COMPARING APPLES AND ORANGES:
EMPIRICAL STUDY ON THE EFFECT OF NEGATIVE CUSTOMER REVIEWS ON WILLIGNNESS TO PAY FOR DIFFERENT PRODUCT TYPES
CONSIDERING VARYING CUSTOMER LOYALTY LEVELS
BSc Thesis by Carmen Willingshofer (10508384) Thesis Supervisor: Dr. F.B. Situmeang
27. June 2016
Abstract
Online customer reviews have an increasingly substantial influence on consumer buyer behaviour. Although there is much research conducted on the relation between negative customer reviews and willingness to pay, this research is first in assessing the difference between cost-leaders and differentiators as moderators of this relationship. Additionally, this study considers customer loyalty as a moderator of the relationship between online reviews and willingness to pay. The airlines easyJet and KLM are used as prime examples for the product types throughout the study. Data from these airlines is gathered by means of a cross-sectional survey design (N=331), and is analysed by performing regressions and a 3-way interaction analysis. The results show that negative online customer reviews, product types, and loyalty have a significant impact on initial willingness to pay. However, there are no interactions discovered between the variables, providing no support for the proposed model. However, when analysing the product types separately it becomes evident that for easyJet there exists a negative moderation effect of customer loyalty on the effect of negative customer reviews on willingness to pay. This implies that higher levels of loyalty to easyJet enhance rather than reduce the negative effect of such customer reviews. This negative effect was discovered for both product types but was not significant for KLM.
Keywords: customer reviews, loyalty levels, differentiation versus
Statement of Originality
This document is written by Student Carmen Willingshofer who declares to take full responsibility for the contents of this document.
I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.
The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.
Table of contents 1. Introduction ... 4 2. Theoretical Framework ... 6 2.1 Negative customer reviews and willingness to pay ... 6 2.2 Product types and willingness to pay ... 7 2.3 Customer loyalty as a moderator ... 10 3. Methodology ... 13 3.1 Design ... 13 3.2 Measurements ... 13 3.2.1 Dependent Variable Willingness To Pay ... 14 3.2.2 Independent Variable Negative Online Reviews ... 14 3.2.3 Moderator Variable Product Type ... 15 3.2.4 Moderator Variable Customer Loyalty ... 15 3.2.5 Control variables ... 17 3.2.6. Test cases ... 17 3.3 Procedure ... 18 3.4 Analyses and Predictions ... 18 3.4.1. Analysis of reliability and correlation ... 18 3.4.2. Analysis of predictions ... 19 4. Results ... 20 4.1 Sample ... 20 4.2 Reliability and Correlations ... 20 4.2.1 Correlations for easyJet ... 21 4.2.2 Correlations for KLM ... 22 4.2.3 Comparing correlations for easyJet and KLM ... 23 4.3.1 Main effects regression analysis ... 24 4.3.2 The 2-way and 3-way interaction effects ... 25 4.3.3 Loyalty as only moderator: separate regression analyses ... 27 5. Discussion ... 33 5.1 Discussion on the results ... 33 5.1.1 Discussion on the first hypothesis: product types as moderator ... 34 5.1.2 Discussion on the second hypothesis: loyalty as moderator ... 35 5.2 Strengths and Limitations and Future Research ... 36 5.3 Contributions and Implications for practice ... 39 6. Conclusion ... 40 References ... 41 Appendix A ... 43 Appendix B ... 47 Appendix C ... 52 Appendix D ... 53
1. Introduction
Customer reviews on the Internet have become increasingly important to customers to verify initial believes and expectations of a product or service offer and are thus instrumental for the final purchase decision. These online reviews provided by customers on independent platforms are nowadays one of the most popular forms of consumer-generated content (Bickart & Schindler, 2012). Consumer platforms have altered the way individuals gather product information, and the way they make purchase decisions by increasingly substituting traditional sources of information, including evaluations provided by the company and official advertising (Li & Hitt, 2008; Chevalier & Mayzlin, 2006; Lee, Park, & Han, 2008). The influence that online customer reviews have on purchase decisions is claimed to be especially significant with regard to experience goods such as air travel, when characterization of the good before consumption is difficult (Li & Hitt, 2008; Zhang & Zhu, 2010). The results of a survey on Internet marketing by comScore (2007) confirms this and suggests that 24% of participating customers seek online reviews before paying for a service delivered offline. Due to this substantial influence of online customer reviews on consumer buyer behaviour, researchers and marketers are becoming increasingly aware of the opportunities and risks associated with such reviews (Zhang & Zhu, 2010). This awareness has led to much research being conducted across various industries and is targeted towards understanding the relationship between customer reviews and company sales. These studies claim that online customer evaluations significantly influence the number of sales (Gu, Law, & Ye, 2009), and some authors even state that word of mouth interactions such as online reviews can predict firm performance even better than traditional measures of performance such as customer satisfaction (Hu, Liu, & Zhang, 2008). Therefore, the increasing numbers of online word of mouth interactions
encountered by customers have fundamental implications for customer management activities and strategy making.
However, no research has yet been conducted on how different product types are affected by the relationship between online reviews on customer’s willingness to pay. Therefore, this research considers cost-leadership and differentiation by using the airline industry as prime example, to investigate the product type as moderator of the relation between negative customer reviews and willingness to pay. Providing insights as to what extent the negative effect of online customer reviews influences the willingness to pay for specific product types helps management to understand how many resources should be devoted to overcome these negative effects. This is essential information since a reduction in willingness to pay after negative word of mouth interactions may result in customers considering alternatives (Hu, Liu, & Zhang, 2008). Contemplating that even a marginal reduction in retention rate resulting from exit or change of patronage is exponential, the long-term revenue can be significantly affected due to negative product reviews (Andreassen & Lindestad, 1998).
Additionally, previous research has, in accordance with the disconfirmation-of-expectation paradigm, regarded willingness to provide positive word of mouth as a function of customer loyalty (Andreassen & Lindestad, 1998). This study however takes a different perspective by considering customer loyalty as a moderator of the relationship between online reviews and willingness to pay. The primary assumption in this approach is that stronger relationships between customers and companies make up for negative encounters (Priluck, 2003). Researching customer loyalty from this perspective revaluates the importance for firms with different strategies to invest in customer loyalty programmes as tool for enhanced customer retention (Borenstein, 1989).
In summary, this study investigates through an empirical study: (a) to what extent product types (differentiation versus cost-leadership) strengthen or weaken the impact of negative online customer reviews on the customer’s willingness to pay; and (b) whether this relation is moderated by different levels of customer loyalty.
2. Theoretical Framework
2.1 Negative customer reviews and willingness to pay
Several empirical studies have identified relationships between online customer reviews and firm performance across various industries. For instance, Gu, Law and Ye (2009) suggest that online customer evaluations significantly influence the number of online hotel bookings, and Hu et al. (2008) state that linking online reviews with sales often yields positive correlation between the average review score and product sales. In this study, the impact of negative customer reviews on willingness to pay is researched, which is a direct function of sales (Skiera & Wertenbroch, 2002).
Negative reviews are perceived as useful benchmarks when consumers are engaged in ruling out decision alternatives (Schindler & Bickart, 2005; Lee, Park, & Han, 2008; Maheswaran & Meyers-Levy, 1990). Additionally, negative word-of-mouth communications have been recognized to be especially valuable since negative information carries more weight and is seen as more informative than positive information (Mizerski, 1982; Charlett, Garland, & Marr, 1995). This claim is based on the prospect theory, which entails that: ‘’the experience of loss appears to be greater than the pleasure associated with gaining an amount equivalent to that which was lost because the value function is steeper for losses than for gains’’ (Lee, Park, & Han, 2008, p. 342). Applying the prospect theory to this study suggests that negative word of mouth
interactions have a stronger impact on a customer’s brand evaluation and willingness to pay, which is in line with previous studies that found that negative reviews influence company sales (Hu, Liu, & Zhang, 2008).
The extent to which this effect manifests itself depends on the strength of the initial believes of the customer and the quality of the review, resulting in two possible occurrences. First, the arguments provided by the online evaluations could be undermined since they are contradictory to the initial positive attitude of individuals about the product (Bambauer-Sachse & Mangold, 2013). This implies that the interaction with negative evaluations would result in an insignificant change of the initial attitude towards the product or service, which is in line with the disconfirmation model (Edwards & Smith, 1996). Second, when the negative review is considered highly informative, this could lead to a strong shift from the initial positive attitude towards a negative view. This is particularly the case when consumers encounter several negative reviews that contradict a consumer's initial positive product evaluation (Lee, Park, & Han, 2008).
2.2 Product types and willingness to pay
Traditional market leaders in all industries whose businesses are built on complex, high-cost models, experience increasing degrees of rivalry by companies with new, simpler ways to manage their operations and reduce costs (Jarach, 2004). These new ways of doing business provide a solution for different customer perceptions of value (Woodruff, 1997). According to Zeithalm (1988, p. 14), perceived value can be defined as follows: ‘’perceived value is the consumer's overall assessment of the utility of a product based on perceptions of what is received and what is given. Though what is received varies across consumers (i.e., some may want volume, others high quality, still
others convenience) and what is given varies (i.e., some are concerned only with money expended, others with time and effort), value represents a tradeoff of the salient give and get components’’. Based on this knowledge, companies position their offerings on the market in a continuum of cost leadership and product differentiation strategies that best fits the target consumer’s needs (Gunasekaran, Mavondo, & Yamin, 1999). The airline industry is a prime example of an industry in which the distinction between differentiation and cost leadership is clearly evident.
In the airline industry competition has rapidly shifted from rivalry between the traditional carriers (also called network-based airlines), to including Low-Cost Carriers as direct competitors, which implies a confrontation between two fundamentally different business models (Jarach, 2004). While traditional carriers are still characterized by expensive and complex services, high marketing expenses and extensive use of technology, it is the Low-Cost Carriers that are gaining significant market share by offering lower prices through reduction of the unit costs and at increasing both output and productivity (Dobruszkes, 2006). According to the European Organization for the Safety of Air Navigation (2013), the European LCCs have gained a market share of up to 26% within Europe in 2013 and these figures have continued to strongly increase at the expense of the market share of traditional airlines (Jarach, 2004).
This study seeks to investigate whether different product types are affected differently by negative reviews in terms of willingness to pay. This relationship has not been studied before and expectations on the outcome rely mainly on theories related to the evaluation of alternatives and price elasticity.
It is established that negative customer reviews might result in customers looking for alternatives, which may lead to reduced retention rates (Hu, Liu, & Zhang, 2008). According to Porter (Porter, 2008) this threat of substitution is diminished for
differentiated products as customers’ become attached to differentiating attributes, such as superior service levels and extensive customer loyalty programs. This attachment might result in a diminished effect of negative reviews on willingness to pay for differentiated products. However, stronger relationships between firms and customers have been argued to amplify customers’ unfavourable responses to negative encounters (Bhattacharya & Sen, 2003; Ganesan, Hess, & Klein, 2003). Gregoire and Fisher developed a theory proposing that: ‘’as a relationship gains in strength, a fairness violation leads to an increased sense of betrayal, which in turn drives strong relationship customers to retaliate with greater intensity’’ (2008, p. 248). In this study their theory could apply when a review from a customer who was unrightfully treated feels to the loyal customer as if the firm wronged him or her personally.
Additionally, since air travel is shifting incrementally from an experience good to a commodity in the mind of the consumer (Brons, Nijkamp, Pels, & Rietveld, 2002), the consumer demand for airline services has become more elastic, resulting in customers becoming more price conscious. Due to this elasticity and the ability of Low-Cost Carriers to compete on price leadership, LCC’s are gradually regarded as alternatives for services provided by traditional carriers (Tretheway, 2004).
The different theories from the literature are inconclusive on the effect of each product type and the extent of the impact of this effect on willingness to pay after negative WOM interactions. However, based on the existing literature the following hypothesis is proposed:
H1: Differentiated product types are less sensitive to negative customer reviews and its impact on willingness to pay than cost-leadership products.
2.3 Customer loyalty as a moderator
Customer loyalty has been defined in many ways. In this study customer loyalty is viewed as the strength of the relationship between an individual’s attitude and repeat patronage (Dick & Basu, 1994). Loyalty can materialize through positive interactions with a firm that affect consumers’ levels of trust and commitment, resulting in an enhanced believe that the value received from the company is greater than that available from other suppliers (Andreassen & Lindestad, 1998). Researching loyalty in terms of the attitude-behavior relationship allows for inferring the consequences that follow from this relationship, including resulting behaviour in terms of willingness to pay after negative word-of-mouth encounters with the firm for specific product types (Dick & Basu, 1994).
Studies have found that through these repeat purchases and a favourable attitude towards the brand, high levels of loyalty positively affect willingness to pay and, therefore also firm revenues (Andreassen & Lindestad, 1998; Hallowell, 1996). Namely, consumers with higher attitudinal loyalty will accept higher prices, which is in accordance with the theory of brand equity (Hill, 1988). This theory implies that associations with and positive behaviour towards the brand allow a company to gain greater market share or earn grater margins than it could without the brand name (Chaudhuri & Holbrook, 2001).
As the negative impact of negative word of mouth interactions on sales is confirmed in previous studies, higher degrees of loyalty are expected to reduce this effect. This assumption is based on the following findings. Loyalty is based on a long-term relationship between firm and customer after repeat purchases, which gives the company the opportunity to build up loyalty and goodwill. This has several implications. Firstly, customers that are more loyal to the firm are less likely to search for information
on alternative products (Dick & Basu, 1994), even after undesirable encounters with a brand, for instance, negative evaluations. This is due to the strong commitments that loyal people hold towards their brand, which results in an enhanced resistance to persuasion and increased likeliness to refute information that is not in line with the customer’s believes (Ahluwalia, 2000). Various cognitive mechanisms explain this phenomenon, such as the confirmation bias, and cognitive consistency (Dick & Basu, 1994; Nickerson, 1998). Additionally stronger relationships with the firm result in increased willingness of customers to overlook negative interactions due to the halo effect (Priluck, 2003). Thus, for more loyal customers search motivation is reduced, and the impact of negative word-of-mouth interactions with the firm is probably less substantial due to strong cognitive commitments to the brand.
Nevertheless, the impact of loyalty is also expected to depend on the product type as it is not researched yet what the maximum degrees of customer loyalty are for each positioning strategy. It is well known from Porters generic strategies that creating brand loyalty is the basic aim of the differentiation strategy as this results in price inelasticity, but that it does not play a significant role in cost-leadership strategies (Miller & Friesen, 1986). It is evident, however, that both strategies use tactics to increase loyalty to retain customers and overcome negative effects. One of the most prominent examples of programs to enhance loyalty in the airline industry are the frequent-flyer programs (FFPs). These programs provide reoccurring customers with gifts such as free seating or even free travel, depending on the customer status after conducting a certain amount of business with the airline. These frequent flyer programs have been effective in securing repeat purchases, and locking in customers by adding extra value (Borenstein, 1989).
As firms with different strategies all invest in loyalty programs, it would be essential to infer for which type of product loyalty enhancement is most beneficial as moderator of negative WOM interactions.
Based on the favourable effects of loyalty on customer behaviour and attitudes, the following hypothesis is proposed:
H2: Higher levels of customer loyalty weaken the negative relationship between negative customer reviews and willingness to pay, of which the degree of moderation depends on the product type
Figure 1 illustrates the conceptual model, based on the hypotheses as proposed in the study.
Figure 1: Conceptual model of the study showing the proposed interrelationship between
3. Methodology 3.1 Design
To test how the relationship between negative customer reviews and the willingness to pay for airfares is affected for different product types and varying levels of customer loyalty, a cross-sectional survey design is used consisting of two surveys. There will be one survey for each of the two airlines, KLM and easyJet, which are representatives for the product types under consideration in this study: differentiation and cost-leadership respectively. Each survey is directed at customers of these specific airlines, as they will provide insights on the effects of negative reviews on their willingness to pay for tickets, while simultaneously considering the level of loyalty for the customers of each airline.
The sample of the study will consist of as many customers as possible, aiming to gather data from customers with varying degrees of customer loyalty to increase the reliability and validity of the study. The participants will be selected through a combination of purposive sampling (including snowball sampling) and convenience sampling, as potential respondents will be approached mainly at and around Amsterdam Airport Schiphol, and through social media and email. The survey will be designed in English, as the sample is likely to be of an international nature consisting of passengers with different nationalities, travelling from and to various destinations.
3.2 Measurements
The conceptual model consists of two moderator variables, implying that there are 4 conditions in the model to be tested: the type of product, categorized as differentiated (traditional airlines) and cost-leadership (Low-Cost Carriers), and varying levels of customer loyalty: loyal and not loyal.
3.2.1 Dependent Variable Willingness To Pay
Willingness to pay will be rated by customers of KLM and easyJet by a 6-item scale, asking the customer to offer a price for a flight ticket, which should equal the highest price thay are willing to pay. Three items in this scale concern hedonic purchase intentions, while the other three items regard a utilitarian intention for purchasing the ticket. An example item looks as follows: ‘‘What is the highest fare you would pay for a ticket with KLM to Rome if you would travel for leisure’’. The answer is to be given in Euros.
The items are presented to the customer in sets of two, hedonic and utilitarian: firstly before reading a review, secondly after reading the negative customer review, and lastly after the negative expert written review.
3.2.2 Independent Variable Negative Online Reviews
Since source credibility has been recognized to play a significant role in online and offline persuasion and influencing buying behaviour (Bambauer-Sachse & Mangold, 2013), the respondents will be presented two negative airline reviews. The first will be a consumer written review derived from an independent consumer platform that is indicated as ‘’very helpful’’ by its readers. The second review will be expert written and derived from what is perceived by consumers as a credible source e.g. a quality newspaper. This information about the source of the review, as well as basic informaiton about the author is provided to the respondents of the survey. Including both types of reviews enables for testing and comparing differences in imact on willingness to pay related to perceived source scredibility by the consumer.
Significant differences between the prices customers are willing to pay before and after the reviews indicate that the negative reviews have a considerable impact on willingness to pay.
3.2.3 Moderator Variable Product Type
For research purposes, easyJet is considered representative of Low-Cost Carriers while KLM is considered representative for traditional airlines. Ticket prices derived from the official airline websites compared with the services offered reflect the different types of value propositions that the companies offer. Comparing the lowest fare prices for destinations within Europe for both airlines indicates that prices for KLM are on average four times the price of easyJet tickets.
2 surveys will be conducted, one for each airline representing different product types. The results will be compared to assess the differences in outcomes of the relationship between negative reviews and willingness to pay for each product type.
3.2.4 Moderator Variable Customer Loyalty
Customer loyalty is in multiple studies described in terms of repeat purchases, and the customer's attitudinal state of intentions to repurchase (Evanschitzky, Plassmann, Iyer, Niessing, & Meffert, 2006; Andreassen & Lindestad, 1998). Therefore, both the behavioural and attitudinal components of loyalty will be measured.
The behavioural loyalty of the customer implies consideration of customer actions and involves the measurement of past purchases of airline tickets, along with the probability of future purchases. Consequently, the survey includes items regarding the behavioral aspects of the relationship between the customer and a company, in terms of an estimation of frequency of interaction (flights per year) (Ganesan, Hess, & Klein,
2003). Additionally, the respondents will be asked to indicate whether they are part of the loyalty program of the airline (e.g. Frequent Flyer Programs), and specify their cardholder status.
According to Evanschitzky et al. (2006) attitudinal loyalty concerns the position of a brand in the mind of the customer, and involves the measurement of consumer satisfaction and perception. To provide a clear measurement of the attitude towards the company, the customer will be asked to indicate their perceived loyalty to the company with a 7-point Likert Scale, which ranges from: (1) not loyal at all to (7) very loyal. Additionally, analyses will be conducted to determine the customer’s attitude towards different attributes that are related to the price-service tradeoff (Skiera & Wertenbroch, 2002). The literature review by Lin et al. (2011) provides a list of the most important evaluation criteria for airlines derived from previous studies, including items related to the distinction between traditional airlines versus Low-Cost Carriers (2011). The criteria are presented to the participants who are asked to indicate their satisfaction with these attributes regarding the respective airline. Each criterion is measured by a 7-point Likert Scale, which ranges from: (1) extremely satisfied to (7) extremely dissatisfied. These scores give an indication of customer attitudes, where higher fit between company offerings and customer values will lead to higher degrees of loyalty (Andreassen & Lindestad, 1998).
In short, the variable for Loyalty will be computed by adding up the scores of the survey items measuring the number of flights a year, loyalty program membership status (distinguishing between the following: no cardholder, ivory, silver, gold, platinum for KLM Flying Blue; and Plus membership for easyJet), the degree of loyalty as perceived by the customer, and customer satisfaction scores.
3.2.5 Control variables
Research conducted on the relation between online reviews and airline performance suggests that it is likely that three factors have a fundamental impact on passenger demand in the airline industry: income, ticket prices, and service quality (Lin, Tseng, & Wang, 2011). As ticket prices and service quality are key factors in the distinction between product types in this model, income is included as control variable, in addition to the demographics age, gender, and occupation. Additionally, studies have found country of origin to be a relevant indication of buyer’s perceptions of quality, and indicate that nationality may reflect presumptions in favor of "home" versus "foreign" country products and services (Douglas, Johansson, & Nonaka, 1985). To control for these biases and cultural differences, nationality and country of residence are added as control variables.
3.2.6. Test cases
Finally, test cases are included in the model to assess the model’s ability to accurately reflect relationships. These test cases will be: (1) source credibility as discussed under the Independent Variable section, consisting of customer reviews versus expert reviews; and (2) hedonic versus utilitarian products.
In general, previous literature has confirmed that demand for business travel tends to be less sensitive to changes in airfare than demand for leisure travel. One explanation provided by Brons et al. (2002) is that leisure travel is generally regarded as discretionary expenditure, and that as a consequence people are willing to consider more different substitutes to air travel to save money than they would for business travel. Due to the difference in price elasticity between leisure travel and business travel, it is
expected that individuals travelling for leisure would be more receptive to perceived price-quality changes, as a result of encounters with negative reviews.
3.3 Procedure
Customers of the representative airlines will be approached personally around the Check-In rows of KLM and easyJet at Amsterdam Airport Schiphol, at public spaces such as train stations, and through social media and e-mail. Basic information will be provided regarding the nature of the study and to inform the participant on where to find additional information and the privacy disclaimer that applies to this study. The link to the survey will be distributed by handing out small cards and sending out emails with both the URL and a QR code to enable easy mobile access for the respondents. This ensures the privacy of the respondents by not requiring any additional personal information for distribution purposes. When gathering data at the airport this method ensures that airport processes will not be disturbed by keeping passengers from proceeding on their way to the gate. The surveys are to be completed individually. The data collection period will consist of four weeks; the customers are able to complete the survey at any point within this time span.
3.4 Analyses and Predictions
3.4.1. Analysis of reliability and correlation
The first step in testing for moderation is to assess the reliability of individual scales. This is measured using Cronbach’s Alpha coefficients. Next, when all scales are confirmed to be reliable, correlations between the variables are measured to identify the direction, and strength between associations. Finally, a robustness test is used to assess the model’s ability to accurately reflect relationships while including test cases. These
test cases will be: (1) source credibility: customer reviews versus expert reviews; (2) hedonic versus utilitarian products.
3.4.2. Analysis of predictions
The model yields a 3-way interaction, which will be tested through multiple regression analysis. The regression equation for this model can be defined as follows:
First, based on the results of previous studies, the relationship between negative reviews and customer willingness to pay is expected to be significant and negative of nature (Gu, Law, & Ye, 2009). This implies that when a customer reads negative reviews, this customer will be willing to pay less for the airline ticket.
Second, the strength of the relation between negative reviews and willingness to pay is expected to be dependent on the moderating effect of the type of product. Meaning that the extent to which willingness to pay is reduced after reading negative reviews depends on whether the type of product under consideration can be categorized as differentiated or as a cost-leader product. It is expected that differentiated products and cost-leaders will influence the relationship between negative customer reviews and willingness to pay in a different way and to a different extent.
Third, customer loyalty is expected to moderate the interaction between product types and negative reviews on willingness to pay. This implies that customers experiencing higher levels of loyalty to a company (resulting from more advanced relationships) would be less receptive of negative word of mouth interactions.Therefore, it is expected that higher levels of loyalty weaken the effect of negative reviews on willingness to pay for airline tickets, in which the extent of the moderating effect
depends on the type of airline (differentiated or cost-leadership). This study will indicate for what product type this interaction effect is stronger in relation to willingness to pay.
4. Results 4.1 Sample
Data is collected from a total of 369 respondents of which 125 respondents are easyJet customers, and 244 are KLM customers. Due to participant errors and partially unfinished survey responses, there are 116 (53% male), and 215 (51% male) valid responses for easyJet and KLM respectively. Participating EasyJet customers are aged 15 to 78 years (M=36, SD=15), with average income between €20,000 to€39,999 on a yearly basis. Most of the participants are employed fulltime (51%), are Dutch nationals and reside in the Netherlands (59% and 88% respectively).
Participating KLM customers are aged between 16 and 73 years old (M=34,
SD=14), with an average income between €20,000 to €39,999 on a yearly basis. Most of
the participants are fulltime employees (47%), and are Dutch nationals and residents of the Netherlands (69% and 92% respectively).
4.2 Reliability and Correlations
The variables used for each product type in the study were derived from mean scales. Their individual Cronbach’s alpha coefficients of willingness to pay and online customer and expert reviews indicate high internal consistency for both easyJet and KLM: willingness to pay (easyJet α = .829; KLM α = .781), online customer reviews (easyJet α = .796; KLM α = .795), online expert reviews (easyJet α = .785; KLM α =
.784), and loyalty (easyJet α = .797;KLM α = .816). For loyalty one item, an estimation of years one has been a customer of the company, was deleted from the scale.
Table 1 and Table 2 contain the means, standard deviations, correlations, and
Cronbach’s alpha coefficients of the variables of the data gathered from easyJet and KLM customers respectively. When testing for multicollinearity analysis shows that the correlation values between the variables are low and do not indicate multicollinearity (see Appendix D).
4.2.1 Correlations for easyJet
For easyJet, the impact of negative online customer reviews, r (113) = -.527, p= .000, and of negative online expert written reviews, r (108) = -.588, p= .000, is, according to the expectations based on existing literature, negatively related to willingness to pay. These results imply that customers of cost-leaders are willing to pay significantly less for airline tickets after encounters with negative online reviews. There is however no significant difference between the impact of negative expert reviews and the impact of customer written reviews on willingness to pay (z = 0.65, p= .258, one-tailed). Additionally, calculations (see Appendix C) based on the data as displayed in
Table 1 reveal that the negative effect of negative online reviews on willingness to pay is
not significantly different when comparing willingness to pay for the purpose of business travel with leisure travel (customer reviews: z = 1.21, p= .113, one-tailed; expert reviews: z = 0.91, p= .181, one-tailed). This result is against expectations and contradicts previous studies that found a difference in price elasticity of business versus leisure travellers.
From the data it can also be concluded that for loyalty r (102) = .114, p= .255 there exists no significant relationship to willingness to pay. Additionally there is no
significant relationship between customer loyalty and negative online customer reviews (r (102) = .132, p= .187) or expert written reviews (r (102) = .102, p=.309). The latter may imply that higher levels of loyalty to the company do not weaken the negative impact of negative customer reviews.
4.2.2 Correlations for KLM
Similar to the results for easyJet, data of KLM customers illuminates that negative online customer reviews, r (179) = -.584, p= .000, and expert reviews, r (170) = -.595, p= .000, are, according to the expectations based on the literature, negatively related to willingness to pay. This implies that customers of companies offering differentiated products like KLM are willing to pay significantly less for airline tickets after encounters with online reviews.
Consistent with the results of easyJet, there is no evidence for a significant difference between the impact of negative expert reviews and the impact of customer written reviews on willingness to pay (z = 0.16, p= .436, one-tailed). Additionally, the data derived from KLM customers suggests that the impact of negative online reviews on willingness to pay in general is not different when one purchases tickets for business purposes compared to tickets for leisure (customer reviews: z = 1.21, p= .113, one-tailed; expert reviews: z = 0.85, p= .198, one-tailed), which is contradictory to the expectations.
Unlike with easyJet, the KLM data points out that for loyalty r (150) = .223, p= .006 there exists a significant positive relationship to willingness to pay, and a significant negative relationship with negative online expert written reviews r (150) = -.199, p= .015. This might suggest that a higher level of loyalty does increase initial willingness to pay, and does moderate the negative impact of online reviews but only for expert written reviews.
4.2.3 Comparing correlations for easyJet and KLM
When comparing the correlation coefficients of data from easyJet and KLM the
z-values suggest that there is no significant difference in the effect of negative customer
reviews (z = 0.68, p= .248, one-tailed) and negative expert written reviews (z = 0.09, p= .464, one-tailed) on willingness to pay. For loyalty however, there exists a significant difference in correlation when comparing the moderating effect of loyalty on online customer reviews (z = 2.12, p= .017, one-tailed) and online expert reviews (z = 2.34, p= .010, one-tailed) for both companies. This result, in combination with the correlation coefficients indicating no significant relationship between loyalty and negative reviews for easyJet, might suggests that there is a difference in the effect of varying levels of loyalty for both product types. Since there is no significant difference in the effect of negative reviews on willingness to pay between the companies, this consequently may result in a lower net effect of negative reviews for differentiated products.
Table 1: easyJet descriptives and correlations (Cronbach’s alphas on diagonal)
M SD 1 2 3 4
1. Willingness to Pay Total
1.a WTP for leisure 1.b WTP for business 337.05 135.12 201.93 212.70 81.61 140.79 (.829)
2. Online Customer Reviews (OCR)
2.a Effect of OCR on leisure 2.b Effect of OCR on business
-76.23 -27.61 -48.62 143.87 44.92 105.48 -.527** -.416** -.541** (.796)
3. Online Expert Reviews (OER)
3.a Effect of OER on leisure 3.b Effect of OER on business
-80.12 -28.19 -51.94 146.45 46.89 107.36 -.588** -.496** -.585** .881** .753** .873** (.785) 4. Customer Loyalty 70.21 14.69 .114 .132 .102 (.797)
Table 2: KLM descriptives and correlations (Cronbach’s alphas on diagonal)
M SD 1 2 3 4
1. Willingness to Pay Total
1.a WTP for leisure 1.b WTP for business 440.10 180.09 259.87 245.20 100.57 163.33 (.781)
2. Online Customer Reviews (OCR)
2.a Effect of OCR on leisure 2.b Effect of OCR on business
-103.81 -42.05 -61.76 205.41 91.11 125.06 -.584** -.549** -.559** (.795)
3. Online Expert Reviews (OER)
3.a Effect of OER on leisure 3.b Effect of OER on business
-110.00 -43.92 -65.42 191.06 84.54 120.38 -.595** -.518** -.583** .781** .688** .767** (.784) 4. Customer Loyalty 81.72 14.38 .223** -.142 -.199* (.816)
Note: N=214(Willingness to Pay), N=187(Online Customer Reviews), N= 170 (Online Expert Reviews), N= 150 (Customer Loyalty).*p<.05. **p<.01.
4.3 Results
4.3.1 Main effects regression analysis
First, the outliers are detected and excluded from the data set. This is done by identifying cases that are outliers according to at least 2 of the following indices: Mahalanobis’ distance (cut-off point 11.35), Cook’s distance (cut-off point 0.012), and the Centered Leverage Value (cut-off point 0.024). Consequently 10 cases were excluded from the data set.
Next a two-step hierarchical linear regression is used to determine the interactions between online customer reviews, product type and customer loyalty to willingness to pay after controlling for demographic variables. Model 1 measures the effects of control variables (age, gender, employment status, income, country of citizenship and of residence) on willingness to pay and Model 2 includes the main
In the regression models, age is a negative significant predictor of willingness to pay, i.e. younger customers are willing to pay more for airline tickets or are less price conscious. Additionally, income is a positive predictor of willingness to pay, suggesting that an increase in income significantly increases willingness to pay for airline tickets, which is in line with results of existing literature. The regression shows no significant effect of the other control variables included in the model. The control variables accounted for about 10% of the variance in the amount of money customers are willing to pay for tickets.
Online negative customer reviews have, as expected, a negative effect on willingness to pay after controlling for demographic variables. Both product type and loyalty are positively related to willingness to pay. This implies that people are willing to pay more for tickets offered by KLM than for easyJet, and that people that are more loyal to a company are initially willing to pay more. The main variables account for 28% of the variance in the amount customers are willing to pay for airline tickets over and above the control variables.
4.3.2 The 2-way and 3-way interaction effects
After establishing the main interactions, the 2-way interactions between: (1) customer reviews and product types, (2) customer reviews and loyalty, (3) and product types and loyalty are computed, together with the full 3-way interaction of the moderated moderator model. The results of the regression and the interactions between variables are displayed in table 4 and Figure 3 (F (13, 195) = 10.188, p = .000, R2 = .400). From the results it becomes evident that the effect of negative customer reviews in the model is significantly negative (b = -.63, t(195) = -4.96, p = .000), which confirms the prior linear regression results. Additionally, loyalty has a significant positive effect in the model (b =
2.26, t(195) = 2.27, p = .024), while product type has no significant effect (b = 47.38,
t(195) = 1.66, p = .098). Furthermore, there is no support for significant 2-way or 3-way
interaction effects between the variables. This implies that, while there are significant effects of each variable on willingness to pay, there does not necessarily exist any interrelatedness between the variables on the effect on willingness to pay. Consequently, from the data derived from the correlation analysis, and the regression analyses it becomes evident that there is no statistical support for H1 stating that differentiated
product types are less sensitive to negative customer reviews and its impact on willingness to pay than cost-leadership products. Additionally, since there is no significant interaction effect, there is no support for H2 that customer loyalty has a different impact on willingness to pay after negative customer reviews for different product types.
Figure 3: Statistical diagram 3-way interaction model
4.3.3 Loyalty as only moderator: separate regression analyses
Since the effect of loyalty is significant but no interaction exists, regressions are performed for each product type separately, introducing loyalty as the direct moderator of the relationship of negative customer reviews on willingness to pay (see Figure 4). This way the effect of different levels on loyalty as moderator can be reviewed individually for both product types after which the results are compared.
Figure 4: Conceptual model for separate regression for easyJet and KLM with Loyalty
as moderating variable
The results of these separate regressions yield that the regression coefficient of loyalty is not significant for either easyJet (b = 1.95, t(88) = 2.75, p = .181) or KLM (b = 3.14, t(118) = 1.37, p = .173), while for easyJet there exists an interaction effect between loyalty and online customer reviews (b = -.041, t(88) = -2.94, p = .004). When analysing the conditional effects of negative customer reviews on willingness to pay for different levels of loyalty, it becomes evident that the effect of negative customer reviews on WTP is significantly stronger for averagely loyal and loyal customers of easyJet (see
than customers who are not loyal. This significant difference between levels of loyalty is not found for KLM. The results are represented visually in Figure 5.
Despite of these results, when comparing the regression coefficients of negative online customer reviews for three levels of loyalty for both product types it is confirmed that the difference of the effect of loyalty for easyJet and KLM is not significant (see
Appendix C for calculated differences). This confirms the result of the 3-way interaction
analysis; the models provide no support for H2: the effect of loyalty on the negative effect of negative customer reviews is not different across the product types.
Table 3: Linear regression of Online Customer Reviews, Product Type, and Loyalty on Willingness To Pay. Control variables included.
Model 1 Model 2
Dependent variable Willingness To Pay Willingness To Pay
Coefficient SE Beta Coefficient SE Beta
Constant 267.75* 112.791 -13.327 117.716 Control variables Age -3.988** 1.346 -.247 -3.248** 1.153 -.227 Gender 41.302 28.203 .091 21.408 24.960 .052 Employment status 6.191 7.814 .059 .704 7.126 .007 Income 40.566** 8.143 .481 32.069** 6.976 .440 Country of citizenship .478 .778 .039 .443 .674 .042 Country of residence .031 1.426 .001 .892 1.170 .049
Online Customer Reviews -5.97** .087 -.400
Product Type 58.262* 25.495 .140
Loyalty 1.714* .868 .123
R2 .102 .383
Note: F(6, 262) = 4.984** for Model 1; F(9, 199) = 13.699**for Model 2. * p < .05; ** p < .01
Table 4: 3-way interaction of: Online Customer Reviews (X), Willingness To Pay (Y), Product Type (M), Loyalty (W) and Control Variables
3-way interaction Model
Dependent variable Willingness To Pay
Coefficient SE t p
Constant 280.29** 94.978 2.951 .004
Product Type 47.384 28.484 1.664 .098
Online Customer Review -.626** .126 -4.961 .000
Interaction 1 - XM .410 .227 1.810 .072 Loyalty 2.261 .996 2.272 .024 Interaction 2 - XW -.017 .010 -1.667 .097 Interaction 3 - MW -.458 1.975 -.232 .817 Interaction 4 - XMW .015 .019 .818 .414 Control variables Age -3.332** 1.175 -2.837 .005 Gender 17.927 25.888 .693 .490 Employment status .231 7.222 .032 .975 Income 34.275** 9.000 3.808 .000 Country of citizenship .404 .697 .580 .563 Country of residence .558 1.273 .438 .662 R2 .400 Note: F(7, 234) = 14.45**. p < .05; ** p < .01
Table 5: Conditional effect of negative Online Customer Reviews on Willingness To Pay at values of the moderator (Customer Loyalty) Regression analysis: single moderator
Values of loyalty easyJet Effect SE t p
Low .014 .326 .041 .967 Average -.506* .215 -2.35 .021 High -1.025** .221 -4.633 .000 Values of loyalty KLM Low -.171 .343 -.499 .619 Average -.574** .181 -3.166 .002 High -.976* .451 -2.165 .033 Note: FeasyJet(9, 78) = 7.54**; FKLM(9, 108) = 5.59**. p < .05; ** p < .01
Figure 5: Impact of OCR on WTP with low, average, and high levels of loyalty
5. Discussion
5.1 Discussion on the results
The main goal of this study is to examine whether and to what extent different types of products are affected by online customer reviews. Specifically, this research assesses whether differentiated product types are less sensitive to negative customer reviews and its impact on willingness to pay than cost-leadership products. Additionally, it is researched whether higher levels of customer loyalty to brands of either product type weaken the negative relationship between negative customer reviews and willingness to pay.
Each of the main interactions between the variables in the model and willingness to pay is confirmed according to expectations in the direction as established in previous studies (Hu, Liu, & Zhang, 2008; Dick & Basu, 1994; Hill, 1988): online negative customer reviews have a negative impact on willingness to pay, while people are willing to pay more for tickets issued by KLM than they would pay for easyJet, and customers that score high on loyalty are initially willing to pay more. Control variables are included and reveal that age and income are significant predictors of willingness to pay, which is in line with the results of Lin et al. (2011), while other control variables such as gender, employment status and cultural differences are not. The data showed no significant difference in willingness to pay between the effect of negative customer written reviews and expert written reviews. Nor was there a difference between willingness to pay for business versus leisure travel throughout the study. The latter contradicts expectations based on literature by Brons et al. (2002) on differences in price-elasticity and demand for business and leisure travel. However, this study emphasizes short distance flights (Amsterdam to Rome as indicated in the survey) for which alternative transport modes
may provide sufficiently similar qualities to be regarded as substitute modes for both business as well as leisure travel, which may explain the lack of difference between these test cases.
5.1.1 Discussion on the first hypothesis: product types as moderator
Of the two proposed hypotheses, the first hypothesis predicting that differentiated products are less sensitive to negative customer reviews and its impact on willingness to pay than cost-leader products is not supported by the data; no significant difference is found between the correlation values or the regression coefficients of negative customer reviews to willingness to pay for easyJet and KLM.This suggests that when (potential) customers encounter negative online customer reviews of the offered product, the reduction of the customer’s willingness to pay for differentiators is comparable to the effect experienced by cost-leaders. This result contradicts the theory of Porter (2008), claiming that the threat of substitution for differentiators is lower than for cost-leaders, and that therefore company sales of differentiators should be less affected than cost-leader sales. An explanation for this contradiction to the literature might be that even tough the strategy of both companies under investigation is close to being the exact opposite: easyJet reduces the sacrifice (offers low prices) and KLM adds value (extra quality services), from the survey data it becomes evident that customers from both airlines value the same aspects in airline travel. The aspects that were rated as most important for customers of both airlines are to a large extent independent of the generic strategy and are mainly points of parity in the aviation industry (Lin, Tseng, & Wang, 2011). The five aspects rated as most important for KLM and easyJet alike are: rational ticket prices, on-time flights, arrangement of flight times, comfortable seats, and accuracy of operations. In addition to the same aspects being valued most among
customers, the same reviews were presented to easyJet and KLM customers. Therefore when the same negative statements about the same topics are presented to people who generally value the same things, this might result in the same impact on willingness to pay, regardless of the airline’s generic strategy.
5.1.2 Discussion on the second hypothesis: loyalty as moderator
The second hypothesis proposed that higher levels of customer loyalty weaken the negative relationship between negative customer reviews and willingness to pay, of which the degree of moderation depends on the product type. This hypothesis was based largely on claims from existing literature, stating that inter alia the strong commitments loyal people hold towards their brand result in customers being more likely to refute information that is not in line with the customer’s believes (Ahluwalia, 2000).
This hypothesis was however not supported by the data. Instead a negative effect of loyalty on this relation was found for higher levels of loyalty, implying that loyalty enhances the negative effect of negative customer reviews. This negative effect is found for both easyJet and KLM, but is only significant for easyJet. Consequently, for cost-leaders average and higher levels of loyalty enhance the negative effect of negative online customer reviews on willingness to pay.
These results indicate that the direction of the effect of loyalty is contradictory to the positive moderating effect that was initially predicted. This implies that the outcome of this study challenges existing literature, stating that people are more receptive for word-of-mouth interactions that are in line with their own believes and attitudes (Edwards & Smith, 1996). There are a few explanations for these contradicting results. According to Anderson (1998), overall extremely satisfied customers more likely initiate and respond to word-of-mouth interactions. In addition to this increased likeliness of
(active) interaction with word-of-mouth, Fisher & Grégoire (2008, p. 248) found that as a relationship gains in strength and customers are more loyal, ‘’a violation of the fairness norm has a stronger effect on the sense of betrayal experienced by the customers’’. These findings indicate that more loyal customers are more likely to interact with negative reviews, and are more likely to feel ‘betrayed’ by the company resulting in lower willingness to pay.
Furthermore, loyalty programs such as KLM’s ‘Flying Blue’ program induce high switching costs: e.g. customers obtain rights and rewards when reaching a higher membership status which can only be maintained by flying with KLM on a regular basis. The presence of these switching costs could imply that some customers seem loyal according to their membership status and the number of flights a year, but are in fact rather dissatisfied and do not defect only because of these switching costs (Lee, Feick, & Lee, 2001). So even though the loyalty variable as computed in this study does not show this dissatisfaction (e.g. different types of loyalty such as behavioral loyalty and attitudinal loyalty and satisfaction are not distinguished but added together), the willingness to pay after encountering negative reviews might be much lower than the initial amount as a result of this dissatisfaction. This contradiction is expected to be present mainly for KLM as the switching costs of its loyalty program is higher than for easyJet, which might explain the non-significant moderation effect of loyalty for the airline.
5.2 Strengths and Limitations and Future Research
In the following paragraphs strengths and weaknesses of this study on which suggestions for future research are based will be elaborated.
First, this study has revealed insights into a previously unknown relationship between the impact of negative reviews on willingness to pay when distinguishing between product types (cost-leadership and differentiators). Additionally, loyalty is included as a separate variable as it was expected that different levels of loyalty directly influence the effect of negative reviews, rather than word-of-mouth communication being considered solely as a measure of loyalty as is the case in existing literature (Hu, Liu, & Zhang, 2008). The revision of the role of customer loyalty has led to new understandings of the role of loyalty and its connection to word-of-mouth communication. However, the study relies on cross-sectional data and does, therefore, not include changes in degrees of customer loyalty over a longer period of time. For this reason, it is not possible to indicate how large the effect of loyalty is, and what the effect is comparing increasing versus decreasing loyalty levels as could be measured at the individual level. Therefore, conducting research on the model and hypotheses as proposed in this study tested in a longitudinal design would be a suggestion for future research and is likely to yield specific answers on the strength of the variable. For such research to be successfully conducted, a reliable foundation is needed, which this study provides by establishing relations and their direction within a general model based on assumptions from existing literature.
Second, a weakness of this study design may be that the data is gathered from a purely hypothetical purchase setting. It would be valuable to gather data from customers at the actual moments of purchase to compare the data of this study with. Gathering such data is a complicated matter, however, since customers may be reluctant to provide a company with information on the maximum price they would be willing to pay for e.g. airline tickets. Especially, since an increasing number of consumers has become aware of pricing techniques such as dynamic pricing, which enables the pricing of products at a
level determined to match particular customer's perceived ability to pay and thus capture a larger share of the consumer surplus (Garbarino & Lee, 2003). This may result in data that is skewed towards lower prices than customers would actually be willing to pay and, therefore, biased data. Additionally, actual purchase settings might be more beneficial when investigating search goods, in which the price/quality ratio can be evaluated rather easily, compared to experience goods (Murray, 1991). Thus, the drawbacks of a hypothetical purchase setting for this study are to certain extent outweighed by the benefits.
Third, the results of the study are based on data gathered from customers of airlines that represent the different product types. Airline tickets, the main product offered by airline companies, are experience goods and evaluation of such goods is, therefore, influenced by the perceived quality and the value of the service (Andreassen & Lindestad, 1998). It is expected that the external validity of this study is high for other experience goods and services throughout different industries. However, it is not likely that the results are directly generalizable to include organizations that offer products where online reviews do not play an equally significant role in the customer decision process. Examples are convenience goods, which are purchased with minimum shopping efforts. However, research shows that customer’s willingness to exert minimum shopping effort for convenience goods can also lead to growing quantities of online purchases, as is the case with e.g. online grocery shopping (Li, Kuo, & Rusell, 1999). The shift from traditional physical stores and shopping environments to online retailers offering convenience goods increases the likelihood for people to encounter online product reviews. Therefore, researching the effect of online customer reviews on convenience goods and its effect on willingness to pay would be a suggestion for future research.
5.3 Contributions and Implications for practice
In addition to confirming the results of previous studies on the subject of online customer reviews, the study expands the existing literature by examining the roles of product types and loyalty in relation to the effect of negative online customer reviews in a unique way. Since the predictions tested in this study were never before investigated the results of this work provide a foundation for future research.
While the analyses yield results that did not confirm the hypotheses as proposed in this study, the findings have interesting implications for practice. The results suggest that: (1) both cost-leaders and differentiators should invest the same amount of resources to handle complaints of customers and deal with negative online reviews since the impact on willingness to pay was equal in size for both companies; and (2) the airlines might want to reconsider their investments in loyalty programs. As the results point out, initial willingness to pay for tickets is positively influenced by customer loyalty. However, once a customer has encountered negative reviews, higher levels of loyalty do impact willingness to pay in a negative manner. Businesses should therefore research the likeliness that customers search for or encounter these negative online reviews to assess whether it is worth investing time and money in loyalty programs. This is especially true for cost-leaders like easyJet where the initial increase in WTP due to loyalty is only
€1.67 (non significant) and the decrease in WTP after negative reviews is €1.03 for loyal
customers (significant). For KLM loyalty increases initial WTP with €3.63 (significant) and higher levels of loyalty decreases WTP with €0.98 (non significant) after encounters with negative customer reviews. These figures suggest that loyalty programs are, in relation to the effect of negative reviews, more lucrative for companies offering differentiated products than for cost-leaders. It is then up to the firm to review what
customer service measures as response to negative reviews are already in place, and to look for long-term trends in the industry and the increasing impact of WOM on the Internet to adjust investments in loyalty programs accordingly.
6. Conclusion
Online customer reviews have an extensive influence on consumer buyer behaviour throughout all industries. Since cost-leaders and differentiators characterize most markets, offering different products based on contradictory value proposition while serving different customers, it is plausible to believe that negative word of mouth communications such as negative online reviews have different consequences for both firm types in terms of company sales. Interestingly, the results of this study indicate that the sales of KLM as representative of differentiators and easyJet for cost-leaders, measured in willingness to pay, suffer equally by the interactions of customers with negative reviews. Even more surprisingly, the negative effect of negative reviews is found to be larger for customers that score higher on loyalty (measured by combining satisfaction, usage, and objective and subjective measures of loyalty status). This effect is found significant for cost-leaders, showing that while loyalty initially increases willingness to pay, the opposite is true when loyal customers encounter negative reviews before purchasing the product or service. Since the reach of online reviews becomes more extensive, the question remains how lucrative loyalty programs are for shopping goods and specialty goods when taking into account the newly discovered negative effects loyalty might enhance, in combination with the high investment costs, versus the financial benefits.
References
1. Ahluwalia, R. (2000). Examination of psychological processes underlying resistance to persuasion. Journal of Consumer Research , 27 (2), 217-232.
2. Andreassen, T. W., & Lindestad, B. (1998). Customer loyalty and complex services: The impact of corporate image on quality, customer satisfaction and loyalty for customers with varying degrees of service expertise. International Journal of service Industry management , 9 (1), 7-23. 3. Bambauer-Sachse, S., & Mangold, S. (2013). Do consumers still believe what is said in online
product reviews? A persuasion knowledge approach. Journal of Retailing and Consumer Services
, 20 (4), 373-381.
4. Bhattacharya, C. B., & Sen, S. (2003). Consumer-company identification: A framework for understanding consumers’ relationships with companies. Journal of marketing , 67 (2), 76-88. 5. Bickart , B., & Schindler, R. M. (2012). Perceived helpfulness of online consumer reviews: the
role of message content and style. Journal of Consumer Behaviour , 11 (3), 234-243. 6. Borenstein, S. (1989). Hubs and high fares: dominance and market power in the US airline
industry. The RAND Journal of Economics , 344-365.
7. Brons, M., Nijkamp, P., Pels, E., & Rietveld, P. (2002). Price elasticities of demand for passenger air travel: a meta-analysis. Journal of Air Transport Management , 8 (3), 165-175.
8. Charlett, D., Garland, R., & Marr, N. (1995). How damaging is negative word of mouth?
Marketing Bulletin , 6, 42–50.
9. Chaudhuri, A., & Holbrook, M. B. (2001). The chain of effects from brand trust and brand affect to brand performance: the role of brand loyalty. Journal of marketing , 65 (2), 81-93.
10. Chevalier, J. A., & Mayzlin, D. (2006). The effect of word of mouth on sales: Online book reviews. Journal of marketing research , 43 (3), 345-354.
11. Dick, A. S., & Basu, K. (1994). Customer loyalty: toward an integrated conceptual framework.
Journal of the academy of marketing science , 22 (2), 99-113.
12. Dobruszkes, F. (2006). An analysis of European low-cost airlines and their networks. 14 (4), 249-264.
13. Douglas, S. P., Johansson, J. K., & Nonaka, I. (1985). Assessing the impact of country of origin on product evaluations: a new methodological perspective. Journal of Marketing Research , 388-396.
14. Edwards, K., & Smith, E. E. (1996). A disconfirmation bias in the evaluation of arguments.
Journal of Personality and Social Psychology , 71 (1), 5.
15. European Organisation for the Safety of Air Navigation. (2013). Market Segments in European
Air Travel 2013. European Commission.
16. Evanschitzky, H., Plassmann, H., Iyer, G. R., Niessing, J., & Meffert, H. (2006). The relative strength of affective commitment in securing loyalty in service relationships. Journal of Business
Research , 59 (12), 1207-1213.
17. Fisher, R. J., & Grégoire, Y. (2008). Customer betrayal and retaliation: when your best customers become your worst enemies. Journal of the Academy of Marketing Science , 36 (2), 247-261. 18. Ganesan, S., Hess, R. L., & Klein, N. M. (2003). Service failure and recovery: The impact of
relationship factors on customer satisfaction. Journal of the Academy of Marketing Science , 31, 127–145.
19. Garbarino, E., & Lee, O. F. (2003). Dynamic pricing in internet retail: effects on consumer trust.
Psychology & Marketing , 20 (6), 495-513.
20. Gu, B., Law, R., & Ye, Q. (2009). The impact of online user reviews on hotel room sales.
International Journal of Hospitality Management , 28 (1), 180-182.
21. Gunasekaran, A., Mavondo, F. T., & Yamin, S. (1999). Relationship between generic strategies, competitive advantage and organizational performance: an empirical analysis. Technovation , 19 (8), 507-518.
22. Hallowell, R. (1996). The relationships of customer satisfaction, customer loyalty, and
profitability: an empirical study. International journal of service industry management , 7 (4), 27-42.
23. Hill, C. W. (1988). Differentiation versus low cost or differentiation and low cost: A contingency framework. Academy of Management Review , 13 (3), 401-412.
24. Hu, N., Liu, L., & Zhang, J. J. (2008). Do online reviews affect product sales? The role of reviewer characteristics and temporal effects. Information Technology and Management , 9 (3), 201-214.
