Talking frames : the assessment, use, and impact of message frames in conversational perspective

Choice behavior is susceptible to slight changes in the representation and formulation of messages (e.g., Tversky & Kahneman, 1981), thus violating the normative requirement of "description invariance". Even logically equivalent message frames (e.g., "80% success rate of a new product launch" vs. "20% failure rate of a new product launch") may invoke substantially different appraisals. In decision making research the social context of message framing has been underappreciated. This dissertation addresses the social context by considering framing effects in a conversational framework (e.g., Schwarz, 1996). In particular, previous research focused on the listener’s perspective. In contrast, this research examined speakers’ assessment and selection of message frames and its corresponding persuasiveness was evaluated using listeners’ responses. In chapter 2, experimental results on risky choice framing showed that speakers exhibit a consistent preference for positive over negative framing. Judged from listeners’ responses, this preference is effective for promoting riskless, but not risky options. The incompatibility between speakers and listeners may be resolved by noting that speakers can usually assess the information and the persuasive qualities of alternative frames jointly (i.e., comparatively). In contrast, listeners are usually exposed only to one of these frames and, consequently, can only assess the information separately (i.e., non–comparatively). Further experiments revealed that no incompatibility exists when both speakers and listeners are either in separate, or in joint evaluation mode. This line of research elaborates and applies Hsee’s (1996) joint– separate distinction to the domain of framing. In chapter 3, subsequent research on speakers’ assessments of message frames suggested that (i) speakers have a default preference for positive over negative frames when jointly considering frames, and (ii) compatibility effects operate to enhance or diminish this positivity effect thereby affecting speakers’ success rate in persuasive communication. For better or worse, the positive formulation of ideas, facts, and opinions serves as a social lubricant in human communication. Positive frames simply make "better stories" (Dawes, 1999). Notwithstanding, negative framing may be employed whenever there is compatibility between frames and elements of the choice situation (e.g., Shafir, 1995). In particular, a distinction was proposed between recommending for preferred choice options (i.e., encourage) and recommending against non–preferred choice options (i.e., discourage). In binary choice cases, both recommendation modes are logically, though not psychologically, equivalent. Most important, empirical results showed that speakers that recommend for preferred options predominantly select positive frames. Positive frames were used less when speakers recommend against non–preferred options. In chapter 4, findings from Meyerowitz and Chaiken (1987) were re–examined and re– interpreted from a conversational viewpoint. Meyerowitz and Chaiken found that breast self– examination (BSE) is more effectively promoted by negatively framed messages that highlight dangers of neglecting to performBSE than by logically equivalent positively framed messages that emphasizes benefits of performing BSE. Experiments in this chapter (i) corroborated our previous finding that speakers prefer positive frames, but surprisingly, (ii) yielded results opposite fromMeyerowitz and Chaiken, i.e., listeners were influenced most by positively framed messages about BSE. Finally, in chapter 5, a framework for framing was developed which stresses the importance of reference information contained in message frames. The framework complements previous classifications (e.g., Druckman, 2001b; Levin et al., 1998; Soman, 2004), and explicates the importance of a central theme of this dissertation, namely the joint–separate destinction.