No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Overcoming Self-Selection Biases in Online Product Reviews
Abstract
Online product reviews help consumers infer product quality, and the mean (average) rating is often used as a proxy for product quality. However, two self-selection biases, acquisition bias (mostly consumers with a favorable predisposition acquire a product and hence write a product review) and underreporting bias (consumers with extreme, either positive or negative, ratings are more likely to write reviews than consumers with moderate product ratings), render the mean rating a biased estimator of product quality, and they result in the well-known J-shaped (positively skewed, asymmetric, bimodal) distribution of online product reviews. To better understand the nature and consequences of these two self-selection biases, we analytically model and empirically investigate how these two biases originate from consumers' purchasing and reviewing decisions, how these decisions shape the distribution of online product reviews over time, and how they affect the firm's product pricing strategy. Our empirical results reveal that consumers do realize both self-selection biases and attempt to correct for them by using other distributional parameters of online reviews, besides the mean rating. However, consumers cannot fully account for these two self-selection biases because of bounded rationality. We also find that firms can strategically respond to these self-selection biases by adjusting their prices. Still, since consumers cannot fully correct for these two self-selection biases, product demand, the firm's profit, and consumer surplus may all suffer from the two self-selection biases. This paper has implications for consumers to leverage online product reviews to infer true product quality, for commercial websites to improve the design of their online product review systems, and for product manufacturers to predict the success of their products.
... When making online purchases, buyers often rely on online reviews to gather more information (Yin et al., 2021). Previous research has shown that online reviews have significant impacts on buyers' expectation and behaviors (Ding et al., 2022;Hu et al., 2017;Li & Hitt, 2008). However, a major concern with online reviews is the presence of self-selection bias (Askalidis et al., 2017;Jha & Shah, 2021;Piramuthu et al., 2012;Wu et al., 2020;Zhou & Guo, 2017). ...
... Together, acquisition bias and self-selection bias can result in a J-shaped distribution of biased online reviews, where there is a high volume of extremely positive reviews, a moderate volume of extremely negative reviews, and a low volume of moderate reviews. As a result, the presence of the two self-selection biases can lead to the unreliability of online reviews (Hu et al., 2017). ...
... Buyer agents rely on online reviews to make purchases, subsequently generating satisfaction based on their expectation and actual experienced product quality (Anderson & Sullivan, 1993). Then, they produce their product reviews and decide whether to submit their reviews according to the generated satisfaction (Hu et al., 2017). By interpreting how buyer satisfaction changes with different self-selection bias and product types, we generate theoretical insights and build a novel theory to explain the effects of self-selection bias on buyer satisfaction. ...
Please cite this article as: Y. Xie, W. Yeoh and J. Wang, How self-selection Bias in online reviews affects buyer satisfaction: A product type perspective, Decision Support Systems (2024), https://doi.org/10.1016/j.dss.2024.114199 ABSTRACT Online reviews play a crucial role in shaping buyers' purchase decisions. However, previous research has highlighted the existence of self-selection biases among buyers who contribute to reviews, which in turn leads to biased distributions of review ratings. This research aims to explore the further influences of self-selection bias on buyer satisfaction through agent-based modeling, considering two product differentiations: search and experience differentiation, as well as vertical and horizontal differentiation. Our findings reveal that self-selection bias can have varying positive and negative effects on the usefulness of online reviews in suggesting product quality (i.e., review utility) to buyers, thus affecting buyer satisfaction. While self-selection bias tends to decrease review utility in most scenarios, interestingly, it can also increase review utility by enabling a "screening" function of online reviews in addition to its normal "measuring" function. We also find that the varying effects of self-selection bias on buyer satisfaction are contingent upon the type of products under scrutiny and the interaction of different types of self-selection bias. This research makes valuable contributions to the existing literature on online reviews by introducing a novel theory to explain the effects of self-selection bias on buyer satisfaction. J o u r n a l P r e-p r o o f Journal Pre-proof 2
... Yuan et al. (2012) recognize self-selection bias in online reviews, manifesting in purchasing and under-reporting bias. In the same manner, Hu et al. (2017) refer to two components of self-selection bias that both affect reporting: Acquisition bias, as consumers with a positive predisposition acquire a product and write reviews, and under-reporting bias, which is related to the intensity of positivity or negativity affecting the decision to post a product or service experience. The effect of self-selection on mean ratings can distort product quality evaluations overall. ...
... Kuan et al., 2015;Liu & Karahanna, 2017;Marinescu et al., 2021;Wu et al., 2018 self-selection biasHu et al., 2017;Shen et al., 2020;Yuan et al., 2012;Zhang et al., 2018 reporting biasChen et al., 2016;de Barra, 2017;Han & Anderson, 2020;Hu et al., 2017;Jurca et al., 2010; Karamana, 2021;Koh et al., 2010;Osman et al., 2019;Yuan et al., 2012 under-reporting biasKoh et al., 2010;Wu et al., 2018;Yuan et al. ...
... Kuan et al., 2015;Liu & Karahanna, 2017;Marinescu et al., 2021;Wu et al., 2018 self-selection biasHu et al., 2017;Shen et al., 2020;Yuan et al., 2012;Zhang et al., 2018 reporting biasChen et al., 2016;de Barra, 2017;Han & Anderson, 2020;Hu et al., 2017;Jurca et al., 2010; Karamana, 2021;Koh et al., 2010;Osman et al., 2019;Yuan et al., 2012 under-reporting biasKoh et al., 2010;Wu et al., 2018;Yuan et al. ...
This study conducts a systematic literature review to identify forms of user behavioral bias, affecting the generation of content, evaluation patterns, and the impact of online reviews. Furthermore, it indicates the resulting implications for organizations' management of customer knowledge. The authors collected studies dating from 2009 to 2023 in Scopus, where a high dimensionality of bias definitions is observed. As a result of this review, a series of thematic distinctions is proposed, distinguishing sentiment and evaluation asymmetries, motivational factors, as well as on-platform effects that can amplify, adjust, or create biased behaviors. The authors draw analogies with cognitive bias literature regarding cognitive limitations, preconceptions and motives, and contextual adaptations that lead to biased behaviors. Suggested methods and analyses aiming at bias understanding and mitigation are also presented in detail. The authors provide a comprehensive framework of bias-centered management in online review platforms, incorporating customer knowledge flows and practices to monitor and address bias effects.
... There has been a significant interest in the biases that can affect the reliability of ratings given by online consumers (e.g., G. Gao et al., 2015;Han & Anderson, 2020;Hu et al., 2017;Jiang & Guo, 2015;X. Li & Hitt, 2008;F. ...
... This phenomenon may lead to a distortion in the representation of average opinions, with published reviews failing to capture the full range of experiences of all users. Hu et al. (2017) have broken down the self-selection bias into two distinct biases: the underreporting bias and the acquisition bias. ...
... Acquisition bias occurs when consumers with positive expectations or exceptional experiences are more inclined to acquire a product or service and, consequently, to leave reviews. This can often lead to inflated ratings, as only consumers with a positive expected net utility tend to write a review (Hu et al., 2017). However, this tendency toward inflated ratings can be ambiguous: These reviews may reflect an authentic and faithful assessment of the consumer experience or overconfidence in their ability to choose good products or services. ...
It has been established in the literature that the number of ratings and the scores restaurants obtain on online rating systems (ORS) significantly impact their revenue. However, when a restaurant has a limited number of ratings, it may be challenging to predict its future performance. It may well be that ratings reveal more about the user who gave the rating than about the quality of the restaurant. This motivates us to segment users into “inflating raters,” who tend to give unusually high ratings, and “deflating raters,” who tend to give unusually low ratings, and compare the rankings generated by these two populations. Using a public dataset provided by Yelp, we find that deflating raters are better at predicting restaurants that will achieve a top rating (4.5 and above) in the future. As such, these deflating raters may have an important role in restaurant discovery.
... In other words, by observing the distribution of numerical ratings, some items stand out as statistical outliers. For example, Extremely Bipolar Items (EBIs) exhibit anomalous inflation in the frequency of maximum and minimum scores of the rating scale, an empirical occurrence that is not observed outside the cluster of EBIs, and that is related to a source of controversy regarding these items (Li and Hitt, 2008;Hu et al., 2009;Anderson and Simester, 2014;Fu et al., 2015;Amendola et al., 2015;Hu et al., 2017;Zhuang et al., 2018;Santos et al., 2019;Janosov et al., 2020;Lu et al., 2020;Schoenmueller et al., 2020;Day and Kim, 2022;Ziser et al., 2023;Li et al., 2023;Cantone et al., 2024). ...
... However, they do not hold for online platforms, which are typically characterised by a bi-modal, concave, shape (J-shape). Hu et al. (2017) explain the J-shape as determined by: ...
Online reviews provide users with the opportunity to rate various types of items such as movies, music, and video games using a combination of numeric scores and textual comments. The study proposes a novel method that applies statistical matching on network-based covariates, with the aim to improve the estimation of the association between words and highly controversial items in online reviews. The application of this method on a sample of 40,665 items from the website Metacritic detects 218 highly controversial items. The application supports the theory that controversies on Metacritic are driven with a sense of self-awareness of participating of an online controversy (‘review bombing’). Typical controversial topics (sexual identities, religious morality, politics) are associated with controversial reviews, too.
... A body of research explores strategies to mitigate various biases in human decision-making, including cognitive and data biases [54], self-selection biases [55], racial bias [43,56] and gender bias [44]. For instance, evidence suggests that mindfulness helps decision-makers stay open to new information and avoid reinforcing initial biases, thereby mitigating the harmful effects of cognitive and data biases [54]. ...
... For instance, evidence suggests that mindfulness helps decision-makers stay open to new information and avoid reinforcing initial biases, thereby mitigating the harmful effects of cognitive and data biases [54]. Similarly, adjusting prices may counteract the effects of self-selection biases in online reviews, which distort the true representation of product quality and influence customers' purchasing decisions [55]. Like our study, [56] and [43] focus on societal biases such as race and gender. ...
Biased human decisions have consequential impacts across various domains, yielding unfair treatment of individuals and resulting in suboptimal outcomes for organizations and society. In recognition of this fact, organizations regularly design and deploy interventions aimed at mitigating these biases. However, measuring human decision biases remains an important but elusive task. Organizations are frequently concerned with mistaken decisions disproportionately affecting one group. In practice, however, this is typically not possible to assess due to the scarcity of a gold standard: a label that indicates what the correct decision would have been. In this work, we propose a machine learning-based framework to assess bias in human-generated decisions when gold standard labels are scarce. We provide theoretical guarantees and empirical evidence demonstrating the superiority of our method over existing alternatives. This proposed methodology establishes a foundation for transparency in human decision-making, carrying substantial implications for managerial duties, and offering potential for alleviating algorithmic biases when human decisions are used as labels to train algorithms.
... The numerical rating and the textual content are two components of the OCRs through which customers express uncertainty and trust towards the product. The numerical rating is often positively biased and customers are generally aware of this (Hu et al., 2017) and this leads to uncertainty. However, if the previous product users express trust in the product in the OCR, this can reduce the uncertainty perceived by the customer to an extent. ...
... The deviation or variance in the rating reflects the reviewer's uncertainty about the product (de Maeyer, 2012).While rating variance for movies has not impacted their box office performance (Chintagunta et al., 2010), it has impacted the customers' expected utility of newly introduced products such as high-end digital cameras (Markopoulos & Clemons, 2013). For books with a lower (higher) average rating, a higher (lower) rating deviation has shown a positive impact on their sales (Hu et al., 2017). One interesting study observed that, in the case of restaurants, customers uncertainty increased with the volume of reviews and decreased after a threshold volume was crossed (Bang & SooCheong (Shawn) Jang, 2024). ...
While the research on online consumer reviews is immense, it has largely focused on products which can also be purchased online. However, the high level of digital engagement of individuals today along with a reported fall in physical store visits indicate that digital content can also affect products available only for purchase offline. This research examines the effects of trust, uncertainty, and topics extracted from online consumer reviews on two outcomes in the India car market, namely the search for online information and sales. The study finds that while uncertainty does not affect sales, and has a negative effect on online information search, trust is positively associated with both. The topics extracted using Latent Dirichlet Allocation from the review corpus fall under the category of experiential or functional. The different topics have direct and mediating impacts on search and sales.
... From an external perspective, firstly, rating inflation makes it laborious for consumers to differentiate products based solely on online reviews [3]. Typically, the distribution of online review ratings is J-shaped, as consumers tend to generate more positive reviews [39]. Besides, the presence of fake reviews further complicates differentiation under these circumstances [51,70]. ...
Online reviews influence consumers’ assessments of product qualities, which in turn can affect their purchase decisions. However, the impact of online reviews is constrained by the information acquisition factor, a process through which consumers obtain information to assess product qualities. This paper examines the diminishing marginal impact of online reviews on consumers’ purchase decisions from the perspective of information acquisition. Leveraging HCI theories of competition for attention and visual search, we study their direct (i.e., on purchase intentions) and indirect (i.e., mediated by information acquisition) impacts on consumer decisions. We conducted empirical experiments and mathematical modeling to test our research hypotheses and framework. The findings reveal that video/image reviews have both direct and indirect diminishing marginal impacts on purchase intentions. This paper contributes to the literature on online reviews and offers practical implications for e-commerce stakeholders.
... Additionally, users are more likely to rate items if they are either satisfied or dissatisfied rather than if they feel neutral about the product [8]. 3) Rating scale bias reflects personal differences in the use of rating scales across users [3,5]. ...
In this study, we partition users by rating disposition - looking first at their percentage of negative ratings, and then at the general use of the rating scale. We hypothesize that users with different rating dispositions may use the recommender system differently and therefore the agreement with their past ratings may be less predictive of the future agreement. We use data from a large movie rating website to explore whether users should be grouped by disposition, focusing on identifying their various rating distributions that may hurt recommender effectiveness. We find that such partitioning not only improves computational efficiency but also improves top-k performance and predictive accuracy. Though such effects are largest for the user-based KNN CF, smaller for item-based KNN CF, and smallest for latent factor algorithms such as SVD.
... Despite the widespread use of online reviews and celebrity endorsements as marketing tools, their specific impacts on the brand reputation of healthcare services remain a significant area of concern, particularly in the diverse and dynamic market of India (Chandra and Tripathi, 2023;Chou et al., 2024). Online reviews have the potential to significantly influence a potential patient's decision-making process by providing firsthand accounts of patient experiences; however, the reliability of these reviews can be compromised due to issues such as fake reviews, biased perspectives, or paid endorsements (Lappas et al., 2016;Hu et al., 2017). These variations add complexity to the task of healthcare providers in effectively managing their online reputation (Jenkins et al., 2020;Patel et al., 2023). ...
Impact of Online Reviews and Celebrity Endorsement on the Brand Reputation of Health Care Services: Evidence from India .
by Saifuddin Ahmad, Ved Srivastava, Raj Kumar Mishra, Yusuf Kamal
Abstract: This research examines the effects of online reviews and celebrity endorsements on the brand reputation of healthcare services. Empirical data were gathered through a survey using 32 scale items that were adapted and refined from previous studies. These items were specifically crafted to assess eight essential variables associated with celebrity endorsement and brand reputation. A purposive sampling methodology was utilised to select 524 online purchasers from e-commerce platforms operating in India. The findings indicate that Celebrity Attractiveness, Celebrity Credibility, and Online Reviews exert a significant influence on consumer perceptions of celebrities within the Indian healthcare industry. Moreover, Online Reviews, Brand Awareness, and Brand Loyalty exert a positive effect on Brand Attitude. Additionally, Attitude Toward Celebrity has a substantial positive impact on both Brand Attitude and Brand Reputation. The results underscore the pivotal importance of online reviews and celebrity endorsements in shaping the brand reputation of healthcare services in India.
Keywords: Celebrity Endorsement; Brand Reputation; Celebrity Attractiveness; Celebrity Credibility; Online Reviews.
DOI: 10.1504/IJPMB.2024.10068318
... In particular, studies (Hu et al. 2017;Jabr and Zheng 2014;Kim and Yoo 2020) indicate that high variance in reviews can lead to decision uncertainty, negatively impacting firm performance. This information consistency complicates consumer choices and can result in cognitive overload, which diminishes the effectiveness of information processing, challenging consumers' ability to make informed decisions (Liu and Karahanna 2017;Kampani and Nicolaides 2023). ...
Understanding search frictions and their impact on market shares is crucial for devising effective business strategies in digital marketplaces. This study introduces a novel controversy metric to proxy search frictions, linking three interconnected research areas: search frictions’ relationship with market share, price signaling by reputable developers, and cross-review controversy’s impact on consumer behavior. Focusing on the PC game industry on the Steam platform, we apply panel data regression analysis to reveal that search friction, as captured by cross-review controversy, exhibits an inverted-U impact on market shares, challenging traditional models. Contrary to prevailing assumptions, our research shows that established publishers do not consistently use price premiums to signal quality. Furthermore, cross-review controversy diminishes the perceived helpfulness of reviews, potentially leading to purchase deferrals. While the volume of reviews and negative sentiments are moderated by controversy, positive comments remain largely unaffected. This research provides insights into how quality signals and user-generated content shape market outcomes in digital marketplaces and highlights how an optimal level of search frictions can maximize market shares, enhancing understanding of information processing and marketing strategies in markets for experience goods.
... Consumers often rely on online reviews to inform their purchase decisions, regardless of the platform or product category [1][2][3][4][5]. This strong correlation between online reviews and sales is well documented [6]. ...
Studies have found that competitive products’ online review ratings (ORRs) have a spillover effect on the focal product’s sales. However, the spillover effect of online review sentiment (ORS) as an essential component of online review analysis has yet to be studied. In this study, we analyze online review content from JD.com using the latent Dirichlet allocation to identify the product attribute topics that consumers are most concerned about. We then construct a baseline regression model of ORS and ORRs to explore the effects of online competitive product reviews on focal product sales. Moreover, we examine how the interaction between ORS and critical factors of online reviews affect sales. Our results indicate that the ORS of competitive products has a negative effect on focal product sales, and the effect is greater than the ORS and ORRs of focal products, respectively. In addition, the ORS of competitive products inhibits the sale of focal products as evaluations of product attributes become more positive or online review usefulness increases. We also find that the effect of ORRs of competitive products is not significant, which may be because clothing, as an experiential product, requires consumers to gain more information about specific usage scenarios before making a decision. This study provides a more accurate basis for consumer decision-making and offers retailers a novel approach to developing marketing strategies.
... Self-injecting favorable online reviews have been proven to corrode review credibility (Jabr 2022), cause rating bias (Hu et al. 2017), and diminish consumer Wang et al.: Self-Injecting Favorable Online Reviews Information Systems Research, Articles in Advance, pp. 1-21, © 2024 INFORMS 3 surplus (Mayzlin et al. 2014). ...
Literature has long assumed that unethical behaviors are fueled by competition. This study challenges this prevailing notion by introducing a conceptual distinction between rivalry and competition. We show that rivalry-induced deterrence triggers the formation of mutual forbearance in self-injecting favorable reviews. Mutual forbearance collapses when one party unilaterally engages in self-injecting behaviors, prompting the other party to retaliate by increasing their self-injecting activities. The presence of an additional rival leads to a 3.2% decrease in self-injecting intensity. Furthermore, self-injecting intensity increases by 0.095% for every 1% rise in that of the rivals and increases by 0.041% for a 1% rise in that of the nonrival competitors. Given the apparent role of rivalry in reducing self-injecting behaviors, platforms should strategically consider how system designs and policies can cultivate firms’ feelings of rivalry and avoid non-rival competition. We recommend platforms consider including evaluations from peer hotels as an alternative index for recommending hotels to potential consumers. Introducing peer evaluations to the current online review systems could attenuate the bias of consumer ratings and facilitate the formation of rivalry, which hinders self-injecting behaviors.
... This is also empirically supported by [BMZ12] which shows that Groupon discounts lead to lower ratings. [HPZ17] study a two-stage model which quantifies both self-selection bias and under-reporting bias (reviews are provided only by customers with extreme experiences); see references within for further related work. ...
We study a model of social learning from reviews where customers are computationally limited and make purchases based on reading only the first few reviews displayed by the platform. Under this bounded rationality, we establish that the review ordering policy can have a significant impact. In particular, the popular Newest First ordering induces a negative review to persist as the most recent review longer than a positive review. This phenomenon, which we term the Cost of Newest First, can make the long-term revenue unboundedly lower than a counterpart where reviews are exogenously drawn for each customer. We show that the impact of the Cost of Newest First can be mitigated under dynamic pricing, which allows the price to depend on the set of displayed reviews. Under the optimal dynamic pricing policy, the revenue loss is at most a factor of 2. On the way, we identify a structural property for this optimal dynamic pricing: the prices should ensure that the probability of a purchase is always the same, regardless of the state of reviews. We also study an extension of the model where customers put more weight on more recent reviews (and discount older reviews based on their time of posting), and we show that Newest First is still not the optimal ordering policy if customers discount slowly. Lastly, we corroborate our theoretical findings using a real-world review dataset. We find that the average rating of the first page of reviews is statistically significantly smaller than the overall average rating, which is in line with our theoretical results.
... User reviews are proven to impact customer behaviors greatly. Products with higher ratings and more positive reviews were likely to be purchased in dramatically larger quantities by customers [2]. On top of that, the study also discovered that the number and the length of reviews were also essential indicators that influences customer behaviors to a great extent [3]. ...
With the rapid growth of e-commerce, accurately capturing buyers' sentiments through their reviews is increasingly vital for online marketplaces. In this paper, we aim to deal with sentiment analysis in these reviews by exploring effective methods to analyze them. We use a review dataset containing user ratings and comments on Amazon products. Applying the two-step methodology of data preprocessing and model building, we intend to employ models like LSTM and SVM to analyze Amazon customer reviews and gain insights into their performance. The findings of this study may also allow e-commerce platforms to provide better service to sellers and buyers.
... Thus, reviews are more likely to be perceived as misleading if they are attributed more to reviewer factors, such as emotions. Furthermore, self-selection biases exist in online reviews, including acquisition bias (where most consumers with a favorable predisposition acquire products and write reviews) and underreporting bias (where consumers with extreme ratings are more inclined to write reviews compared with those with moderate product ratings; Hu et al., 2017). These biases are related to reviewer attribution. ...
Purpose
The study examines the potential moderating effects of repeating purchase cues and product knowledge on the relationship between the varying consistency of the review set and causal attribution. This study also investigates how causal attribution correlates with the perceived misleadingness of the review set.
Design/methodology/approach
A scenario-based experiment was conducted with 170 participants to explore the relationship between the consistency of the review set and causal attribution and how repeating purchase cues and product knowledge moderates this relationship.
Findings
Findings suggest that inconsistent review sets lead to more product (vs reviewer) attribution than consistent review sets. The repeating purchase cues mitigate the negative relationship between the consistency of the review set and product attribution, whereas product knowledge mitigates the positive relationship between the consistency of the review set and reviewer attribution. Furthermore, the results indicate that high product attribution and low reviewer attribution are associated with low perceived misleadingness.
Originality/value
This study is novel because it examines the moderating effects of repeating purchase cues and product knowledge on the relationship between the consistency of the review set and causal attribution. It adds to the literature by shedding light on the causal attribution process underlying the formation of perceived misleadingness of online reviews. The findings of this study provide valuable insights for managers on how to enhance the positive effects of consistent review sets and mitigate the negative effects of inconsistent review sets.
... Some customer has their own biases when writing a review and these biases in rating may result in inconsistent reviews. Hu et al. [3] also claimed that customers are not entirely rational, and that affects selfselection rating biases. www.ijacsa.thesai.org ...
... However, user-generated content has inherent biases. People with extreme opinions (either positive or negative) are more likely to leave their comments than those with moderate opinions, resulting in an underreporting bias [42]. The analytic results can thus be biased toward the extremes. ...
In the early stages of the design process, designers explore opportunities by discovering unmet needs and developing innovative concepts as potential solutions. From a human-centered design perspective, designers must develop empathy with people to truly understand their needs. However, developing empathy is a complex and subjective process that relies heavily on the designer’s empathetic capability. Therefore, the development of empathetic understanding is intuitive, and the discovery of underlying needs is often serendipitous. This paper aims to provide insights from artificial intelligence research to indicate the future direction of AI-driven human-centered design, taking into account the essential role of empathy. Specifically, we conduct an interdisciplinary investigation of research areas such as data-driven user studies, empathetic understanding development, and artificial empathy. Based on this foundation, we discuss the role that artificial empathy can play in human-centered design and propose an artificial empathy framework for human-centered design. Building on the mechanisms behind empathy and insights from empathetic design research, the framework aims to break down the rather complex and subjective concept of empathy into components and modules that can potentially be modeled computationally. Furthermore, we discuss the expected benefits of developing such systems and identify current research gaps to encourage future research efforts.
... In addition, while numerical ratings are easy to collect and process, previous research has shown that they may not fully capture consumer behavior or may be subject to biases in review distributions (e.g., Hu, Pavlou, and Zhang 2017;Karaman 2021). Ludwig et al. (2013) discuss the inconclusive results regarding the inference of numerical ratings, which might stem from the inability of numerical ratings to deliver full information such as the nuanced, fine-grained, and expressive nature of text reviews. ...
The authors examine the effects of a firm’s and its competitors’ online reviews on its demand within the hotel industry. The authors leverage a unique data set of actual bookings from properties of a major hotel chain in six different markets in the United States, supplemented with online reviews garnered from a popular social media platform. The findings indicate that not only a hotel’s own reviews but also its competitors’ reviews have a significant impact on the hotel’s booking performance. The impact of review sentiment is amplified if the focal hotel also charges higher prices or when the volume of reviews is high. The authors establish heterogeneous effects across consumer segments (business vs. leisure travelers) and by the type of review content (objective vs. subjective attributes to assess quality). Specifically, both a hotel’s own reviews and its competitors’ reviews have a larger impact on bookings for business travelers compared with leisure travelers, and for reviews that mainly discuss subjective attributes, for which consumers need to rely on the experiences of others to assess the quality of a hotel prior to their stay. The study provides a set of comprehensive insights on the impact of both own and competitors’ online reviews on a focal hotel’s bookings.
... While online reviews offer advantages in terms of the population of participants as well as the representations of thousands of companies in our sample, which will be very difficult to achieve with survey data, at the same time, the predefined scales found in employee online review platform fail to directly measure constructs such as job stress, role ambiguity, role conflict or role overload. Another limitation inherent to the nature of online reviews is the existence of several bias (Hu et al., 2017;Li & Hitt, 2008) or the probability of online reviews manipulation (Mayzlin et al., 2014). To some extent, our empirical design addressed such biases. ...
Online platforms increasingly utilize technologies like artificial intelligence (AI)-empowered tools to reduce consumers’ search costs and simplify decision making. However, these tools often target specific types of information, leading to what we term "search cost reduction for partial information." Although designed to assist consumers, our study highlights their unintended consequence: these tools can induce "cognitive miser" behavior, where consumers focus on easily accessible information while neglecting other critical details. This behavior can ultimately result in poorer decision making. Using a natural experiment on Yelp, we evaluated the impact of its AI-powered image categorization feature, introduced in 2015 to reduce the search costs of review images. Through a difference-in-differences design and text analysis of consumer complaints, we found that this feature negatively affected decision quality. These findings carry important implications for platform managers and policymakers. Although search cost reduction tools can improve efficiency, they also risk biasing consumer attention toward easily accessible information at the expense of holistic decision making. Online platforms could mitigate these effects by complementing AI-empowered search cost reduction features with tools that emphasize information requiring greater cognitive effort, thereby ensuring balanced consumer awareness. We recommend that platform designers carefully evaluate the broader impacts of such tools to better support consumer decision making.
Adding explanations to recommender systems is said to have multiple benefits, such as increasing user trust or system transparency. Previous work from other application areas suggests that specific user characteristics impact the users’ perception of the explanation. However, we rarely find this type of evaluation for recommender systems explanations. This paper addresses this gap by surveying 124 papers in which recommender systems explanations were evaluated in user studies. We analyzed their participant descriptions and study results where the impact of user characteristics on the explanation effects was measured. Our findings suggest that the results from the surveyed studies predominantly cover specific users who do not necessarily represent the users of recommender systems in the evaluation domain. This may seriously hamper the generalizability of any insights we may gain from current studies on explanations in recommender systems. We further find inconsistencies in the data reporting, which impacts the reproducibility of the reported results. Hence, we recommend actions to move toward a more inclusive and reproducible evaluation.
Electronic word-of-mouth (eWOM) dispersion often increases perceived risk and reduces purchase intention; however, a high dispersion of eWOM regarding local food may have different effects on tourists. This study explores how eWOM dispersion influences tourists’ purchase intentions for local food. We find that high eWOM dispersion actually stimulates tourists’ curiosity, thereby enhancing their intention to purchase local food. This effect is moderated by the dimensions of eWOM and regulatory focus. Specifically, high dispersion regarding taste increases purchase intention, whereas high dispersion regarding quality decreases it. For tourists with a promotion focus, high dispersion strengthens purchase intention, but it is vice versa for those with a prevention focus. This study extends the literature on eWOM dispersion in the marketing of local food and reveals the mechanism of curiosity, providing practical guidance for local food restaurants.
Adding explanations to recommender systems is said to have multiple benefits, such as increasing user trust or system transparency. Previous work from other application areas suggests that specific user characteristics impact the users' perception of the explanation. However, we rarely find this type of evaluation for recommender systems explanations. This paper addresses this gap by surveying 124 papers in which recommender systems explanations were evaluated in user studies. We analyzed their participant descriptions and study results where the impact of user characteristics on the explanation effects was measured. Our findings suggest that the results from the surveyed studies predominantly cover specific users who do not necessarily represent the users of recommender systems in the evaluation domain. This may seriously hamper the generalizability of any insights we may gain from current studies on explanations in recommender systems. We further find inconsistencies in the data reporting, which impacts the reproducibility of the reported results. Hence, we recommend actions to move toward a more inclusive and reproducible evaluation.
We study the impact of consumer heterogeneity on online ratings. Consumers differ in their experience, which can affect both their choices and ratings. Thus, biases in average ratings can arise when the opinions of experienced and novice users are aggregated. We first build a two-period model to characterize the biases’ drivers and consequences. We test our theory combining data from IMDb and MovieLens, two well-known movie ratings platforms. We proxy users’ experience with the total number of ratings posted on the platforms. First, using external measures of quality, such as the Academy awards and nominations, we show that, on both platforms, experienced users, on average, rate movies of higher quality compared with novices. Moreover, they post more stringent ratings than novices for more than 98% of movies. Combined, these imply a compression in aggregate ratings, and thus a bias against high quality movies. We then propose a simple, fixed-point algorithm to debias ratings. Our debiased ratings demonstrate the presence of ranking reversals for more than 8% of comparisons in our sample. As a result, our debiased ratings better correlate with external measures of quality.
This paper was accepted by Eric Anderson, marketing.
Supplemental Material: The online appendices and data files are available at https://doi.org/10.1287/mnsc.2022.03281 .
This paper investigates how the presence of selection bias affects the interplay between customer reviews and price when marketing experience goods with uncertain quality to consumers.
We developed a security guard robot that is specifically designed to manage queues of people and conducted a field trial at an actual public event to assess its effectiveness. However, the acceptance of robot instructions or admonishments poses challenges in real-world applications. Our primary objective was to achieve an effective and socially acceptable queue-management solution. To accomplish this, we took inspiration from human security guards whose role has already been well-received in society. Our robot, whose design embodied the image of a professional security guard, focused on three key aspects: duties, professional behavior, and appearance. To ensure its competence, we interviewed professional security guards to deepen our understanding of the responsibilities associated with queue management. Based on their insights, we incorporated features of ushering, admonishing, announcing, and question answering into the robot’s functionality. We also prioritized the modeling of professional ushering behavior. During a 10-day field trial at a children’s amusement event, we interviewed both the visitors who interacted with the robot and the event staff. The results revealed that visitors generally complied with its ushering and admonishments, indicating a positive reception. Both visitors and event staff expressed an overall favorable impression of the robot and its queue-management services. These findings suggest that our proposed security guard robot shows great promise as a solution for effective crowd handling in public spaces.
This study examines how unexpected exogenous events, labelled as suprises, affect the utility of experience goods reported in online rating systems. Using over 300,000 reviews of accommodation services listed on Booking.com, the research investigates whether online ratings capture the impact of surprises related to meteorological conditions and whether they create additional biases in service evaluation. The study finds that sudden changes in weather conditions have a significant impact on experienced utility, with the effect varying based on the direction of the surprise. Additionally, in line with the hedonic adaption theory, we find that the duration of consumption moderates the surprise effect, reducing its impact on reported utility.
Purpose
This study aims to examine the impact of consumer risk appetite, biases (specifically negative recency bias), and the importance of reviews in enhancing information quality. By analyzing these variables, the authors gain insights into their role in enriching the overall information spectrum available to consumers. The findings contribute to a better understanding of how risk appetite, biases and consumer reviews shape the quality of information.
Design/methodology/approach
The questionnaire assessed the relationship between dependent and independent variables by asking participants to rate their experiences in relevant scenarios. Variance-based structural equation modeling with the ADANCO program was used to examine the data. ADANCO software is used explicitly for variance-based structural equation modeling. To evaluate research models and test hypotheses, partial least square path modeling is used.
Findings
The efficiency of reviews and ratings is greatly influenced by consumer risk appetite. Businesses should focus on clients who are willing to take risks and balance positive and negative feedback. It is essential to comprehend how customers understand reviews. Credibility is increased by taking biases into account and encouraging unbiased criticism. Promoting thorough reviews strengthens influence. Monitoring and making use of these elements improve online reputation and commercial success.
Research limitations/implications
The research has limitations due to the simplicity of the attributes taken into account and the requirement for a larger sample size. Overcoming barriers to promote consistent client feedback is essential, and tailored emails can help with assessment generation. Increased customer participation in writing evaluations can be achieved by removing obstacles and highlighting the advantages of participation.
Originality/value
Businesses and buyers rely on this “organically” generated content as the basis of their promotional strategy and buying decisions. Most of the research is related to consumer reviews, their behavior and the importance of social validation. However, some critical aspects related to this need further investigation.
Purpose
Online reviews serve as valuable sources of information, reflecting tourists’ attentions, preferences and sentiments. However, although the existing research has demonstrated that incorporating online review data can enhance the performance of tourism demand forecasting models, the reliability of online review data and consumers’ decision-making process have not been given adequate attention. To address the aforementioned problem, the purpose of this study is to forecast tourism demand using online review data derived from the analysis of review helpfulness.
Design/methodology/approach
The authors propose a novel “identification-first, forecasting-second” framework. This framework prioritizes the identification of helpful reviews through a comprehensive analysis of review helpfulness, followed by the integration of helpful online review data into the forecasting system. Using the SARIMAX model with helpful online review data sourced from TripAdvisor, this study forecasts tourist arrivals in Hong Kong during the period from August 2012 to June 2019. The SNAÏVE/SARIMA model was used as the benchmark model. Additionally, artificial intelligence models including long short-term memory, back propagation neural network, extreme learning machine and random forest models were used to assess the robustness of the results.
Findings
The results demonstrate that online review data are subject to noise and bias, which can adversely affect the accuracy of predictions when used directly. However, by identifying helpful online reviews beforehand and incorporating them into the forecasting process, a notable enhancement in predictive performance can be realized.
Originality/value
First, to the best of the authors’ knowledge, this study is one of the first to focus on the data issue of online reviews on tourism arrivals forecasting. Second, this study pioneers the integration of the consumer decision-making process into the domain of tourism demand forecasting, marking one of the earliest endeavors in this area. Third, this study makes a novel attempt to identify helpful online reviews based on reviews helpfulness analysis.
The research was motivated by the growing significance of online courses and the need to understand their success factors through the lens of user experience. Addressing a gap in existing knowledge, the study aims to provide insights into the determinants of course success, specifically focusing on user ratings, and to offer recommendations for enhancing online learning experiences. By analyzing Udemy course data, the study seeks to uncover the factors influencing the success of a course, as perceived by students. The study utilizes logistic regression models to analyze data from Udemy courses, focusing on variables that contribute to the success of a course from the user experience standpoint. The conceptual framework incorporates elements from literature to explain unusual distributions in continuous variable values. The research reveals crucial insights into the determinants of online course success, emphasizing the impact of user ratings on perceived success. The model sheds light on the nuanced dynamics influencing the user experience. The article concludes with practical recommendations for stakeholders, offering a roadmap for future research endeavors in the field of online education.
Online hotel reviews on platforms like TripAdvisor are crucial in shaping customer choices and steering business strategies in the hospitality sector. However, the effectiveness of these platforms is partially hindered by the self-selection bias found in voluntary reviews. This bias can create false expectations and unsatisfactory experiences, mainly as the feedback generally comes from a non-representative group of self-motivated reviewers (SMRs). A common strategy to mitigate this bias is increasing the number of reviews through customer surveys, generating retailer-prompted reviews (RPRs). However, these RPRs, despite reducing selection bias, tend to lack the depth and insight of SMRs, resulting in a credibility gap and reduced representativeness. To address this, our study presents a novel approach using the propensity score adjustment (PSA) technique. This method leverages the distribution of RPRs to refine the accuracy of text data from SMRs, aiming to enhance the reliability and representativeness of online reviews. By combining the strengths of both RPRs and SMRs, we aim to create an online review environment that is both accurate and reliable. In conclusion, this research marks an important step toward improving online review platforms, aiming for a more transparent and trustworthy environment for reviews.
Purpose
The purpose of this study is to test the impact of time and price sensitivity on consumer satisfaction and purchase intention on online-to-offline (O2O) takeout platforms and explore the moderating effect of purchase preference on time sensitivity and satisfaction, as well as price sensitivity and satisfaction, in order to guide market pricing.
Design/methodology/approach
A structural equation model (SEM) of customer purchase intention was constructed, and the relationships between the variables (time sensitivity, price sensitivity, satisfaction and purchase intention) were examined. The completed questionnaires of 349 respondents were collected from the Questionnaire Star platform in China. The research model and hypotheses were then tested. Analytic hierarchy procedure was used to determine the moderating effect of purchase preference. Finally, the study proposes a pricing strategy for customer-active selective services.
Findings
Satisfaction positively influences purchase intention, and price sensitivity significantly increases satisfaction and further increases purchase intention; however, time sensitivity negatively affects satisfaction. Specifically, purchase preference has strongly moderated the relationship between time, price sensitivity and satisfaction. In addition, the findings show that when purchase preference is high, the effect of price sensitivity on satisfaction is stronger, suggesting the importance of purchase preference in strengthening purchase intentions. The research work recommends a pricing strategy involving value-added pricing primarily for time-sensitive customers, which can help build a high-end brand image and reduce price competition. Reduced pricing is mainly for price-sensitive customers, which is conducive to stimulating consumption within a specific time. This pricing strategy is important for adjusting market sensitivity and flexibility.
Originality/value
This research provides new ideas for related disciplines and guidance for the differentiated pricing and promotion of takeout platforms, as well as a theoretical basis for the diversified development of takeout platforms, improvement of personalized service quality and enhancement of customer stickiness. This study fills gaps in the existing literature on the moderating effect of purchase preference on time sensitivity and satisfaction and price sensitivity and satisfaction.
Online platforms have emerged as a dominant business model in numerous industries in the new millennium. In light of the substantial and burgeoning body of empirical platform research, this article synthesizes extant studies and identifies the evolution of underlying research methodologies and topics. Building upon a database of 860 empirical online platform papers in premier journals during the first two decades of the new millennium, this article presents a categorization framework based on the online platform type (including search platforms, e‐commerce platforms, online communities, and mobile platforms) and research perspective (including platform participants, platform orchestrators, and platform ecosystems). We provide a critical review of noteworthy trends and highlight directions for future research in each category of the proposed framework. A comprehensive bibliometric analysis is then conducted to visualize and track scholarship in empirical online platform research. Lastly, we adopt an interdisciplinary lens to synthesize our critical review of empirical online platform research into lessons and research opportunities that emerge from multiple disciplines.
In the early stages of the design process, designers explore opportunities by discovering unmet needs and developing innovative concepts as potential solutions. From a human-centered design perspective, designers must develop empathy with people to truly understand their experiences and needs. However, developing empathy is a complex and subjective process that relies heavily on the designer's empathic capability, and is often subject to the experiences of a small group of people. Therefore, the development of empathic understanding is intuitive, and the discovery of underlying needs can be serendipitous and unrepresentative. This paper aims to provide insights from artificial intelligence research to indicate the future direction of AI-driven human-centered design, considering the essential role of empathy. Specifically, we conduct an interdisciplinary investigation of research areas such as data-driven user research, empathic design, and artificial empathy. Based on this foundation, we discuss the role that artificial empathy can play in human-centered design and propose an artificial empathy framework for human-centered design. Building on the mechanisms behind empathy and insights from empathic design research, the framework aims to break down the rather complex and subjective process of developing empathic understanding into modules and components that can potentially be modeled computationally. Furthermore, we discuss the expected benefits of developing such systems and identify research opportunities to suggest future research efforts.
Gossiping and its reputation effects are viewed as the most powerful mechanism to sustain cooperation without the intervention of formal authorities. Being virtually costless, gossiping is highly effective in monitoring and sanctioning norm violators. Rational individuals cooperate in order to avoid negative reputations. But this narrative is incomplete and often leads to wrong predictions. Goal Framing Theory, a cognitive-behavioral approach anchored in evolutionary research, provides a better explanatory framework. Three overarching goal frames (hedonic, gain, and normative) constantly compete for being in our cognitive foreground. This Element argues that for gossip to have reputation effects, a salient normative goal frame is required. But since the hedonic mindset usually trumps gain and normative concerns, most gossip will be driven by hedonic motives and therefore not have strong reputation effects. Propositions on cultural, structural, dispositional, situational, and technological gossip antecedents and consequences are developed and illustrated with evidence from the empirical record.
Purpose
While there has been extensive research on understanding the effects of online reviews on product sales, there is not enough investigation of the inter-relationships between online reviews, online search and product sales. The study attempts to address this gap in the context of the Indian car market.
Design/methodology/approach
The research uses text mining and considers six important review features volume, valence, length, deviation of valence, sentiment and readability within the heuristic and systematic model of information processing. Panel data regression is used along with mediation analysis to study the inter-relationships between features of reviews, online search and sales.
Findings
The study finds that numerical heuristic features significantly affect sales and online search, numerical systematic feature affects sales and the textual heuristic and systematic features do not affect sales or online search in the Indian car market. Further, online search mediates the association between features of reviews and sales of cars.
Research limitations/implications
Although only car sales data from India is considered in this research, similar relationships between review features, online search and sales could exist for the car market of other countries as well.
Originality/value
This research uncovers the unique role of online search as a mediator between review features and sales, whereas prior literature has considered review features and online search as independent variables that affect sales.
ResearchGate has not been able to resolve any references for this publication.