Browse By Day
Browse By Time
Browse By Person
Browse By Division
Browse By Session or Event Type
According to the media-indexing hypothesis, first articulated by W. Lance Bennett in 1990, news media discourse on contentious political issues tends to follow the contours of debate among political elites, rather than intervening in elite discourse to serve as a check on governmental authority. In Bennett’s words (1990, p. 106):
Mass media news professionals, from the boardroom to the beat, tend to “index” the range of voices and viewpoints both in news and editorials according to the range of views expressed in mainstream government debate about a given topic.
While some scholars have contested Bennett’s original formulation of the indexing hypothesis in various ways—ranging from complete dismissal to minor revision (e.g., Althaus, Edy, Entman, & Phalen, 1996; Entman, 2003; Rowling, Jones, & Sheets, 2011)—others have found empirical support for the general hypothesis (Alexseev & Bennett, 1995; Bennett, Lawrence, & Livingston, 2006; Hickerson, Moy, & Dunsmore, 2011; Lawrence, 1996; Livingston & Bennett, 2003; Livingston & Eachus, 1996). This body of literature, which has broadly found that official (usually Congressional) debate “sets the parameters of media debate, establishing the boundaries or agenda of public discussion” (Althaus et al., 1996, p. 408).
As an additional corollary of the core hypothesis, Bennett (1990) further suggested that the inclusion of non-elite voices in news media discourse would nonetheless be constrained in their range of opinions expressed to those viewpoints already circulating within elite debate. That is, in the words of Althaus et al. (1996, p. 408), “the proportions” of support for and opposition to specific issues represented in news media “closely reflect the distribution of expressed views among officials.”
Yet, while the foundational work on media indexing largely focused on issues of inherent mainstream political relevance (e.g., political scandal, military action, and foreign policy; Althaus et al., 1996; Bennett, 1990; Bennett, Lawrence, & Livingston, 2006; Livingston & Eachus, 1996; Rowling, Jones, & Sheets, 2011), topics relating to transgender issues and individuals have only recently emerged from social prominence into legitimate political debate at all levels of governance (Billard, 2016). As such, transgender issues may pose an interesting challenge to the indexing hypothesis in that the centrality of institutional politics to transgender concerns is relatively novel, despite the rapid proliferation of attention in policy and legislation. Furthermore, investigating how media indexing occurs in the transgender context may illuminate the dynamics of media and governance that drive both public opinion and policymaking on transgender issues.
We thus conducted a quantitative content analysis of both elite discourse and news media content focusing on issues relevant to the transgender community in the United States, following the example of Althaus and colleagues (1996). Like Althaus et al. (1996), two samples—one from the Congressional Record and one from the New York Times—were drawn for this study. For our purposes, the Congressional Record served as an independent measure of elite debate against which we could compare news content to assess the indexing hypothesis. For both samples, the sampling timeframe was 1 January 2015 to 31 December 2016—a timeframe we selected because of the numerous national political debates about transgender rights that emerged in the period. The Congressional Record sample was drawn from the official record at Congress.gov, and all search results for the keywords within the time frame of interest (N = 119) were downloaded. The New York Times sample was drawn from the LexisNexis database on 3 October 2017. Again, all search results for the keywords within the time frame of interest (N = 1,741) were downloaded. However, we were interested only in those articles likely to pertain to political issues, and thus restricted our sample to articles published in the Foreign Desk, National Desk, Politics, and US sections of the Times (N = 397).
Overall, our results suggest that the media-indexing hypothesis does not fully explain patterns of new media discourse on transgender rights issues. Not only did news media cover a broader range of issues than elite political debate, but political elites were also underrepresented as sources in news media coverage relative to citizens and other non-governmental sources. Finally, the relative proportions of pro- to anti-transgender valence expressed in news media did not mirror the proportions expressed in elite political debate, indicating a rather high level of press independence in the context of transgender rights issues. These results might suggest that the indexing hypothesis is limited in its application to certain policy/issue areas, and that, accordingly, the press maintains different levels of independence for political elites in different issue domains.