Analytics in smart tourism design concepts and methods part 2
➤ Gửi thông báo lỗi ⚠️ Báo cáo tài liệu vi phạmNội dung chi tiết: Analytics in smart tourism design concepts and methods part 2
Analytics in smart tourism design concepts and methods part 2
Estimating the Effect of Online Consumer Reviews: An Application of Count Data ModelsSangwon Park1IntroductionThe advent of information technology has Analytics in smart tourism design concepts and methods part 2s resulted in the development of a new form of web communication, known as eWOM (electronic word-of-mouth), operated by consumer participation (Tussyadiah & Fcscnmaicr, 2009). Online consumer reviews have become one of the vital information sources which allow people to gather sufficient and reliabl Analytics in smart tourism design concepts and methods part 2e information about products and services (Liu & Park, 2015). In particular, due to the characteristics of tourism products (e.g. intangibility and peAnalytics in smart tourism design concepts and methods part 2
rishability), online reviews provide substantial benefits to current travellers, enabling them to obtain authentic and indirect consumption experienceEstimating the Effect of Online Consumer Reviews: An Application of Count Data ModelsSangwon Park1IntroductionThe advent of information technology has Analytics in smart tourism design concepts and methods part 2tality, a number of researchers have investigated the effects of consumer review's, essentially in terms of product sales (Ye. Law, Gu. & Chen, 2011) and lhe decision-making process (Sparks, Perkins, & Buckley, 2013). These studies conclude that online reviews have positive influences on increasing Analytics in smart tourism design concepts and methods part 2revenues and assisting with purchase decisions.Importantly, easily accessible online reviews facilitate consumers in finding plentiful information (loAnalytics in smart tourism design concepts and methods part 2
w search costs); however, they also make it difficult for people to determine helpful information (high evaluation costs). Overall, the important quesEstimating the Effect of Online Consumer Reviews: An Application of Count Data ModelsSangwon Park1IntroductionThe advent of information technology has Analytics in smart tourism design concepts and methods part 2& Johnson, 1992), consumers are likely to focus on heuristic information cues when the size of information to be evaluated is larger than their cognitive abilities. With regard to the context of online consumer reviews, it has beens. Park (X)University of Surrey. Guildford. UK e-mail: sangwon.park@s Analytics in smart tourism design concepts and methods part 2urrey.ac.uk© Springer International Publishing Switzerland 2017147z. Xiang, D.R. Fesenmaier (eds.). Analytics in Smart Tourism Design, Tourism on tknAnalytics in smart tourism design concepts and methods part 2
Vorno not IO im7/O7«.7.7I o148s. Parkidentified that star rating is a key clement of heuristic information, which is regarded as an explanatory variabEstimating the Effect of Online Consumer Reviews: An Application of Count Data ModelsSangwon Park1IntroductionThe advent of information technology has Analytics in smart tourism design concepts and methods part 2e reviews. In order to address the research question, over 5000 reviews were collected from Yelp (yelp.com). a well-recognised consumer review website for tourism and hospitality products. This study then employed negative binomial regression, a type of count model (Allison & Waterman, 2(X)2). Analy Analytics in smart tourism design concepts and methods part 2sing secondary data obtained with an unstructured formal commonly violates the assumptions of the ordinary least square (OLS) regression, or general cAnalytics in smart tourism design concepts and methods part 2
ount models such as the Poisson regression (Hox & Boeije, 2005). For instance, there can be skewed distribution of the data, zero inflation problems, Estimating the Effect of Online Consumer Reviews: An Application of Count Data ModelsSangwon Park1IntroductionThe advent of information technology has Analytics in smart tourism design concepts and methods part 2 aim of this chapter is to discuss count models and. in particular, provide evidence of the usability of negative binomial models in analysing the online review data.2Online Consumer Reviews in Tourism and HospitalityOnline travellers like to obtain detailed and up-to-date information and examine in Analytics in smart tourism design concepts and methods part 2direct experiences of tourism products in order to make a belter decision on them (Xiang. Wang, O’Leary, & Fesenmaier. 2015). In this sense, online reAnalytics in smart tourism design concepts and methods part 2
views developed by other consumers have relatively higher reliability and bring about more attention from other consumers. Based on the important roleEstimating the Effect of Online Consumer Reviews: An Application of Count Data ModelsSangwon Park1IntroductionThe advent of information technology has Analytics in smart tourism design concepts and methods part 2o the three areas of product sales, the decision-making process and evaluation of the information sources (Park & Nicolau. 2015).Following a statement that the number of consumer reviews written on the social media websites reflects product sales, previous studies have identified a positive relation Analytics in smart tourism design concepts and methods part 2ship between online reviews and revenues in hotels (Xie, Chen, & Wu. 2012) and restaurants (Zhang. Ye. Law. & Li, 2010). For example. Ye et al. (2011)Analytics in smart tourism design concepts and methods part 2
found that a 10% increase in travel review ratings improves the volume of hotel bookings by more than 5 %. A study conducted by Ogut and Tas (2012) cEstimating the Effect of Online Consumer Reviews: An Application of Count Data ModelsSangwon Park1IntroductionThe advent of information technology has Analytics in smart tourism design concepts and methods part 2ality and service of restaurants, as well as the volume of reviews, also have positive relationships with restaurant popularity (Zhang cl al., 2010). Additionally, high ratings of online reviews tend to generate price premiums (Yacouel & Fleischer. 2012; Zhang. Ye, & Law, 2011). Online reviews, pote Analytics in smart tourism design concepts and methods part 2ntially representing service quality, lead consumers to have increased confidence in their decisions. This increase in trustworthiness encourages travAnalytics in smart tourism design concepts and methods part 2
ellers to pay higher prices when purchasing tourism products.Estimating the Effect of Online Consumer Reviews: An Application of Count...149With regarEstimating the Effect of Online Consumer Reviews: An Application of Count Data ModelsSangwon Park1IntroductionThe advent of information technology has Analytics in smart tourism design concepts and methods part 2avel planning process, including pre-, during- and post-consumption. Specifically, positive reviews with numerical ratings improve altitudes toward travel products, being associated with the formation of consideration sets (Vermeulen & Seegers. 2009) and purchasing intentions (Sparks & Browning. 201 Analytics in smart tourism design concepts and methods part 21). Filicri and McLcay (2014) attempted to identify the factors that bring about the adoption of online information by consumers w ith regard to the eAnalytics in smart tourism design concepts and methods part 2
laboration likelihood theory, including the central route (c.g. information accuracy, value-added information, information relevance, information timeEstimating the Effect of Online Consumer Reviews: An Application of Count Data ModelsSangwon Park1IntroductionThe advent of information technology has Analytics in smart tourism design concepts and methods part 2 to online reviews concerning the trustworthiness, helpfulness and usefulness of the reviews (Racherla & Friskc, 2012; Wei, Miao, & Huang, 2013). Il has been recognised in this research that positive review's are likely to be more favourable than negative comments, and heuristic cues of online revie Analytics in smart tourism design concepts and methods part 2ws lead readers to enlarge the perceived helpfulness of the review's. A recent research by Liu and Park (2015) concluded that the messenger characteriAnalytics in smart tourism design concepts and methods part 2
stics (c.g. disclosed photo, reviewers’ expertise) and message characteristics (number of W'ords. star ratings readability) of the online reviews affeEstimating the Effect of Online Consumer Reviews: An Application of Count Data ModelsSangwon Park1IntroductionThe advent of information technology has Analytics in smart tourism design concepts and methods part 2od or experimental design approach to estimate the effect of online comments on consumer behaviours (Schuckcrt Ct al.. 2015). Importantly, however, this study uses data reflecting actual user behaviours collected from a real tourism review website. Thus, it is suggested that an alternative method of Analytics in smart tourism design concepts and methods part 2 count models—the negative binomial model—better addresses the research question, as discussed in the following section.3Count ModelsCount models dealAnalytics in smart tourism design concepts and methods part 2
with specific types of data, which are discrete, using a non-negative integer (e.g. 0. 1.2...). which stand for counts rather than rankings. In otherEstimating the Effect of Online Consumer Reviews: An Application of Count Data ModelsSangwon Park1IntroductionThe advent of information technology has Analytics in smart tourism design concepts and methods part 2er of occurrences of an event. Since count data is distinct from binary data consisting of two values (‘0’ or ‘ I ’), alternative estimations have been suggested for use. such as the Poisson and negative binomial models (Castéran & Roedcrcr. 2013; Czajkowski. Giergiczny, Kronenbcrg, & Tryjanowski, 2 Analytics in smart tourism design concepts and methods part 2014; Hellerstein & Mendelsohn, 1993). While the linear least square regression coping with continuous variables is applicable, the estimated results cAnalytics in smart tourism design concepts and methods part 2
an be inefficient, inconsistent and biased (Cameron & Trivcdi. 2013). This is because the response variable is categorical or discrete, which often prEstimating the Effect of Online Consumer Reviews: An Application of Count Data ModelsSangwon Park1IntroductionThe advent of information technology has Analytics in smart tourism design concepts and methods part 2nThe Poisson model is useful when the outcome is count with which the large count becomes rare occurrences (Kulner. Nachtsheim. Neler. & Li. 2004). The Poisson function predicts the number of occurrences of events (Y 0. I, 2 ...) during an interval of lime. The Poisson distribution can be expressed Analytics in smart tourism design concepts and methods part 2as follows:where Y refers to a Poisson distribution with parameter (or intensity) pTherefore it can be said that p exp (x',0).Importantly, one of propAnalytics in smart tourism design concepts and methods part 2
erties of the Poisson estimation is the equality of mean and variance for p >0, known as cquidispersion (Cameron & Trivedi, 2013).E(y|z) = vw(y|/) = pGọi ngay
Chat zalo
Facebook