|
PRIMING EFFECT OF MEDIA ON POLITICAL PERCEPTIONS.
Term Paper ID:26723
|
|
|
Essay Subject:
Analyzes article on the theory that media coverage affects perceptions of President's performance, focusing on Iran-Contra scandal.... More...
|
9 Pages / 2025 Words
1 sources, 13 Citations,
APA Format
$36.00
Return to List of Papers
|
Paper Abstract:
Analyzes article on the theory that media coverage affects perceptions of President's performance, focusing on Iran-Contra scandal.
Paper Introduction:
Krosnick and Kinder (1990) conducted a test of their theory of "priming" in a study based on data gathered by the 1986 National Election Study (NES). The data was compiled in the period preceding and following the 25 November 1986 announcement by the Reagan White House that funds secured from the sale of arms to Iran had been diverted by members of the President's National Security Council (NSC) to the Contra forces fighting to overthrow the Sandinista government of Nicaragua. Priming deals with the effect of news media attention to certain stories on individuals' assessments of the President's performance and had, according to the literature reviewed here, been supported by experimental tests. The authors felt that the fortunate chance of a major scandal occurring in the midst of a major information gathering initiative provided them with a rare opportunity to
Text of the Paper:
The entire text of the paper is shown below. However, the text is somewhat scrambled. We want to give you as much information as we possibly can about our papers and essays, but we cannot give them away for free. In the text below you will find that while disordered, many of the phrases are essentially intact. From this text you will be able to get a solid sense of the writing style, the concepts addressed, and the sources used in the research paper.
R. Yet the present test is, supposedly, based on a single incident andKrosnick and Kinder lay special emphasis on the fact that their sampleconsisted of subjects tested before and after this single announcement.Since their theory seems to involve changes in the impact of the mediaattention dependent on duration and intensity of attention to a particularevent or circumstance it is all the more surprising that they do notacknowledge this seeming discrepancy between the statement of the theoryand the circumstances of the test. Nor do they indicate whether "roughly midway" refers to atime span or to the number of interviews completed. This occurred on 3 November. In the first place, the authorsclearly lay out the history of the scandal which began with accusationsthat the President had traded guns for hostages based on the revelation ofthe sale of arms to Iran. 5 4). 498). Yet, far more than the events in CentralAmerica, these situations had surely been important determinants ofpresidential approval ratings throughout the President's tenure--especiallysince the hostage situation, and supposed Republican superiority ininternational affairs, had been major factors in his election. 498). 5 1). The basic question posed by a test of the authors' theory is open-ended and nonbiasing: whether "public assessments of the president'sperformance . But the effect of thelack of clarity about prior priming effects and of the failed attempt tostudy a supposedly distinct event may have had a strong influence on theresearchers' secondary claims. (199 ). international strength, national economic assessments,and the desirability of programs that provide assistance to AfricanAmericans. But there issimply no way to tell because Krosnick and Kinder disregard all theseconsiderations in their effort to create the impression that they trulyhave a single "event" whose priming effects can be studied in relativeisolation. On the one hand, the hiddensupport for the Iranian war effort would have surprised those who had longbeen "primed" to view Iran as a kind of ultimate enemy of the United Statesand those who assumed that U.S. They do not, however, demonstrate thisin any form except the secondary analysis of the data which may be alreadyhave been so distorted by the enormous priming effect of the earlierrevelations as to produce little useful information about the primingeffects of their supposed single event. But the use of such a singular event and, indeed, the authors'identification of the 25 November announcement as a single event alsocontribute to their definitional difficulties and to problems with thesecondary analysis of the NES data. strength rather than guile was the basis ofthe President's successful negotiation of the hostages' release--and thesetwo groups were not necessarily exactly the same. An ideal test of the theory would, of course,be surveys of the same population before and after an event. 5 4). In thereview of the literature, which is limited (of course) to Kinder's ownstudies, the authors also indicate that the experimental tests of thetheory were based on the effects of media attention to aspects of socialconditions rather than events per se. The last two questions were chosen as a control measure becausethey have been found to be "highly relevant to presidential evaluations andbecause they are utterly unrelated to the Iran-Contra revelation" (p. But they alsoacknowledged that they could not actually determine whether or not the twogroups "still differ[ed] from each other in consequential ways that [they]missed" (p. 498). . The President madestrong denials of this charge (while admitting that some arms had been soldto the Iranian government) on 13 November and 19 November. 497) and"the more attention the news pays to a particular domain--the morefrequently it is primed--the more citizens will . Thecircumstances, however, do not appear to have been quite so fortuitous asthe authors seem to believe; largely because the "event" that they studiedwas part of an ongoing scandal and they could not control for possibleeffects caused by the initial stages of the affair. In the formercase the "character" factor may have been far more heavily influenced bythe initial revelation of the President's apparent duplicity regarding thesale of arms to Iran and the possible relation of this to the release ofthe hostages. One could infer from these remarks that the authors held that thefrequency (e.g., number of days the story received focused attention) andintensity of that attention (e.g., placement of stories, number ofheadlines, length of time on the front page, secondary attention by newsanalysts, and so on) create the priming effect. This would have meant that the control population (thoseinterviewed prior to 25 November) had already revised their estimates ofthe President's character to a considerable extent and, merely as wildspeculation, the second major revelation could even have reversed somepeople's estimates of his character upward. But thenature of the "event" is problematic in this study and this is related toKrosnick's and Kinder's statement of their theory. They do not, however, explainwhether the remainder of the subjects were all interviewed on 26 Novemberor whether the remaining interviews were stretched out over the next two orthree weeks. The graph features the findingsof all three of the polls mentioned yet the detail is insufficient todemonstrate any differences in the approval ratings relative to the eventsof 3, 13, 19, and 25 November. As to the other secondaryclaim, the priming effect on "experts" could have been so intense in the 21days that preceded the event of 25 November as to eliminate any appearanceof a substantial effect on those "experts" after that date. Just as it is impossible to determine from theinformation provided whether the bulk of the extremely sharp decline inapproval graphed by Krosnick and Kinder relates to one or all of theseevents it is also impossible to say what effects the earlier events had onthe sample population as a whole. ReferenceKrosnick, J. The effect of their report is to give theimpression that the responses of their sample group were given within a dayof the event, although, on reflection, this seems unlikely to be the case. Based on the strength of the correlation (r = .42) between favoring oropposing aid to the Contras and favoring or opposing intervention inCentral America the authors "averaged attitudes" toward these factors "intoa single measure" (p. Within these limitations, however, the perfectly straightforwardanalysis of a priming effect of some sort (since duration of priming andrelation to previous events are unclear) demonstrated that the subjectssurveyed tended to hold the President to a standard that placedconsiderably greater emphasis on Central America and the Contras than didthose who were interviewed prior to 25 November. This would not be a terribly serious omission except that it issomewhat misleading because the priming effect they feel they havedemonstrated should at least be clearly shown to have occurred either in asingle day, or two, or over time. incorporate what theyknow about the domain into their overall judgment of the president" (p.5 ). Polls conducted byGallup, ABC with the Washington Post, and CBS with the New York Times"register sharp declines in public support for President Reagan'sperformance, roughly coincident with the Iran-Contra revelation" (emphasisadded, p. 5 9). They say, for instance that "the more attention media pay toa particular domain--the more the public is primed with it" (p. The data was compiled in the period preceding and followingthe 25 November 1986 announcement by the Reagan White House that fundssecured from the sale of arms to Iran had been diverted by members of thePresident's National Security Council (NSC) to the Contra forces fightingto overthrow the Sandinista government of Nicaragua. Part of the motivation for the study was thesudden, precipitous decline in the President's approval ratings identifiedby other, unrelated, surveys conducted by news media pollsters and theauthors state that they wished to discover whether this decline in approvalwas, in fact, caused by individuals' tendency to respond to priming. This isfurther complicated by the fact that the Iran-Contra scandal as a wholecreated a certain amount of confusion because the administration appearedto have acted in a very ambiguous manner. depend upon which pieces of political memory come mostreadily to mind" (p. The authors disregarded data collected on25 November and determined that the segments of the sample tested beforeand after 25 November, were "essentially indistinguishable" from each otherin terms of demographic characteristics (p. Nor do Krosnick and Kinder account for possible effects that mighthave resulted from the passage of time and intensity of news coveragewithin their sample. . 5 2). . It is confusingbecause they appear to refer only to the 25 November "revelation" but stilluse the overall term "Iran-Contra" which would seem to encompass a numberof startling revelations made over the course of November 1986. . Once again, this would not be an extremely serious omission if itwere not for the fact that the authors' own theory seems to state that theeffects of priming are cumulative and yet, in trying to distinguish the 25November announcement as a single event, they ignore its relationship tothe preceding events. Theauthors also provide a graphic representation of the decline in thePresident's approval ratings relative to his ratings over a periodstretching from January 1985 to June 1987. In addition, theauthors' statement of their theory is somewhat confusing and raisesquestions about what, exactly, they were demonstrating in this study. In the discussion of the theorythey note that the information that is most accessible to individuals whenthey make their judgments of presidential performance "is determined by theprevailing economic, social, and political conditions of the times [and]for their knowledge about such conditions, most citizens of course rely oninformation and analysis provided by mass media" (pp. In their design the questionsfrom the NES data selected as forming a model of approval of presidentialperformance were those dealing with U.S. American Political Science Review, 84, 497-512. Krosnick and Kinder performed a secondary analysis of the NES dataand demonstrated to their satisfaction that their priming theory did indeedexplain changes in individuals' responses to questions about thePresident's performance. Even more importantlythey did not study the correlation between attitudes regarding Iran andattitudes toward Central America. This becomes problematic because of the approach the researchers taketo the secondary analysis of the NES data. On the other hand, thosewho supported the President's public position on Iran might also have beenlikely to support his covert actions in Nicaragua. Altering the foundations of support for the president through priming. There is no way todetermine, therefore, whether the implications of the authors' theoryregarding the possible cumulative effect of priming relative to durationand intensity of coverage might not have had an effect on their samplepopulation. The newsabout the secret, illegal support for the Contras "immediately took overthe national news: suddenly, and dramatically, Nicaragua and aid to theContras were the focus of front-page stories" (p. To offer only two examples, they comparedthe effect of the event on evaluations of the President's competence and onhis character and they studied the difference in the effect of priming on"novices" and "experts" (as determined by subjects' ability to identify amajority of important political figures in a brief list). They describe the priming in theexperiments as "television news stories focusing on national defense" and"stories about inflation" and these types of priming experiences would seemto differ considerably from the scandal-generating announcement made on 25November 1986. Then, in theauthors' words, "On 25 November, the focus of the scandal abruptly shiftedaway from Iran and the arms-for-hostages question" (p. Thesampling methods of the NES, "a national probability sample of 2,176 U.S.citizens of voting age," presented no problems for secondary analysis andKrosnick's and Kinder's design broke the survey population into threesections: individuals who responded to face-to-face interviews before, on,and after 25 November (p. The single extremely sharp decline in theratings in November 1986 is all that is available in Krosnick's andKinder's study. 5 2). Finally, it should be noted that because the authors' theory has suchgreat intuitive power (that is, it seems to assert a commonsense principal)this should cause them to be even more scrupulous about definitions and thedetails of data accumulation and interpretation. Here Krosnick and Kinder have, unknowingly perhaps, made aswitch, within a single paragraph, from their account of an isolated event(the defining event around which their study is constructed) to seeming todefine it as a second or third major event in an ongoing scandal.Tellingly, Krosnick and Kinder refer repeatedly, throughout their study, tothe "Iran-Contra" scandal and their reference to an "Iran-Contrarevelation" in the singular is extremely confusing. The potentialconfusions generated by the ambiguities of the Iran-Contra scandal,therefore, make it a particularly poor choice for the type of studyconducted by Krosnick and Kinder; not least because the priming effects ofthe 3, 13, and 19 November revelations may, therefore, have already beenresponsible for the sharp decline in the President's approval ratings eventhough the researchers' seem to suggest that there was a separate effectfor the 25 November announcement. Krosnick and Kinder (199 ) conducted a test of their theory of"priming" in a study based on data gathered by the 1986 National ElectionStudy (NES). Theauthors felt that the fortunate chance of a major scandal occurring in themidst of a major information gathering initiative provided them with a rareopportunity to perform an empirical test of their theory. Their findings, however, are far more tentative than they wouldappear from such a superficial reading. 5 1). It also appears that theirtheory was primarily concerned with ongoing circumstances or situationsrather than specific, time-limited events. aid for the Contras, U.S.involvement in Central America generally, attitudes toward isolationism,perceptions of U.S. A., & Kinder, D. Priming deals withthe effect of news media attention to certain stories on individuals'assessments of the President's performance and had, according to theliterature reviewed here, been supported by experimental tests. 499-5 ). This is all the more true because suchfactors also seem to be relevant to their theory but appear, whether or notthis is the case, to have been conveniently disregarded by the researchersfor this particular test. Unless readers payscrupulous attention themselves to the difficulties generated by Krosnick'sand Kinder's research design they will be readily persuaded that anempirical test of their priming theory has, indeed, been successfullyperformed. The authors did not, however, include any survey items that dealtwith Iran, the Middle East, or the Iran-Iraq war. The authors only mention in passing (and not in thesection describing data collection) that the event occurred "roughly midwaythrough the 1986 [NES] study" (p. In both their summaryof the study and in the text of the article the authors seem to emphasizethe intensity and frequency of news media attention to the event orcircumstance.
If this paper is not what you are looking for, you can search again:
or
Click here to request an essay written just for you.
|
|
|
