The Media in Elections: Methods ol Study
Some (Further) Issues and Difficulties
in Methods of Studying the Media in Elections
Let us now discuss one problem in the method of content analysis that has been common to most studies using that method, including the Democracy '92 study the author was involved in. A central question in studies of the news media's coverage of election campaigns has been the extent to which policy issues are covered versus coverage of the "horserace" (which candidate is ahead and which behind) and of hoopla and events. Clearly, this is an important question if we care about the effectiveness of the democratic process. Unfortunately, most studies have simply counted the number of "issue mentions" in each news story— or even in each message or "statement" for those that have used the more minute content analysis. The total of such mentions is then computed and con¬clusions are reached regarding "how much issue coverage" was carried in the news. Frankly, this is not legitimate; in fact it is downright misleading, because any mention by a candidate, reporter, or other source in a Story of even just the name of an issue is counted as an "issue mention." For example, if a news story during the 1992 election said, '"George Bush talked about the environment today," and then moved on to other topics, this would have been content coded as an issue mention. But there is no meaningful issue content in that story from which the public can learn something. Most importantly, when an aggregate total is presented registering that X percent of all news stories included such "issue mentions," we have no way of knowing how many in that total were actu¬ally just meaningless mentions of the name of an issue or other such meaningless reference. This also means that, in practice, even the low totals of stories (or messages) with "issues" in them found in many media and election studies are probably overstatements of real issue coverage.
A content category used in the Democracy' '92 study does provide a more meaningful approach: the coding of what was the "main focus" of the story— that is, was the news story principally about one or more issues or principally about a candidate (such as a biographical profile), the campaign process, or other matter? In general, to paraphrase a comment by Thomas Patterson, we need to move beyond a focus on the simple quantity of issue coverage in the news to analyzing the quality of the issue coverage; this is what is most meaning¬ful for the democratic process.35
Finally, let us take note of some methodological issues raised in an interest¬ing and significant study by Michael Robinson and Margaret Sheehan.36 They sought to add depth to analysis of what election news communicates to the pub¬lic in a presidential election.
In their study of the media in the 1980 campaign, they did not seek to ana¬lyze the impact of communications on the public, only the content of what news was communicated. They were especially interested in comparing print with
285
Chapter 10 The Media in Elections I
TV coverage to see if and how they differed. They selected CBS to represent the network news shows (at that time, the most watched and respected network news), and, to represent "traditional" print news, they selected a wire service, UPI (AP refused to rent them a daily wire). By using a wire service, they sought to overcome the limitations of selecting one or a few particular newspapers to represent what people throughout the nation receive from print. But did they also raise problems by this choice? The wire itself is the general feed to news¬rooms; which stories among the many transmitted are actually selected by edi¬tors for inclusion in the day's newspaper is a different matter. Thus, content analysis of the wire service stories does not tell us what specific communications actually reached the public. It does provide a rough "average" of what is used throughout the nation, but only a rough average. Also lost in using the wires are any locally covered stories or national stories covered with a local angle—a loss that is more telling with respect to the major papers, which have one or more of their own reporters on the presidential campaign trail.
The most interesting aspect of the Robinson-Sheehan study is the topics on which they chose to focus analysis. The prime goal of the study was to evaluate the performance of the news media in their coverage of the election. They used five criteria to accomplish that laudable task, which they "distilled" from several lists of principles enunciated by news media organizations. Those criteria are: (1) "objectivity"; (2) "equity of access"—did candidates for the same office receive essentially equal time on and in the news media?; (3) "fairness"—beyond simple amounts of time or space received, how balanced and fair was the treat¬ment of candidates for the same office in the "tone" of the stories?; (4) "serious¬ness"—how much of the news involved the substance of what is at stake in elec¬tions, such as policy issues and candidates' qualifications, rather than election hoopla, speculations about who's ahead, personality stories, and the like?; (5) "comprehensiveness"—how thoroughly were all elections covered, including major elections beyond the presidential one?
That is certainly a sensible list of criteria; it gives promise of taking us a step beyond simply describing amounts of coverage, allowing construction of a com¬posite view of the adequacy of coverage and thereby an evaluation of it. But how are these criteria spelled out so that they can be used in the analysis—how are they operationalized? For the present purposes we will concentrate on the first and third of those criteria.
The criterion of objectivity raises some interesting questions. Robinson and Sheehan suggest the following as the "best" definition of objectivity:
Objective news. (1) relates only observable facts of an overt event; (2) cites others on mat-ters of opinion; (3) refuses to allow one's own beliefs, principles or inclinations—"or even his own knowledge"—to color raw, overt material for the story. Objectivity includes, above all, a considerable reluctance to go beyond what was actually observed by the journalist,
286
The Media in Elections: Melhods oi Study
and an even greater reluctance to draw any explicit conclusions or inferences about the events being covered, unless the conclusion or inference comes from another legitimate source, not the journalist him/herself.37
From those standards they derived four ways to test for objectivity in news coverage. First, they analyzed what news people said to gauge the extent that CBS and UPI "drew explicit and unsupported conclusions about the personal qualities of the candidates." Second, they identified any personal opinions expressed by reporters on the policy issues. Third, they "measured the degree to which both media broke away from 'description' and moved toward 'analysis' in covering candidates or issues, defining objectivity here as a function of basic descriptiveness." Fourth, they analyzed the verbs used by reporters to help determine how much "insinuation" was included in reports.38
Is a definition of objectivity adequate if it restricts reporting to "observable facts of an overt event" and, at the core of the "test" for objectivity, defines this as "a function of basic descriptiveness" and disallows use of "even [a reporter's] own knowledge" to "color" material for a story? The discussion early in Chapter 6 explains why this is not an adequate definition. The control of news media access by some presidents, combined with their intensive and comprehensive strategy for presenting a positive picture of the president and for manipulating the news in general, as discussed in Chapter 9, provides further reason for con¬cluding that such a notion of objectivity is inadequate. The 1972, 1984, and 1988 elections are dramatic illustrations of the problem with such a narrow def-inition of objectivity. In each of those elections, the incumbent simply did not allow himself to be questioned or challenged by newspeople but did appear at a series of carefully staged overt events as part of a general carefully crafted image purveyance. As David Broder has said: "The lessons of the 1972 Nixon cam¬paign—in bypassing the press and refusing to talk about the substantive policy choices ahead—were applied by Ronald Reagan in his reelection campaign in 1984, with even more lopsided and politically satisfying results."39 An objective coverage of the 1984 campaign would have centrally pointed out that the demo¬cratic process was being short-circuited and that the campaign-for-the-cameras was calculated image-building, not a presentation of the policy record. To rate news as "objective" because it described a series of carefully staged events is to suggest the "news" need not relate to the substance of the electoral process or the actual function of governing. As pointed out in Chapter 6, one cannot be objective without perspective.
That discussion brings us to another problem with how Robinson and Sheehan operationalized the criterion of objectivity and with their method more generally. Strikingly, the visual dimension of the news was ignored in their con¬tent analysis; only verbal messages were considered. Robinson, along with Maura Clancy, did a study of 1984 elecdon coverage in which the visual dimen-
287