Assessing customers’ perceived value of the online channel of multichannel retailers: A two country examination

Journal of Retailing and Consumer Services 27 (2015) 90–102

Contents lists available at ScienceDirect

Journal of Retailing and Consumer Services journal homepage: www.elsevier.com/locate/jretconser

Assessing customers’ perceived value of the online channel of multichannel retailers: A two country examination Jamie Carlson, Aron O’Cass, Dennis Ahrholdt Newcastle Business School, University of Newcastle, Level 3, University House, Corner King and Auckland Streets, Newcastle 2300, Australia School of Management, Faculty of Business, University of Tasmania, Hobart, 7001 Tasmania, Australia Department of Marketing & Sales, HSBA Hamburg School of Business Administration, Alter Wall 38, 20457 Hamburg, Germany

art ic l e i nf o

a b s t r a c t

Article history: Received 22 January 2015 Received in revised form 21 July 2015 Accepted 21 July 2015

Given the constant and often dramatic technological advancements that facilitate retailing, especially online channel adoption by industry and customers, research focusing on customers perceived value of the online channel of multi-channel retailer’s requires conceptual and empirical elaboration. This study advances understanding of customer perceived online channel value and how customer perceptions of value effect online channel satisfaction and online channel loyalty. Using data from a multi-country study, we provide a deeper understanding for multi-channel retailers of how to balance investments in various value drivers to enhance online channel satisfaction and customer loyalty. & 2015 Elsevier Ltd. All rights reserved.

Keywords: Customer perceived value Online channel Multi-channel retail Satisfaction Loyalty Multi-country research

1. Introduction With the continued adoption of disruptive technologies by firms and customers in retailing, bricks and mortar retail firms with e-commerce capabilities must give more attention to the performance of their online channel in order to enhance the multichannel customer experience (Ahrholdt, 2011; Badrinarayanan et al., 2014). Online sales grew 21.1% for 145 of the ‘Top 250’ retailers globally with e-commerce operations (Deloitte, 2015) and U.S. retail e-commerce sales (in 2014 representing 8.9% of total sales) are estimated to increase from $263 billion in 2013 to $414 billion in 2018 (Forrester Research, 2014). It is projected that as online sales accelerate, leading bricks and mortar retailers (Walmart, Nordstrom, Woolworths) are viewing e-commerce as a key element of their expansion (Deloitte, 2015) with the “brickand-click” business model growing because of advances in the integration of retail processes across multiple channels. Such seamless integration across channels now allows retailers, more than ever before, the ability to provide greater benefits and innovative services to customers (Oh et al., 2012). In line with these developments, researchers studying retail channel management continue to argue that multi-channel retailers need to better understand consumer decision making in E-mail addresses: [email protected] (J. Carlson), [email protected] (A. O’Cass), [email protected] (D. Ahrholdt). http://dx.doi.org/10.1016/j.jretconser.2015.07.008 0969-6989/& 2015 Elsevier Ltd. All rights reserved.

technology mediated environments across country markets (Badrinarayanan et al., 2012). In particular, researchers have called for research to understand how consumers perceive value by taking into account the idiosyncratic nature of the online channel and cross-channel effects (Maity and Dass, 2014; Yang et al., 2013; Zhang et al., 2010) when making purchases across channels. This has become a central issue for practitioners and researchers since consumers make judgements regarding the presentation of a coherent brand across other retail channels (Carlson and O’Cass, 2011; Badrinarayanan et al., 2012), as well as how the online channel provides advantage (e.g. convenience) in comparison to other channels of the retailer (Banerjee, 2014; Verhoef et al., 2009). Despite the growth in retail firms moving to “bricks-andclicks”, explicit research on customer perceptions of value relating to the online channel of multi-channel retailing has remained underexposed. For instance, perceived value research within the online environment so far has adopted a single-channel mindset restricted to understanding customer value assessments in an e-retail context only (e.g. Barrutia and Gilsanz 2013; Chiu et al., 2014; Wetzels et al., 2009) which inherently does not consider cross-channel interactions that consumers value with a multichannel retailer’s physical channel. Therefore, there is a need for a more in-depth understanding of the specific drivers of value creation in the online channel of multi-channel retailers. Advancing understanding will help multi-channel retailers of all size

J. Carlson et al. / Journal of Retailing and Consumer Services 27 (2015) 90–102

effectively allocate their resources to enhance online channel value perceptions. In order to overcome this gap in the literature, this study advances customer perceived value theory in online environments within a multi-channel retail context addressing three key areas. First, we extend the multi-dimensional approach to the formulation of online channel value in a multi-channel business-toconsumer retail context to capture the distinct components that reflect benefits the customer receive that facilitates the achievement of goals in this setting, particularly benefits which create value for customers. Specifically, our investigation focuses on value assessments by the customer arising from the performance of the online channel together with important cross-channel interactions which considers value judgements regarding the presentation of a consistent retail brand across channels, and how the online channel provides convenience in comparison to other channels of the retailer. Accordingly, we clarify the conceptual components to advance a model of perceived online channel value (labeled POCVAL), grounded in perceived value-in-use theory, which is a formative second-order construct comprising five first-order components which together constitute POCVAL. Second, given the growing focus on integrating the role of customers and technology for value measurement and optimization to enhance customer centricity objectives (Ostrom et al., 2015; Rust and Huang, 2014; Verhoef and Lemon, 2013), we identify outcomes of POCVAL and examine the effect of customers POCVAL evaluations on the customer satisfaction and online channel loyalty intention in multi-channel retailing. We also investigate the mediating role of customer satisfaction intervening between POCVAL and online channel loyalty to provide a more complete picture of the value creation process and its consequences. Finally, we conduct our study in a cross-country setting to validate the proposed conceptualization of POCVAL. In an effort to better understand and model how consumers make value assessments when engaging with online channels within multichannel retailing which increasingly transforms retailing into a global retail landscape, we study two developed economies. As such, data from a cross-country setting is used to substantiate the theory and our developed model, respectively, which is supported by empirical data from two country market settings.

2. Theoretical framework The development of the theoretical framework and relevant hypotheses are discussed in three phases. First, the background of customer perceived value theory is discussed which acts as the theoretical underpinning of the study. Second, the dimensionality of POCVAL in the context of multi-channel retailing is examined followed by its conceptualization and operationalization. Third, the interrelationships between POCVAL, online channel satisfaction, and online channel loyalty intentions are examined. 2.1. Customer perceived value Although perceived value has received growing attention, researchers seem to hold divergent views and definitions of it. Value has been defined as “consumer’s overall assessment of the utility of a product based on perception of what is received and what it is given” (Zeithaml, 1988, p. 4). In this context it has been treated as a uni-dimensional construct including give (e.g., payment/effort) and take (e.g., quality) aspects together. However, studies addressing customer value have suggested that the construct is too complex to be operationalized as uni-dimensional and solely utilitarian-based (e.g. O’Cass and Ngo, 2011; Woodall, 2003) where it

91

loses conceptual richness. Consequently, an approach which considers customer value as a concept with multiple components (e.g. Sheth et al., 1991; Holbrook, 1994; De Ruyter et al., 1997; Sweeney and Soutar, 2001) seems more adequate. Many conceptualizations consider functional, as well as hedonic components which helps overcome the overly restrictive concentration on economic value. In moving away from the traditional value conceptualization (cost-benefit), the multi-dimensional approach echoes the growing relevance of emotional and experience facets in consumer behavior research (Sánchez-Fernández et al., 2009). Taking the multidimensional view of value, Woodruff (1997, p. 142) defines customer value as “a customer’s perceived preference for and evaluation of those product attributes, attribute performances, and consequences arising from use that facilitate (or block) achieving the customer’s goals and purposes in use situations”. Others also view customer value as the customer’s perception of the benefit or advantage arising out of their association with an organization’s offering (Bradley and Sparks, 2012; Holbrook, 1999; Woodall, 2003). As such, value is directly related to the benefits one receives from a product or service and encompasses two domains – outcomes and processes. Customer perceived value encompasses outcomes which reflect the nature of the end-state – what an individual wants or their ultimate goal (i.e., benefits received). An outcome is valued in this sense to the extent that the object is useful, satisfies a need, or solves a problem. It also encompasses process which reflects the experience during the activity driving goal pursuit. Perceived value is seen a more broad, including any factor that affects the individual’s experience during goal pursuit (Higgins and Scholar, 2009). This view is also consistent with Grönroos (2011) who states that, value for the consumer is created or emerges during the consumption experience arising from direct interactions with the firm. Therefore, value is accumulated (or destroyed) within the value creation process. In this sense, value (and its components representing the benefits derived from service) is not determined solely at the end of the process, but during the interactions between the customer and the firm. Other frameworks have been advanced on the premise that value is complex and consists of multiple components that deliver benefits to customers. Sheth et al. (1991) propose five components of customer value – epistemic, social, functional, emotional, and conditional where a decision may be influenced by one or all of the consumption values. Their study serves as a framework for research conducted by De Ruyter et al. (1997) and later by Sweeney and Soutar (2001) who advance the notion of perceived value or PERVAL. This collective work provides the foundation for extending customer perceived value focusing on the benefits perceived by customers in the Internet retailing environment, which has been advanced over the past decade. Other views on value have also been advanced into the literature. For instance, the work of Hirschman and Holbrook (1982) and later Holbrook (1994) differentiates people based on their consumption motives. Hirschman and Holbrook (1982) describe consumers as either problem solvers or seekers of fun and enjoyment, and thus refer to utilitarian vs. hedonic consumption. The hedonic view highlights the importance of a fun experience in contrast to the effective achievement of a utilitarian goal. Holbrook (1994) further postulates that consumption experiences most likely involve more than one type of value simultaneously. Mathwick et al. (2002) were the first to introduce a multidimensional scale of experiential value in an online purchasing context. They define online experiential value as “a perceived, relativistic preference for product attributes or service performances arising from interaction within a consumption setting that facilitates or blocks achievement of customer goals or purposes” (p. 53) which may be experienced as the consequence of active or

92

J. Carlson et al. / Journal of Retailing and Consumer Services 27 (2015) 90–102

reactive interaction with the product, service or marketing entity. In their work, Mathwick et al. (2001) also distinguish playfulness, aesthetics, customer return on investment (CROI), and service excellence as sources of value which serve as retail performance indicators. In a similar fashion research has shown that online shoppers value both utilitarian and hedonic experiences from an internet retailer (Chiu et al., 2014; Overby and Lee, 2006). Further, other studies in the internet retailing domain have advanced a more expansive set of value components beyond the dual concepts of utilitarian and hedonic value. Cheng et al. (2009) incorporate functional, social, emotional, and epistemic value components which explain purchase intentions of internet retailing. Wetzels et al. (2009) use Mathwick et al.’s (2001) conceptualization to empirically illustrate the assessment of a reflective, fourth-order hierarchical model where hedonic value (reflecting the secondorder latent variables aesthetics and playfulness) and utilitarian value (reflecting the second-order latent variables service excellence and CROI) are specified as third-order latent variables and experiential value as a fourth-order latent variable reflecting utilitarian and hedonic value. While these studies make important contributions to knowledge on customer perceived value in the Internet only retailing environment (i.e. the online channel), several shortcomings remain. In particular, the lack of consensus over the number and nature of value components, and the absence of studies which consider value judgements by consumers which account for those online channels of retailers which also have a physical retail store presence. In this sense, greater conceptual clarity and robust modeling is needed which articulates customer perceived value dimensions critical in their overall assessment of the online channel’s delivery of value set within the multichannel retail context. Therefore, adopting and extending a multidimensional approach to online channel value in the context of multichannel retailing in a cross-country setting, the present study advances a conceptualization of POCVAL-comprising five components which are perceived by the customer and which provide benefits to them. This being the case, extending and integrating the work of Woodruff (1997), Sweeney and Soutar (2001), and Bradley and Sparks (2012) where value is derived from an interaction in a consumption setting, with the work of Mathwick et al. (2001) and Wetzels et al. (2009) where value is perceived and experienced in an online consumption setting, we advance a definition of POCVAL as “A personal perception of benefits or advantages arising from interactions with technology driven service processes in the online channel of a multichannel retailer that facilitates achievement of customer goals or purposes”. The following section describes each value component of the POCVAL conceptualization which comprises both cognitive and affective components. 2.1.1. Perceived online channel value 2.1.1.1. Service performance value. In services, the delivery of specific service attributes is critical for delivering superior performance value for the customer (Mittal and Sheth, 2001). For example, in the context of multichannel retailing, this evaluation includes assessment of the superiority of functional performance relating to obtaining dependable and accurate information to make smart purchase decisions, the breadth and depth of product assortment, receiving personalized attention, and efficient purchasing processes in the online channel which takes place specifically via a variety of technology driven service processes on the website platform (i.e. the online channel) (Barrutia and Gilsanz, 2013; O’Cass and Carlson, 2012). On this basis, this study conceptualizes online channel service performance value as the utility derived by the customer arising from the online channel delivering high quality, dependable and innovative service features.

2.1.1.2. Emotional value. Consumption emotions are affective responses occurring during consumption experiences such as those within shopping environments (Mehrabian and Russell, 1974; Walsh et al., 2011). Mehrabian and Russell (1974) identify two key affective responses (i.e. arousal and pleasantness) to the shopping environment, arising from environmental cues which impact on shopping behavior. Studies have also shown that customers shop not only because they value the items they buy or consume in a functional sense, but also because they derive pleasure and/or excitement from the shopping experience (Babin et al., 1994). The theory of consumption values identifies a component of value embedded in a product or service which is derived from specific feelings that a customer associates with it (Sheth et al., 1991; Sweeney and Soutar, 2001). Play, fun, and escapism gained by using the service for its own sake by the consumer are some aspects of emotional value (Holbrook, 1999). Literature across varying domains reports evidence that customers experience fun, entertainment, and enjoyment from interacting with technology (e.g. Kim and Han, 2009) including service processes which take place via the online channel (e.g. Chiu et al., 2014; Rose et al., 2012). As such, this study conceptualizes emotional value as the benefit derived by the customer from affective responses generated via the online channel during consumption. 2.1.1.3. Monetary value. In this study, monetary value is underpinned by the work of Mittal and Sheth (2001) and O’Cass and Ngo (2011) who argue that monetary value refers to the utility perceived by the customer where the pricing policies of products/ services in terms of price levels provided by the firm are perceived fair and reasonable against competitive offerings. In a multichannel retailing context, we argue the monetary value of offerings (i.e. merchandise and services) provided in the online channel is important as customers look for products/services that are affordable, fair, and reasonable priced in comparison to prices provided in alternate channels of the retailer, as well as prices offered by competitors. Empirical research has shown that similar merchandize and price policies, across channels ease consumers’ uncertainty, enhance their trust in the retailer and increased sales through them overtime (Avery et al., 2012; Cao and Li, 2015). Therefore, monetary value is defined here as the utility derived by the customer from the pricing policies and pricing levels of offerings found in the online channel considered acceptable and fair to the customer. 2.1.1.4. Brand integration value. Designing an online channel that effectively integrates a consistent brand image across with the firm’s physical service channels is a major challenge confronting multi-channel providers (Carlson and O’Cass, 2011; Banerjee, 2014). Studies in the multichannel context have shown that a uniform service experience for customers include the consistent portrayal, and reinforcement of brand information, including marketing messages, pricing, providing consistent merchandise, and customer service across channels (Banerjee, 2014; Badrinarayanan et al., 2014). Importantly, misalignment of a multichannel service experience not only negatively impacts customer perceptions of integration quality but also make customers susceptible to terminating the relationship with the service provider (Banerjee, 2014). On this basis, a consumer will derive brand integration value when they perceive brand image consistency between the online channel service experience and the physical service experience. Therefore, brand integration value is conceptualized as the utility derived by the customer arising from a consistent, seamless, and coherent brand experience across retail channels. 2.1.1.5. Channel convenience value. Research on the convenience benefits and adoption of online channels for online shopping is

J. Carlson et al. / Journal of Retailing and Consumer Services 27 (2015) 90–102

well established. For instance, it is known that consumers will adopt online channels if they perceive its relative benefits, such as the convenience over existing traditional channels (Choudhury and Karahanna, 2008; Yang et al., 2013). According to Kushwaha and Shankar (2013), customers who adopt multiple channels from the same retail provider seek greater convenience associated with the adoption of the electronic channel as they allow the saving of time and money where they can shop whenever they want. In terms of online shopping, convenience has been found to be the most compelling benefit in terms of being able to shop anywhere at any time which involves utilities such as location (place utility), expanded store hours and quick, efficient checkouts (time utility) (Rohm and Swaminathan, 2004). For multi-channel retailers, the online channel is able to provide location and time flexibilities compared to single (physical) channel shopping, leading to the fulfillment of customers’ need for accessibility, reduced waiting time, and increased service efficiency. As such, the present study conceptualizes channel convenience value as the utility derived by the customer of space and time of the online channel relative to the other channels delivered by the retailer. 2.2. Model development and hypotheses Based on literature regarding customer value components, the research model depicted in Fig. 1 is developed. This model indicates that perceived online channel value (POCVAL) is set within a multi-dimensional approach and follows customer perceived value theory literature, especially the work of Sheth et al. (1991) and Sweeney and Soutar (2001), which has been adopted in later research (e.g. Blocker, 2011; Turel et al., 2010). Furthermore, the framework in Fig. 1 reflects an influence of POCVAL on online channel satisfaction and online channel loyalty intentions (see Section 2.2.2). In this study, customer satisfaction with the online channel refers to customers’ post-purchase evaluation to the consumption experience (Carlson and O’Cass, 2011). Online channel loyalty intention refers to the future continuation in using the online channel and its recommendation of customers who derive value from the online channel and are satisfied. st

POCVAL 1 Order Components

2.2.1. Conceptualization and operationalization of POCVAL Following the methodological guidelines of Becker et al. (2012), Lee and Cadogan (2013) as well as perceived value research by Ruiz et al. (2008) and Wetzels et al. (2009), we conceptualize POCVAL as a second-order construct which is constituted formatively by the previously discussed first-order components. A formative conceptualization is appropriate because the five identified POCVAL components are distinct in nature, and are not interchangeable (Becker et al., 2012; Edwards, 2001; Jarvis et al., 2003; Lee and Cadogan, 2013; Ruiz et al., 2008). The notion of a perceived value with distinct components is analogous to Sheth et al. (1991) who propose components of customer value, where a decision may be influenced on a continuum by one or all of the value components. Correspondingly, a change in one component is not necessarily accompanied by changes in other components (e.g. reflected in high pairwise construct correlations). For example, a decrease in brand integration value can occur separately and would not necessarily be accompanied by a change in perceptions of service performance value, convenience value or any of the other salient value components as perceived by the customer in a multi-channel retail context. All together, the components do not satisfy the conditions for reflective factor modeling of POCVAL which would mean that the components do share a common cause (Becker et al., 2012; Chin, 1998; Edwards, 2001; Jarvis et al., 2003; Lee and Cadogan, 2013). Instead, POCVAL is understood as a combination of the five specific (latent) components into a general concept (Lee and Cadogan, 2013; Wetzels et al., 2009). As such—and as for every higher-order conceptualization—it has to be highlighted that our predictions linking the lower-order constructs in a formative way with the general concept POCVAL are “not a question of causality but rather a question of the nature of the hierarchical latent variable, as the higher-order construct (the general concept) does not exist without its lower-order constructs (dimensions)” (Becker et al., 2012, p. 362). Thus, “…the higher-order construct [POCVAL] … is a combination of several [five] specific (latent) dimensions into a general concept” (Becker et al., 2012, p. 363) that fully mediates the influence on subsequent endogenous variables (Becker et al., 2012; Chin, 1998). The lower-order components are especially important for management as they represent actionable

POCVAL 2nd Order Component*

POCVAL Outcomes

Service Performance Value H1a + Online Channel Satisfaction

Emotional Value H1b + Monetary Value

H1c +

H3+ H4+

Perceived Online Channel Value

H2+ Brand Integration Value

93

H1d +

Online Channel Loyalty Intention

H1e + Convenience Value

Control variables: Age, Gender, Income, Involvement

Note: Perceived online channel value is configured as a hierarchical component, second-order construct. Fig. 1. Conceptual model of the study. Note: Perceived online channel value is configured as a hierarchical component, second-order construct.

94

J. Carlson et al. / Journal of Retailing and Consumer Services 27 (2015) 90–102

drivers (Becker et al., 2012; Chin, 1998; Wetzels et al., 2009). Correspondingly, we predict that each of the previously elaborated value dimensions contributes to (i.e. is a component of) the general concept POCVAL: H1a:. Service performance value is a formative first-order component of POCVAL, H1b:. Emotional value is a formative first-order component of POCVAL, H1c:. Monetary value is a formative first-order component of POCVAL, H1d:. Brand integration value is a formative first-order component of POCVAL, H1e:. Channel convenience value is a formative first-order component of POCVAL. 2.2.2. The interrelationship between POCVAL, online channel satisfaction and online channel loyalty intentions It is now widely recognized that customer perceived value is a powerful force in high velocity, competitive markets (O’Cass and Ngo, 2011; Ostrom et al., 2015) and that firms that can create value possess an underlying competitive advantage (Woodruff, 1997) for example through the positive effect of perceived value on customer satisfaction and loyalty (Babin et al., 1994; Cronin et al., 2000; Parasuraman and Grewal, 2000). According to Lempinen and Rajala (2014), future technology-oriented service management studies need to investigate the connection between user-perceived value and output business value (e.g. satisfaction or loyalty) in IT delivered services. In response, customer perceived value, satisfaction, and loyalty are introduced in the POCVAL model. In this investigation, POCVAL is considered as an antecedent of satisfaction, and satisfaction in turn, along with POCVAL, as key antecedents of online channel loyalty intention [OCLI] (Parasuraman et al., 2005). Thereby loyalty intentions are indicated by an inclination to perform a diverse set of behaviors that signal a motivation to enhance an ongoing relationship with a firm (Oliver, 1999; Augustín and Singh, 2005). The interrelationships between OCVAL, online channel satisfaction and OCLI are emphasized in Fig. 1 and explained below ending with the articulation of formal hypotheses. As such, our approach to model POCVAL as secondorder construct allows for more theoretical parsimony and reduced model complexity. We would otherwise need to model the effects of five lower-order dimensions, on satisfaction and OCLI and could not make conclusions about which OCLI driver—POCVAL or satisfaction—on an abstract level, is more important for enhancing OCLI. The present study conceptualizes POCVAL in a framework which is formed by a collection of customer benefits that accounts for cross-channel perceptions that reflect a consumer’s consideration of his or her relationship with the retailer’s offline channel. Our POCVAL conceptualization is similar to Augustín and Singh’s (2005) understanding of value, grounded on need-gratification and dual-factor motivation theory, that value is a necessary as well as a sufficient condition for loyalty. This leads them to the conceptualization of a direct relationship with loyalty intention (Augustín and Singh, 2005). Similarly, Babin et al. (1994) suggest that value is an important outcome influencing future consumer decisions (e.g. loyalty intentions) through feedback loops into the consumer decision processes. Other studies drawing from the theory of goal-directed behaviors in that consumers seek value as the higher-order goal in marketplace exchanges and that this goal regulates their future loyalty intentions also show that loyalty related variables, for example advocacy in terms of favorable word-of-mouth referrals, strength of preference or (intended)

repeat visitation and purchase intentions (Oliver, 2010; Augustín and Singh, 2005) are a direct function of perceived value assessments (e.g. Ruiz et al., 2008; Wetzels et al., 2009). Thus, H2:. POCVAL is positively related to online channel loyalty intention. Similarly, the concept of customer value brings together service and relationship benefits, and posits that customer satisfaction, which can be understood as the degree of fulfillment of some need, desire, goal, or other pleasurable end state that results from a specific exchange transaction between the consumer and a firm (Oliver, 2010; Augustín and Singh, 2005), is a key outcome. Specifically, customer value frameworks suggest that customers’ evaluation of the value they receive may lead directly to the formation of satisfaction (Bradley and Sparks, 2012; Woodruff, 1997). The service management literature argues that customer satisfaction is the result of a customer’s perception of the value received (e.g. Cronin et al., 2000; Kim et al., 2013), and that customers’ perceived value that lacks their overall satisfaction may not provide an adequate explanation of their future/continued behavior (Woodruff, 1997). This conceptualization of value differs to customer satisfaction which is viewed in this study as a summary judgement of active reactions to a service incident (Oliver, 1980). This reasoning mimics the notion within the service performance-satisfaction literature, that service performance exogenously initiates a sequence which leads to satisfaction (and finally loyalty intentions) (Oliver, 2010). In the context of the online shopping environment, empirical studies have shown that customer value assessments impact on satisfaction judgements with the online channel (e.g. Yoo et al., 2010). As such, we argue as POCVAL assessments (derived from the online channel across the five value components) increase, customers will become more satisfied with the retailer’s online channel. Thus, H3:. POCVAL is positively related to online channel satisfaction. Satisfaction is understood as a necessary condition to establish loyalty, reflected in a direct effect on loyalty (Oliver, 2010; Augustín and Singh, 2005). This is confirmed in a variety of marketing studies (Chandrashekaran et al., 2007; Kumar et al., 2013). The finding that satisfaction serves as an antecedent of online loyalty intentions has been also acknowledged in the online environment literature (Barrutia and Gilsanz, 2013; Carlson and O’Cass, 2010; Chen, 2012). As such, we argue that highly satisfied customers with the online channel are likely to spread positive word-of-mouth, revisit the website and purchase from the channel in the future. Thus, H4:. Online channel satisfaction is positively related to online channel loyalty intention.

3. Research methodology Data were collected from customers of retailers who had a recent shopping and purchase experience with an online channel and the offline channel of the same multi-channel retailer. Specifically, the investigation focused on multi-channel service providers that had a physical offline service operation, and an online presence with e-commerce capability. Furthermore, Walsh et al. (2014) argue that consumer perceived value may exist as a ubiquitous shopping-related influence that permeates many aspects of consumption for many individuals worldwide. On this basis, following a similar argument of Truong (2013), to test if online channel value perceptions are somewhat universal and substantiate our theory, we tested the robustness of the framework

J. Carlson et al. / Journal of Retailing and Consumer Services 27 (2015) 90–102

across two developed countries (1) Australia and (2) France. Australia and France served as our study countries because they share similar cultural values and norms (Hofstede, 2001) and are advanced countries in terms of e-commerce readiness. For instance, according to Kearney’s (2015) e-Commerce Retail Index which considers the top 30 developed and developing countries for online retail (including pure online and brick and mortar retailers) across factors including online market size, technology adoption and consumer behavior, growth potential and available logistical and telecommunications infrastructure into an overall online market attractiveness score, the analysis revealed that France ranked 6th (index score 59.3) and Australia 10th (index score 43.6) respectively. Thus, finding differences or similarities across countries with a similar cultural background and e-commerce readiness would allow us to make a more convincing argument than conducting the study in a single country or across countries that differ greatly in cultural background. Moreover, using countries with highly disparate cultural backgrounds and e-commerce readiness could introduce significant biases into our sample, which may limit the possibility to generalize the findings. Finally, even though the countries are culturally close, differences in consumption habits still exist and warrant further investigation (Truong, 2013). Data collection was conducted through an online survey administered to a consumer panel in France and Australia with a market firm to randomly selected members 18 years and over. Respondents had to have made a purchase in the bricks and mortar outlet (i.e. retail store), as well as the online presence of the same retailer in the previous 6 months. The instruction to respondents was “Please consider an online retailer with an offline retail store and answer all the questions with this particular retailer and specifically relating to the online experience of this retailer”. Retrospective sampling was adopted for the study which is characterized by allowing respondents to reflect on their recent shopping and purchase experiences with the online channel in responding to the survey instrument in relation to the multichannel retailer. In this sense, the data set is neither firm nor product category specific, and it captures customer purchases across a comprehensive set of retail product categories (such as fashion apparel and accessories, groceries, hobby items, electronics, sporting equipment, and musical instruments) and competing firms. Such an approach is considered representative of the population and allows for empirical generalizability (Kushwaha and Shankar, 2013). For the Australian survey, the instrument was in English. However for the French sample, the questionnaire was in French. To ensure the validity of the translation the survey was developed in English and then all original items were translated into French by a researcher whose native language was French. Then, another researcher independently translated the items back to English (O’Cass and Sok, 2013). Further, the two researchers confirmed the meaning of the French version by comparing the two English versions. On this basis, the wording of some items was modified to make them clear, consistent, and understandable. 3.1. Measures and survey instrument To measure the constructs in the research model, validated items by other researchers in the extant literature were adapted and multi-item scales were used. All items were modified to fit the online context. Three reflective items each were used to measure the POCVAL components; this leads – together with the formative type of the hierarchical latent variable POCVAL (see H1a–H1e) – to a reflective-formative Type II second-order model according to the specification typology of Jarvis et al. (2003). Service performance value and monetary value were measured building on the

95

approach of O’Cass and Ngo (2011); emotional value items focused on capturing key affective responses of arousal and pleasantness which have been adopted to measure emotion in past research with items drawn from Walsh et al. (2011) and Eroglu et al. (2003); brand integration value measurement was adapted from Wang et al. (2009) and channel convenience value from Loiacono et al. (2007) (see Table 2).1 Three items drawn from Carlson and O’Cass (2010) were used to measure online channel satisfaction with three items measuring OCLI drawn from Parasuraman et al. (2005). Each of the scale items was measured by a 1–7 Likert-type scale ranging from strongly disagree (1) to strongly agree (7). The final items used in the questionnaire are listed in Table 2. To provide a more robust test of the POCVAL framework, consistent with prior research, this study used (1) three demographic variables (age, gender, and income) (e.g., Schlosser et al., 2006), and (2) the level of involvement with the retailer to control for user heterogeneity (Chiu et al., 2014). All control variables are measured by single-item questions. 3.2. Profile of the respondents In total, 448 completed surveys were received. In the French sample, 262 responses were received with 45% male and 55% female with ages between 19 and 66 years, with the average age of 35 years. Respondents had indicated that they had predominately access the website from home (90%), and on average have visited the website 7.18 times and made 1.14 purchases in the past four week period. Respondents are in administration (34.7%) and technical profession (18.3%) occupations, and predominately High School educated (32.8%), followed by Advanced Diploma level (25.6%) and Bachelor level (10.7%). In the Australian sample, 186 responses were received with 49% male and 51% female with ages between 18 and 64 years, with the average age of 44 years. Respondents had indicated that they had predominately access the website from home (88.7%), and on average have visited the website 9.18 times and made 1.56 purchases in the past four week period prior to completing the survey. Respondents are in professional (22%) and administration (16%) occupations, and predominately High School educated (32.3%), followed by Certificate level (23.7%) and Bachelor level (19.9%). The respondents across the two samples shared similar characteristics where they were generally well educated, relatively affluent, predominately accessed the Internet from home to purchase from the online channel, experienced with using their preferred retail website prior to completing the survey, and skilled in using the Internet.

4. Research results 4.1. Estimation procedure Structural equation modeling (SEM) represents a suitable multivariate data analysis method to analyse the hypotheses of the study in the context of the theoretical model (Fig. 1) and the available empirical data. Two SEM methods – the covariancebased and the variance-based SEM approaches – can be used to estimate cause-effect models with latent variables. While both approaches are suitable methods for estimating the POCVAL model, the PLS-SEM approach is more appropriate for the purpose of this study. PLS-SEM supports the study’s prediction-oriented goal (i.e., the prediction of satisfaction and OCLI). It also suits 1 The indicators of the first-order reflective constructs are repeatedly used to operationalize POCVAL (see section Estimation Procedure) (Becker et al., 2012).

96

J. Carlson et al. / Journal of Retailing and Consumer Services 27 (2015) 90–102

complex model structures (Becker et al., 2012) as found in this study and requires a comparatively smaller sample size (Hair et al., 2012). To analyse the data, the statistical software SmartPLS 2.0 (Ringle et al., 2005), and the path weighting scheme to estimate the PLS-SEM. To compute the construct score for the reflectiveformative Type II second-order hierarchical latent variable POCVAL, the repeated indicator approach Mode A (reflective measurement) was used (Becker et al., 2012). The repeated indicator approach with Mode A is an appropriate proceeding, particularly since the POCVAL dimensions have equal numbers of indicators. In a two-step approach, the study first evaluated the reflective measurement models and tested the construct reliability and validity for each of the two samples (e.g. Australia and France) which is consistent with Ruiz et al. (2008) in the presentation of cross-country empirical results. Second, the structural models were examined (Hair et al., 2012). 4.2. Assessment 4.2.1. Measurement mode All constructs in our model (see Fig. 1) draw on reflective measurement models. The qualitative criteria provided by Jarvis et al. (2003) support this a priori theoretical conceptualization of the measurement mode (see also Diamantopoulos and Winklhofer, 2001). The formative conceptualization of the higher-orderconstruct POCVAL – belonging to the structural model – is assessed in Section 4.2.3. 4.2.2. Reflective measurement models To examine the construct reliability, its internal consistency was examined using both Cronbach α and composite reliability (Hair et al., 2012; Jarvis et al., 2003). The measurement models’ internal consistencies exceed the recommended benchmark with values clearly above the common threshold of .70 across both, the Australian (AU) and French (FR) data sets (see Table 1). Likewise, the ‘Average Variance Extracted’ [AVE] which captures the portion of the reflective indicator block’s variance that the associated construct can explain (Hair et al., 2012), exceed the common threshold of .50 for all reflective measurement models across both data sets (see Table 1). All in all, a good internal consistency of all scales received support. Table 1 Internal consistency criteria of reflective constructs. Latent variable

Service performance value Emotional value Monetary value Brand integration value Convenience value Satisfaction OCLI

AU retail sample (n¼ 186) α – CR – AVE

FR retail sample (n ¼262) α – CR – AVE

.89/.93/.82 .86/.91/.78 .89/.93/.82 .97/.98/.94 .88/.93/.80 .96/.98/.93 .94/.96/.89

.85/.91/.77 .83/.89/.72 .90/.94/.84 .96/.98/.93 .78/.87/.69 .95/.97/.90 .93/.95/.87

Note: α ¼Cronbach α, CR ¼Composite Reliability, AVE ¼ Average Variance Extracted.

Convergent validity which detects whether the measures of a construct are more correlated with one another than with measures of other constructs is also established. All indicator loadings for the assigned reflective constructs offer significant and high values ( 4.80) which are above the conservative threshold of .70 across both data sets (see Table 2). Finally, and despite a few high correlations of the constructs’ latent variable scores, discriminant validity which determines whether the constructs are distinct constructs is clearly confirmed

according to the Fornell–Larcker criterion (see Tables 3 and 4) (Hair et al., 2012; Jarvis et al., 2003). With the Fornell–Larcker criterion being satisfied, the ‘square root of the AVE’ of each reflective construct has to be greater than its correlation coefficients with each other constructs; this means, the shared variance for each latent variable and their associated indicators is confirmed as higher than the shared variance with other latent variables. As such, the formative type of the hierarchical latent variable POCVAL is substantiated through the confirmation of discriminant validity (of the first-order components) (Becker et al., 2012). In addition, the investigation of the cross loadings established the discriminant validity further as each measurement item loaded the highest on its assigned constructs in comparison to its cross loadings on other constructs (see Appendix Tables A1 and A2) (Hair et al., 2012). 4.2.3. Structural model After confirming POCVAL’s and satisfaction’s predictive relevance on the basis of the Stone–Geisser Q² criterion which was in each case well above 0, we evaluated the structural model regarding the size and significance of the path coefficients and R² values of satisfaction and OCLI (see Table 5) (Hair et al., 2012). The paths-coefficients (i.e. weights) presented in Table 5 were significant (α level o1%, two-sided test) and greater than the .10 benchmark with both data sets (Hair et al., 2012). The results indicate that the formation of POCVAL by the five components has a positive direct effect on satisfaction and OCLI and satisfaction’s direct effect on OCLI is also confirmed. Consequently, the results support all hypotheses. To assess POCVAL’s indirect effect on OCLI (i.e. mediated via satisfaction) on OCLI, we calculated the ‘variance accounted for’ (VAF) (Hair et al., 2012). The VAF results indicate that 63.6%2 of POCVAL’s total effect on OCLI is mediated via satisfaction with the Australian data, and 45.8% with the French data.3 For the formative type second-order construct POCVAL, the weights of the lower-order components are especially important as they represent actionable drivers of the higher-order construct. Hence, they represent strategic key components that managers can influence to improve the overall value perception and, thus, satisfaction and OCLI. To obtain a more detailed insight into the individual POCVAL components’ efficiency with regard to the OCLI construct, their total effects are shown in Fig. 2. The variance explained by the model (R²) is a key criterion for evaluating the structural model’s quality in PLS-SEM (Hair et al., 2012). With the Australian data, 75% of satisfaction’s variance is explained through the model and with the French data, 65%. OCLI’s variance is explained through the model of 78% with the Australian data and 75% with the French data (see Table 5). The R²values are strong, reflecting that the POCVAL model explains a substantial amount of the variance of the endogenous variables satisfaction and OCLI within both data sets (Chin, 1998). Finally, to provide a more robust test of our results, the control variables were included as direct antecedents of OCLI. User heterogeneity with regard to the included variables seems not to be a critical issue, since the effects of all control variables across both data sets were not significant; with the exception of the effect of ‘age’ with the French data which had a weak negative effect (pathcoefficient
.08; t-value 2.45). 2

VAF ¼(.86n.61)/(.86n.61þ .30) We note, that while each structural model’s path-coefficients differs in magnitude between the Australian and the French sample (see Table 5), no statistically significant differences were found between the Australian and the French sample (α levelo 10%; two-sided test) (Chin, 2004). Thus, when comparing the computed results between the Australian and French sample, this is provided in a descriptive manner only. 3

J. Carlson et al. / Journal of Retailing and Consumer Services 27 (2015) 90–102

97

Table 2 Loadings of reflective constructs’ indicators across contexts. Construct and indicator

AU retail sample (n¼ 186) Loading

Service performance value PV1: The Website delivers a service that is exactly what I want PV2: The Website delivers quality services PV3: The Website delivers services with innovative performance features Emotional value In general, when I use this Website I feel: EV1: Happy EV2: Stimulated EV3: Excited Monetary value MV1: The pricing policies on the Website are fair MV2: The Website provides me with consistent and accurate pricing policies MV3: The pricing policy on the Website are more beneficial for me than that of competitors Brand integration value BIV1: The Website’s image matches that of this retailer BIV2: This Website projects an image consistent with the retailer’s image BIV3: The image portrayed of the retailer is the same as this Website Convenience value CV1: This Website is easier to use than calling on the phone or visiting in-store for service CV2: It is easier to use this Website for information/purchasing rather than other channels (e.g. visit the store, phone, or catalogue) CV3: This Website is an alternative to calling customer service or visiting the retailer Online channel satisfaction SAT1: I am satisfied with this Website SAT2: My choice to use this Website was a wise one SAT3: The Website does a good job of satisfying my needs Online channel loyalty intention OCLI1: I will purchase from this retailer’s Website in the future OCLI2: I will revisit this retailer’s Website in the future OCLI3: I will say positive things about this retailer’s Website to others

t-value

FR retail sample (n¼ 262) Loading

t-value

.92 .93 .87

57.56 67.49 28.20

.91 .90 .83

59.70 58,16 32.93

.87 .88 .91

47.45 29.06 36.53

.88 .87 .78

46.13 31.80 15.96

.94 .92 .85

79.82 52.57 22.81

.93 .96 .84

91.01 219.49 30.36

.95 .97 .98

80.71 114.50 236.23

.95 .97 .97

78.60 143.85 193.91

.89 .91

45.31 46.33

.81 .83

19.26 27.23

.89

34.26

.86

31.94

.97 .97 .96

148.06 139.10 96.29

.96 .94 .95

117.50 87.94 82.04

.94 .95 .95

64.11 68.45 92.62

.93 .96 .92

38.15 144.25 69.98

Notes: All loadings are significant o 1% α-level as all vastly exceed the threshold (t4 2.60); significance results are based on bootstrapping. Sample size equals the original sample size of the respective data set; 1000 subsamples. There is no sign change option.

5. Discussion and implications

context.

Scholars have called for empirical insights into strategies and performance metrics that capture how retailers can assess and measure customer value in multi-channel environments that takes into account the idiosyncratic nature of each channel and crosschannel effects (e.g. Maity and Dass, 2014; Zhang et al., 2010). In response, this study further conceptualizes and examines the notion of POCVAL in a multi-channel retail context by using data across two diverse countries from 448 customers and examining online channel value’s strategic relevance for explaining online customer satisfaction and loyalty intentions. In doing so, this investigation makes four substantive theoretical and managerial contributions to the customer perceived value literature in the context of online channel management within a multi-channel retail setting in three areas. First, this study contributes to the literature by supporting theory and practice in relation to a multichannel retailer’s capability to calibrate and deliver on a range of POCVAL components as perceived by the customer that can serve as a retail performance indicator which considers the performance of the online experience as well as important cross-channel interactions. Second, the study contributes to theory and practice in relation to a parsimonious measurement of POCVAL. Third, it contributes to the literature by identifying the extent to which POCVAL supports a firm’s value creation effort in order to enhance channel satisfaction and loyalty, which have major impacts on ultimate retail firm performance. Fourth, the study contributes to the development of generalizable theory of customer perceived value in online channels by being shown to be robust across two country market settings. Our findings give credence to a universal nature and application of POCVAL within the multichannel retail

5.1. Evaluation and performance of POCVAL components The findings suggest customer perceptions of online channel value depend significantly on a range of different components. For an online channel, IT, or multi-channel manager these components should be considered as an integral part of the value creation effort in delivering online channel value, and as such, strategy should be developed to ensure value is supported through their delivery. Building primarily upon the work on customer perceived value of Bradley and Sparks (2012), O’Cass and Ngo (2011), Sweeney and Soutar (2001) and the work of Wetzels et al. (2009) and Mathwick et al. (2001) in the online retail channel context in theorising the POCVAL framework and its first-order dimensions as actionable drivers for managers, the results of this suggest – reflected in the size of the path coefficients to POCVAL as well as the POCVAL components total effects on OCLI – that service performance value of the online channel has the largest contribution to POCVAL and OCLI. Furthermore, these results were found to be consistent across each country studied. This finding supports previous literature that highlights service utility/quality in the online environment as an essential pillar of the value creation process (e.g. Barrutia and Gilsanz, 2013; Cheng et al., 2009; Overby and Lee, 2006). Thus, channel managers should focus on strengthening their functional online channel service performance capability because such a capability significantly affects customer value perceptions and should be maximized in all customer encounters. Channel managers should also take note of the importance of monetary value of product offerings and brand integration value

98

J. Carlson et al. / Journal of Retailing and Consumer Services 27 (2015) 90–102

Table 3 Correlation matrix for Australian retail sample.

Test criterion (AVE1/2) 1) Service performance value 2) Emotional value 3) Monetary value 4) Brand integration value 5) Convenience value 6) Online channel satisfaction 7) OCLI

Table 5 Structural model results across contexts.

1

2

3

4

5

6

7

r .91 –

r .88

r .91

r .97

r.90

r .96

r .94

.57 .83 .60 .69 .85

– .45 .37 .41 .48

.81

– .48 .64 .77

.43

– .54 .60

.71

– .73

.63

–

.70

.87

–

Table 4 Correlation matrix for French retail sample.

Test criterion (AVE1/2) 1) Service performance value 2) Emotional value 3) Monetary value 4) Brand integration value 5) Convenience value 6) Online channel satisfaction 7) OCLI

1

2

3

4

5

6

7

r.88 –

r .85

r .92

r .96

r.83

r .95

r .93

.48 .74 .57 .58 .79

–

.79

.42 .27 .19 .34 .41

.71

Service performance value Emotional value Monetary value Brand integration value Convenience value Structural paths POCVAL-Online channel satisfaction R2 of SAT POCVAL-OCLI Online channel satisfaction-OCLI R2 of OCLI

AU retail sample (n¼ 186)

FR retail sample (n¼ 262)

Path coefficient

T-value

Path coefficient

T-value

.29

21.75

.31

21.76

.16 .26 .26

8.68 16.14 13.79

.14 .30 .30

7.23 22.06 20.29

.25

15.91

.22

15.14

.86

42.28

.81

26.55

3.64 7.94

.45 .47

.75 .30 .61 .78

.65 6.39 6.65 .75

Notes: All path coefficients are significant o1% α-level as all exceed the threshold (t 42.6); significance results are based on bootstrapping. Sample size equals the original sample size of the respective data set; 1000 subsamples. There is no sign change option.

– .52 .52 .66

OCVAL weights

– .53 .61 .57

– .61 .62

–

variables .82

–

(i.e. congruency of the retail brand between online and physical channels). Particularly in the French retail sample, online channel monetary value and brand integration rival online channel service performance value as the second and third most important POCVAL component and OCLI driver; in comparison to the Australian retail sample, the path coefficients to POCVAL and the total effects of those three most important drivers are stronger in the French retail sample. Consequently, the presentation of prices that are perceived as fair and accurate or at best more beneficial with regard to competitors is thus advisable for managers. This holds as well for the portrayal of a consistent image through the online channel. Retrieval cues in the online-store could be for example fonts, colors, slogans, and testimonials used in the offline store environment as well as the portrayal of a consistent user and usage imagery (cf. Keller, 1993). Convenience value of the online channel relative to other channels is the fourth important component in both samples. What is particularly pertinent about these findings is that these utilitarian-based components are largely under the control of the multi-channel retailer. Despite the study findings in relation to emotional value of the online channel showing it to have the smallest contribution to the POCVAL index, it is however significant and relevant nonetheless across both samples. This is a somewhat surprising finding given the number of Internet retailing studies reporting the strong presence of hedonic effects in the process of online shopping due to the incoming sensory data from a range of interactive stimuli on the website such as text-based information, visual imagery, video, or audio delivery (e.g. Rose et al., 2012; Wetzels et al., 2009). However, the findings of this study in a multi-channel retail context support prior online only retailing-based studies (c.f. O’Brien, 2010; Overby and Lee, 2006) which have shown that components relating to utilitarian value are more strongly related than hedonic value components to preference and behavioral intentions towards the Internet retailer. 5.2. Online channel value measurement and the effects on relational

The findings suggest that a simple, direct, uni-dimensional measure of customer judgements of perceived value which focus on monetary terms (‘value for money’) is inadequate for capturing the richness of online channel value perceptions in multi-channel retailing. By modeling POCVAL as a formative second-order construct, this study achieves a higher level of theoretical depth and breadth, offering a unique and meaningful model for examining the over-arching effects of those first-order value components on online channel value, and, ultimately online channel satisfaction and online channel loyalty intention.4 The development in this study of a POCVAL index (articulated as a Type II multi-dimensional model) implies that, for the measure to be comprehensive, it should contain several value components in order to assess and optimize customer perceived value in the online channel. With regard to the analysis of POCVAL’s effects, the study confirmed the strong and direct influence of POCVAL on OCLI parallel to that of online channel satisfaction and POCVAL’s effect on online channel satisfaction. Whereas POCVAL’s effect on OCLI is similar to that of satisfaction in the French sample (see pathcoefficients); it is weaker but nevertheless sizable in the Australian sample. Generally, the study can confirm that the dominant paradigm to invest only in customer satisfaction (cf. the criticism of Kumar et al., 2013) is not adequate since value plays an important parallel role to enhance OCLI, and is thus, essential for the long-term survival of firms in today’s competitive environment which builds on customer loyalty (see also Wang, 2011). In this sense, POCVAL plays a key role in driving and unlocking customer satisfaction. Importantly, omitting any of the value components of the POCVAL model prevents a complete understanding of the multidimensionality of the value construct and its effects on online channel satisfaction and OCLI. From a practitioner perspective, a formative model facilitates a channel manager’s decision process, in that it determines which value components are the most 4 From a modeling perspective instead of using only single-item (constructs) the use of multi-item-scales for the dimensions of POCVAL is advisable particularly due to higher reliability and predictive validity; as such the set of circumstances, that would favor the use of a small set or single-item measures rather than multiple items is very unlikely to be encountered in practice (Hair et al., 2014).

J. Carlson et al. / Journal of Retailing and Consumer Services 27 (2015) 90–102

99

0.254 0.240

Service Performance Value 0.110

Emotional Value

0.134 0.248

Monetary Value

0.219

FR Retail Sample AU Retail Sample

0.243

Brand Integration Value

0.213 0.180

Convenience Value

0.207

Fig. 2. Total effects of POCVAL dimensions on OCLI.

influential in forming customers’ value perceptions. As such, managers can then allocate resources to improve and optimize the components within the POCVAL framework which are needed to deliver and maximize value fulfillment to customers, and to realize customer satisfaction and loyalty goals for the long-term survival of multi-channel firms. In doing so, technology investments, service provider skills, and capabilities may need to be considered to effectively deploy the POCVAL components. 5.3. Online channel value for global service firms For multi-channel retail practitioners with online channel operations serving these countries could be improved through assessment of POCVAL where channel managers can measure online channel value across their country markets with increased confidence. The findings indicates that the POCVAL model is robust across the Australian and French retail markets where the (importance) order of weights are identical and the magnitude of the weights are fairly similar for each component across the two countries. Thus, the POCVAL model offers managers a tool to measure perceived value in international retailing contexts. While complete generalization requires further research and validation, multi-channel retailers can begin to develop improved programs and measurement instruments with the expectation that customers in different markets may define or view online channel value in a similar fashion.

6. Limitations and future research directions As in any empirical research, there are limitations with the study which open opportunities for further research. First, the research model is validated using data collected from Australian and French customers’ retrospective experiences of multi-channel retailers. Thus, caution should be observed in generalizing the findings beyond countries with similar characteristics (technology, retail environment, culture etc.) at this stage. Nevertheless, this study has provided some useful insights into the role of customers’ perceived value of the online channel in multi-channel retailing to support the development of customer relationships that help drive channel satisfaction and loyalty intentions. However, more research is needed in different product categories (e.g. banking, finance, tourism, service industries). Furthermore, even though a cross-country retail setting was adopted where each had developed multi-channel infrastructure in their economies, replication in different country markets, including developing countries at varying stages of IT readiness (c.f. Cyr, 2013), may elucidate further insights to the generalizability of the POCVAL model. This work may also map more of the customer culture issues to provide greater insight into POCVAL across culture and retail

environments. Second, the conceptualization of POCVAL components may not be exhaustive. In this study, a primary objective was to examine a group of basic online channel value components of multi-channel retailers. However, other POCVAL components may be salient in specific situations or for some types of customers; and analogously the few high construct correlations of POCVAL’s distinct first-order components might indicate that some of them work together. For instance, Lempinen and Rajala (2014) note that contemporary organizations are turning their attention to jointly creating value with a variety of stakeholders, including customers and that investigating the ways for orchestrating value co-creation in technology driven service processes remains an under-explored research area. Furthermore, extending the emotion component beyond affective states aroused by environmental stimuli examined in this study (e.g. perceived dominance such as perceptions of freedom and control c.f. Hsieh et al., 2014) and/or exploring additional value components along with our five components may more deeply identify consumption values that have been overlooked in this study. Third, while it is acknowledged that loyalty intention is behaviorally focused and is only one, yet still very important, aspect of customer loyalty (Oliver, 1999), future research could expand the dependent variable to include outcomes of perceived value such as consumer attitudes towards the retailer as well as other loyalty behaviors such as cross buying, purchase frequency and other actual buying behaviors. Fourth, the impact of individual characteristics on value perceptions and the research framework advanced here warrants further research, since consumer traits, culture and consumer knowledge-related resources have been suggested to impact the evaluations of technologybased services across country markets (Barrutia and Gilsanz, 2013; Cyr, 2013). Finally, researchers could extend and broaden the research in this study to investigate the concept of customer perceived value by considering value components associated with a firm’s multichannel strategy and its brand. The development and widespread market acceptance of recent technological devices, (such as smartphones and tablets), across digital channels and applications (e.g. such as social media, mobile applications to access service and the emergence of location-based services) poses new challenges and great opportunities for innovative retail service designs, value creation and customer experience management (Ström et al., 2014). Given this emerging retail operating environment, it can be assumed that a valuable experience will be influenced by the performance of the multi-channel service interface to deliver compelling and differentiated branded customer experiences, and more importantly, the ability of customers to seamlessly use multiple modes of contact with the retailer that reflects a coherent brand image. We suggest future research to look into these matters.

100

J. Carlson et al. / Journal of Retailing and Consumer Services 27 (2015) 90–102

Appendix A Tables A1 and A2.

Table A1 Cross-correlations Australian retail sample.

PV1 PV2 PV3 EV1 EV2 EV3 MV1 MV2 MV3 BIV1 BIV2 BIV3 CV1 CV2 CV3 SAT1 SAT2 SAT3 OCLI1 OCLI2 OCLI3

Performance value

Emotional value

Monetary value

Brand integration value

Convenience value

Online channel satisfaction

OCLI

.919 .929 .871 .588 .432 .464 .786 .776 .688 .571 .573 .596 .568 .670 .64 .819 .827 .808 .755 .721 .802

.472 .496 .592 .866 .875 .907 .395 .399 .426 .333 .370 .370 .386 .348 .365 .449 .443 .496 .368 .420 .438

.784 .811 .660 .507 .300 .347 .940 .917 .848 .472 .455 .475 .576 .598 .538 .765 .726 .745 .680 .623 .712

.529 .592 .501 .381 .309 .270 .426 .482 .394 .953 .974 .981 .482 .485 .491 .582 .579 .581 .602 .602 .579

.669 .672 .520 .455 .277 .315 .575 .606 .542 .534 .513 528 .893 .905 .892 .710 .690 .703 .669 .674 .648

.841 .815 .639 .528 .344 .363 .739 .733 .613 .587 .579 .582 .630 .675 .648 .967 .970 .959 .822 .810 .840

.765 .796 .617 .489 .311 .312 .695 .693 .530 .600 .620 .601 .601 .660 .626 .849 .844 .832 .943 .945 .949

Table A2 Cross-correlations French retail sample.

PV1 PV2 PV3 EV1 EV2 EV3 MV1 MV2 MV3 BIV1 BIV2 BIV3 CV1 CV2 CV3 SAT1 SAT2 SAT3 OCLI1 OCLI2 OCLI3

Performance value

Emotional value

Monetary value

Brand integration value

Convenience value

Online channel satisfaction

OCLI

.905 .896 .828 .529 .404 .180 .700 .746 .560 .530 .555 .576 .415 .486 .548 .761 .724 .779 .734 .748 .748

.423 .392 .467 .881 .873 .783 .355 .408 .407 .246 .253 .273 .099 .168 .203 .307 .292 .369 .375 .337 .430

.671 .680 .582 .446 .369 .168 .937 .959 .843 .460 .496 .532 .389 .391 .512 .653 .607 .626 .645 .666 .675

.533 .583 .379 .275 .240 .105 .520 .529 .345 .953 .969 .971 .403 .473 .454 .573 .570 .586 .487 .576 .526

.539 .548 .443 .240 .161 .011 .477 .551 .388 .509 .508 .529 .811 .828 .856 .568 .557 .605 .585 .629 .521

.761 .744 .572 .406 .285 .049 .621 .669 .515 .567 .581 .608 .456 .473 .578 .956 .944 .948 .740 .803 .766

.753 .794 .521 .482 .313 .117 .668 .717 .543 .531 .548 .562 .440 .469 .621 .792 .761 .794 .926 .955 .923

References Ahrholdt, D., 2011. Empirical identification of success-enhancing website signals in e-tailing. J. Market. Theory Pract. 19, 441–458. Kearney, A.T., 2015. The 2015 Global Retail e-Commerce Index. Accessed at: https:// www.atkearney.com/documents/10192/5691153/Global þ Retail þ E-Commer ce þKeeps þ On þClicking.pdf/abe38776-2669-47ba-9387-5d1653e40409. Augustín, C., Singh, J., 2005. Curvilinear effects of consumer loyalty determinants in relational exchanges. J. Market. Res. 42, 96–108. Avery, J., Steenburgh, J., Deighton, J., Caravella, M., 2012. Adding bricks to clicks:

predicting the patterns of cross-channel elastic-ities over time. J. Market. 76 (3), 96–111. Babin, B., Darden, W., Griffin, M., 1994. Work and/or fun: measuring hedonic and utilitarian shopping value. J. Consum. Res. 20, 644–656. Badrinarayanan, V., Becerra, E., Kim, C., Madhavaram, S., 2012. Transference and congruence effects on purchase intentions in online stores of multi-channel retailers: initial evidence from the U.S. and South Korea. J. Acad. Market. Sci. 40, 539–557. Badrinarayanan, V., Becerra, E., Kim, C., Madhavaram, S., 2014. Influence of congruity in store-attribute dimensions and self-image on purchase intentions in online stores of multichannel retailers. J. Retail. Consum. Serv. 21, 1013–1020.

J. Carlson et al. / Journal of Retailing and Consumer Services 27 (2015) 90–102

Banerjee, 2014. Misalignment and its influence on integration quality in multichannel services. J. Serv. Res. 17 (4), 460–474. Barrutia, J., Gilsanz, A., 2013. Electronic service quality and value do consumer knowledge-related resources matter? J. Serv. Res. 16, 231–246. Becker, J., Klein, K., Wetzels, M., 2012. Hierarchical latent variable models in PLSSEM: guidelines for using reflective-formative type models. Long Range Plan. 45, 359–394. Blocker, C., 2011. Modeling customer value perceptions in cross-cultural business markets. J. Bus. Res. 64, 533–540. Bradley, G., Sparks, B., 2012. Antecedents and consequences of consumer value: a longitudinal study of timeshare owners. J. Travel Res. 51, 191–204. Cao, L., Li, L., 2015. The impact of cross-channel integration on retailers’ sales growth. J. Retail. 91 (2), 198–216. Carlson, J., O’Cass, A., 2010. Exploring the relationships between e-service quality, satisfaction, attitudes and behaviours in content-driven e-service web sites. J. Serv. Market. 24, 112–127. Carlson, J., O’Cass, A., 2011. Managing web site performance taking account of the contingency role of branding in multi-channel retailing. J. Consum. Market. 28 (7), 524–531. Chandrashekaran, M., Rotte, K., Tax, S., Grewal, R., 2007. Satisfaction strength and customer loyalty. J. Market. Res. 44, 153–163. Chen, S., 2012. The customer satisfaction–loyalty relation in an interactive e-service setting: The mediators. J. Retail. Consum. Serv. 19 (2), 202–210. Cheng, J., Wang, E., Lin, J., Vivek, S., 2009. Why do customers utilize the internet as a retailing platform? A view from consumer perceived value. Asia Pac. J. Market. Logist. 21 (1), 144–160. Chin, W., 1998. Issues and opinion on structural equation modelling. MIS Q. 22, 7–16. Chin, W., 2004. Multi-Group analysis with PLS. Frequently Asked Questions – Partial Least Squares & PLS-Graph, from http://disc-nt.cba.uh.edu/chin/plsfaq/ multigroup.htm. Chiu, C., Wang, E., Fang, Y., Huang, H., 2014. Understanding customers’ repeat purchase intentions in B2C e‐commerce: the roles of utilitarian value, hedonic value and perceived risk. Inf. Syst. J. 24, 85–114. Choudhury, V., Karahanna, E., 2008. The relative advantage of electronic channels: a multidimensional view. MIS Q. 32, 179–200. Cronin, J., Brady, M., Hult, G., 2000. Assessing the effects of quality, value, and customer satisfaction on consumer behavioral intentions in service environments. J. Retail. 76 (2), 193–218. Cyr, D., 2013. Website design, trust and culture: an eight country investigation. Electron. Commer. Res. Appl. 12 (6), 373–385. Deloitte, 2015. Global Powers of Retailing 2015: Embracing innovation. http:// www2.deloitte.com/content/dam/Deloitte/au/Documents/consumer-business/ deloitte-au-cb-gpor-120115.pdf (accessed 19.01.15). De Ruyter, K., Wetzels, M., Bloemer, J., 1997. The dynamics of the service delivery process: a value-based approach. Int. J. Res. Market. 14 (3), 231–243. Diamantopoulos, A., Winklhofer, H.M., 2001. Index construction with formative indicators: an alternative to scale development. J. Market. Res. 38 (2), 269–277. Edwards, J., 2001. Multidimensional constructs in organizational behavior research. Organ. Res. Methods 4 (2), 144–192. Eroglu, S., Machleit, K., Davis, L., 2003. Empirical testing of a model of online store atmospherics and shopper responses. Psychol. Market. 20 (2), 139–150. Forrester Research, 2014. Forrester Research Online Retail Forecast, 2013 To 2018 (US). https://www.forrester.com/For resterþ Researchþ Onlineþ Retail þ Forecast þ2013 þTo þ2018þ US/fulltext/-/ E-RES115941 (accessed 19.01.15). Grönroos, C., 2011. Value co-creation in service logic: a critical analysis. Market. Theory 11 (3), 279–301. Hair, J., Sarstedt, M., Ringle, C., Mena, J., 2012. An assessment of the use of partial least squares structural equation modeling in marketing research. J. Acad. Market. Sci. 40 (3), 414–433. Hair, J.F., Hult, G.T.M., Ringle, C.M., Sarstedt, M., 2014. A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM). Thousand Oaks, CA: Sage. Higgins, E., Scholer, A., 2009. Engaging the consumer: the science and art of the value creation process. J. Consum. Psychol. 19, 100–114. Hirschman, E., Holbrook, M., 1982. Hedonic consumption: emerging concepts, methods and propositions. J. Market., 92–101. Holbrook, M., 1994. The nature of customer value, an axilogy of services in the consumption experience. Thousand Oaks, CA: Sage. Hofstede, G., 2001. Culture’s Consequences: Comparing Values, Behaviors, Institutions, and Organizations Across Nations, 2nd ed., Sage Publications, Thousand Oaks, CA. Holbrook, M., 1999. Introduction to consumer value. In: Holbrook, M.B. (Ed.), Consumer Value: A Framework for Analysis and Research. Routledge, London, pp. 1–28. Hsieh, J., Hsieh, Y., Chiu, H., Yang, Y., 2014. Customer response to web site atmospherics: task-relevant cues, situational involvement and PAD. J. Interact. Market. 28 (3), 225–236. Jarvis, C., MacKenzie, S., Podsakoff, P., 2003. A critical review of construct indicators and measurement model misspecification in marketing and consumer research. J. Consum. Res. 30 (2), 199–218. Keller, K., 1993. Conceptualizing, measuring, and managing customer-based brand equity. J. Market. 57 (1), 1–22. Kim, B., Han, I., 2009. What drives the adoption of mobile data services & quest: an approach from a value perspective. J. Inf. Technol. 24 (1), 35–45. Kim, Y.H., Kim, D.J., Wachter, K., 2013. A study of mobile user engagement (MoEN):

101

engagement motivations, perceived value, satisfaction, and continued engagement intention. Decis. Support Syst. 56, 361–370. Kumar, V., Pozza, I., Ganesh, J., 2013. Revisiting the satisfaction–loyalty relationship: empirical generalizations and directions for future research. J. Retail. 89 (3), 246–262. Kushwaha, T., Shankar, V., 2013. Are multichannel customers really more valuable? The moderating role of product category characteristics. J. Market. 77 (4), 67–85. Lee, N., Cadogan, J., 2013. Problems with formative and higher-order reflective variables. J. Bus. Res. 66, 242–247. Lempinen, H., Rajala, R., 2014. Exploring multi-actor value creation in IT service processes. J. Inf. Technol. 29 (2), 170–185. Loiacono, E., Watson, R., Goodhue, D., 2007. WebQual: an instrument for consumer evaluation of websites. Int. J. Electron. Comm. 11, 51–87. Maity, M., Dass, M., 2014. Consumer decision-making across modern and traditional channels: E-Commerce, m-commerce, in-store. Decis. Support Syst. 61, 34–46. Mathwick, C., Malhotra, N., Rigdon, E., 2001. Experiential value: conceptualization, measurement and application in the catalog and Internet shopping environment. J. Retail. 77 (1), 39–56. Mathwick, C., Malhotra, N., Rigdon, E., 2002. The effect of dynamic retail experiences on experiential perceptions of value: an Internet and catalog comparison. J. Retail. 78 (1), 51–60. Mehrabian, A., Russell, J., 1974. An approach to environmental psychology. Cambridge, MA. Mittal, B., Sheth, J., 2001. Value Space Winning the Battle for Market Leadership: Lessons from the World’s Most Admired Companies. McGraw-Hill, New York. O’Brien, H., 2010. The influence of hedonic and utilitarian motivations on user engagement: the case of online shopping experiences. Interact. Comput. 22 (5), 344–352. O’Cass, A., Carlson, J., 2012. An empirical assessment of consumers’ evaluations of web site service quality: conceptualizing and testing a formative model. J. Serv. Market. 26, 419–434. O’Cass, A., Ngo, L., 2011. Examining the firm’s value creation process: a managerial perspective of the firm’s value offering strategy and performance. Br. J. Manag. 22 (4), 646–671. O’Cass, A., Sok, P., 2013. Exploring innovation driven value creation in B2B service firms. J. Bus. Res. 66 (8), 1074–1084. Oh, L., Teo, H., Sambamurthy, V., 2012. The effects of retail channel integration through the use of information technologies on firm performance. J. Oper. Manag. 30 (5), 368–381. Oliver, R.L., 1980. A cognitive model of the antecedents and consequences of satisfaction decisions. J. Market. Res., 460–469. Oliver, R., 1999. Whence customer loyalty? J. Market. 63, 33–44. Oliver, R.L., 2010. Satisfaction: A Behavioral Perspective on the Consumer, second ed. M.E. Sharpe, Armonk. Ostrom, A.L., Parasuraman, A., Bowen, D.E., Patrício, L., Voss, C.A., Lemon, K., 2015. Service research priorities in a rapidly changing context. J. Serv. Res. 18 (2), 127–159. Overby, J., Lee, E., 2006. The effects of utilitarian and hedonic online shopping value on consumer preference and intentions. J. Bus. Res. 59 (10), 1160–1166. Parasuraman, A., Grewal, D., 2000. The impact of technology on the quality-valueloyalty chain: a research agenda. J. Acad. Market. Sci. 28 (1), 168–174. Parasuraman, A., Zeithaml, V., Malhotra, A., 2005. E-S-QUAL: a multiple-item scale for assessing electronic service quality. J. Serv. Res. 7 (3), 213. Ringle, C., Wende, S. and Will, A., 2005. SmartPLS 2.0. Hamburg. Rohm, A., Swaminathan, V., 2004. A typology of online shoppers based on shopping motivations. J. Bus. Res. 57, 748–757. Rose, S., Clark, M., Samouel, P., Hair, N., 2012. Online customer experience in e-Retailing. J. Retail. 88 (2), 308–322. Ruiz, D., Gremler, D., Washburn, J., 2008. J. Bus. Res. 61, 1278–1291. Rust, R., Huang, M., 2014. The service revolution and the transformation of marketing science. Market. Sci. 33 (2), 206–221. Sánchez-Fernández, R., Iniesta-Bonillo, M., 2009. Efficiency and quality as economic dimensions of perceived value: Conceptualization, measurement, and effect on satisfaction. J. Retail. Consum. Serv. 16 (6), 425–433. Schlosser, A., White, T., Lloyd, S., 2006. Converting web site visitors into buyers: how web site investment increases consumer trusting beliefs and online purchase intentions. J. Market. 70, 133–148. Sheth, J., Newman, B., Gross, B., 1991. Why we buy what we buy: a theory of consumption values. J. Bus. Res. 22 (2), 159–170. Ström, R., Vendel, M., Bredican, J., 2014. Mobile marketing: a literature review on its value for consumers and retailers. J. Retail. Consum. Serv. 21 (6), 1001–1012. Sweeney, J., Soutar, G., 2001. Consumer perceived value: the development of a multiple item scale. J. Retail. 77, 203–207. Truong, Y., 2013. A cross-country study of consumer innovativeness and technological service innovation. J. Retail. Consum. Serv. 20 (1), 130–137. Turel, O., Serenko, A., Bontis, N., 2010. User acceptance of hedonic digital artifacts: a theory of consumption values perspective. Inf. Manag. 47 (1), 53–59. Verhoef, P., Lemon, K., 2013. Successful customer value management: key lessons and emerging trends. Eur. Manag. J. 31 (1), 1–15. Verhoef, P., Lemon, K., Parasuraman, A., Roggeveen, A., Tsiros, M., Schlesinger, L., 2009. Customer experience creation: determinants, dynamics and management strategies. J. Retail. 85, 31–41. Walsh, G., Shiu, E., Hassan, L.M., Michaelidou, N., Beatty, S.E., 2011. Emotions, storeenvironmental cues, store-choice criteria, and marketing outcomes. J. Bus. Res.

102

J. Carlson et al. / Journal of Retailing and Consumer Services 27 (2015) 90–102

64 (7), 737–744. Walsh, G., Shiu, E., Hassan, L., 2014. Replicating, validating, and reducing the length of the consumer perceived value scale. J. Bus. Res. 67 (3), 260–267. Wang, S., Beatty, S., Mothersbaugh, D., 2009. Congruity’s role in web site attitude formati. J. Bus. Res. 62, 609–616. Wang, X., 2011. The effect of unrelated supporting service quality on consumer delight, satisfaction, and repurchase intentions. J. Serv. Res. 14, 149–163. Wetzels, M., Odekerken-Schroeder, G., van Oppen, C., 2009. Using PLS path modeling for assessing hierarchical construct models: guidelines and empirical illustration. MIS Q. 33 (1), 177–195. Woodall, T., 2003. Conceptualising ‘value for the custome. Acad. Market. Sci. Rev. 12, 1–42.

Woodruff, R., 1997. Customer value: the next source for competitive advantage. J. Acad. Market. Sci. 25, 139–153. Yang, S., Lu, Y., Chau, P., 2013. Why do consumers adopt online channel? An empirical investigation of two channel extension mechanisms. Decis. Support Syst. 54, 858–869. Yoo, W., Lee, Y., Park, J., 2010. The role of interactivity in e-tailing: creating value and increasing satisfaction. J. Retail. Consum. Serv. 17 (2), 89–96. Zeithaml, V., 1988. Consumer perceptions of price, quality and value: a means-end model and synthesis of evidence. J. Market. 52, 2–22. Zhang, J., Farris, P.W., Irvin, J., Kushwaha, T., Steenburgh, T., Weitz, B., 2010. Crafting integrated multichannel retailing strategies. J. Interact. Market. 24 (2), 168–180.