Portal de Periódicos da CAPES

Framing effects in decisions with completely and incompletely described alternatives

1986; Elsevier BV; Volume: 38; Issue: 1 Linguagem: Inglês

10.1016/0749-5978(86)90025-7

1095-9920

Irwin P. Levin, Richard D. Johnson, Patricia J. Deldin, Laura M Carstens, LuAnne J Cressey, Charles R. Davis,

Forecasting Techniques and Applications

Earlier research has shown that a variety of judgments depend upon how key sources of information are framed. Framing effects have been extended to include relative evaluations of stimuli with complete and incomplete information. The present study was designed to see if these effects are predictive of discrete choices and to isolate the locus of the framing effect. In Experiment 1 subjects were asked to indicate whether they would take each of a series of gambles described by payoff and probability information or by only one of these values. As predicted, subjects in the positive condition were more apt than subjects in the negative condition to take gambles when probability information was included, but not when it was missing. Subjects in Experiment 2 chose between gambles with complete and incomplete information. Consistent with the earlier findings, subjects in the positive condition were less apt than subjects in the negative condition to choose the gamble with missing probability information. Experiment 3 replicated the results of Experiment 2 while eliminating various psychophysical factors as causes of the framing effect. The way stimuli are framed appears to affect their subjective scale values and this determines both how they are responded to in an absolute sense and how they are compared to other stimuli.