DICHIARAZIONE SOSTITUTIVA DI CERTIFICAZIONE
DICHIARAZIONE SOSTITUTIVA DI CERTIFICAZIONE
(ai sensi dell’Art. 46, D.P.R. 445/2000)
DICHIARAZIONE SOSTITUTIVA DELL’ATTO DI NOTORIETA’
(ai sensi degli Artt. 19 e 47, D.P.R. 445/2000)
Il sottoscritto Xxxxxx Xxxxxxxx, nato a Termoli, prov. CB, il 01/03/1975, residente a TERMOLI prov. CB c.a.p. 86039
indirizzo Via INDIA 2 tel. 0000000000,
consapevole delle sanzioni penali, nel caso di dichiarazioni non veritiere o contenenti dati non più rispondenti a verità, di formazione o uso di atti falsi, richiamate dall’art. 76 del D.P.R. 445 del 28 dicembre 2000
DICHIARA
1) che l’articolo:
Xxxxxx, G. L., Xxx, Y., Presenza, A., Xxxxx, C. L. (2020). How does familiarity shape destination image and loyalty for visitors and residents?. Journal of Vacation Marketing, xxxxx://xxx.xxx/00.0000/0000000000000000
è frutto del lavoro comune degli Autori. Tuttavia, i diversi paragrafi sono da attribuirsi come segue:
✓ Casali: Instrument and measures; Measurement invariance between residents and non- residents; Discussion
✓ Xxx: Theoretical model; Analysis strategy; Descriptive statistics of measures
✓ Presenza: Introduction; Study area and sample; Conclusion
✓ Xxxxx: The role of familiarity in shaping destination image; Structural differences between residents and non-residents
2) che la copia dell’articolo sopra richiamato è conforme all’originale.
Il sottoscritto è a conoscenza che, ai sensi dell’art. 10 della legge 31 dicembre 1996, n. 675, i dati personali saranno trattati esclusivamente per le finalità di gestione della carriera
Termoli, 06.11.2020
IL DICHIARANTE
Article
How does familiarity shape destination image and loyalty for visitors and residents?
Xxxx Xxxx Xxxxxx and Xxxxx Xxx
Queensland University of Technology, Australia
Xxxxxx Xxxxxxxx
Universita degli Studi del Molise, Italy
Xxxx-Xxx Xxxxx
Queensland University of Technology, Australia
Journal of Vacation Marketing 1–17
ª The Author(s) 2020 Article reuse guidelines:
xxxxxxx.xxx/xxxxxxxx-xxxxxxxxxxx DOI: 10.1177/1356766720969747
Abstract
Destination familiarity is thought to critically influence tourists’ decision-making processes. Yet the role of familiarity in shaping tourists’ and residents’ image of, and loyalty to, a destination remains uncertain. This research tests a complex and holistic model of familiarity, affective, cognitive and overall images, and the conative behavioural intentions of visiting and recommending the destination for both resi- dents and visitors in the context of the emerging tourism destination of Molise, Italy. The results reveal that residents and visitors differ in terms of their familiarity and intention to visit a place, with familiarity being less likely to influence residents’ intentions. There is heterogeneity between residents and visitors’ affective image and intention to visit, as well as between their overall image and intention to recommend Molise. Hence, unlike visitors, residents are more likely to respond to factual cognitive imaging, rather than emotional messaging, suggesting that shifting residents’ perceptions of place image requires a different approach to that of visitors. Future research should seek to confirm the relation- ships in a multi-destination study.
Keywords
Destination familiarity, destination image, emerging destinations, intention to recommend, intention to visit, loyalty
Introduction
Tourism is one of the fastest growing sectors of the global economy and competition between destina- tions has intensified with an upsurge in new entrants globally. The highly competitive environ- ment requires deep understanding of the funda- mental drivers of destination image and how this translates into loyalty behaviours, particularly like- lihood to visit and recommend the destination (Xxxxxxxx et al., 2018; Xxxx and Xxxx, 2014; Wea- ver and Xxxxxx, 2011). Visitors’ destination image is often formed from various information sources they are exposed to, as well as their prior knowl- edge and experience of the destination. This means
that destination image can be influenced by sec- ondary sources, such as destination websites, travel guides, the internet and social media (Sharifpour et al., 2014; Xxxxx et al., 2015), as well as destina- tion familiarity, which arises from their exposure to education, travel guides, mass media, and per- sonal contact with other individuals knowledge- able about the destination (Gursoy, 2011).
Corresponding author:
Xxxx Xxxx Xxxxxx, School of Management, Xxxxxxxxxx Xxxxxxxxxx xx Xxxxxxxxxx, Xxxxxxxx, XXX 0000, Xxxxxxxxx.
The literature purports that destination famil- iarity is central in shaping tourists’ decision- making regarding a destination (Xxxxxxx et al., 2017; Xxxx and Xxxx, 2013; Xxxxxx et al., 2018; Xxxxxxxxxx et al., 2014). For example, Xxxxxxx (2001) finds that greater familiarity leads to more positive images of a destination. Familiarity causes a person to develop new thoughts and feelings about the destination that can reshape their image of the destination and sense of place (Xxxxxxx et al., 2006). Conse- quently, a visitor who is personally familiar with a destination may have certain opinions and beliefs about a destination and be less likely to draw on secondary sources of information, meaning critical marketing messages are not received or need to be presented differently to these visitors. Importantly, familiarity can criti- cally affect an individual’s destination choice (intention to visit) and their word-of-mouth behaviour (intention to recommend) (Xxxxxxx et al., 2012; Xxxx, 2012).
Prior research considers differences between visitors and residents perceptions of a destination (Xxxxxxx et al., 2014; Xxxxxxxx and Xxxx, 2010) and their attitude and behaviours relating to the destination (Xxxx et al., 2011; Xxxxxxxx-Xxxxx and Xxxxxx, 2012). Residents are often very familiar with a destination and perceive the des- tination differently (Xxx and Xx, 2016). Indeed, both residents and visitors exist along a familiar- ity spectrum, with varying degrees of familiarity that manifest in different perceptions and beha- vioural intentions with regards to a particular destination. So, familiarity likely explains differ- ences between visitors and residents’ perceptions and behaviours. In the tourism literature, famil- iarity has been defined in a variety of ways (Baloglu, 2001) and intersects with a variety of other concepts, such as experience, awareness and prior knowledge (Sharifpour et al., 2014). While originally conceptualised as a prior visit or number of prior visits (Xxxxxx and Pizam, 1995; Sun et al., 2013), the literature now acknowledges that familiarity is a multi- dimensional construct, including experiential, informational, self-described, primate, self- assured, educational, and expected familiarity (Prentice, 2004). Indeed, researchers argue that familiarity does not only originate from experi- ence, but also from the information search, edu- cation and networks.
Consequently, this research aims to study the role of destination familiarity in shaping destina- tion image and loyalty, as measured through
intention to recommend and visit Molise. The Molise region in Southern Italy is the smallest and youngest Italian region. The region is around 4460 km2 in size and had a population of 310,449 in 2017 (ISTAT, 2018). Molise has vast tourism potential as, to date, the region has not been aggressively marketed. According to the Italian National Institute of Statistics (ISTAT, 2018), Molise has the lowest share of overnight visitor arrivals of all Italian regions. Unlike some regions in central-northern Italy that attract large flows of cultural tourists, the Molise territory has a peripheral position in southern Italy and relies on 3S tourism. As a result, residents are able to undertake day and overnight trips within their own region, but they also have an opinion of the place within which they live. This provides an ideal context to consider the relationship between destination familiarity, image and xxx- alty from the perspective of both the visitor and the resident. The model presented in this paper is more complex and holistic than those previously presented in the literature as it compares multiple types of respondents (non-residents and resi- dents) and tests multiple relationships, including the relationships between destination familiarity, affective, cognitive and overall images, and the conative behavioural intentions of visiting and recommending the destination.
The role of familiarity in shaping destination image
Destination familiarity is defined as one’s ‘abil- ity to describe or even map a place based on images, memories and perceptions of locations, size, distance, physical attributes and site experi- ences’ (Xxxxxxx et al., 2009: 25). As familiarity is enhanced by frequent visits or a period of resi- dence, it can develop cognitive and affective images more likely shaped by experiences for residents and information for visitors (Manyiwa et al., 2018). The early literature on destination familiarity tended to place familiarity and novelty at opposite ends of a spectrum (Xxxxx, 1972; Xxxxxxxxx, 1987). But while novelty has been a central topic in the tourism literature, familiarity has been relatively disregarded (Pre- ntice, 2004). However, increasingly studies are finding that familiarity influences the tourism decision-making process, particularly because it means that visitors may not undertake an infor- mational search (Xxxxxx and XxXxxxxx, 2004). Familiarity has been equated to the sum of infor- mational familiarity, combined with experiential
familiarity (Xxxxxxx, 2001; Xxxx and Xxx, 2012). Often researchers find that visitors with greater familiarity of a destination have a more positive overall image of the destination (Xxxxxxx, 2001; Xxxx and Xxx, 2012; Xxxxxxxx, 2004) and greater intention to visit (Xxx and Xx, 2016). The pos- itive relationship between familiarity and favour- ability has been explained by the destination choice-sets model (Xxxx and Xxx, 2012; Xxxxxxxx and Xxxxxxxx, 2000), which suggests that people funnel down their potential destinations through intensive information-processing resulting in vis- itors selecting from a choice-set of familiar and favourable destinations. Xxxx et al. (2017) sug- gest that familiarity creates a persistent image that is difficult to alter through marketing.
The importance of cognitive attributes in destination image formation can vary based on a person’s knowledge about the destination. Thus, destination familiarity can explain differ- ences in behaviour between residents and tourists because it reflects ‘a key marketing variable in segmenting and targeting certain groups and developing a marketing action plan, including product, distribution, pricing and promotion decisions’ (Baloglu, 2001: 127). Studies by Xx Xxxxx et al. (2015) and Xxxxx and Xxxxxxx (2014) find that familiarity generally improves destination image for visitors. However, Xxxxxx et al. (2011) found a non-significant positive effect of familiarity on destination image, sug- gesting that the relationship may not always hold. Xxxx et al. (2017) found that people’s eva- luation of a destination is influenced by their general (and often stereotypical) images, as well as their tourism specific images, and that this is influenced by their familiarity with the destina- tion (with greater familiarity leading to more positive images and evaluations).
Destination image is a multidimensional con- struct (Xxxxxx et al., 2013) that is the sum of ideas, impressions and beliefs people have of the various attributes, aspects and activities of a des- tination (Xxxxx et al., 2014). Many recent studies measure overall destination image with cognitive image and affective image (e.g., Xxxxxxxx et al., 2015; Xxx, 2018; Xxxxxxxxx et al., 2018; Xxxxxx- Xxx and Mart´ın-Xxxxxxx, 2015; Xxxxxxxx et al., 2017). For this study, overall image is composed of the cognitive image, which consists of beliefs and knowledge about a destination’s attributes (Xxxx and Xxxx, 2004), as well as the affective image, which represents feelings about a destina- tion (Xxx and Xx, 2016). Cognitive images are often measured using catalogues of functional
and psychological attributes, while affective images are measured via an affective grid scale (Xxxxxx and Xxxx, 2012; Xxxxxx et al., 1981). Generally, tourists need a positive image of a destination to consider visiting (Xxxxx et al., 2013; Xxxxxxxxxx and Nunkoo, 2011). Finally a third interrelated component of destination image is the conative or behavioural image, which influences and predicts tourists’ behaviour (Xxxxxxx et al., 2018).
Several studies consider residents and visitors place attachment, finding key differences in per- ceptions and images of a destination between residents and visitors (Braun et al., 2013; Xxxxxx and Xxxxxxxx, 2013). The literature finds that both residents and visitors establish place attach- ment and form images of the destination and its attributes (Xxxxx and Xxxxx, 2008; Xx and Xxxx, 2008). Thus, there is a need to more fully under- stand the link between destination image and resident and visitors’ attitudinal and behavioural intentions (Xxxxx et al., 2001; Xxxxxxxx et al., 2017). Importantly, residents’ perceptions and attitudes towards tourism are pivotal for tourism planning as they market their place to others and provide support for tourism development activi- ties (Stylidis et al., 2017). Hence residents’ des- tination image can influence tourists’ image formation, decision-making and purchasing behaviour, as well as destination development (Xxxxx et al., 2001; Xxxxx et al., 2011). Yet Sty- lidis et al. (2017) argues that few studies compare the images of tourist destinations formed by res- idents to those held by tourists, even if significant differences between the two groups have been found (Henkel et al., 2006).
Prior studies (i.e. Xx and Xxxxxxx, 1993; Mil- man and Pizam, 1995) explain differences in per- ceived image using tourists’ previous experience, with destination familiarity usually leading to positive images of place (Baloglu, 2001). Residents often have more accurate per- ceptions of, and stronger attachment to, their place of residence (Xxxxxxxx et al., 2017). Yet there is little understanding of how residents’ familiarity with a place differs from tourists and how this influences their cognitive and affective images, and their behavioural intentions. Cul- tural factors and place attachment often underlie differences in an individual’s perceptions of a destination and influence destination loyalty (Xxxxxx and Xxxxxx, 2004; Xx and Xxxx, 2008; Imada and Xxxxxxxxx, 0000; Xxx, 2018). There is also a dynamic interplay between residents’ per- ceptions of place and their perceptions of the
impacts of tourism, which can thereby influence their own travel behaviour (Su et al., 2016). Ela- borating on this concept further, Xue and Xxxxx (2020) discuss the role of distance in influencing tourist behaviour between long-haul, short-haul and local travellers, particularly their motiva- tions, travel patterns and willingness to pay.
Henkel et al. (2006) compared the perceptions of Thai residents with international visitors finding several significant differences relating to terrorism and disease, whereby visitors were more concerned than residents about the potential threats. This sug- gests that resident’s familiarity shapes their image of the destination. Similarly, Xxxxx et al. (2013) pur- port that residents are more likely to view a destina- tion positively and become more emotionally attached as they are more closely associated with the destination. Likewise, Xxxxxxx et al. (2018) investigate differences in resident and visitors’ per- ceptions of image and emotional attachment to a destination finding that affective and cognitive image positively influence emotional attachment confirming findings of Xxx et al. (2015). In fact, Xxxxxxx et al. (2018) find that affective image has a larger effect on emotional attachment for residents than for visitors’, but cognitive image has a similar impact. Hence, prior research suggests that while residents and visitors may have similar images of a destination, familiarity plays a role in determining how wide the gap between the two groups percep- tions and images.
The literature recognises that consumer loyalty is a more important factor than satisfaction when it comes to strategic marketing as it is a better pre- dictor of behaviour (Chi and Qu, 2008; Xxxxxx, 1999). Consumer loyalty indices tend to be com- posed of behaviour-based measures, including intention to visit, repeat visit and recommend a destination, but also overall satisfaction (Xxxxxx, 1998). There is substantial literature arguing that overall image of a destination impacts intention to revisit a destination (Qu et al., 2011). Indeed, more broadly many studies find that destination image is an antecedent of loyalty (Xxxxxxxx et al., 2008; Xxxxxx and Xxxx, 2012). Some tourism researchers equate place attachment to increased loyalty (Xxxx and Xxx, 2014; Xxxx, 2012). Others also argue that destination evaluations are a predictor of travel propensity (Xxxxxxx and XxXxxxxx, 1999; Xxxxxx and Mart´ın, 2004; Xxx et al., 2012). Moreover, the strength of the relationships between image and future behavioural intentions can vary depending on the context (San Mart´ın et al., 2013) and per- sonal characteristics (Xxxxxx, 2009). Yet XxXxxxxxx et al. (2012) point out that loyalty has
been studied quite simplistically in the literature, often via similar indicators using single case stud- ies. Instead they argue that a consumer can be hor- izontally loyal, whereby the tourist is loyal to several tourism suppliers or destinations. While there have been significant studies into tourist xxx- alty, very few consider multi-destination loyalty, with the exception of Xxxxxxx-Xxxxxxx and Xxxxxx-Xxx (2018). Moreover, factors that may reduce loyalty and intention to visit have been under-examined with Xxxx and Xxxxxxx (2016) suggesting the need to consider the effect of limited time, money and travelling companion(s).
Prior research shows that loyalty is influenced by income and gender (Xxxxxxx, 2005; Xxxxx et al., 2008). For example, older travellers and those with lower incomes are more likely to revi- sit (Xxxxxxx et al., 2015). Moreover, Xxxxxxxxx et al. (2018) find that tourist involvement, or perceived relevance of the destination, positively impacts cognitive and affective impact leading to intention to visit. Loyalty is also influenced by frequency of holidaying, with those who travel more being more loyal, although this also increases horizontal destination loyalty (Xxxxxxx-Xxxxxxx and Xxxxxx-Xxx, 2018). Xxxxxxx-Xxxxxxx and Xxxxxx-Xxx (2018) find that sea, sun and sand (3S) destinations are nega- tively related to loyalty, possibly due to the num- ber of these types of destinations resulting in them being easily substitutable. They find that affective image is negatively related to destina- tion loyalty, particularly for 3S destinations, as the type of destination has a more generic image thereby reducing loyalty. Put another way, the more loyal a visitor, the lower their affective image, but destinations with unique attributes encourage loyalty. Hence the literature supports the need to consider the complex interplay between familiarity, distance and loyalty and their image on an individual’s destination image and subsequent behavioural intentions.
Theoretical model
Many tourism studies consider familiarity, image, satisfaction and loyalty individually, but often fail to establish the relationship between the various factors (Chi and Xx, 2008). Xxxxxx and Xxxxxx (2004) and Xxxx´ıguez Xxxxxx et al. (2013) argue that better evaluating a destination attributes does not always lead to better destination image, as the significance of the dimensions vary among market segments. To further explore this aspect, we developed our model distinguishing between
Services
Environment
H8
H1
Cognitive Image
H4
H6
Intention to Recommend
Destination Familiarity
H3
Overall Image
H10
H7
H2
H5
Intention to Visit
Affective Image
H9
Emotional
Original
Figure 1. Research model.
non-residents and residents. Hypothesised effects among destination familiarity, cognitive image, affective image, overall image, intention to rec- ommend and intention to visit the destination are collected in Figure 1.
Formally, our hypotheses are:
●
H1 and H2: destination familiarity has positive effects on cognitive and affective images of Molise.
●
H3: Cognitive image predicts affective image of the destination (both are second-order factors reflected by first- order factors: service, environment, emo- tional, and original).
●
●
●
H4 and H5: Cognitive and affective images predict overall image of the place. H6: Overall image of the destination pre- dicts the intention to visit the destination. H7: Overall image of the destination pre- dicts the intention to recommend Molise to others.
●
H8 and H9: the effects of destination familiarity on the two outcomes are par- tially mediated through the perceptions of destination images.
●
H10: Intention to visit and intention to recommend are both related to each other.
Method
Study area and sample
In October 2017, a mixed-mode sample (McLen- nan et al., 2014) of Italians completed a three-
part on-line questionnaire in Italian. Initial respondents were selected from 500 students of the local University using email invitation. Con- currently, snowball sampling (Xxxxx et al., 2007) was employed whereby the students were asked to forward the survey to their friends, rela- tives and contacts (18 years or older) who live in Italy. Concurrently, the survey was distributed on Facebook as social media platforms are very popular and increasingly replacing other forms of communication, including emails, surface mails and telephone (Phan and Airoldi, 2015). This mixed-mode referral approach is inexpen- sive and efficient in providing researchers with an increasingly-expanding set of respondents (Xxxxxxxxxx et al., 2009). The choice of online survey based on social media is accentuated by its reach, recruitment of stigmatised and ‘hard- to-reach’/’hard to involve’ populations (Baltar and Brunet, 2012), and cost reduction (Xxxx and Xxxxxxx, 2000). Although a limitation of the approach is the possibility of missing, or being biased against, certain groups within the popula- tion, such as older people who are less inclined to use social media. However, as age is not a key consideration for this study, this possible bias is considered to have a relatively minimal impact on our overall analysis.
Among 1,091 respondents to the question- naire survey, 116 respondents (10.6% of the sam- ple) did not answer at least a quarter of the questions (excluding demographics). These low engagement respondents were removed from fur- ther analysis. The remaining sample included 975 respondents with an overall 4.48% of scale
¼
¼ ¼
item values missing not completely at random (Little’s MCAR test: χ2 26223.06, df 24276, p < .001). With the assumption of con- ditionally random missing data, the missing val- ues were then imputed using the expectation– maximisation (EM) algorithm in SPSS v.25. No significant multivariate outliers were identified in the sample based xx Xxxxxxxxxxx distances on the items. The final sample (n 975) was composed of 295 Molise residents and 680 non-residents living in Italy – or 975 respondents in total. Such sample sizes are sufficiently large for the models to test in this study, which usually require a minimum of 300 to 500 observations (Hair et al., 2013: 574). The two sample groups shared similar demographic profiles in terms of gender, age, marital status, level of education, type of occupation, and income level (Table 1).
Instrument and measures
The questionnaire comprised two sections, with the first being general socio-demographic char- acteristics (e.g., gender, age, education, etc. in Table 1) and the second measuring the constructs involved in the hypothetical model as illustrated in Figure 1. The aim was to investigate residents’ and non-residents image perceptions of, and familiarity with, Molise, as well as their intention to visit and recommend Molise as a holiday des- tination. For the purposes of this research and the questionnaire implemented in this study, the def- inition of a resident was an inhabitant of the Molise region, while a visitor is someone from outside of Molise.
A multi-factor scale of destination image was implemented because image is a complex and multifaceted concept, which follows prior research that has offered a number of destination image dimensions (Xxx et al., 2007). Drawing on the literature (Xxxxxxx et al., 2014; Xxxxxxxx et al., 2017; Xxxx and Hsu, 2010), multiple aspects of cognitive image were considered: natural envi- ronment, amenities, accessibility and social envi- ronment. The items reflecting two dimensions, services and environment, were rated on a five- point Likert-type scale, ranging from ‘1’ (poor) to ‘5’ (excellent). Based on prior studies the attri- butes selected to evaluate the affective image comprised the following five-point bipolar semantic differential items such as chaotic- tranquil and boring-intriguing (Xxx and Xxxxxxx- son, 2003; Xxx Xxxxxx and xxx Xxxxxx, 2008; Xxxx and Xxx, 2010). Overall destination image was measured using a single-item on a five-point
Likert-type scale from ‘1’ (very dissatisfied) to ‘5’ (very satisfied), following Xxxxxx and Xxxxxx (2004) and Xxxx and Hsu (2010). The set of items relevant to each image dimension was con- structed by focusing on ‘universal attributes’ (i.e., scenery, weather, accommodation) and excluding those not relevant to the context of Molise, based on a critical review by a select number of residents and tourists.
Familiarity with the tourism area was mea- sured with three items of perceived knowledge of Molise, on a five-point Likert rating scale of agreement ranging from ‘1’ (strongly disagree) to ‘5’(strongly agree). Respondents were also asked to rate their intentions to recommend Molise as a holiday destination and to spend holi- days in Molise in the next 24 months, each on a five-point Likert-type scale ranging from ‘1’ (no) to ‘5’ (yes). Table 2 presents the key variables with descriptive statistics, with the items of the latent construct scales being presented in Appen- dix 1. The questionnaire was piloted with 20 residents and tourists to verify instrument valid- ity. No major concerns were reported in the pilot.
Analysis strategy
This study employed multi-group structural equation modelling (SEM) to statistically test the afore-mentioned hypotheses regarding the moderation role of tourist identity (i.e., resi- dents vs. non-residents) in the relationships between destination familiarity, destination image, and tourist intentions (both intention to visit and to recommend). Initially, descriptive statistics of all items were calculated separately for xxx xxxxxxxxx’ and non-residents’ sample. Univariate normality of the items was examined because it was a prerequisite for the use of the SEM framework. Items with five ordinal levels and absolute values of skewness and kurtosis lower than three were considered to generally follow normal distributions, supported by Xxxxx (2016) and Hair et al. (2013) who argued that moderate departures from normality do not sig- nificantly impact the results of SEMs when sample sizes are sufficient, as was the case in this study.
Second, the various constructs were assessed in both samples (residents and visitors) for relia- bility and validity using confirmatory factor analysis (CFA) in SEM framework. The first- order CFA was verified first, followed by the second-order CFA model, where two seconder order factors – cognitive image and affective
Table 1. Demographic profiles of local resident and visitor samples.
Demographics | N | % | N | % |
Gender* | ||||
Female | 141 | 47.8 | 376 | 55.3 |
Male | 153 | 51.9 | 303 | 44.6 |
Age Group | ||||
18–24 | 67 | 22.7 | 167 | 24.6 |
25–34 | 76 | 25.8 | 162 | 23.8 |
35–44 | 76 | 25.8 | 184 | 27.1 |
45–54 | 41 | 13.9 | 101 | 14.9 |
55–64 | 29 | 9.8 | 49 | 7.2 |
65+ | 6 | 2.0 | 17 | 2.5 |
Single | 170 | 57.6 | 377 | 55.4 |
Married with young children living | 57 | 19.3 | 146 | 21.5 |
at home | ||||
Married with no children or | 64 | 21.7 | 147 | 21.6 |
older children no longer living | ||||
at home | ||||
Education Level | ||||
Middle school diploma, or lower | 10 | 3.4 | 13 | 1.9 |
High school diploma | 108 | 36.6 | 169 | 24.9 |
Post-graduate education | 42 | 14.2 | 158 | 23.2 |
(master – PhD) | ||||
Master’s degree | 77 | 26.1 | 163 | 24.0 |
3-year degree | 30 | 10.2 | 78 | 11.5 |
I am university student | 28 | 9.5 | 99 | 14.6 |
Occupation | ||||
Executive | 12 | 4.1 | 26 | 3.8 |
Professor | 24 | 8.1 | 88 | 12.9 |
Employee | 65 | 22.0 | 166 | 24.4 |
Entrepreneur | 13 | 4.4 | 21 | 3.1 |
Freelance | 50 | 16.9 | 86 | 12.6 |
Worker | 16 | 5.4 | 22 | 3.2 |
Unemployed | 38 | 12.9 | 69 | 10.1 |
I am retired | 9 | 3.1 | 14 | 2.1 |
Student | 68 | 23.1 | 187 | 27.5 |
Income* 0€–19,999€ | 99 | 33.6 | 206 | 30.3 |
20,000€–39,999€ | 91 | 30.8 | 217 | 31.9 |
40,000€–59,999€ | 36 | 12.1 | 97 | 14.2 |
60,000€–79,999€ | 7 | 2.3 | 37 | 5.4 |
80.000€–99.999€ | 3 | 1.0 | 22 | 3.2 |
100,000€ or more | 7 | 2.4 | 14 | 2.1 |
n.a. | 43 | 14.6 | 74 | 10.9 |
Locals (N = 295) Visitors (N = 680)
Marital Status*
*Having a trivial amount of missing values.
image – were tested. The SEM models in this study were statistically evaluated by multiple most commonly used absolute, incremental, and parsimony global goodness-of-fit indices (cut- offs) (Xxxxx, 2016; Hair et al., 2013; Xxxxx, 2012), including normed Chi-square (the ratio of Chi-square statistic to the degrees of freedom for a model, 1 < χ2/df < 3), root mean square error of approximation (RMSEA < .05), the
upper limit of 90% confidence interval of RMSEA (90% UL < .10), comparative fit index (CFI > .95), Xxxxxx-Xxxxx Index (TLI > .92), parsimony normed fit index (PNFI > .50), and Akaike information criterion (AIC, lower value preferred). Measurement reliability was examined by the scale’s Cronbach’s alpha (a) coefficient and composite reliability (CR) with .70 as the cut-off value; convergent validity was evaluated with the
Table 2. Descriptive statistics of measures in both local and visitor samples.
Residents (N = 295) Non-residents (N = 680)
Scale items M SD Skew Xxxx Alpha M SD Skew Xxxx Alpha
Destination Familiarity .94 .94
— —
DF1 4.14 1.08 1.11 .41 2.98 1.44 .04 1.32
— — —
DF2 3.83 1.12 .79 .07 2.62 1.34 .33 1.08
— — —
DF3 3.74 1.13 .70 .21 2.56 1.31 .41 .95
Original .77 .81
— —
Original3 4.34 .98 1.83 3.37 4.10 1.02 1.16 1.06
— —
Original2 4.41 .99 2.07 4.09 4.29 .96 1.59 2.50
Emotional .78 .78
— — — —
Emotional4 3.04 1.18 .09 .80 2.88 1.07 .05 .54
— — — —
Emotional2 3.06 1.08 .09 .44 2.97 .97 .02 .05
— —
Emotional1 4.04 .94 .94 .75 3.74 .93 .48 .25
Services .87 .91
— —
Attraction3 2.40 1.14 .49 .51 2.76 1.02 .15 .48
— —
Amenities2 2.45 1.09 .32 .72 2.66 1.04 .23 .36
— —
Amenities3 2.69 1.11 .26 .65 2.76 1.03 .19 .44
— —
Amenities1 2.60 1.18 .40 .56 2.58 .99 .37 .10
Environment .90 .93
Natural1 3.93 1.17 —.83 —.35 3.18 1.16 —.04 —.83
Natural2 3.95 1.13 —1.02 .34 3.62 | 1.10 —.57 —.31 | ||||
Natural3 4.17 1.09 —1.14 .36 3.70 | 1.15 —.56 —.55 | ||||
Accessability3 3.93 1.19 —.96 —.03 3.50 | 1.16 —.34 —.71 | ||||
Accessability4 4.18 1.02 —1.16 .66 3.69 | 1.14 —.59 —.44 | ||||
SocialEnv1 4.10 1.07 —1.05 .27 | 3.68 | 1.23 —.56 —.70 | |||
SocialEnv2 3.73 1.26 —.77 —.39 | 3.46 | 1.20 —.35 —.77 | |||
Overall image | 3.57 | 1.07 —.53 | —.15 3.46 | .87 —.48 .33 | |
Intention to recommend | 3.44 | 1.32 —.43 | —.89 2.59 | 1.18 .42 —.63 | |
Intention to visit | 2.71 | 1.42 .29 | —1.20 2.42 | 1.34 .67 —.71 | |
M: mean; SD: standard deviation; Skew: skewness; Xxxx: kurtosis; Alpha: Xxxxxxxx’x a.
factor’s average variance extracted (AVE) with .50 as the minimum; and discriminant validity was confirmed when a factor’s square root of AVE was stronger than the factor’s correlations with other factors (Xxxxxxx and Xxxxxxx, 1981).
Further, to reach valid comparison of the hypothesised relationships between two groups, the scales must measure identical con- structs across different groups, namely mea- surement invariance. It means that the respondents across groups interpret the indi- vidual items, as well as their underlying latent factors, in the same way (Van de Schoot et al., 2012). This study first assessed configural invariance, which required theoretically oper- ationalised factor structure to be the same for two groups of respondents, whereas the values of parameter estimates could vary. This model and the parameters estimated in the model were used as the baseline for comparing other more restrictive models. For both groups the items for their cognitive domains were the same (Xxxxxx and Rensvold, 2002).
With configural invariance established, the CFA model was further constrained with equal factor loadings between groups, i.e. metric invar- iance. This was testing whether the two groups attributed the same meaning to the latent con- struct (Xxx xx Xxxxxx et al., 2012) or the con- structs manifested in the same way across groups (Xxxxxx and Rensvold, 2002). Full metric invariance is difficult to satisfy in practice, and some researchers (e.g., Xxxxx, 0000) propose par- tial invariance, suggesting that if non-invariant items constitute only a small part of the model then cross-group comparisons are still relevant and meaningful. The classical approach to testing for multi-group invariance in SEM is χ2 differ- ence test, which is an excessively stringent test of invariance (Xxx xx Xxxxxx et al., 2012). Xxxxxx and Xxxxxxxx (2002) proposed a practical criter- ion for evidence of invariance, namely ∆CFI <
.01. Xxxx (2007) also recommended that ∆CFI <
.01 and ∆RMSEA < .015 for tests of invariance. This paper presents testing results using both χ2 test and change in CFI.
Table 3. Fit indices and tests for multi-group confirmatory factor analysis model and structural equation models.
Model | Multi-Group Specification | w2/df | RMSEA | 90% UL | CFI | TLI | PNFI | AIC | Dw2 | Ddf | p | DCFI |
CFA | Configural invariance | 2.175 | .035 | .039 | .978 | .969 | .669 | 801.618 | – | – | – | – |
Full metric invariance | 2.196 | .035 | .039 | .977 | .969 | .712 | 809.849 | 40.231 | 16 | .001 | .001 | |
Partial metric | 2.140 | .034 | .038 | .978 | .970 | .702 | 795.117 | 17.499 | 12 | .132 | .000 | |
invariance | ||||||||||||
SEM | Free structural paths | 2.358 | .037 | .040 | .957 | .940 | .677 | 1569.492 | – | – | – | – |
Full constrained paths | 2.351 | .037 | .040 | .955 | .941 | .696 | 1571.654 | 32.162 | 15 | .006 | .002 | |
Partial Constrained | 2.327 | .037 | .040 | .956 | .942 | .693 | 1558.955 | 13.463 | 12 | .336 | .001 |
paths
w2/df: Chi-square to the degrees of freedom ratio; RMSEA: root mean square error of approximation; 90% UL: the upper limit of 90% confidence interval of RMSEA; CFI: comparative fit index; TLI: Xxxxxx-Xxxxx Index; PNFI: parsimony normed fit index; AIC: Akaike information criterion; Dw2: difference in w2 values compared with the unconstrained CFA or SEM model; Ddf: difference in number of degrees of freedom compared with the unconstrained CFA or SEM model; p: significance of the Chi-square likelihood ratio test; DCFI: difference in CFI values compared with the unconstrained CFA or SEM model.
Lastly, with measurement invariance estab- lished, the cross-group equality constraints were applied to the hypothesised ‘causal’ relationships to test the moderation effect of tourist identity. All ‘causal’ paths were freely estimated for each group without any equality constraints, and then all were constrained to be equal across groups. Given that the fully constrained model has sta- tistically lower fit than the model with no con- straints, a set of increasingly constrained SEM models were tested to locate the paths that were significantly different between residents and vis- itors. Demographic variables were included as controls. All descriptive statistics were calcu- lated using SPSS v.25 and all CFA and SEM models were estimated using AMOS v.25.
Results
Descriptive statistics of measures
Descriptive statistics (Table 2) and distribution visualisation affirmed univariate normality of the items in each of the two samples, which allowed for a plausible assumption of multivariate nor- mality in the data for further CFA and SEM. All five scales also showed high internal consistency with Cronbach’s a exceeding .77 (Table 2).
Measurement invariance between residents and non-residents
= = =
The good fit of second-order CFA model (χ2/df 2.175; RMSEA .035; 90%UL .039;
=
CFI .978) verified that the construct scales dis- played theorised factor structure – cognitive image and affective image as seconder-order factors each being reflected by two first-order destination image
factors (Figure 1). As shown in Table 3 (CFA mod- els), full metric invariant CFA model (χ2/df 2.196; RMSEA .035; 90%UL .039; CFI
= = =
=
= = = =
.977) was found to have significantly less goodness of fit (∆χ2 40.231, ∆df 16, p .001; ∆CFI
= = =
=
.001) than the baseline configural invariant model. Therefore, partial metric invariance (χ2/df 2.140; RMSEA .034; 90%UL .038; CFI .978) was
= = =
sought and reached (∆χ2 17.499, ∆df 12, p
=
.132; ∆CFI .000; lowest AIC) when freeing the factor loadings on four items DF3, Aminities_1, Accessibility_4, and SocialEnv_1). Standardised factor loadings of partial metric invariance CFA are shown in Appendix 1.
The construct reliability and validity of all first- and second-order factors were assessed based on the results of the partial metric invar- iance CFA model. CR values over .70 imply the shared variance between each construct and its indicators is greater than the error variance (Table 4). The AVE values over .50 suggests the quantity variance derived from the indicators, was higher than the quantity variance due to measure- ment error. The square root of each AVE was greater than the inter-construct correlations, sug- gesting that the constructs differed from each other. In summary, the results suggest that the constructs are reliable and valid in each group.
Structural differences between residents and non-residents
With measurement invariance being confirmed, the hypothesised ‘causal’ relationships were tested with multi-group SEM models. The
full cross-group equality constrained model (χ2/df = 2.351; RMSEA = .037; 90%UL =
.040; CFI = .955) was found to have
Table 4. Construct reliability and validity for first-order and second-order factors in local sample (N = 295) and visitor sample (N = 680).
Locals Visitors Correlations
Factor | CR | AVE | CR | AVE | OR | EM | SV | EV | DF | CI | AI | ||
Original (OR) | .78 | .64 | .82 | .69 | .80/.83 | .47 | .21 | .40 | .26 | – | .41 | ||
Emotional (EM) | .75 | .50 | .76 | .51 | .54 | .71/.72 | .55 | .70 | .44 | – | .89 | ||
Services (SV) | .88 | .64 | .91 | .72 | .11 | .60 | .80/.85 | .71 | .22 | .85 | – | ||
Environment (EV) | .91 | .58 | .93 | .64 | .26 | .57 | .59 | .76/.80 | .56 | .82 | – | ||
Destination Familiarity (DF) | .94 | .83 | .94 | .83 | .15 | .26 | .21 | .55 | .91/.91 | .67 | .40 | ||
Cognitive (CI) | .77 | .63 | .82 | .70 | – | – | .80 | .78 | .70 | .79/.83 | .79 | ||
Affective (AI) | .88 | .81 | .74 | .64 | .59 | .90 | – | – | .21 | .62 | .90/.80 |
CR: composite reliability; AVE: average variance extracted.
=
Note: Correlations for locals sample below the diagonal (for visitors sample below the diagonal), the square root of AVE for locals/visitors on the diagonal; all correlations were statistically significant (p < .001) except for those between OR and SV (p
=
.12) and between OR and DF (p .02) in locals sample; second-order factors including cognitive image and affective image displaying their standardised loadings on relevant first-order factors.
= = =
=
significantly lower goodness of fit (∆χ2 32.162, ∆df 15, p .006; ∆CFI .002) than
= = =
=
the freely estimated model (χ2/df 2.358; RMSEA .037; 90%UL .040; CFI .957).
= = =
=
Partial cross-group equivalence (χ2/df 2.327; RMSEA .037; 90%UL .040; CFI .956) in
= =
= =
the paths was achieved (∆χ2 13.463, ∆df 12, p .336; ∆CFI .001; lowest AIC) after a series of tests (Table 3, SEM models). Finally, three paths were found to significantly differ between residents and non-residents (dashed paths in Figure 2), specifically, the effect of familiarity of place onto affective image; the direct effect of destination familiarity onto inten- tion to visit; and, the effect of overall image onto intention to recommend the destination.
Discussion
This research empirically tests the relationship between destination familiarity, destination image and the behavioural intentions of visiting and recommending the destination and juxta- poses the results across residents and tourists in the context of the emerging tourist destination of Molise, Italy. The results support all the overall tested hypotheses, with one exception: H4, which states that cognitive image predicts overall image of the place. Instead of predicting overall image, cognitive image partially mediates the relationship between destination familiarity and affective image. To confirm that cognitive image predicts affective image we applied a sensitivity test by reversing the direction of the relationship and testing the alternative structural model that affective image predicts cognitive image on the H3 path. We found that the alternative model had
significantly worse fit than that of hypothesised model in Figure 1 and its H3 path was not statis- tically significant thereby confirming the litera- ture that cognitive image predicts affective image (Xxxxxx et al., 2011). Thus, a person requires a cognitive image, prior to forming an affective (or emotional) image of a destination. This is intuitive, as generally it would take time to understand and develop an emotional image of a destination. Hence, this research supports and extends the findings of Chi and Xx (2008) sug- gesting a positive relationship between destina- tion familiarity, image, satisfaction and loyalty. Xxxxx et al. (2017) also found a complicated relationship between the characteristics of a des- tination’s slogan and a visitor’s destination familiarity, attitudes and intentions to visit. This is important as it suggest that information expo- sure, as well as primary (own) and secondary (others) experiences, form a person’s familiarity that then influences a person’s cognitive and affective images and ultimately determines their loyalty to a destination.
Xxxx and Xxxxxx (2010) and Xxxxx et al. (2010) argue that it is important to understand whether visitors and residents have congruency in their perception of a destination as it reflects their value systems and ultimately impacts their perceptions of tourism impacts. The present study finds several significant differences in the relationship between visitors and residents for Molise, Italy. Firstly, H2 varies significantly by residents and visitors, suggesting that destination familiarity has stronger negative effect on affec- tive image for residents. The most loyal of all people in a destination are residents and being a resident reduces the novelty of a destination,
R: .378*** NR: .520***
Service
Environment
Intention to Recommend
R: .801*** NR: .852***
R: .784*** NR: .815***
R: .540*** NR: .629***
Cognitive Image
R: -.037 ns.
NR: -.048
ns.
R: .228*** NR: .069**
Destination familiarity
L: .729*** V: .910***
Overall Image
R: .520*** NR: .505***
R: -.238*** NR: -.206***
Affective Image
R .500*** NR: .518***
R: .091** NR: .080**
R: .901***
NR: .889***
R: .592***
NR: .408***
Intention to visit
Emotional
Original
R: .242*** NR: .539***
Statistically equal relationship between groups
*** p < .001
** p < .05 Statistically varying relationship between
ns. p >.10 groups at the 0.01 level
Groups: Residents (R) vs. Non-Residents (NR) (NR)
Figure 2. Comparison of the hypothesised effects between residents and non-residents.
which may reduce emotional images. Thus, res- idents who are very familiar with a destination are less likely to rely on affective image to make decisions and instead rely more on cognitive image and functionality. This aligns with finds by Xxxxxxx-Xxxxxxx and Xxxxxx-Xxx (2018), who determined that loyalty is negatively related to affective image. However, these findings con- trast to those of Xxxxxxx et al. (2018) who con- clude that affective image has a stronger significant positive effect on emotional attach- ment for residents than visitors, but cognitive image has a similar impact. These contrasting results suggest the need to consider differences between emotional attachment and overall image between residents and visitors in a single multi- destination study. The implications are that mar- keting needs to differ for those very familiar with a destination compared with those less familiar and also that those residents that are promoting a destination to visitors may focus on cognitive images and functionality, rather than affective images hence marketing to visitors should be biased towards affective images to counteract the effect. Prior research has suggested affective image is appropriate for marketing to tourists (Baloglu and Xxxxxxxx, 1997), but that both cog- nitive and affective are appropriate for residents
(Xxxxxxxx et al., 2017). This research, however, finds that only cognitive marketing will be effec- tive for residents.
The group comparisons found that H6 varied significantly between residents and visitors, sug- gesting that residents’ overall image of the des- tination and intention to recommend the destination is stronger than for visitors. This is likely because residents feel more informed and able to recommend a destination than visitors, and they may also be more emotionally attached to the destination leading to a greater desire to recommend the destination; although this needs formal confirmation in future research. Another interesting discovery from the present research is the positive direct effect between familiarity and intention to recommend (H8). This result repre- sents a valuable contribution to the current area of research on intention to recommend, or what is also known as ‘Word of Mouth’ referral, because it suggests that familiarity is a potential influen- tial force for delivering destination recommenda- tions (Papadimitriou et al., 2018). The results indicate that familiarity is a positive antecedent to intention to recommend and the relationship is stronger for visitors, although still positive for residents. Indeed, the results reveal that resi- dents’ behavioural intentions are more likely to
be influenced by their destination image, whereas visitors tend to have a stronger direct relationship between familiarity and intention to recommend and visit a destination.
This study confirms prior research that finds destination familiarity positively affects tourists’ intention to travel (Xxxxxxx et al., 2017). However, this study also discovers a significant moderating effect of tourist identity between familiarity and intention to visit (H9). H9 varies significantly by residents and visitors suggesting that the relation- ship between destination familiarity and the beha- vioural intention of visiting the destination, partially mediated by destination image, is much stronger for visitors than residents. This suggests that familiarity is less likely to influence residents’ intention to visit the destination, which is plausi- ble if they already live there, and also suggests that novelty may play a role in people’s decision-making regarding destination selection. Tourists are likely to place greater weight on emo- tional images as they need to form a closer con- nection when evaluating and selecting a destination because they have greater direct and opportunity costs associated with visiting than a resident. Indeed, tourists are more likely to mini- mise or avoid risks associated with their travel decision-making forcing them to be more careful and dependent on their image and more likely to visit an established, rather than emerging destina- tion (Xxxxxxxxxx et al., 2014; Xxxxx et al., 2017). Lastly, prior research has found that beha- vioural intentions are influenced by gender, age and income (Xxxxxxx et al., 2015; Xxxxxxx, 2005; Xxxxxxx and Xxxxxxx, 2001; Xxxxx et al., 2008), but we found no significant effect for age, gender and income on intention to recom- mend the destination. However, females and older visitors were significantly more likely to intend to spend future holidays in Molise, while income had no effect on intention to visit. Prior research confirms that older travel- lers are more likely to repeat visit (Xxxxxxx et al., 2015), but the other differences suggest demographics are likely contextually linked. We found no between-group differences for gender, age, and income on the intentions, meaning residents and visitors did not differ in terms of the demographic control variables’
effects on behavioural intentions.
Conclusion
This research aimed to understand the role of destination familiarity in shaping destination
image and loyalty for tourists and residents, as measured through intention to recommend and visit the destination. Understanding the differ- ences is important as residents can determine the path of destination development as well as influ- ence visitors perceptions and behaviour. Theore- tically, we find that the relationship between familiarity, image and the behavioural intentions of recommending and visiting the destination are positive, the relationship between destination familiarity and affective image is negative. We find heterogeneity between residents and visi- tors’ affective image and intention to visit asso- ciated with their familiarity with the destination, as well as differences between the groups overall image and intention to recommend the destina- tion. Notably, we find that familiarity is nega- tively related to affective image in the case of Molise, which contrasts to that of Manyiwa et al. (2018) in the context of the emerging des- tination of Bratislava, Slovakia.
Measuring the destination image perceived by tourists and understanding how this relates to res- idents’ perceptions of the destination is essential for the proper strategic management of destina- tions (San Mart´ın and Del Bosque, 2008). In fact, understanding the image of a destination enables destination managers to understand how tourists and residents perceive the destination and identify factors that affect their attitude and behaviour towards the destination (Xxxxxxx and Xxxxxxx, 1993). This is important as it means that tempting visitors to a region requires a different marketing strategy to that aimed at encouraging residents to experience local attractions. This is particularly true as residents are likely to have a strong and persistent image of the destination that is difficult to change and are less likely to pay attention to information that contradicts their pre-existing per- ceptions (Xxxx et al., 2017). Our findings show that residents are less likely to be positively influenced by emotional messaging, but are more likely to respond to factual cognitive imaging. Shifting res- idents’ perceptions of place image is therefore more likely to be effective through community consultation that enables community identified tar- geted infrastructure development, inclusive plan- ning processes and involvement in the imaging campaign, for example through competitions to select logos (Avraham, 2004). Hence, our findings have important strategic implications for destina- tion managers and can guide their use of marketing to enhance both residents and visitors’ image and intention to visit.
The current research is limited to understand- ing similarities and differences between domestic tourists and residents. It would be anticipated that valuable insights into vertical destination loyalty would be elicited by considering international tourists and non-visitors, including potential cul- tural differences, as well as geographic, cultural and touristic distance. A limitation of our study is a relatively low sample size among older residents hence future research could further explore age effects in different contexts. Moreover, a multi- destination study that also considered both verti- cal and horizontal loyalty would be a significant leap forward within the literature. Furthermore, future research should include a mixed-method approach using both quantitative and qualitative analysis to elicit greater theoretical insights. For example, it would enable extending the analysis to other groups of stakeholders, such as the per- ceptions of policymakers and destination market- ers. Lastly, an in-depth exploration of the effect of novelty on vertical loyalty would add value to the literature and identify if marketers should focus on the uniqueness of the destination when pro- moting to various types of people (residents, domestic/international visitors and non-visitors).
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.
ORCID iD
Xxxx Xxxx Xxxxxx xxxxx://xxxxx.xxx/0000-0000- 6889-0615
Xxxxx Xxx xxxxx://xxxxx.xxx/0000-0000-0000- 6724
References
Albayrak T, Xxxxx M, Gonza´lez-Rodr´ıguez MR, et al. (2018) Analysis of destination competitiveness by IPA and IPCA methods: the case of Costa Brava, Spain against Antalya, Turkey. Tourism Manage- ment Perspectives 28: 53–61.
Xxxxxxx-Xxxxxxx A and Xxxxxx-Xxx X (2018) Under- standing tourism loyalty: horizontal vs. destination loyalty. Tourism Management 65: 245–255.
Xxxxxxx X (2004) Media strategies for improving an unfavorable city image. Cities 21(6): 471–479.
Baloglu S (2001) Image variations of Turkey by famil- iarity index: informational and experiential dimen- sions. Tourism Management 22: 127–133.
Baloglu S and Xxxxxxxx D (1997) Affective images of tourism destinations. Journal of Travel Research 35(4): 11–15.
Xxxxxxx S, Xxxxxxxxx TL and Xxxxx S (2014) Destina- tion image and brand personality of Jamaica: a model of tourist behavior. Journal of Travel & Tourism Marketing 31(8): 1057–1070.
Xxxxxxx S and XxXxxxxx KW (1999) A model of des- tination image formation. Annals of Tourism Research 26(4): 868–897.
Baltar F and Brunet I (2012) Social research 2.0: vir- tual snowball sampling method using Facebook. Internet Research 22(1): 57–74.
Xxxxxx A and Mart´ın JD (2004) Factors influencing destination image. Annals of Tourism Research 31(3): 657–681.
Xxxxxxx C, Xxxxxxx S and Cu´neo A (2017) Under- standing travellers’ intentions to visit a short versus long-haul emerging vacation destination: the case of Chile. Tourism Management 59: 312–324.
Xxxxx X, Xxxxxxx M and Xxxxxxx J (2001) Tourism
image, evaluation variables and after-purchase behaviour inter-relationships. Tourism Manage- ment 22(6): 607–616.
Xxxxx E, Xxxxxxxxxx M and Xxxxxx S (2013) My city– my brand: the different roles of residents in place branding. Journal of Place Management and Development 6(1): 18–28.
Xxxxx XX (1989) Suppressing valid inferences with conditionals. Cognition 31: 61–83.
Xxxxx RM (2016) Structural Equation Modeling with AMOS: Basic Concepts, Applications, and Pro- gramming, 3rd ed. London: Routledge.
Campo S and Xxxxxxx MD (2014) Can tourism promo- tions influence a country’s negative image? An experimental study on Israel’s image. Current Issues in Tourism 17(3): 201–219.
Xxxxxxxx MM, Aran˜a JE, Xxx´n CJ, et al. (2015) Eco- nomic valuation of tourism destination image. Tourism Economics 21(4): 741–759.
Xxxx X and Xxx Y (2012) Segmenting mainland Chi- nese tourists to Taiwan by destination familiarity: a factor-cluster approach. International Journal of Tourism Research 14(4): 339–352.
Xxxx CC, Xxxxx JY, Xxx J, et al. (2017) Destination familiarity and favorability in a country-image context: examining Taiwanese travelers’ percep- tions of China. Journal of Travel & Tourism Mar- keting 34(9): 1211–1223.
Xxxx C-C and Xxxxxxx JF (2016) The roles of perceived travel benefits, importance, and constraints in
predicting travel behavior. Journal of Travel Research 55(4): 509–522.
Xxxx CF and Xxxx S (2013) A closer look at destina- tion: image, personality, relationship and loyalty. Tourism Management 36: 269–278.
Xxxx FF (2007) Sensitivity of goodness of fit indexes to lack of measurement invariance. Structural Equation Modeling 14(3): 464–504.
Xxxxxx GW and Xxxxxxxx RB (2002) Evaluating goodness-of-fit indexes for testing measurement invariance. Structural Equation Modeling 9: 233–255.
Chi CG and Qu H (2008) Examining the structural relationships of destination image, tourist satisfac- tion and destination loyalty: an integrated approach. Tourism Management 29(4): 624–636.
Xxxx JG, Xxxxxxxxx T and Xxx S (2011) On the desti- nation image of Korea by Russian tourists. Tourism Management 32: 193–194.
Xxxxx X (1972) Toward a sociology of international tourism. Social Research 39(1): 164–182.
Xxxxxxx X, Xxxx XX and Xxxxx F (2015) Why do tour- ists persist in visiting the same destination? Tour- ism Economics 21(1): 205–221.
De Nisco A, Xxxxxxxx G, Xxxxxx V, et al. (2015) Tour- ism satisfaction effect on general country image, destination image, and post-visit intentions. Jour- nal of Vacation Marketing 21(4): 305–317.
Xxxxxxx CM and Xxxxxxx XXX (1993) The measurement of destination image: an empirical assessment. Journal of Travel Research 31(4): 3–13.
Xxxxxx X, Xxxxxxxxxxxx N and Xxx SS (2011) An inte- grative model of place image: exploring relation- ships between destination, product, and country images. Journal of Travel Research 50(5): 520–534.
Faullant R, Xxxxxxx K and Xx¨ller J (2008) The impact of satisfaction and image on loyalty: the case of Alpine ski resorts. Managing Service Quality: An International Journal 18(2): 163–178.
Xxxxxxx C and Xxxxxxx D (1981) Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research 18(1): 39–50.
Xxxxxxxxxx J, Xxx S, Xxxxxxx DR, et al. (2009) The role of hubs in the adoption process. Journal of Marketing 73(2): 1–13.
Xxxxx MJ and Xxxxx G (2008) An empirical structural model of tourists and places: progressing involve- ment and place attachment into tourism. Tourism Management 29(6): 1141–1151.
Xx X and Xxxx X (2008) Place attachment, identity and community impacts of tourism – the case of a Beijing hutong. Tourism Management 29(4): 637–647.
Gursoy D (2011) Modeling Tourist Information Search Behavior: A Structural Modeling Approach. Saarbrucken: Xxxxxxx Academic Publishing.
Gursoy D, Xxx Xxxxxxx G and Xxxxx Y (2018) Impact of destination familiarity on external information source selection process. Journal of Destination Marketing & Management 8: 137–146.
Gursoy D and XxXxxxxx KW (2004) An integrative model of tourists’ information search behavior. Annals of Tourism Research 31(2): 353–373.
Hair JF, Xxxxx WC, Xxxxx BJ, et al. (2013) Multivari- ate Data Analysis, 7th ed. Upper Saddle River, NJ: Xxxxxxx Education Limited.
Xxxxxxx WE, Xxxxxxxx EA and Xxxxxx RD (2006) Place bonding for recreation places: conceptual and empirical development. Leisure Studies 25(1): 17–41.
Xxxxxxx XX, Xxxx GT and Oh CO (2009) Compari- son of place bonding models in recreation resource management. Journal of Leisure Research 41(1): 57–72.
Henkel R, Henkel P, Xxxxxx W, et al. (2006) Thailand as a tourist destination: perceptions of international visitors and Thai residents. Asia Pacific Journal of Tourism Research 11(3): 269–287.
Xxxxx RH (2012) Path analysis and structural equation modelling with latent variables. In: Xxxxxx X (ed) APA Handbook of Research Methods in Psychol- ogy: Vol. 2. Research Designs. Washington, DC: American Psychological Association, pp. 333–367. Xx Y and Xxxxxxx XXX (1993) Measuring destination attractiveness: a contextual approach. Journal of
Travel Research 32(2): 25–34.
Imada T and Xxxxxxxxx PC (2011) Proud Americans and lucky Japanese: cultural differences in apprai- sal and corresponding emotion. Emotion 11(2): 329–345.
ISTAT (2018) Tourist movement in Italy. Available at: xxxxx://xxxxx.xx/xx/xxxxx/./0000/00/xxxxxxxxx_ 2016_rettifica-1.pdf (accessed 3 November 2018). Xxx X and Xxxxxxxxxx SL (2003) Motion picture impacts on destination images. Annals of Tourism
Research 30(1): 216–237.
Xxx X-X (2018) The impact of memorable tourism experiences on loyalty behaviors: the mediating effects of destination image and satisfaction. Jour- nal of Travel Research 57(7): 856–870.
Xxx X and Xx X (2016) Tourism destination image: conceptual problems and definitional solutions. Journal of Travel Research 55(8): 1065–1080.
Xxx X, Xxxxxx J and Xxxx J (2015) Exploring the dimensional relationships among image formation agents, destination image, and place attachment
from the perspectives of pop star fans. Journal of Travel & Tourism Marketing 32(6): 730–746.
Xxx CH, Xxxxxx DB, Xxxxxxxxxx DL, et al. (2007) Examining the role of cognitive and affective image in predicting choice across natural, devel- oped, and theme-park destinations. Journal of Travel Research 46(2): 183–194.
Xxx YH, Xxxx CC and Park CW (2012) The salient and organic images of Taiwan as perceived by Main- land Chinese tourists. Asia Pacific Journal of Tour- ism Research 17(4): 381–393.
XxXxxxxxx B, Xxxxxxx-Xxxxxxx B and Xx E (2012) Rethinking loyalty. Annals of Tourism Research 39(2): 708–734.
XxXxxxxx XX, Xxxxxxx BW, Xxxxxxx XX, et al. (2014) An institutional assessment of three local government-level tourism destinations at different stages of the transformation process. Tourism Man- agement 41: 107–118.
Xxxx C and Xxxxxxx X (2000) Internet Communication and Qualitative Research. London: SAGE.
Manyiwa S, Priporas CV and Wang XL (2018) Influ- ence of perceived city brand image on emotional attachment to the city. Journal of Place Manage- ment and Development 11(1): 60–77.
Xxxxxxx X, Xxxxx X and Xxxxxxx X (2018) Australia’s cognitive, affective and conative destination image: an Emirati tourist perspective. Journal of Islamic Marketing 9(1): 36–59.
Xxxxxx A and Pizam A (1995) The role of awareness and familiarity with a destination: the Central Flor- ida case. Journal of Travel Research 33(3): 21–00. Xxxxxxxxx X, Xxxxxxxx-Xxxxxxxxxx X, Xxxxx-Xxxxxxxx R, et al. (2018) DMO online platforms: image and intention to visit. Tourism Management 65:
116–130.
Xxxxxx-Xxx X and Mart´ın-Xxxxxxx JD (2015) Under- standing the image of self-contained and serviced apartments: the case of sun and beach destinations. Journal of Hospitality & Tourism Research 39(3): 373–400.
Xxxxx B, Xxxx Xxxx W and Xxxxxx B (2010) Commu- nity perceptions of tourism: Bruny and Magnetic islands, Australia. Asia Pacific Journal of Tourism Research 15(3): 353–366.
Xxxxxx XX (1999) Whence consumer loyalty? The Journal of Marketing 63: 33e44.
Ozdemir B, Xxxx A, Xxxxxxx R, et al. (2012) Relation- ships among tourist profile, satisfaction and desti- nation loyalty: examining empirical evidences in Antalya region of Turkey. Journal of Hospitality Marketing & Management 21(5): 506–540.
Papadimitriou D, Xxxxxxxxxx K and Apostolopoulou A (2018) Destination image components and word- of-mouth intentions in urban tourism: a multigroup
approach. Journal of Hospitality & Tourism Research 42(4): 503–527.
Xxxxxxx JF (2005) Segmenting cruise passengers with price sensitivity. Tourism Management 26(5): 753–762.
Xxxxxxx JF and Xxxxxxx SJ (2001) An examination of golf travelers’ satisfaction, perceived value, xxx- alty, and intentions to revisit. Tourism Analysis 6(3/4): 223–237.
Xxxx XX and Airoldi EM (2015) A natural experiment of social network formation and dynamics. Pro- ceedings of the National Academy of Sciences 112(21): 6595–6600.
Xxxxxxxx WJ and Xxxx X (2010) Destination image dif- ferences between visitors and non-visitors: a case of New York City. International Journal of Tour- ism Research 12(5): 642–645.
Xxxx S and Page SJ (2014) Destination marketing organizations and destination marketing: a narra- tive analysis of the literature. Tourism Manage- ment 41: 202–227.
Xxxx S and Xxxx X (2004) Destination positioning analysis through a comparison of cognitive, affec- tive, and conative perceptions. Journal of Travel Research 42(4): 333–342.
Prayag G (2009) Tourists’ evaluations of destination image, satisfaction, and future behavioural inten- tions – the case of Mauritius. Journal of Travel & Tourism Marketing 26(8): 836–853.
Xxxxxx X and Xxxx X (2012) Antecedents of tourists’ loyalty to Mauritius: the role and influence of des- tination image, place attachment, personal involve- ment, and satisfaction. Journal of Travel Research 51(3): 342–356.
Xxxxxxxx X (2004) Tourist familiarity and imagery.
Annals of Tourism Research 31(4): 923–945.
Xxxxxxxx R and Xxxxxxxx V (2000) Evoking Ireland: modeling tourism propensity. Annals of Tourism Research 27(2): 490–516.
Xxxx X and Xxx X (2014) Chinese urban migrants’ sense of place: Emotional attachment, identity formation, and place dependence in the city and community of Guangzhou. Asia Pacific Viewpoint 55(1): 81–101.
Xx X, Xxx LH and Im HH (2011) A model of destina- tion branding: integrating the concepts of the branding and destination image. Tourism Manage- ment 32(3): 465–476.
Xxxxxxxxxx X and Xxxxxx X (2011) City image and perceived tourism impact: evidence from Port Xxxxx, Mauritius. International Journal of Hospi- tality & Tourism Administration 12(2): 123–143.
Riscinto-Xxxxx K and Xxxxxx N (2012) Conversion of local winery awareness: an exploratory study in visitor vs non-visitor attitude and perception. International Jour- nal of Wine Business Research 24(4): 287–301.
Rodr´ıguez Xxxxxx MA´ , Fr´ıas-Jamilena DM and Cas- tan˜eda-Garc´ıa JA (2013) The moderating role of past experience in the formation of a tourist desti- nation’s image and in tourists’ behavioural inten- tions. Current Issues in Tourism 16(2): 107–127.
Xxxxxx JA, Xxxx LM and Xxxxx G (1981) Affective quality attributed to environments: a factor analytic study. Environment and Behavior 13(3): 259–288. Xxxx X and Xxxxxx M (2010) The destination image gap – visitors’ and residents’ perceptions of place: evidence from Xxxxxxx Island, New Zealand. Cur-
rent Issues in Tourism 13(6): 541–561.
San Mart´ın H, Xxxxxxx J and Xxx Xxxxxx IAR (2013) An exploration of the effects of past experience and tourist involvement on destina- tion loyalty formation. Current Issues in Tourism 16(4): 327–000.
Xxx Xxxxxxx X and Xxx Xxxxxx IAR (2008) Exploring the cognitive–affective nature of destination image and the role of psychological factors in its forma- tion. Tourism Management 29(2): 263–277.
Xxxxxxxxxx X, Xxxxxxx G, Xxxxxxx BW, et al. (2014) Investigating the role of prior knowledge in tourist decision making: a structural equation model of risk perceptions and information search. Journal of Travel Research 53(3): 307–322.
Snepenger DJ (1987) Segmenting the vacation market by novelty-seeking role. Journal of Travel Research 26(2): 8–14.
Stylidis D, Xxxxx A and Xxxxxxxxx Y (2017) Testing an integrated destination image model across resi- dents and tourists. Tourism Management 58: 184–195.
Su MM, Wall G and Xx X (2016) Heritage tourism and livelihood sustainability of a resettled rural community: Mount Sanqingshan World Heritage Site, China. Journal of Sustainable Tourism 24(5): 735–757.
Sun X, Xxx CG and Xx X (2013) Developing destina- tion loyalty: The case of Hainan Island. Annals of Tourism Research 43: 547–577.
Xxx WK and Xx CE (2016) An investigation of the relationships among destination familiarity, desti- nation image and future visit intention. Journal of Destination Marketing & Management 5(3): 214–226.
Xxxxxx TB (1998) Better loyalty measurement leads to business solutions. Marketing News 32(22): 41–42. Tsai SP (2012) Place attachment and tourism market- ing: investigating international tourists in
Singapore. International Journal of Tourism Research 14(2): 139–152.
Xxxxx POD, Xxxxxxx A and Rebelo E (2008) Determi- nants of tourism return behaviour. Tourism and Hospitality Research 8(3): 205–219.
Xxx xx Xxxxxx R, Xxxxxx P and Xxx J (2012) A check- list for testing measurement invariance. European Journal of Developmental Psychology 9(4): 486–492.
Veasna S, Xx WY and Xxxxx CH (2013) The impact of destination source credibility on destination satisfaction: the mediating effects of destination attachment and destination image. Tourism Man- agement 36: 511–526.
Xxxxx AR, Xxxxxx F, Xxxx YR, et al. (2011) An epis- temological view of consumer experiences. Interna- tional Journal of Hospitality Management 30(1): 10–21.
Xxxx CY and Xxx MK (2010) The relationships of destination image, satisfaction, and behavioral intentions: an integrated model. Journal of Travel & Tourism Marketing 27(8): 829–843.
Xxxxxx DB and Xxxxxx LJ (2011) Visitor loyalty at a private South Carolina protected area. Journal of Travel Research 50(3): 335–346.
Xxxxx B, Xxxxxxx RE and Xxxxxx DL (2007) Market- ing Research: Text and Cases. Philadelphia, PA: The Haworth Press.
Xxxxx X, Xxxxxxx VP and Xxxxxxxxxx DR (2015) Information technology and consumer behavior in travel and tourism: insights from travel planning using the Internet. Journal of Retailing and Con- sumer Services 22: 244–249.
Xue L and Xxxxx Y (2020) The effect of distance on tourist behavior: a study based on social media data. Annals of Tourism Research 82: 10291b.
Xxxxxx S and Xxxxxxxx SC (2013) My place is not your place – different place brand knowledge by different target groups. Journal of Place Manage- ment and Development 6(1): 6–17.
Xxxxx H, Xx X, Cai LA, et al. (2014) Destination image and tourist loyalty: a meta-analysis. Tourism Management 40: 213–223.
Xxxxx HD, Xxxxxx D and Xx X (2017) The effects of associative slogans on tourists’ attitudes and travel intention: the moderating effects of need for cogni- tion and familiarity. Journal of Travel Research 56(2): 206–220.
Appendix 1. Construct scales and extended CFA results
Table 1A. Items of construct scales and standardised factor loadings of partial metric invariance CFA.
Standardised factor loading
Scale Item statement (Name) | Residents (N = 295) | Non-residents (N = 680) |
Destination Familiarity I know holiday destinations in Molise. (DF1)a | .877 | .855 |
I know the natural resources of Molise. (DF2) | .918 | .955 |
I know the cities in Molise. (DF3)b | .937 | .911 |
Original Chaotic – Tranquil (Original3)a | .896 | .906 |
Risky – Safe (Original2) | .710 | .755 |
Emotional Lack of interest – With a character (Emotional4)a | .600 | .618 |
Boring – Intriguing (Emotional2) | .650 | .654 |
Sad – Pleasant (Emotional1) | .861 | .835 |
Services Hospitality services (Attraction3)a | .789 | .838 |
Accommodation facilities (Amenities2) | .763 | .826 |
Outdoor sports facilities (Amenities3) | .893 | .935 |
Entertainment activities (Amenities1)b | .813 | .797 |
Environment Scenic beauty (Natural1)a | .835 | .819 |
Weather/climate (Natural2) | .729 | .758 |
Natural environment (Natural3) | .802 | .775 |
Local cultural/historic sites (Accessability3) | .771 | .809 |
Orderly environment (Accessability4)b | .762 | .803 |
Personal safety (SocialEnv1)b | .722 | .752 |
Local people’s friendliness (SocialEnv2) | .756 | .791 |
Note: All unfixed loadings are statistically significant at 0.1%; loadings of the second-order factors (i.e., cognitive image, affective image) on their first-order factors are displayed in Table 4.
aItem with loading fixed as 1.
bUnconstrained loadings between two group.