The Effects Of Question Backgrounds And Self-esteem On Framing Effects In Risky Choice

The “Framing Effect” is observed when decision makers respond differently todifferent but objectively equivalent descriptions of some same problems. Kahnemanbelieved that men not only accepted the frame provided by external task, but alsoformed internal frame according to their own subjective reference points, which canbe named as the self-frame. Wang developed pie charts to explore the self-frame andfound that self-frame had the same effect as the external frame. In other previousresearches, we found that characteristics of decision makers and backgrounds of taskshave significant influence on framing-effects. So in this study, we hope to make anexperiment to explore the relationship among question backgrounds, self-esteem andframes in risky choice context. In study1, we set a stock investment problem indifferent situations by using pie charts and a complementary sentence-completion taskto test the self-frame they imposed by themselves. And in study2, participants wouldmake a choice under the offered positive or negative frame between sure option and agamble by using the same stock investment problem.The whole study totally investigated312college students from ZhejiangProvince. They were asked to complete a self-frame task, IAT, SES and externalframe task orderly. After collecting all the phrases in self-frame task, we invited30participants in previous experiment to help make a rate of each phrase on seven-pointscale in order to get the hedonic tone and divide the positive self-frame and negativeself-frame.Results:(1) The hedonic tone of self-frames had a significant difference in two situations(tall=2.82，p<0.01). The hedonic tone was more positive in good situation than in badone.(2) No matter in self-frame task nor in external frame task, framing effects was obviously observed in bad situation (χself2=5.79，p<0.05；χexternal2=7.47，p<0.01). Therewas no significant difference in risk preference in good situation.(3) Self-esteem (SE) had an effect on editing self-frame. In good situation, highimplicit high explicit SE would describe the options by using more positive hedonictone and form a more positive self-frame. Whereas low implicit low explicit SEwould be more positive and has a high ratio of positive self-frame in bad situation.(4) In self-frame task, low implicit low explicit SE became risk averse andpreferred to choose a sure option no matter in what situation. There was no riskpreference in other kind of SE. But this outcome was limited in self-frame task, nodifference was found in all kinds of SE under external frame.(5) No matter in self-frame task nor in external frame task, a significantdifference in risk preference of high implicit low explicit SE was found according to atwo by two chi-square analysis (χself2=4.21，p<0.05；χexternal2=4.15，p<0.05). Same ashigh implicit high explicit SE (χself2=7.04， p<0.01； χexternal2=7.00， p<0.01). Achi-square analysis also showed the difference was not significant to low implicit lowexplicit SE and low implicit high explicit SE.Conclusion:(1) Self-framing effect and external framing-effect were only observed in badsituation.(2) Self-framing effect and external framing-effect were only found among highimplicit SE.