Re-examining the relationships among perceived quality,value,satisfaction,and destination loyalty

Abstract

Drawing on the literature on service quality, marketing, and tourism, the present study tests a comprehensive model of the effects of perceived quality (PQ) on loyalty in the context of a tourism destination. Using a sample of 249 residents from the United Kingdom and the United States who visited Australia between 2008 and 2012, this research applies partial least squares–structural equation modeling to examine these relationships. PQ is operationalized as a multidimensional construct determined by six destination dimensions: natural and well-known attractions, variety of tourist services, quality of general atmosphere, entertainment and recreation, general environment, and accessibility. Our results support the conceptualization of PQ as a “reflective first-order, formative second-order” model (also referred to as a molar, or type II, higher order model, see Diamantopoulos A, Riefler P, and Roth KP (2008) Advancing formative measurement models. Journal of Business Research 61(12): 1203–1218). Results from this study found that the six reflective first-order dimensions of PQ form the higher (second)-order PQ construct. Moreover, PQ has a stronger effect on loyalty (both direct and indirect through satisfaction) compared to perceived value, which only exercises an indirect effect on loyalty. The study presents new insights on the operationalization of PQ and the network of causal relationships among PQ, value, satisfaction, and loyalty in tourism destinations.

Keywords

Perceived quality loyalty partial least squares–structural equation modeling (PLS-SEM)hierarchical constructs.

Introduction

Perceptions of quality, value, and satisfaction in a service encounter can ultimately influence consumer loyalty and create positive word of mouth (Yoon et al., 2010). Consumer loyalty enables firms to “secure future revenues, reduce the costs of future transactions, decrease price elasticity, and minimize the likelihood that customers will defect if quality falters” (Anderson et al., 1997: 129). Similarly, positive word of mouth from satisfied customers is a cost-effective marketing strategy that attracts new customers, and builds brand equity and shareholder value (Luo, 2009). In a tourism context, tourists’ perceptions of service quality and value at a destination affect satisfaction, generate customer referrals, and repeat visits. This not only affects the destination but also impacts on the long-term performance of tourism-related enterprises (Bigné et al., 2001).

Tourism research examining the relationships among perceived quality (PQ), perceived value (PV), satisfaction, and loyalty has adopted diverse approaches with multiple, often contradictory findings. For example, models developed from the services literature and later adapted in tourism research posit that consumer satisfaction affects perceptions of value, and value will consequently impact on loyalty (see Yang and Lau, 2015). Others argue differently positing PV to be a direct antecedent of satisfaction (Cronin et al., 2000; Oh, 1999; Tam, 2000). Quality and PV are represented as “cognitive responses” to a service experience, whereas satisfaction is an emotional response, and thus, cognitive responses act as precursors to emotional responses (Bagozzi, 1992; Cronin et al., 2000; Petrick, 2004). An alternative model suggests that customer satisfaction, specifically, is the main predictor of both behavioral intentions and loyalty (see Fornell et al., 1996; Yoon et al., 2010; Ranjbarian and Pool, 2015). However, this model also has its limitations in that “quality is not only mediated by perceived value and satisfaction in the prediction of loyalty and behavioral intentions, but also has a direct relationship on behavioral intentions” (Hutchinson et al., 2009: 301). Another model, referred to as the “interrelationship model” (Cronin et al., 2000), assumes that it is a combination of quality, value, and satisfaction that, collectively, impact consumers’ behavioral intentions and loyalty (Chen, 2008; Hutchinson et al., 2009).

The manner in which “PQ” is defined and conceptualized in empirical studies has also received much attention. Research from the services literature operationalizes PQ as a multidimensional, “higher order” construct formed by a number of first-order dimensions (latent factors), with each dimension measured, in turn, by a set of individual quality items (observed variables) (see Brady and Cronin, 2001; Grönroos, 1984; Parasuraman et al., 1988).

In contrast, tourism studies have often operationalized PQ as a first-order reflective construct, where the observed variables used to measure the latent construct are aggregated together into the first-order dimensions and the aggregate scores (of each dimensions) are used as manifestations of PQ (e.g. Bigné et al., 2001; Chen and Tsai, 2007; Petrick, 2004). Using the first-order reflective constructs to operationalize quality in this case can lead to empirical bias and create measurement errors (see Little et al., 2002). This, in turn, can bias possible causality among quality, value, satisfaction, and loyalty.

Using the first-order reflective constructs can also be problematic from a theoretical point of view as consumers’ perceptions of quality occur at multiple levels. For example, customers first evaluate the quality at the individual attribute level (e.g. assessing natural and scenery attractions at the destination, tourist sites/activities, beaches, and wildlife), the quality of the interaction is then evaluated at the dimensions level (e.g. attributes of natural attractions form the overall quality of the natural environment at the destination). These eventually lead to overall PQ (destination level) evaluation (see Dabholkar et al., 2000; Clemes et al., 2011).

Tourism research has often omitted the higher order PQ construct, choosing instead to examine the first-order subdimensions as separate latent constructs, using them as exogenous variables (predictors) of PV, satisfaction, and loyalty (Yoon et al., 2010; Zabkar et al., 2010). Although using the first-order dimensions of PQ might overcome the problems associated with analyzing higher order models in structural equation modeling (SEM), this approach may compromise the model’s validity and cause construct misconceptualization. Researchers may be inappropriately matching constructs at different levels of “abstraction” when examining the relationships among different constructs; the underlying dimensions of PQ (lower level of abstraction) are used to predict value and satisfaction, rather than examining the effects of a higher order PQ construct (equal level of abstraction) (Edwards, 2001).

A review of previous studies in tourism demonstrates the problems associated with the operationalization of PQ (Table 1). Of the 16 studies found (dating back to 2000), 13 used a first-order reflective construct to operationalize PQ, two used first-order subdimensions as separate latent constructs, and one study adopted a single-item measure. Such practices have led to mixed results and have obfuscated the causal relationships among quality, value, satisfaction, and loyalty in a tourism context (Table 2). For example, studies 5–9 from Table 2 provide evidence that PQ and “PV” have a direct, positive effect on “satisfaction” and that satisfaction has a direct, positive effect on “loyalty.” In other words, PQ and PV have an indirect relationship with loyalty with satisfaction acting as the mediating variable. Other studies (paper 4 and 16) provide different results and argue that PQ and satisfaction have a positive, indirect relationship with “loyalty” with PV acting as the mediating variable. Several cases have found PQ to have both direct and indirect effects on “loyalty” (Table 2). Thus, there is a dissimilitude among the extant research with regard to the network of relationships among PQ, value, satisfaction, and loyalty. The contrasts in empirical findings are often due to the operationalization of the constructs (measurement models) creating problems in validating the structural model.

Table 1.

Past operationalization of PQ in tourism research.

Authors	Single-item construct	First-order subdimensions	First-order latent construct	Higher order formative construct
1. Baker and Crompton (2000)	—	—	√	—
2. Murphy, Pritchard, and Smith (2000)	—	—	√	—
3. Bigné, Sanchez, and Sanchez (2001)	√	—	—	—
4. Petrick and Backman (2002)	—	—	√	—
5. Petrick (2004)	—	—	√	—
6. Gallarza and Saura (2006)	—	—	√	—
7. Chen and Tsai (2007)	—	—	√	—
8. Chen (2008)	—	—	√	—
9. Hu, Kandampully, and Juwaheer (2009)	—	—	√	—
10. Hutchinson, Lai, and Wang (2009)	—	—	√	—
11. Chen and Chen (2010)	—	—	√	—
12. Yoon, Lee, and Lee (2010)	—	√	—	—
13. Zabkar, Brencic, and Dmitrovic (2010)	—	√	—	—
14. Jo, Lee, and Reisinger (2014)	—	—	√	—
15. Ranjbarian and Pool (2015)	—	—	√	—
16. Yang and Lau (2015)	—	—	√	—

PQ: perceived quality.

Table 2.

Summary of identified links among PQ, value, satisfaction, and loyalty in tourism research.

	Endogenous constructs
	PQ	PV	Satisfaction	Loyalty
Exogenous constructs
PQ	—	2, 5, 6, 7, 8, 9, 10, 11, 12, 14, 15, 16	1, 3, 5, 9, 11, 13, 14, 15, 16	1,2, 3, 5, 13
PV	—	—	5, 6, 7, 8, 9, 10, 11, 12, 14, 15	2, 4, 5, 8, 9, 10, 11, 14, 16
Satisfaction	—	4, 16	—	1, 3, 5,6, 7, 8, 9, 10, 11, 12, 13, 14, 15

PQ: perceived quality; PV: perceived value.

Note: Each cell in the table corresponds to the relationship between the exogenous variable (presented in the rows) and the endogenous variable (presented in the columns), with the number shown in the cell corresponding to the paper in which this relationship was supported. For example, paper number 12 (by Yoon, Lee, and Lee, 2010) found that PQ has a direct effect on value, which in turn has a direct effect on satisfaction; finally, satisfaction has a direct effect on loyalty (PQ → value → satisfaction → loyalty). Authors: 1, Baker and Crompton (2000); 2, Murphy, Pritchard, and Smith (2000); 3, Bigne, Sanchez, and Sanchez (2001); 4, Petrick and Backman (2002); 5, Petrick (2004); 6, Gallarza and Saura (2006); 7, Chen and Tsai (2007); 8, Chen (2008); 9, Hu, Kandampully, and Juwaheer (2009); 10, Hutchinson, Lai, and Wang (2009); 11, Chen and Chen (2010); 12, Yoon, Lee, and Lee (2011); 13, Zakbar, Brencic, and Dmitrovic (2010); 14, Jo, Lee, and Reisinger (2014); 15, Ranjbarian and Pool (2015); 16, Yang and Lau (2015).

In an attempt to address these discrepancies, this study develops and empirically examines a theoretically derived “reflective first-order, formative second-order” model (also referred to as a molar, or type II, higher order model, see Diamantopoulos et al., 2008), where the six reflective first-order dimensions of PQ form (arrows pointing towards) the higher (second)-order PQ construct. Using data collected from 249 visitors to Australia, we delineate the primary (first-order) dimensions that form destination PQ as well as identify the specific destination attributes (observed variables) within each dimension that are most influential in forming visitors’ overall quality perceptions. Drawing on theory, we examine the higher order PQ model in a structural network of relationships among PQ, value, satisfaction, and loyalty, presenting new directions of its future use in tourism research. The research presents destination managers with an in-depth understanding of the factors that constitute “quality” in a destination and the drivers of visitor satisfaction. This informs destination managers on the destination attributes that are most critical for the tourism experience, informing new product development and directing marking activities in a targeted manner. The research also presents empirical evidence on the antecedents of consumer loyalty to a destination.

Proposed hypothetical model

Figure 1 depicts the full hypothesized model of this study. It presents the underlying dimensions for each construct (PQ, PV, satisfaction, and loyalty) and the theorized causal relationships among them. PQ is conceptualized as a higher order formative construct determined by six first-order dimensions: natural and well-known attractions, variety of tourist services and culture, quality of general tourism atmosphere, entertainment and recreation, general environment, and accessibility. The 18 destination attributes (observed variables) used to measure the six first-order factors of PQ are explained in Appendix Table 1A.

Figure 1.

Hypothesized higher order model of PQ, PV, and satisfaction on loyalty. PQ: perceived quality; PV: perceived value.

Theoretical framework and hypotheses

Perceived service quality in tourism

Much of the recent studies on service quality follow on from the work of Parasuraman et al. (1988) and Grönroos (1984). The SERVQUAL instrument (Parasuraman et al., 1988) includes five dimensions: tangibles, responsiveness, assurance, empathy, and reliability. Tangibles refers to “physical facilities, equipment, and the appearance of personnel.” Responsiveness is “the willingness to help customers and provide prompt service.” Assurance refers to “knowledge and courtesy of employees and their ability to inspire trust and confidence.” Empathy is “the caring, individualized attention the firm provides to customers” (Petrick, 2004: 399). Reliability, the only outcome dimension in this case, is defined as the ability to perform the promised service dependably and accurately (Zeithaml et al., 2009). The model by Grönroos (1984) defines service in terms of two quality dimensions: “technical quality”—which refers to the outcomes and “functional quality”—which refers to the processes. The functional quality dimension (describing the process dimension) includes subdimensions, such as relational quality (staff–customer interactions) and physical quality (tangibles such as facilities and equipment), which are similar to the process dimensions proposed in the SERVQUAL model (Parasuraman et al., 1988). Finally, in Brady and Cronin’s (2001) model, nine first-order subdimensions (e.g. tangibles, waiting time, and attitude of employees) compose three second-order dimensions: interaction (relational) quality, physical environment quality, and outcome quality. Brady and Cronin’s (2001) “interaction quality” and “physical environment quality” are similar to Parasuraman et al.’s (1988) process dimensions and the functional dimension of Grönroos (1984, 2005).

Despite its popularity, the SERVQUAL instrument for quality assessment neglects to consider important factors specific to tourism destinations (e.g. attractions, entertainment, and cultural experiences) (Zabkar et al., 2010). A tourism product is a “bundle of components” including accommodations, travel, food, entertainment, and so on and research must capture the different elements of this experience in order to evaluate PQ at the destination (Zabkar et al., 2010). In their study of PQ in Slovenia, Zabkar et al. (2010) adapted the six A’s framework—“attractions,” “access,” “amenities,” “ancillary services,” “available packages,” and “activities” (see also Buhalis, 2000; Cooper et al., 1993) to operationalize a PQ construct comprising nine destination attributes (i.e. accessibility, cleanliness, diversity, local people, rest, safety and security, nature, and local cuisine). The authors argued that the “relevant destination attributes are highly contextual and that the measurement of quality should reflect the specificity of a destination’s features” (Zabkar et al., 2010: 538).

In this study, we adapt the six A’s framework to examine Australia’s PQ and identify selected attributes specific to Australia. In particular, six destination attributes are selected: (1) natural and well-known attractions, (2) variety of tourist services and culture, (3) quality of the general tourist atmosphere, (4) entertainment and recreation, (5) general environment, and (6) accessibility (see Son and Pearce, 2005; Wang and Davidson, 2008, 2010). These dimensions are measured through 18 attribute items (observed variables). We propose and examine PQ as a formative construct where the construct dimensions form the higher level PQ construct (see Clemes et al., 2011). We conceptualize PQ as formative by following the criteria presented by Jarvis et al. (2003): “(1) the direction of causality between the construct and its indicators; (2) the interchangeability of indicators; (3) co-variation among the indicators; and (4) nomological net of the indicators.¹” PQ should be formative as a change in one of the PQ quality dimensions affects overall PQ of the destination. Moreover, dimensions that affect PQ do not necessarily correlate with one another (although they might actually covary in practice). For example, the measures of the “natural and well-known attractions” do not necessarily correlate with the measures of “accessibility” (e.g. price, communication, and infrastructure). As explained by Zabkar et al. (2010: 539), “the dimensions/attributes that jointly influence a tourist’s perception of a destination’s quality do not necessarily share a common conceptual domain and there is no reason to expect they are correlated. Further, a destination’s dimensions/attributes do not necessarily share the same set of antecedents.”

Perceived value

PV is commonly defined as the trade-off between perceived benefits and perceived costs (Lovelock, 2000). The literature provides a lack of convergence on a single definition of PV, because the PV construct can be analyzed using a unidimensional measure or a multidimensional scale (Chen and Chen, 2010; Sanchez-Fernandez and Iniesta-Bonillo 2007). The unidimensional measure stems from neoclassical economic theory and assumes that consumers are rational beings that are governed by maximizing their utility when it comes to choosing between different products/services (Chiu et al., 2005). Under this utilitarian perspective, consumers’ PV of a product/service is generally determined by the difference between performance (benefits consumers receive from the product/service) and the sacrifices they make to acquire that product/service. Along that vein, previous studies have defined sacrifices in terms of the monetary price incurred by the customer to acquire the product (Monroe, 1979, 1990). For example, Brady et al. (2005) proposed a three-item scale for PV of costs, while other researchers have used two-item measures of the value of the price paid (Petrick et al., 1999). In addition, other studies (Gutman, 1982 and Zeithaml, 1998) viewed sacrifices as involving both monetary and nonmonetary price (such as time and effort expended). Thus, the benefits consumers receive from the product are inferred in terms of the quality of the attributes of that product, whereby quality perception stems from the consumers’ cognitive evaluation of the attributes of the product to infer/derive the benefits they will obtain from it (Sanchez-Fernandez and Iniesta-Bonillo, 2007).

The multidimensional (complex) approach to measure PV has its roots in consumer behavior psychology, it brings richness and complexity to the construct, but there remains little consensus on the components of this multidimensional construct or how these components are related (Sanchez-Fernandez and Iniesta-Bonillo, 2007). A multidimensional measure of PV encompasses “both the cognitive and affective facets of a product/service, for example, using a five-dimensional construct consisting of social, emotional, functional, epistemic, and conditional responses” (Chen and Chen, 2010: 30). Alternatively, a scale proposed by Petrick and Backman (2002) introduced the SERV-PERVAL based on the following dimensions: quality, monetary price (value perceived in contrast to the price paid), nonmonetary price (value perceived in return for costs, such as time and effort expended), reputation, and emotional response. Finally, a more comprehensive scale proposed by Holbrook (1994) considers eight realms of PV: efficiency (convenience), excellence (quality), politics (success), esteem (reputation), play (fun), aesthetics (beauty), morality (virtue), and spirituality (faith or ecstasy). Given that the main focus of this study is on the conceptualization of PQ, we measured PV following the unidimensional (simplified) utilitarian perspective (Boksberger and Melsen, 2011; Zeithaml, 1988) through which value is measured in terms of the value of monetary and nonmonetary (i.e. time) costs (Brady et al., 2005).

Satisfaction

Customer satisfaction is relative to the extent that a service “provides a pleasurable level of consumption-related fulfillment” (Oliver, 1997: 13). The expectation/disconfirmation paradigm (Oliver, 1980) posits that consumers develop expectations about a product/service prior to purchase, satisfaction occurs when post-purchase experiences match or exceed prepurchase expectations. Equity theory (Oliver and Swan, 1989) argues that satisfaction is based on customers receiving value relative to their investment (cost, time, and effort). Meanwhile, Norm theory (LaTour and Peat, 1979) uses a comparison standard where consumers compare their purchase with other alternative purchases. Finally, the perceived overall performance theory (Tse and Wilton, 1988) treats consumer dissatisfaction as a function of actual performance that is independent from consumer expectations.

The expectation/disconfirmation paradigm has been widely used to measure levels of tourist satisfaction (Petrick et al., 2001). The problem with this approach, however, is that it makes it difficult to interpret the link between expectations and satisfaction (Barsky, 1992). For instance, low expectations in the prepurchase stage can lead to higher probability of satisfaction, even in the case of poor performance. As a result, several studies suggest the use of a combination of items for overall satisfaction based on the expectation/disconfirmation paradigm, equity theory, norm models, and perceived overall performance (Li and Petrick, 2010; Zabkar et al., 2010). This approach is adopted in the current study.

Loyalty

Building customer loyalty is an important objective for organizations. Repeat purchases from existing customers is highly cost-effective and loyal customers can act as free advertising agents (word of mouth) (Zeithaml et al., 2009). From a tourism perspective, loyalty is a critical factor for destinations as (1) loyal visitors are less price sensitive, (2) it leads to reduced marketing costs, and (3) it ensures a constant flow of inbound travelers and revenues (Assaker and Hallak, 2012). Destination loyalty is commonly assessed in tourists’ behavioral intentions which are measured in terms of tourists’ intention to revisit the destination, their willingness to spread positive word of mouth, and intention to recommend the destination to family and friends (Zeithaml et al., 1996; Chen and Tsai, 2007). Accordingly, this study adopts these three measures of destination loyalty.

Research hypotheses

The literature emphasizes the importance of empirically examining the “antecedent, mediating, and consequent relationships” among quality, value, satisfaction, and loyalty (Rust and Oliver, 1994). Conflicting arguments have been made with regard to the causal order of the relationships—namely, which variables have mediating effects or directly impact behavioral intentions and loyalty. However, there is a consistent argument that quality leads to both satisfaction and PV (Petrick, 2002; Zeithaml, 1988), and the two are direct predictors of behavioral intentions (Petrick and Backman, 2002; Tam, 2000). Furthermore, “quality is not only moderated by perceived value and satisfaction in the prediction of loyalty and behavioral intentions, but also has a direct relationship on behavioral intentions” (Petrick, 2004: 400).

Studies examining the relationship between value and satisfaction have produced mixed findings. In some studies, satisfaction is a direct predictor of PV (Chang and Wildt, 1994; Petrick and Backman, 2002), whereas others have found the reverse, arguing that it is PV, per se, which drives satisfaction (Cronin et al., 2000; Oh, 1999; Tam, 2000). The argument stands that “quality and perceived value are cognitive responses to a service experience, while satisfaction is an affective response” (Petrick, 2004: 397). Based on theory, cognitive responses intuitively precede emotional responses (Bagozzi, 1992), and thus, PV (cognitive) should be driving feelings of satisfaction (emotional). The extensive literature conducted on value, quality, satisfaction, and loyalty form the bases for the following hypotheses:

H1: PQ has a direct positive effect on PV.

H2: PQ has a direct positive effect on satisfaction.

H3: PQ has a direct positive effect on loyalty.

H4: PV has a direct positive effect on satisfaction.

H5: PV has a direct positive effect on loyalty.

H6: Satisfaction has a direct positive effect on loyalty.

Research methodology

Research design and data collection

Data for this research were collected as part of a larger study on Australia’s inbound tourism (see Assaker, 2014). An online questionnaire was administered to residents in the United Kingdom and the United States from June to August 2012. These two countries are major market segments of Australia’s inbound visitors (Tourism Research Australia, 2013). The two countries are also representative of the English-speaking visitors and are combined in our sample. Data was deliberately collected from the origin countries to ensure the reliability and credibility of data across countries (Echtner and Ritchie, 1993). The sample was drawn from established panels of individual’s representative of the population in each country.

Research instrument and construct measures

The questionnaire sought information about individuals’ demographics, PQ of Australia’s destination attributes, PV, level of satisfaction with the experience, and loyalty.

PQ of Australia as perceived by the visitors was operationalized as a higher order formative construct adapted from a validated scale from Wang and Davidson (2010). The PQ scale comprises six latent factors: (1) natural and well-known attractions, (2) variety of tourists’ services and culture, (3) quality of general tourists’ atmosphere, (4) entertainment and recreation, (5) general environment, and (6) accessibility. The six factors (dimensions) were measured using 18 attribute items (observed variables). Participants were asked to respond to a list of statements based on how they perceived the various destination’s attributes using a five-point Likert-type scale (1 = much lower than my expected quality level, 5 = much higher than my expected quality level; see Appendix Table 1A).

PV was measured in terms of value for monetary and nonmonetary (time) costs based on the three-item scale proposed by Brady et al. (2005). The three-item scale included: “I think it was worthwhile to go to Australia”; “Australia is very good value for the money”; and “I consider the time spent in Australia to be a good investment” (1 = strongly disagree, 7 = strongly agree). Unidimensional operationalization of PV (also referred to as the “simplified” approach, see Sanchez-Fernandez and Iniesta-Bonillo, 2007) is the widely adopted method in both tourism and other industries including retail (He et al., 2012) and public transport (Lai and Chen, 2011).

Satisfaction with the visit to Australia was measured using four observed variables: (1) “How satisfied were you with your overall experience?” (2) “Based on the expectations you had of Australia prior to your visit, how satisfied were you with your stay in Australia?” (3) “Based on all of your previous travel experiences, how satisfied were you with your stay in Australia?” and (4) “Based on what you spent in terms of price, time, and effort, how satisfied were you with your stay in Australia?” (1 = not at all satisfied, 7 = very satisfied). The four items were adapted from previously validated scales (Cronin et al., 2000; Gallarza and Saura, 2006; Petrick and Backman, 2002). In particular, item 1 of the satisfaction scale aligns with the overall perceived performance theory (Tse and Wilton, 1988). Item 2 aligns with the expectation/disconfirmation theory (Oliver, 1980), which stipulates that consumers develop expectations about a product before purchasing it and compare actual performance following the purchase. Item 3 aligns with the norm theory (LaTour and Peat, 1979) and the comparison standard in that consumers compare a purchase to previous purchases. Finally, item 4 aligns with the equity theory (Oliver and Swan, 1989), which stipulates that satisfaction occurs when customers receive more value than from what they actually spent in terms of price, time, and effort.

Loyalty was measured using three items adapted from Zeithaml et al. (1996): (1) How likely are you to visit Australia again in the future, (2) How likely are you to recommend Australia to friends and relatives, and (3) How likely are you to say good things about Australia (1 = very unlikely, 7 = very likely).

Data analysis

Data collected from the survey was analyzed using descriptive statistics, exploratory factor analysis and partial least squares-SEM (PLS-SEM) analysis through XL-STAT/PLSPM. PLS was adopted due to the higher order formative conceptualization for PQ (see Hulland et al. 2010). In line with the two-step process for PLS-SEM analysis (Chin, 1998), we first examined the outer model by testing the reliability, convergent, and discriminant validity for the reflective factors as well as examining the content validity for the higher order formative PQ. Following the outer model analysis, we then examine the inner (structural) model to assess the hypothesized relationships among the constructs.

Results

The 1625 surveys delivered resulted in 500 responses (250 from each country, 31% response rate). Of these responses, 249 had visited Australia in the past 5 years, including 133 from the United Kingdom (53.4%) and 116 from the United States (46.6%). As such, only data from the 249 respondents who had visited Australia were used to test the proposed model in the present study. The sample consisted of 122 (51%) females and 127 (51%) males. The mean age of participants was 40 years and 42% of respondents identified themselves as “professionals/managers” (42%), followed by skilled and unskilled workers (15%), retirees (10%), and students (9%).

The subject-to-item ratio (249/28) = 9.2:1 is sufficient to for our analysis purposes and is within the range needed to achieve sample size adequacy at the 95% confidence level (Hatcher, 1994). Missing values on variables from the data set were imputed through the nearest neighborhood approach (Olinsky et al., 2003). Finally, no further checks on the normality of the data were needed as PLS-SEM does not make assumptions about the distribution properties of the data set (see do Valle and Assaker, 2016).

Principal component analysis and reliability tests

Results of the principal component analysis on the unstandardized data of the nine blocks of variables (i.e. the dimensions PQ, PV, satisfaction, and loyalty) found that all loadings for the nine constructs were >0.7, eigenvalue >1, supporting unidimensionality (Table 3). Furthermore, Cronbach’s α and Dillon–Goldstein’s ρ were all > 0.6 indicating good scale reliability and further supporting the unidimensionality and reflective scheme of these factors (Nunnally and Bernstein, 1994) (Table 3). As the higher order PQ was specified as a formative construct, traditional tests of dimensionality and reliability were inappropriate; we thus proceeded with two-step PLS-SEM analysis (see Chin, 1998) to (1) confirm how these indicators load on their underlying construct at the population level and (2) examine the causal relationship among the constructs as hypothesized in our model (Figure 1).

Table 3.

Factor matrix, Cronbach’s α, composite reliability, and eigenvalues by variable blocks with component analysis extraction method.

Constructs	Variables	Factor 1	Cronbach’s α	Dillon Gold ρ	Critical value	Eigenvalues
Natural and well-known attractions	A1	0.80	0.82	0.88	1.00	2.59
	A13	0.79				0.50
	A14	0.82				0.47
	A16	0.81				0.44
Variety of tourists services and culture	A7	0.83	0.75	0.86	1	2.00
	A8	0.84				0.56
	A9	0.78				0.44
Quality of general tourists atmosphere	A2	0.81	0.74	0.85	1	1.98
	A4	0.82				0.51
	A15	0.81				0.50
Entertainment and recreation	A5	0.83	0.75	0.86	1	2.01
	A6	0.82				0.52
	A12	0.81				0.47
General environment	A10	0.87	0.77	0.87	1	1.52
	A11	0.87				0.48
Accessibility	A3	0.78	0.69	0.81	1	1.78
	A17	0.71				0.74
	A18	0.84				0.48
PV	PV1	0.88	0.86	0.91		2.34
	PV2	0.86				0.38
	PV3	0.90				0.28
Satisfaction	SAT1	0.90	0.92	0.95		3.25
	SAT2	0.92				0.33
	SAT3	0.91				0.23
	SAT4	0.88				0.19
Loyalty	LOY1	0.87	0.82	0.89		2.21
	LOY2	0.87				0.44
	LOY3	0.83				0.35

PV: perceived value.

PLS-SEM analysis

We used PLS-SEM on the whole data set of the unstandardized data. As discussed earlier, PQ is operationalized as a “reflective first-order, formative second-order” construct. Mode A (reflective scheme) was specified for the six first-order constructs of PQ, PV, satisfaction, and loyalty, whereas Mode B (formative scheme) was specified for the higher (second)-order PQ (Fornell and Bookstein, 1982). We estimated the weights of the inner model following the centroid approach of Vinzi and Russolillo (2013).

Outer model analysis

The convergent validity of the first-order constructs of PQ dimensions, PV, satisfaction, and loyalty was supported as all factor loadings >0.7 with the bootstrap test supporting significance (the 95% confidence interval does not go through zero; Table 4). In addition, the average variance extracted (AVE), “which measures the variance in the indicators due to the construct relative to the amount due to the measurement error” (Huang et al., 2015: 352) for all six PQ dimensions, was >0.5. This was also the case for value (AVE = 0.78), satisfaction (AVE = 0.812), and loyalty (AVE = 0.737). Discriminant validity is supported “when the average shared variance of a construct and its indicators exceed the shared variance with every other construct in the model” (Fornell and Larcker, 1981: 435). This was the case in the model outlined in the present study (Table 5), in which the AVE for each of the first-order constructs of PQ, PV, satisfaction, and loyalty (which in this case measure the average shared variance of a construct and its indicators) was greater than the squared correlation coefficient of that construct with every other construct of the model (which in this case measures the shared variance of each construct with every other construct in the model).

Table 4.

Outer model: first-order latent variables with reflective indicators and formative higher order PQ.

Latent variable	Manifest variables label	Standardized loadings	Standardized loadings (bootstrap)	CR	Lower bound (95%)	Upper bound (95%)	AVE
Natural and well-known attractions	A1	0.818	0.818	20.956	0.716	0.878	0.675
	A13	0.807	0.802	28.895	0.750	0.857
	A14	0.819	0.817	29.332	0.749	0.876
	A16	0.843	0.841	30.183	0.768	0.886
Variety of tourists services and culture	A7	0.816	0.809	23.451	0.715	0.871	0.694
	A8	0.865	0.862	42.187	0.800	0.899
	A9	0.817	0.813	26.227	0.722	0.865
Quality of general tourists atmosphere	A2	0.849	0.843	31.466	0.764	0.894	0.709
	A4	0.823	0.822	29.388	0.749	0.865
	A15	0.854	0.853	37.751	0.804	0.895
Entertainment and recreation	A5	0.820	0.818	27.323	0.748	0.876	0.705
	A6	0.829	0.826	29.462	0.750	0.871
	A12	0.869	0.872	40.990	0.828	0.915
General environment	A10	0.882	0.878	34.989	0.811	0.920	0.796
	A11	0.902	0.900	50.962	0.856	0.924
Accessibility	A3	0.859	0.860	44.093	0.814	0.893	0.658
	A17	0.727	0.733	15.229	0.623	0.824
	A18	0.842	0.840	30.909	0.756	0.892
PV	PV1	0.885	0.882	45.057	0.825	0.922	0.780
	PV2	0.867	0.865	38.498	0.811	0.913
	PV3	0.897	0.896	46.479	0.847	0.930
Satisfaction	SAT1	0.900	0.895	45.129	0.858	0.930	0.812
	SAT2	0.916	0.915	56.483	0.876	0.939
	SAT3	0.906	0.906	48.463	0.859	0.939
	SAT4	0.882	0.881	44.728	0.830	0.912
Loyalty	LOY1	0.863	0.858	27.140	0.766	0.914	0.737
	LOY2	0.875	0.878	36.159	0.832	0.927
	LOY3	0.837	0.843	25.911	0.754	0.905
PQ	Natural and well-known attractions	0.214	0.195	6.231	0.139	0.274	–
	Variety of tourists services and culture	0.176	0.154	5.227	0.098	0.220
	Quality of general tourists atmosphere	0.147	0.136	3.996	0.076	0.217
	Entertainment and recreation	0.176	0.149	5.022	0.079	0.199
	General environment	0.182	0.142	6.156	0.094	0.193
	Accessibility	0.270	0.238	7.947	0.164	0.297

PQ: perceived quality; PV: perceived value; AVE: average variance extracted; CR: critical ratio.

Table 5.

Discriminant validity: first-order latent variables with reflective indicators (squared correlations for any pair of latent variables < AVE).

	Natural and well-known attractions	Variety of tourist services	Quality of general tourists atmosphere	Entertainment and recreation	General environment	Accessibility	PV	Satisfaction	Loyalty	Mean communalities (AVE)
Natural and well-known attractions	1	0.538	0.515	0.634	0.539	0.323	0.405	0.293	0.288	0.675
Variety of tourist services	0.538	1	0.561	0.527	0.417	0.434	0.428	0.302	0.230	0.694
Quality of general tourists atmosphere	0.515	0.561	1	0.545	0.456	0.495	0.411	0.280	0.256	0.709
Entertainment and recreation	0.634	0.527	0.545	1	0.463	0.409	0.403	0.333	0.246	0.705
General environment	0.539	0.417	0.456	0.463	1	0.302	0.308	0.274	0.253	0.796
Accessibility	0.323	0.434	0.495	0.409	0.302	1	0.416	0.365	0.272	0.658
PV	0.405	0.428	0.411	0.403	0.308	0.416	1	0.445	0.330	0.780
Satisfaction	0.293	0.302	0.280	0.333	0.274	0.365	0.445	1	0.628	0.812
Loyalty	0.288	0.230	0.256	0.246	0.253	0.272	0.330	0.628	1	0.737
Mean communalities (AVE)	0.675	0.694	0.709	0.705	0.796	0.658	0.780	0.812	0.737	0

PV: perceived value; AVE: average variance extracted; All values in italics signify the .05 significance level.

The content validity of the higher order PQ construct was assessed at both the individual and construct levels. The results of the bootstrap tests for the individual level showed that loadings of the six first-order dimensions of PQ are all significant, where the bootstrap-based empirical 95% confidence interval did not include zero (Table 4). Moreover, the variance inflation factor values for the six PQ dimensions were all <2.0, demonstrating that these dimensions are not highly correlated with one another. Furthermore, the PQ: R ² = 0.994 confirms that 99.4% of the variances in PQ could be explained by its determined first-order dimensions (Figure 2). These results empirically support the operationalization of PQ as a first-order reflective, higher order formative construct.

Figure 2.

Results of hierarchical model (standardized estimates for inner model).

Inner model analysis

The path coefficients among the constructs were examined using bootstrapping with 1000 iterations of resampling (Davison and Hinkley, 1997). Figure 2 depicts the results of the inner model with the results of the bootstrapping, indicating that five of the six hypotheses were supported. In particular, PQ had a significant positive effect on PV, satisfaction, and loyalty (reg. coeffs. std. = 0.746, 0.373, and 0.142, respectively), thereby supporting hypotheses 1–3. Value also had a significant positive effect on satisfaction (reg. coeff. std. = 0.389), thereby supporting hypothesis 4. Satisfaction had a significant positive effect on loyalty (reg. coeff. std. = 0.695), thereby supporting hypothesis 6. However, PV was found to have no significant effect on loyalty (reg. coeff. std. = 0.005), leaving hypothesis 5 unsupported.

Table 6 reports the direct, indirect, and total effects of PQ, value, and satisfaction variables on loyalty. PQ and satisfaction had the highest effects on loyalty, followed by value (total effect = 0.603, 0.695, and 0.275, respectively). The R ² of the model also demonstrated that PQ, value, and satisfaction explain almost 63% of the variance in loyalty, supporting the homological validity of the model (Chin, 1998). Further assessment of the structural model involved computing the Stone–Geisser Q2 values (referred to as cross-validated redundancy measures; see Jöreskog and Wold, 1982). These values are used to “measure predictive relevance in terms of the indicators, not just the constructs, for each of the endogenous constructs in the model” (do Valle and Assaker, 2016: 9). The Stone–Geisser Q2 values for the PV, satisfaction, and loyalty variable indicators, computed using blindfolding procedures, were found to be larger than zero, supporting predictive relevance in explaining the endogenous latent variables (Vinzi and Russolillo, 2013).

Table 6.

Standardized direct, indirect, and total effects for inner model from PLS-SEM.

Path	Direct effect	Indirect effect	Total effects
PQ → satisfaction	0.373	0.290	0.663
PQ → loyalty	0.142	0.461	0.603
PV → satisfaction	0.389	0	0.389
PV → loyalty	0.005n.s.	0.270	0.275
Satisfaction → loyalty	0.695	0	0.695

PLS-SEM: partial least squares–structural equation modeling; PQ: perceived quality; PV: perceived value; n.s.: non-significant effects at the 0.05 confidence level.

Conclusions, discussion, and limitations

Drawing on the conflicting research on the relationships among PQ, satisfaction, value, and loyalty in tourism, the research empirically examined a comprehensive higher order model in the context of Australia as a tourism destination. PQ of Australia as a tourism destination is developed through six destination dimensions (latent first-order constructs). These dimensions are (1) natural and well-known attractions, (2) variety of tourist services and culture, (3) quality of general tourism atmosphere, (4) entertainment and recreation, (5) general environment, and (6) accessibility. The dimensions were measured using 18 attribute items (observed variables) as identified in the previous literature (Son and Pearce, 2005; Wang and Davidson, 2008, 2010). The model was analyzed on data collected from 249 UK and US residents who had previously visited Australia. Five of the six hypotheses were supported, presenting new empirical evidence on the complexity of the structural relationships among quality, value, satisfaction, and loyalty.

Specifically, our results make an important contribution to existing knowledge on PQ at the destination level, given that previous research has focused only on reflective approaches, with little significance being attached to destination-specific attributes. In particular, our findings show support for a higher order PQ model, with all six dimensions contributing significantly in forming the higher order construct of PQ and explaining 99.4% of its variance. This suggests that travelers form their overall evaluation of quality experience at a destination based on how they evaluate the different dimensions and attributes of that destination. These results support the higher order formative scheme for conceptualizing destination PQ, as suggested in the service marketing literature (e.g. Brady and Cronin, 2001; Cronin et al., 2000; Rust and Oliver, 1994). In particular, the “accessibility” of the destination (including items on costs/price levels, accessibility, and tourist information/communication) and “natural and well-known attractions” (including items on scenery/natural attractions, tourist sites/activities, beaches, and wildlife) were found to have the strongest influence on the higher order PQ construct. We recommend future studies in tourism to think carefully about the operationalization of PQ. Our results provide strong support for a “reflective first-order, formative second-order” model (also referred to as a molar, or type II, higher order model, see Diamantopoulos et al., 2008), where the six reflective first-order dimensions of PQ form (arrows pointing toward) the higher (second)-order PQ construct (see Hair et al., 2014). As argued previously, the accurate operationalization of the measurement construct is critical for the validity of the overall structural model.

Second, by providing a more (theoretically and empirically) accurate approach in terms of the conceptualization of the PQ construct compared to previous tourism studies (e.g. Chen and Chen, 2010; Zakbar et al., 2010), the hierarchical model tested in the present study overcomes possible measurement errors in the operationalization of PQ. Such measurement errors usually lead to bias in the relationships among the constructs in the structural model (see Edwards, 2001; Zabkar et al., 2010). Thus, our hierarchical model allows for a better investigation of the relationships among constructs in light of diverging results (e.g. relationships among PQ, PV, satisfaction, and loyalty) from previous studies that have assumed either reflective or multifactor quality models (Chen and Tsai, 2007; Chen, 2008). Most importantly, when operationalized as a higher order construct, PQ was shown to have a direct positive effect on loyalty; tourists’ perceptions of the quality of destination dimensions and attributes are strong and significant predictors of behavioral loyalty (Bigné et al., 2001; Zakbar et al., 2010).

This study’s findings corroborate with those of Cronin et al. (2000), who argue that visitor satisfaction alone is insufficient to predict behavioral responses; one also must examine the effects of quality and value on loyalty. The results from this study show that PQ and satisfaction have a stronger influence on loyalty (total direct and indirect effects of 0.603 and 0.695, respectively) compared to PV which only has an indirect effect of 0.270 on loyalty through satisfaction. In particular, the low effect of value on satisfaction (total direct effect = 0.389) and loyalty (indirect effect = 0.270), compared to quality (total effects = 0.663 and 0.603 on satisfaction and loyalty, respectively), indicates that satisfaction and loyalty are largely defined by perceptions of quality.

The validated relationships among PQ, value, satisfaction, and loyalty in the present study provide a comprehensive framework for understanding destination loyalty and tourists’ behavioral intentions. Destination managers need to pay close attention to the six dimensions (accessibility, natural and well-known attraction, general environment, variety of tourist services, entertainment and recreation, and quality of general tourism atmosphere—listed in order of importance) which affect the PQ of destination’s offerings. Identifying and ranking which destination has the most significant impact on overall quality perceptions can allow decision makers to effectively concentrate their efforts on creating differentiated and enhanced offers along these attributes.

The higher order scheme for the PQ construct validated in our study enables destination managers to measure service quality on three levels—overall quality, destination attributes, and individual attribute level items. For example, using the first-order quality dimensions enables managers to identify major operational areas such as accessibility or well-known attractions that they need to act upon in order to enhance PQ of the destinations. Detailed diagnostics at the individual attribute level can further help destination managers identify the relative influence of attribute items on a specific dimension. For example, the strongest drivers of accessibility are cost/price levels (reg. coeffs. std. = 0.859), accessibility (reg. coeffs. std. = 0.842), and tourist information/communication (reg. coeffs. std. = 0.727). In addition, as seen in our model, PQ has a direct and indirect impact with behavioral intentions (word of mouth, repurchase (revisit) intentions, and recommendations), indicating that a comprehensive managerial approach is needed to maintain and enhance destination loyalty.

In terms of the limitations of these findings, we acknowledge that conceptualizing PQ as a formative construct may limit the generalizability of the findings across various destinations. Data was collected for the specific case of Australia as a tourism destination, and future research should examine whether the list of attributes identified also hold true in other destinations. In addition, another limitation of the present article is the simplified (unidimensional) utilitarian conceptualization of PV; future research could make use of a more complex multidimensional scale for PV that incorporates both cognitive and affective/hedonic facets of PV (see Sanchez-Fernandez and Iniesta-Bonillo, 2007). This may affect the model relationships especially in terms of a more significant effect of a comprehensive (complex) PV construct on the variables of satisfaction and loyalty. Moreover, the sample used for this study included people who visited Australia between 2008 and 2012. The 5-year time period was needed to allow for a substantial sample frame and response rate, especially to serve the purpose of our structural modeling analysis. We acknowledge the limitations of a 5-year period with regard to an individual’s ability to recall travel experiences. Despite these limitations, the validated relationships among higher order schemes for the PQ construct, value, satisfaction, and loyalty in the present study provide a comprehensive framework for understanding destination loyalty and tourists’ behavioral intentions. Future research should incorporate other antecedents of tourists’ behavioral intentions, such as destinations’ image and its relationship with other service evaluation constructs. Such additional constructs might add more refinement to the conceptual framework and extend the results tested in the current model.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Funding for this research came through a grants scheme from the School of Management, University of South Australia.

Note

Appendix 1

Table 1A.

Measurement items.

Constructs	Variable labels	Measurement items	Scale	Mean	Std. deviation	Skewness	Kurtosis
Natural and well-known attractions	A1	Australia has spectacular scenery and natural attractions.	Each item is measured on a five-point Likert-type scale with 1 = much lower than my expected quality level to 5 = much higher than my expected quality level	4.350	0.870	−1.675	3.272
	A13	Australia is a country with many well-known tourists sites.		4.365	0.706	−1.071	1.694
	A14	Australia has magnificent sunny beaches.		4.494	0.713	−1.590	3.151
	A16	Australia has fascinating native animals and vegetation.		4.386	0.770	−1.214	1.404
Variety of tourists services and culture	A7	Australia offers a food variety of souvenirs and duty-free goods for travelers.		4.145	0.830	−0.660	−0.074
	A8	Australia has wonderful historical sites and excellent museums/art galleries.		4.141	0.903	−0.978	0.736
	A9	Australia has a unique aboriginal culture.		4.333	0.796	−1.300	2.059
Quality of general tourists atmosphere	A2	Australia service staff are qualified, helpful and friendly.		4.169	0.811	−0.959	0.972
	A4	Australia is a safe destination for travelers.		4.225	0.705	−0.626	0.210
	A15	The environment in Australia is very clean.		4.373	0.714	−1.158	1.952
Entertainment and recreation	A5	Australia has a variety of entertainment/night life activities for travelers.		4.233	0.784	−0.942	0.937
	A6	Australia offers many opportunities for sports and adventurous activities.		4.265	0.799	−0.944	0.662
	A12	Australia has good tourism infrastructure facilities, for example, restaurants, accommodations, and so on.		4.309	0.770	−1.182	1.991
General environment	A10	Australia climate is good.		4.289	0.786	−1.615	4.181
	A11	Australia is a good place for rest and relaxation.		4.373	0.736	−1.392	3.102
Accessibility	A3	Australia is a value for money destination.		3.924	0.937	−0.708	0.247
	A17	Communication is not a serious problem for non-English speaking tourists.		4.120	0.872	−0.679	−0.196
	A18	Australia is easy to access.		4.056	0.957	−0.836	0.036
Perceived value	PV1	I think it was worthwhile going to Australia.	Each item is measured on a seven-point Likert-type scale with 1 = strongly disagree to 7 = strongly agree	6.229	1.191	−2.312	6.502
	PV2	Australia is very good value for the money.		5.486	1.409	−0.918	0.385
	PV3	I consider the time I spent in Australia to be a good investment.		5.884	1.184	−1.374	2.618
Satisfaction	SAT1	How satisfied were you with your overall experience?	Each item is measured on a seven-point Likert-type scale with 1 = not at all satisfied to 7 = very satisfied	6.229	1.191	−2.312	6.502
	SAT2	Based on the expectations you had of Australia prior to your visit, how satisfied were you with your stay in Australia?		6.064	1.176	−1.700	3.431
	SAT3	Based on all of your previous travel experiences, how satisfied were you with your stay in Australia?		5.486	1.409	−0.918	0.385
	SAT4	Based on what you spent in terms of price, time, and effort, how satisfied were you with your stay in Australia?		5.884	1.184	−1.374	2.618
Loyalty	LOY1	How likely are you to visit Australia again in the future?	Each item is measured on a seven-point Likert-type scale with 1 = very unlikely to 7 = very likely	5.969	1.403	−1.605	2.342
	LOY2	How likely are you to recommend Australia to friends and relatives?		6.263	1.157	−1.983	4.652
	LOY3	How likely are you to say good things about Australia?		6.413	0.954	−2.231	6.914

References

Anderson

Fornell

Rust

(1997) Customer satisfaction, productivity, and profitability: differences between goods and services. Marketing Science 16(2): 129–145.

Assaker

(2014) Examining a hierarchical model of Australia’s destination image. Journal of Vacation Marketing 20(3): 195–210.

Assaker

Hallak

(2012) Examining a supply-side predictive model in tourism using partial least squares path modeling: an empirical analysis at the country aggregate level. Tourism Analysis 17(5): 587–599.

Bagozzi

(1992) The self-regulation of attitudes, intentions, and behavior. Social Psychology Quarterly Special Issue: Theoretical Advances in Social Psychology 55(2): 178–204.

Baker

Crompton

(2000) Quality, satisfaction and behavioral intentions. Annals of Tourism Research 27(3): 785–804.

Barsky

(1992) Customer satisfaction in the hotel industry: meaning and measurement. Hospitality Research Journal 16(1): 51–73.

Bigné

Sanchez

(2001) Tourism image, evaluation variables and after purchase behaviour: inter-relationship. Tourism Management 22: 607–616.

Boksberger

Melsen

(2011) Perceived value: a critical examination of definitions, concepts and measures for the service industry. Journal of Services Marketing 25(3): 229–240.

Brady

Cronin

Jr (2001) Some new thoughts on conceptualizing perceived service quality: a hierarchical approach. Journal of Marketing 65(3): 34–49.

10.

Brady

Knight

Cronin

. (2005) Removing the contextual lens: a multinational, multi-setting comparison of service evaluation models. Journal of Retailing 81(3): 215–230.

11.

Buhalis

(2000) Marketing the competitive destination in the future. Tourism Management 21(1): 97–116.

12.

Chang

Wildt

(1994) Price, product information, and purchase intention: an empirical study. Journal of the Academy of Marketing Science 22(1): 16–27.

13.

Chen

(2008) Investigating structural relationships between service quality, perceived value, satisfaction, and behavioral intentions for air passengers: evidence from Taiwan. Transportation Research Part A 42(4): 709–717.

14.

Chen

(2010) Experience quality, perceived value, satisfaction and behavioral intentions for heritage tourists. Tourism Management 31: 29–35.

15.

Chen

Tsai

(2007) How destination image and evaluative factors affect behavioral intentions? Tourism Management 28: 1115–1122.

16.

Chin

(1998) The partial least squares approach for structural equation modeling. In: Marcoulides

(ed.) Modern Methods for Business Research. London: Lawrence Erlbaum Associates, pp. 236–295.

17.

Chiu

Hsieh

. (2005) Relationship marketing and consumer switching behavior. Journal of Business Research 58(1 2): 1681–1689.

18.

Clemes

Brush

Collins

(2011) Analysing the professional sport experience: a hierarchical approach. Sport Management Review 14(4): 370–388.

19.

Cooper

Fletcher

Gilbert

. (1993) In Tourism: Principles and Practice. Harlow: Longman Scientific & Technical.

20.

Cronin

Brady

Hult

GTM

(2000) Assessing the effects of quality, value and customer satisfaction on consumer behavioral intentions in service environments. Journal of Retailing 76(2): 193–218.

21.

Dabholkar

Shepherd

Thorpe

(2000) A comprehensive framework for service quality: an investigation of critical conceptual and measurement issues through a longitudinal study. Journal of Retailing 76(2): 139–173.

22.

Davison

Hinkley

(1997) Bootstrap Methods and their Application. Cambridge: University Press.

23.

Diamantopoulos

Riefler

Roth

(2008) Advancing formative measurement models. Journal of Business Research 61(1 2): 1203–1218.

24.

Do Valle

Assaker

(2016) Using partial least squares structural equation modeling in tourism research a review of past research and recommendations for future applications. Journal of Travel Research 55(6): 695–708.

25.

Echtner

Ritchie

JRB

(1993) The measurement of destinations image: an empirical assessment. Journal of Travel Research 31(4): 3–13.

26.

Edwards

(2001) Multidimensional constructs in organizational behavior research: an integrative analytical framework. Organ Res Methods 4(2): 144–192.

27.

Fornell

Larcker

(1981) Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research 18: 39–50.

28.

Fornell

Bookstein

(1982) Two structural equation models: LISREL and PLS applied to consumer exit-voice theory. Journal of Marketing Research 19: 440–452.

29.

Fornell

Johnson

Anderson

. (1996) The American customer satisfaction index: nature, purpose and findings. Journal of Marketing 60(4): 7–18.

30.

Gallarza

Saura

(2006) Value dimensions, perceived value, satisfaction and loyalty: an investigation of university students’ travel. Tourism Management 27(3): 437–452.

31.

Grönroos

(1984) A service quality model and its marketing implications. European Journal of Marketing 18(4): 36–44.

32.

Grönroos

(2005) What can a service logic offer marketing theory?. In: Lusch

Vargo

(eds). The Service-Dominant Logic of Marketing: Dialog, Debate, and Directions. Armonk, NY: M.E. Sharpe, pp. 354–364.

33.

Gutman

(1982) A means-end chain model based on consumer categorization processes. Journal of Marketing 46(2): 60–72.

34.

Hair

JFJ

Hult

GTM

Ringle

. (2014) A Primer on Partial Least Squares Structural Equation Modelling (PLS-SEM). Thousand Oaks, CA: SAGE.

35.

Hatcher

(1994) A Step-by-Step Approach to Using the SAS^® System for Factor Analysis and Structural Equation Modeling. Cary: SAS Institute, Inc.

36.

Harris

(2012) Social identity perspective on brand loyalty. Journal of Business Research 6: 648–657.

37.

Holbrook

(1994) The nature of customer value: an axiology of services in the consumption experience. In: Rust

Oliver

(eds). Service quality: new directions in theory and practice. Thousand Oaks, CA: SAGE, pp. 21–71.

38.

Kandampully

Juwaheer

(2009) Relationships and impacts of service quality, perceived value, customer satisfaction, and image: an empirical study. Service Industries Journal 29(2): 111–125.

39.

Huang

Weiler

Assaker

(2015) Effects of interpretive guiding outcomes on tourist satisfaction and behavioral intention. Journal of Travel Research 54(3): 344–358.

40.

Hulland

Rayner

(2010) Modeling customer satisfaction: a comparative performance evaluation of covariance structure analysis versus partial least squares. In: Esposito Vinzi

Chin

Henseler

Wang

(eds) Handbook of Partial Least Squares: Concepts, Methods and Application. Germany: Springer, pp. 307–325.

41.

Hutchinson

Fujun

Wang

(2009) Understanding the relationships of quality, value, equity, satisfaction, and behavioral intentions among golf travelers. Tourism Management 30: 298–308.

42.

Jarvis

MacKenzie

Podsakoff

(2003) A critical review of construct indicators and measurement model misspecification in marketing and consumer research. Journal of Consumer Research 30: 199–218.

43.

Lee

Reisinger

(2014) Behavioral intentions of international visitors to the Korean hanok guest houses: quality, value and satisfaction. Annals of Tourism Research 47: 83–86.

44.

Jöreskog

Wold

(1982) The ML and PLS techniques for modeling with latent variables: historical and comparative aspects. In: Jöreskog

Wold

(eds) Systems Under Indirect Observation. Amsterdam: North-Holland, Vol. Part 1, pp. 263–270.

45.

Lai

Chen

(2011) Behavioral intentions of public transit passengers—the roles of service quality, perceived value, satisfaction and involvement. Transport Policy 18: 318–325.

46.

LaTour

Peat

(eds) (1979) Conceptual and Methodological Issues in Consumer Satisfaction Research. Bloomington: Indiana University Press.

47.

Little

Cunningham

Shahar

. (2002). To parcel or not to parcel: Exploring the question, weighing the merits. Structural Equation Modeling 9: 151–173.

48.

Petrick

(2010) Revisiting the commitment-loyalty distinction in a cruising context. Journal of Leisure Research 42(1): 67–90.

49.

Lovelock

(2000) Service Marketing, 4th ed. New Jersey: Prentice Hall.

50.

Luo

(2009) Quantifying the long-term impact of negative word of mouth on cash flows and stock Prices. Marketing Science 28(1): 148–165.

51.

Monroe

(1979) Pricing: Making Profitable Decisions. New York: McGraw-Hill.

52.

Monroe

(1990) Pricing: Making Profitable Decisions. New York: McGraw-Hill.

53.

Murphy

Pritchard

Smith

(2000) The destination product and its impact on traveller perceptions. Tourism Management 21(1): 43–52.

54.

Nunnally

Bernstein

(1994) Psychometric Theory. New York: McGraw-Hill.

55.

(1999) Service quality, customer satisfaction, and customer value: a holistic perspective. Hospitality Management 18: 67–82.

56.

Olinsky

Chen

Harlow

(2003) The comparative efficacy of imputation methods for missing data in structural equation modeling. European Journal of Operational Research 151(1): 53–79.

57.

Oliver

(1980) A cognitive model of the antecedents and consequences of satisfaction decisions. Journal of Marketing Research 17: 46–49.

58.

Oliver

(1997) Satisfaction: A Behavioral Perspective of the Consumer. New York: Irvin/McGraw-Hill.

59.

Oliver

Swan

(1989) Consumer perceptions of interpersonal equity and satisfaction in transactions: a field survey approach. Journal of Marketing 53: 21–35.

60.

Parasuraman

Zeithaml

Berry

(1988) SERVQUAL: a multiple-item scale for measuring consumer perceptions of service quality. Journal of Retailing 64(Spring): 12–40.

61.

Petrick

Backman

Bixler

(1999) An investigation of selected factors’ impact on golfer satisfaction and perceived value. Journal of Park and Recreation Administration 17(1): 40–59.

62.

Petrick

Morais

Norman

(2001) An examination of the determinants of entertainment vacationers’ intentions to revisit. Journal of Travel Research 40(1): 41–48.

63.

Petrick

(2002) Development of a multi-dimensional scale for measuring the perceived value of a service. Journal of Leisure Research 34(2): 119–134.

64.

Petrick

Backman

(2002) An examination of the construct of perceived value for the prediction of golf travelers’ intentions to revisit. Journal of Travel Research 41(1): 38–45.

65.

Petrick

(2004) The roles of quality, value and satisfaction in predicting cruise passengers’ behavioral intentions. Journal of Travel Research 42(4): 397–407.

66.

Ranjbarian

Pool

(2015) The impact of perceived quality and value on tourists’ satisfaction and intention to revisit Nowshahr city of Iran. Journal of Quality Assurance in Hospitality & Tourism 16(1): 103–117.

67.

Rust

Oliver

(1994) Service Quality: New Directions in Theory and Practice. Thousand Oaks, CA: SAGE.

68.

Sanchez-Fernandez

Iniesta-Bonillo

(2007) The concept of perceived value: a systematic review of the research. Marketing Theory 7(4): 427–451.

69.

Son

Pearce

(2005) Multi-faceted image assessment: international students’ views of Australia as a tourist destination. Journal of Travel & Tourism Marketing 18(4): 21–35.

70.

Tam

JLM

(2000) The effects of service quality, perceived value and customer satisfaction on behavioral intentions. Journal of Hospitality and Leisure Marketing 6(4): 31–43.

71.

Tourism Research Australia (2013) International Visitors to Australia 2013. Available at: http://www.tra.gov.au/ (accessed 5 October 2015)

72.

Tse

Wilton

(1988) Models of consumer satisfaction: an extension. Journal of Marketing Research 25: 204–212.

73.

Vinzi

Russolillo

(2013) Partial least squares algorithms and methods. Wiley Interdisciplinary Reviews: Computational Statistics 5(1): 1–19.

74.

Wang

Davidson

MCG

(2008) Chinese student travel market to Australia: an exploratory assessment of destination perceptions. International Journal of Hospitality and Tourism Administration 9(4): 405–426.

75.

Wang

Davidson

MCG

, (2010), ‘Pre- and post-trip perceptions: An insight into Chinese package holiday market to Australia’, Journal of Vacation Marketing, vol. 16(2), pp. 111–124.

76.

Yang

Lau

(2015) “LuXurY” hotel loyalty—a comparison of Chinese Gen X and Y tourists to Macau. International Journal of Contemporary Hospitality Management 27(7): 338–352.

77.

Yoon

Lee

(2010) Measuring festival quality and value affecting visitors’ satisfaction and loyalty using a structural approach. International Journal of Hospitality Management 2: 335–342.

78.

Zabkar

Brenc

Dmitrovic

(2010) Modelling perceived quality, visitor satisfaction and behavioural intentions at the destination level. Tourism Management 31(4): 537–546.

79.

Zeithaml

(1988) Consumer perceptions of price, quality and value: a means-end model and synthesis of evidence. Journal of Marketing 52: 2–22.

80.

Zeithaml

Berry

Parasuraman

(1996) The behavioral consequences of service quality. Journal of Marketing 60: 31–46.

81.

Zeithaml

Bitner

Gremler

(2009) Services Marketing: Integrating Customer Focus Across the Firm. New York: Mcgraw-Hill/Irwin.