As online reviews become a major factor in the consumer decision making process, firms have started seeking ways to help consumers make informed decisions. To avoid potential information overload, most online retailers provide review helpfulness features by allowing consumers to cast a vote on the (un)helpfulness of a review. Previous research has extensively focused on identifying the drivers of review helpfulness while little attention has been paid to review unhelpfulness. Using an empirical analysis of a large review dataset from a major online retailer, this research investigates the drivers of both review helpfulness and unhelpfulness. By revealing the asymmetry between the consumers’ review helpfulness and unhelpfulness voting behaviors, this study provides practical implications for review platforms.

This research aims to examine the interplay of fluency and framing in consumers’ perception of online review helpfulness. We propose to conduct three experiments to study how consumers evaluate a product review as presented in easy- vs. hard-to-read font and promotion vs. prevention frame. INTRODUCTION Nowadays consumers are often exposed to and influenced by online product reviews. These reviews are generated and posted by consumers on online shopping websites, such as Amazon.com. Obviously, the content of an online product review, either appraisal or criticism, would offer valuable information for consumers to make purchase decisions. The almost immediate and strong impact of product reviews on a prospective consumer’s decision would not be easily achieved by other marketing attempts (Daugherty, Eastin, & Bright, 2013; Ghose & Ipeirotis, 2006; Mahajan, Muller, & Kerin, 1984). In particular, past research has proposed that, if a review is believed to be more helpful it is more likely to add value for a future customer (Mudambi & Schuff, 2010; Yin, Bond, & Zhang, 2014). In this research, we would like to research what makes a review helpful. According to consumer psychology literature, metacognitive cues, such as fluency, would play an important role in consumer judgments (Lee, 2004; Schwarz, 2004). So, we first propose that fluency, the ease of information processing, could influence perceived review helpfulness. Moreover, a growing body of research has suggested some factors that could interact with fluency when influencing consumer decisions and judgments (Shah, Alter, & Oppenheimer, 2010; Tsai & McGill, 2010). For example, framing effect has been extensively studied in consumer decision making literature (Block & Keller, 1995; Rothman, Salovey, Antone, Keough, & Martin, 1993). Accordingly, we would like to study the effect of framing, emphasizing on gain vs. loss, on perceived review helpfulness. Specifically, we research whether framing a review as promotion- vs. prevention-focused would moderate fluency effect on consumers’ perception of online product reviews. To sum up, in this research we investigate how consumers’ evaluation on online product reviews is influenced by manipulating the review to be promotion- vs. preventionoriented and easy- vs. hard-to-read. To our best understanding, little of past research has directly examined the interplay of fluency and framing in consumer judgments, especially in social media contexts. We wish to collect further evidence regarding the underlying rationale involving fluency and framing effects on consumers’ perception of review helpfulness. THEORETICAL DEVELOPMENT AND HYPOTHESES Researchers have defined online reviews as peer-generated product evaluations posted on a company or third party website (Mudambi & Schuff, 2010). To better understand how consumers evaluate online reviews, we propose to examine the effects of fluency and framing on perceived review helpfulness. First, fluency, defined as the ease of information processing, could make a strong impact on various judgments, including perceived risk, liking, and confidence (Alter & Oppenheimer, 2009; Novemsky, Dhar, & Schwarz, 2007; Reber, Winkielman, & Schwarz, 1998). We notice that limited empirical work has yet addressed the relationship between fluency and perceived review helpfulness, although fluency is easy to manipulate and could have overwhelming effect on judgments. Rather, marketing and social media researchers have long been focused on content- and emotion-based features of online reviews in relation to review helpfulness (Moore, 2015; Mudambi & Schuff, 2010). Therefore, this research aims to address this limitation by exploring fluency effect on review helpfulness. Second, we suggest that framing, by focusing on promotion vs. prevention, could influence consumers’ perception of online product reviews. In particular, regulatory focus literature has revealed that, message framing would influence the effectiveness of persuasion (Higgins, 1998; Meyerowitz & Chaiken, 1987; Tykocinski, Higgins, & Chaiken, 1994). To emphasize, past research has gained mixed results about promotionvs. prevention-focused messages in persuasion (Block & Keller, 1995; Rothman et al., 1993). In this research we will have a close look at framing and explore its effect on online review helpfulness. Third, we would like to test whether a gain (vs. loss) frame is more effective when disfluency (vs. fluency) is experienced. In a past study, Shah, Alter, and Oppenheimer (2010) argue that fluency (vs. disfluency) makes people feel they are close to (vs. distant from) a stimulus, and then become more likely to be engaged in low (vs. high) level thinking. According to construal level theory, when people think concretely they are more likely to focus on concrete, specific details; by contrast, when they think abstractly they are more likely to focus on abstract, global properties (Trope & Liberman, 2003). In addition, past research has found that mind-set, or thinking concretely vs. abstractly, could interact with framing in consumer judgments (White, MacDonnell, & Dhal, 2011). Regarding the interaction between fluency and framing, mixed results are found in literature. On the one hand, Shah and colleagues (2010) have suggested that, disfluency is related to high-level thinking, and thus would make a promotion frame more effective. On the other hand, researchers have proposed that, disfluency induces negative feelings, and thus would make a prevention frame more effective (White, MacDonnell, & Dhal, 2011). Given the disagreement in past findings, we would like to further examine the underlying mechanisms for consumers to perceive online product reviews presented in easy vs. hard font and promotion vs. prevention frame. Following these studies, we will conduct three experiments to test fluency and framing effects on perceived review helpfulness. Our hypotheses are listed below: H1: Consumers perceive an online product review in easy-to-read font (vs. hard-to-read) as more helpful. H2-a: Consumers perceive an online product review in easy-to-read font and a promotion (vs. prevention) frame as more helpful. H2-b: Consumers perceive an online product review in easy-to-read font and a prevention (vs. promotion) frame as more helpful. H3-a: Consumers perceive an online product review in hard-to-read font and a prevention (vs. promotion) frame as more helpful. H3-b: Consumers perceive an online product review in hard-to-read font and a promotion (vs. prevention) frame as more helpful. RESEARCH METHOD We propose three experiments to examine fluency and framing effects on perceived review helpfulness, by using an actual product review collected from Amazon.com. Two hundred undergraduate students from a public university in mainland China will participate in our experiments. In Experiment 1, we will investigate the main effect of fluency on review helpfulness. It is expected that, fluency would have significant effect on perceived review helpfulness. A review on a home speaker – Amazon Echo Dot is selected, and then rated by three individual researchers as relatively neutral and objective. We manipulate fluency by displaying the review in an easy- vs. hard-to-read font, respectively. After reading the review, using 7-point scales participants rate fluency (1 = very difficult, 7 = very easy) and review helpfulness (1 = not helpful at all, 7 = very helpful). To test fluency effect as predicted in H1, we will conduct a t-test. Also, using t-test we will check if the manipulation is successful. In Experiments 2 and 3, we will examine the main effect of framing on review helpfulness, as well as the interaction between fluency and framing. As discussed earlier, past research suggests two opposing expectations. As stated in H2-a and H3-a, promotion (vs. prevention) frame is more effective when a review is shown in easy (vs. hard) font. However, H2-b and H3-b are holding the contradictory predictions. Here, the review is modified to be more focused on promotion vs. prevention respectively, shown in either easy- or hard-to-read font. In each experiment, half of the participants read the Amazon Echo Dot’s review in a promotion frame and the other half will process the review in a prevention frame. In both promotion and prevention conditions, participants are randomly assigned to read the review in an easy- or hard-to-read font. To test our hypotheses, we will employ a 2 (easy vs. hard) x 2 (promotion vs. prevention) ANOVA. In addition, to understand the rationale of consumers’ making evaluation on reviews, we ask participants to use information contained in the review to write a short description of the product. So that, we will be able to capture participants’ thinking mode. Also, we ask participants to rate review valence (1 = very negative, 7 = very positive) and describe their feelings after reading the review, in order to examine if framing would result in different feelings. Then, we analyze these self-reported responses, in attempts to determine either participants’ mind-set (concrete vs. abstract) or their feelings (positive vs. negative) would be more prevalent when they process and evaluate the review. A mediational analysis will be conducted to explore the mechanisms underlying consumers’ perception of review helpfulness. DISCUSSION The aim of this research is to test whether fluency and framing would influence perceived online review helpfulness. Also, our research is designed to explore the interplay of fluency and framing in review perception. The completed work will present empirical results to demonstrate the roles of fluency and framing in consumers’ review evaluations, through the three experiments as proposed. Theoretically, our research will contribute to existing literature by addressing the effects of fluency and framing on consumer decision making, in the context of online marketing communications. Practically, our research will provide insights for marketers and consumers to engage in communications via online review systems. Given that fluency and framing are considered as salient cues to influence consumer decisions, marketers should learn how to leverage these factors while shaping the framework in which online product information is created and shared by consumers.

The purpose of this study was to measure respondent's demographic characteristics, respondent's attitudes toward chicken, and factor influencing on the level of perceived helpfulness of country of origin in predicting the quality of chicken. The data was collected through a consumer survey during the March 2006. A total number of 250 meat consumers living in Suncheon, the eastern part of Chonnam, were randomly selected as respondents. Eleven respondents did not complete the survey instrument, resulting in a final sample size of 239. All estimations were carried out using chi-square, correlation, and logistic procedure of SAS package. The results are as follows. The level of perceived helpfulness of country of origin in predicting the quality of chicken was significantly different by age and occupation of demographic variables, and was significantly correlated with respondent informed of attitude variables. The proportional odds assumption of model was not violated at p<0.05. The effects of income, occupation and respondent informed on the level of perceived helpfulness of country of origin in predicting the quality of chicken. The results from this study could be useful in developing marketing and health promotion strategies, as well as government trade policy.

인터넷 환경의 발달로 소비자들 사이에 상품정보에 대한 의견이 교환되기 시작하면서 다양한 형식의 온라인 리뷰들이 급속도로 생성되고 있다. 이러한 추세에 따라, 기업들은 온라인 리뷰들 을 분석하여 마케팅, 세일즈, 제품개발 등의 다양한 기업 활동에서 그 결과를 활용하려는 노력 을 진행하고 있다. 그러나 대표적인 경험재인 ‘게임’과 관련된 산업에서의 온라인 리뷰에 대한 연구는 매우 부족한 실정이다. 이에 본 연구는 머신러닝 모델을 활용하여 스팀(STEAM)게임의 커뮤니티 데이터를 분석하였다. 이를 통해 타 사용자의 게임 리뷰를 유용하다고 판단하는데 영 향을 미치는 요인을 분석하고, 리뷰의 유용성을 예측하는데 있어 가장 우수한 성능을 보인 모 델과 변수들을 도출하여 사용자의 충성도와 사용성을 증대시키기 위한 제안을 하고자 한다.