A Research of Customer’s Choice of Reward Program to ...



A Research of Customer’s Choice of Reward Program to Online Travel IntermediatesBy Xuan QiuSchool of EconomicsErasmus UniversityRotterdam, NetherlandsAbstractAs the competition of the customer market and the cost of retaining customers increasing having a loyal group of customer becomes one of the import strategies for a business to be successful. The customer reward programs emerge under the circumstance. As the effective tool of cultivating, developing and retention loyalty customers, customer reward programs are increasingly implemented in marketing, and related issues have also aroused the concern of academics. Though researchers have studied the reward programs from different aspects, most of the academic literatures focus on the impact on consumer buying patterns or the corporate profits from the program, there still is a relative lack of published literature of putting the affecting elements of reward program take into account.The design of a reward program has to consider many factors such as the unique characteristics of an industry, and the cost to establish the program etc. The structure of each customer reward program is also different. In China, customer reward programs are often copied from their competitors or directly brought back from the other countries. Such programs can hardly reach the expectation. Therefore, we have to dig out the special affecting factor for customers reward program based on the circumstances of the special market.This article reviews a lot of related costumer reward programs, collects the data from various existing programs and interviews. It then summarizes the affecting factors and establishes a model of the factors for the costumer reward programs, and verifies the model by using a survey of the Ctrip’s reward programs. The result has shown that: The reward value, the reward type, convenience, possibility and the customer fit are the 5 fundamental factors of designing a reward program. These factors affect the willingness of the customers ‘participation to the program. The effectiveness of these five factors is different. Among them, the convenience of the program plays the most important role. Based on the empirical research, the paper gives out some conclusions and suggestions which are instructive for enterprises to design and implement a reward program.Keywords：Reward Program, Online Travel Agency, Affecting FactorsTable of ContentsIntroduction 1Background1Content and Significance 3Research Content3Research Significance 3Research Framework 4Innovative Points 5Literature Review 6The connotation of the reward program6The importance of the reward program 7For customer retention7For customer loyalty 8The study on reward program’s elements 10The limitation of the existing research 14Model and assumptions 15Research hypotheses derived 15Existing literatures summary 15Interviews 16Existing reward programs summary 17The reward program influencing factors 19The mediating role of customer perceptive value assumptions 19Modeling 20Hypotheses 21Empirical research 23The research object selection 23Research variables and scale design 23Independent variables 23Intermediate variable 25Dependent variable 25Demographic variables 25Questionnaire design 25Questionnaire measures 25Questionnaire structure 26Preliminary test 26Questionnaire revise 27Data analysis 28Sample description 28Research method and research object 28Sample size 28Sample description 28Reliability analysis 30Validity analysis 33Correlation analysis 35Structural Equation Model 36Structural Equation Model analysis 36Model-fitting testing 37Hypothesis testing 39The revised model 40Conclusions 40Conclusions and limitations 44Conclusions 44Suggestions 44Research contributions 45Research limitations 46Further study 47List of TablesTable 2.1 Reward types 11Table 3.1 Literature summary 15Table 3.2 Existing reward program summary 18Table 4.1 Variables 23Table 4.2 Independent variables measurement 24Table 4.3 Intermediate variable measurement 25Table 4.4 Dependent variable measurement 25Table 4.5 Removed measure items 27Table 4.6 Changed measure items 27Table 5.1 Descriptive statistics 28Table 5.2 Means and Std. Deviations 31Table 5.3 Cronbach’s Alpha standards 32Table 5.4 Cronbach’s Alpha coefficient 32Table 5.5 KMO - Bartlett test 33Table 5.6 Varimax rotation 34Table 5.7 Extraction Sums of Squared Loadings 35Table 5.8 Correlation 35Table 5.9 Model-fitting 37Table 5.10 Hypothesis testing 39List of FiguresFigure 2.1 Reward classifications 13Figure 3.1 Research model21Figure 5.1 Structural Equation Model 37Figure 5.2 Revised model40§1. IntroductionThe purpose of this research is to examine the critical factors that influence consumers’ choice towards reward programs.This chapter is a general introduction of the research contents. This chapter first states the research background, which is followed by the research content, significance, plan and framework of this paper. The chapter ends with illustration of the innovative points of this paper.§1.1 BackgroundAs marketing shifts from traditional marketing to relationship marketing (Sheth, 1994), Customer-centric marketing strategy is widely used by more and more enterprises in order to cultivate loyal customers and expects to make more profits. As a powerful tool to cultivate, develop and keep loyal customers, customer reward program is widely used in the field of marketing.The reward program is widely used in airlines, telecom, finance, retail and tourism. Data shows that corporate profits will decrease by 25% if the customer loyalty decreases by 5%, and corporate profits might increase by 85% if the customer loyalty increases by 5% because that 60% ??of corporate customers come from loyal customers’ recommendations. Thus, according to Kim (2001), the implementation of customer reward programs is an important guarantee to establish business relationships with customers and maintain long-term consumers. At the same time they defined customer reward programs as those that offer incentives to consumers on the basis of cumulative purchases of a given product or service from a firm.Studies have shown that in the seven biggest business areas in the United States, more than 50% of the top ten companies are using reward programs, while the proportion in UK is also high. On the other hand, the reward programs are very popular among customers too. According to data from Catuity, a developer of smart card-based loyalty software for retailers, about 70 % of U.S. households participated in various forms of reward programs, while 83% of households regularly use the rewards program (MacSmith, 2002). Additional data shows that 53% of the European grocery buyers participated in reward programs, and the corresponding figure in the field of fashion is 21% (Cigliano, 2000)In China, an increasing number of companies are using reward programs as a promotional tool to develop customer loyalty. In retailing, Parkson also offers a "loyalty card " to its customers, which gives reward to members who spent more than 4000yuan($640) a year; in financial industry, the "Dragon Card " (credit card, debit card) issued by Chinese construction Bank Shanghai Branch, introduced credit incentives in 1997, after which almost all commercial banks have introduced a rewards program based on credit; in aviation, Air China introduced "Air Salon credit card" while China Southern Airlines launched "Sky Pearl Club", and other airlines also set up Member Clubs to attract customers.In tourism, Ctrip (ctrip.tom, China's first tourism company who is listed on NASDAQ) set up a "loyalty program" for its members, which allows their members to accumulate points through a variety of ways, and redeems the attractive bonus gifts according to reward points.With the growth of both number and the information literacy of Internet users, as well as the rapid development of Internet, e-commerce based online travel service websites have also developed rapidly. According to China e-Business Research Centre and iResearch Research Center, the number of online travel booking has been growing steadily. IResearch statistics shows that more than CNY131.39 billion was spent on online travel in 2011, increased 38.5% from a year earlier. In 2012, more than CNY172.97 billion was spent on online travel, increased 31.6% from a year earlier. Ctrip, Elong and LY were the top three online travel OTA (online travel agency) (source: iResearch, 2012). Such data indicates that China's online travel market develops extremely fast at present, and travel website undoubtedly plays a very important role.As enterprise's concerns shift from products to customers, analysis of the influence factors for customer to participate in reward program and the structure of the conceptual model for customer to participate in reward program has important theoretical significance and practical value for enterprise to drive customers to participate in the rewards program more effectively, especially in promoting customer to participate in the rewards and strengthen customer relationship.§1.2 Content and Significance§1.2.1 Research ContentCustomers’ participation in reward programs is affected by many factors, and it is impossible for enterprises to invest human, material and financial resources in all of the factors. To ensure the rational use of the limited resources, the key point is to determine the key factors. This paper adopts Literature Research method on structural factors of reward program to analyze the structural factors of reward program, and summarizes the design method of each structural element. Then the paper selects several critical factors that influence consumers’ choice towards reward programs and determined the measurements of each factor. In addition, through the empirical analysis of Crip’s reward program, the paper establishes the concept model based on the customers’ participation of reward program, and proposes improvements and extensions of reward program.§1.2.2 Research SignificanceBuilding conceptual model of customers’ participation in reward programs can help us further understand that there are many factors that affect customers’ participation in a reward program, and through theoretically analyzing the complex factors, we can better understand the real reasons for customers’ choosing a reward program.On the one hand, the results of this thesis is expected to help enterprises to recognize the affect factors of a reward program, therefore making their reward program more effective; For enterprises, the affecting factor discussed in this paper can offer them a reference in designing a reward program. On the other hand, because of the differences between industries, it is necessary to establish an appropriate reward program conceptual model based on online travel website.§1.3 Research FrameworkThe content of this paper is divided into six chapters, the summaries are as follows:Chapter 1: Introduction. This chapter discusses the research background, contents as well as framework, and describes the innovate points of this research.Chapter 2: Literature review. By reading about the research concerning the reward program, this paper further put forward the affect factors of customers’ participation in reward programs and discusses the drawbacks of the existing research.Chapter 3: Model and assumptions. This chapter put forward the relevant elements of the reward program, and builds a conceptual model as well as proposed assumptions, followed by the questionnaire design.Chapter 4: Empirical research. Statistical analysis software is used to analyze the data and test the model, after which the model would be evaluated and improved.Chapter 5: Data analysis. Use data reliability analysis, descriptive statistical analysis, correlation analysis and regression analysis to test hypothesis and obtain a result.Chapter 6: Conclusions and limitations. This chapter gives the conclusions of the paper, summarizes the previous studies, and establishes the conceptual model and assumptions, proposes limitations and future research direction.This study collects its problems through empirical observation, and then determines the research subject with questions. After literature study, hypothesis is derived and the conceptual model is constructed. The next step is to prepare the questionnaire and do the preliminary experiment. On the basis of preliminary experiment, questionnaire is revised and issued on a large scale. After the research data is obtained, statistical analysis is done and the research hypothesis is tested. Finally we obtain the conclusion. The concrete research methods are:Consumer interviewThe purpose of consumer interviews is to understand consumers’ views on research topic. The interview also plays an important role in determining the research hypothesis and constructing the model.Questionnaire?methodThis paper takes online travel websites as research objects, and adopts?questionnaire to collect relevant data to verify the proposed theoretical model and put forward improvement measures on this basis.Statistical analysisAfter the reliability and validity analysis, the author uses descriptive analysis, correlation analysis, regression analysis (Refer to appendix A) and SEM method to examine the research hypothesis.§1.4 Innovative PointsThe main innovative points of the paper are:1) This paper takes online travel websites as research objects, making the study more representative and specific. 2) A new theoretical basis for online travel websites is provided to initiate a reward program.§2. Literature ReviewThe goal of this chapter is to provide support and arguments for developing the research question and build a theoretical framework. Our goal of this research is to understand the real reasons for customers’ choosing a reward program through theoretically analyzing the complex factors. There are a few researches done on different factors that influence customers’ participation in reward programs. Therefore, it is essential to examine the results of a research stream that is relevant to this subject.§2.1 The connotation of the reward programWith the advent of the age of product homogeneity and the development of e-commerce, more and more enterprises realize that customer loyalty is the source of competitive advantage. Existing research shows that, reducing the defection rate just by 5% generates 85% more profits in one bank's branch system, 50% more in an insurance brokerage, and 30% more in an auto-service chain（Reichheld, 1990）. Loyal customers tend to buy more products, and have low price sensitivity (Reichheld, 1996). Therefore, enterprises that have long-term loyal customers gain more competitive advantage compared with those who have low unit cost, high market share but high customer return rate.Reward Programs, also known as Loyalty Programs, first appeared in 1981 with American airlines launching frequent flyer program. After a period of rapid development, the customer reward program has been widely used in all walks of life. 75% of American consumers are participated in at least one customer reward program; and 98% of Canadian consumers are participated in a customer reward program (The 2009 Colloquy Loyalty Marketing Census). The major means used by enterprises including, membership card, customer clubs, and bonus point which is the most popular one.Kim (2001) argued that customer reward program is used in the lucrative market segments to maintain a higher customer retention rate, by delivering more value and customer satisfaction to customers. Reward program gives rewards to quality customers, in order to sustain profits on the base of long-term relationship with customers (Kannan & Bramlett, 2000). Reward programs are initiated to achieve some sort of financial pay-off or strengthening of their long-term competitive position. (Sharp, 1997)In this study, Reward program is de?ned as a marketing method to maintain a higher customer retention rate, and mainly refers to Bonus Point Scheme.§2.2 The importance of the reward program§2.2.1 For customer retentionCustomer retention is the process of keeping a client's business and preventing the client from using a competitor's services or product. Customers are the most important asset for an enterprise, and the higher the customer retention is, the stronger profitability is, Compared to the cost of acquiring a new customer that of keeping an existing customer is much lower. What is more, long-term customers generally have such characteristics as repeat purchase ability and low price sensitivity, which will help improve profits. So in order to obtain long-term competitive advantage, enterprises have to use a customer-centric CRM strategy to retain customers that are most valuable.Reward program has a positive influence on customer retention and market share (Verhoef, 2003). Bolton (2004) argued that marketing activities (such as reward program, sales promotion, channel expansion and advertising) lack corresponding understanding of the length, depth and width of relationship between customer and the company. However, the reward program, an important means of relationship marketing, will first influence customer value perception, satisfaction and commitment, namely perception of relationship between companies and clients, and then affect the customer behavior.The customer retention can be divided into three dimensions, the length, width and depth. Relationship length refers to the possibility for a customer to continue the relations with the enterprise; Width refers to the number of time that customers purchase or use other products or services from the enterprises when keeping a relationship with the enterprise; Depth refers to the number of time that customers purchase or use products or services from the enterprises when keeping a relationship with the enterprise (Bolton, 2004). The three dimensions of customer retention are not independent. In fact, using the enterprise’s products or services is an essential element of customer retention, while the use of additional products or services and maintaining relations with the organization is based on the continued purchase or use of the product or service.Marketing plans represented by reward program has two common factors. First, they have made efforts to maintain customer. Second, they become more and more impotent for both strategies and industry development. In all markets and industries, marketing efforts are concentrated on getting closer to the customer, providing customers with customized products and services, as well as paying attention to feedback from the market and selecting valuable information.§2.2.2 For customer loyaltyFrom the perspective of consumer behavior, customer loyalty is the repeat purchase of a particular product or service in a period of time (Sharp, 1997). From the perspective of consumer attitudes, customer loyalty is the high level of commitment to repeat purchase their favorite products or services in the future, which will not switch to other enterprises because of changes in market and competition (Oliver, 1999). Dick & Basu (1994) believes that real customer loyalty only occurs when repeat purchase behavior is accompanied by a high emotional attitude. When customers decide to participate in a reward program, whether the plan itself is worth attending comes to the first consideration (O'Brien & Jones, 1995). Since the customers’ perception towards reward program is subjective to each individual, different customers have different perception. If the value gain is perceived as having greater value than the value loss, it is much more possible for customer to participate in the program; therefore the possibility for enterprise to cultivate customer loyalty through reward program will also increase. As mentioned earlier, the original intention of reward program is to cultivate customer loyalty, so a reasonable designed reward program that is effectively implemented should promote the formation of customer loyalty (Michael, 2009).Sharp (1997) studied the loyalty program in Australia. The purpose of their study is to investigate the ability for loyalty program to create “extra loyalty”. Based on the investigating and analyzing data of nine weeks before and after the Christmas, the study shows that only two(of six) brands’ loyalty program had significant influence on brand loyalty. The empirical conclusion showed that loyalty programs can change the consumers’ buying behavior, but the change is not obvious and hard to achieve.Dreze & Hoch (1998) did an empirical analysis of loyalty programs for baby products in a supermarket, which showed that: the effect is very significant during the reward program implementations, and there is a significant increase in both passenger volume and sales for all products (including non-infant products).Bolton (2000) argued that customer reward programs can create positive impact on customer evaluation, customer behavior and repeat purchase intention, due to the fact that reward program members feel they get more cost-effective services.Kopalle & Neslin (2003) believed that customer reward program is a powerful tool to increase sales, strengthen consumer relationship and increase consumer brand loyalty.Yi & Jeon (2003) distinguished the influence between program loyalty and the brand loyalty exerted by retailers’ reward programs in their study. Research indicated that as long as the customer reward program is considered worthwhile, costumers are willing to keep a long-term relationship with the enterprise.Lacey (2003) studied the strategic value of loyalty programs for retailers. The results showed that compared with customers who do not participate in the reward program, customers who participate in the reward program, especially those who use the reward program frequently, have a higher level of commitment and trust towards the enterprise. However, enterprises need to spend more to keep these customers.Chunqing Li & Yanfeng Xu (2004) established a dynamic CRM model with reward programs which use the return on reward program as the main variable. The results showed that the appropriate reward programs can promote consumer purchasing.Xiao Xiao (2008) pointed out that the state of the customer reward program in Chinese enterprises is not optimistic. It mainly shows in lacking of strategic thinking when implement a customer reward program, and the vast majority of reward programs are limited in a low level (such as price discount or coupon). Moreover, many enterprises were even forced to take action due to industry competition. These reasons lead to the result that the value of customer reward program is not significant. Customers did not get the appropriate value, they just got a few packages of tissues or washing powder, that results in a lot of reward program members are not loyal.The essence of the reward program is to give the customer a certain value, which needs to accumulated by constant buying behavior (or recommendation), and after reaching a certain line these values ??can be realized. For customers, the initial purchase may just be a process to get the desired products to meet the functional needs, and they did not think this process also has added value. Hence, this unexpected?bonus itself can please customer. However, once the customer emotionally considered all these values belonging to ??him, he will find a way to protect it, or even looking for ways to gain the added value, and this process will follow the rules of procedure of reward program to restrict their purchase choice. It is the customer loyalty desired by customer rewards program.Now that shopping online for travel has become prevalent in China, online travel shoppers have higher expectations of travel intermediates than before. Therefore the problem of retaining customers has become increasingly challenging for online travel agencies. In that context, reward program can be a useful tool for companies to maintain customer and gain customer loyalty, and the research on reward program can obtain realistic significance.§2.3 The?study?on reward program’s elementsVery few studies focus on affecting factors for customer to join a reward program, so the author needs to find another way to determine the affecting factors. And it makes sense to look into the study on rewards program’s elements in order to figure out the affecting factors for customer to join a reward program.Dowling & Uncle (1997) classified different types of reward program according to the reward’s support of the product or service value proposition and the reward’s timing.Reward type can be divided into two types: direct and indirect reward. Direct reward refers to the explicit rewards strongly associated with the product or service offered to customer, namely directly enhanced product or service value. And an incentive provided by indirect reward is not associated with the product or service offered to customer.Reward’s timing is also divided into two types: immediate and delayed reward. Immediate reward provides rewards for every purchase, while delayed reward only provides rewards after several times purchase.Timing of RewardImmediateDelayedType of RewardDirectly Supports the Product's Value PropositionRetailer/Brand Manufacturer Promotions(Price Promotions)Airline Frequent-Flyer Clubs.Coupons and Tokens(GM card)Other Indirect Types of RewardCompetitions and Lotteries(Instant Scratches)Multiproduct Frequent-Buyer Clubs(Fly Buys)Table 2.1 Reward typesThe immediate reward defined by Dowling & Uncle cannot distinguish between short-term promotion and long-term reward program, which means they did not differentiate short-term promotion from reward program. They think price promotion is a type of reward program. However, these two should be distinguished from the consideration of the enterprises’ original intention. Customer reward program concerns about loyal customers, so it is suitable to measure the customer's behavior from a long-term perspective; while the price promotion just a method to solve excess inventory, it is aimed at price-sensitive customers rather than loyal customer, and cannot cultivate loyal customers in a long-term.Kim (2001) considered reward program can weaken price competition by offering incentives for repeat purchases, thus resulting in higher profits; while price-promotion-oriented firms gain less from undercutting their prices. Considering that the reward program plays a role as "competitive leverage" or "exit barriers"(Klemperer, 1987), it makes sense to distinguish reward program from short-term promotion.For this reason, Yi & Jeon (2003) redefined and reclassified the type of reward program based on Dowling & Uncle’s study by adding repeated reinforcements to immediate rewards, and emphasizing that immediate reward is an immediately return for loyal customers’ repeat purchase. For example, if a supermarket always offers special lower prices to their rewards program members, this belongs to immediate and repeated rewards. Unlike simple price promotion, immediate and repeated rewards can help explain successive reinforcements of customer behavior and select target customers (provide rewards for members only), so that it could control value sharing toward loyal customers. (Yi & Jeon, 2003)Dowling & Uncle also classified reward program from the perspective of reward amount and pointed out that different reward amount can influence the buyer’s motivation to make the next purchase. The reward amount (or discount amount) offered by a typical reward program is equal, which means customer can become a member when up to a certain amount of the accumulated consumption is achieved, and after that for each dollar spent a participant gains the same number of points. However, Dowling & Uncle considered it is better to use a differential reward to enhance customer’s repeat purchase motivation, namely to offer more reward points for each additional dollar spent, so that the next purchase is increasingly more valuable to the customer.From the above we can classify the reward program from three dimensions (reward type, reward time and reward amount), as shown in figure 2.1: Figure 2.1 Reward classificationsChunqing Li consider that a complete reward scheme should answer four questions: who to reward (Who), when to reward (When), what to reward (What) and how to reward (How). Who is the problem about who is the target customer for reward program, which usually includes target market segmentation, target market positioning and target market decision. When to reward refers to the problem about reward timing, if it is an immediate reward, or a delayed reward. What to reward, in fact, is a problem about the value provided to the customer that is created by reward program, namely what can be redeemed from reward point when a customer meets the requirements of reward. There are usually two ways (Dowling & Uncle, 1997): direct reward (refers to the reward which is strongly associated with the product or service, namely direct enhancing product or service value) and indirect reward (refers to the incentive provided by reward program is not associated with the product or service.) How to reward has three aspects of meaning: the problem about how to calculate points, the issue of reward ratio, and the specific method of rewarding. Only when a reward scheme can answer the above four questions, can we judge that the structure of the reward program to be completed.From the discussion above, the reward program’s elements is defined and approached from very diverted perspectives and on different levels. So far there is no complete agreement reached over the classification of it. Therefore, further summary of the reward program’s elements is needed for finding the affecting factors.§2.4 The limitation of the existing researchThe most common reasons for stopping a customer reward program includes the rising cost and the reward program’s rejection by the target customer, and such consequence usually results from the fact that the reward program does not provide enough value to customers. The required financial and organizational costs for forming, initiating and maintaining a customer reward program are often underestimated, thus leading to the failure of a good reward program.Current documents and analyses mainly focus on how a reward program can lead to customer loyalty, and do not pay enough attention on how to attract customer to join a reward program. So this paper aims at understanding the real reasons for customers’ choosing a reward program and providing a reference standard for online travel agency to implement a reward program successfully.§3. Model and assumptions§3.1 Research hypotheses derivedIn this study, the main purpose of the exploratory study is to preliminary determine the influencing factors of reward program based on the existing literature, the typical user interviews and the collection of enterprises’ reward program. Based on content analysis, the paper established empirical analysis model and raised assumptions.§3.1.1 Existing literature summaryAccording to the literature analysis, the result is shown on table 3.1, the code number refers to the project frequency appeared in literatures.ProjectDefinitionCode numberRelated articlesDirect rewardRewards directly support the value proposition of the product or service offered to customers2Dowling & Uncle, 1997；Chunqing Li, 2007Indirect rewardRewards that are designed to motivate loyalty by a more indirect route2Dowling & Uncle, 1997；Chunqing Li, 2007Tangible rewardReward that can be measured by cash value(such as cash, products and coupon)1O'Brien & Jones, 1995Intangible rewardRefers to the sense of belonging. Trying to give the consumer feelings of being recognized or make them feel special difference with other customers1O'Brien & Jones, 1995Aspirational valueThe degree of attracting customers for reward program1O'Brien & Jones, 1995Equal amountFor each dollar spent a participant gains the same number of points2Dowling & Uncle, 1997；Kiveze & Simonson, 2003Differential rewardOffer more reward points for each additional dollar spent, so that the next purchase is increasingly more valuable to the customer2Dowling & Uncle, 1997；Kiveze & Simonson, 2003Immediate rewardImmediate reward provides rewards for every purchase3Dowling & Uncle, 1997; Yi & Jeon, 2003; Chunqing Li, 2007Delayed rewardDelayed reward only provides rewards after n times purchase3Dowling & Uncle, 1997; Yi & Jeon, 2003; Chunqing Li, 2007Reward rateThe ratio of cash value to reward threshold2O'Brien & Jones, 1995; Chunqing Li, 2004Reward thresholdThe requirements in order to get reward2O'Brien & Jones, 1995; Chunqing Li, 2004Time?horizonThe time constraints for Calculating the total purchase2O'Brien & Jones, 1995; Chunqing Li, 2004Limited rewardA limited loyalty program cannot be joined by just anybody.2Butscher, 2005; Chunqing Li, 2007Open rewardOpen loyalty program can be joined by anybody.2Butscher, 2005; Chunqing Li, 2007Target customer groupsSelect and segment customers1Chunqing Li, 2007Reward wayThe way to give customer reward1Chunqing Li, 2007Table 3.1 Literature summary§3.1.2 InterviewsIn general, interview is a way of collecting market information by visiting, symposia, etc. Strictly speaking, interview method belongs to field survey.Despite its high average cost per unit, the limited of investigation and quantity, interview method has incomparable advantages such as like in-depth communication, interactive communication, as well as reliable and abundant information sources.Interview is not only a marketing method, but also a working style for consultants, business leaders and managers that can be used in many occasions. In many cases, interview also can be used at any time, in order to understand and master the local situation, and to accumulate information.The interviewees of this study mainly include internal staff and regular customers. And the interview goals are: fully understand Ctrip’s reward program through the interview, find the difference between each reward characteristics, and excavate the influencing factors of reward program through the interview.The influencing factors of reward program from the interview include: personalized products; differential reward; the validity of the points.§3.1.3 Existing reward programs summaryOnline Travel is an emerging industry in China, so we do not have many research findings on it. For this reason, we start from the overall situation.At present, the enterprises that use reward program in China concentrate on four industries: banking, aviation industry, mobile telecommunication industry and retail.The existing reward programs in this research mainly includes: supermarket’s reward programs which are easy to participate in, such as Darunfa supermarket and Hualian supermarket; mobile communications industry’s reward programs, whose characteristics are high public participation, no access conditions, automatic accumulate points after consumption; bank’s reward programs, whose characteristic is high transparent information such as those by ICBC, CBC, BOC ; airline’s reward programs whose characteristic is well designed compared to other industries because airline’s reward program is the earliest one in China.Bank’s reward points accumulating methods includes: credit card consumption/withdrawal/installment, bank loans settlement, personal foreign exchange trading, national debt, trust and fund, etc.Airline’s reward points accumulating methods includes: take flight; take partner airlines flight; stay in well-known hotel; pay by credit card; use car rental services; use telecommunications services, etc.Mobile communications industry’s reward points accumulating methods includes: communication consumption, duration of use, etc.Retail’s reward points accumulating methods includes: consumption amount.Air miles reward scheme has the biggest number of reward choice in these four industries, namely it provides more options to the customer's: not only can they swap points in the partner airlines, but they can also choose other industries’ reward. On the contrary, there is no points swap cooperation between the banks or the communication companies, either in- or outside the industry. And the absolute value of reward is low. This entire reward program did not integrate the opinions of the members, nor did they find value driving factors. Banks and airlines provide the same service, such as hotels discount and car rental discount, so customer's perception of value became lower due to the lack of originality. The companies just provide these services, and do not increase the value of the reward program by providing special or creative benefit. Comparison banking, aviation, telecommunications, and retail reward program reveals that they all adopt similar cumulative rules, that is, use portfolio or transactions as a measurement. Customer reward marketing pattern has entered a stage of alliance, not only within the same industries, but also across industry borders. Alliance’s biggest benefit for customers is to expand the range of point’s usage, therefore improve customer satisfaction. For companies, alliance overcomes the limitations of enterprises set up reward program on their own and shares the costs.ProjectCode numberExamplesDirect reward245 airlines, 3 telecoms, 5 supermarkets, 11 banksIndirect reward195 airlines, 3 telecoms, 11 banksReward rate245 airlines, 3 telecoms, 5 supermarkets, 11 banksEqual amount75 airlines, 2 supermarketsDifferential reward173 telecoms, 3 supermarkets, 11 banksImmediate reward53 telecoms, 2 supermarketDelayed reward245 airlines, 3 telecoms, 5 supermarkets, 11 banksReward threshold245 airlines, 3 telecoms, 5 supermarkets, 11 banksTime?horizon245 airlines, 3 telecoms, 5 supermarkets, 11 banksOpen reward245 airlines, 3 telecoms, 5 supermarkets, 11 banksTarget customer groups235 airlines, 3 telecoms, 4 supermarkets, 11 banksReward way245 airlines, 3 telecoms, 5 supermarkets, 11 banksTotal245 airlines, 3 telecoms, 5 supermarkets, 11 banksTable 3.2 Existing reward program summary§3.1.4 The reward program influencing factorsChunqing Li’s study shows that a complete reward program should contain four structure factors: who, when, what, and how. O'Brien & Jones (1995) argued that customers can measure a rewards program from five aspects: Cash Value, Aspirational Value, Redemption choice, Relevance, and Convenience.Based on O'Brien & Jones’s study, we summarized four reward program influencing factors: Reward Value, Reward Type, Convenience, and Possibility. And according to the exploratory research above, we obtained one more influencing factors: Customer Fit.Therefore we put forward the following five reward program influencing factors: Reward Value, Reward Type, Convenience, Possibility, and Customer Fit. The following hypothesis is proposed:H1 Reward value has a positive impact on customer choice.H2 Reward type has a positive impact on customer choice.H3 Convenience has a positive impact on customer choice.H4 Possibility has a positive impact on customer choice.H5 Customer fit has a positive impact on customer choice.§3.1.5 The mediating role of customer perceptive value assumptionsZeithaml (1998) believes that customer perceived value typically involves a tradeoff between what the consumer receives and what he or she gives up to acquire and use a product or service, which is a general evaluation of the utility of the product or service. Monroe (1990) interpreted customer perceived value as a ratio of perceived benefits to perceived cost, which is almost the same with Zeithaml's definition. Woodruff (2002) also noted that customer perceived value should be a comparison process related to competitors’ products or services value. Grewal (1998) considered perceived value as a dynamic concept, which includes four types of value: acquisition value, transaction value, in-use value, and redemption value. Acquisition value refers the consumer’s benefits (or tradeoff) from acquiring the product or service. Transaction value refers to the pleasure consumers experienced at the point of purchase when getting a good financial deal. In-use value involves the utility derived from using the product or service. Redemption value relates to the residual benefit at the time of disposing the product or terminating the service.Dodds & Grewal (1991) proved that customer perceived value as an intermediary variable factor between price and customers’ purchasing intension.O'Brien & Jones (1995) considered that customers can determine a program's value from a customer's perspective from five elements. Therefore, the author also concluded that the customer perceived value is an intermediary variable factor between influencing factors and customers participating intension. The following hypothesis is proposed:H6 Customer perceived value has a positive impact on customer choice.H6a Reward value has a positive impact on perceived value.H6b Reward type has a positive impact on perceived value.H6c Convenience has a positive impact on perceived value.H6d Possibility has a positive impact on perceived value.H6e Customer fit has a positive impact on perceived value.§3.2 ModelingCombining the above analysis, the research model is proposed and is shown on figure 3.1.Figure 3.1 Research model§3.3 HypothesesResearch model describes the relationship between independent variables (Reward Value, Reward Type, Convenience, Possibility, and Customer Fit), the mediate variable (Perceived value) and the dependent variable (Customer choice).These relationships are the hypotheses of this study that need to be tested.H1 Reward value has a positive impact on customer choice.H2 Reward type has a positive impact on customer choice.H3 Convenience has a positive impact on customer choice.H4 Possibility has a positive impact on customer choice.H5 Customer fit has a positive impact on customer choice.H6 Perceived value has a positive impact on customer choice.H6a Reward value has a positive impact on perceived value.H6b Reward type has a positive impact on perceived value.H6c Convenience has a positive impact on perceived value.H6d Possibility has a positive impact on perceived value.H6e Customer fit has a positive impact on perceived value.§4. Empirical research§4.1 The research object International is the biggest consolidator of hotel accommodations and airline tickets in China. Ctrip is the first Chinese tourism company to be listed on NASDAQ. Founded in early 1999, Ctrip is headquartered in Shanghai, China, with Beijing, Guangzhou, Shenzhen, Hong Kong four branches, and has branches in more than 20 large and medium-sized cities in China; the existing staff is more than 1,500. Ctrip provides travel related services including hotel reservation, air-ticketing, packaged tour services, internet advertising and other related services.Ctrip as the top online travel OTA (online travel agency) among China has a relatively complete reward program, so it is a representative object for the author to understand the online OTA’s reward program in China.§4.2 Research variables and scale designFive independent variables, one intermediate variable and one dependent variable are used in this study.VariablesVariable typeNumbers of itemsReward valueIndependent variable4Reward typeIndependent variable5PossibilityIndependent variable3ConvenienceIndependent variable6Customer fitsIndependent variable4Value perceptionIntermediate variable4PreferenceDependent variable4Table 4.1 Variables§4.2.1 Independent variablesOn the basis of literature research combined with the actual interviews, the variable measure items are shown in table 4.2.VariablesMeasure itemsReward valueI think the rewards provided by Ctrip have a relatively high value.Relative to my spending, I was satisfied with the return pared with my spending amount, I think the discounted value of the goods or services I obtained in return are too low. I think the value of the rewards provided by Ctrip is higher than other similar sites.Reward typeThe reward program provided by Ctrip offers a large variety of reward.I am very concerned about what products can I get as a reward.I would prefer to get products or services directly related to the website like discounted hotel rates.I would prefer to get products or services not directly related to the website like Daily Necessities.I would prefer to get tangible things like vouchers& commodities.PossibilityIt is easily for me to reach the amount of consumption required by Ctrip (the minimum amount of consumption to participate in reward program) I think there is a big possibility for me to get reward. After using this website, I will soon be able to participate in reward program.ConvenienceI prefer permanent reward program. I think it is better for reward program to have a pared to delayed reward, I prefer to get immediate reward or pared to immediate but lower reward, I prefer high-value rewards that need time to cumulative.I think reward program provided by Ctrip is easy to check. (Like Points query) I think reward program provided by Ctrip is easy to use. (Like Redemption)Customer fitsIn contrast, I believe it is appropriate that different accumulate points can be converted to a products have different values.I think the higher the spending amount is, the greater the reward ratio should be. If my points are relatively high, I hope I can get special products that others cannot have.I think my membership can be recognized and respected.Table 4.2 Independent variables measurement§4.2.2 Intermediate variableVariablesMeasure itemsValue perceptionCompared to other traveling website I prefer reward program provided by Ctrip.The reward product form Ctrip‘s rewards program is exactly what I want. As far as the money, time, effort I spent, the rewards program is worth it.I think it is a good choice to participate in this reward program.Table 4.3 Intermediate variable measurement§4.2.3 Dependent variableVariablesMeasure itemsValue perceptionI really like Ctrip’s reward programs.The rewards program will encourage me to spend more.I would recommend the program to others.I have a strong preference towards Ctrip’s reward programs.Table 4.4 Dependent variable measurement§4.2.4 Demographic variablesIn order to have a preliminary understanding for the characteristics of Chinese OTA website customers, this paper chose the following demographic variables:Gender: defined as a binary variable (male or female).Age: decided by participants’ age.Income: referred to the customers’ total revenue per month.Education situation: the highest level of education that customers have completed. Time of usage: referred to the years that customers use the website since the first time.§4.3 Questionnaire design§4.3.1 Questionnaire measuresThis research adopted three kinds of measure methods: nominal scale, ordinal scale and interval scale. Nominal and ordinal scale are mainly used in the survey to measure participants’ demographic indicators, for example using a nominal scale to investigate gender, using an ordinal scale to investigate education situation. And the interval scale is used to measure survey participants’ viewpoints on a particular issue or tendency.The interval scale uses 7 point Likert Scale, with 1 means “Totally disagree” and 7 refers to “Totally agree”. All questions are closed questions, no open questions are used. In general, the authors often use five or seven points Likert Scale, more points means more detailed, therefore theoretically the measurement are more precise. So the 7 point scale is adopted.However, from the author's research experience, 7 point scale also has a lot of defects. Influenced by Chinese traditional culture, the participants tend to choose the median 4, thus can cause a central tendency error, which means that many participants’ attitudes are not showed. Therefore, the author would like to explore the 6 point Likert Scale by deleting the median 4, to forcing participants to show their stances, but 6 point Likert Scale is rarely seen in the research, so 7 point scale is adopted in this study.§4.3.2 Questionnaire structureIn this study, the organizational structure of the questionnaire’s first draft is arranged as following: first part is to introduce the purpose and significance of the questionnaire to participants, and asks for participants’ serious answers, in order to get higher quality research data; Then comes the main part of the questionnaire, 7 Likert Scale are used to test various research variables, and this part is ended with an either-or question about if they will choose the specific reward program; the last part of the questionnaire is about participants’ background data, mainly used for the analysis of samples’ characteristics.§4.3.3 Preliminary testAfter the first questionnaire draft was designed, a small-scale experiment has been done in order to prevent the experiment questionnaire cannot be fully understand by the participants. The author invited 10 participants to read and do the questionnaire and asked them to point out the expression problem and the structure problem in the questionnaire which need to be revised.This preliminary test focused on the following aspects: (1) if the scale’s test items are semantically correct; (2) if the test items can be understood by the participants; (3) if there are some questions that are difficult to answer; (4) other difficulties that may occur in the process of doing experiments.§4.3.4 Questionnaire reviseQuestionnaire expression is corrected and a few variables are removed based on preliminary experiment.VariablesRemoved measure itemsReward valueI think the rewards provided by Ctrip have a relatively high value.PossibilityI think there is a big possibility for me to get reward.Table 4.5 Removed measure itemsVariablesChanged measure itemsCustomer fitsIf my points are relatively high, I hope I can get special treatment (products) that others cannot have.Table 4.6 Changed measure items§5. Data analysis§5.1 Sample description§5.1.1 Research method and research objectThis research mainly used an internet-based questionnaire. The author put the questionnaire on the internet and request Ctrip’ reward program members to fill out the questionnaire.§5.1.2 Sample size245 questionnaires were collected, ant the author rejected 15 invalid questionnaires (e.g., all the answers are the same, or there are questions unanswered), acquired 230 effective questionnaires. The effective return ratio is 93.8%.§5.1.3 Sample descriptionThe questionnaire’s sample characteristics are shown on table 5.1.VariablesOptionsFrequencyPercentValid PercentCumulative PercentGenderMale12554.354.354.3Female10545.745.7100.0Total230100.0100.0　AgeUnder 1820.90.90.918-3010545.745.746.531-4010445.245.291.741-50187.87.899.6Above 5010.40.4100.0Total230100.0100.0Income（Yuan/month）Under 2000187.87.87.82001-500017676.576.584.35001-80003314.314.398.78001-1000010.40.499.1Above 1000020.90.9100.0Total230100.0100.0　EducationUnder junior college156.56.56.5junior college9440.940.947.4bachelor10344.844.892.2master187.87.8100.0Total230100.0100.0　Using time(years)Under 0.53816.516.516.50.5-18737.837.854.31-27934.334.388.72-3198.38.397.0Above 373.03.0100.0Total230100.0100.0　Table 5.1 Descriptive statisticsFrom the above-mentioned basic features we can see:GenderIn all 230 effective questionnaires, the male account for 54.3% of the samples, and female account for 45.7%, the male to female ratio is 1.19, in this study man samples are more than woman samples.AgeIn all 230 effective questionnaires, age distributions are ?centralized on 18 to 30 years old samples (45.7%) and the 31 to 40 years old samples (45.2%); 41 to 50 years old samples (7.8%); under 18 years old samples (0.9 %); And above 50 years old samples (0.4%).In this study, most participants are between 18 to 40 years old, accounting for 90.9% of the research object. The crowd is mainly composed of young adults, which are the main users of OTA websites, and also the main target group of reward program.EducationIn all 230 effective questionnaires, most samples are junior college students or university graduates, 40.9% and44.8% of total samples respectively. These two kinds of samples accounted for 85.7% of the total research object. And they are the main users of new business, they are also the major customers in the future, due to they are interested in internet business, also has the ability to use more internet business in the future.IncomeIn all 230 effective questionnaires, the samples are intensive in 2001-5000Yuan per month (76.5%). And these samples are the main users of reward program because they are affordable for a travel or ticket while they pay more attention on rewards because of their income is not very high.Using timeIn all 230 effective questionnaires, most participants are using the for at least half a year.It can be seen that most of the research objects are regular customers of , they are familiar with the website, and they use the reward program provided by Ctrip more frequently. Therefore through their feedback we can see the actual application situation of the reward program and its impact on customers.§5.2 Reliability analysisBefore doing reliability analysis, the author first did a descriptive statistical analysis on each measurement term, and mainly gave out the mean and standard deviation of the test item, which can roughly reflect the attitude of the participants.From table 5.2, it is easy to see that most of the test items’ mean are above 4, and all test items’ standard deviation are above 1.2, in line with the Nunally’s (1978) requirements on the Likert Scale that all standard deviations should be greater than 0. 5.VariableMinimumMaximumMeanStd. DeviationReward value1175.211.3702173.061.5073174.631.291Reward type1175.011.2692175.641.1913174.191.8324175.001.2455275.721.134Possibility1174.861.2782174.311.385Convenience1275.541.0092173.731.8723175.451.1084174.202.2255275.031.0516174.501.128Customer fit1375.720.8972475.820.9493375.340.9754174.941.235Perceived value1175.101.2232174.961.1443174.891.3074174.841.175Loyalty1174.961.2362175.221.4593174.931.3214174.691.416Table 5.2 Means and Std. DeviationsReliability is the degree to which an assessment tool produces stable and consistent results (Camines& Zeller, 1979). Scores in the same scale measured by different terms are affected by errors. However, the higher the reliability is, the smaller the influence will be. Thus answers have a consistent change between different respondents, and it can reflect the real situation.Reliability has two types: external reliability and internal reliability. External reliability usually refers to the consistency of the scale when measured in different times; retest reliability is commonly used in testing external reliability. This research adopts the cross section data, so there is no need to test the external reliability. Internal reliability assesses the consistency of results across items within a test. Cronbach’s Alpha coefficient is used the most to test internal consistency. This paper uses the Cronbach’s Alpha coefficient to evaluate the reliability of questionnaire.Cronbach's alphaInternal consistencyα ≥ 0.9Excellent (High-Stakes testing)0.7 ≤ α < 0.9Good (Low-Stakes testing)0.6 ≤ α < 0.7Acceptable0.5 ≤ α < 0.6Poorα < 0.5UnacceptableTable 5.3 Cronbach’s Alpha standardsVariablesNumbers of itemsCronbach's alphaReward value30.703Reward type50.710Possibility20.847Convenience60.658Customer fits40.686Value perception40.756Preference40.792Table 5.4 Cronbach’s Alpha coefficient From table 5.4, we can see that the reliability of all variables is above 0.6, the reliability is acceptable. Therefore the scale has good stability and consistency.§5.3 Validity analysisValidity is the extent to which an instrument does indeed measure what it is supposed to measure in order to be valid; it reveals the relationship between the structure variables and its measurement terms (Zikmund, 1995). So the inferences made from scores need to be “appropriate, meaningful, and useful” (Gregory, 1992). Validity generally falls into four categories: content validity, construct validity, criterion validity and consequential validity (Messick, 1995)In this paper, the Factor analysis (use principal component method to extract influence factors) in SPSS20 is used to determine the validity. Since factor analysis is based on correlation coefficient, Bartlett spherical analysis can be used to test whether the correlation coefficient is greater than 0 and a significant result of spherical analysis shows that correlation coefficient is enough to extract factors. KMO coefficient refers to the ratio of all correlation coefficient related to the variable to net correlation coefficient, so the bigger the ratio is, the stronger the correlation is. And KMO should be greater than 0.5 for factor analysis. Therefore a KMO - Bartlett test was done before a factor analysis.VariablesKaiser-Meyer-OlkinBartlett's Test of SphericityApprox. Chi-SquareDf.Sig.Reward value0.696231.4303.000Reward type0.59750.2566.000Possibility0.60035.0531.000Convenience0.651239.1063.000Customer fits0.638161.1036.000Value perception0.753210.7506.000Preference0.775264.3786.000Table 5.5 KMO - Bartlett testAll variables’ KMO are more than 0.5, the Bartlett test is significant, so we consider the scale is suitable for factor analysis.The orthogonal solution used a varimax rotation. The analysis results are shown on table 5.6. We can see that the vast majority of the terms of different scale in the model are loaded on the same factor, so the evaluation criteria are satisfied. Therefore the scale can achieve good quality on the convergent ponentPreferenceReward valueReward typePossibilityConvenienceCustomer fitPerceived valueL1.796　　　　　　L2.670　　　　　　L3.584　　　　　　L4.535　　　　　　RV1　.622　　　　　RV2　.779　　　　　RV3　.714　　　　　RT1　　.719　　　　RT2　　.817　　　　RT3　　.744　　　　RT4　　.614　　　　RT5　　.429　　　　P1　　　.716　　　P2　　　.666　　　C1　　　　.685　　C2　　　　.858　　C3　　　　.794　　C4　　　　.735　　C5　　　　.358　　C6　　　　.584　　F1　　　　　.715　F2　　　　　.725　F3　　　　　.474　F4　　　　　.683　PV1　　　　　　.730PV2　　　　　　.761PV3　　　　　　.657PV4　　　　　　.834Table 5.6 Varimax rotationVariables’ Extraction Sums of Squared Loadings are listed in the table below; it can be seen that the variables can be explained very well.VariablesExtraction Sums of Squared LoadingsReward value43.154%Reward type58.455%Possibility62.989%Convenience48.843%Customer fits60.401%Value perception68.791%Preference63.929%Table 5.7 Extraction Sums of Squared LoadingsSummary: the former analysis of the reliability and validity of the questionnaire shows that the quality of the questionnaire is quite good and suitable for further analysis.§5.4 Correlation analysisAs can be seen from the table, there is a strong relationship between independent variables and dependent variable, independent variables and intermediate variable in the model.Variable namesReward valueReward typePossibilityConvenienceCustomer fitPerceived valuePreferenceReward value1Reward type.135*1Possibility.024.0421Convenience.133*.085.0491Customer fit.082.122.115.199*1Perceived value.379**.438**.496**.311**.222**1Preference.460**.370**.495**.291**.166*.771**1**. Correlation is significant at the 0.01 level (2-tailed).*. Correlation is significant at the 0.05 level (2-tailed).Table 5.8 Correlation§5.5 Structural Equation ModelThis article uses the Structural Equation Model (SEM) to analyze the data. SEM is a new developing method in the field of statistical analysis, and has been extensively used since the early 90s. SEM does not have a very strict requirement, while allowing the existence of independent variable and dependent variable measuring errors. So it performs better in the quantitative study of the interactive relationship between multivariate comparing with multiple regressions, factor analysis and other methods.As for analysis software, the AMOS 17 is used to study whether the structural equation model can be supported.§5.5.1 Structural Equation Model analysisThe calculation results of model path coefficient index, the error term and the load are illustrated in figure 5.1.Figure 5.1 Structural Equation Model§5.5.2 Model-fitting testingAbsolute Fit IndexComparative Fit IndexIndexDFx2x2/DFGFIRMRRMSEANFICFIStandard model128195.12(p=0.000)1.5230.8980.04160.0540.9130.937Table 5.9 Model-fittingFrom the model-fitting index, the degree of model fitting is good.The Chi Square Test: For models with around 75 to 200 cases, the Chi Square test is a reasonable measure to test fit. If the Chi-Square is not significant, the model is regarded as acceptable. The x2=195.12 and the p=0.000 is not significant in this case. If relative Chi-Square is less than 2 or 3 the model is regarded as acceptable. (Kline, 1998; Ullman, 2001). As for x2/df=1.523＜2, so we consider the model-fitting is good from these two indexes. However, Chi Square is easily affected by the correlations in the model: the larger the correlations, the poorer the fit, so alternative measures of fit have been developed.Goodness of Fit Index (GFI): GFI typically summarize the difference between observed values and the expected values under the model. If the Goodness of Fit Index exceeds 0.90, the model is regarded as acceptable. (Byrne, 1994) In this case, the GFI=0.898 is approximately equal to 0.9, so the model-fitting is good from this index. However, this measure is influenced by sample size, so we need more method to confirm it.Root Mean Square Residual (RMR): The RMR is defined as the difference between the observed correlation and the predicted correlation, and RMR is an absolute measure of fit. Therefore, a value closer to zero indicates a more perfect fit. If the value is less than 0.08, the model is regarded as good fit. (Hu & Bentler, 1999). In this case, the RMR=0.0416＜0.08, so the model-fitting is good from this index.Root Mean Square Error of Approximation (RMSEA): This is an absolute measure of fit and the RMSEA is currently the most popular measure of model fit. MacCallum, Browne & Sugawara (1996) has used 0.01, 0.05, and 0.08 to indicate excellent, good, and mediocre fit, respectively. That is, RMSEA values <0.01 are considered to indicate a good fit, RMSEA values <0.05 are considered to indicate a suitable fit, and RMSEA＜0.08 are considered to indicate a suitable fit. However, Hu & Bentler (1999) has suggested that RMSEA less than 0.06 the model-fitting is not good. In this case we adopted the measurement form MacCallum et al, the RMSEA=0.054＜0.08, so we consider the model-fitting acceptable.Bentler-Bonett Index or Normed Fit Index (NFI): It is an incremental measure of fit. The index value ranges from 0 to 1,and when the value is between 0.90 and 0.95, the model is considered marginal; When the value is above 0.95, the model is good fit; and when the value is below 0.90, the model is considered to be a poor fitting model and needs to be reset. In this case, the NFI=0.913＞0.90, so the model-fitting is suitable from this parative Fit Index (CFI): This is an incremental measure and is less affected than other indices by sample size and model complexity (Bollen & Long, 1993). The index value ranges from 0 to 1. If the index is greater than one, it is set at one; and the index is less than zero, it is set to zero. When the value is between 0.90 and 0.95, the model is considered a good fit; when the value is above 0.95, the model is a perfect fit; and when the value is below 0.90, it is considered non-satisfactory model fit. In this case, the CFI=0.937＞0.90, so the model-fitting is good from this index.Bentler & Chou (1987) pointed out: for a model that contains several variables, it is difficult to fully achieve the theoretical goodness-of-fit.This model includes 7 variables and 28 measuring terms, so some fitting indexes cannot reach 0.9 is acceptable, and the results are approximately equal to 0.9. What is more, the rest of the fitting indexes shows that the model fitting is good. Therefore we consider the degree of model fitting is good in general.§5.5.3 Hypothesis testingThere are 11 hypotheses in total in this paper, combined with model-fitting indexes and significance test index P value, we can know that 6 hypotheses are confirmed and 5 hypotheses are rejected. The hypothesis testing result is shown in table 5.5(A hypothesis is accepted when coefficient is significant at the 0.05 level): HypothesisEstimateS.E.C.R.PTest ResultsH1:Y<---X10.5840.1234.768***AcceptedH2:Y<---X2-0.470.032-1.4520.146RejectedH3:Y<---X30.8980.1177.71***AcceptedH4:Y<---X4-0.4280.314-1.360.174RejectedH5:Y<---X50.4730.1982.3850.017AcceptedH6:Y<---M0.980.1019.693***AcceptedH6a:M<---X10.9140.4192.180.029AcceptedH6b:M<---X20.5140.3071.6730.094RejectedH6c:M<---X3-0.0370.081-0.4610.899RejectedH6d:M<---X40.3670.0692.4260.015AcceptedH6e:M<---X50.5690.0220.430.668Rejected***. Coefficient is significant at the 0.01 level (2-tailed).P＜0.05. Coefficient is significant at the 0.05 level (2-tailed).Table 5.10 Hypothesis testing§5.6 The revised modelThe customer reward program’s influence factor model proposed by figure 3.1 can be modified through analysis. We keep confirmed model assumptions and remove the rejected hypotheses, hence finally determining the conceptual model of customer reward program’s influence factor, which is shown in figure 5.1. Overall, the customer reward program conceptual model includes 6 elements.Figure 5.2 Revised model§5.7 ConclusionsReward value:Hl&H6a are confirmed, there is a significant positive?correlation?between reward value and customer choice, and there is a significant positive?correlation between reward value and perceived value. Reward value is often a major determinant for customer to participate in a reward program, especially considering the high homogeneity and the similar price in the OTA’s products, the influence is even greater. Reward value can attract customers’ attention of the reward program, and promote the consumption. In the current Chinese OTA purchasing environment, customers generally believed that the reward value is the most worthy reward in customers’ perception, and reward value is a key point for customers to evaluate whether the reward program is worth to attend or not.Reward type: H2&H6b are rejected, there is no significant?correlation?between reward type and customer choice. Also, there is no significant?correlation?between reward type and perceived value. At present, Ctrip did not provide what they want to their members and make them feel satisfied. Any single factors, such as offering more reward type, cannot play a decisive role, and listening to the customers’ voice and giving them the right reward type may lead to a different result. At present, the majority of the members did not really feel this kind of special advantage, so they feel these aspects did not bring them extra perceived value, let alone to participate in the reward program due to this factor. In the future, Ctrip should provide more reward types to customers and try to understand the important reward types, thus enabling the customers to feel real difference from reward type, and feel the reward program can give them value, so that they would like to participate in the scheme more.Convenience:H3 is confirmed but H6c is rejected, which means there is a significant positive?correlation?between convenience and customer choice but there is no significant correlation?between convenience and perceived value. A reward program is easier to participate in if it more convenient. Although customers did not feel the true value provided by Ctrip’s reward program, they would participate in a reward program if it is convenient and would not take much time, because the costs are relatively low and a gift is better than none. In short, customers get benefits from the reward program but not real perceived value. But as a result of low cost (convenience), customer will participate in the reward program.Possibility:H4 is rejected but H6d is confirmed, so there is a significant positive?correlation?between possibility and perceived value but there is no significant?correlation?between possibility and customer choice. The higher possibility for customer to get reward, the higher the perceived value they considered. On the contrary, the lower possibility for customer to get rewards, the lower the perceived value. The possibility for customer to get reward can affect the choice of customer to participate in reward program indirectly by influencing perceived value. Which is to say, a high possibility can indirectly improve the possibility of customers to participate in a reward program.Customer fit:H5 is confirmed but H6e is rejected, which means there is a significant positive?correlation?between customer fit and customer choice but there is no significant?correlation?between customer fit and perceived value. The costumer will be interested in the reward program if the customer fit is high, and would be willing to participate. However, the motivation for customer to participate in the reward program is just because the want to get the rewards the want and they will not be loyal to the reward program after they get what they want. In short, customers get benefits from the reward program but not real perceived value. But as a result of desirable gifts (customer fit), customer will participate in reward program. So Ctrip should collect and analysis the members’ data and launch the appropriate gifts based on these data. By continually providing the humanized and customized reward, it can let customers participate in the reward program and try to increase the customer perceived value at the same time.Perceived value:H6 shows that there is a significant positive?correlation?between perceived value and customer choice. When customers choosing to participate in a reward program, the first thing to consider is whether the scheme itself is worth to attend (O 'Brien and Jones, 1995) .Customer perception of reward program is very subjective. Different customers have different perception of reward programs. But if customers think the benefits for participate in a reward program outweighs the costs significantly, the likelihood of customers to participate in the plan is high, and therefore the possibilities for enterprises to cultivate customer loyalty by reward program is high.To sum up, this study can be the following main conclusions:Reward value has a positive impact on customer choice and reward value has a positive impact on perceived value.Convenience has a positive impact on customer choice.Possibility has a positive impact on perceived value.Customer fit has a positive impact on customer choice.Perceived value has a positive impact on customer choice.§6. Conclusions and limitations§6.1 ConclusionsThrough the above analysis, two main conclusions are obtained:Reward value, Convenience and Customer fit are the three decisive factors for customers to choose whether they would participate in a reward program or not. They all have a significant direct positive influence for customers to participate in a reward program, as for the reward type and possibility, the influence of these two factors are not significant.Even these three factors all have significant influence on customers’ choice, the weights of each factors are different. Among them, the convenience is the most powerful factor, followed by the reward value, and customer fit.Perceived value has significant direct positive influence for customers to participate in a reward program. And reward value and possibility have indirect positive influence for customers to participate in a reward program by influencing perceived value.Even these two factors all have significant influence on perceived value, the weights of each factors are different. Among them, the reward value is the most powerful factor, followed by possibility.§6.2 SuggestionsThe research conclusion in this paper for reward program has the following significance:First of all, a reasonable designed reward program can promote customers’ psychological intentions to participate in a reward program. The conclusion of this paper points out that convenience is the most significant factor in promoting customers to participate in a reward program. It shows that enterprises should pay more attention to convenience of customers to participate in the reward program when designing a reward program, for instance, providing multiple consulting methods, providing more redeem way, and providing more convenient in terms of customer service, etc. Customers can thus perceive more value from participating in the reward program. Further they are likely to produce more identity and attachment feeling towards the enterprise and become one of the most enthusiastic advocates and supporters of the enterprise, which is the basis for customers to bring long-term value for the enterprise.Secondly, the customer fit is as important as the reward value. This shows that the enterprise can consider more about providing the rewards that customers are interested in, rather than only pay attention to reward value in the process of designing a reward program, This point happens to be in line with the enterprise's actual need, because the higher the reward value, the higher the enterprise’s costs. In the process of implement reward programs, enterprise should launch customer survey from time to time, in order to understand customer fit and enable customers to maintain long-term relationship with the enterprise.Finally, reward type is not significant when customer choose to participate in the program. This shows that customer does not care about what kind of reward offered by enterprises or whether he can eventually get the reward. This may be associated with customers’ psychological benefits from participate a reward program. Such as customers think they have a sense of belonging, or have a membership card is the embodiment of the identity etc. So when design reward program conditions, enterprises should fully consider the customer's perception, arouse the enthusiasm of the participating possibility from psychological, and give customers appropriate reward, and then can get maximal customer perceived benefits with minimal economic cost. §6.3 Research contributionsBased on relevant literature review, the paper recognized that most of the past studies are focused on the influence of the reward program and reward program’s impact on corporate profits, and there is few past studies that focus on influence elements of reward program. On the basis of past studies, further study on the influence of five elements, which are the reward value, reward type, convenience, possibility and customer fit, has been done. Whether each factor will exert positive effects on customers is the purpose of the rewards program design. So in this paper, we focus on influence elements of reward program and established the concept model of customer’s participation of reward program. Although many literatures argued perceived value has impact on customers’ choice on reward program and many literatures put forward the affecting factors for perceived value, there are few literatures talking about the influence elements on customers’ participation on reward program. These factors are combined in this paper. Although there are literatures proposed the five elements that influence customer perceived value, they did not test the theory by empirical investigations (O'Brien & Jones, 1995). Based on their research, the new proposed five elements are tested. On the basis of above research, the influence factor-customer fit has been added according to the interview and the existing reward program analysis. The new element makes the research more in line with China's actual situation, and fills the research deficiency on this aspect in China.This study also makes clear the relationship between the affecting factors and the customer’s participation of reward program; this is the basis for the further study of relationship marketing and service marketing. Empirical study results show that only part of the elements have positive influence on the measurement variables of customer’s participation of reward program. This is not exactly the same with what we usually think "once you have a good customer relationship, the customer will participate in the reward program".In practice, this research provides a measuring tool for online travel agency to test customer’s participation towards reward program, and provides a reference standard for online travel agency to implement reward program successfully. So the enterprise can make reasonable use of limited resources, improve the quality of service, and attract customers to participate in reward program.§6.4 Research limitationsThe paper did some research and exploration of the reward program’s influence factors according to the relevant theories about customer reward program, practical results, combined with the author’s own knowledge and cognitive ability. However, because of the limitation of research conditions and resources, as well as that of the author’s own knowledge and research level, although this paper has obtained some research results, there are still a lot of limitations.When summing up the research results, we need to pay attention to see if the study conclusion can is applicable to other industries, aided by further research. Based on the study of , this paper mainly studies the problem of online travel agency. The benefit is the ability to have a thorough understanding of this industry, but the disadvantages can be overgeneralization. To better understand the problem in other industry, further studies are needed.In this study interview survey of reward program is insufficient. Due to the limitation of time and conditions, this article only carried on four interview investigations. And due to the small number of interviews, the subjectivity of the conclusion is strong.As for the research object, the online questionnaire is used in this paper, so samples are mainly concentrated in young people who often surf the Internet. At the same time, the sample size is not very big (230 valid questionnaires), so it is difficult to guarantee the representation of the sample. Based on the above reasons, the external validity of this research conclusion is difficult to guarantee. In addition, because measurement error is widespread, and the existence of error can affect the relevance of factors, there may have deviations for the results of the study.§6.5 Further studyResults of this study provide a possibility for further study of the influence factors of reward program and customers’ participation of reward program:The study of customers’ participation of reward program, under B to B situation can be a further direction. This paper is based on B to C situation and the research objects are individual customers. And the problem can be different in B to B situation.This paper mainly studies the problem of online travel agency, so further study can choose more industry to verify the research conclusion of this article. Comparing many industries at the same time can explore the influence of product category and consumer's attributes on the research conclusion and provide research support for market segmentation and knowledge integration.Further study should increase the sample size in different levels, increase the number of samples and provide sample representativeness, in order to make the model more universal.References：Bolton Ruth N, Kannan P &Matthew D. Bramlett (2000). Implications of Loyalty Program Membership and Service Experiences for Customer Retention and Value. Journal of the Academy of Marketing Science, 28, 95-108.Bolton R.N., Lemon K.N. & Verhoef P.C. (2004). HYPERLINK ""The theoretical underpinnings of customer asset management: a framework and propositions for future research. Journal of the Academy of Marketing Science, 32, 271-292.Cigliano, Margaret Georgiadis, Datcen Pleasance & Susan Whalley (2000), The Price of Loyalty, The Mckinsey Qumerly, 40, 68-77.Dodds WB, Monroe KB & Grewal D (1991). The effects of price, brand and store information on buyers' product evaluations. Journal of Marketing Research, 28, 307-319.Dowling G.R. & M. Uncles (1997). Do Customer Loyalty Programs Really Work? Sloan Management Review, 38, 71-82.Dick Alan & Kunal Basu(1994).Customer Loyalty: Toward an Integrated Conceptual Framework. Journal of the Academy Marketing Science, 22, 99-113.Drèze & Stephen J. Hoch (1998). Exploiting the Installed Base Using Cross- Merchandising and Category Destination Programs. International Journal of Research in Marketing, 15, 459-471. Grewal D, Monroe K.B. & Krishnan R (1998). The effects of price-comparison advertising on buyers’ perceptions of acquisition value, transaction value, and behavioral intentions. Journal of Marketing, 62, 46-59. Kim Byung Do, Mengze Shi & Kannarl Srinivasan (2001). Reward Programs and Tacit collusion. Marketing Science, 20, 99-120.Kivetz Ran & Itamar Simonson (2002). Earning the Right to Indulge: Effort as a Determinant of Customer Preferences toward Frequency Program Rewards. Journal of Marketing Research, 39, 155-170.Kivetz Ran & Itamar Simonson (2003). The Idiosyncratic Fit Heuristic: Effort Advantage as a Determinant of Consumer Response to Loyalty Programs. Journal of Marketing Research, 40, 454-467.Klemperer, P. (1987a). Markets with Consumer Switching Costs. The Quarterly Journal of Economics, 102, 375-394.Klemperer, P. (1987b). The Competitiveness of Markets with Switching Costs. Journal of Economics, 18, 138-150.Kopalle P & S. A. Neslin (2003). The economic viability of frequency reward programs in a strategic competitive environment. Review of Marketing Science, 1(1), 1-39.Lacey Russell Wayne (2003). Customer Loyalty Programs: Strategic Value to Relationship Marketing. Doctoral dissertation, The University of Alabama.Lewis Michael V (2001). Applications of Dynamic Programming to Customer Management. Evanston, Illinois, Northwestern University. Li Chunqing, Xu Yinfeng & Li Hongyi (2005). An Empirical Study of the Dynamic Customer Relationship Management. Journal of Retailing and Consumer Services, 12(6), 431-441.Li Chunqing, Zhao Ping & Ma Junping (2006). The theoretical underpinnings of the influence of customer reward programs on customer retention: A framework and propositions for future research. Research and Practical Issues of Enterprise Information Systems IFIP International Federation for Information Processing, 205, 615-621.MacSmith D (2002). The US Customer Loyalty Market. Card Technology Today, 14, 13-14Monroe Kent (1990), Pricing: Marketable Decisions (2nd. ed.), New York: McGraw-Hill Publishing Co.O'Brien L & C. Jones (1995). Do Rewards Really Create Loyalty? Harvard Business Review (May-June), 75-82.Oliver L (1999). Whence Consumer Loyalty. Journal of Marketing, 63(4), 33-34.Reichheld Frederick & Earl W. Sasser (1990). Zero defections: quality comes to Services. Harvard Business review, September-October, 105-111.Reichheld Frederick & Thomas Teal (1996). The Loyalty Effect. Harvard Business School Press, Boston, Massachusetts.Rciehheid Frederick (2000). Loyalty Effect---the relationship between Loyalty and Profits. European Business Journal, 12(3), 173-179.Sharp B. & A. Sharp (1997). Loyalty Programs and Their Impact on Repeat-Purchase Loyalty Patterns. International Journal of Research in Marketing, 14 (5), 473-486.Sheth J.N. & Parvatiyar A. (1994). Relationship Marketing: Theory, Methods and Applications. Atlanta Emory University.Stephan A.Butscher (2005). Customer loyalty programmes and Clubs (2end). Dianzigongye press, Beijing.Verhoef Peter C (2003). Understanding the Effect of Customer Relationship Management Efforts on Customer Retention and Customer Share Development. Journal of Marketing, 67(4), 30-45.Woodruff R. B. (1997). Customer Value: The Next Source for Competitive Advantage. Journal of the Academy of Marketing Science, 25 (2), 139-154.Wotruba T. R. (1996). The Transformation of Industrial Selling: Causes and Consequences. Industrial Marketing Management, 25 (5), 327-338. Youjae Yi & Hoseong Jeon (2003). Effects of Loyalty Programs on Value Perception, Program Loyalty, and Brand Loyalty. Journal of the Academy of Marketing Science, 31(3), 229-240.Zeithaml V. A. (1988). Consumer Perceptions of Price, Quality and Value: A Means-End Model and Synthesis of Evidence. Journal of Marketing, 52 (3), 2-21.Zeithaml V.A., Berry L. & Parasuraman A. (1996). The Behavioral Consequences of Service Quality. Journal of Marketing, 60 (2), 31-46. Appendix A (Regression results):Model SummaryModelRR SquareAdjusted R SquareStd. Error of the Estimate1.647a.418.405.82374a. Predictors: (Constant), customer fit, reward value, convenience, possibility, reward typeCoefficientsaModelUnstandardized CoefficientsStandardized CoefficientstSig.BStd. ErrorBeta1(Constant)4.759E-005.054.001.999reward value.432.058.4477.444.000reward type-.035.111-.017-.320.749possibility-.299.090-.205-3.324.061convenience.595.094.3386.309.000customer fit.322.105.1893.078.002a. Dependent Variable: choiceModel SummaryModelRR SquareAdjusted R SquareStd. Error of the Estimate1.822a.675.666.61678a. Predictors: (Constant), perceived value, customer fit, convenience, reward value, reward type, possibilityCoefficientsaModelUnstandardized CoefficientsStandardized CoefficientstSig.BStd. ErrorBeta1(Constant)6.624E-006.041.0001.000reward value.328.073.1864.464.000reward type.072.081.042.895.372possibility.217.046.2254.681.000convenience.314.067.2154.659.000customer fit.013.063.0063.155.007perceived value.738.056.63713.287.000a. Dependent Variable: choiceModel SummaryModelRR SquareAdjusted R SquareStd. Error of the Estimate1.871a.695.693.68150a. Predictors: (Constant), perceived valueCoefficientsaModelUnstandardized CoefficientsStandardized CoefficientstSig.BStd. ErrorBeta1(Constant)-2.190E-005.045.0001.000perceived value.893.049.77118.287.000a. Dependent Variable: choiceModel SummaryModelRR SquareAdjusted R SquareStd. Error of the Estimate1.705a.566.552.74230a. Predictors: (Constant), customer fit, reward value, convenience, possibility, reward typeCoefficientsaModelUnstandardized CoefficientsStandardized CoefficientstSig.BStd. ErrorBeta1(Constant)5.554E-005.049.001.999reward value.363.085.3494.263.000reward type.339.094.2303.594.323possibility.291.052.2395.572.000convenience.020.081.016.248.805customer fit-.065.100-.036-.657.512a. Dependent Variable: perceived valueAppendix B：Questionnaire about Ctrip’s reward programPurpose：For thesis writing, we stage an investigative questionnaire. If you have interest in it, please help us to finish the questionnaire and write your ideas. The answers you give will be kept confidential and used seriously for research purposes only.?Thank you.Part 1: First, according to your own feelings on Ctrip rewards program, please determine the extent you agree or oppose for each question, and fill in the corresponding figures in parentheses. 1234567Totally disagreeDisagreePartlydisagreeNot agree nor disagreePartlyagreeAgreeTotally agree1234567Relative to my spending, I was satisfied with the return pared with my spending amount, I think the discounted value of the goods or services I obtained in return are too low. I think the value of the rewards provided by Ctrip is higher than other similar sites.The reward program provided by Ctrip offers a large variety of reward.I am very concerned about what products can I get as a reward.I would prefer to get products or services directly related to the website like discounted hotel rates.I would prefer to get products or services directly related to the website like Daily Necessities.I would prefer to get tangible things like vouchers& commodities.It is easily for me to reach the amount of consumption required by Ctrip (the minimum amount of consumption to participate in reward program) After using this site, I will soon be able to participate in reward program.I prefer permanent reward program. I think it is better for reward program to have a pared to delayed reward, I prefer to get immediate reward or pared to immediate but lower reward, I prefer high-value rewards that need time to cumulative.I think reward program provided by Ctrip is easy to check. (Like Points query) I think reward program provided by Ctrip is easy to use. (Like Redemption)In contrast, I believe it is appropriate that different accumulate points can be converted to a products have different values.I think the higher the spending amount is, the greater the reward ratio should be. If my points are relatively high, I hope I can get special treatment that others cannot have.I think my membership can be recognized and pared to other traveling website I prefer reward program provided byCtrip.The reward product form Ctrip‘s rewards program is exactly what I want. As far as the money, time, effort I spent, the rewards program is worth it.I think it is a good choice to participate in this reward program.I really like Ctrip’s reward programs.The rewards program will encourage me to spend more.I would recommend the program to others.I have a strong preference towards Ctrip’s reward programs.I am willing to participate in the rewards program provided by Ctrip:Yes b. NoPart 2: Background InformationGender: a. male b. femaleAge: A. under18 B. 18-30 years old C. 31-40 years old D. 41-50 years old E. over51Do you consider your income per month between: ＜2000 Yuan b. 2001-5000 Yuan c. 5001-8000 Yuan d. 8001-10000 Yuane．＞10000 Yuan4．Education situations (already obtained or are studying): under junior college (excluding junior college) B. junior college C. Bachelor D. Master E. doctor and above 5．Years of using Ctrip: A. under 0.5year B. 0.5- 1 year C. 1-2 years D. 2-3 years E. over 3 years ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download