Abstract
Using a long-duration service experience, this study proposed and tested a servicescape transference effects phenomenon of two nested multidimensional servicescape satisfaction constructs—labeled perceived nestscape and surroundscape satisfaction—on one another and on loyalty intentions. Both perceived servicescape satisfaction constructs positively affected loyalty intentions. The direction of effects between the two satisfaction constructs was found to emanate from satisfaction with the larger surroundscape to satisfaction with the smaller nestscape rather than the opposite direction. All dimensions of the servicescapes were significant, but the design dimension contributed the most to perceived nestscape satisfaction, while the social dimension had the greatest influence on surroundscape satisfaction. These findings suggest that managers of servicescapes should proactively respond to all dimensions of both the larger and the smaller contiguous servicescapes to heighten customer satisfaction and loyalty. In addition, servicescape engagement positively affected perceived surroundscape satisfaction, but not perceived nestscape satisfaction. Servicescape engagement moderated the nestscape and surroundscape effects by weakening the impacts of both constructs on loyalty intentions. Consequently, managers should provide superior levels of engagement activities for seasonal migrants less concerned with the servicescape and enhance perceived servicescape satisfaction for less engaged consumers to induce loyalty.
There are no more ubiquitous servicescapes than the places in which we live and play. The myriad of residential options and amenities—from urban to suburban to rural dwellings—comprise servicescapes that are embedded or nested in cities, towns, metropolitan statistical areas, or counties with servicescapes of their own, planned and developed to provide services to residents and visitors. The power of each of these servicescapes to influence our feelings, thoughts, activities, and lifestyle is undeniable (Jackson 2003). Walking paths encourage us to walk more (Brownson et al. 2001), restaurants encourage us to dine out, parks encourage us to gather with others (Kuo et al. 1998), neighborhoods increase residents’ social capital, and so on (Jackson 2003). The characteristics—or servicescape—of each of the places within which we live may affect us differently, depending on how we view the servicescape at each level; yet, little research exists about residential servicescapes, whether permanent or temporary, or about the effects of one location nested within another.
A large body of research has been devoted to understanding commercial servicescapes (e.g., Baker 1986; Ezeh and Harris 2007; Harris and Ezeh 2008; Rosenbaum and Massiah 2011), with most of this inquiry focused on dimensions and effects as proposed in Bitner’s (1992) groundbreaking framework. Concurrently, a number of studies have examined image transfer effects in various contexts, including brands, events, destinations, and celebrities (e.g., Gwinner 1997; Keller 1993; McCracken 1986), implying that the underpinning theory may also apply to contiguous or nested servicescapes, such as with a store within a mall. While some researchers have proposed a nested image transference effect (Chebat, Sirgy, and St-James 2006), no known studies have empirically tested and reported these possible effects.
To begin this discourse about nested servicescapes, we draw from Baker’s (1986) and Bitner’s (1992) frameworks, related research, and image transfer theory to expand our knowledge of servicescapes in five ways. First and foremost, this research uniquely investigates the directionality of the relationship between satisfaction with two different servicescapes, one nested in another. This objective is achieved by exploring the interrelationship between the servicescape satisfaction effect of extended stay, temporary residential communities specifically designed for wintering migrants, and that of the larger, regional area in which the communities exist. This directional nested effect is important to servicescape research, since the impact of one servicescape nested within another is likely to influence both consumer intentions and behavior regarding both servicescape providers. Furthermore, one servicescape may mediate the other’s influence on outcome variables, which may affect future marketing efforts for both servicescapes.
Second, the combined effect of satisfaction with the two nested servicescapes on loyalty intentions is examined. Third, the impact of servicescape engagement on satisfaction with the nested servicescapes is also explored. Specifically, servicescape engagement is conceptualized to both directly affect servicescape satisfaction and moderate the effect of servicescape satisfaction on loyalty intentions. Engagement in activities should generally increase satisfaction with the servicescape in which the activities occur, since the individual is actively rather than passively consuming servicescape offerings. This specific relationship has received almost no attention in a broad range of literatures.
Fourth, this study augments servicescape knowledge by addressing the call for research to better understand extended, relatively long duration service experiences (e.g., Arnould and Price 1993; Ezeh and Harris 2007; Mari and Poggesi 2013). With a few notable exceptions (Arnould and Price 1993; Gamefeld and Steinhoff 2013; Price, Arnould, and Tierney 1995; Rosenbaum and Smallwood 2011), prior research has focused on short-duration experiences.
Finally, this study is conducted among older, retired consumers. Research related to this market is growing in importance, as the number of seniors in proportion to the total population soars (Meiners and Seeberger 2010). Currently, retirees represent a massive, underappreciated market with twice the discretionary income of younger cohorts and a demonstrated willingness to splurge on purchases for themselves (e.g., Meiners and Seeberger 2010; Moschis 2002). Futurists predict that the retiree market will remain significant for decades and will be a primary determinant of success for relevant industries (e.g., The Wall Street Journal 2013; Wilson 2009).
Study Context: Active Adult Communities
This study was conducted among the approximate 100,000 older consumers who routinely spend the winter in the subtropical climate of the Rio Grande Valley (RGV) in South Texas. In 2014, these seasonal migrants had journeyed to the RGV an average of 11 years, and about 90% reported living in amenity-rich, 55+, active adult mobile home and recreational vehicle (RV) communities where just over half of the migrants owned a mobile home and 34% owned an RV (Simpson 2014). These consumers are considered “permanent migrants” (Hall and Williams 2013) rather than traditional tourists by virtue of their extended stay and long history of returning to the destination.
As long ago as the early 1900s, seasonal migrant communities evolved from the annual winter migration of retirees from the midwestern and northern parts of the United States to Florida (Trolander 2011). Then, as now, these seasonal migrant communities were manufactured home parks and were popular because of the low cost of housing and numerous amenities. These communities were generally designed to facilitate activity and socializing among seniors and, according to Trolander (2011, p. 952), were “experiments in social engineering.” Today, such design features generally include the proximity of neighbors; golf courses; tennis courts; club houses for dances, crafting, and so on; and gated perimeters with little-trafficked roadways suitable for walking and travel by golf cart. Most of these communities also have activity coordinators who develop comprehensive itineraries to promote an active, interactive lifestyle (Trolander 2011) while fostering an image of “relatively youthful retirees socializing with roughly the same number of the opposite sex, enjoying golf, and relaxing by the pool” where “there are a lot of people just like us” (Trolander 2011, pp. 969, 964).
While some seasonal migrants form a strong attachment to their winter or second homes and return each season (e.g., Hall and Williams 2013; Stedman 2006), they can readily move between locations or never return if they are unhappy with their neighbors, the weather, and area attractions or if the community or destination conditions deteriorate (Franklin and Raadschelders 2007; Hall and Williams 2013). Indeed, 39.7% of the respondents in this study had moved from one mobile home/RV community to another during their time in the RGV. Furthermore, the place attachment of seasonal migrants is different from that of permanent residents. Permanent residents’ attachment to an area is based on “social networks and community meanings,” while the attachment of seasonal migrants is rooted in the “meaning of environmental quality and escape from day-to-day cares” (Stedman 2006, p. 201). Permanent residents create a sense of a place, while seasonal migrants consume the place through participation in recreational and relaxation activities (Tuan 1977).
Seasonal migrants—also known as snowbirds, gray nomads, and winter migrants—are a global phenomenon. Over one million people seasonally journey to Spain from Britain, Norway, and other countries (Haug, Dann, and Mehmetoglu 2007); thousands travel to Morocco from France (Viallon 2012); and hundreds of thousands visit destinations around Australia (Mahadevan 2014). While current statistics are elusive, early studies estimated that 10 million U.S. seniors migrate in the winter, primarily to Arizona, Florida, and Texas (Rush 1980; Smith and House 2006), thus significantly contributing to those local economies. For example, seasonal migrants spent almost US$1 billion in Texas in 1997 (Report: Snowbirds Bring 2001) and “billions” in Florida (“Florida Snowbirds” 2014).
Considering the substantial economic impact of seasonal migrants, mobile home/RV communities and the broader regional destinations often design activities, programs, and features that specifically target the group to behaviorally engage them and encourage their return. In the RGV, the seasonal migrants are called Winter Texans to make them feel appreciated. Winter Texan festivals, events, and signs are numerous; Winter Texan newspapers and websites have been created (e.g., www.wintertexantimes.com and www.welcomehomergv.com) and businesses offer Winter Texan coupons. These considerations, as well as the subtropical climate, welcoming people, and various activities, have resulted in a return intention rate of 96.6% among Winter Texans (Simpson 2014).
Theoretical Background and Conceptual Framework
In the RGV, mobile home/RV communities and the region have created servicescape environments to meet the experiential needs and wants of seasonal migrants. Thus, Bitner’s (1992) theoretical framework is an appropriate model for understanding how the dimensions of a service’s environment will, holistically, affect perceptions of the servicescape, which in turn influence behaviors as moderated by constructs such as personality traits and situational factors. Since its inception, the framework has been tested in various contexts and expanded to include social and nonbuilt dimensions. What has not been examined heretofore, however, are effects of perceptions of one servicescape on another, when one servicescape resides within the other as with departments within retail stores, retail stores within malls, medical clinics within hospitals, cities within states, and so on.
Because of the nested, geographical proximity of a community within a region, servicescape perceptions of one entity are likely to affect servicescape perceptions of the other, according to the law of contiguity and image transfer theory. The law of contiguity maintains that when two ideas or events occur in spatiotemporal proximity, individuals form associations between those ideas or events such that the subsequent occurrence of one tends to elicit memories of the other (Lachnit 2003; Mill 1882/2005), which now serves as the basic premise underlying “most scientific theories of learning, memory, and knowledge” (Lachnit 2003, p. 3). As an example of the contagion effects of contiguity, an experiment by Rozin, Millman, and Nemeroff (1986) found that the placement of a disgusting object, a dead roach, near one drink encouraged respondents to select a drink located further away from the roach.
In marketing research, image transfer theory, derived from the law of contiguity, explains that the image or associations of an object, place, or idea may be transferred to another entity depending on various factors, including the nexus between entities that are formed in consumers’ minds. This theory has been used primarily to explain image transference effects from brands to family brands (e.g., Keller 1993; Martin, Stewart, and Matta 2005), from events to event sponsors (e.g., Gwinner 1997; Smith 2004) and destinations (e.g., Deng and Li 2014; Kim and Morrison 2005; Lee, Lee, and Lee 2005), from country-of-origin to brands (e.g., Diamantopoulos, Schlegelmilch, and Palihawadana 2011), and from celebrity endorsers to brand image (e.g., McCracken 1986). Image transfer theory was used in a retail context by Chebat, Sirgy, and St-James (2006) to explain effects of mall image on store image and by Dion and Arnould (2011) to explain the transfer of ideological meanings embodied by art work to a retail brand.
While these works demonstrate the importance of the law of contiguity and image transfer theory, they are not specific to experiential servicescape environments as is the case with lived-in spaces, where consumers are fully immersed in and experienced with all dimensions of a community servicescape and its surrounding servicescape. Also, these prior investigations do not explore the idea that both entities, one nested in another, will likely affect outcomes that are important to both entities. Hence, this study melds image transfer theory with servicescape theory by proposing that, in the case of nested servicescapes, the image, impressions, perceptions, and associations of one physical and biotic environment may be transferred to, or affect, another contiguous one. This phenomenon is hereafter termed servicescape effects transference. Underpinned by image transfer and servicescape theories, servicescape effects transference suggests that a servicescape’s effects should not be considered in isolation but rather that the management of any servicescape must also consider the servicescapes of contiguous entities.
The notion of nested servicescape effects transference has been alluded to without a theoretical reference in some prior research. Examples include studies of malls and their stores (Chebat, Sirgy, and Grzeskowiak 2010), wine regions and wine tasting rooms (Bruwer and Lesschaeve 2012), and overlaps between country image and destination image with regard to products and objects (Mossberg and Kleppe 2005). A few other studies testing nested relationships with some theoretical positioning, but without specific linkages to servicescapes, include Chebat, Sirgy, and St-James’ (2006) investigation of store image on mall image effects and Kim, Kang, and Kim’s (2014) examination of Olympic Games’ effects on destination and country image, where the authors noted that country image should be examined because the nation is comprised of all destinations. Finally, Nadeau et al. (2008) found nested effects of product country image and destination image on travel intentions.
Accordingly, based on Bitner’s (1992) servicescape framework and the newly proposed servicescape effects transfer, Figure 1 illustrates our proposed theoretical structure, whereby the perceived servicescape of a nested entity and its surrounding servicescape both affect loyalty intentions. This perspective suggests an expanded definition of servicescape as comprised of all controllable and uncontrollable dimensions of the immediate environment, labeled nestscape, in which a service is performed, delivered, and consumed, and the encompassing environs, labeled surroundscape, which together define the total customer experience with and response to the service. As with traditional conceptualizations of servicescape, both servicescapes consist of the built and nonbuilt environment incorporating design and ambiance as previously conveyed by Bitner (1992) and Baker (1986) and the social environment, consisting of all human interactions, including those with employees and other customers as described by Baker (1986) and Rosenbaum (2006). The model also includes servicescape engagement, a variable proposed to directly affect perceived servicescape satisfaction and moderate the relationship between perceived servicescape satisfaction and loyalty intentions.
Servicescape framework. Adapted from Baker (1986) and Bitner (1992). Bold-faced constructs and dotted lines indicate relationships examined in this study.
Servicescape Dimensions and Research Hypotheses
Servicescape Dimensions
Despite more than two decades of research, little consistency in defining the dimensions of servicescape exists. Some selected marketing and hospitality studies—chosen simply to illustrate the diversity of servicescape dimensions—utilize ambient, design, and social factors (Baker 1986); layout accessibility, facility aesthetics, seating comfort, facility cleanliness, and electronic equipment displays (Wakefield and Blodgett 1996); ambience, design and color, cleanliness, and layout (Kearney, Coughlan, and Kennedy 2013); ambient, spatial layout/functionality, signs/symbols/artefacts, and restorative dimensions (Rosenbaum and Smallwood 2011); ambience, space/functionality, signs, symbols, and artifacts (Brüggen, Foubert, and Gremler 2011); physical, social, socially symbolic, and natural environments (Rosenbaum and Massiah 2011); and physical setting, service providers, and other customers (Wang and Mattila 2015). Despite this broad range of dimensions, there is now a general consensus that servicescapes are context specific and encompass both the observable, physical, or built environment and the biotic environment—the “world of living things” (Hightower 2013). We adopt Baker’s (1986) servicescape dimensions—ambient, design, and social—as they appear particularly relevant given Baker’s sociological approach “which acknowledges the influence of human characteristics in shaping the service experience” (Harris and Ezeh 2008, p. 392).
The ambient dimension
Ambient elements of a servicescape have been defined as “background conditions that exist below the level of immediate awareness” (Aubert-Gamet 1997, p. 29). This dimension has been operationalized in different ways including, among others, music and lighting for retail spaces (Baker, Levy, and Grewal 1992); humidity, circulation, temperature, light, scent, and noise for an international airport (Jeon and Kim 2012); music, olfaction, lighting, and cleanliness for grocery stores (Kearney, Coughlan, and Kennedy 2013); and acoustics, lighting, air quality, temperature, furniture, and visual attractiveness for a health-care facility (Lee 2011).
The design dimension
Design elements are functional or aesthetic features intentionally constructed into the service environment (Brady and Cronin 2001; Nilsson and Ballantyne 2014) with a focus on creating added value for people through their interactions with the design (Juozulynas 2014). Examples of the design dimension include architecture, color, scale, style, shape, layout, comfort, signage, and pattern (Aubert-Gamet 1997; Jeon and Kim 2012; Wakefield and Blodgett 1999).
The social dimension
The social dimension is the human component and includes the number and type of people found in a servicescape (Aubert-Gamet 1997; Brady and Cronin 2001) and represents “how people feel (e.g., comfortable) in the service setting” (Baker and Cameron 1996, p. 340). The social dimension may be especially important in the servicescape of places. Unlike generations ago, in today’s world “the ideal norm has become that of a home where everything that concerns the upkeep of the family takes place between four walls” (Aubert-Gamet and Cova 1999, p. 40), leaving little opportunity for frequent, casual socializing. Thus, Aubert-Gamet and Cova (1999, p. 43) suggested that contemporary services “be designed as postmodern places of creation and development of social links and communities.” In other words, services should create “linking values” of consumers, one to another, through servicescapes by, for example, creating places and experiences that encourage interaction.
Unfortunately, little academic work has been conducted concerning the social dimension—concepts of community and interaction—in servicescapes since the Aubert-Gamet and Cova (1999) article was first published, despite offering the most promise (Mari and Poggesi 2013). Some servicescape-related works have proposed theoretical models that incorporate a social dimension. For instance, Tombs and McColl-Kennedy’s (2003) conceptual model included purchase occasion, social density, displayed emotions of others, and customers’ affective and cognitive responses as elements of the social servicescape. Rosenbaum and Massiah (2011) conceptualized the social dimension as comprising factors related to employees, customers, social density, and displayed emotions of others. Beyond these theoretical developments, two notable studies have empirically tested the social dimension: Harris and Ezeh (2008) examined the influence of a servicescape’s social elements, measured by customers’ interactions with employees and employee image, on loyalty intention, and Jeon and Kim (2012) investigated social dimension effects—other customers and service personnel in terms of number, appearance, and behavior—on customers’ emotions.
Research Hypotheses
Servicescapes and loyalty intentions
Marketers have generally found that consumers’ positive experiences with servicescapes yield positive cognitive, affective, and psychological responses that facilitate firm-desired behaviors, including loyalty, as Bitner (1992) proposed. Within various contexts, research has revealed that customers’ comprehensive views of a servicescape influence affect (e.g., Babin, Chebat, and Michon 2004; Jeon and Kim 2012), desire to stay and recommend (e.g., Dong and Siu 2013; Harris and Ezeh 2008), and loyalty (Brüggen, Foubert, and Gremler 2011; Spangenberg et al. 2006). As such, the individual, as well as the combined influence of satisfaction with the nestscape (in this case, the servicescape of a mobile home/RV community) and the surroundscape (in this case, the community’s surrounding regional servicescape), should affect recommendation likelihood as an indication of loyalty intentions:
Nested servicescapes
As previously noted, image transfer theory has been employed in various marketing research contexts (e.g., Chebat, Sirgy, and St-James 2006; Gwinner 1997; Martin, Stewart, and Matta 2005; Pappu, Quester, and Cooksey 2007; Smith 2004). The theory has also been tested in destination image studies. Deng and Li (2014) found that an event image significantly influenced destination image, and Heslop et al. (2013) substantiated relationships between mega-events and corporate reputation and host country reputation. Other research has shown that a tourist destination image can be transferred to products associated with the destination (Lim and Weaver 2012) or from a country’s image to restaurant preferences (Lee, Ham, and Kim 2015).
These empirical findings suggest that a smaller and larger entity can each impact the image, perceptions, impressions, and associations of the other, although the directionality of transfer is unclear. Therefore, we propose two competing hypotheses to determine the directionality of the nested relationship between the two perceived servicescape satisfaction constructs as follows:
Servicescape engagement
The actual engagement level of consumers in activities and design is likely to influence perceived servicescape satisfaction, as well as the relationships between perceived servicescape satisfaction and loyalty, especially where numerous opportunities for activities and interaction are created to foster a participative lifestyle among a specific target market. However, merely having servicescape features is an insufficient condition to evoke loyalty. Consumers must experience the servicescape elements for the attributes to induce loyalty. A resident who is satisfied with a destination’s hiking trails but never uses them will likely not have the same outcomes as residents who routinely experience this servicescape element.
Customer engagement research is nascent; yet, a variety of definitions already exist. Brodie et al. (2011) conceptualized customer engagement as a psychological state, Van Doorn et al. (2010, p. 254) operationalized such behaviors as “customer’s behavioral manifestations,” and Vivek, Beatty, and Morgan (2012, p. 133) defined it as “the intensity of an individual’s participation in and connection with an organization’s offerings or organizational activities.” Adopting this latter definition, we conceptualize servicescape engagement as the active participation in or experience with the nestscape’s and surroundscape’s offerings and activities. Examples include walking on city sidewalks, swimming in hotel pools, using gym equipment, or listening to music at a bar.
The relatively few studies on the effects of customer engagement in activities have found engagement is a significant predictor of consumption behavior (Calder, Isaac, and Malthouse 2013) and activity participation leads to affective commitment to and positive evaluations of service offerings (e.g., So et al. 2016; Vivek, Beatty, and Morgan 2012; Wakefield and Blodgett 1994). While the effect of activity on perceived servicescape satisfaction has not been directly investigated, Björk et al. (2008) found that residents of Swedish neighborhoods with nearby recreational options engaged in more physical activities and rated their satisfaction with the neighborhood significantly higher than did residents of neighborhoods lacking nearby recreational options. Similarly, de Jong et al. (2012) reported that the presence of green spaces in Sweden was positively associated with neighborhood satisfaction and physical activities. Therefore:
The RGV and most seasonal migrant communities provide many opportunities for their residents to engage in socializing, activities, and events. Prior research suggests that the more the residents participate in these activities, the more involved and engaged they will be (e.g., Björk et al. 2008; So et al. 2016; Vivek, Beatty, and Morgan 2012; Wakefield and Blodgett 1994). Thus, when compared to those who are less participative, seasonal migrants who are more engaged in servicescape activities will likely have stronger ties to the nestscape and surroundscape in which they live and the greater the likely impact of the nestscape and surroundscape on loyalty intention. Following this logic:
Hypotheses 1–8 are depicted in Figure 2.
Research model.
Method
Servicescape and Servicescape Indicator Selection
A pilot study was conducted by disseminating a short survey to seasonal migrants during a trade show targeting Winter Texans. The survey asked respondents open-ended questions about their likes and dislikes of the mobile home/RV community and the RGV city in which they currently lived as well as activities they participated in while in the region. Results from the 1,603 completed questionnaires were used to create a list of community and regional attributes and activities. The listed attributes, as indicators of each servicescape satisfaction dimension and servicescape engagement, were supplemented by Winter Texan experts, and the destination image literature which indicated that consumer impressions are determined by cognitive or attribute-based image, holistic perceptions, and affective impressions (e.g., Alcañiz, García, and Blas 2009; Baloglu and McCleary 1999; Echtner and Ritchie 1991; Nghiêm-Phú 2015). Attribute examples gleaned from the literature included visible functional attributes (e.g., activities, architecture, attractions, infrastructure), invisible psychological or felt attributes (e.g., friendliness, service quality), and mixed attributes, both seen and felt (e.g., safety, cleanliness). This process yielded 17 nestscape and 18 surroundscape attributes as measures of perceived nestscape satisfaction (PNS) and perceived surroundscape satisfaction (PSS), respectively, and 13 servicescape engagement (SE) activities (see Table 1).
Attributes for Constructs With Formative Measurement Model Specifications.
Note. RGV = Rio Grande Valley; RV = recreational vehicle; VIF = variance inflation factor.
aFriendliness of residents and social activities were dropped as indicators for perceived nestscape satisfaction due to nonsignificant loadings and weights.
**p < .01. *p < .05; critical t-values for a two-tailed test: 1.96 (p = .05); 2.58 (p = .01).
Data Collection and Sample
A survey designed to determine mobile home/RV community residents’ satisfaction with various community and regional attributes was created based on the pilot study and then posted online. The survey link was promoted to Winter Texans on the home pages of seven large mobile home/RV communities and through flyers, e-mails, newspaper items, Facebook, and an online advertisement. Hard copies of the survey were also distributed to mobile home/RV communities. All respondents were offered entry into a drawing for a Kindle Fire. Of 1,337 total responses, 159 were eliminated from the analysis because they were from permanent residents. The average respondent age was 68, 50.3% were male, 47.7% reported having a bachelor or higher degree, and 62.1% reported having an annual household income between US$30,000 and US$69,999.
Measures
Following Jarvis, MacKenzie, and Podsakoff’s (2003) criteria for measurement model operationalization and Becker, Klein, and Wetzels’s (2012) guidelines for using hierarchical latent variable models in partial least squares structural equation modeling (PLS-SEM), we employed a formative-formative hierarchical component model (Petter, Straub, and Rai 2007) and the repeated indicator approach (Lohmöller 1989) to measure PNS, PSS, and SE and their lower order components as did Klarner et al. (2013) and Rai, Patnayakuni, and Seth (2006). PNS and PSS each comprised three first-order components, that is, satisfaction with the ambient, design, and social dimensions of the nestscape and the surroundscape, respectively. To measure satisfaction with each servicescape dimension, study participants were asked to indicate their satisfaction with each of the community and regional attributes on a 5-point scale, with end points 1 representing very dissatisfied and 5 representing very satisfied. Likewise, SE comprised three first-order components, that is, interaction, entertainment, and discovery, following the method used by Calder, Isaac, and Malthouse (2013, 2015). To assess SE, respondents were asked to indicate how often they participated in each servicescape activity on a 5-point scale with end points 1 representing never and 5 representing a lot. Loyalty intentions (LI), a first-order reflective construct, was assessed by 2 items: “How likely are you to recommend your current park community to a friend or family member?” and “How likely are you to recommend the Valley as a winter home to a friend or family member?” on a 10-point scale with end points 1, not at all likely, and 10, extremely likely. Demographic characteristics including age, gender, education, and household income were collected along with other variables of interest, such as number of years journeying to the RGV and moves to different mobile home/RV communities in the RGV.
Results
SmartPLS 3.2.3 (Ringle, Wende, and Becker 2016) was used in the data analysis. In evaluating and reporting the PLS-SEM results, Hair et al.’s (2014) and Hair, Ringle, and Sarstedt’s (2011) guidelines were followed by first assessing the measurement models, then the structural model.
Measurement Model
Measure validity for LI, the only reflectively measured construct, was supported by the substantial factor loadings of .91 and .82 for the 2 measurement items, a composite reliability value of .86, and the average variance extracted value of .75. For PNS, PSS, and SE, constructs with formative-formative hierarchical component model specifications, indicators’ outer weights, and loadings were first examined, and then a bootstrapping procedure was run to determine significance. Eleven indicators were found statistically appropriate for the PNS construct, 16 for the PSS construct, and 6 for the SE construct (see Table 1). The variance inflation factor calculations revealed values between 1.02 and 2.97, with none greater than 5, indicating that multicollinearity was not a concern (see Table 1).
Subsequent analyses indicated that satisfaction with each first-order servicescape dimension had a significant, positive impact on the second-order construct of overall perceived servicescape satisfaction. As seen in Table 1, satisfaction with the design dimension had the largest impact on PNS (β = 0.60, t = 61.43, p < .001), followed by satisfaction with the social (β = 0.33, t = 53.75, p < .001), and then ambient dimensions (β = 0.20, t = 19.13, p < .001). Satisfaction with the social dimension had the largest impact on PSS (β = .54, t = 72.79, p < .001), followed by satisfaction with the design (β = 0.42, t = 57.48, p < .001) and ambient dimensions (β = 0.17, t = 20.03, p < .001). For SE, the interaction dimension had the largest impact (β = 0.48, t = 59.07, p < .001), followed by the entertainment (β = 0.44, t = 46.65, p < .001), then the discovery dimensions (β = 0.39, t = 53.21, p < .001). These results confirmed the hierarchical component structure of all three formative constructs and demonstrated the relative contribution of each first-order component in forming its second-order construct.
Harman’s single-factor test was also run by conducting an exploratory factor analysis on all measurement items. The unrotated factor solution yielded eight factors with eigenvalues greater than 1, explaining 57.22% of the total variance. The first factor accounted for 22.97% of the total variance, the second factor explained 10.54%, and the remaining factors explained 2.91–6.31% of total variance. These results indicated that common method bias was not a concern.
Structural Model
As summarized in Table 2, results from testing the structural model indicated that PNS (β = 0.49, t = 15.39, p < .001) and PSS (β = 0.26, t = 8.16, p < .001) positively affected LI, with an R2 value of .39, supporting Hypotheses 1 and 2.
Summary of Hypothesis Testing Results.
Note. LLCI (lower limit confidence interval): 95%, bias-corrected lower bound confidence interval; ULCI (upper limit confidence interval): 95%, bias-corrected upper bound confidence interval.
aCohen’s path analysis results rejected Hypothesis 4.
**p < .01. *p < .05. Critical t-values for a two-tailed test: 1.65 (p = .1); 1.96 (p = .05); 2.58 (p = .01).
Next, the directionality of the relationship between PNS and PSS was tested using Cohen’s path analysis approach (Cohen and Bacdayan 1994; Cohen et al. 1993; Sattler et al. 2010). Regression coefficients and correlations obtained from the structural model analysis were used to calculate the estimated correlations, total squared error (TSE), and Cohen’s d value. Results, summarized in Table 3, indicated that TSE was 0.0069 for PSS → PNS and 0.0242 for PNS → PSS, resulting in a 250.72% change. The positive sign of the TSE change meant that when the direction of effects between PNS and PSS changed from PSS → PNS to PNS → PSS, TSE was increased. The absolute value of Cohen’s d estimated for the TSE difference between PSS → PNS and PNS → PSS was 1.0690, a large change according to Cohen’s (1992) criteria. This result indicated that the PLS path model relationship from PSS → PNS was preferable; thus, we rejected Hypothesis 4 and proceeded to test Hypothesis 3, using PROCESS v2.16 (www.processmacro.org) and following Hayes’ (2013) guidelines and the mediation analysis steps detailed in Zhao, Lynch, and Chen (2010).
Summary of Cohen’s Path Analysis Testing Results.
Results from the bootstrap analysis using 10,000 bootstrap samples, shown in Table 2, indicated that the indirect effect of PSS on LI through PNS was positive and significant (M = 0.15), with a 95% bias-corrected confidence interval excluding zero (0.1–0.19). Likewise, the direct effect of PSS on LI was positive and significant (M = 0.27, p < .001). These results supported Hypothesis 3 that PNS partially mediates effects of PSS on LI.
We then tested the direct and moderating effects of SE as proposed in Hypotheses 5–8, using the two-stage calculation method and standardized data for computing the product terms of the interaction effect. Findings indicated that SE positively affected PSS (β = 0.32, t = 10.44, p < .001), supporting Hypothesis 6. However, Hypothesis 5, the positive effect of SE on PNS, was not supported (β = 0.03, t = 0.80, p = .43). Additionally, while SE did indeed moderate the effects of PNS and PSS on LI, the negative β coefficient of the interaction term indicated that SE weakened rather than strengthened the positive effect of PNS on LI (β = −0.10, t = 2.74, p = .007) and PSS on LI (β = −0.06, t = 2.06, p < 0.04), contrary to Hypotheses 7 and 8.
Finally, we tested whether demographic variables (i.e., age, gender, education level, and income level), years of migration to the RGV, and moving to a different community in the RGV impacted the research model. Only gender and years of migration had any effects. The PSS of female respondents had a stronger effect on LI than males (βFemale = 0.40 vs. βMale = 0.27, t = 1.96, p = .05), but SE had a weaker effect on PSS for females than for males (βFemale = 0.24 vs. βMale = 0.37, t = 2.09, p = .04). Also, effects of PSS on PNS were stronger for respondents who had migrated to the RGV for five or fewer winters and weaker for respondents who had migrated for 11 years or more (βFewer years = 0.43 vs. βMore years = 0.24, t = 2.22, p = .03) as were the effects of SE on PSS (βFewer years = 0.45 vs. βMore years Stay = 0.24, t = 2.67, p = .008).
Discussion
This study adopted a broadened, holistic view of servicescapes as consisting of all controllable and uncontrollable dimensions of the immediate environment, or nestscape, in which a service is performed, delivered, and consumed, and the encompassing environs, or surroundscape, which together define the total customer experience with and response to the service. The test of this conceptualization found, first and foremost, that a servicescape effects transference between two nested servicescapes, the nestscape of the mobile home/RV community and the surroundscape of the region, was supported with the direction of effects from the surroundscape to the nestscape. This finding of transfer effects corroborated and extended prior research, especially regarding contiguous contexts (Chebat, Sirgy, and St-James 2006; Dion and Arnould 2011). Second, while perceived satisfaction with both servicescapes positively affected loyalty intentions directly, perceived surroundscape satisfaction also impacted loyalty intention through a partial mediating effect of perceived nestscape satisfaction, which demonstrated the compounding effect of both the surroundscape and nestscape on loyalty behaviors. Thus, this research further confirmed the importance of servicescape theory as proposed in Bitner’s (1992) seminal work and added to the extensive prior research establishing servicescapes’ influence on consumers’ loyalty behaviors such as positive recommendations (e.g., Brüggen, Foubert, and Gremler 2011; Dong and Siu 2013; Spangenberg et al. 2006).
Third, engagement in activities enhanced perceived surroundscape satisfaction, adding support to other customer engagement studies and emerging theories (e.g., Bowden 2009; Van Doorn et al. 2010). However, engagement did not improve perceived nestscape satisfaction. A possible explanation is that the engagement measure was not specific to the community/nestscape because respondents were asked about their activity participation in the RGV. This result highlighted the importance of being “out and about” and engaged in both the broader surroundscape and the smaller nestscape because engagement in surroundscape alone was not sufficient to create nestscape satisfaction as our study showed. Of the engagement dimensions examined in this study, the interaction dimension contributed most to forming the second-order servicescape engagement construct, highlighting the significance of human connectedness in servicescapes. Unexpectedly, engagement was found to weaken, rather than strengthen, the effect of perceived servicescape satisfaction on loyalty intentions. Perhaps high levels of activity become all-consuming, making perceived servicescape satisfaction itself less important in loyalty intention. For example, consumers may be willing to overlook a rundown beach villa if the beach activities are exceptional. On the other hand, lower levels of engagement strengthened the impact of perceived servicescape satisfaction on loyalty intentions, conceivably because consumer attention is less distracted by activity involvement and, therefore, more focused on servicescape factors. These unanticipated findings lend credence to Henderson, Steinhoff, and Palmatier’s (2014) findings of a negative impact of customer engagement when the customer-company bond was built on habits and relationships.
This study also confirmed the significance of the social dimension of servicescapes, thereby filling a gap in extant servicescape research where the importance of a servicescape’s social component had been conceptualized (e.g., Rosenbaum and Massiah 2011; Tombs and McColl-Kennedy 2003) but rarely tested. The social dimension was found important, especially regarding the surroundscape, which supports the idea that customers seek and appreciate surroundscapes that encourage socializing opportunities as Aubert-Gamet and Cova (1999) proposed for postmodern consumers. In contrast, the finding that the social dimension was a less important component of perceived nestscape satisfaction than the design dimension was somewhat unexpected, given the resources that the communities typically devote to creating daily socializing opportunities for residents. This result is not to say that the social dimension was unimportant but that the design element was simply more important. Seasonal migrants placed a high priority on nestscape design, a desire likely fueled by the long-standing, aging RV/Mobile home communities found across the RGV, but as also explained by previous hospitality research findings reporting that travelers often place a greater emphasis on the design elements (“the tangibles”) of lodging facilities than on any other servicescape element (e.g., Akan 1995; Ladhari 2012). These results further suggested that the social experiences offered in the surroundscape may have had as much or more impact on loyalty than the social experiences offered in the nestscape. Nevertheless, the overall findings about the social dimension of servicescapes underscored the importance of human contact and interaction and the need for managers to facilitate such connections. Although significant, the result that the ambient dimension was the least important component of both perceived nestscape and surroundscape satisfaction was interesting but not surprising, given that customers may not be immediately aware of ambient factors as suggested by other research (e.g., Aubert-Gamet 1997).
Finally, only two control variables, gender and years migrating to the RGV, had significant impacts on the research model. The positive effect of perceived surroundscape satisfaction on loyalty intentions was stronger for females, while servicescape engagement had a stronger impact on perceived surroundscape satisfaction for males. Moreover, the finding that the impact of perceived surroundscape satisfaction on perceived nestscape satisfaction and that of servicescape engagement on surroundscape satisfaction were weaker for those who had been wintering in the RGV for more years indicates the presence of a decaying effect. Possibly, as seasonal migrants grow accustomed to the region over time, the influence of the regional surroundscape becomes less pronounced as suggested by the process of habituation, defined as a “decreased response to repeated stimulation” (Groves and Thompson 1970, p. 419).
Conclusions
Theoretical and Empirical Contributions
This research contributed to theory by melding image transfer theory with servicescape theory to yield the phenomenon of servicescape effects transference. As formally proposed and tested, the servicescape was not defined solely by its own dimensions and boundaries but, instead, was affected by other contiguous servicescapes within which the servicescape was nested. This augmented theoretical version suggesting that the image, perceptions, impressions, and associations of one servicescape are transferred to a contiguous, nested servicescape was supported in this study’s context of nestscapes and surroundscapes. The conceptualization of servicescape effects transference was borrowed from destination research streams that explicate how consumers are cognitively and emotionally affected by a location (e.g., Alcañiz, García, and Blas 2009; Baloglu and McCleary 1999; Echtner and Ritchie 1991; Nghiêm-Phú 2015) and how location or place may become important to self-identities that are formed through interactions with the physical environment (Dixon and Durrheim 2004). The importance of places to self-identity has been explored somewhat in the services literature through the study of third places (e.g., Rosenbaum 2006; Rosenbaum and Montoya 2007; Rosenbaum and Smallwood 2011) but deserves much more scrutiny.
This study also contributed to servicescape research in a number of ways. First, the findings of nested servicescape effects transference encourage the adoption of a broadened view of servicescapes, where the surrounding environment in which a servicescape exists is also considered when examining servicescape effects. Second, this study substantiated the influence of two servicescapes on loyalty intentions and demonstrated the importance of the individual, as well as the combined effects of both servicescapes, in increasing customer loyalty. Third, the role of servicescape engagement as both an antecedent to perceived servicescape satisfaction and a moderator that diminished the effects of perceived servicescape satisfaction on loyalty intentions was established. Fourth, this research answered recent pleas to expand servicescapes beyond just the built facility (e.g., Johnstone 2012; Rosenbaum and Massiah 2011) by empirically testing and substantiating an expanded servicescape model consisting of ambient, design, and social dimensions.
Fifth, this study and its extended-stay context addressed the call for more research to better understand relatively long duration service experiences (e.g., Arnould and Price 1993; Ezeh and Harris 2007; Mari and Poggesi 2013). Relatedly, we found that an increased number of years journeying to the RGV had a decaying effect on the impact of perceived surroundscape satisfaction on perceived nestscape satisfaction and that of servicescape engagement on perceived surroundscape satisfaction, in support of habituation effects. Finally, this study provided some insights into nested residential servicescapes and their effects on older consumers, a highly underresearched market in the academic literature, but one of great importance in terms of numbers worldwide.
Methodologically, this study contributed to the literature by confirming the multidimensional, higher order structure of perceived servicescape satisfaction and of servicescape engagement as composed of lower order, context-specific dimensions, supporting the gestalt, holistic view of servicescape as conceptualized by Bitner (1992) and that of engagement as conceptualized by Calder, Isaac, and Malthouse (2013). Additionally, this research highlighted the absolute, as well as relative, contribution of each servicescape and engagement dimension in forming its respective higher order construct. Finally, this study used Cohen’s path analysis approach to test the directionality of the relationship between the two nested perceived servicescape satisfaction constructs, thus providing additional evidence of the applicability and usefulness of this statistical method in untangling the direction of effects between theoretical constructs.
Managerial Implications
The ability of one servicescape to affect another has broad implications, such as the impact of malls on retail outlets, cities on hotels and entertainment venues, departments on stores, and so on. Subject to testing in various contexts, our findings suggest that managers of servicescapes should consider and proactively respond to all dimensions of both the larger and the smaller contiguous servicescapes, where possible, in order to heighten customer satisfaction and loyalty. For example, residential community owners/managers can inform city managers of roads in need of repair, help tourism managers implement social events, and contribute to surrounding landscapes. Moreover, events that may diminish satisfaction with the surroundscape might be mitigated by high levels of satisfaction with the nestscape. That is, the effects of a region’s spike in crime, a disaster, or other negative events, might be offset somewhat by a high level of community servicescape satisfaction. As another example, a mall may be able to defer updates somewhat if tenant stores’ servicescapes successfully or exceptionally satisfy customers’ preferences.
More specifically, these results demonstrated that nestscape and surroundscape managers have a range of factors—design elements, social activities, and ambient components—that can be manipulated to influence perceived servicescape satisfaction, which then affects customer loyalty intentions. In this study, the finding that the nestscape was less socially imbued than the surroundscape may indicate that community social activity levels could be reduced without harming satisfaction levels and loyalty, thus potentially decreasing nestscape management costs. At the same time, surroundscape managers should capitalize on social interaction activities to ensure that seasonal migrants return each year; and, given the more defining importance of the region’s servicescape, nestscape managers should vigorously collaborate with surroundscape managers to improve social interaction offerings, including helping to sponsor regional events and activities.
The study findings also suggested managerial implications based on gender and on activity engagement level. Managers should emphasize servicescape performance factors in targeting females, whereas a focus on engagement activities is more important for male seasonal migrants. Also, because high levels of involvement in activities might significantly lessen the importance of perceived servicescape satisfaction on loyalty intentions as found in this study, managers should focus on providing superior levels of engagement activities for seasonal migrants less concerned with the servicescape and enhance perceived servicescape satisfaction for less engaged consumers to induce greater loyalty, if resources allow the pursuit of both tactics.
Because design was the most important element in perceived nestscape satisfaction and the second most important dimension in perceived surroundscape satisfaction, managers should also strive to offer attractive, up-to-date, and well-maintained facilities, amenities, and infrastructure. These tangible elements create first impressions and visual impact, thus heavily influencing the approach-avoidance behaviors of visitors to the area.
Limitations and Suggestions for Future Research
As always, there are limitations that could be addressed by future investigators. First, the higher order construct of overall perceived servicescape satisfaction was measured with the repeated indicator approach. Although widely used in prior research involving formative-formative constructs, in hindsight, gathering data and having separate measures of overall servicescape satisfaction for both regional and community servicescapes would have allowed for a more robust statistical analysis. Second, the data are cross-sectional. A longitudinal analysis would provide greater insight into servicescape effects, given the extended duration service context, as would an analysis based on residence ownership status. Along these same lines, the servicescape effects of short- or long-term residents may be different from servicescape effects of consumers simply visiting a mall store, even if frequently. Third, this study examined only seasonal migrants in the RGV. Data collected from seasonal migrants in other parts of the world may yield different results. Also, this research was conducted among a convenience sample of volunteer, senior participants rather than a randomly selected sample among the general population, as needed to improve generalizability of results.
Despite the limitations of this study, the findings provide interesting insights, which we hope will inspire further research. For example, additional work about nested servicescapes is needed, since those relationships are pervasive but vastly underresearched. Another potentially fruitful and closely related area of research is the testing of servicescape effects transference within other contexts. For instance, what happens when one contiguous servicescape has a negative consumer perception and the other a positive one? Will loyalty to the positively perceived servicescape diminish? At what physical point does the transference effect cease? Future research could also examine whether the servicescape effects transference exists between contiguous, albeit not nested, servicescapes. Will the servicescape effects of one store transfer to its adjacent store? Additionally, while this research explored the combined influence of perceived surroundscape and nestscape satisfaction on loyalty intentions, the joint effects on other outcome variables may be of equal interest.
The relationship of nested servicescapes and the impact of servicescape engagement also deserves greater exploration, especially given the largely unexpected findings revealed by this data. For example, future research could reexamine the unsupported effect of servicescape engagement on perceived nestscape satisfaction using a separate nestscape engagement measure. Moreover, considering the diminishing impacts of perceived surroundscape satisfaction and servicescape engagement uncovered in this investigation, additional research in other contexts seems warranted to help further understand this decaying effect. Perhaps frequent customers of a service become inured to the servicescape over time in accordance with habituation effects, which begs the question: Does novelty trump familiarity with regard to servicescapes? Finally, as the most ubiquitous of all servicescapes, much more research is needed regarding the servicescapes of places we live, especially as they affect loyalty and well-being and how servicescape perceptions and effects may differ by sociodemographic characteristics including age. Indeed, we hope this study will encourage researchers to consider less conventional servicescapes, such as regions, towns, and neighborhoods, in their work. Arnould and Price (1993) pioneered this stream of research by examining the servicescape of wilderness rafting trips, and there is much territory to be covered and knowledge to be gained in the investigation of places in which we live, work, and play.
Footnotes
Acknowledgment
The authors are most appreciative of Dr. Christian Ringle who offered advice and guidance concerning the statistical analysis. The authors also wish to thank The University of Texas Rio Grande Valley, the Business and Tourism Research Center, and the Rio Grande Valley Tourism Cooperative officials for assistance in this research.
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) received no financial support for the research, authorship, and/or publication of this article.