Bridging service dominant logic and the concept of customer value through higher order indexes: Insights from hospitality experiences

Service-dominant logic (SDL) and the concept of customer value (CCV) are both phenomenological approaches to value creation, deeply applied to tourism services. Conceptual and empirical models linking CCV value types (efficiency, excellence, status, esteem, play, aesthetics, ethics, escapism) and SDL actors (firm, employees

These discussions have encouraged service theory scholars to build and elaborate on the value cocreation process ("how"), but less on the notion of value itself, that is on the "what" is co-created (Helkkula et al., 2012). In tourism literature, some authors (Mohammadi et al., 2021) have highlighted a major attention to the co-creation process (antecedents and consequences) and minor to the cocreation context (actors and types).
There is no doubt that the inexorable breakthrough of SDL has restored the notion of value to its rightful place, where value is "ultimately, the most central concept" (Vargo and Lusch, 2012, p. 1). However, among the phenomenal SDL expansion, few authors (except works by Helkkula and co-authors) have elaborated on value as an experience, which is the major basis of seminal conceptualizations of value (Holbrook, 1999;Leroi-Werelds, 2019). Indeed, "value as an experience" (or the experiential approach to value) is an extensive, broad, and seminal stream of research for marketing and consumer behaviour (Arnould, 2007;Leroi-Werelds, 2019;Zeithaml et al., 2020). After early works in mid 80s Holbrook and Hirschman, 1982), the essence of value as experience conceptualization was embodied in Holbrook's well-known concept of customer value (hereafter CCV). The concept of experience is fully contextual (Akaka et al., 2015;Holbrook, 1999); accordingly the expansion of the experiential approach across all business and management settings has been wide over decades (Jaziri, 2019); but although tourism and hospitality services are fully experiential services (Brunner-Sperdin et al., 2012;Gallarza and De Diego-Velasco, 2018;Kandampully et al., 2018;, some authors (Barnes et al., 2020) have pointed out a research gap on how specifically experiential value affects tourism encounters.
Considering SDL as an "open source" (Vargo and Lusch, 2008, p. 1) and CCV potentially "serving as a paradigm for researchers¨ (Smith, 1999, p. 157), we throw attention on unresearched connections between both frameworks. Precisely, the paper aims at understanding value creation drivers in hospitality through a comprehensive model of value dimensions (from CCV) and value co-creating actors (from SDL) such as the firm, the employees, and the guest.
The paper seeks to progress into the characterization of value-in-experience (Heinonen, 2022;Helkkula et al., 2012); by integrating the phenomenological typology of value as experience (Holbrook, 1999). Accordingly, the paper looks for a cross-fertilization derived from the richness and complementarity of the SDL and CCV perspectives. As a difference from others, the paper focusses on value as an experience where the customers integrate different value types from different actors, by proposing both conceptual and empirical models: these models depict the analytical view of the CCV 2*2*2 typology of values and connect these values to the different agents co-creating value-in-use for the consumer. We follow a twofold approach (deductive and inductive) and (a) theoretical and (b) empirical objectives.
(a) Theoretically, we build a conceptual model linking postulates from SDL (different actors and different values) with those from CCV (different value dimensions as elements of experiential value).
(b) Empirically, following others applying experiential value and/or SDL to tourism and hospitality services (e.g., Barnes et al., 2020;Gallarza et al., 2017;O'Cass and Sok, 2015;Prebensen et al., 2013;Prebensen and Xie, 2017;Shulga and Busser, 2020) we study value co-creation in a hospitality experience, with a higher-order structural model. The model gives rise to an experiential value co-creation index which is tested as: (b.1) Interactive because it regroups different value co-creation dimensions into three actors (the firm, the employees, and the customer) according to literature on SDL in tourism (namely O'Cass and Sok, 2015), and (b.2) Contextual, that is, it varies according to individuals and circumstances (tested for significant differences by gender, educational level, and travel frequency).
With these objectives (theoretical and empirical) and this methodological approach (a construction and analysis of partial and global indexes), this work is relevant for both scholars and managers. Theoretically and empirically, it contributes to the value co-creation literature by illustrating how consumers holistically co-create the value-in-use of a hospitality experience with multiple dimensions of experiential value. We illustrate how the creation of value occurs, by integrating value(s) which, in keeping with other works (O'Cass and Sok, 2015;Prebensen et al., 2013;Shulga and Busser, 2020) correspond to the resources provided by the firm ("value proposition"), the employees ("value offering") and the consumers themselves ("perceived value-in-use"). In this sense, our paper answers relevant calls (Holmqvist et al., 2020, p. 114) which have pointed out that "The literature on the entire value process remains limited, however, with a lack of empirical evidence on how value is (co)created and what the different activities of service providers and customers are". Managerially, the paper helps hotel managers by providing them with knowledge (models) and tools (indexes) for addressing and measuring the contribution to value creation of different resources (infrastructures, employees, and the consumers themselves as co-creators). Any tourism marketer is supposed to assure, especially after the sanitary crisis, value co-creation through context-based marketing, and assure the highest customization (Andersson, 2007, Gallarza andGil-Saura, 2020) through competitive combinations of values (Goldsmith and Tsiotsou, 2012;Suni et al., 2022;Bayraktaroğlu and Kozak, 2022). A more analytical understanding of both tourist experiences and value creation in terms of dimensions and actors is therefore necessary in the contemporary tourism industry.
The paper is structured as follows. To motivate and position our study, we first review experiential value according to Holbrook, and value co-creation in SDL. Second, a theoretical model is proposed, and research questions are derived. Third, the empirical study is introduced: questionnaire, data collection procedure and sample description. The analysis and results present different indexes as higher-order models and test them as interactive and contextual. We conclude by signalling contributions and outlining future lines of research.

Conceptual framework 2.1. Experiential marketing and experiential value
The experiential approach arouses in mid-80s in Academia to broaden the scope of previous explanations of consumer behaviour, taking off from the rational (and restrictive) information processing paradigm: main postulate of this perspective is that consumers are guided by subjective, emotional and/or symbolic appreciations in their consumption decisions Holbrook and Hirschman, 1982). The experiential approach shifted from a purely scholar perspective to the business-oriented literature, mainly in works by Pine and Gilmore (1999) and the Experience Economy (based on four axes: Education, Entertainment, Aesthetics and Escapism) or to Schmidt (1999) and the strategic experiential modules (SEMs): sensory experiences (SENSE); affective experiences (FEEL); cognitive experiences (THINK); physical experiences (ACT); and social-identity experiences (RELATE) Both frameworks have been applied to tourism settings (see respectively e.g., Pine, 2002 andSantos et al., 2019).
Customer experiences have been a continuously favoured area of study (Grönroos, 2020), always conceiving consumption as fully multidimensional (Chen et al., 2020;Suni et al., 2022) and therefore its study as multidisciplinary (Arnould, 2007;Bayraktaroğlu and Kozak, 2022). Nowadays, the contemporary challenges for customer experience researchers are the integration of offline and online experiential knowledge (Jaziri, 2019), and bringing together digital, physical, and social realms (Bolton et al., 2018).
This broad and comprehensive view of consumption is fundamental for marketing as a science (Gallarza et al., 2011), and for characterizing tourism as a field of inquiry (Chen et al., 2020;Gallarza and De Diego-Velasco, 2018;Gilmore and Pine, 2002;Goldsmith and Tsiotsou, 2012;Kandampully et al., 2018;Walls et al., 2011). Managerially, similar importance is applicable to the tourism industry: as clearly stated by Goldsmith and Tsiotsou, 2012, p. 207) "the concepts described by 'experiential marketing' theory are highly relevant to the task of managing tourism services". Indeed, experiences are also a competitive leverage for tourism as they "provide a prospective basis for value-adding strategies" (Chen et al., 2020(Chen et al., , p. 1407. The experiential approach underlines the intrinsic subjectivity of consumption (Holbrook, 1999;Woodard and Holbrook, 2013) together with the prominent role of senses and sensations (Schmitt, 1999). Considering consumers as "feelers" "thinkers" and "doers", this perspective highlights the symbolism of consumption and the necessary pleasure derived from it; that is, nothing or very little escapes the experiential approach. For that reason, it is continually revisited, in many settings as hospitality or tourism (Kandampully et al., 2018;Walls et al., 2011). Indeed, empirically, it has been adopted in tourism literature for studying very diverse and varied settings: some of the latest being fishing competitions (e.g., Suni et al., 2022), pilgrimage (Chang et al., 2020) or wine tourism (Santos et al., 2019).
The concepts of value and experience are inextricably linked. Value has a highly complex nature that resides in experiences (Tynan et al., 2014). Although different approaches to value conceptualization exist, as known and categorized in works such as Gallarza et al. (2017) or Zeithaml et al. (2020), out of our understanding, there is ontologically one that predominates which considers value as an experience; this is the perspective which is common to both the SDL and the CCV frameworks. In words by Helkkula et al. (2012, p. 60) it is "the epistemological and ontological foundations of value in the experience which allows authors to characterize value as an experience"; this is what Holbrook clearly states in his conceptualization of CCV as a "interactive relativistic preference experience" (1999, p. 5), In addition to an axiological definition of value, Holbrook's typology of value considers eight types (efficiency, excellence, play, aesthetics, status, esteem, ethics, and spirituality) resulting from a 2*2*2 matrix. These key dimensions of consumption (cf. Table 1) are: intrinsic versus extrinsic (when consumption is understood as "an end in itself" or it has a functionality); self-oriented versus otheroriented (when consumption adopts a social dimension, at a meso or macrolevels); and active versus reactive (depending on the sense of the interaction between the subject and the object).
Holbrook's conceptualization of value has raised many conceptual discussions (e.g., Leroi-Werelds, 2019;Smith, 1999). It has also served as the conceptual basis for multidimensional measures of experiential value (e.g., Gallarza et al., 2017;Mathwick et al., 2001). Recent research (Leroi-Werelds, 2019) has re-launched interest in Holbrook's conceptualization, by updating and enlarging some of the names for value dimensions, encompassing new phenomena such as technologies, human contact, collaborative consumption, service ecosystems and transformative service research. And Holbrook himself has been updating and integrating new aspects of his approach in recent compilations of works on customer value (e.g., Holbrook et al., 2020;Holbrook, 2020). Distinctions from his 2*2*2 framework have been used in tourism literature for analysing destinations (e.g., Bayraktaroğlu, and Kozak, 2022), vegetarian restaurants (Sánchez-Fernandez et al., 2009) and hotel stays (Gallarza et al., 2017). Indeed, Holbrook's framework on customer value seeks to explicitly offer a "systematic and integrated approach" (Holbrook, 1999, p. 3), which is also the idiosyncrasy of the SDL as a mindset for understanding value creation in markets and interchanges (Akaka et al., 2021), as explained in the following section.  (Holbrook, 1999, p. 5) Source: Holbrook (1999)

SDL and value co-creation
As known, SDL provides a strategic vision for organizations, based on the creation of greater value with -rather than for -customers (Karpen et al., 2012). This value co-creation process revolves around service as the center of exchange, integrating all actors (Akaka et al., 2021; and generating unique value for each customer, combining capacities, resources, and knowledge (Vargo and Lusch, 2008). It involves suppliers, consumers, and intermediaries, highlighting the role of consumers as cocreators of value (Ballantyne and Varey, 2008).
The SDL development (Vargo and Lusch, 2012; 2017) generated a particularly interesting debate with others (e.g., Grönroos, 2011) helping to enrich the perspective, with extensive debate over the production, creation and/or co-creation of value (e.g., Grönroos and Voima, 2013;Gummerus, 2013). Undoubtedly, one of the qualities of SDL is its "open-source" nature, to accept and integrate criticism and proposals contributed by others with significant reformulations. Given its bibliometric impact (Vinhas et al., 2017) the SDL is with no doubt a "necessary step" Vargo, 2011, p. 1298) for both scholars and practitioners. But to the best of the authors' knowledge, this "open source" mindset has not produced synergies with the CCV concept, as posted in this paper, nor inside nor outside the tourism literature.
At the epicentre of SDL is the concept of value, or rather the notion of value co-creation. The genesis of the value co-creation concept corresponds to Prahalad and Ramaswamy (2000) in the Harvard Business Review; later in 2004, they elaborated on value, experience, and co-creation, stating that "the co-creation experience of the consumer becomes the very basis of value" (2004, p. 7). Later, Ramaswamy (2011) goes even further as a response to SDL authors, and adopts this approach of value as a co-creation experience and encourages to adapt it as a human lens that applies to any service consumption-that is, value is a function of service.
Vargo and Lusch further developed the concept of value co-creation (2008; 2012; 2017): by reflecting the classic distinction between transaction value and acquisition value (Thaler, 1985), value creation is considered from two different perspectives: value-in-exchange and value-in-use, the latter being the one closer to CCV, and therefore the one chosen in our study. SDL considers the roles of supplier and consumer jointly and value is co-created by means of the interactions between the two . As a result, value co-creation in SDL emerges as a process which is greatly affected by the user's characteristics (Grönroos and Voima, 2013). In this sense, as explicitly expressed by Andersson (2007, p. 46) for the case of tourists' experiences, "value created by consuming tourism experiences depends not only upon the objective experience but also upon the tourist and the tourist's state of mind at that particular moment". This is what is under the rationale of posting our second objective, related to the contextuality of value in terms of differences by personal circumstances (gender, travel frequency and level of education), which has had so far had scarce empirical evidence.
Considering the basis of experiential value presented above (Holbrook, 1999), value co-creation is necessarily experiential; as Vargo and Lusch make clear: "value is uniquely and phenomenologically determined by the beneficiary" (Vargo and Lusch, 2008, p. 7).
To us, because of their validity, scope, and depth, both frameworks should be better integrated for a more comprehensive understanding of value creation. And we therefore think there is support for bridging both approaches in a conceptual model as proposed in the following sections.

Experiential value dimensions as resources to integrate in the light of the SDL
To achieve our theoretical objective (i.e., conceptual connection between SDL and CCV), we propose a theoretical model ( Figure 1). Literature on value has sometimes been criticized because of lacking theoretical and conceptual inquiry, before undertaking any empirical work (Gallarza et al., 2017;Leroi-Werelds, 2019;Zeithaml et al., 2020); we support this critic, for such a complex and elusive concept as value creation. Figure 1 outlines fundamental synthetical elements that are core to the experiential value literature (Holbrook, 1999) and to SDL in general (Vargo and Lusch, 2008), leading to value co-creation in tourism (O'Cass and Sok, 2015).
Firstly, the model shows Holbrook's conceptual framework which identifies eight types of value (in the dashed-line circles), reorganized into four groups (from top to bottom: hedonic, economic, altruistic and social values) and secondly, SDL elements (in thick line), such as the creation of value-in-exchange and value-in-use (Grönroos, 2011), "operand" resources (inanimate: decoration, layout…) and "operant" (knowledge and employees' capabilities (Vargo and Lusch, 2008; and the three origins of said value creation, that is, firm, employees and consumers themselves (O'Cass and Sok, 2015). Accordingly, the model encompasses three types of value in the columns, following O'Cass and Sok's (2015) nomenclature: A) value proposition, which the firm provides from the back office; B) value offering in the interaction with employees from the front office and C) perceived value-in-use generated by the consumer's own experience.
Holbrook's (1999) eight dimensions of value are regrouped as follows into three types: -Value proposition. Aesthetics, efficiency, and ethics are designed and implement by the firm (i.e., value proposition); indeed, the environment or servicescape (aesthetics) and the introduction of processes (which contribute to efficiency), and the CSR and pricing policies (helping to deliver ethical value) are decisions made by the company and its managers, providing (mostly operand) resources to co-create value with the consumer. -Value offering. At a second level, employees (i.e., value offering) are responsible for the provision of excellence (as an interpersonal service quality) and of a potential social value (esteem as a reactive value) perceived by customers in their interaction with employees. -Perceived value-in-use. Consumers themselves are directly responsible for creating entertainment or play (having fun and enjoyment as an active value derived from the experience), escapism (as the way to get relaxation and to escape from routine) and status (to achieve an active social positioning through the hospitality experience).

Research questions
To answer our first empirical objective (b1) on the existence of a regrouping of the dimensions of value co-creation into three origins (firm or company, employees, and customer), we propose the first research question (RQ1). As seen in Figure 1, value co-creators in SDL are different agents, that is, "company, employees, and consumers" (O'Cass and Sok, 2015, p. 190), and they play different roles within a "reciprocal nature of value propositions" (Truong et al., 2012, p. 197). The firm is considered as "value facilitator" (Vargo and Lusch, 2008) providing access to main operant (e.g. facilities, layout…) and operand resources (e.g., the CSR and HR policies, brand image…); our model adopts the value-inexperience conceptualization, stands for the value-in-use emerging in the consumer (here the guest)'s sphere (Grönroos and Voima (2013). Similar argument regarding the interactions happening in consumer experiences from an experiential perspective are also stated by Smith (1999) for depicting the CCV framework, as different value(s) with different intensities can be identified in the same consumption experience.
Therefore, we propose our first research question: RQ1 Can (Holbrook's) experiential dimensions of value be empirically regrouped into a threefold scheme of value proposition, value offering, and perceived value-in-use as depicted in Figure 2?
To explore our second empirical objective (b2) of testing value co-creation as contextual (i.e., it varies according to individuals and circumstances), we propose a second research question. Exploring differences in perceptions of value is to go deeper into the most basic principles of experiential consumption. The essential nature of value as highly "relativistic" is key for Holbrook's definition of Resource integration (operant and operand) Source Value facilitators CCV: "relativistic because it is comparative (involving preferences among objects), personal (it varies across people), and situational (specific to the context)" (Holbrook, 1999, p. 6).
Moreover, the relativity of value also affects SDL because value is always "contextually bound" (Grönroos and Voima, 2013, p. 146), or to quote Vargo and Lusch (2013), "value must be understood in the context of the beneficiary's world" (p. 91). Some works have already highlighted interest in the context within SDL (Jaziri, 2019, p. 243) by stating that "the social aspects of the subject have assumed increasing significance in respect of the S-D logic paradigm. Consequently, the service experience data are simultaneously individual (intra-subjective) and social (inter-subjective)". Our second research question stands for measuring value contextuality.

RQ2 Value co-created by different agents is contextual because it varies according to individuals and circumstances (gender, educational level, and travel frequency).
This relativity/contextuality of value is explicitly posted as a future research avenue in Leroi-Werelds (2019, p. 667): "Relationships between value types personal and situational factors". But although most theory agrees on the contextual nature of value (Akaka et al., 2015;Grönroos and Voima, 2013;Holbroook, 1999;Vargo and Lusch, 2013), this contextuality has scarcely been proven in (hospitality) services. Recent state of the art on value co-creation in tourism (Mohammadi et al., 2021, p. 305) evidenced this gap, with a systematic mapping study of 137 academic papers from 2006 to 2019, highlighting that: "the studies focused on the antecedents and prerequisites for co-creation and paid less attention to the co-creation context".

Empirical study
Differing from other works on value and co-creation that cover different agents such as managers, employees and consumers (e.g., Nasution and Mavondo, 2008;O'Cass and Sok, 2015) or other stakeholders such as suppliers, customers and sharing platforms (e.g., Andreassen et al., 2018), in our case, we adopt a phenomenological approach (characteristic of both SDL and CCV) to analyse customers 'perceptions of value-in-use. We therefore follow previous theoretical works where customer value emerges solely in the customer sphere (e.g., Grönroos and Voima, 2013;Grönroos and Ravald, 2011;Heinonen et al., 2010). The service context chosen is hotel stays (proven by others as a "paradigmatic experiential service settings" (Gallarza et al., 2017, p. 737)).

Designing the questionnaire
To transfer Holbrook's conceptual framework into scales, we follow a similar procedure to other empirical approaches based on CCV (e.g., Gallarza et al., 2016;Gallarza et al., 2017;Leroi-Werels et al., 2014;Mathwick et al., 2001;Sánchez et al., 2009), also suggested in Leroi-Werelds (2019) for value measurement: we adapt existing scales from tourism literature, the ones which can be interpreted from the dual perspective of CCV and SDL For efficiency and excellence, we adapted respectively Tsai's (2017) utilitarian scale, and Brunner-Sperdin et al.'s (2012) "human ware" scale (with service quality dimensions like empathy, competence or response capability). Chen and Hu's (2010) work on coffee outlets was used for both social values: symbolic value scale for status (active other-oriented benefits of staying at this hotel) and social benefits (a subscale from relational benefits) for esteem. The choice of these scales reflects the status/esteem dichotomy and is in accordance with Holbrook's statements on what is active vs. reactive values (see Holbrook, 1999, pp. 15-17): the symbolic value of status is that which is actively achieved, while the relational value of esteem is produced reactively through interacting with employees as operant resources. Hedonic values' scales (aesthetics and play) correspond respectively to Brunner-Sperdin et al.'s (2012) "hardware" dimension (design, lights, colour, and sound) and to Tsai's (2017) experiential value dimension. Finally, as regards altruistic values, which are more difficult to generalize because less frequently researched, two indicators of ethical value in Gallarza et al. (2016) were chosen for ethics, concerning ecological orientation and price transparency in hotels. Spirituality is translated into escapism, in line with other studies (e.g., Gallarza et al., 2017;Mathwick et al., 2001;Sparks et al., 2008), choosing to adapt Wu and Liang's escapism (2009) The result is a set of 24 indicators, all measured on 7-point Likert scales.

Data collection
Data were collected through personal interviews in hotels in the region of Valencia (Mediterranean coast of Spain). The main population under study comprised hotel guests located in this geographical area. The province of Valencia is a benchmark for Spain's hospitality industry as it is the second largest autonomous community in terms of accommodation offer and employees (INE, 2022). Two criteria were applied to select the hotels whose guests were surveyed: geographical areas, and number of stars. According to first criterion hotels in two cities in Valencia province were considered: Valencia and Gandía. Valencia is the third biggest city and one of Spain's top tourist destinations. It represents the country's general profile, being both a leisure and business destination. Gandía is a top destination for "sun and beach" tourism in the province. An initial list was drawn up from Spain's official hotel guide and the hotel directory of the Valencian Tourism Agency. This list was completed with information from the SABI (Iberian Balance Analysis System) and DUNS100,000 databases. Following the previous studies, from the final list, various categories of hotels (second criterion) were considered: three-, fourand five-star hotels were selected (two five-star hotels, 22 four-star hotels and 18 three-star hotels). Moreover, considering our research objective and according to several studies (e.g., Heerwegh and Loosveldt, 2008), the fieldwork was carried out through face-to-face surveys with the aim to achieve a representative sample of the study population.
For the purposes of completing the questionnaire, authorisation was requested from the management to conduct interviews in the hotel reception area. The questionnaire was conducted by a survey company in hotel lobbies during mornings and evenings. A non-probabilistic convenience sampling approach was used. 1175 respondents were intercepted, obtaining 340 complete and valid questionnaires (28.9% response rate). The main characteristics of the sample are shown in Table 2.

Construction of value indexes
The methodology used to integrate both our objectives into measurement, was Leroi-Werelds' (2019, p. 663) advice on building indexes, which is in line with our search for multidimensional interaction (objective b1) and contextuality (objective b2):

"the use of a formative measurement index is recommended … not only does it take into account the situational nature of the construct … by adapting the content of the index (i.e. the relevant value types) to the context under study, it also considers the multidimensional nature of customer value … by operationalizing customer value as a formative construct formed by the relevant positive and negative value types … as first-order constructs".
We therefore built an index of experiential value co-creation (EVcC), based on a higher-order structural model, as a way of making Figure 1 operative. Other works on value (e.g., Gallarza et al., 2017;Lin et al., 2005;Martín-Ruiz et al., 2008;Yi and Gong, 2013) have also used hierarchical orders for value, often in causal models, but few propose the construction of indexes (See Figure 2).

Figure 2. Structural Model of Value Co-Creation
In our case, indicators from the eight value scales of CCV are observed variables, and the three types of value from O'Cass and Sok's (2015) (value proposition, value offering and perceived-value-in-use) are second-order constructs which in turn conform a third-order index: the experiential co-created value index. To march our conceptual model from Figure 1, a reflective-formative-formative structure is proposed, where efficiency, ethics and aesthetics make up the second-order construct value proposition (VP), excellence and esteem form value offering (VO) and finally, play, escapism, and status form perceived value-in-use (PViU). These three second-order constructs jointly make up experiential cocreated value (EcCV) as a third-order construct. In all cases, the higher-order constructs are created following the hierarchical components approach (HCA) initially suggested by Wold (1982), known as the repeated indicators method (Lohmöller, 1989;Wold, 1982) or superblock method (Tenenhaus et al., 2005). First-, second-and third-order constructs have reflective indicators, but the higher-order constructs are formative in their preceding constructs (Leroi-Werelds, 2019).

Psychometric properties
The reliability of the eight scales (Table 3) was tested with Cronbach's alpha. Only ethics (0.57) is below the 0.70 threshold (Nunnally and Bernstein 1994). Composed reliability (CR) is over the 0.70 threshold (Werts et al., 1974) in all cases, including ethics; and the first eigenvalue is clearly higher than the rest, and the second eigenvalue is less than one, for ethics as well. Therefore, we keep the ethics scale and confirm the one-dimensionality of the eight scales.   (Fornell and Larcker, 1981); the square root (over 0.7), and correlation of each indicator with its construct is over 0.7 (loadings in bold in Table 4). Discriminant validity is also confirmed (Table 4) with loadings greater than cross-loadings (Barclay et al., 1995;Chin, 1998), and a square root of AVE for each construct (cf. Table 3) greater than its correlation with the other constructs (Fornell and Larcker, 1981). In addition, the Heterotrait-Multitrait ratios approach suggested by Churchill (1979, p. 66) was used to calculate (HT/MT) correlation ratios (Henseler et al., 2015), all being beneath the 0.90 threshold (Clark and Watson, 1995;Kline, 2011).

Value co-creation: higher-order models
The model in Figure 3 depicts value co-creation through an overarching experiential co-created value index (EcCV) and three partial value indexes: value proposition (PV), value offering (VO) and perceived value-in-use (PViU). Firstly, the contribution of the three partial indexes to the overall EcCV index is both significant and well balanced: 0.40 for value proposition (corresponding to customers' perception of the company's efforts to integrate operand resources (layout) and policies (prices); 0.401 for value offering (reflecting the role of employees as operant resources); and 0.360 for perceived value-in-use (as the value generated by consumers themselves in the experience).
Considering the second-order model, the results allow to corroborate Holbrook's claim (1999, p. 9) that "one dimension of value cannot be understood in isolation from the rest". Furthermore, the results --significant structural model loadings, and descriptive values for each dimension (cf. Appendix)--endorse Smith's (1999, p. 149) premise on Holbrook's CCV, that "different types of value occur simultaneously and in various degrees of the consumption experience", and that these "types of value may be subject to a higher order classification". This is evidenced here by links between the dimensions of value and the three partial measures. We compare results with the higher-order models for value dimensions proposed by Yi and Gong (2013) and Martín-Ruiz et al. (2008). The latter also obtained balanced contributions, although all of them are less than 0.4 (cf. Figure 1 in this work). In our case, the specific loadings of the three partial indexes in the EcCV index are better balanced (between 0.41 and 0.36). Nevertheless, we cannot say the same of the contribution of each dimension to the formation of those partial indexes.
In the case of value proposition, efficiency (0.454) and aesthetics (0.601) show stronger contributions than ethics (0.187). This weak effect of ethics as a value dimension is similar, for example, to the findings of Gallarza et al. (2016) and endorses the difficulties of implementing this variable (Smith, 1999). Results on ethics are difficult to discuss because there are limited studies on an ethical dimension of value (Gallarza et al., 2017;Leroi-Werelds, 2019). Replicas are needed so as not to conclude that in consumers' eyes the choice of processes (efficiency), furniture and decoration (aesthetics) take precedence in value creation over the company's ethical responsibility.
In the case of value offering, excellence (0.652) is more relevant than esteem (0.443), highlighting the predominant role of employees as operant resources in value co-creation. Customers integrate the employees' knowledge and skills, by generating reactive values in Holbrook's nomenclature (see Table  1). In the same regard, the fourth indicator of excellence (i.e., "In this hotel, employees know how to take an active part in service provision") has the highest mean (5.79/7), depicting a a prominent role of employees in value co-creation in hospitality services (Gallarza et al., 2017).
Finally, for perceived value-in-use, play (0.492) and escapism (0.466) present the greater contribution to value co-creation, in comparison to status (0.237), which is logical for a hedonistic experience. Similar prominence of hedonic values (and more precisely escapism) in value creation with different actors is found in Holmquist et al. (2020) for luxury services.
All the above furthers our first empirical objective, answering RQ1 in the affirmative.

Value contextuality: results and discussion
To answer our second research question, we studied the effect of gender, educational level and travel frequency on the higher-order index (EcCV) and on the partial indexes (VP, VO and PViU). Thus, four multiple linear regression models were estimated (see coefficients and p-value in Table 5). Gender has no significant effect on any index; so, attention is focused on the other two factors (see Figure 4).

Figure 4. Contrasting the Effects of Gender, Educational Level and Travel Frequency on the four indexes
In all cases (VP, VO, PViU and EcCV), a higher educational level is associated with a lower-value index, whereas greater travel frequency corresponds to a higher-value index just for VP and PViU but to a lower-value index for EcCV. The effect of travel frequency in the case of VO index is not significant (the corresponding lines in Figure 4 are close to each other). We therefore reply affirmatively to our second research question, but only partially. Remarkably, educational level has an opposite effect, suggesting therefore that a less educated tourists perceive higher levels of experiential co-created value (and higher levels of value proposition, value offering and perceived-value-in-use). As a significant result of this study, tourists with a higher educational level tend to be more demanding towards all value co-creation agents, including themselves.
The effect of travel frequency on two indexes (VP and PViU) shows better value co-creation results for those who travel more. However, this is not the case for the VO index, that is for the contribution made by employees. Moreover, more frequent travellers perceived a lower experiential co-created value, as the overall EcCV index decreases with travel frequency.
This result describes an "experiential experience curve" (forgive the redundancy) in value co-creation during the tourist career: the more we travel, the better we value our own contribution (in terms of play, escapism and less of status) and that provided by the firm (mainly efficiency and aesthetics and less ethics). But this cumulative experience has no effect on the perception of the role of employees in value creation. This idea, in accordance with the experiential approach, is indeed reminiscent of Schmitt's (1999, p. 53) proposal that experiential marketing is based on an ongoing uninterrupted process of storing past experiences, "to create holistic experiences that integrate individual experiences into a holistic Gestalt"; but this gestalt can be positive or negative. This storage of experiences is not mere compilation but involves ongoing improvement and relatedness for collaboration (Shulga and Busser, 2020;Santos et al., 2019;Tsai, 2017), which our results endorse. The exception of employees' contribution in this positive "experience curve" highlights the complex role of staff in providing experiential value in hotels (Barnes et al., 2020). Being a more experienced consumer (higher travel frequency) makes us more critical of the role of employees in our own value perception.

Conclusion
To add to the literature on value co-creation in tourism literature, this work aimed, both theoretically and empirically, to unite the longstanding tradition of experiential value (in Holbrook's conceptualization of CCV) with the contemporary SDL mindset (using O'Cass and Sok's (2015) proposal). At the level of theory, we have proposed a conceptual model (Figure 1) to integrate Holbrook's eight dimensions of value (efficiency, excellence, status, esteem, play, aesthetics, ethics, and escapism) with the SDL literature on value co-creation. Empirically, we have tested an experiential value cocreation (EVcC) index built from the three indexes of partial values: value proposition (VP), value offering (VO) and perceived value in-use (PViU).
The results from 340 hotel guests surveyed in the region of Valencia (Spain) suggest that: a) Experiential value co-creation in Holbrook's dimensions is interactive, integrating three different origins (following O'Cass and Sok (2015, p. 190): the firm, the employees, and the consumers themselves). The firm (value proposition) and the employees (value offering) present a balanced value contribution, while resources from the customer (perceived value-in-use) are slightly less relevant in the process. Employees assume the major role as operant resources from value offering in the overall EcCV index, while ethics and status are the least important dimensions respectively for value proposition and for perceived value-in-use. b) Value co-creation is contextual, to the extent that it varies according to individuals and circumstances: contrary to the level of studies, directly with travel frequency and not significantly with gender. These results are in line with theoretical calls on value contextuality in general (Leroi-Werelds, 2019), and specifically for tourism services (Mohammadi et al., 2021): a less important value creation happens for well-educated guests, and the more these guests travel, the better they perceive the value provided by the hotel (the firm) and that experienced by themselves, but not necessarily the one offered by staff (employees). Accordingly, our findings have evidenced that value creation is significantly influenced by consumer characteristics, in terms of previous experience (travel frequency) and socioeconomic characteristics (level of education), but not regarding demographic ones (gender).

Contributions
Several contributions can be viewed on both the concept of value co-creation and the nature of the process in terms of actors and contexts.
Firstly, our findings contribute to extant service literature by analysing the customer's value process when integrating resources as skills and knowledge provided by different actors in terms of different values. Our results have added to the complementary role of concepts of co-creation and experiences, both in general (Gummerus, 2013;Helkkula et al., 2012) and in the case of tourists . The results provide empirical evidence to previous literature which states that the value process consists of multi-actor interactions (Holmquist et al., 2020) which in our case have been proven to be balanced between three actors: the firm, the employees, and the customer. Indeed, both our models (conceptual and structural) help to interpretate literature on value by attributing roles to each cocreation actor, in terms of dimensions of value that are perceived holistically by the consumer. Results confirm the relevance of the experiential value in hotels to produce personalized and emotional encounters (Barnes et al., 2019).
Secondly, our work can be added to the broad and deep stream of research on "the experiential or phenomenological perspective … (which) identified different value dimensions and contended that their relative importance varies across customers" (Zeithaml et al., 2020, p. 417). In our case, in addition to the relative importance (seen in the weights of the different value dimensions for each partial index), our findings go beyond this and depict the contextual nature of four co-creation processes (three partial and one total): these processes, for a hospitality experience, are similar for men and women, but differ according to consumers' expertise or assiduity (guests' travel frequency) and educational level. In this sense, our work has provided empirical evidence to the main axiom of literature on value creation, that stands that value is contextual (Grönroos and Voima, 2013;Holbroook, 1999;Vargo and Lusch, 2013).
Thirdly, our approach (index construction) answers methodological issues on value measurement, which have been signalled for decades as problematic (Gallarza et al., 2011;Leroi-Werelds, 2019;Zeithaml et al., 2020). With the construction and testing of a global index (EcCV) and three partial indexes (VP, VO and PViU), we have progressed into what Holbrook himself recently claimed was needed: "measurement-related empirical issues… are still embryonic and in need of further development" (Holbrook, 2020, p. 9). Moreover, our proposal of both a conceptual model (Figure 1) and the construction of indexes (Figures 3 and 4) has also offered practical and specific answers to Zeithaml et al.'s (2020, p. 410) third research question: "How can the positivist, interpretive, and social constructionist views on customer value be reconciled to motivate future research and assist managers?" The paper is also bringing managerial contributions in terms of value-creation in hospitality services. First, the theoretical model can help managers to visualize the many sources of value creation in the complex ecosystem of actors and relationships that takes place in hotels; the tourists' interaction with employees (operant) and attractions (operand resources) are facilitators of the (positive or negative) experience co-creation in terms of different value outcomes (VP, VO and PViU). Second, and consequently, the construction of indexes is a tool for guiding practitioners in how to monitor valuecreation efforts. Results here show that all three values are integrated in an uppermost level into EcCV, by the beneficiary. Tourist as end-users might not be conscious of the role of each of the partial values, but managers should be able to identify, monitor and monetize each of them to produce competitive advantages. Third, for the case of hospitality managers, our results on value indexes variation bring insights into customer orientation policies (towards frequent travellers vs. first-time ones, and towards less or more educated tourists): value-based segmentation policies should take the contextual effect of sociodemographic aspects into consideration in value creation. Indeed, hotel managers can consider frequent guests as prominent co-creators of value, inviting them to participate in brand ambassadorship programs. The educational level is also a leverage for data-driven segmentation policies. All these analyses are useful for developing a more nuanced view of the value creation that emerges from the tourists' experience in general (Heinonen, 2022).

Limitations and future research directions
Both the theoretical postulates and the empirical findings of this study should be considered in the light of specific limitations. We first recognize that the regrouping of the eight values from the CCV into the three types of value according to O'Cass and Sok (2015) is the authors' criterion; we aim to seek greater bibliographic support for this in future works. Given the quality of current research on SDL with the profusion of works revising its scope and perspectives, both by its creators (e.g., Vargo and Lusch, 2017) and by others (e.g., Pohlmann and Kaartemo, 2017;Vinhas et al., 2017) we think the present work could invite future replicas, with other taxonomies of experiential values, like those of Schmitt (1999) or Mathwick et al. (2001), to enrich the empirical literature on the experiential approach; but also integrating offline and online experiential knowledge (Bolton et al. 2018;Jaziri, 2019), or adding epistemic and conditional values (Suni et al. 2022). The validity of our results could also be enhanced by extending data collection into other geographies, assuming cultural differences may also affect in value creation processes (Gallarza et al., 2011), but a general overview of actors and interrelationship should remain constant, to support robustness of both models.
Second, although results have proven how value creation processes may vary by "subject" (in Holbrook's terminology) or by "actor" (that is the beneficiary, in SDL lexicon), differences by "object", that is here, by type of hotel, haven't been addressed. This effect on the value creation of the different hotel categories could be tested in future works; in this sense, this comparative works could provide hotel managers of hotel chains with relevant consumer insights, since a five-star hotel would propose different value(s) from a three-star hotel (value proposition by managers' mission and value offering by employees). Moreover, to progress into this contextuality of value creation, situational aspects (effects of time in tourism seasons and space in worldwide chains) could also be measured, with replications in different countries and cultures.
Third, O'Cass and Sok's (2015) original work considers the VP, VO and PViU as value creation phases; that is, it adopts a diachronic comprehension of the different value dimensions and value creators. Our adaptation of their work was synchronic, which must be acknowledged as a limitation: there is no specific test of how the actors involved in the experience were interacting in one moment; this is an assumption made in accordance with the nature of the experience as an interaction (Ramaswany, 2011). But to enhance the worth of service design for understanding the processes of both value creation (Andreassen et al., 2016) and experience (Teixeira et al., 20212, further research should adopt dynamic models, in terms of longitudinal studies. These dynamic models could integrate new insights into the value co-creation process from the spheres of both the service provider and the customer in light of Holmqvist et al.'s (2020) work for luxury retailers, which considers value(s) before, during and after service interaction; in this sense, value anticipation and value learning are two concepts which could help to read the contextuality of value through travel frequency in our results. This dynamicity of the value co-creation process could also be envisaged by integrating interactions with other customers in the model, generators of relationship values; this would help to endorse the significance of the different resources that are part of the complex process of value co-creation.
Fourth, in our empirical study, we assume relative problems in the validity of ethics scale, and given its descriptive scores, also for status one (see Appendix). We call for future studies with the consideration of other scales for these two dimensions. For the case of ethics, the dimensions validated here correspond to two ethical benefits acknowledged for an ethical consumption: ecological and societal benefits (Sudbury-Riley and Kohlbacher, 2016): first item corresponds to ecological benefits as positive effects on the environment, but second item does not correspond to any of the three societal benefits: fair trade, community support or employee fairness. It must be said that a larger scale considering contributions to NGOs was initially included and deleted for a gain of reliability (results not shown). Similarly, price fairness here is considered the only societal driver for value co-creation, which is different from other works considering ethical values (e.g., Gallarza et al., 2017). Changes might be expected for hotel stays after the COVID-19 pandemics, when compliance with sanitary rules would be integrated in the value proposition and will probably be thoroughly considered among the customers' expectation of ethical value provision.
Last, both the conceptual model and the indexes can be enlarged by: a) either adding to the three origins (the firm, the employees, and the customer) other actors such as governments as institutions, robots of chatbots as digital resources, or other customers co-creating at the same time; they should all find their places in one of the eight value dimension, to assure the validity of Holbrook's framework, or/and b) adding extra value dimensions (from the updated taxonomy form Leroi-Werelds (2019) to any of the columns (e.g., authentic value, epistemic value, …). In this sense, the largest the better, in the seminal idea of Ramaswamy (2011)'s on co-creation, that is, to favour "a shift to co-creation of value through human experiences". Studies such as the present one, add evidence to tourism as a "human experience", rich, complex, and multidimensional, being potentially a major vehicle to this expanding process. This virtue of value as a notion and mindset pushes the boundaries of value-in-experience, more specifically in tourism, where value "can be expanded into the customer lifeworld domain" (Heinonen, 2022, p. 1) assuming tourism value transcends the limits of the tourist experience.

Acknowledgement
This research has been developed within the framework of the project Grant ECO2016-76553-R funded by the State Research Agency of the Spanish Ministry of Science and Innovation, the project Grant PID2020-112660RB-I00 funded by MCIN/AEI/10.13039/501100011033 and the consolidated research group AICO/2021/144/GVA funded by the Conselleria d'Innovacio, Universitats, Ciencia i Societat Digital of the Generalitat Valenciana.