Anchoring Induced Biases in Internet Buyers’ Price Estimates Hsin-Hui Lin Department of Information Management, National Sun Yat-Sen University, Taiwan hhlin@mis.nsysu.edu.tw Chin-Shan Wu Department of Information Management, National Sun Yat-Sen University, Taiwan Department of Information Management, WuFeng Institute of Technology, Taiwan cswu@mis.nsysu.edu.tw Fei-Fei Cheng Department of Information Management, National Sun Yat-Sen University, Taiwan fei@mis.nsysu.edu.tw Abstract One of the heuristics adopted by human beings when they are making estimation under uncertainty is anchoring and adjustment, which refers to the situation in which people anchor on a specific value and then adjust it to yield a final judgment. Because the adjustment is usually insufficient, the estimates can be biased toward the anchor value and result in anchoring effect. Anchoring effect has been proved to be robust in many domains. However, the study on anchoring effect in electronic commerce context has remained largely unexplored. Moreover, we postulate that the repeated occurrence of anchor will moderate the effect of anchor on participants’ price estimates. One experiment was conducted to investigate anchoring effect as well as the moderating role of repeated anchor in electronic commerce context. The results showed that anchoring effect is robust and the repeated anchor did not influence the occurrence of anchoring effect. 1. Introduction With the proliferation of Internet, it has become one of the important transaction platforms. Internet shoppers cannot see and touch the physical products, therefore, the advertisements, product description and many other relevant or irrelevant cues provided on web pages can have potential to influence buyers’ decisions. Anchoring effect is one of the possible results in which people’ s price estimates are biased toward the anchor to be appeared in virtual store. Scholars in marketing area are also interested in exploring the anchoring effect in consumer behavior. For example, Yadav [14] suggested that buyers anchored their evaluation on the item perceived as most important and then made adjustments on the basis of their evaluations of the remaining bundle items. Wansink et al. [11] proposed a simple anchoring and adjustment model describing how consumers make purchase quantity decisions. Further, Kristensen and Garling [3] demonstrated that how the buyers’ price estimates was influenced by anchor in negotiation. However, the evidence for anchoring effect in electronic commerce context is limited. Accordingly, this study aims to investigate the anchoring effect in Internet era, and to understand whether the anchor embedded in web pages influences Internet buyers’ price estimates. Although the anchoring effect has been empirically well established in many domains, there is still considerable debate regarding how anchoring effect occurred. Not all uninformative numbers produce anchoring effects. Instead, certain features of the anchor, target and judgmental task are required. For example, the extremity of anchor value, the amount of attention paid to the anchor, and the relevancy between anchor and target. Experimental study conducted by Wilson et al. [12] is an example that considered the issue of attention in anchoring effect. They suggest that the amount of attention people paid to the anchor value is the key to obtain anchoring effect. However, this issue has remained largely unexplored. Therefore, the investigation of the relationship between attention paid on anchor and the induced anchoring effect in e-commerce context is another goal of current study. 2. Theoretical Background The basic description of anchoring effect refers to the situation in which an arbitrarily chosen reference point (anchor) significantly influence the decision makers’ value estimates, and the anchor that was insufficiently adjusted toward the true value of the object to be estimated yield the final estimates [6]. The insufficient adjustment of the estimate away from the anchor provides the source of decision bias, which was the so called anchoring effect. 2.1 Experimental design of anchoring effect The traditional experimental design of anchoring effect was suggested by Tversky and Kahneman [10] which involves two separate judgment tasks: a comparative judgment followed by an absolute estimate. In the comparative judgment, an anchor was provided explicitly as the standard of comparison. In the next stage, participants were asked to estimate the target value.