This research used a series of qualitative measurements of media coverage to investigate how differences in characteristics of a terrorist related event correlate with qualitative differences in media coverage. The first part of this study determined that there were tools to measure differences in the quality of coverage. Three variables showed significant differences in coverage. Coverage differed in the structure of the news account- in whose shoes the reader enters the story. The differences between entering through the perspective of the victim, the perpetrator or the context have been correlated by Cerulo (1998) with different messages of the legitimacy of the actors. Victim sequences signal deviant (illegitimate) violence, perpetrator sequences signal legitimate violence and contextual sequences signal ambiguous violence. Coverage also differed in the extent to which an article provided contextual information or focused strictly on the details of the event. Providing contextual information is important for terrorist groups because it includes information on the grievances of the group as well as the history of the conflict.
This variable was measured as an episodic or thematic frame. Explanations of motivation for participation in terrorism also differed based on characteristics of an event. As with contextual coverage, presenting themes of causation or motivation for the account is a way for terrorist groups to present grievances and history of the conflict. Combining these three variables into a favorable coverage variable helped makes sense of competing trends in the data. This first section set up a system for evaluating the qualitative impact on media coverage of choices that terrorist groups and governments make. What stands out is a paradox for a terrorist group around the use of violence.
Both here and in other studies, violence has been shown to be an effective means of getting through the media gatekeeping and achieving coverage, but it is also associated with a decrease in favorable coverage. Number of casualties is also negatively associated with favorable coverage. Hence the paradox that, in order to achieve coverage, based on criteria of newsworthiness, violence may often be necessary, but violent action actually decreases the number of articles presenting the kind of information terrorist groups want to get across. Looking at the paired cases, what was most significant was the lack of change in the favorability of coverage before and after events.
The implication is that while terrorist groups may have some control over whether or not their actions get covered, media organizations develop fairly resilient patterns for covering those actions, irrespective of the nature of the action. Terrorist groups essentially have less capacity to actually manipulate the type of coverage they receive than is commonly believed. While there were some very small effects, the favorability of coverage immediately following an event is essentially the same as before it. The difference lies in the actual amount of coverage. While short-term impacts were slight, there are substantial differences both in quantity and quality over the life of the conflict, a longer term view may allow for better understanding of changes in media coverage.