A large body of word-of-mouth (WOM) research has shown that ratings, rating volume, and text comments function as determinants to assessing quality of products (Berger, 2014; Chen & Lurie, 2013). However, little attention has been paid to the reviewers as a significant source of product evaluation (He & Bond, 2013). To be specific, there is a lack of understanding whether a product is preferred because it was recommended by an associative group or because it was not preferred by a dissociative reference group (dissimilar to self). The results from Study 1 demonstrated that participants are more influenced by similarity-attraction than by dissimilarity-repulsion. To be specific, African American participants rated product a higher, which was recommended by their associative reference group. Similar pattern was revealed in the male group. In Study 2, we found an underlying mechanism to explain the effects from Study 1. We tested the perceived difficulty of processing as a potential mediator for this phenomenon and utilized a choice/reject task. The analysis of moderated mediation (Hayes, 2013) revealed that the significant indirect effect of the number of options on purchase intention was conditional on the deleting approach but not on the choosing approach. By rejecting multiple alternatives, participants were more likely to buy the product since it is easier to make decision. In conclusion, this finding provides insights regarding the decisionmaking process resulting from social influence based on a reference-dependence approach.