(A) Comprehensive Approach to Online Restaurant Reviews : Examining Posters, Postings, and Readers

Li, Lin 경희대학교 대학원 2021 국내박사

The purpose of this dissertation is to provide a profound understanding of components on online restaurant reviews (i.e., posters, postings, and readers) by investigating the relationships between these components and further examining how these components influence review evaluation. Drawing on the theoretical framework of Jacobson’s theory of communication, Study 1 employed a 2 x 2 x 2 scenario-based experiment to examine the roles of posters’ expertise, postings’ quality, and readers’ prior knowledge on review evaluation on online restaurant reviews. Moreover, the double-moderation effects among the features of components on readers’ voting intention is innovatively examined by using the moderated moderation model. Results of Study 1 suggest that the textual quality is significantly related to intention to vote, and expertise and visual presence do not relate to intention to vote significantly. As to the two-way interaction effect, there is a negative effect between textual quality and expertise on intention to vote. The negative two-way interaction also existed marginally between textual quality and prior knowledge. However, the two-way interaction does not show between textual quality and visual presence. By using the moderated moderation model, the three-way interaction effect is examined. The three-way interaction existed marginally between textual quality, expertise, and prior knowledge. There is no three-way interaction between textual quality, expertise, and visual presence. Lastly, there is no three-way interaction between textual quality, visual presence, and prior knowledge. Based on the findings of Study 1, the mere absence or presence of the visual content has been proven to be insignificant for review evaluation, which leaves a research question that why the visual presence is not as important as we initially assumed. Study 2 and Study 3 is thus inspired to examine the semantic aspects of visual contents. Study 2 exploratorily examines features of posters and postings, including the textual-visual contents similarity using both text mining (i.e., topic modeling) and image mining (i.e., Google Vision Application Programming Interface (API)) techniques to extract semantic meanings from textual and visual contents, which has not been conducted in the context of online restaurant reviews. It is found in Study 2 that review readability in the textual content and salient objects in images in the visual content have significant influences on review usefulness and review enjoyment. Moreover, textual-visual similarity is found to be a vital factor in determining review usefulness but not review enjoyment. As to reviewer information, expertise, reputation, and residence location are found to be significantly related to review enjoyment. Furthermore, Study 3 conducted a confirmative study to identify the textual (i.e., textual description elaborateness and textual narrative readability) and visual features (i.e., visual feature complexity and visual semantic complexity) of online restaurant reviews that influence review usefulness. In addition, Study 3 investigates the moderating role of textual-visual similarity in increasing the votes of review usefulness. It is found in Study 3 that for the textual features, readers might consider that the textual narrative readability is more important than textual description elaborateness (i.e., number of words) in terms of review usefulness. However, when the moderator, the textual-visual similarity is taken into consideration, it is interesting that both textual narrative readability and textual description elaborateness are not interacted with similarity. On the other hand, both readability and number of words are not salient factors for review enjoyment. However, when the similarity is considered, textual description elaborateness (i.e., number of words) has an interaction effect with similarity, meaning that readers begin to check relatively more features of textual content, and the review length is reckoned so that whether there is a similarity between textual and visual content could be better measured in a review posting. For the visual features, visual semantic complexity (i.e., the image labels) is significantly related to review usefulness. On the contrary, the relatively heuristic cues of the visual content (i.e., pixel values) have a significant relationship with review enjoyment. When the moderator is considered, it is interesting that the visual feature complexity (i.e., pixel values) start to be measured by readers in terms of review usefulness, the visual semantic complexity (i.e., image labels) start to matter for readers in terms of review enjoyment, because that when they have to consider both the texts and images, not only how colorful and brightness of images matter, but also the actual meaning of image become meaning to readers, so that the image labels are significantly related to review enjoyment. This dissertation contributes to the body of knowledge on the comprehensive examination of components of online restaurant reviews, especially on readers’ prior knowledge, image mining and text mining, and the double moderated effects of posters, postings, and readers on review evaluation. This dissertation also provides practical implications for online restaurant reviews managers and users and restaurant owners in the hospitality and tourism industry.

Exploring the Impact of Heuristic Attributes of Online Reviews on the Popularity of Shared Accommodation

WANG YUWAN 경희대학교 대학원 2021 국내석사

The sharing economy is a new economic field that relies on the rapid growth of the Internet. The sharing economy revitalizes idle resources and realizes efficient allocation. The popularization of the sharing economy has brought about tremendous changes in the way of travel and accommodation. More and more people are willing to try shared travel. Some people say that shared accommodation has brought them a whole new experience. The shared accommodation platform integrates restricted housing resources and provides landlords with a convenient housing management model. It is also convenient for consumers to search for shared accommodation and provide relevant reference information. Due to the geographical distance, consumers cannot fully and objectively feel the real situation of shared accommodation before purchasing like offline shopping. At this time, online reviews, as an important reference indicator for consumers in purchasing decisions, exert an important influence. Online reviews are an important way to reflect the business situation, sales volume and customer response of a product or service. It is also an important data source for analyzing consumer behavior and psychology. There have been much research related to online reviews and consumer trust, sales, purchase intention, and motivation, but few scholars pay attention to the relationship between online reviews and popularity. Especially in the field of shared accommodation, related research has been sorted out. Some scholars have paid attention to the impact of hotel online reviews on popularity, but there is no research on shared accommodation and the popularity of shared accommodation. In order to fill this gap, based on cognitive miser theory, this research first sorted out the relevant research on online reviews of tourism in recent years and defined the heuristic factors of online reviews. Incorporating the characteristics of shared accommodation, the original single-dimensional reviews are expanded into multi-dimensional online reviews related to house, host, and posting. And one by one verified the relationship with the popularity of shared accommodation. And in order to highlight the difference between shared accommodation and traditional hotels, highlight the characteristics of sharing. Added rental type (entire houses and shared houses) and landlord type (individual hosts and merchant hosts) as adjustment variables to explore whether these two adjustment variables are related to house-related attributes and host-related attributes and popularity influential. This research is based on the evaluation on the Chinese Tujia shared accommodation website crawled through data mining. Considering the characteristics of non-zero distribution of popularity, negative binomial regression is used to analyze the data. This study found that in the dimensions related to posting, the number of comments and the ratio of comments with pictures have a significant positive impact on popularity. The rate of negative reviews has a significant negative impact on popularity. Among the dimensions related to house, location and cleanliness have a significant positive impact on popularity. Describe the impact of scores on popularity is not significant. In the dimensions related to host, the response rate and acceptance rate have a significant positive impact on the popularity of accommodation. This shows that consumers will pay attention to relevant reviews on the platform before booking shared accommodation, and different online review information will have different effects on the popularity of shared accommodation. The second step is to add adjustment variables. Consumers are more concerned about the location of the entire houses, and more concerned about the cleanliness and description scores of shared houses. For accommodations operated by merchant houses, consumers are more concerned about the response rate and acceptance rate. This research theoretically defines the multi-dimensional heuristics of online reviews in the context of shared accommodation. And explored its relationship with popularity. The second point is to adopt a new number of likes as a measure of popularity. In practice, the operators of shared accommodation and the managers of shared accommodation platforms need to pay attention to online reviews to create highly popular shared accommodation that truly meets the needs of consumers.

Online Review Studies on Smart Tourism Destinations : Applying a n Heuristic and Systematic Model

이선영 경희대학교 대학원 2017 국내박사

Hospitality and tourism online reviews (HTORs) are the outcomes of the consumer co- creation process within smart tourism destinations (STDs). HTORs have been rapidly adopted for tourists’ decision making to assist travel-related products/services. Due to the competitive and dynamic environment of the tourism industry, on the one hand, tourism consumers prefer to see previous experiences and stories, while on the other hand, tourism companies wish to use appropriate online review methods to attract possible as well as potential consumers. Although some studies have investigated the impact of HTORs (see Liu & Park, 2015; Racherla & Friske, 2012), the literature on the impact of HTORs is fragmented, and little effort has been made to investigate the influence factors of HTORs and integrate the findings of previous studies in the hospitality and tourism area. In addition, some studies have investigated diverse types of online reviews, including the interaction effects of systematic characteristics (i.e., review content including review sentiment and review readability) as well as heuristic characteristics (i.e., review rating and review length), in line with contextual situations on peer review evaluation. This dissertation examines restaurants, attractions and hotels represented by top ranked travel-related websites, namely, TripAdvisor and Yelp, to gain an understanding of various issues related to HTORs. This dissertation aims to develop suitable theories, not only in a comprehensive integrated model, but also with theoretical contributions related to three studies, to explain the current online review context. Further, this dissertation works to adapt these theories and modify them in the context of this research. The study will also empirically examine the suggested research model by applying web harvesting, computational linguistics and statistical techniques to collect and analyze the real-world secondary data. This dissertation seeks to fill a gap between a survey based approach and directly earned, actual data from websites. This dissertation shows conclusively how heuristics and systematic processing are formulated by theinteraction effects among the characteristics of sources, reviews and contextual factors on peer review evaluation in HTORs. In addition, this dissertation applies the heuristic.systematic model (HSM) in HTOR research, which was developed by integrating the impact of interactions among source-related heuristics, review-related heuristics , review-related systematic and contextual factors of HTORs on peer review evaluation. The HSM helps researchers identify the scope of the extant HTORs literature and indicates possible directions for future studies. To achieve this research goal, this dissertation conducted a systematic review of HTOR research. Building upon the literature analysis, the influence factors of HTORs are classified. This dissertation further empirically conducted three studies by using and interpreting the so called big data through major HTOR platforms, namely Yelp and TripAdvisor. Study 1 investigates the interaction effects produced by the heuristic processing of review factors (i.e., review rating, review length and number of images) and source factor (i.e., number of friends and elite badge) on peer review evaluation (“funny” votes as well as “useful” votes) in the online restaurant review (ORR) setting. Study 2 categorizes the interaction effects produced by the systematic processing of review sentiment (i.e., affective, cognitive, positive and negative language) on review peer evaluation in terms of hedonic or utilitarian destinations in the online attraction review (OAR) setting. Study 3 investigates the interaction effects among the reviewer level and hotel classification (three star independent vs. four star chain vs. five star luxury hotels) on the review factor containing heuristic characteristics (i.e., review ratings and review length), systematic characteristics (i.e., affective language and cognitive language) and peer review evaluation in the online hotel review (OHR) setting. The results show that the heuristic-systematic processing of HTORs has some influence on peer review evaluation. Specifically, heuristic processing through source factors includes identity, reputation, expertise, bandwagon and authority heuristics, and heuristic processing through review factors involves attribute, visual and textual heuristics. A systematic lexical cues developed through the review content is also defined to explain the results. Further, the interaction effects of moderator (Study 1: restaurant type, Study 2: attraction/destination type and Study 3: hotel classifications) have influences on peer review evaluation. By using the classical communication-persuasion paradigm in accordance with the heuristic-systematic model for four important dimensions (i.e., source-related heuristics, review-related heuristics, review-related systematics and contextual factors) in the HTOR setting, this dissertation expands the scope of conventional communication theories used in the context of studying the HTOR system. Further, the results of the study presented in this dissertation provide not only insights for hospitality and tourism research but also practical implications for HTOR management.

온라인 쇼핑몰의 추가리뷰가 소비자 반응에 미치는 영향에 관한 연구

란야원 韓國外國語大學校 大學院 2021 국내석사

본 연구에서는 온라인 쇼핑몰 중의 온라인 리뷰는 최초리뷰와 추가리뷰로 구분하여 두 리뷰의 차이와 소비자 심리 및 행동에 미치는 영향을 고찰한다. In online shopping, online consumer reviews have an impact on sales and other consumers' purchase intentions. The online shopping review system is a platform for collecting and sharing online consumer reviews, acting as both "intelligence receiver" and "intelligence sender". In daily online shopping, consumers have been inseparable from the online reviews system, an information platform provided by shopping websites. This study classifies online consumer reviews and investigates the influence of two online reviews systems and two evaluations on consumers' trust and purchase intention respectively. In this study, the online reviews system is divided into the review system with additional reviews and the review - 66 - system without additional reviews. Experiment 1 was designed to investigate the effects of the review system with additional reviews and the review system without additional reviews on consumer trust and purchase intention respectively. The online consumer comments are divided into initial reviews and additional reviews. Experiment 2 is designed to investigate the impact of initial reviews and additional reviews on consumer trust and purchase intention respectively in the review system with additional reviews. The results of experiment 1 show that the trust degree and purchase intention of the review system with additional reviews are significantly higher than those without additional reviews. In other words, the review system with additional reviews can make consumers trust more and increase their purchase intention. The results of Experiment 2 show that in the review system with additional reviews, there is no trust difference between the initial review published with the same ID and the additional review. However, additional reviews has a significant positive impact on purchase intention, which means additional reviews can greatly increase consumers' purchase intention. The results of this study add theoretical significance to online consumer review and online review system, and also provide practical implications for marketing managers.

Exploring the impact of contradictory online reviews with different structures on consumer perceptions from the perspective of attribution theory - a moderated mediation model

황방한 호서대학교 일반대학원 2022 국내석사

In the era when online shopping is gradually becoming the mainstream shopping method, online reviews are a reflection of consumers' subjective consumption feelings after consuming products or experiencing services, so online reviews have become an important standard for consumers to measure the online reputation of products. With the emergence of the additional review function, consumers can get additional review opportunities within a certain time frame after making initial reviews, thus reflecting the post-purchase experience more truly. Since such reviews are a delayed experience after the consumer has used the product or service, they imply that the purchaser has a more comprehensive and in-depth understanding of the product and the service provided by the merchant. In addition, initial and additional reviews by the same reviewer in a contrasting relationship are more likely to reveal similarities and differences in opinions than a single review, and have greater reference value for consumers. Online reviews containing additional reviews can be classified into consistent online reviews and contradictory online reviews based on the same and different directions from the initial review. Contradictory online reviews, as a type of online reviews, are an important channel for consumers to obtain product information in the online shopping environment. There is a large amount of literature and studies that illustrate that contradictory additional reviews have a stronger impact and are more valuable to study than consistent additional reviews. Most of the current research on contradictory online reviews focuses on the comparison of contradictory online reviews with consistent online reviews, but there is a lack of discussion on the different structures of contradictory online reviews in real life. What are the different effects on consumers' perceived risk and perceived usefulness of reviews when initial and additional reviews appear in different orders and produce contradictions? Few scholars have investigated this aspect. Current studies lack research on the structure of Contradictory online reviews and the perceived mental activity of observers, and there is a lack of insight into the case where the observers' own mental changes and the inherent properties of the goods are used as variables. Therefore, based on the theoretical foundations such as the Elaboration Likelihood Model (ELM) and the Information Adoption Model (IAM), this study takes contradictory online reviews of different structures and consumer perceptions as the main research objects, takes attribution to reviews at the consumer psychological level as the entry perspective, and uses product involvement as the moderating variable of this process, so as to explore the effect of contradictory online reviews of different structures on consumer This paper explores the impact of different structures of contradictory online reviews on consumers' perceived risk and perceived usefulness. This paper explores the following three questions: 1. Does the influence of contradictory online reviews on consumers' perceptions differ across different structures; 2. Does attribution to reviews play a mediating role in the influence of contradictory online reviews on consumers' perceptions across different structures; 3. Does product involvement play a moderating role in the different structures of contradictory online reviews influencing consumer perceptions through attribution to reviews?

온라인 리뷰가 소비자의 구매의도에 미치는 영향 : 인터넷 쇼핑몰을 중심으로

최영 경희대학교 대학원 2022 국내석사

인터넷의 발전하기 전까지는 단순한 입소문, 즉 전통적인 구전으로 제품이나 브랜드의 이미지에 대해 영향을 주고 있었다면 인터넷의 발전은 전통적인 구전에서 온라인 구전으로, 즉 소비자들의 제품을 구매한 후 자신의 사용경험과 후기를 작성한 리뷰로 다른 잠재적 소비자들의 구매의도에 영향을 주게 되었다. 또한 이러한 후기는 빠르고, 신속하게, 광범위한 전달력, 오랜기간 저장이 되고 언제든지 소비자들이 참고하여 제품을 평가 할 수 있는 특점을 지니고 있다. 따라서 소비자들은 자신의 원하는 제품을 구매하기 전 다른 소비자들의 제품에 대한 사용후기를 참고하는 것은 당연한 것으로 자리매김하였다. 많은 소비자의 구매의도에 관한 연구가 진행되어 왔지만 온라인 리뷰의 유형과 소비자의 쾌락적 소비와 실용적 소비, 즉 제품을 쾌락재와 실용재로 나뉘었을때의 구매의도에 대한 연구가 필요하다. 본 연구에서는 선행연구의 이론을 바탕으로 온라인 리뷰의 유형이 소비자들의 구매의도에 주는 영향을 확인하고자 한다. 따라서 온라인 리뷰의 유형을 리뷰의 유용성, 리뷰의 신뢰성, 리뷰의 전문성, 리뷰의 수로 구성을 하였다. 또한 소비자들의 쾌락적 소비와 실용적 소비과정으로 제품을 쾌락재와 실용재로 구분을 하여 조절변인으로 사용을 하였다. 이와 같이 총8개의 가설을 수립하였다. 가설의 검증을 위해서는 다양한 인터넷 쇼핑업체가 존재하지만, 소비자의 사용량이 비교적 많고 대표적인 업체 중 하나라고 생각되는 쿠팡을 선택하였고 쿠팡을 사용한 경험이 있는 소비자들을 대상으로 인터넷 설문조사를 실시 하였다. 자료에 의하면 2020년 기준으로 2000만명 이상의 한국인들이 쿠팡을 사용하고 있는것으로 나타났다. 연구의 결과로는 온라인에서의 리뷰의 유용성, 리뷰의 신뢰성, 리뷰의 전문성, 리뷰의 수는 소비자들의 구매의도에 유의미한 영향을 미치는것으로 나타났다. 또한 리뷰의 유형이 소비자의 구매의도에 대한 관계에서 제품의 성격이 조절효과를 살펴보면 리뷰의 유용성과 리뷰의 전문성이 소비자의 구매의도와의 관계에서 조절효과를 가지는것으로 나타났다. 따라서 본 연구에서는 온라인 리뷰의 유형이 잠재적 소비자들의 제품에 대한 구매 의사결정과의 관계를 검증하고 기업이 판매하고 있는 제품에 대해 새로운, 효과적인 마케팅 전략을 구현하는데 기초자료로서 도움을 줄 수 있는데 의의가 있다. Until the development of the Internet, simple word of mouth, that is, traditional word of mouth, influenced the image of a product or brand. The reviews you write have an impact on the purchase intentions of other potential consumers. In addition, these reviews have the characteristics of being able to evaluate products with reference to consumers at any time, as they are quickly, extensively delivered, stored for a long time. Therefore, it has become natural for consumers to refer to the reviews of other consumers' products before purchasing their desired product. Although many studies have been conducted on the purchase intention of consumers, as a type of online review, it is necessary to study consumers' hedonic and practical consumption. i.e, purchase intention when products are divided into pleasure goods and utility goods. In this study, based on the theories of previous studies, the purpose of this study is to check the effect of online review types on consumers' purchase intentions. Therefore, the types of online reviews were composed of review usefulness, review reliability, review expertise, and number of reviews. In addition, according to consumers' hedonistic consumption and practical consumption process, products were classified into pleasure goods and practical goods, and used as control variables. In this way, a total of eight hypotheses were established. To test the hypothesis, there are various internet shopping companies, but Coupang, which is considered to be one of the representative companies with relatively high usage by consumers, was selected. According to the data, as of 2020, more than 20 million Koreans are using Coupang. As a result of the study, it was found that the usefulness of online reviews, the reliability of the reviews, the professionalism of the reviews, and the number of reviews had a significant effect on the purchase intention of consumers. Also, looking at the moderating effect of product characteristics on the relationship between review type and consumer's purchase intention, it was found that the usefulness of reviews and the professionalism of reviews had a moderating effect on the relationship between the usefulness of reviews and the professionalism of reviews. Therefore, in this study, it is meaningful that the type of online review can help as basic data to verify the relationship between potential consumers' product purchase decision-making and to implement a new and effective marketing strategy for the product the company is selling.

(An) exploratory study on online review filtering algorithm : based on Yelp's reviews

Liu, Yansha Graduate School, Yonsei University 2024 국내석사

온라인 리뷰가 일상생활에서 중요성이 증가함에 따라, 온라인 리뷰를 조종하려고 하고 있다, 이에 따라 리뷰의 품질과 신뢰성에 대한 우려가 커지고 있다. 이에 대응하여 많은 온라인 리뷰 플랫폼들은 가짜 리뷰로 의심되는 리뷰를 필터링하기 위해 알고리즘을 사용하고 있다. 이 연구는 리뷰 필터링 알고리즘의 기준을 이해하고 가짜 리뷰와 일반 리뷰의 차이점을 탐구하고자 한다. 이 연구에서는 데이터 스크래핑을 통해 유명한 리뷰 웹사이트인 Yelp의 미국 50개 도시 중 750개 레스토랑의 리뷰 데이터를 수집하였다. 수집한 데이터는 리뷰뿐만 아니라 리뷰를 게시한 레스토랑 및 리뷰를 작성한 사용자에 대한 정보도 포함되어 있다. 분석 결과, 리뷰 수가 더 많고 평균 평점이 낮은 레스토랑은 가짜 리뷰를 가질 가능성이 더 높다는 것을 발견하였다. 텍스트적 특징 측면에서 가짜 리뷰는 일반 리뷰에 비해 언어적 풍부성이 낮고 단어 사용 감정에서 상당한 차이를 보인다. 또한 가짜 리뷰의 평점은 더 극단적인 경향이 있다. 덜 활동적인 사용자가 작성한 리뷰는 가짜로 분류될 가능성이 더 높다는 것도 관찰되었다. 특히 본 연구에 포함된 요인 내에서 리뷰가 가장 영향력 있는 결정 요인은 텍스트 풍부성을 의미하는 'word_count' 변수이며, 사용자가 작성한 리뷰 수를 의미하는 'user_review_count' 변수가 그 뒤를 잇는다는 것을 알게 되었다. 마지막으로, 본 논문은 리뷰 필터링에 영향을 미치는 요인들을 사용하여 머신러닝 모델을 개발했다. 실제 리뷰 필터링 시스템을 시뮬레이션하는 것을 목표로 하여 개발된 이 모델은 정확도 93%, F1-score 0.89의 결과를 달성하여 기존 연구보다 우수한 성능을 보였다. 이 연구는 가짜 리뷰의 특성을 더 잘 이해하고 온라인 리뷰 웹사이트의 필터링 알고리즘 기준을 밝히는 데 도움이 될 것이다. As online reviews play an increasingly important role in daily life, businesses have begun to try to manipulate the online review environment, leading to concerns about the quality and credibility of reviews. In response, many online review platforms have begun to use algorithms to filter reviews that are suspected to be fake reviews. To understand the mechanisms of review filtering algorithms and to explore the distinctions between fake and genuine reviews, this study analyzes restaurant review data from the well-known review website Yelp. In contrast to most studies that utilize publicly available data containing only the reviews themselves, this research, through data scraping, has gathered review data up to September 2023 from 750 restaurants across 50 cities in the United States. The data encompasses not only the reviews but also information about the restaurants and the users who posted the reviews. Our analysis reveals several key findings. At the restaurant level, we found that restaurants with a larger number of reviews and lower ratings are more likely to have fake reviews. Regarding the textual features of the reviews, fake reviews exhibit poorer linguistic richness and significant differences in word usage, tense, and sentiment compared to genuine reviews. Additionally, the ratings in fake reviews tend to be more extreme. It was also observed that reviews posted by less active users are more likely to be classified as fake. Notably, within the factors included in this study, we discovered that the most influential determinant of a review being fake is the 'word_count' variable, which represents textual richness, followed by the 'user_review_count' variable, indicating the number of reviews posted by a user. Finally, this paper constructs machine learning models using the validated key factors that significantly impact review filtering. These model aims to simulate actual review filtering systems. The results demonstrate that our proposed model outperforms some existing studies with more limited variable dimensions. It achieves an accuracy of 93% in determining whether a review should be considered fake, with an F1-score of 0.89. This research helps to further understand the characteristics of fake reviews and partially reveals the standards of online review websites' filtering algorithms, demystifying them to an extent.

Language Abstraction in Negative Online Consumer Reviews : the Choice of Firm Response Strategies and Voice

FENG WEI 가천대학교 일반대학원 2022 국내박사

Negative online consumer reviews can observably affect consumer attitudes and behaviors, and companies and businesses attach great importance to managing and responding to negative reviews. In the context of hotel services and functional products, this study examines the impact of language abstraction in negative online reviews on potential customers’ attitude and purchase intention, as well as the moderating effects of managerial responses, response strategies, and response tone on the persuasiveness of negative reviews. Three experiments were conducted in this study. The experimental results show that, compared with abstract negative reviews, concrete negative reviews can significantly enhance potential customers’ attribution of responsibility towards service or product providers, and their perception of stability and controllability of negative events, which leads to a more negative view of the company, although not lower purchase intention. It’s also found that corporate responses can significantly reduce potential customers’ attribution of responsibility towards the disputing company and improve their evaluation of it. Accommodative responses to concrete negative reviews and defensive responses to abstract negative reviews are shown to be successful in inducing potential customers to make judgments favorable to the firm. Furthermore, the results also demonstrate the importance of managerial response voice. Regardless of the level of language abstraction in negative reviews, a personal voice can dramatically diminish criticism and negative evaluations in comparison to a corporate one. Implications of these findings for sellers and marketing managers are discussed to assist them in implementing more effective negative word-of-mouth intervention.

화장품의 온라인 리뷰 정보 특성이 소비자 행동에 미치는 영향

신지유 영산대학교 미용예술대학원 2022 국내석사

With the development and widespread Internet, many areas of consumer life are shifting from offline to online. As information search and product purchase activities using online shopping malls spread significantly, consumers collect information on the desired product before purchasing cosmetics and decide their purchase intention based on it, and online reviews of cosmetics are the industry. The necessity is being emphasized as a marketing strategy. This study conducted a survey focusing on men and women citizens in their teens to 50s to find out the effect of online distribution cosmetics reviews on purchasing behavior, and recently studied the effect of consumers' online purchases and online review characteristics. As the influence of the online market is growing more than offline in the era of Untact, I would like to study the characteristics of reviews and consumers' direct influence in the online cosmetics market and present marketing strategies accordingly. The summary of the results is as follows. First, the difference according to the characteristics of the study subjects is that the purchase ranking information and the photo and video information are in their 20s or younger than those in their 50s.Those in their 30s showed a higher purchase ranking, but there was no significant difference in content information. There was no significant difference in reliability in demographic differences, and those in their 20s or younger had higher utilization than those in their 50s. It was found that consumers in their 50s had the lowest intention to purchase online, and consumers with less than 2 to 3 million won had higher intention to purchase compared to those with more than 3 to 5 million won and more than 5 million won. Second, among the differences in usage status according to the characteristics of the study subjects, the average daily Internet usage was the most for women over 5 hours, but for men, it was the most for 1 to 2 hours. In the past month, women also used the most frequently to purchase pictorials more than five times, but men had the most frequent one to three times. Both had the largest number of purchases of less than one to three times, but there was a difference in the percentage of women responding less than three to five times compared to men. Women used Internet shopping malls the most, with 100,000 won to 150,000 won for women, but less than 30,000 won for men. Among them, women had the highest purchase amount of cosmetics of 50,000 won or more, and men had the highest purchase amount of less than 30,000 won. Both men and women use smartphones more as purchasing tools, but it was men who used PCs and smartphones more. Internet usage differed according to age. The most common answer was that those in their 20s or younger use more than 5 hours, and those in their 30s use 2 to 3 hours and 5 hours, those in their 40s use less than 1 to 2 hours, and those in their 50s use less than 1 hour, showing interesting results. There was a difference in the amount of purchase depending on the income. Those with less than 2 million won were the highest than those with more than 2 to 500 won. There was no significant difference in the use status depending on the educational background. For jobs, there was a difference between the amount used and the purchase tool. As for the amount of use, office workers, full-time housewives, professionals, and public officials in the occupational group were the highest at 100,000 won to 150,000 won, while self-employed and other (unemployed, job seekers, etc.) and students were less than 30,000 won. Third, looking at the relationship between online review information and consumer behavior, reliability, utilization, and purchase intention all had a positive correlation with purchase ranking information, review content information, and photo video information, which are sub-factors of online review information. Fourth, looking at the effect of online review information on reliability in consumer behavior, it can be seen that purchase ranking information, review content information, and photo video information all affect reliability. That is, it may be seen that the reliability also increases as the degree of purchase ranking information, review content information, and photo video information increases. In addition, the review content information has a greater influence on reliability than the purchase ranking information and the photo video information. Looking at the effects of purchase ranking information, review content information, and photo video information on utilization, it can be seen that purchase ranking information, review content information, and photo video information all affect utilization. That is, it may be seen that the use also increases as the degree of purchase ranking information, review content information, and photo video information increases. In addition, through the standardization coefficient, the review content information has a greater influence on the usability than the photo video information and purchase ranking information. Looking at the effect of purchase ranking information, review content information, and photo video information on usability, it can be seen that purchase ranking information and photo video information affect purchase intention. That is, it may be seen that as the degree of purchase ranking information and photo-video information increases, the purchase intention also increases. In addition, the photo and video information has more influence on the purchase intention than the purchase order information. Based on the above research results, it can be seen that reviews of online distribution cosmetics affect consumer behavior. In order to meet the consumer needs of these online cosmetics, the cosmetics industry and online platform operators will have to understand the constant development and consumer needs. Therefore, we hope that this data will help come up with specific measures for marketing strategies of cosmetics brands and online distribution companies.

소셜 미디어 상호작용의 동력: 현대 기술 환경에서의 커뮤니케이션 구조 콘텐츠 특성 및 소비자 저항 분석

왕밍 순천향대학교 일반대학원 2023 국내박사