased in favour of the religious group to which the newspaper is affiliated.
Study 4 was designed to examine the extent to which Indonesians display the intergroup attribution bias when reading reports of atrocities between Muslims and Christians. Previous research has shown that people attribute events in a way that protects or enhances the integrity of their own group. However, because this research has asked people to reflect on real historical events, it is not clear the extent to which the bias is in the minds of perceivers or whether it is a response to the attributional biases that exist within the media to which they are exposed. To overcome this problem. Study 4 (N = 282) used a description of an event that was fictional but which the participants themselves believed to be real. An added benefit of this paradigm is that it becomes possible to conduct a fully controlled design where people are participants are exposed to the same events, but are led to believe that the negative actions were committed either by Muslims or by Christians. Consistent with the intergroup attribution bias. Christians (but not Muslims) used stronger situational attributions for the events described when the perpetrators of the violence were ingroup than when they were outgroup members. On ratings of dispositional attributions, both Muslims and Christians displayed an effect in line with the intergroup attribution bias.
Finally, two studies were conducted to examine the extent to which reactions to negative media commentary are affected by the group membership of the source of the message. These studies were based on the intergroup sensitivity effect, which describes the tendency for people to be more defensive in the face of group-directed criticism when it comes from outgroup members than when it comes from ingroup members. The goal of Study 5 was to examine the extent to which the intergroup sensitivity effect was generalizable to a context where the groups are engaged in high-level conflict. Muslim participants (N= 191) were exposed to criticism of their religious group stemming from either another Muslim or from a Christian. Prior to reading the criticism, participants were either primed with a newspaper article describing inter-religious conflict or they were not. Where conflict was not made salient, the intergroup sensitivity effect emerged. Where the conflict was made salient, however, defensiveness in the face of ingroup criticisms increased to the point where the intergroup sensitivity effect was eliminated. The goal of Study 6 (N = 191 Muslims) was to examine also whether responses to criticism are affected by the group membership of the media through which the comments are reported. Although there is a general tendency for people to be surprisingly tolerant toward ingroup critics, it is questionable whether this tolerance would remain even if the criticisms were made to an outgroup audience. To examine this we exposed Muslims to criticism of their religion from either another Muslim or a Christian. Furthermore, the criticisms were made to either a Muslim or a Christian newspaper. Although the intergroup sensitivity effect was again replicated, there was limited evidence that the group membership of the audience played a role in influencing participants' responses to the comments.
Overall, these 6 studies demonstrate how group identities influence how intergroup conflict is reported in Indonesia, and also how media reports on intergroup conflict are interpreted. Practical and theoretical implications of these findings are discussed.