![Sample our Politics & International Relations journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5947/TOC-Subject-Sample-2021-Politics-International-Relations.jpg"})
Abstract
Most scholars agree that news coverage of politics is the product of complicated interaction between journalists and politicians. Yet, we know little about how the interaction affects the coverage. Our analysis examines U.S. senators' press events and subsequent national network coverage from 1980–1996. Our evidence suggests that all senators can increase journalists' interest in their press events by carefully choosing the type of event and which politicians attend. In turn, such interest often translates into actual news stories, although that coverage is not guaranteed. Thus, senators can structure press events in order to increase the likelihood of coverage, but reporters understandably resist their attempts to do so. As a result, the most newsworthy press events require senators to give up control over content, creating more potential for revealing unexpected information.
Notes
1. See CitationKovach and Rosenstiel (2001) for a comprehensive discussion of newsworthiness.
2. The SRTG records contain two additional types of events, which we excluded. First, “special events” include many nonpolicy events, such as receptions for sport teams. Special events made up only 1% of all events in the SRTG records. Second, data on committee hearings only list the senator chairing the hearing, omitting other senators who attended.
3. Unfortunately, the SRTG records do not allow us to determine who organizes each event. This question is important for understanding the dynamics of Senate press coverage: Does a non-leader try to win coverage by trying to get leaders to attend her or his event, or do leaders try to win even more coverage by getting nonleaders to attend “cattle call” press conferences for the party or committee? Given the limitations of this data set, we investigate these questions in other work focusing on a smaller number of events in greater depth.
4. Each record in the Archive provides only a summary of the relevant news story. The summary does not identify whether the story's information came from a particular press event, file footage, or another source. Senators use press events to call attention to themselves. That attention may enhance both their reputations for newsworthiness and their actual coverage, regardless of whether that coverage uses footage from senators' events or other sources.
5. Searches for coverage using other time periods after an event yielded nearly identical results.
6. To address the generational change argument mentioned above, we also created a measure of the average year that senators participating in an event were elected to the Senate. This measure is not identical to the seniority measure. A senator serving in 1980 and one serving in 1996 could each have 10 years of seniority, but the years that they entered the Senate would be very different (1970 versus 1986, respectively). Nonetheless, the two measures are highly correlated (−.74) and produced substantively similar results in the models below. We also tried two alternative estimations to seniority—the total seniority of the senators in attendance and the seniority of the most senior member attending. Both measures produced similar results to the variable for average seniority, and including these alternative measures did not alter the results.
7. Reestimations using each event type as the omitted category did not change substantive results.
8. The rate of attracting cameras or stories may differ between low and high levels. Senators may find it difficult to change from failing to attract any cameras to attracting one camera to an event. But once an event has attracted one camera, it may be easier to attract another camera.
9. The controls produced significant coefficients, largely in expected directions. The exception was the measure of the total number of events: More overall press activity by senators on a particular day is positively associated with greater media interest in an individual event on that day. Many senators may hold events when the national networks are likely to pay attention.
10. We used Clarify (CitationKing et al., 2000) to calculate the predicted values and standard errors. For independent variables that were dummy variables, we calculated predicted values for each independent variable value (i.e., 0 and 1). For all other independent variables, we used a “low” value of one standard deviation below the measure's mean, and a “high” value of one standard deviation above that mean. When calculating the predicted values for a particular independent variable, we held all other independent variables at their means, unless otherwise noted.
11. It is important to note that senators' institutional positions may interact with their attempts to win news coverage. Senators such as Tom Daschle or Bill Frist may win leadership posts precisely because of their ability and efforts to win press coverage.