An empirical investigation of the factors motivating Japanese repeat consumers to review their shopping experiences

Journal of Business Research xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Journal of Business Research journal homepage: www.elsevier.com/locate/jbusres

An empirical investigation of the factors motivating Japanese repeat consumers to review their shopping experiences Emi Moriuchia,⁎, Ikuo Takahashib a b

Rochester Institute of Technology, Saunders College of Business, 107 Lomb Memorial Drive, Bldg12, Rochester, NY 14623, United States Keio University, Faculty of Business and Commerce, 2-15-45 Mita, Minato-ku, Tokyo 108-8345, Japan

A R T I C L E I N F O

A B S T R A C T

Keywords: Japanese consumer Online reviews Search attributes Online supermarket Repeat online consumers

Consumers purchase items (e.g., food) online due to today's rapidly changing markets and to improvements in online and mobile technology. Online supermarkets have been gaining popularity among Japanese consumers. As the population experiences a hectic lifestyle and is aging, many Japanese consumers are seeing the benefits of shopping on an online supermarket website. However, as with any e-commerce activity, reviews are critical for the success of these e-vendors. Despite the importance of reviews, little is known about what motivates repeat online supermarket consumers to review their purchasing experiences. This paper examines consumers' willingness to review and the relationship between consumers' search attributes such as price, promotion and service, e-satisfaction and trust on their online supermarket purchase experiences. The research findings support the results of earlier studies that search attributes are determinants of consumers' e-satisfaction. In addition, esatisfaction affects e-trust. Interestingly, e-trust has a negative effect on the willingness to review.

1. Introduction Japan's consumer market is constantly driven by societal changes, and this has brought about significant shifts in cultural identity. Japanese consumers are seeing an increase in Western influence as a result of globalization. Japan is often described as a collective nation of ethnic homogeneity that it is culturally unique, which factors toward the building of a successful modern nation (Tsutsui, 2009). Japan's investment in Asia and its dependence on international trade and on a large number of foreign workers have forced the nation to recognize that its survival is dependent on understanding the importance of ethnic diversity. The consumer market in Japan is constantly driven by new product demands and by images to promote and sell. Despite efforts made to promote and protect Japanese cultural identity, Japanese consumers are inevitably exposed to products and cultural influences from other countries. While Japanese marketers accommodate their culturally diverse consumer market, Western exposure has influenced Japanese marketers to be challenged in their development and creation of products. Furthermore, with the presence of the Internet, news and trends on the latest products reach consumers more quickly. As asserted by Atchariyachanvanich et al. (2007) “electronic commerce is at its core of expansion, and its scale of growth rate varies among countries” (p.47). Japan is a country with a strongly collectivistic culture that is

⁎

embedded in everything it does, and this is one of the reasons why Japan is known to have a complex distribution system (Itoh, 1991, 2000). Japan's e-commerce endeavors are beginning to override its multichannel distribution systems, which many foreign companies view as essential to establishing businesses in Japan. The Japanese consumer market has not been widely explored since the latest long-term recession hit and since the country's two most recent earthquake disasters. Although the Japanese economy has begun to regenerate, a major challenge facing the Japanese economy is its long-term demographic shift. This has altered opportunities for and threats to marketers in determining the consumption patterns of Japanese consumers. To capitalize on a growing market such as Japan's, it is important to first recognize the lifestyles of average Japanese consumers. According to Takahashi and Fluch (2009), Japan is an Internet-driven society and Japanese consumers “use diverse Internet retail shops and Internet auction sites, creating a variety of purchasing channels available to them” (p. 158). Salsberg (2010) reported that “online shopping is central to both economizing and nesting trends” (para. 8). With such heavy reliance on the Internet, many retail outlets are inevitably forced to adapt to this emerging trend to remain competitive in the market. Companies and retail stores in particular are finding ways to satisfy their customers by paying attention to feedback and complaints and especially when various social media platforms allow consumers to post their comments instantly. In this paper we examine some of the

Corresponding author. E-mail addresses: [email protected] (E. Moriuchi), [email protected] (I. Takahashi).

http://dx.doi.org/10.1016/j.jbusres.2017.07.024 Received 1 April 2015; Accepted 1 July 2017 0148-2963/ © 2017 Elsevier Inc. All rights reserved.

Please cite this article as: Moriuchi, E., Journal of Business Research (2017), http://dx.doi.org/10.1016/j.jbusres.2017.07.024

Journal of Business Research xxx (xxxx) xxx–xxx

E. Moriuchi, I. Takahashi

Osmonbekov, & Czaplewski, 2006; Liu, 2006). While online reviews are considered important to varying degrees, most consumers who write reviews are often first time purchasers of a particular item. Thus, there is still a lack of understanding of what factors encourage repeat customers to review their products after making second and subsequent purchases. According to Peterson, Balasubramanian, and Bronnenberg’s (1997) framework for consumer decision sequences, performance and competition for shopping media are mediated by three factors: 1) a consumer's choice of communication, transaction, and distribution channels; 2) product and service offerings marketed and 3) the sequence of decisions that follows from a purchasing intention. With the theoretical underpinning of this theory and accordance with the Web Trust model, this study seeks to investigate consumers' search attributes. As the consumers examined in this study have chosen online channels as a shopping mode, we focus on search attributes related to intentions to review and on degrees of online shopping satisfaction and trust.

different factors that motivate repeat Japanese online consumers to shop at online supermarkets and how these factors affect their intentions to post online reviews on the products and services that they have received from online stores. 2. Literature review 2.1. Japanese online shoppers Japan is the third largest economy in the world after the U.S. and China (BBC News, 2015). Japan has played a major role in the international community and has served as a major aid donor and as a source of global capital and credit. As Japan continues to recover from its 2011 tsunami, it has become evident that Japanese consumers are playing an active role in shopping both at brick-and-mortar stores and online. Furthermore, with the introduction of Abenomics, increases in sales tax have encouraged price-sensitive customers to shop online. Another area witnessing an increase in online shopping rates is the online food and grocery product industry.

3. Research framework

2.2. E-commerce and eWOM

3.1. Search attributes and online review intentions

Consumers can use a variety of different retail channels (e.g., brickand-mortar stores, mail orders, TV shopping channels and the Internet). Each retail channel is characterized by a different combination of multiple attributes that has an impact on a consumer's choice of a retail channel (Childers, Carr, Peck, & Carson, 2002; Ohanian, Tashchian, & Beard, 1992). With the expansion of e-commerce, social commerce and third party review sites, many consumers are sharing their experiences through these platforms via word of mouth (hereafter WOM). WOM communication can occur at different stages of a product's life cycle depending on the customer's experience with a product and the life stage of a product. A product life cycle refers to the sequence of stages that a product undergoes from its introduction to its decline (Day, 1981). However, not every product undergoes the same cycle or each stage of the cycle. For instance, consumers classified as innovators may seek more information at early stages of a product's life cycle. It is however difficult for marketing managers to actually recognize when and how they should change their means of providing information, as stages are not fully discrete. Should marketers thus present all related information at once? This is not advisable because the volume of information available maybe too large to consume, causing consumers to abandon a purchase. Word of mouth (WOM) communication serves an effective way of overcoming such challenges (Mahajan, Muller, & Kerin, 1984), as WOM communication involves the delivery of product information from a user's perspective at each stage (Park & Kim, 2009). Thus, WOM communication is regarded as an effective form of communication that can deliver the right information to different consumer segments. However, many traditional marketers may not have the knowledge needed to apply WOM communication to their products, as the presence of WOM communication can at times be untraceable (Park & Kim, 2009). With the emergence of the Internet, WOM communication has evolved to become electronic WOM communication (hereafter eWOM), which is measurable due to its digital footprint on websites. Marketers can apply their marketing strategies according to eWOM communication that they receive from their consumers. The retrieval of eWOM information requires that consumers are willing to spend time reviewing and sharing their shopping experiences online. Cantallops and Salvi (2014) reasoned that people contribute to online reviews because they are self-directed, want to help other consumers for social benefits and consumer empowerment and want to help companies. Research has shown that online reviews (e.g., eWOM) influence product sales (Decker & Trusov, 2011; Zhu & Zhang, 2010), customer value and loyalty (Chevalier & Mayzlin, 2006; Gruen,

When searching online for information in an external environment such as an online store, consumers often focus on relevant attributes that are easily accessible and distinguishable (Dick, Chakravarti, & Biehal, 1990). However, when the costs of acquiring relevant information are higher than the expected benefits, Ratchford (1982) asserts that consumers rely on their prior experiences for such information. In online media channels, it is not unusual for certain attribute information necessary for decision making is not readily accessible (Degeratu, Rangaswamy, & Wu, 2000). Depending on whether a product has a large number of sensory attributes (e.g., edible food) or is an industrial product that is non-sensory, the availability of information will vary. In other words, when there is much information on sensory and non-sensory attitudes and when the price is high, consumers find attributes to play a larger role in their evaluations (Degeratu et al., 2000). These evaluations are often observed in the form of online reviews on a company's website, on a social media networking site (e.g., Facebook) or on a third-party website (e.g., Yelp). Prior studies (Moriuchi, 2016; Park & Kim, 2009; Zhu & Zhang, 2010) have touched on the topic of word of mouth online reviews. Online reviews shape consumer behavior. Previous studies argue that consumers' intentions to shop online are influenced by several variables, including levels of convenience, prices and product types (Burke, 1997; Chiang & Dholakia, 2003; Peterson et al., 1997). Although these studies examine intentions to shop online and not specifically intentions to review online, it is arguable that as intentions are linked to control beliefs,1 which are applicable to consumers' review intentions. In addition, the willingness to review is influenced by initial use and/or purchase experiences. From these arguments, we develop the following hypotheses: H1. Service and fulfillment have a positive impact on the intent to review. H2. Prices have a positive impact on the intent to review. H3. Promotions have a positive impact on the intent to review.

3.2. Services, fulfillment levels and prices as antecedents for e-satisfaction Satisfaction, according to Oliver (1997), is the overall psychological state that results when a consumer's prior feelings about his or her 1 Control beliefs refer to beliefs on the presence of factors that may facilitate or impede the performance of behavior.

2

Journal of Business Research xxx (xxxx) xxx–xxx

E. Moriuchi, I. Takahashi

3.4. The perceived relationship between e-satisfaction and e-trust

experience are coupled with surrounding disconfirmed emotional expectations. In this study, e-satisfaction is defined as the gratification of a customer with respect to his or her prior purchasing experiences on electronic shopping sites. In this study, the term “service and fulfillment” is used to better describe the construct that this research attempts to assess and to prevent confusion. Wolfinbarger and Gilly (2003) defined customer service (e.g., service and fulfillment) as the helpfulness of service representatives and as levels of willingness to address customers' inquiries quickly. They suggested that e-satisfaction is based on online shoppers' evaluations of products and services offered online and on the fulfillment of shoppers' needs through service support.

Every consumer harbors some expectations about a product or service. However, if such expectations are not met in a positive manner, the consumer will be dissatisfied. The e-satisfaction construct has been used as a dependent variable (Chang, Wang, & Yang, 2009), as a mediator (Yang & Peterson, 2004) and as a moderator (Mittal & Kamakura, 2001; Keh & Lee, 2006) in research models. It is a very important construct because corresponding results can influence levels of e-satisfaction, which predicate future behavior. These future behaviors include customer retention, which subsequently leads to customer loyalty (Storbacka, Strandvik, & Grönroos, 1994). In addition, Cronin, Brady, and Hult (2000) argued that consequences of consumer satisfaction lead to positive word of mouth effects, price premiums and repurchase intentions. Alternatively, dissatisfied customers discontinue their subsequent purchases and may make an effort to write a negative review about the e-vendor. According Yoon (2002) and others (Flavián et al., 2006; Horppu, Kuivalainen, Tarkiainen, & Ellonen, 2008), e-satisfaction appears to be a prerequisite for e-trust. Furthermore, according to other studies conducted in the field (Gefen, Karahanna, & Straub, 2003; Jarvenpaa, Tractinsky, et al., 2000; McKnight et al., 2002), trust serves as a central aspect of many economic transactions made online and especially due to the complexities of social surroundings. In other words, we are constantly pressured to “identify what, when, why and how others behave” (Gefen et al., 2003). Luhmann (1979) further explained that when the social environment is beyond the control of rules and regulations, people tend to use trust as a central means of mitigating social complexity. This argument holds true when applied to the online environment. Given the presence of high levels of flexibility, the importance of trust is more pronounced between consumers and e-vendors. Moreover, Reichheld and Schefter (2000) argued that “Price does not rule the Web, trust does” (p.107). Studies have revealed a link between website satisfaction and website trust (Horppu et al., 2008). Furthermore, Flavián et al. (2006) proposed that satisfaction is based on consumers' experiences. Yoon (2002) and Ribbink, Van Riel, Liljander, and Streukens (2004) also supported this argument, stating that there is a positive correlation between e-satisfaction and e-trust. Thus, we hypothesize the following:

H4. Service and fulfillment have a positive impact on e-satisfaction. There are consistent results on the impacts of prices on satisfaction levels. Marshall (2009) argued that a consumer will exhibit a higher level of price tolerance when whatever he pays is the economic measure of his satisfaction surplus. In other words, consumer satisfaction is based on the difference between what consumers are willing to spend on a product relative to its market price. Anderson and Srinivasan (2003) and others (Degeratu et al., 2000) argued that price seeking information has an impact on satisfaction. In addition, Burke, Harlam, Kahn, and Lodish (1992) noted that offline and online shopping channels differ in that online consumers can obtain information on price attributes. Wang (2003) and others (Evanschitzky, Iyer, Hesse, & Ahlert, 2004) agreed that the level of consumer satisfaction is dependent on not only the quality of services delivered but also price levels. H5. Price has a positive impact on e-satisfaction.

3.3. The Web Trust Model (WTM) and the perceived link between services, fulfillment and e-trust In this study, McKnight, Choudhury, and Kacmar's (2002) Web Trust Model was used. This model supports the notion that trusting beliefs (perceived web vendor attributes) lead to trusting intentions, which in turn influence consumers' intentions to review. This behavior includes an intention to review an e-vendor online and consumers' overall shopping experiences. Gefen and Pavlou (2006) claimed that trust has varying effects on online purchases. An earlier study conducted by McKnight et al. (2002, 2004) argued that trust is focused primarily on initial online purchases and implied that institutional mechanisms do not play a key role in evaluating trustworthiness in online repurchasing contexts. However, in contrast, Kim, Xu, and Koh (2004) found in their study that trust still plays a role in repeat customer purchase decisions. Furthermore, they added that customer satisfaction is a driving force behind trust building for these repeat customers. Trust is generally a crucial element of many e-commerce activities and can result in undesirable opportunistic behaviors (Fukuyama, 1995). In this study, we adopt the concept of e-trust to describe online trust. Coulter and Coulter (2002) added that trust serves as a key factor in the establishment of long-term relationships between service representatives and their customers. Aurier and N'Goala (2010) found that trust has a direct influence on service usage and cross buying. They added that trust is an important factor when considering how a company can improve on their service relationships and company profits. Heintzman and Marson (2005) stated that Citizens First 3 found (through structural equation modeling) that this relationship does exist, that service satisfaction is a strong driver of trust, and that this relationship only occurs in the service-trust direction and not vice versa. This finding is important, as it shows that service satisfaction boosts trust levels and is not just a reflection of it as some researchers have suggested. Thus, the following hypothesis was developed:

H7. E-satisfaction has a positive effect on e-trust. Trust is a multidimensional construct that is often related to characteristics such as integrity and competence (Gefen et al., 2003). It was highlighted by Luhmann (1979) that trust serves as the foundation of all social life. In fact, trust is a fundamental consideration in the commercial world where people are influenced by the attitudes and actions of consumers and sellers (McCole et al., 2010). Trust is largely an important consideration because people are often faced with uncertainties. Consumers often go through a standard learning hierarchy of thought processes when there are two factors present: 1) uncertainty and 2) the complexity of the situation at hand. With the rise of online shopping, the e-commerce environment has been challenged with various risks (e.g., security risks) involved. Thus, consumers rely on their trust in a specific vendor, in the Internet or even in eWOM communication to mitigate the effects of uncertainties in the relationship between buyers and sellers in the online environment (Ha & Stoel, 2012). 3.5. E-trust as an antecedent of the intent to review In the world of e-commerce, trust can be conceptualized as existing between a consumer and retailer (Ranaweera & Prabhu, 2003). McCole et al. (2010) argued that “trust in the transacting vendor is important for the consumer to accept risk which is associated with or inherent in a given transaction” (p. 1020). The present study focuses on trust developed through a relationship between individuals and specifically between customers and e-retailers. We operationalize trust consistent with Morgan and Hunt (1994) as “existing when one party has

H6. Service and fulfillment have a positive impact on e-trust. 3

Journal of Business Research xxx (xxxx) xxx–xxx

E. Moriuchi, I. Takahashi

NTTCom Online Marketing Solutions Corporation, a panel service company, in Japan in November 2012. The online survey was used as a tool to gather data due to its advantages over email surveys and owing its higher levels of data quality. In total, 1208 responses were collected through an online survey. These responses were received from respondents who have purchased product(s) online. Purchases made ranged from groceries to electronics and travel packages. As this study focuses on repeat shoppers who have purchased a product from an online supermarket, 264 responses were deemed usable as these respondents reported to be online supermarket customers. The remaining responses were given by customers who had purchased electronics, tourism packages, musical instruments, and other product/services from an online vendor. Two screening questions were used to filter out respondents who do not fit the sampling requirements. Respondents who do not shop online for groceries were not included in the analysis, and in turn 264 responses were used in the analysis. The respondents spent 10.15 min completing the survey on average with the fastest completing it in 2.70 min and the slowest completing it in 117.6 min. The standard deviation for time spent on the survey for the 264 respondents was recorded as 9.18 min.

confidence in an exchange partner's reliability and integrity” (p. 23). Thus, our construct can be denoted as trust in e-tailers. Ranaweera and Prabhu (2003) argued that the use of word of mouth communication is a key facet of truly loyal customers. In their study they found that higher levels of trust lead to higher levels of positive word of mouth communication. Gefen (2002) asserted that trust can develop directly or indirectly through an overall assessment of how trust influences relevant behavioral intentions. Through his study he found that trust, integrity and trustworthiness in a vendor affect purchase intentions. Although Gefen's (2002) study focused on purchasing intentions, we argue that as review intentions are behavioral, they are impacted by trust as well. Furthermore, Garbarino and Johnson (1999) found that trust is related to customers' future intentions. Although Garbarino and Johnson's (1999) study did not consider WOM communication, the act of saying positive things about a company to others is consistent with the behavior that they measured. Thus, following the logic of Gremler, Gwinner, and Brown's (2001) study, a customer's trust will have a directly positive influence on the propensity to engage in positive WOM communication, which in this study is defined as the intention to review. Furthermore, Ganesan (1994) and others (Garbarino & Johnson, 1999; Morgan & Hunt, 1994) claim that the trust construct has been associated with many positive behaviors in both empirical and conceptual research.

5. Results 5.1. Preliminary data analysis

H8. Trust positively affects a customer's intention to review online shopping experiences.

The subjects were asked to respond to various general statements that assessed the importance of obtaining product information as a repeat consumer as well as factual information about the given product. Search attributes were measured on a 5-point scale (1 = strongly disagree; 5 = strongly agree). Subjects were asked to respond to various general statements that assessed the importance of obtaining product information as a repeat consumer and of previous experiences with the selected online vendor. Male respondents represented 41.6% of the respondents. The three largest respondent age groups were 30–39 (18.6%), 40–49 (39.9%) and 50–59 years of age (36.4%). All of the respondents were repeat online consumers: 6.8% shop at an online supermarket at least several times a week; 36.4% shop at an online supermarket at least several times a month; 18.9% shop at an online supermarket several times every six months; and 37.9% shop at an online supermarket at least several times each year. Prior to the final data analysis, a maximum likelihood exploratory factor analysis was conducted on the effects of consumers' search attributes on their online supermarket shopping habits. A total of 13 items on search attributes, four items on e-satisfaction, four items on etrust and six items on review intentions were analyzed. All factors had a loading of (> 0.40), and all items were considered in the final analysis. AMOS software version 22 was employed to perform structural equation modeling through a two-staged analysis (Anderson & Gerbing, 1988). The delineation of relationships between the constructs was the primary focus of the study; therefore, a correlation matrix (see Table 5) was used to estimate the structural model (Hair, Black, Babin, Anderson, & Tatham, 2006). We first developed a measurement model consisting of three exogenous and three endogenous constructs by conducting a confirmatory factor analysis on a multi-item scale (i.e., search attributes of online supermarket shopping, trust in online cybermediaries, satisfaction with online supermarket shopping, and intentions to review products online). Cronbach's Alpha was used to evaluate scales for internal consistency. All measures demonstrated reliability with alpha values of 0.80 or greater. A confirmatory factor analysis (CFA) was conducted to test the overall validity of the measurement model. The CFA results show a good model fit for the 27-item model with χ2 = 282.79, df = 174, p = 0.000; CFI = 0.98; RMSEA = 0.05; NFI = 0.95; and TLI = 0.97. To assess the convergent validity of the measurements, Fornell and Larcker (1981) propose examining three metrics: the item reliability of

4. Method 4.1. Measurement development An online survey was developed to test the research hypotheses. The intent of this survey was to investigate the effects of repeat online consumers' intentions to review their online experiences. The intention to review was measured based on three search attributes: price, promotion and service and fulfillment. Measures for trust in a vendor were inspired by Balabanis, Reynolds, and Simintiras (2001). Measures for esatisfaction were operationalized based on two items inspired by Lau and Lee (2000) and using two items adopted from Flavián et al. (2006). The antecedents of e-satisfaction (i.e., price, promotion and service and fulfillment) were measured on four-, three- and four-item scales, respectively. All measurements were based on a 5-point Likert scale (1 = Strongly Disagree; 5 = Strongly Agree). Price was measured on the following four-item scale: inexpensive, value for money, higher discounts and comparative pricing. Promotion was measured on a three-item scale that included the following: exposure to ads on communication channels other than TV and Internet search sites, permission marketing (e.g., emails), and the persuasiveness of an ad. Service and fulfillment were measured on a four-item scale that included the following: the reliability of the delivery service, convenience, product availability, and post-purchase services. E-satisfaction was measured on a three-item scale that included the following: product variety, satisfaction with purchases, and satisfaction with the shopping experience. E-trust was measured on a three-item scale that included the following: trust in the e-vendor, the transaction process, and confidence in the monetary transaction. Trust was measured on a three-item scale that considered the trustworthiness of an online transaction's safety precautions. Review intentions were measured on a four-item scale that included the following: the intention to review the probability of repeat purchases, the e-vendor's product, the e-vendor's website, the overall e-shopping experience, and the transaction experience. 4.2. Data collection Data were collected via an online survey administered by the 4

Journal of Business Research xxx (xxxx) xxx–xxx

E. Moriuchi, I. Takahashi

Table 1 Measurement items and internal consistency.

PROMO1 PROMO2 PROMO4 SERVICE1 SERVICE2 SERVICE3 SERVICE4 PRICE1 PRICE2 PRICE3 PRICE4 SATIS2 SATIS3 SATIS4 TRUST1 TRUST2 TRUST4 REVIEW4 REVIEW 5 REVIEW 6

Aside from TV commercials and on the Internet, I do see ads for this supermarket at other places as well. I often see a lot of flyers and ads through postal mail and emails. This online supermarket is very active in their advertising efforts. As there isn't any nearby stores, it is convenient to have this online supermarket as a shopping alternative. The time it takes from placing an order to reviewing the product is short. The delivery service time is reasonable. The delivery service is very reliable The product prices in this online supermarket is lower than other online supermarkets. The products in this online supermarket is well priced. This online supermarket has a wide variety of discounted products. In comparison to other online supermarket, this online supermarket has the lowest prices on their products. I am satisfied with the products I received from this online supermarket I am satisfied with my current purchase I am satisfied with my shopping experience This online supermarket is trustworthy I feel that I can trust their safety precautions when I make an online transactions. I feel that I can trust this online supermarket's security precautions for any online payment. I intend to write a review on this online supermarket's website on its main website. I intend to write a review on the products I purchased on this online supermarket's website. I intend to write a review on my shopping experience on this online supermarket's website.

Coefficient alpha

CRa

AVEb

0.87

0.92

0.79

0.77

0.85

0.59

0.90

0.93

0.77

0.88

0.93

0.81

0.94

0.96

0.89

0.98

0.99

0.96

PROMO: Promotional activities; Price: Perception on price, price fairness, SATIS: Level of satisfaction with purchase, SERVICE: pre- and post-service quality, TRUST: level of trust in the evendor, REVIEW: review intentions, e-WOM. a CR: Composite Reliability refers to composite reliability is less biased estimate of reliability than Cronbach Alpha. b AVE: average variance extracted measures the level of variance captured by a construct versus the level due to measurement error.

model of 20 items exhibited an improved fit relative to the full model of 27 items with χ2 = 217.975, df = 155, p = 0.000; CFI = 0.99; RMSEA = 0.04; NFI = 0.95; and TLI = 0.98.

each measure; the composite reliability of each construct; and the average variance extracted for each construct and for inter-construct correlations. Convergent and discriminant validity levels were tested using the partial least square (PLS) path modeling method (Hair, Sarstedt, Hopkins, & Kuppelwieser, 2014) (see Table 5). Further, discriminant validity levels were measured using the Heterotrait-Monotrait Ratio (HTMT) quality criteria. A complete bootstrapping procedure was also used to check the HTMT inference. The maximum HTMT value was measured as 0.70, which is below the threshold of 0.90 (Henseler, Ringle, & Sarstedt, 2015). HTMT values of < 0.90 (vertically) denote that the true correlation between two constructs should differ. A confidence interval containing a value of 0.90 denotes a lack of discriminant validity. The structural model was then estimated to test the hypotheses (see Table 3). In this present study, a jackknife approach was applied whereby individual items were removed after the full model was estimated. The jackknife procedure was applied in five steps (Larwin & Harvey, 2012; Rensvold & Cheung, 1999). In the first step, fit statistics were calculated for the full dataset with all items considered. In step 2, the model was re-estimated K times (Larwin & Harvey, 2012) with each estimate based on the full model minus one of the items. With a different item removed, the model was re-estimated. In the third step, the resulting models were ranked and the best fit relative to the original full-item model was selected. From the CFI and RMSEA values, the best fit model was determined. In step 4, the procedure was repeated using the model selected for step 3. Finally, the final model was selected when the following conditions were met: when there were at least three observed variables (Bagozzi, 1980); when items were removed as long as the structural strength of the model was not violated (Bollen, 1989); and when the chosen model demonstrated a good fit (Bollen, 1989). For the 27-item model, this process involved the execution of 55 separate AMOS runs. Items derived from the item-deletion procedure that created the best fit model were based on the CFI and RMSEA estimates. The model was reduced seven items. The following items were removed: Promotions (PRO) 3, Service and Fulfillment (SERVICE) 5, Satisfaction (SATIS) 1, Trust (TRUST) 3 and Intentions to Review (REVIEW) 1, 2, and 3 (see Table 4). Each factor continued to converge on at least three observed variables (Bagozzi, 1980) and the integrity of the original model was maintained with the reduced model demonstrating a good fit. The final

5.2. Measurement model fit Factor loadings of the indicators for each construct were found to be statistically significant and high enough to demonstrate that the indicators and their underlying constructs were acceptable (Table 2). All loading estimates were found to be significant (p < 0.00) with the lowest measured at 0.73 and the highest measured at 0.99. Reliabilities and variances extracted for each latent variable show that the measurement model is reliable and valid. Computed from indicator standardized factor loadings and measurement errors (Hair et al., 2006), the average variance extracted (AVE) was found to range from 0.59 to 0.96 (Table 1). Table 2 Standard factor loading estimates. Promo Promo1 Promo2 Promo4 Price1 Price2 Price3 Price4 Service1 Service2 Service3 Service4 Satis2 Satis3 Satis4 Trust1 Trust2 Trust4 Review4 Review5 Review6 Variance extracted Constructed reliability

5

Price

Service

Satis

Trust

Review

0.87 0.91 0.88 0.89 0.89 0.84 0.90 0.82 0.72 0.81 0.73 0.94 0.83 0.93 0.95 0.95 0.93

79.3% 0.92

77.4% 0.93

59.2% 0.85

80.8% 0.93

88.5% 0.96

0.98 0.98 0.99 96.3% 0.99

Journal of Business Research xxx (xxxx) xxx–xxx

E. Moriuchi, I. Takahashi

factor (VIF) was calculated at 1.76, which is below the threshold of 5 (Hair, Ringle, & Sarstedt, 2011), and the tolerance level was measured at above 0.2. These results show strong evidence for the reliability and validity of the construct measures. The coefficient of determinations (R2) of the endogenous latent variables (Henseler, Ringle, & Sinkovics, 2009) was also examined. Percentages of the explained variance (R2) for intentions to review, satisfaction and trust were measured as 26.2, 32.8, and 48.7, respectively. Fig. 2 shows the results of the structural model test. In running the bootstrap analysis, we followed the procedure proposed by Hair et al. (2011). Our nonparametric bootstrap analysis of 5000 subsamples and 264 cases revealed the proposed relationships. All hypotheses except for H8 were supported. Structural Model parameters. Hypotheses H1, H2 and H3, which predicted the existence of a positive relationship between search attributes (promotions, price and services and fulfillment) and intentions to review, were supported. The results show that the path between these three constructs (services and fulfillment, promotions, and price) is indeed positive (β = 0.21; β = 0.32; β = 0.23, respectively) and significant (p < 0.01). In the analysis, we tested covariates such as gender and the frequency of purchases, but neither was found to have a significant effect on consumers' intentions to review. The fourth hypothesis (H4), which predicted the existence of a relationship between service and fulfillment and e-satisfaction, was supported (β = 0.54; p < 0.01). Hypothesis 5 (H5), which predicted the existence of a positive relationship between search attributes of prices and e-satisfaction, was supported (β = 0.09; p < 0.10). Hypothesis 6 (H6), which predicted the existence of a positive relationship between services and fulfillment and e-trust, was also supported (β = 0.35; p < 0.01). The seventh hypothesis (H7), which predicted the existence of a positive relationship between e-satisfaction and e-trust, was supported (β = 0.44; p < 0.01). The last hypothesis (H8), which predicted the existence of a positive relationship between e-trust and intentions to review, was not supported (β = − 0.15; p < 0.05). This hypothesis was not supported because the path shows a negatively significant relationship. Considering effects of the constructs on the intent to review, search attributes such as promotions exhibit the strongest direct impact

Table 3 Heterotrait-Monotrait ratio (HTMT).

Price Promotion Review Satis Service Trust

Price

Promotion

Review

Satis

Service

0.282 0.363 0.267 0.342 0.154

0.451 0.282 0.404 0.191

0.266 0.331 0.063

0.677 0.693

0.701

Trust

HTMT is the average of the heterotrait-heteromethod correlations relative to the average of the monotrait-heteromethod correlations (Henseler et al., 2015). HTMT values smaller than 0.90 (vertically) means that the true correlation between the two constructs should differ. The confidence interval containing the value 0.90 indicates a lack of discriminant validity.

Table 4 Online review intention model change in model fit.

All items (27 items) Reduced model (20 items) Δ online review intention model

×2

df

CFI

TLI

RMSEA

988.170 217.975 967.395

309 155 217

0.8895 0.985 0.104

0.880 0.979 0.115

0.091 0.045 0.046

5.3. Overall model fit Structural equation modeling for the theoretical model generated a value of 229.248 (df 158; p < 0.00), producing a GFI of 0.921, an adjusted GFI of 0.896, a CFI of 0.985, an RMSEA of 0.041, an NFI of 0.952, a TLI 0.982, and an x2/df of 1.44. The x2/df ratio of 1.44 denotes a good model fit. All relationships proposed by the theoretical model are significant except for one path, which is marginally significant (p = 0.10) for price and e-satisfaction (see Fig. 1). 5.4. Structural model fit To check for multi-collinearity problems, the variance inflation Table 5 Inter-construct correlations and the square root of AVE (Fornell-Larcker criterion).

Price Promotion Review Satisfaction Service Trust

SD

Mean

Price

Promotion

Review

Satisfaction

Service

Trust

3.20 2.88 2.55 2.17 2.61 2.17

2.92 2.86 3.41 2.98 3.55 3.93

0.880 0.252 0.343 0.249 0.288 0.147

0.891 0.420 0.244 0.331 0.171

0.981 0.245 0.288 0.060

0.899 0.566 0.635

0.770 0.597

0.941

On a scale of “1” Strongly disagree; “5” Strongly agree.

Fig. 1. Research framework.

H1 Services and Fulfillment

H6

H4 e-trust

e-satisfaction

Promotions

H8

H7

H2

H5 Price H3

6

Intention to Review

Journal of Business Research xxx (xxxx) xxx–xxx

E. Moriuchi, I. Takahashi

0.21*** Services and Fulfillment

0.35*** 0.54***

0.39***

e-trust

e-satisfaction

Promotions 0.21***

Intention to Review

-.15**.

0.44*** 0.32*** 0.26*** 0.09* Price 0.23*** Fig. 2. Research framework.

Table 6 Path coefficients and hypotheses. Hypothesis

Hypothesis 1

Hypothesis 2 Hypothesis Hypothesis Hypothesis Hypothesis Hypothesis Hypothesis

3 4 5 6 7 8

Service fulfillment ➔ Review intention Promotion ➔ Review intention Price ➔ Review intention Services ➔ e-satisfaction Price ➔ e-satisfaction Service ➔ e-trust e-satisfaction ➔ e-trust e-trust ➔ Review Intention

Table 7 Hypotheses summary and mediation results. Path coefficient

t-Value

Supported?

0.21⁎⁎⁎

3.116

Yes

0.32

⁎⁎⁎

⁎⁎⁎

0.23 0.54⁎⁎⁎ 0.09⁎ 0.35⁎⁎⁎ 0.44⁎⁎⁎ −0.15⁎⁎

5.832

Yes

3.998 11.547 1.613 4.883 7.031 2.303

Yes Yes Yes Yes Yes No

Mediation

Evidence

Mediation

β and p value Service Fulfillment- ➔ esatisfaction ➔ e-trust ➔ review intention

Price - ➔ e-satisfaction ➔ etrust ➔ review intention

< 0.10⁎, < 0.05⁎⁎, < 0.01⁎⁎⁎.

Service and Fulfillment ➔ etrust ➔ review intention

X ➔ M1

0.54⁎⁎⁎

M1 ➔ M2 M2 ➔ Y X➔Y X ➔ M1

0.44⁎⁎⁎ − 0.15⁎⁎ 0.21⁎⁎⁎ 0.10⁎

M1 ➔ M2 M2 ➔ Y X➔Y X ➔ M2

0.44⁎⁎⁎ − 0.15⁎⁎ 0.23⁎⁎⁎ 0.54⁎⁎⁎

M2 ➔ Y X➔Y

(β = 0.32). Prices (β = 0.23) as a search attribute were found to be slightly more influential than service fulfillment (β = 0.21). The indirect effect of e-satisfaction through e-trust on the intent to review was found to be strong (β = −0.15) (see Table 6).

Partial mediation

Partial mediation

Partial Mediation

⁎⁎

− 0.15 0.21⁎⁎⁎

X- Independent variable (service fulfillment, price); M1- 1st mediator (e-satisfaction); M22nd mediator (e-trust); Y- Dependent variable (review intention).

6. Mediation findings

This suggests the presence of partial mediation. The result shows that service fulfillment is directly related to e-satisfaction (std. β = 0.54, p < 0.01). E-satisfaction in turn leads to e-trust (β = 0.44, p < 0.01). For both service fulfillment (std. β = 0.54, p < 0.01) and prices (std. β = 0.09, p < 0.10), direct effects were found to be significant and even when mediators were included. This confirms that the relationship between services and promotions and intentions to review is partially mediated by e-satisfaction and e-trust. Finally, e-trust negatively predicts review intentions (std. β = −0.15, p < 0.05). Services and fulfillment were found to be mediated by e-trust (std. β = 0.35, p < 0.01), leading to intentions to review. Direct effects were found to be significant for this relationship (std. β = 0.21, p < 0.01) and to remain significant in the presence of a mediator (std. β = −0.15, p < 0.05). Furthermore, indirect effects between services and fulfillment and e-trust through intentions to review were found to be significant. This shows that e-trust partially mediates this relationship.

For mediation, Mathieu and Taylor's (2006) bootstrap method was used. We used 2000 bias-corrected (BC) bootstrapping samples at the 95 BC confidence level to determine the chain mediation effects (Shrout & Bolger, 2002). Two paths were hypothesized to affect intentions to review, both of which originated with services and fulfillment and prices. These paths were found to be mediated by two variables (e-satisfaction and e-trust) in a serial configuration. To achieve complete mediation, direct effects of the independent variable on the dependent variable must be significant; however, significance is eliminated in the presence of mediators. This indicates that the independent variable influences the dependent variable through mediators. On the other hand, to observe partial mediation, direct effects of the independent variable on the dependent variable must be significant and this significance is not eliminated in the presence of mediators. The mediated results are summarized in Table 7. Using the search attributes of services and fulfillment, promotions and prices as exogenous variables, the results show that serial mediation is present. Direct effects of the independent variables on the dependent variables were found to be significant for service fulfillment and promotions and this significance remained when mediators were introduced.

7. Discussion The purposes of this study were as follows: to examine relationships between antecedents of e-vendor satisfaction and to investigate the integrated model on repeat online consumers. The findings support this relationship and the relationship between e-satisfaction and e-trust. 7

Journal of Business Research xxx (xxxx) xxx–xxx

E. Moriuchi, I. Takahashi

However, the findings do not support the relationship between e-trust and review intentions. A positive relationship between the three search attributes (services and fulfillment, prices and promotions) and the intent to review was substantiated. Positive relationships between the search attributes services and fulfillment, e-satisfaction and price were supported. Although services and fulfillment had a stronger impact on e-satisfaction, both search attributes are the motivating factors behind e-satisfaction. This result is consistent with Werben and Verkaufen (2003), who found that price information is an important aspect of e-satisfaction. Park and Kim (2003) also support this argument in finding that prices constitute an important factor that shape purchasing behaviors online. This suggests that even though these online customers are repeat customers, search attributes such as service fulfillment and prices still play an important role in consumers' levels of e-satisfaction. Furthermore, Frank (1962) added that when consumers are engaged in habitual purchasing behavior such as repeat purchases, this can be viewed as a sign of brand loyalty. We can also assume that even when price information is not available, as long as good service quality levels are present, consumers will continue to shop through an online supermarket. Although we must take into consideration consumers' price sensitivities, the results generally imply that there is a correlation between consumer satisfaction and the perceived fairness of pricing. The relationship between e-satisfaction and e-trust was also supported. This causal relationship proposed in the literature on offline parent branding was supported for our sample based on the online supermarket context. Surprisingly, e-trust was found to have a negative effect on intentions to review. This negative relationship does not involve contributing a negative comment and rather is characterized by the fact that as a shopper develops more e-trust, they are less inclined to review their shopping experiences. In our attempts to explain the causes of these unexpected results, we turn our focus to three issues: purchasing behaviors (i.e., habitual), the norm of using eWOM communications as a communication tool and the Japanese cultural habits of the study respondents. First, the respondents who participated in this survey are repeat customers. Consequently, while they were likely more active at first, their motivations to write reviews had likely declined as they made habitual purchases. Alternatively, as consumer trust increased, the respondents' need to review an e-vendor either positively or negatively likely decreased. Second, to better understand this negative relationship, we ran a frequency analysis to determine the number of consumers who were willing to review their shopping experiences (REVIEW 6) via eWOM. These consumers accounted for approximately 20% of the respondents. Interestingly, the results show that respondents who were willing to spread positive information about the e-vendor (REVIEW 1) accounted for 45% of the respondents. This 45% is mostly categorized as general WOM communication. Scores of 4 and 5 (agree and strongly agree, respectively) were used for both variables. This discrepancy suggests that Japanese customers do not use online review platforms or modalities to spread the word. This may be because in brick-and-mortar Japanese grocery stores, staff and consumers do not frequently interact. Thus, we argue that these Japanese consumers may not (yet) be comfortable with using online review sites to interact with online grocery stores (e.g., through online reviews). They may still prefer to spread information (e.g., WOM) through other means (e.g., face-to-face communication or social media chat apps). This suggests that online review tools are yet to be fully utilized to satisfy communication between e-vendors (e.g., online supermarkets) the consumers. Finally, we considered the fact that the consumers studied are Japanese. In Japan, individuals are not comfortable giving positive feedback. Kopp (2013) noted that the Japanese view praise as a form of sarcasm. The term homegoroshi, which literally means “to kill with compliments,” is used to describe this perception. In other words, positive feedback could be viewed as effusive and negative. In a similar vein, Japanese consumers may not be motivated to compliment an e-vendor because

they may receive this gesture as sarcastic rather than positive. This unexpected negative regression coefficient is likely the result of a suppression effect. A suppression effect is likely to be present when direct and mediated effects of an independent variable on a dependent variable have opposite signs (Cliff & Earleywine, 1994; Tzelgov & Henik, 1991). According to MacKinnon, Krull, and Lockwood (2000), in a mediation model the relationship between the independent variable and dependent variable is reduced because the mediator explains part or all of the relationship because it occupies the causal path between the independent and dependent variables. 8. Managerial implications We can make a couple of practical recommendations to marketing managers and practitioners based on the results of this study. First, for online contexts it is important for managers to be aware of the usefulness of consumers' intentions to review (e.g., eWOM), as reviews can facilitate consumers' abilities to obtain information, which decreases search costs (Liu & Park, 2015). Online reviews have dual roles: 1) they not only allow consumers to obtain information but also allow consumers to make anonymous comments on products that they have purchased. 2) These reviews empower e-vendors with information on ways to improve. When dual communication is used, consumers are more likely to think highly of a particular e-vendor. In turn, by developing a strong reputation, a company can be made less vulnerable to competitive market actions can gradually increase its market share through an increase in repeat purchases (Chaudhuri & Holbrook, 2001; Lau & Lee, 2000). Second, we found e-trust to be negatively related to the intent to review. While we at first wondered whether this result could have stemmed from Japanese cultural values of humility, there is another explanation for such an unexpected result: the nature of habitual purchases and the use of eWOM. We suggest that e-vendors be cognizant of the feedback that they seek and of the culture of the target market. Feedback could be requested with incentives to educate online consumers that feedback is integral to a company's growth. It is thus reasonable to assume that once consumers have made repeat purchases, a level of familiarity has been established with the e-vendor, thus maybe more reluctant to review a product when it meets their expectations. On the other hand, due to this established level of trust (i.e., an expectation has been set), when a product that repeat consumers have purchased does not meet expectations, trust levels will decrease, thus motivating such consumers to vent their dissatisfaction by writing negative reviews online. The results of this study suggest that e-vendors may choose to encourage reviews by providing extrinsic benefits such as discount coupons or even exclusive sale offers to those who review items that they have purchased. With fake reviews on the rise (Streitfeld, 2013), evendors may also need to apply verification systems through which only verified customers are able to review items that they have purchased. 9. Limitations and future research Limitations of the present study must be acknowledged. First, the study results are based on Japanese online consumers. Before generalizations can be made without caution, this research must be extended to other ethnic consumer markets. Another limitation concerns the examination of an alternative model. Given the unexpected negative regression coefficient found for the relationship between e-trust and intentions to review, the reliability of the results must be reconsidered. Thus, an alternative model based on a different variable should be used to reaffirm the reliability of the results. Our model of Japanese repeat online consumers has been validated, and it documents the online review intentions of Japanese online consumers. However, this model should be validated in reference to other ethnic groups (e.g., Chinese). For example, Werben and 8

Journal of Business Research xxx (xxxx) xxx–xxx

E. Moriuchi, I. Takahashi

customer satisfaction and loyalty on e-marketing: Moderating effect of perceived value. Total Quality Management, 20(4), 423–443. Chaudhuri, A., & Holbrook, M. B. (2001). The chain of effects from brand trust and brand affect to brand performance: The role of brand loyalty. Journal of Marketing, 65(2), 81–93. Chevalier, J. A., & Mayzlin, D. (2006). The effect of word of mouth on sales: Online book reviews. Journal of Marketing Research, 43(3), 345–354. Chiang, K. P., & Dholakia, R. R. (2003). Factors driving consumer intention to shop online: An empirical investigation. Journal of Consumer Psychology, 13(1), 177–183. Childers, T. L., Carr, C. L., Peck, J., & Carson, S. (2002). Hedonic and utilitarian motivations for online retail shopping behavior. Journal of Retailing, 77(4), 511–535. Cliff, N., & Earleywine, M. (1994). All predictors are ‘mediators’ unless the other predictor is a ‘suppressor’. unpublished manuscriptUniversity of Southern California. Coulter, K. S., & Coulter, R. A. (2002). Determinants of trust in a service provider: The moderating role of length of relationship. Journal of Services Marketing, 16(1), 35–50. Cronin, J. J., Brady, M. K., & Hult, G. T. M. (2000). Assessing the effects of quality, value, and customer satisfaction on consumer behavioral intentions in service environments. Journal of Retailing, 76(2), 193–218. Day, G. S. (1981). The product life cycle: Analysis and applications issues. Journal of Marketing, 60–67. Decker, R., & Trusov, M. (2011). Estimating aggregate consumer preferences from online product reviews. International Journal of Research in Marketing, 27(4), 293–307. Degeratu, A. M., Rangaswamy, A., & Wu, J. (2000). Consumer choice behavior in online and traditional supermarkets: The effects of brand name, price, and other search attributes. International Journal of Research in Marketing, 17(1), 55–78. Dick, A., Chakravarti, D., & Biehal, G. (1990). Memory-based inferences during consumer choice. Journal of Consumer Research, 17, 82–93. Evanschitzky, H., Iyer, G. R., Hesse, J., & Ahlert, D. (2004). E-satisfaction: A re-examination. Journal of Retailing, 80(3), 239–247. Flavián, C., et al. (2006). The role played by perceived usability, satisfaction and consumer trust on website loyalty. Information Management, 43(1), 1–14. Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 39–50. Frank, R. E. (1962). Brand choice as a probability process. The Journal of Business, 35(1), 43–56. Fukuyama, F. (1995). Social capital and the global economy: A redrawn map of the world. Foreign Affairs, 74(5), 89–103. Ganesan, S. (1994). Determinants of long-term orientation in buyer-seller relationships. Journal of Marketing, 1–19. Garbarino, E., & Johnson, M. S. (1999). The different roles of satisfaction, trust, and commitment in customer relationships. Journal of Marketing, 70–87. Gefen, D. (2002). Reflections on the dimensions of trust and trustworthiness among online consumers. ACM Sigmis Database, 33(3), 38–53. Gefen, D., & Pavlou, P. (2006). The moderating role of perceived regulatory effectiveness of online marketplaces on the role of trust and risk on transaction intentions. ICIS 2006 proceedings (pp. 81). . Gefen, D., Karahanna, E., & Straub, D. W. (2003). Trust and TAM in online shopping: An integrated model. MIS Quarterly, 27(1), 51–90. Gremler, D. D., Gwinner, K. P., & Brown, S. W. (2001). Generating positive word-ofmouth communication through customer-employee relationships. International Journal of Service Industry Management, 12(1), 44–59. Gruen, T. W., Osmonbekov, T., & Czaplewski, A. J. (2006). eWOM: The impact of customer-to-customer online know-how exchange on customer value and loyalty. Journal of Business Research, 59(4), 449–456. Ha, S., & Stoel, L. (2012). Online apparel retailing: Roles of e-shopping quality and experiential e-shopping motives. Journal of Service Management, 23(2), 197–215. Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E., & Tatham, R. L. (2006). Multivariate data analysis. Vol. 6. Upper Saddle River, NJ: Pearson Prentice Hall. Hair, J. F., Ringle, C. M., & Sarstedt, M. (2011). PLS-SEM: Indeed a silver bullet. Journal of Marketing Theory and Practice, 19(2), 139–152. Hair, J., Jr., Sarstedt, M., Hopkins, L., & Kuppelwieser, G. V. (2014). Partial least squares structural equation modeling (PLS-SEM) an emerging tool in business research. European Business Review, 26(2), 106–121. Heintzman, R., & Marson, B. (2005). People, service and trust: Is there a public sector service value chain? International Review of Administrative Sciences, 71(4), 549–575. Henseler, J., Ringle, C. M., & Sinkovics, R. R. (2009). The use of partial least squares path modeling in international marketing. Advances in International Marketing, 20(1), 277–319. Henseler, J., Ringle, C. M., & Sarstedt, M. (2015). A new criterion for assessing discriminant validity in variance-based structural equation modeling. Journal of the Academy of Marketing Science, 43(1), 115–135. Horppu, M., Kuivalainen, O., Tarkiainen, A., & Ellonen, H. K. (2008). Online satisfaction, trust and loyalty, and the impact of the offline parent brand. Journal of Product & Brand Management, 17(6), 403–413. Itoh, M. (1991). The Japanese distribution system and access to the Japanese market. In P. Krugman (Ed.). Trade with Japan: Has the door opened wider? (pp. 175–190). University of Chicago Press. Itoh, M. (2000). In T. Ito, & A. O. Krueger (Eds.). Competition in the Japanese distribution market and market access from abroad (pp. 139–156). University Chicago Press. Jarvenpaa, S. L., Tractinsky, N., et al. (2000). Consumer trust in an internet store. Information Technology & Management, 1(1), 45–71. Keh, H. T., & Lee, Y. H. (2006). Do reward programs build loyalty for services?: The moderating effect of satisfaction on type and timing of rewards. Journal of Retailing, 82(2), 127–136 (Chicago). Kim, H. W., Xu, Y., & Koh, J. (2004). A comparison of online trust building factors between potential customers and repeat customers. Journal of the Association for

Verkaufen (2003) showed that price information has a stronger impact on the purchase intentions of German consumers than product information. On this basis, it can be hypothesized that financial risk overshadows social risk. Moreover, are our respondents were Japanese and thus exhibit consumption behaviors that are vastly different from consumers of other societies (Synodinos, 2001), a comparative study on repeat online consumers' intentions to review in other countries could be conducted to extend this study. Differences found based on online consumers' ethnicities regarding intentions to review online will yield insights that can help e-vendors better acquire and retain consumers on a global scale. Finally, the proposed model did not take into consideration other important individual or external factors in terms of degrees of willingness to review online shopping experiences. These individual levels are inner- or other-oriented (e.g., inter vs. intra) (e.g., a consumer's need to gain social acceptance versus self-approval or to be an opinion leader in order to become a market maven) are related to the levels of product involvement (e.g., low vs. high). As another limitation, future research could examine the frequency of complaints made on products purchased online. This measure may serve as a strong indicator of online shoppers' degrees of willingness to review in the future. For external factors, a construct that examines the organization of websites may prove beneficial. For example, a research framework may include a construct that measures online shoppers' perceptions of the design of a website (i.e., whether it facilitates, promotes and implements a rewards system for online shoppers). These factors may be included in a research model to help explain the negative relationship found between etrust and intentions to review. As discussed above on the use of eWOM communication in the form of online reviews (e.g., company websites), future research could also examine the thought processes of Japanese online consumers. In other words, it would be beneficial to investigate the psychological attributes that affect consumers' decisions to write reviews on their shopping experiences online. In understanding these psychological attributes, online grocery stores can better serve their customers by becoming more aware of their customers' needs and wants through these written “voices” (i.e., online reviews). Due to the presence of a suppression effect in the mediation model, a replicated study could be conducted to clarify whether the existing variable is a suppressor or mediator. If the variable is a mediator, as stated by MacKinnon et al. (2000) changes in the dependent variable should be specific to changes in the mediator and not to changes in others. If changes are not specific to the mediator, then a Type I error will have been identified. References Anderson, J. C., & Gerbing, D. W. (1988). Structural equation modeling in practice: A review and recommended two-step approach. Psychological Bulletin, 103(3), 411. Anderson, R. E., & Srinivasan, S. S. (2003). E-satisfaction and e-loyalty: A contingency framework. Psychology and Marketing, 20(2), 123–138. Atchariyachanvanich, K., et al. (2007). What keeps online customers repurchasing through the internet? ACM SIGecom Exchanges, 6(2), 47–57. Aurier, P., & N'Goala, G. (2010). The differing and mediating roles of trust and relationship commitment in service relationship maintenance and development. Journal of the Academy of Marketing Science, 38(3), 303–325. Bagozzi, R. P. (1980). Performance and satisfaction in an industrial sales force: An examination of their antecedents and simultaneity. Journal of Marketing, 65–77. Balabanis, G. M., Reynolds, N., & Simintiras, A. (2001). Bases of e-store loyalty: Perceived switching barriers and satisfaction. Journal of Business Research, 59(2), 214–224. BBC News. Japan country profile- overview. (2015). Retrieved from https://ustr.gov/ countries-regions/japan-korea-apec/japan (February 17). Bollen, K. A. (1989). A new incremental fit index for general structural equation models. Sociological Methods & Research, 17(3), 303–316. Burke, R. R. (1997). Do you see what I see? The future of virtual shopping. Journal of the Academy of Marketing Science, 25(4), 352–360. Burke, R. R., Harlam, B. A., Kahn, B. E., & Lodish, L. M. (1992). Comparing dynamic consumer choice in real and computer-simulated environments. Journal of Consumer Research, 19(1), 71–82. Cantallops, A. S., & Salvi, F. (2014). New consumer behavior: A review of research on eWOM and hotels. International Journal of Hospitality Management, 36, 41–51. Chang, H. H., Wang, Y. H., & Yang, W. Y. (2009). The impact of e-service quality,

9

Journal of Business Research xxx (xxxx) xxx–xxx

E. Moriuchi, I. Takahashi

Research and Applications, 7(4), 399–410. Peterson, R. A., Balasubramanian, S., & Bronnenberg, B. J. (1997). Exploring the implications of the internet for consumer marketing. Journal of the Academy of Marketing Science, 25(4), 329–346. Ranaweera, C., & Prabhu, J. (2003). On the relative importance of customer satisfaction and trust as determinants of customer retention and positive word of mouth. Journal of Targeting, Measurement and Analysis for Marketing, 12(1), 82–90. Ratchford, B. T. (1982). Cost-benefit models for explaining consumer choice and information seeking behavior. Management Science, 28(2), 197–212. Reichheld, F. F., & Schefter, P. (2000). E-loyalty. Harvard Business Review, 78(4), 105–113. Rensvold, R. B., & Cheung, G. W. (1999). Identification of influential cases in structural equation models using the jackknife method. Organizational Research Methods, 2(3), 293–308. Ribbink, D., Van Riel, A. C., Liljander, V., & Streukens, S. (2004). Comfort your online customer: Quality, trust and loyalty on the internet. Managing Service Quality: An International Journal, 14(6), 446–456. Salsberg, B. (2010). The new Japanese consumer. Retrieved from http://www.mckinsey. com/industries/consumer-packaged-goods/our-insights/the-new-japaneseconsumer. Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and nonexperimental studies: New procedures and recommendations. Psychological Methods, 7(4), 422. Storbacka, K., Strandvik, T., & Grönroos, C. (1994). Managing customer relationships for profit: The dynamics of relationship quality. International Journal of Service Industry Management, 5(5), 21–38. Streitfeld, D.. The New York Times. Give yourself 5 stars? Online, it might cost you. (2013). Retrieved from: http://www.nytimes.com/2013/09/23/technology/give-yourself-4stars-online-it-might-cost-you.html?_r=0 (Sept. 22) . Synodinos, N. E. (2001). Understanding Japanese consumers: Some important underlying factors. Japanese Psychological Research, 43(4), 235–248. Takahashi, I., & Fluch, H. (2009). Retailing in Japan: Overview and key trends. European Retail Research, 23(2), 145–162. Tsutsui, W. M. (2009). A companion to Japanese history. John Wiley & Sons. Tzelgov, J., & Henik, A. (1991). Suppression situations in psychological research: Definitions, implications, and applications. Psychological Bulletin, 109(3), 524. Wang, M. (2003). Assessment of e-service quality via e-satisfaction in e-commerce globalization. Electronic Journal on Information Systems in Developing Countries, 11, 10. Werben and Verkaufen (2003). Verbraucher in Deutschland sind mit Service unzufrieden. Retrieved from http://www.wuv.de/news/artikel/2003/10/18684/index.html. Wolfinbarger, M., & Gilly, M. C. (2003). eTailQ: Dimensionalizing, measuring and predicting retail quality. Journal of Retailing, 79(3), 183–198. Yang, Z., & Peterson, R. T. (2004). Customer perceived value, satisfaction, and loyalty: The role of switching costs. Psychology & Marketing, 21(10), 799–822. Yoon, S. (2002). The antecedents and consequences of trust in online-purchase decisions. Journal of Interactive Marketing, 16(2), 47–63. Zhu, F., & Zhang, X. (2010). Impact of online consumer reviews on sales: The moderating role of product and consumer characteristics. Journal of Marketing, 74(2), 133–148.

Information Systems, 5(10), 392–420. Kopp, R. (Mar 28, 2013). Homegoroshi-Japanese don't always like positive feedback. Retrieved from http://www.japanintercultural.com/en/news/default.aspx?newsID= 258. Larwin, K., & Harvey, M. (2012). A demonstration of a systematic item-reduction approach using structural equation modeling. Practical Assessment, Research & Evaluation, 17(8), 1–19. Lau, G. T., & Lee, S. H. (2000). Consumer's trust in a brand and the link to brand loyalty. Journal of Market Focused Management, 4(4), 341–370. Liu, Y. (2006). Word-of-mouth for movies: Its dynamics and impact on box office revenue. Journal of Marketing, 70(3), 74–89. Liu, Z., & Park, S. (2015). What makes a useful online review? Implication for travel product websites. Tourism Management, 47, 140–151. Luhmann, N. (1979). Trust and power. Chichester: Wiley. MacKinnon, D. P., Krull, J. L., & Lockwood, C. M. (2000). Equivalence of the mediation, confounding and suppression effect. Prevention Science, 1(4), 173–181. Mahajan, V., Muller, E., & Kerin, R. A. (1984). Introduction strategy for new products with positive and negative word-of-mouth. Management Science, 30(12), 1389–1404. Marshall, A. (2009). Principles of economics: Unabridged (8th edition). Inc: Cosimo. Mathieu, J. E., & Taylor, S. R. (2006). Clarifying conditions and decision points for mediational type inferences in organizational behavior. Journal of Organizational Behavior, 27(8), 1031–1056. McCole, P., et al. (2010). Trust considerations on attitudes towards online purchasing: The moderating effect of privacy and security concerns. Journal of Business Research, 63(9–10), 1018–1024. McKnight, D. H., Choudhury, V., & Kacmar, C. (2002). The impact of initial consumer trust on intentions to transact with a web site: A trust building model. Journal of Strategic Information Systems, 11(3), 297–323. McKnight, D. H., et al. (2004). Shifting factors and the ineffectiveness of third party seals: A two-stage model of initial trust in a web business. Electronic Markets, 14(3), 252–266. Mittal, V., & Kamakura, W. A. (2001). Satisfaction, repurchase intent, and repurchase behavior: Investigating the moderating effect of customer characteristics. Journal of Marketing Research, 38(1), 131–142. Morgan, R. M., & Hunt, S. D. (1994). The commitment-trust theory of relationship marketing. Journal of Marketing, 58(July), 20–38. Moriuchi, E. (2016). Social media marketing: Strategies in utilizing consumer-generated content. New York: Business Expert Press. Ohanian, R., Tashchian, A., & Beard, L. H. (1992). Constructing marketing examinations that are reliable and valid measures of student performance. Marketing Education Review, 2(1), 14–20. Oliver, R. L. (1997). Satisfaction: A behavioral perspective on the customer. New York: IrwinMcGraw-Hill. Park, C., & Kim, Y. (2003). Identifying key factors affecting consumer purchase behavior in an online shopping context. International Journal of Retail & Distribution Management, 31(1), 16–29. Park, D. H., & Kim, S. (2009). The effects of consumer knowledge on message processing of electronic word-of-mouth via online consumer reviews. Electronic Commerce

10