![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
In this study, I examine how the local news media covered members of Congress tied to the Jack Abramoff scandal during the 2006 congressional elections. Previous research suggests heavy coverage of any politicians tied to scandals, particularly during the campaign season. Using a multilevel data analysis approach, I show the local news media strategically considered whether to cover members under suspicion of scandal during the election season taking into account race competitiveness and challengers’ actions. Specifically, local newspapers paid the most attention to the Abramoff scandal when the incumbent was in a competitive race and the challenger was actively pushing the scandal.
Notes
1. Scholarship has established that not all members of Congress involved in scandals suffer electorally. For example, research has shown that the check-kiting scandal had only a small effect on electoral margins in the 1992 congressional elections (CitationAlford, Teeters, Ward, & Wilson, 1994; CitationBanducci & Karp, 1994; CitationDimock & Jacobson, 1995; CitationGroseclose & Krehbiel, 1994; CitationJacobson & Dimock, 1994). The scandal did increase the number of members who retired and tripped up in primaries, though as CitationGroseclose and Krehbiel (1994) point out, the effect was tempered by the closing of a loophole in the Federal Election Campaign Act that allowed members to pocket campaign funds.
2. Since the Department of Justice does not list members of Congress under investigation, it is difficult to precisely identify who was being considered for prosecution for dealing with Abramoff. The members in this study were identified using information and reports from the Center for Responsibility and Ethics in Washington, ProPublica.org, Sourcewatch.org (of the Center for Media and Democracy), and the Washington Post. Dennis Hastert (R-IL) and Richard Pombo (R-CA) were also considered for inclusion in the study. Hastert's name came up in an ABC News report in May 2006, but the Justice Department explicitly stated that Hastert was not under investigation for dealings with Abramoff (CitationRoss, 2006). Hastert's ethics troubles in 2006 came instead from his handling of the Mark Foley page scandal as speaker. Pombo worked with Native American tribes on protecting their casino interests during this same time period. However, Pombo appears not to be tied to Abramoff's lobbying efforts (CitationRichman, 2004).
3. It is possible that the reporting strategies inside and outside of the election season were markedly different. To examine this possibility, local news coverage of the seven target members was also collected from January 1, 2006, to Labor Day 2006 (Abramoff pleaded guilty on January 4, 2006). Local Abramoff coverage linked to these members did spike in January, with a majority of information sourcing coming from the incumbents and journalists (63% and 59% of the stories, respectively). While Abramoff coverage died down in the following months, it did not disappear. Members expected to be involved in competitive races (e.g., Conrad Burns and John Doolittle) continued to have the Abramoff scandal covered, with an increasing amount of sourcing shifting to challengers and their supporters. This source shift began in the early spring and continued through the primary season and into the election season. For example, by March challengers and their supporters became the dominant information source for Abramoff-related stories. Clearly, events do matter for explaining reporting strategies, but for sustained scandal coverage, in order to assuage the costs for local journalists, strong and vocal opponents appear necessary.
4. This technique has been used in other studies on local news coverage of members of Congress (e.g., CitationFogarty, 2008).
5. Blunt and Akin are the most spatially disparate House members in the data. The reasons for their pairing include that they had similar political profiles and Blunt's adjoining districts included a Democrat (Ike Skelton) and a woman (Jo Ann Emerson). Further, the Springfield News-Leader is only read in Blunt's district, and so trying to compare coverage in the paper for adjacent members would be fruitless.
6. To date, none of these members have been convicted of crimes related to Abramoff. Trevor Blackann, a former aide to Roy Blunt, was convicted of lying on tax returns about gifts and money from lobbyists, but Blunt, now junior senator from Missouri, has not been affected (CitationKravitz, 2008).
7. The coding was performed by the author and two graduate assistants. On a random sample of stories, coding reliabilities (Krippendorff's alpha) were .78 on the issues in the stories, .76 on the sources in the issues, .85 on whether the member played a major role or not in the story, and .77 on the word count (CitationKrippendorff, 2004).
8. Journalist sourcing refers to journalists including their own research and making statements not attributable to an outside source.
9. Although a somewhat arbitrary cutoff point, most stories either had the member as the featured and central subject or had the member as tangentially involved.
10. In this study, letters to the editor were excluded in the analysis as they provide less information about the strategies and choices made by the local news media in covering the campaigns.
11. As previously mentioned, many of the incumbents spoke on their role in the Abramoff scandal in January 2006. Past research suggests no added electoral benefit for incumbents to continue to speak on the matter.
12. Roy Blunt as majority party whip had the strongest connection with Foley among the members in the data. The Republican leadership was criticized for poor handling of the Foley scandal, but this had little effect on Blunt's coverage or reelection.
13. I also considered papers’ congressional candidate endorsements and circulation levels as paper-level explanatory variables. On the paper level, there is even less variation, as only 10 papers were included in the data. Paper endorsements had a statistically significant negative effect on the probability of a paper reporting on Abramoff. However, when controlling for any second-level variable (paper or challenger level), this effect disappears. To keep the models parsimonious, these two variables were dropped in predicting coverage of the four issue areas.
14. That is, challenger press releases linking the incumbent to Abramoff were used as a predictor of Abramoff-related coverage, challenger press releases bringing up the Foley scandal were used as a predictor of Foley-related coverage, and so forth. Challenger press releases come from the candidates’ Web sites, accessed via the Internet archive.
15. Ideally, all potentially pertinent independent variables on the candidate level would be included. This includes incumbents’ seniority, committee and leadership status, and other campaign information. Due to the small number of races and based on previous research (i.e., CitationArnold, 2004), the focus was on the challengers’ actions instead of the incumbents’ actions. The ads and event schedules of incumbents’ and challengers’ campaigns were also considered. However, not all candidates included this information on their campaign Web sites or, in some cases, the candidates did not have any campaign advertising.
16. Challenger press releases on Abramoff were also statistically significant when controlling for the explanatory variables that were excluded. Hence, the relationship between challenger Abramoff-related press releases and Abramoff-related coverage does not appear to be spurious.
17. Thanks to a reviewer for pointing out the potential temporal nature of coverage and press releases.