Abstract: Review helpfulness, which refers to the extent to which an online review is perceived by consumers to facilitates a consumer's purchase decision making, has captured considerable attention both from scholars and practitioners. Prior studies examining the effect of review valence on review helpfulness have produced mixed results. Most of past studies have shown that people tend to weight negative reviews more than positive reviews under certain circumstances. However, the reviews in previous studies were mostly general reviews without the recognition of the review platform. Therefore, this study focuses on the platform-recognized reviews of experienced products.
Based upon the attribution theory, the effects of "negativity bias" on the on perceived helpfulness of platform-recognized reviews are examined. Furthermore, the interaction effect between reviewer level and "negativity bias" on perceived helpfulness of platform-recognized reviews and the intrinsic causal relationship between them are examined. A number of undergraduates are invited to participate in our studies. The reviews we adopted come from the real online review environment such as Dazhongdianping APP as the experimental data.
The conclusions are as follows:First, for the platform-recognized reviews, the perceived usefulness of negative reviews is higher than that of positive reviews, that is, the negative bias effect is still significant. Secondly, for the platform-recognized reviews, reviewer level will adjust the influence of "negativity bias" on perceived usefulness of reviews. The higher the reviewer level, the more helpful the positive review is. However, the difference is not significant for negative reviews. Thirdly, these effects will be mediated by review attribution. That is to say, positive reviews with high-level reviewer (vs. low-level reviewer) can be more likely to be attributed to a product itself rather than reviewers. Consequently, they are perceived to be more helpful. However, for negative reviews, there was no significant difference in the effect of reviewer level on the review attribution. So, no matter the level of the reviewer is high or low, there is no significant difference in perceived usefulness of the platform-recognized negative reviews.
It is assumed this study can contribute to not only the research framework of online reviews and the influence factors of perceived helpfulness, but also expanding and enriching the range of application and interpretation signal theory and of attribution. According to the research conclusions, some targeted countermeasures and suggestions for the enterprise management are provided.

Key words: platform-recognized reviews, reviewer level, negativity bias, perceived helpfulness