It is well known that the origin of product has an impact on consumer product evaluation and consumers typically treat products made in developing countries less favorable. Therefore, it is important for firms to develop effective strategies to reduce possible damage caused by negative country-of-origin (COO). Previous studies have found that many attributes of product could influence the strength of COO, especially brand and price. Some researchers pointed out that strong brand and low price could effectively reduce the influences of negative COO. However, consistence about this conclusion has not been reached..According to theory of constructive preferences, the formation of consumer preference is more like architecture, rather than like archaeology. Consumers prefer to build defensible set of values, instead of uncovering values that are already there. There are two important factors which can influence the process of constructing preferences: (1) methods for eliciting preferences; (2) the characteristics of attributes. Based on the theory of constructive preference, this study tried to explore (1) can the relationship between brand and COO and the relationship between price and COO really be affected by the join-separate evaluation mode and the characteristics of attributes? In which evaluation mode, consumers usually weigh brand and price more heavily, which can reduce the effects of COO. (2) Can the amount of evaluable information and the changes of attribute influence the weight assignment? To examine the hypotheses, two studies with five experiments were designed. In study one, a 2 (COO: good and bad) × 2 (product price: high and low) between-subject design was adopted. ANOVA analysis was conducted and the results showed that in joint evaluation mode, consumers weight price more heavily, which reduces the negative influences of a bad COO. Price differential moderates the relationship between price and the effects of COO. In separate evaluation mode, the result was reverse that consumers weight COO more heavily and think the product with a good COO is more favorable. In study two, a 2 (COO: good and bad) × 2 (product brand: strong and weak) between-subjects design was adopted. ANOVA analysis was conducted and the results showed that brand cannot reduce the influences of COO on product evaluation in both evaluation modes. However, increasing the amount of evaluable information of brand or using sorting instead of specific name to present the country of origin of product would make consumers weight brand more heavily in separate mode.