Recipient Organization
UNIV OF WISCONSIN
21 N PARK ST STE 6401
MADISON,WI 53715-1218
Performing Department
LIFE SCIENCES COMMUNICATION
Non Technical Summary
Partisans in debates over such controversial issues as GM food appear prone to see information in the mass media - however evenhanded it may be - as hostile to their own point of view. This study will examine the hostile media effect in the context of the GM food debate in an effort to focus more closely on the conditions underlying this perceptual bias.
Animal Health Component
40%
Research Effort Categories
Basic
60%
Applied
40%
Developmental
(N/A)
Goals / Objectives
The debate over genetically modified foods is raging, and promises to continue. Constructive public debate on controversies such as this depends at least in part on the supply of accurate and informative public information. This notion is especially pertinent to science-related conflicts, which often involve greater complexity, greater uncertainty about appropriate solutions, and high levels of citizen involvement. But a problem for mass media, and consequently for public debate, is that partisans (of which there are many in the GM foods arena) appear prone to see information in the mass media - however evenhanded it may be - as hostile to their own point of view. This tendency in highly involved partisans to perceive news coverage as biased against their own position is known as the hostile media effect. This study will examine the hostile media effect in the context of the GM foods debate in an effort to focus more closely on the conditions underlying this perceptual
bias. Goals include (1) determining to what extent this perceptual bias is distinctive to mass media information, (2) exploring underlying mechanisms, and (3) identifying audiences most prone to this effect.
Project Methods
Two field experiments will be conducted in sequence, employing news article treatments compiled from actual media coverage. The hostile media effect hypothesis will be tested using partisan participants and appropriate control groups. The first experimental design, testing the effect of reach, will include a passage of information about GM foods presented in one of four sources: a student class composition, a journalism class exercise, a limited-circulation newsletter or a mainstream media - e.g., Time magazine - news article (text will be identical in each condition). Partisans from both groups, and neutral non-partisans, will be randomly assigned to condition. The treatment stimulus will be a simulated news article (or essay), compiled from actual reports and containing a reasonable balance of information and argument on both sides of the issue. The article will be pretested with neutral judges to reinforce the effort to create content that contains controversial
material but is balanced overall. It will be presented as if it had been published (or written) shortly before administration of the experiment. The article/essay will be composed graphically, with appropriate typefaces, logos, datelines or page headings, to reflect the cosmetic features of these various sources. The second experiment will focus on psychological processes, primarily to distinguish between the selective recall and selective categorization mechanisms. Subject selection will be essentially the same, but manipulations will be confined to two conditions - student essay vs mainstream news article. Initial attitude assessments will also be the same, but after reading the stimulus information, participants will be asked to list five facts or arguments they can recall from the article/essay. After completing that task, participants will be asked to go back and rate their recalled items as favorable, opposed or neutral toward GM foods - what I call a 'subjective' rating. For an
'objective' rating, these free recall items will later be classified by an independent coding team using the same scale but blind to subjects' ratings. Later in the questionnaire, subjects will be provided with ten excerpts - one- or two-sentence passages - from the article/essay text. Again, subjects will be asked to rate the excerpts as favorable, opposed or neutral.