Kuo, Y. -F. and Wu, C. -M., Satisfaction and post-purchase intentions with service recovery of online shopping websites: Perspectives on perceived justice and emotions, International Journal of Information Management, doi:10.1016/j.ijinfomgt.2011.09.001.
Satisfaction and Post-Purchase Intentions with Service Recovery of Online Shopping Websites: Perspectives on Perceived Justice and Emotions
Abstract
This study explores post-recovery satisfaction and post-purchase intentions with service recovery of online shopping customers from the perspectives on perceived justice and emotions based on twenty scenarios that combine five service failures and four recoveries. Analytical results indicating that in addition to enhancing post-purchase intentions and post-recovery satisfaction among customers, distributive justice increases positive emotions and decreases negative ones. Additionally, procedural justice enhances post-recovery satisfaction as well as increases positive emotions and decreases negative ones, while interactional justice only increases post-recovery satisfaction of customers. Moreover, while positive emotions increase post-recovery satisfaction, and vice versa. Post-recovery satisfaction positively affects post-purchase intentions. Results of this study provide a valuable reference for operators and managers of online shopping websites.
Keywords: Post-recovery satisfaction; Post-purchase intention; Perceived justice; Emotions; Online shopping websites
1. Introduction
The inseparable and intangible nature of services makes it difficult for service providers to avoid service failures during service delivery. Customers experiencing a service failure may convey their dissatisfaction to others through negative word-of-mouth and a negative sentiment towards the offending service provider, adversely impacting customers, profits, and even company reputation (Bitner et al., 2000). Most customers whom encounter a service failure anticipate service recovery (Goodwin & Ross, 1992; Bloggett et al., 1997; Holloway & Beatty, 2003). Via effective recovery strategies, service providers can still appease unsatisfied customers, increase the customer retention rate (McCollough et al., 2000) and even foster a long lasting relation with dissatisfied customers (Kelley et al., 1993), ultimately making them loyal ones (Boshoff, 1997).
Despite the numerous studies on service failure and service recovery, most focus on physical providers of services. Rapid advances of e-commerce in recent years have increased the prevalence of online shopping as a vital business model. Even so, online shopping failure is still inevitable. When service failure occurs, service providers must take appropriate recoveries to return dissatisfied customers to a state of satisfaction, enhance customer
retention rate and even help build long-term relationships that make customers loyal. Therefore, online shopping service failure and recovery strategies have become important issues for academics and practitioners (Holloway & Beatty, 2003). While many researchers (e.g., Holloway & Beatty, 2003; Forbes et al., 2005; Collier & Bienstock, 2006; Holloway and Beatty, 2008; Wang et al., 2010; Kuo et al., 2011) have extended service failure and service recovery research in cyberspace, most have focus on the typology of online retailing service failures and recoveries as well as seldom explore the relationships between customer post-recovery satisfaction and post-purchase intentions with service recoveries in the context of online retailing.
Justice theory has received many attentions as a theoretical framework for service recovery studies (Sparks & McColl- Kennedy, 1998; Tax et al., 1998; Smith et al., 1999; Hoffman & Kelley, 2000; Maxham III, 2001; Wirtz & Mattila, 2004; Schoefer, 2008; Ha & Jang, 2009). Prior studies have pointed out that perceived justice is identified as a key cognitive influence in the formation of subsequent customer satisfaction and post-purchase intentions in service recovery situations (McColl-Kennedy & Sparks, 2003; Schoefer, 2008; del Río-Lanza et al., 2009; Ha & Jang, 2009). Additionally, customer emotions triggered by service recovery encounters affect satisfaction (Smith & Bolton, 2002; Chebat & Slusarczyk, 2005; Schoefer & Ennew, 2005; del Río-Lanza et al., 2009). Although emotions are also considered as playing an important role in customers evaluating service failures and perceptions of service recovery experiences (Bagozzi et al., 1999; Weiss et al., 1999; Schoefer & Diamantopoulos, 2008), empirical investigations of emotional responses of customers to service recovery encounters are seldom evaluated (del Río-Lanza et al., 2009). Additionally, perceived justice usually represents a cognitive appraisal concept, whereas its effects have been shown to be both emotional and behavioral following service recovery experiences (Chebat & Slusarczyk, 2005; Schoefer & Ennew, 2005). The integration of emotions within the perceived justice of service recovery seems a necessary step to better understand what drives customers’ evaluative judgments in a recovery situation. In the e-shopping context, to the best of our knowledge, no empirical study has specifically addressed how customers’ cognitive justice perceptions and emotional responses to service recovery encounters combine to influence subsequent customer satisfaction and post-purchase intentions with service recovery. Therefore, this study investigates post-recovery satisfaction and post-purchase intentions with service recovery of online shopping customers from the perspectives on perceived justice and emotional responses to fill this important gap. Based on those results, suggestions are provided to operators and managers of online shopping websites.
The rest of this paper is organized as follows. Section 2 reviews relevant literature on service failures and recoveries of online shopping and constructs of this study, and then develops research hypotheses. Next, Section 3 describes the research methodology and research design. Additionally, Section 4 summarizes the analytical results. Discussion and
conclusions are finally drawn in Section 5.
2. Literature review and hypotheses development
This section reviews the theoretical framework for this study, including factors such as online retail service failure and recovery, post-purchase intentions, perceived justice, emotions, and post-recovery satisfaction as well as develops the research hypotheses as follows.
2.1 Online retail service failure and recovery
As a real or perceived service-related calamity, service failures occur during interaction with a customer’s experience with a firm (Maxham III, 2001). Service recovery refers to how service providers respond to a service failure (Gronroos, 1988). As for classifying service failures, the typology of Bitner et al. (1990) has received the most attention in subsequent research (e.g., Kelley et al., 1993; Hoffman et al., 1995; Forbes et al., 2005), including three main groups such as employee response to service delivery system failures, employee response to customer needs and requests, and unprompted and unsolicited employee actions. Service recovery strategies can be classified tangibly into psychological and tangible strategies (Miller et al., 2000). Psychological recovery strategies refer to actions that can directly improve customer psychological dissatisfaction, such as apology and explanation, while tangible recovery strategies provide tangible compensation, such as a free service, refund, gift, discount, and coupon, to reduce customer practical loss.
With the popularity of online shopping, many researchers have engaged in service failure research. For instance, Holloway and Beatty (2003) classified service failures in online shopping into delivery, web site design, customer service, payment, security, as well as miscellaneous and others. They further classified the dimensions of dissatisfiers and satisfiers in an online environment into website design, interaction, fulfillment/reliability, customer service, and security/privacy (Holloway & Beatty, 2008). The dimension implying the greatest dissatisfaction is fulfillment/reliability, while the dimension implying the most satisfaction is website design/interaction. By surveying customers on service failure and recovery in online shopping, Forbes et al. (2005) classified service failures of online retailers into two groups and ten categories, including (1) response to service delivery system/product failure: slow/unavailable service, system pricing, packaging errors, out of stock, product defect, bad information, and web site system failure; and (2) response to customer needs and requests: special order/request, customer error, and size variation, as well as identification of eleven types of service recovery including discount, correction, correction plus, replacement, apology, refund, store credit, unsatisfactory correction, failure escalation, nothing, and replacement at brick-and-mortar. Kuo et al. (2011) proposed three groups and eighteen categories of service failures as well as ten recovery strategies to be aimed at online auction. Compared with shopping websites (Forbes et al., 2005), online auction revealed a new group
of failures, “unprompted and unsolicited seller actions”, and four different categories, including “policy failure”, “hold disaster”, “alterations and repairs” and “gap between expectation and perception”. However, “website system failure” in shopping websites is absent in an online auction. Additionally, online auction recoveries are similar to the recoveries used in shopping websites.
2.2 Post-purchase intentions
Post-purchase intentions have been frequently used as a basis for predicting customers’ future behaviors (Kuo et al., 2009). It can be defined as customers’ intentions to repurchase products or services from the same retailer and spread their experience of buying and using the product or service to their friends (Zeithaml et al., 1996; Wang et al., 2006). Post-purchase intentions can be classified into economic behavioral intentions and social behavioral intentions (Smith et al., 1999). Economic behavioral intentions refer to customers’ behavioral reactions in the financial aspect, such as repurchase intention (Anderson & Mittla, 2000; Maxham III, 2001; Maxham III & Netemeyer, 2002, 2003). Repurchase intention is an expression of customer loyalty, which is critical if the vendor is to succeed (Kim & Son, 2009; Qureshi et al., 2009; Zhang et al., 2011). While social behavioral intentions refer to cognitive reactions of customers to the delivery of services for service providers, such as complaining behavioral intentions (Tax et al., 1998) and word-of-mouth communication intentions (Maxham III, 2001; Maxham III & Netemeyer, 2002, 2003).
2.3 Perceived justice
Justice theory originates from social exchange (Homans, 1961) and equity theory (Adams, 1965). In an exchange, the cost or price of product or service must be equivalent to the gains. If the cost is higher than the gains, actions can be taken to reduce the level of unfairness. Konovsky (2000) mentioned that perceive justice is indispensable to research on how individuals react to a conflict. Because service failures are typical conflicts, service recovery actions taken to cope with a service failure can be evaluated based on perceived justice.
In practice, recovery strategies of a firm are normally evaluated using the three components of justice, namely distributive justice, procedural justice, and interactional justice (Maxham III & Netemeyer, 2003; Wirtz & Mattila, 2004; Chebat & Slusarczyk, 2005; Schoefer & Ennew, 2005; Collier & Bienstock, 2006; Schoefer, 2008; del Río-Lanza et al., 2009). “Distributive justice” refers to whether the failed customer has received monetary compensation. Most failed customers can perceive distributive justice of a recovery action after they have received a discount, coupons, refund, free giveaways or alternative goods as compensation from the offending service provider (Goodwin & Ross, 1992; Blodgett et al., 1997; Tax et al., 1998; Hoffman & Kelley, 2000). Distributive justice can be evaluated by
customer perception of the fairness, equality, necessity, and value of the compensation (Blodgett et al., 1997; Tax et al., 1998; Smith et al., 1999; Maxham III & Netemeyer, 2003; Wirtz & Mattila, 2004).
Defined as customer perceptions of the recovery process, procedural justice focuses on the flexibility and efficiency of the recovery policies or rules. Failed customers can perceive procedural justice of a recovery action when the offending company admits the failure, attempts to rectify the mistake timely, and adjusts its recovery strategy in line with customer demands. Procedural justice can generally be evaluated as to whether customers can freely express their opinions, recovery efficiency of the offending company, dominance over the outcome, easiness of making complaints, flexibility, instantaneity, transparency of the recovery process, and appropriateness of the recovery action or policy (Blodgett et al., 1997; Tax et al., 1998; Smith et al., 1999; Maxham III & Netemeyer, 2003; Wirtz & Mattila, 2004; Chebat & Slusarczyk, 2005).
Interactional justice refers to the extent of fairness in which service providers communicate with and treat failed customers. Most failed customers perceive interactional justice of a service recovery action when the offending service providers are willing to communicate with them courteously, honestly, and empathetically while attempting to solve a problem and communicate with customers (Goodwin & Ross, 1992; Tax et al., 1998; Maxham III & Netemeyer, 2003; Wirtz & Mattila, 2004). Interactional justice is generally evaluated by reliability, clear explanation of the problem, sincerity, apologetic attitude, communication, politeness, respect, detailed attention to problems, willingness to hear complaints, and resolve to solving the problem (Blodgett et al., 1997; Smith et al., 1999; Wirtz & Mattila, 2004).
Service recovery is essential owing to the inevitability of service failures. Improperly handling service complaints may yield negative word-of-mouth and low repurchase intentions (Tax & Chandrashekaran, 1992). Additionally, customers do not easily forget or forgive unfair handling of service failures (Seiders & Berry, 1998). However, a situation in which service providers offer refund or discounts to redress a service failure and handle a failure politely allows them to induce customers’ word-of-mouth and increase their repurchase intention (Blodgett et al., 1997). According to research on service failures in online retailing, a higher perceived procedural justice of service recovery implies a higher motivation for positive word-of mouth, while a higher perceived interactional justice implies a higher repurchase intention (Maxham III & Netemeyer, 2003). Ha and Jang (2009) also suggested that perceived justice positively affects the post-purchase intentions of customers. Hypothesis 1 is thus proposed as follows:
H1a Perceived distributive justice positively influences post-purchase intentions in online shopping websites.
shopping websites.
H1c Perceived interactional justice positively influences post-purchase intentions in online shopping websites.
2.4 Emotions
Emotions refer to affective states of an individual that are specific to certain events or one’s own thoughts (Bagozzi et al., 1999). Studies (Watson et al., 1988; Cacioppo & Gardner, 1993; Cacioppo et al., 1997) show that emotions consist of two dominant dimensions including positive and negative. Positive emotion is related to contentment, happiness, love, and pride, whereas negative emotion is related to anger, fear, sadness, and shame (Laros & Steenkamp, 2005). Positive and negative emotions are independent across a range of time frames (Watson, 1988) and have distinct and asymmetrical effects on behavior (Cacioppo & Gardner, 1993; Cacioppo et al., 1997). Customers tend to become intensely emotional during service recovery, emotional responses that will largely determine their future relation with the service provider (Smith & Bolton, 2002). Emotions are increasingly viewed as essential in researching how customers evaluate service failure and recovery (Bagozzi et al., 1999; Weiss et al., 1999; Schoefer, 2008; Schoefer & Diamantopoulos, 2008). Schoefer and Dianantopoulos (2008) mentioned that emotions are seldom addressed in service recovery research, and a valid and reliability measurement scale for emotions is lacking. That study developed experienced emotions during service recovery encounters (ESRE) scale to evaluate emotional responses among customers during and after service recovery.
Customers often experience emotions when perceiving the justice of a service recovery strategy (Schoefer, 2008). Many researchers (e.g., Weiss et al., 1999; William, 1999; Chebat & Slusarczyk; 2005; Schoefer & Ennew, 2005; Schoefer, 2008; del Río-Lanza et al., 2009) studied post-recovery customer emotions under the framework of perceived justice. Their findings suggest that low perceived justice leads to highly negative emotions (anger, fury, and unhappiness) and low positive emotions (happiness, pleasure, and joy) (Schoefer & Ennew, 2005). Customers may display furious emotions if they are displeased with the recovery outcome or perceive injustice of the recovery process (Weiss et al., 1999). Increasing the level of distributive justice raises pleasant emotions, while decreasing that of distributive justice makes customers angry and disappointed (William, 1999). Chebat and Slusarczyk (2005) also found a similar phenomenon in service recovery measures adopted in the banking industry. Increasing distributive justice, procedural justice, and interactional justice can reduce negative emotions in customer, while increasing distributive justice and interactional justice can heighten positive emotions in customers. Additionally, a higher procedural justice that customers perceive imply greater positive emotions (Schoefer, 2008) and lower negative ones (del Río-Lanza et al., 2009). Hypothesis 2 is thus proposed as follows:
websites.
H2b Perceived distributive justice negatively influences negative emotions in online shopping websites.
H2c Perceived procedural justice positively influences positive emotions in online shopping websites.
H2d Perceived procedural justice negatively influences negative emotions in online shopping websites.
H2e Perceived interactional justice positively influences positive emotions in online shopping websites.
H2f Perceived interactional justice negatively influences negative emotions in online shopping websites.
2.5 Post-recovery satisfaction
In this customer-oriented era, all enterprises pursue customer satisfaction as essential to gaining sustainable growth and competitive advantages (Deng et al., 2010; Udo et al., 2010). Customer satisfaction also plays a crucial role in service recovery and directly affects customers’ attitude and intentions (Holloway et al., 2005). Post-recovery satisfaction refers to customers’ overall satisfaction with the secondary service (remedial action) of a service provider after a service failure; it is different from customers’ satisfaction with the first service encounter. Therefore, post-recovery satisfaction also refers to as secondary satisfaction (Tax et al., 1998; Smith et al., 1999; McCollough et al., 2000; Harris et al., 2006).
Post-recovery satisfaction has been extensively adopted to evaluate the level of perceived justice (Maxham III & Netemeyer, 2002, 2003; Mattila & Patterson, 2004; Schoefer, 2008; del Río-Lanza et al., 2009). While surveying the service recovery of banks, Maxham III and Netemeyer (2002) found that post-recovery satisfaction of customers increases with a rising of perceived distributive justice and interactional justice. This finding is consistent with the study of Mattila and Patterson (2004) on service recovery measures in different cultures. While examining online shopping of electronic devices, Maxham III and Netemeyer (2003) found that distributive justice and procedural justice increase post-recovery satisfaction. Schoefer (2008) and del Río-Lanza et al. (2009) suggested that distributive justice, procedural justice, and interactional justice increase post-recovery satisfaction. A situation in which service providers view the components of perceived justice from customer perspective allow them to more thoroughly understand customer perceptions of justice and enhance post-recovery satisfaction among customers. Hypothesis 3 is thus proposed as follows:
H3a Perceived distributive justice positively influences post-recovery satisfaction in online shopping websites.
H3b Perceived procedural justice positively influences post-recovery satisfaction in online shopping websites.
H3c Perceived interactional justice positively influences post-recovery satisfaction in online shopping websites.
Various customer emotions may arise during consumption (including before, during, and after consumption); such emotions influence their assessment (such as satisfaction) of the service provider (Oliver, 1997). Clarifying emotional responses among customers thus facilitates service providers in improving post-recovery satisfaction of customers (Smith & Bolton, 2002) and recovery from service failures (del Río-Lanza et al., 2009). Schoefer and Ennew (2005) asserted that customer emotions are an antecedent of post-recovery satisfaction. Other studies also suggest that positive emotions facilitate increased post-recovery satisfaction (Schoefer, 2008), while negative emotions only reduce it (Smith & Bolton, 2002; Schoefer, 2008; del Río-Lanza et al., 2009). Hypothesis 4 is thus proposed as follows:
H4a Positive emotion positively influences post-recovery satisfaction in online shopping websites.
H4b Negative emotion negatively influences post-recovery satisfaction in online shopping websites.
Post-purchase intentions can be viewed as a result of customer satisfaction (Anderson & Mittal, 2000). While examining the service failures of online retailers, Collier and Bienstock (2006) found that customer dissatisfaction with recovery measures affects their future behavioral intentions, such as switching and negative word-of-mouth. However, customers having experienced satisfactory recovery from a service failure not only engage in positive word-of-mouth communications (Holloway et al., 2005) but also have persistent and more faith in online shipping (Montoya-Weiss et al., 2003) and higher repurchase intentions (Holloway et al., 2005). Hypothesis 5 is thus proposed as follows:
H5 Post-recovery satisfaction positively influences post-purchase intentions in online shopping websites.
Fig. 1 shows the model relationships and hypotheses.
3. Research methodology
3.1 Experiment and questionnaire design
Simulated service failure and recovery scenarios were used in our survey. Via these virtual scenarios, respondents’ reactions in each service failure and recovery context can be understood as if they were realistically involved in it. In contrast with the conventional method of asking respondents to answer questions based on their memory, this method can more effectively avoid biases caused by fading of memory, reasonable tendency, and consistency factors, subsequently yielding a representative sample more efficiently and at a lower cost (Smith et al., 1999). Given the infrequency of service failures, simulating the scenarios allows us to more accurately present common service failures and recovery strategies of online shopping websites.
This study designed the scenarios by considering various service failures and recovery
strategies and manipulating variables. According to related literature (Holloway & Beatty, 2003; Forbes et al., 2005; Holloway & Beatty, 2008), this study classified common service failures of online shopping websites into delivery failure, system failure, product quality failure, website security failure, and customer support failure. Based on the tangibility of recovery strategies, four recovery strategies were identified, including tangible recovery strategy, psychological recovery strategy, strategy with neither tangible nor psychological recovery, and strategy with both tangible and psychological recovery. Totally, 20 scenarios were designed based on the above five failures and four recovery strategies. Each scenario was designed based on previous literature and practical circumstances. Given that the ratios of female and male online shoppers vary across different categories of goods (Chan, 2008), this study attempted to reduce the interference of product information differences by selecting a good from the top five popular wears products among both female and male shoppers (Chan, 2008). Finally, a shoes brand priced at NT$999 was selected as the product for testing in our scenarios.
An attempt was also made to verify the adequacy of experimental design and the amount of tangible compensation by inviting twenty experts and long-term online shoppers to evaluate not only the highest possible amount of compensation that online shopping websites would offer for the designed scenarios, but also the possibility that online shopping websites would also adopt the same measure as described in each of our scenarios by using a scale ranging from 1-5 (1: highly impossibly; 5: highly possibly). According to their evaluation results, the average possibility of the 20 scenarios was 3.825, implying the high possibility that online shopping websites adopt the same measure as described in each of our scenarios (all with a possibility value significantly greater than 3, p<0.001). That is, these experts and online shoppers conferred not only that the simulated scenarios would very likely occur in practice, but also that our manipulation of service failures and recovery strategies in each
scenario was appropriate. In terms of determining the compensation amount, most of the participants agreed that online retailers were more likely to offer NT$100 as a compensation for their failure in selling or delivering a product priced below NT$999. Besides, they also performed a subjective classification of service failures and recovery strategies. According to those results, our attribution of the five service failures reached an accuracy 90%, while that of the four recovery strategies reached 95%. Above findings suggest that our experimental design conformed to our definition of manipulation.
The draft questionnaire consisted of valid and reliable questions extracted from previous literature. To ensure the adequacy and clarity of each question and identify potential problems of the questionnaire, this work invited three experts to review the questionnaire, helping us to finalize the questionnaire. The final questionnaire consisted of three parts. The first part surveyed the basic data of the respondents individually using nominal scales. The second part evaluated the perceived realism of the scenarios using two multi-item scales ranging from 1 (highly impossible) to 5 (highly possibly). First, respondents were asked how strongly that these incidents could happen in real life described in the scenarios. Second, respondents were asked how possible that online shopping websites would adopt the same measure as described in each of our scenarios. Last part evaluated the subject’s perception of each construct in the model. A five-point Likert scale ranging from 1 (very disagreed) to 5 (very agreed) was adapted for the above instrument. Table 1 lists the research constructs and items included in the questionnaire. The Appendix contains complete scenarios of this study.
Table 1 Constructs and items included in the questionnaire
Construct Item Measurement References
Perceived Justice (PJ) Perceived Distributive Justice (PDJ)
PDJ1 Considering the trouble caused and the time lost, the compensation I received from this shopping website was acceptable.
del Río-Lanza et al.(2009) PDJ2 This shopping website took good
compensation measures to solve the problem.
PDJ3 This shopping website's efforts were sufficient to offer a satisfactory compensation.
PDJ4 I think this shopping website was quite fair when compensating me for the problem that occurred.
PDJ5* In general, this shopping website was able to compensate me adequately to solve the problems it had in the delivery of the service.
Perceived Procedural
Justice (PPJ)
PPJ1 I think my problem was resolved in the right way.
PPJ2* I think this shopping website has good policies and practices for dealing with problems.
PPJ3 Despite the trouble caused by the problem, this shopping website was able to respond adequately.
PPJ4 This shopping website proved flexible in solving the problem.
PPJ5* This shopping website tried to solve the problem as quickly as possible.
Perceived Interactional
Justice (PIJ)
PIJ1 The employees in this shopping website showed interest in my problem.
PIJ2 The employees in this shopping website did everything possible to solve my problem. PIJ3* The employees in this shopping website
were honest when dealing with my problem. PIJ4 The employees in this shopping website
proved able and to have enough authority to solve the problem.
PIJ5* The employees in this shopping website dealt with me courteously when solving the problem.
PIJ6 The employees in this shopping website showed interest in being fair when solving the problem.
PIJ7* The treatment and communication with employees of this shopping website to solve the problem were acceptable.
Emotions (EM)
Positive Emotion (PEM)
PEM1 I feel happy after service recovery encounters of this shopping website.
Schoefer and Ennew (2005); Schoefer and Dianantopoulos (2008)
PEM2 I have a warm feeling after service recovery encounters of this shopping website.
PEM3 I am being valued after service recovery encounters of this shopping website.
Negative Emotion (NEM)
NEM1 I feel angry after service recovery encounters of this shopping website.
NEM2 I am in a bad mood after service recovery encounters of this shopping website.
NEM3 I feel upset after service recovery encounters of this shopping website.
Post-Recovery Satisfaction (PRS)
PRS1 Overall, I felt that this service recovery encounters would have been good.
Goodwin and Ross (1992) PRS2 Overall, I was satisfied with the way this
complaint was resolved.
PRS3 Overall, I was pleased with the service recovery encounters I experienced.
Post Purchase Intentions (PPI)
PPI1 I intend to purchase from this shopping website next time.
Kuo et al. (2009)
PPI2 I will recommend this shopping website to my friends or relatives.
PPI3 I intend to continue to purchase from this shopping website in the future.
*: Items have been dropped after EFA
3.2 Research subjects and sampling method
According to Market Intelligence & Consulting Institute (MIC) of Institute for Information Industry (2008), most online shoppers in Taiwan range between 20-29 years old
(52.8%), and are university or graduate school students. Hence, in this study, university and graduate school students were selected as our subjects; responses from volunteers were also obtained to increase the sample reliability. The upcoming survey details were posted on the electronic BBS and announced in classes of two universities. Interested participants filled out the questionnaire at designated locations. Additionally, our sample size was increased by holding a lucky draw of gift certificates. A total of 289 students participated in the study and were randomly assigned to one of the 20 scenarios. Due to missing important data and outliers, 252 responses were valid. The age of the participants ranges between 18 and 26, with a heavy concentration (54.8%) between 20 and 21, and 48.5% were males and 51.5% females.
4. Results
4.1 Manipulation check
A mean possibility of the incidents occurrence (4.139) showed statistically significant great than 4 (p<0.010), implying that the respondents perceived the scenarios to be highly realistic. A mean possibility of the recoveries (3.516) showed statistically significant great than 3 (p<0.001), implying the high possibility that online shopping websites adopt the same measure as described in each of our scenarios. That is, these participants conferred not only that the simulated scenarios would very likely occur in practice, but also that our manipulation of service failures and recovery strategies in each scenario was appropriate.
4.2 Verification of the proposed model and hypotheses
This study employed structural equation modeling (SEM) to verify the proposed model and hypotheses. Notably, multivariate normally is required for most SEM approaches. According to those results, most variables used deviated significantly from normality, thus limiting the validity of maximum likelihood estimation approaches. Given these distributional characteristics, this study adopted a partial least squares (PLS) estimation approach since it makes no distributional assumptions. Hence, both the measurement model and the structural model were estimated using PLS (Hulland, 1999; Walczuch et al., 2007). The software SmartPLS version 2.0 (Ringle et al., 2005) was adopted for the analysis. The bootstrapping procedure was implemented to provide additional confidence that the results are not sample-specific by using repeated random samples drawn from the data. In this instance, the bootstrapping procedure was repeated until it reached 500 bootstrap samples (Chin, 1998). The analytical results of testing the measurement model and structural model are summarized as follows.
1. Measurement model analysis
Past studies (Agarwal & Prasad, 1998; Farrell, 2010) suggested using both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) for assessing construct validity. An EFA was first conducted to purify the scale and assess the dimensionality of the constructs
(perceived justice and emotional responses) used, as shown in Table 2. Before EFA, two measures were used to test the appropriateness of factor analysis. The Kaiser-Meyer-Olkin (KMO) overall measure of sampling adequacy (MSA) was 0.919 and 0.856 for perceived justice and emotional responses, respectively, and thus fell within the acceptable level. Additionally, significance levels of 0.000 were obtained for both constructs using Bartlett’s sphericity test, showing a significant correlation among the variables (Hair et al., 1998). For EFA, Principal Component Analysis, with varimax rotation and eigenvalue greater than 1 and factor loadings exceeding 0.4 was used (Kaiser, 1958). Additionally, any items which were cross-loaded on two factors with a factor loading greater than 0.4 were removed. For item analysis, corrected item to total correlation coefficient of less than 0.40 was used as the criterion for item deletion, and whether the removal of the item could significantly enhance the reliability of the questionnaire was considered. This process was iterated until an optimal result was obtained. Six cross-loading items (PDJ5, PPJ2, PPJ5, PIJ3, PIJ5, and PIJ7) were dropped from the items pool. The Cronbach’s α coefficients ranged from 0.770 to 0.862, and the cumulative variance explained was 69.01% and 78.63% for perceived justice and emotional responses, respectively.
Reliability, convergent validity, and discriminant validity of the scale were examined using CFA. According to Tables 3 and 4, the composite reliability (CR) values of all constructs exceeded the recommended level of 0.8, demonstrating the reliability of all measures (Bagozzi & Yi, 1988; Hair et al., 1998). Moreover, the completely standardized factor loadings all reached the level of significance; all constructs had a CR exceeding 0.8, and the average variance extracted (AVE) values for all constructs exceeded the suggested threshold value of 0.50, thus demonstrating the convergent validity of the scale (Fornell & Larcker, 1981; Anderson & Gerbing, 1988). AVE of each latent variable was larger than the squared correlation between each pair of latent variables. Hence, the discrimination validity was adequate (Fornell & Larcker 1981). Overall, the evidence of good reliability, convergent validity, and discriminant validity indicates the adequacy of the measurement model in testing the structural model at a subsequent stage.
Table 2 EFA results of perceived justice and emotions
Construct Item Factor Loading Item-to-total correlation Eigenvalue Variance explained (%) Cronbach’s α PJ PDJ PDJ1 0.800 0.704 2.846 25.872 0.862 PDJ2 0.795 0.766 PDJ3 0.769 0.756 PDJ4 0.698 0.622 PPJ PPJ3 0.784 0.583 2.121 19.282 0.770 PPJ4 0.730 0.648 PPJ1 0.657 0.587 PIJ PIJ1 0.785 0.662 2.624 23.851 0.818 PIJ2 0.715 0.712 PIJ6 0.708 0.661 PIJ4 0.695 0.540
Cumulative variance explained: 69.01%
EM PEM PEM3 0.861 0.677 2.392 39.859 0.845 PEM2 0.838 0.744 PEM1 0.747 0.717 NEM NEM6 0.885 0.684 2.326 38.771 0.861 NEM5 0.842 0.803 NEM4 0.723 0.734
Table 3 Standardized factor loadings, SMC and CR of the measurement model
Construct Item Factor
loading t-value CR PJ PDJ PDJ1 0.836 38.765*** 0.906 PDJ2 0.878 52.890*** PDJ3 0.874 59.714*** PDJ4 0.774 27.924*** PPJ PPJ1 0.857 57.325*** 0.866 PPJ3 0.772 18.147*** PPJ4 0.848 41.092*** PIJ PIJ1 0.807 31.294*** 0.881 PIJ2 0.859 48.070*** PIJ4 0.738 20.668*** PIJ6 0.818 29.200*** PEM PEM1 0.874 42.856*** 0.906 PEM2 0.897 72.649*** PEM3 0.851 34.447*** NEM NEM1 0.894 53.488*** 0.915 NEM2 0.922 84.494*** NEM3 0.835 31.203*** PRS PRS1 0.909 73.818*** 0.931 PRS2 0.924 88.469*** PRS3 0.879 47.560*** PPI PPI1 0.925 70.030*** 0.948 PPI2 0.922 102.677*** PPI3 0.933 82.816*** Note: CR= 2 2 (Σλ)
[(Σλ) +Σ(θ)];λ=factor loading; θ=measurement error of each measured variable. *** p<0.001
Table 4 The mean, standard deviation, and inter-variable correlations
Mean S. D. AVE PDJ PPJ PIJ PEM NEM PRS PPI
PDJ 3.184 0.758 0.708 0.841a PPJ 3.442 0.717 0.683 0.634 0.826 PIJ 3.440 0.750 0.651 0.621 0.656 0.807 PEM 3.082 0.870 0.764 0.719 0.607 0.516 0.874 NEM 2.978 0.864 0.782 -0.516 -0.516 -0.448 -0.666 0.884 PRS 3.241 0.828 0.817 0.766 0.696 0.635 0.743 0.663 0.904 PPI 2.960 0.911 0.859 0.629 0.560 0.514 0.616 -0.514 0.676 0.927
Note: aDiagonal elements show the square root of average variance extracted (AVE). AVE=
2
2 (Σλ )
[(Σλ )+Σ(θ)]; λ=factor loading; θ=measurement error of each measured variable.
2. Structural model analysis
The estimated path coefficients of the structural model were examined to evaluate the hypotheses. Figure 2 shows the standardized path coefficients, t values and coefficients of determination (R2) of the latent variables. Those results (Table5 and Table 6) are as follows.
-0 .27 6* ** (-3 .89 2) 0.2 26** (3.1 84) 0.1 88 ** (3 .3 97) -0.2 22 ** * (-4.5 87 ) -0 .28 6* ** (-4 .1 19 ) 0.2 50 *** (3.9 47 ) 0.0 06 (0.0 88) 0.11 2* (2.05 8) -0.089 (-1.24 5)
(1) Effect of perceived justice on post-purchase intentions
Among the three components of perceived justice, only distributive justice (γ=0.226, t=3.184) increases positively influences post-purchase intentions significantly, while procedural justice (γ=0.107, t=1.259) and interactional justice (γ=0.054, t=0.723) affect post-purchase intentions insignificantly. Thus, only H1a is supported.
(2) Effect of perceived justice on emotions
Our findings on how perceived justice and emotions are related suggest that distributive justice has significantly positive effects on positive emotions (γ=0.557, t=9.400) and significantly negative effects on negative ones (γ=-0.286, t=-4.119). Hence, H2a and H2b are supported. That is, a higher perceived distributive justice implies more positive emotions and less negative emotions. Procedural justice also positively affects positive emotions (γ=0.250, t =3.947) and negatively affects negative ones significantly (γ=-0.276, t=-3.892). Therefore, H2c and H2d are supported; suggesting that a higher perceived procedural justice can increase positive emotions and less negative ones. Interactional justice significantly affects neither positive emotions (γ=0.006, t=0.088) nor negative emotions (γ=-0.089, t=-1.245). Thus, H2e and H2f are not supported. Among the three components of justice, distributive justice influences positive emotions and negative emotions the most.
(3) Effect of perceived justice on post-recovery satisfaction
As for how perceived justice and post-recovery satisfaction are related, our findings suggest that distributive justice (γ=0.329, t=5.672), procedural justice (γ=0.186, t=3.155) and interactional justice (γ=0.112, t=2.058) all positively affect post-recovery satisfaction significantly. H3a, H3b and H3c are thus supported. We can infer that a higher perceived justice implies a higher post-recovery satisfaction.
(4) Effect of emotions on post-recovery satisfaction
As for how emotions and post-recovery satisfaction are related, our results suggest that post-recovery satisfaction will be positively affected by positive emotions (β=0.188, t=3.397) and negatively affected by negative emotions (β=-0.222, t=-4.587) significantly. H4a and 43b are thus supported. That is, customers show a higher post-recovery satisfaction if the recovery process or outcome evokes positive emotions and a lower post-recovery satisfaction if the recovery process or outcome evokes negative emotions.
(5) Effect of post-recovery satisfaction on post-purchase intentions
As for how post-recovery satisfaction and post-purchase intentions are related, our results support H5, which states that post-recovery satisfaction positively affect post-purchase
intentions significantly (β=0.394, t=4.490).
Additionally, according to R2 results, 55.6% variance of positive emotion can be explained by perceived distributive justice, perceived procedural justice, and perceived interactional justice; 33.0% variance of negative emotion can be explained by perceived distributive justice, perceived procedural justice, and perceived interactional justice; 74.0% variance of post-recovery satisfaction can be jointly explained by perceived distributive justice, perceived procedural justice, perceived interactional justice, positive emotion, and negative emotion; 49.6% variance of post-purchase intentions can be jointly explained by perceived distributive justice, perceived procedural justice, perceived interactional justice, and post-recovery satisfaction.
Above results demonstrate that the proposed model exhibited a high level of reliability and validity in both the test of measurement model and the test of a structural model. All proposed causal paths were in a reasonable direction, and most of the proposed hypotheses were supported. Overall, the proposed model has an acceptable level of explaining how perceived justice, positive/negative emotions, post-recovery satisfaction, and post-purchase intentions are related, and can provide a valuable reference for online retailers.
Table 5 Summary of hypothesis test results
Hypothesis Relationship Coefficient t-value Conclusion
H2a PDJ PEM(+) 0.557*** 9.400 Support
H2b PDJ NEM(-) -0.286*** -4.119 Support
H2c PPJ PEM(+) 0.250*** 3.947 Support
H2d PPJ NEM(-) -0.276*** -3.892 Support
H2e PIJ PEM(+) 0.006 0.088 No Support
H2f PIJ NEM(-) -0.089 -1.245 No Support
H3a PDJ PRS(+) 0.329*** 5.672 Support H3b PPJ PRS(+) 0.186** 3.155 Support H3c PIJ PRS(+) 0.112* 2.058 Support H4a PEM PRS(+) 0.188*** 3.397 Support H4b NEM PRS(-) -0.222*** -4.587 Support H1a PDJ PPI(+) 0.226** 3.184 Support H1b PPJ->PPI(+) 0.107 1.259 No Support H1c PIJ->PPI(+) 0.054 0.723 No Support H5 PRS -> PPI(+) 0.394*** 4.990 Support * p<0.05; ** p<0.01; *** p<0.001
5. Discussion and implications
5.1 Discussion
Our results further demonstrate that online shoppers will raise positive emotions and decrease negative ones if they perceive distributive justice and procedural justice in a service recovery action. But perceived interactional justice has no significant effect on emotions. This may be due to the design and environment of online shopping websites, consumers are not easy to feel the interactional manner with offending service providers, therefore, the offender’s dedication in this aspect does not significantly induce positive or negative emotions from customers. All three perceived justices as well as both positive and negative emotions significantly affect post-recovery satisfaction. Additionally, distributive justice influences post-recovery satisfaction the most, which corresponds to the findings of Mattila (2001) and Maxham III and Netemeyer (2002).
Among the three components of justice, only distributive justice significantly affects post-purchase intentions. Procedural justice and interactional justice have no significant effect on post-purchase intentions. It could be explained in the following ways. First, it is possibly owing to that customers shop online to procure desired products. Moreover, discounts or coupons offered by offending service providers can effectively reduce the cost of the products they purchase in the future. However, a timely reaction of offenders or friendly attitude towards service failure does not reduce the cost of the products they purchase next time and, thus, cannot effectively induce repurchase intentions of customers. Furthermore, for customers, evaluating recovery outcomes is easier than evaluating recovery strategies or the interaction process with the offender, explaining why distributive justice influences post-purchase intentions than procedural justice and interactional justice do. Our finding further demonstrates that post-recovery satisfaction enhances post-purchase intentions, which reflects the finding of Montoya-Weiss et al. (2003) and Holloway et al. (2005).
5.2 Theoretical implications
To the best of our knowledge, no empirical study has specifically explored how customers’ perceived justice and emotions aroused by service recovery combine to affect subsequent customer satisfaction and post-purchase intentions with service recovery in the e-shopping context. The proposed model in this study shows an acceptable level of explaining how perceived justice, emotions, post-recovery satisfaction, and post-purchase intentions are related of online shopping to fill this important gap.
Prior researches typically find that perceived justice plays as a direct antecedent to post-recovery satisfaction of customers (Maxham III & Netemeyer, 2002, 2003; Mattila & Patterson, 2004). Recent studies indicate that perceived justice elicits emotions during service recovery encounters (e.g. Chebat and Slusarczyk, 2005; Schoefer and Ennew, 2005). This study provides evidence that perceived justice influences customer post-recovery satisfaction
both directly and indirectly (through emotions) as well as emotions have effects on post-recovery satisfaction directly. The present findings emphasize the importance of emotions in service recovery encounters. It also shows that the integration of emotions within the perceived justice of service recovery is a necessary way to better understand what drives customers’ evaluative judgments in a recovery situation.
5.3 Managerial implications
The findings of this study reveal that only distributive justice among the three
components of justice significantly affects post-purchase intentions. Additionally, distributive justice influences positive emotions and negative ones as well as post-recovery satisfaction the most. Therefore, distributive justice can be viewed as the most important component of perceived justice for online shoppers. The ability of service providers to more heavily emphasize distributive justice in service recovery, e.g., by offering discounts, refunds or change for alternative products at a higher grade, to avoid increase of costs born by customers, would allow them not only to evoke the positive emotions of customers but also enhance their satisfaction, positive post-purchase intentions, and loyalty.
Our finding also confirms that post-recovery satisfaction and post-purchase intentions are positively related. Additionally, all perceived justices increase post-recovery satisfaction. Therefore, service providers should provide customers with sufficient information on the tangible compensation for service failures and service recovery procedure, inform customers of the latest progress of service recovery, and properly react to their questions to increase their satisfaction with the recovery effort and outcome, ultimately affecting their post-purchase intentions.
Additionally, both positive and negative emotions significantly affect post-recovery satisfaction, indicating that positive emotions increase post-recovery satisfaction and negative emotions lower post-recovery satisfaction. That is, customer satisfaction may be reduced if their negative emotions are evoked during service recovery. Service providers should thus observe customers’ emotional responses, e.g., tones or phrasing, and adopt necessary corrective measures to appease their anger in the handling of a service failure. Inducing more positive emotions and mitigating negative emotions in offended customers would allow them to enhance customer satisfaction and positive post-purchase intentions.
6. Conclusion
This study provides empirical support for a new approach to understanding the antecedents of post-recovery satisfaction and post-purchase intentions with service recovery of online shopping by incorporating perceived justice and emotions perspectives. Contrary to previous findings, procedural justice and interactional justice were found to have no significant relationships with post-purchase intentions. This highlights the potential
importance of taking perceived distributive justice for online shoppers to predict post-purchase intentions with service recovery. Additionally, both positive and negative emotions significantly affect post-recovery satisfaction. That is, understanding customers’ emotions is helpful for service providers in increasing post-recovery satisfaction of customers and recovery from service failures.
Despite its contributions, this study has certain limitations. Even though the study succeeds in validating the measurement scale of perceived justice of service recovery in the context of online shopping, but the results obtained in the EFA developed to test the measurement model of perceived justice indicating some weaknesses in the elaboration of the measurement scales for distributive, procedural, and interactional justices. Particularly, the need to remove six items from the scales originally proposed for this construct may raise doubts about content validity of the scales. This could be due to various reasons. One of the reasons is that the measurement scale of perceived justice in online shopping context is designed according to the related literatures in the context of brick-and-mortar retail stores, but the attributes of e-tailing are still a little different from the brick-and-mortar retail stores. This situation indicated that when rashly applying the perceived justice scale in the context of brick-and-mortar retail stores to measure perceived justice in e-shopping is inappropriate. The validated scale, the eleven items across three dimensions, in this study can serve a useful instrument to measure the perceived justice in the context of e-shopping. Consequently, we can use a second-order factor model to test the stability of the scale of perceived justice in the context of e-shopping. In addition, other large samples should be gathered to confirm and refine, the factor structure of the perceived justice scale in the context of e-shopping, and to assess its reliability and validity in the future.
From the coefficients of determination in our model, we can infer that other factors not consider here also affect emotions and post-purchase intentions. Therefore, future studies should considerable other variables of emotions and post-purchase intentions, including degree of involvement or service quality, or role of the attribution theory in elucidating the factors affecting customer emotions and post-purchase intentions. Besides, our research sample was obtained from only undergraduate and graduate school students, thus limiting the generalizability of the research results to all groups of online shoppers. Future studies should expand the scope of research subjects to other customer groups or accumulate longitudinal data to re-examine our model and compare the results obtained from different customer groups. As for the design of scenarios, this study limited service failures to only those stemming from mistakes of online retailers and defined recovery strategies as those offered by online retailers. Because service failures may also be attributed to partners of online retailers (e.g. late delivery of goods may be caused by problems with the online order system or the delivery system of the express delivery company) or customers themselves (e.g. customers provide a wrong address), future studies should also thoroughly investigate the causes of
different service failures to obtain more comprehensive results.
Acknowledgement
This research is supported by the National Science Council (NSC 97-2410-H-390-014-MY2), Taiwan.
References
1. Adams, J. S. (1965). Inequity in social exchange. Advances in Experimental Social Psychology, 2, 267-269.
2. Anderson, E. W., & Mittal, V. (2000). Strengthening the satisfaction-profit chain. Journal of Service Research, 3(2), 107-120.
3. Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: A review and recommended two-step approach. Psychological Bulletin, 103(3), 411-423. 4. Agarwal, R. & Peasad, J. (1998). A conceptual and operational definition of personal
innovativeness in the domain of information technology. Information Systems Research, 9(2), 204-215.
5. Bagozzi, R. P., Gopinath, M., & Nyer, P. U. (1999). The role of emotions in marketing. Journal of the Academy of Marketing Science, 27(2), 184-206.
6. Bagozzi, R. P., & Yi, Y. (1988). On the evaluation of structural equation models. Journal of the Academy of Marketing Science, 16(1), 74-94.
7. Bitner, M. J., Booms, B. H., & Tetreault, M. S. (1990). The service encounter: diagnosing favorable and unfavorable incidents. Journal of Marketing, 54(1), 71-84. 8. Bitner, M. J., Brown, S. W., & Meuter, M. L. (2000). Technology infusion in service
encounters. Journal of Academy of Marketing Science, 28(1), 138-149.
9. Blodgett, J. G., Granbois, D. H., & Walters, R. G. (1993). The effects of perceived justice on complainants' negative word-of-mouth behavior and repatronage intentions. Journal of Retailing, 69(4), 399-428.
10. Blodgett, J. G., Hill, D. J., & Tax, S. S. (1997). The effects of distributive, procedural, and interactional justice on postcomplaint behavior. Journal of Retailing, 73(2), 185-210. 11. Boshoff, C. (1997). An experimental study of service recovery options. International
Journal of Service Industry Management, 8(2), 110-130.
12. Cacioppo, J. T., & Gardner, W. L. (1993). What underlies medical donor attitudes and behavior? Health Psychology, 12(4), 269–271.
13. Cacioppo, J. T., Gardner, W. L., & Berntson, G. G. (1997). Beyond bipolar conceptualizations and measuress: The case of attitudes and evaluative space. Personality and Social Psychology Review, 1(1), 3–25.
14. Chan, C. Y. (2008). 2008 suffers behavior and consuming development trends of B2C in Taiwan. Report of Market Intelligence and Consulting Institute, Institute for Information
Industry, Taiwan.
15. Chebat, J. C., & Slusarczyk, W. (2005). How emotions mediate the effects of perceived justice on loyalty in service recovery situations: an empirical study. Journal of Business Research, 58(5), 664-673.
16. Chin, W. W. (1998). The partial least squares approach for structural equation modeling. In G. A. Marcoulides, G. A. (ed.), Modern Methods for Business Research, Lawrence Erlbaum Associates Hillsdale, NJ.
17. Collier, J. E., & Bienstock, C. C. (2006). Measuring service quality in e-retailing. Journal of Service Research, 8(3), 260-275.
18. del Río-Lanza, A. B., Vázquez-Casielles, R., & Díaz-Martín, A. M. (2009). Satisfaction with service recovery: perceived justice and emotional responses. Journal of Business Research, 62(8), 775-781.
19. Deng, Z., Lu, Y., Wei, K. K., & Zhang, J. (2010). Understanding customer satisfaction and loyalty: an empirical study of mobile instant messages in China. International Journal of Information Management, 30(4), 298-300.
20. Farrell, A. M. (2010). Insufficient discriminant validity: a comment on Bove, Pervan, Beatty, and Shiu (2009). Journal of Business Research, 63(3), 324-327.
21. Forbes, L. P., Kelley, S. W., & Hoffman, K. D. (2005) Typologies of e-commerce retail failures and recovery strategies. Journal of Services Marketing, 19(5), 280-292.
22. Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39-50.
23. Goodwin, C., & Ross, I. (1992). Customer responses to service failures: influence of procedural and interactional fairness perceptions. Journal of Business Research, 25(2), 149-163.
24. Gronroos, C. (1988). Service quality: the six criteria of good perceived service quality. Review of Business, 9(3), 10-13.
25. Ha, J., & Jang, S. C. (2009). Perceived justice in service recovery and behavioral intentions: the role of relationship quality. International Journal of Hospitality Management, 28(3), 319-327.
26. Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1998). Multivariate Data Analysis, Prentice-Hill, Englewood Cliffs.
27. Harris, K. E., Grewal, D., Mohr, L. A., & Bernhardt, K. L. (2006). Consumer responses to service recovery strategies: the moderating role of online versus offline environment. Journal of Business Research, 59(4), 425-431.
28. Hoffman, K. D., & Kelley, S. W. (2000). Perceived justice needs and recovery evaluation: a contingency approach. European Journal of Marketing, 34(3/4), 418-433.
employee recovery efforts. The Journal of Service Marketing, 9(2), 49-61.
30. Holloway, B. B., & Beatty, S. E. (2003). Service failure in online retailing: a recovery opportunity. Journal of Service Research, 6(1), 92-105.
31. Holloway, B. B., Wang, S., & Parish, J. T. (2005). The role of cumulative online purchasing experience in service recovery management. Journal of Interactive Marketing, 19(3), 54-66.
32. Holloway, B. B., & Beatty, S. E. (2008). Satisfiers and dissatisfiers in the online environment: a critical incident assessment. Journal of Service Research, 10(4), 347-364. 33. Homans, G. C. (1961). The humanities and the social sciences: joint concern with ”individual” and values the arts distinct from social science distinctions of social status. The American Behavioral Scientist, 4(8), 3-6.
34. Hulland, J. (1999). Use of partial least squares (PLS) in strategic management research: a review of four recent studies. Strategic Management Journal, 20(2), 195–204.
35. Kaiser, H. F. (1958). The varimax criterion for analytic rotation in factor analysis. Psychometrika, 23(2), 187-200.
36. Kelley, S. W., Hoffman, K. D., & Davis, M. A. (1993). A typology of retail failures and recoveries. Journal of Retailing, 69(4), 429-452.
37. Kim, S. S., & Son, J. Y. (2009). Out of dedication or constraint? a dual model of post-Adoption phenomena and its empirical test in the context of online services. MIS Quarterly, 33(1), 49-70.
38. Konovsky, M. A. (2000). Understanding procedural justice and its impact on business organizations. Journal of Management, 26(3), 489-511.
39. Kuo, Y. F., Wu, C. M., and Deng, W. J. (2009). The relationships among service quality, perceived value, customer satisfaction, and post-purchase intention in mobile value-added services. Computers in Human Behavior, 25(4), 887-896
40. Kuo, Y. F., Yen, S. T., & Chen, L. H. (2011). Online auction service failures in Taiwan: typologies and recovery strategies. Electronic Commerce Research and Applications, 10(2), 183-193.
41. Laros, F. J. M., & Steenkamp, J. B. E. M. (2005). Emotions in consumer behavior: a hierarchical approach. Journal of Business Research, 58(100), 1437-1445.
42. Lee, Y. K., Young, K., Son, M. H, & Lee, D. J. (2011). Do emotions play a mediating role in the relationship between owner leadership styles and manager customer orientation, and performance in service environment? International Journal of Hospitality Management, doi:10.10.1016/j.ijhm.2011.02.002.
43. Mattila, A. S. (2001). The effectiveness of service recovery in a multi-industry setting. Journal of Services Marketing, 15(7), 583-596.
44. Mattila, A. S., & Patterson, P. G. (2004). Service recovery and fairness perceptions in collectivist and individualist contexts. Journal of Service Research, 6(4), 336-346.
45. Maxham, J. G. III. (2001). Service recovery’s influence on consumer satisfaction, positive word-of-mouth, and purchase intentions. Journal of Business Research, 54(1), 11-24.
46. Maxham, J. G., III, & Netemeyer, R. G. (2002). Modeling customer perceptions of complaint handling over time: the effects of perceived justice on satisfaction and intent. Journal of Retailing, 78(4), 239-252.
47. Maxham, J. G., III, & Netemeyer, R. G. (2003). Firms reap what they sow: the effects of employee shared values and perceived organizational justice on customer evaluations of complaint handling. Journal of Marketing, 67(1), 46-62.
48. McColl-Kennedy, J. R., & Sparks, B. A. (2003). Application of fairness theory to service failure and service recovery. Journal of Service Research, 5(3), 251-266.
49. McCollough, M. A., Berry, L. L., & Yadav, M. S. (2000). An empirical investigation of customer satisfaction after service failure and recovery. Journal of Services Research, 3(2), 121-137.
50. Miller, J. L., Craighead, C. W., & Karwan, K. R. (2000). Service recovery: a framework and empirical investigation. Journal of Operations Management, 18(4), 387-400.
51. Montoya-Weiss, M. M., Glenn, B. V., & Grewal D. (2003). Determinants of online channel use and overall satisfaction with a relational, multichannel service provider. Journal of the Academy of Marketing Science, 31(4), 448-458.
52. Oliver, R. L. (1997). Satisfaction: A Behavioral Perspective on the Consumer, Singapore: McGraw-Hill.
53. Qureshi, I., Fang, Y., Ramesy, E., McCole, P., Ibboston, P., & Compeau, D. (2009).Understanding online customer repurchasing intention and the mediating role of trust - an empirical investigation in two developed countries. European Journal of Information Systems, 18(3), 205-222
54. Ringle, C. M., Wende, S., & Will, S. (2005). SmartPLS 2.0 (M3) Beta, Hamburg, http://www.smartpls.de.
55. Schoefer, K. (2008). The role of cognition and affect in the formation of customer satisfaction judgments concerning service recovery encounters. Journal of Consumer Behaviour, 7(3), 210-221.
56. Schoefer, K., & Diamantopoulos, A. (2008). Measuring experienced emotions during service recovery encounters: construction and assessment of the ESRE scale. Service Business, 2(1), 65-81.
57. Schoefer, K., & Ennew, C. (2005). The impact of perceived justice on consumer emotional responses to service complaints experiences. Journal of Services Marketing, 19(5), 261-270.
58. Seiders, K., & Berry, L. L. (1998). Service fairness: what it is and why it matter. Academy of Management Executive, 12(2), 8-20.
59. Smith, A. K., Bolton, R. N., & Wagner, J. (1999). A model of customer satisfaction with service encounters involving failure and recovery. Journal of Marketing Research, 36(3), 356-372.
60. Smith, A. K., & Bolton, R. N. (2002). The effects of customers, emotional responses to service failures on their recovery effort evaluations and satisfaction judgments. Journal of the Academy of Marketing Science, 30(1), 5-23.
61. Sparks, B. A., & McColl-Kennedy, J. R. (2001). Justice strategy options for increased customer satisfaction in a services recovery setting. Journal of Business Research, 54(3), 209-218.
62. Tax, S. S., & Chandrashekaran, M. (1992). Consumer decision making following a failed service encounter: a pilot study. Journal of Consumer Satisfaction, Dissatisfaction and Complaining Behavior, 5, 55-68.
63. Tax, S. S., Brown, S. W., & Chandrashekaran, M. (1998). Customer evaluations of service complaint experiences: implications for relationship marketing. Journal of Marketing, 62(4), 60-76.
64. Udo, G. J., Bagchi, K. K., & Kirs, P. J. (2010). An assessment of customers’ e-service quality perception, satisfaction and intention. International Journal of Information Management, 30(6), 481-492.
65. Walczuch, R., Lemmink, J., & Streukens, S. (2007). The effect of service employees' technology readiness on technology acceptance. Information and Management, 44(2), 206-215.
66. Wang, H. C., Pallister, J. G., & Foxall, G. R. (2006). Innovativeness and involvement as determinants of website loyalty: III. theoretical and managerial contributions. Technovation, 26(12). 1374-1383.
67. Wang, Y. S., Wu, S. C., Lin, H. H., and Wang, Y. Y. (2010). The relationship of service failure severity, service recovery justice and perceived switching costs with customer loyalty in the context of e-tailing. International Journal of Information Management, doi:10.1016/j.ijinfomgt.2010.09.001.
68. Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: the PANAS scales. Journal of Personality and Social Psychology, 54 (6), 1063–1070.
69. Weiss, H. M., Suckow, K., & Cropanzano, R. (1999). Effects of justice conditions on discrete emotions. Journal of Applied Psychology, 84(5), 1999, 786-194.
70. William, S. (1999). The effects of distributive and procedural justice on performance. Journal of Psychology Interdisciplinary and Applied, 133(2), 183-194.
71. Wirtz, J., & Mattila, A. S. (2004). Consumer responses to compensation, speed of recovery and apology after a service failure. International Journal of Service Industry Management, 15(2), 150-166.
