How does familiarity shape destination image and loyalty for visitors and residents?

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 residents 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 relationships 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 destinations has intensified with an upsurge in new entrants globally. The highly competitive environment requires deep understanding of the fundamental drivers of destination image and how this translates into loyalty behaviours, particularly likelihood to visit and recommend the destination (Albayrak et al., 2018; Pike and Page, 2014; Weaver and Lawton, 2011). Visitors’ destination image is often formed from various information sources they are exposed to, as well as their prior knowledge and experience of the destination. This means that destination image can be influenced by secondary sources, such as destination websites, travel guides, the internet and social media (Sharifpour et al., 2014; Xiang et al., 2015), as well as destination familiarity, which arises from their exposure to education, travel guides, mass media, and personal contact with other individuals knowledgeable about the destination (Gursoy, 2011).

The literature purports that destination familiarity is central in shaping tourists’ decision-making regarding a destination (Bianchi et al., 2017; Chen and Phou, 2013; Gursoy et al., 2018; Sharifpour et al., 2014). For example, Baloglu (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 (Hammitt et al., 2006). Consequently, 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 critically affect an individual’s destination choice (intention to visit) and their word-of-mouth behaviour (intention to recommend) (Ozdemir et al., 2012; Tsai, 2012).

Prior research considers differences between visitors and residents perceptions of a destination (Baloglu et al., 2014; Phillips and Jang, 2010) and their attitude and behaviours relating to the destination (Choi et al., 2011; Riscinto-Kozub and Childs, 2012). Residents are often very familiar with a destination and perceive the destination differently (Tan and Wu, 2016). Indeed, both residents and visitors exist along a familiarity spectrum, with varying degrees of familiarity that manifest in different perceptions and behavioural intentions with regards to a particular destination. So, familiarity likely explains differences between visitors and residents’ perceptions and behaviours. In the tourism literature, familiarity 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 (Milman 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 experience, but also from the information search, education and networks.

Consequently, this research aims to study the role of destination familiarity in shaping destination 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 loyalty 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 residents) 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 ‘ability to describe or even map a place based on images, memories and perceptions of locations, size, distance, physical attributes and site experiences’ (Hammitt et al., 2009: 25). As familiarity is enhanced by frequent visits or a period of residence, 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 (Cohen, 1972; Snepenger, 1987). But while novelty has been a central topic in the tourism literature, familiarity has been relatively disregarded (Prentice, 2004). However, increasingly studies are finding that familiarity influences the tourism decision-making process, particularly because it means that visitors may not undertake an informational search (Gursoy and McCleary, 2004). Familiarity has been equated to the sum of informational familiarity, combined with experiential familiarity (Baloglu, 2001; Chen and Lin, 2012). Often researchers find that visitors with greater familiarity of a destination have a more positive overall image of the destination (Baloglu, 2001; Chen and Lin, 2012; Prentice, 2004) and greater intention to visit (Tan and Wu, 2016). The positive relationship between familiarity and favourability has been explained by the destination choice-sets model (Chen and Lin, 2012; Prentice and Andersen, 2000), which suggests that people funnel down their potential destinations through intensive information-processing resulting in visitors selecting from a choice-set of familiar and favourable destinations. Chen et al. (2017) suggest 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 differences 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 De Nisco et al. (2015) and Campo and Alvarez (2014) find that familiarity generally improves destination image for visitors. However, Elliot et al. (2011) found a non-significant positive effect of familiarity on destination image, suggesting that the relationship may not always hold. Chen et al. (2017) found that people’s evaluation 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 destination (with greater familiarity leading to more positive images and evaluations).

Destination image is a multidimensional construct (Veasna et al., 2013) that is the sum of ideas, impressions and beliefs people have of the various attributes, aspects and activities of a destination (Zhang et al., 2014). Many recent studies measure overall destination image with cognitive image and affective image (e.g., Carballo et al., 2015; Kim, 2018; Molinillo et al., 2018; Moreno-Gil and Martín-Santana, 2015; Stylidis 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 (Pike and Ryan, 2004), as well as the affective image, which represents feelings about a destination (Lai and Li, 2016). Cognitive images are often measured using catalogues of functional and psychological attributes, while affective images are measured via an affective grid scale (Prayag and Ryan, 2012; Russel et al., 1981). Generally, tourists need a positive image of a destination to consider visiting (Braun et al., 2013; Ramkissoon and Nunkoo, 2011). Finally a third interrelated component of destination image is the conative or behavioural image, which influences and predicts tourists’ behaviour (Michael et al., 2018).

Several studies consider residents and visitors place attachment, finding key differences in perceptions and images of a destination between residents and visitors (Braun et al., 2013; Zenker and Beckmann, 2013). The literature finds that both residents and visitors establish place attachment and form images of the destination and its attributes (Gross and Brown, 2008; Gu and Ryan, 2008). Thus, there is a need to more fully understand the link between destination image and resident and visitors’ attitudinal and behavioural intentions (Bigne et al., 2001; Stylidis 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 activities (Stylidis et al., 2017). Hence residents’ destination image can influence tourists’ image formation, decision-making and purchasing behaviour, as well as destination development (Bigne et al., 2001; Walls et al., 2011). Yet Stylidis et al. (2017) argues that few studies compare the images of tourist destinations formed by residents to those held by tourists, even if significant differences between the two groups have been found (Henkel et al., 2006).

Prior studies (i.e. Hu and Ritchie, 1993; Milman and Pizam, 1995) explain differences in perceived image using tourists’ previous experience, with destination familiarity usually leading to positive images of place (Baloglu, 2001). Residents often have more accurate perceptions of, and stronger attachment to, their place of residence (Stylidis 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. Cultural factors and place attachment often underlie differences in an individual’s perceptions of a destination and influence destination loyalty (Beerli and Martin, 2004; Gu and Ryan, 2008; Imada and Ellsworth, 2011; Kim, 2018). There is also a dynamic interplay between residents’ perceptions of place and their perceptions of the impacts of tourism, which can thereby influence their own travel behaviour (Su et al., 2016). Elaborating on this concept further, Xue and Zhang (2020) discuss the role of distance in influencing tourist behaviour between long-haul, short-haul and local travellers, particularly their motivations, 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 suggests that resident’s familiarity shapes their image of the destination. Similarly, Braun et al. (2013) purport that residents are more likely to view a destination positively and become more emotionally attached as they are more closely associated with the destination. Likewise, Manyiwa et al. (2018) investigate differences in resident and visitors’ perceptions of image and emotional attachment to a destination finding that affective and cognitive image positively influence emotional attachment confirming findings of Lee et al. (2015). In fact, Manyiwa 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 perceptions 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 predictor of behaviour (Chi and Qu, 2008; Oliver, 1999). Consumer loyalty indices tend to be composed of behaviour-based measures, including intention to visit, repeat visit and recommend a destination, but also overall satisfaction (Taylor, 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 (Faullant et al., 2008; Prayag and Ryan, 2012). Some tourism researchers equate place attachment to increased loyalty (Qian and Zhu, 2014; Tsai, 2012). Others also argue that destination evaluations are a predictor of travel propensity (Baloglu and McCleary, 1999; Beerli and Martín, 2004; Lin et al., 2012). Moreover, the strength of the relationships between image and future behavioural intentions can vary depending on the context (San Martín et al., 2013) and personal characteristics (Prayag, 2009). Yet McKercher et al. (2012) point out that loyalty has been studied quite simplistically in the literature, often via similar indicators using single case studies. Instead they argue that a consumer can be horizontally loyal, whereby the tourist is loyal to several tourism suppliers or destinations. While there have been significant studies into tourist loyalty, very few consider multi-destination loyalty, with the exception of Almeida-Santana and Moreno-Gil (2018). Moreover, factors that may reduce loyalty and intention to visit have been under-examined with Chen and Petrick (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 (Petrick, 2005; Valle et al., 2008). For example, older travellers and those with lower incomes are more likely to revisit (Correia et al., 2015). Moreover, Molinillo 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 (Almeida-Santana and Moreno-Gil, 2018). Almeida-Santana and Moreno-Gil (2018) find that sea, sun and sand (3S) destinations are negatively related to loyalty, possibly due to the number of these types of destinations resulting in them being easily substitutable. They find that affective image is negatively related to destination 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 Qu, 2008). Beerli and Martin (2004) and Rodríguez Molina 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 non-residents and residents. Hypothesised effects among destination familiarity, cognitive image, affective image, overall image, intention to recommend and intention to visit the destination are collected in Figure 1.

Figure 1.

Research model.

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, emotional, and original).

H4 and H5: Cognitive and affective images predict overall image of the place.

H6: Overall image of the destination predicts the intention to visit the destination.

H7: Overall image of the destination predicts the intention to recommend Molise to others.

H8 and H9: the effects of destination familiarity on the two outcomes are partially 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 (McLennan 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. Concurrently, snowball sampling (Wrenn et al., 2007) was employed whereby the students were asked to forward the survey to their friends, relatives 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 inexpensive and efficient in providing researchers with an increasingly-expanding set of respondents (Goldenberg 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 (Mann and Stewart, 2000). Although a limitation of the approach is the possibility of missing, or being biased against, certain groups within the population, 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 questionnaire survey, 116 respondents (10.6% of the sample) did not answer at least a quarter of the questions (excluding demographics). These low engagement respondents were removed from further analysis. The remaining sample included 975 respondents with an overall 4.48% of scale item values missing not completely at random (Little’s MCAR test: χ² = 26223.06, df = 24276, p < .001). With the assumption of conditionally random missing data, the missing values were then imputed using the expectation–maximisation (EM) algorithm in SPSS v.25. No significant multivariate outliers were identified in the sample based on Mahalanobis 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).

Table 1.

Demographic profiles of local resident and visitor samples.

Demographics	Locals (N = 295)		Visitors (N = 680)
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
Marital Status*
Single	170	57.6	377	55.4
Married with young children living at home	57	19.3	146	21.5
Married with no children or older children no longer living at home	64	21.7	147	21.6
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 (master – PhD)	42	14.2	158	23.2
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

* Having a trivial amount of missing values.

Instrument and measures

The questionnaire comprised two sections, with the first being general socio-demographic characteristics (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 destination. For the purposes of this research and the questionnaire implemented in this study, the definition 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 (Lin et al., 2007). Drawing on the literature (Baloglu et al., 2014; Stylidis et al., 2017; Wang and Hsu, 2010), multiple aspects of cognitive image were considered: natural environment, amenities, accessibility and social environment. 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 attributes selected to evaluate the affective image comprised the following five-point bipolar semantic differential items such as chaotic-tranquil and boring-intriguing (Kim and Richardson, 2003; San Martin and del Bosque, 2008; Wang and Hsu, 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 Beerli and Martin (2004) and Wang and Hsu (2010). The set of items relevant to each image dimension was constructed 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 measured 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 holidays 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 Appendix 1. The questionnaire was piloted with 20 residents and tourists to verify instrument validity. No major concerns were reported in the pilot.

Table 2.

Descriptive statistics of measures in both local and visitor samples.

Scale items	Residents (N = 295)					Non-residents (N = 680)
Scale items	M	SD	Skew	Kurt	Alpha	M	SD	Skew	Kurt	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; Kurt: kurtosis; Alpha: Cronbach’s α.

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., residents 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 the residents’ 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 Byrne (2016) and Hair et al. (2013) who argued that moderate departures from normality do not significantly 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 reliability 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 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) (Byrne, 2016; Hair et al., 2013; Hoyle, 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), Tucker-Lewis 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 (α) coefficient and composite reliability (CR) with .70 as the cut-off value; convergent validity was evaluated with the 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 (Fornell and Larcker, 1981).

Further, to reach valid comparison of the hypothesised relationships between two groups, the scales must measure identical constructs across different groups, namely measurement invariance. It means that the respondents across groups interpret the individual 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 operationalised 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 (Cheung and Rensvold, 2002).

With configural invariance established, the CFA model was further constrained with equal factor loadings between groups, i.e. metric invariance. This was testing whether the two groups attributed the same meaning to the latent construct (Van de Schoot et al., 2012) or the constructs manifested in the same way across groups (Cheung and Rensvold, 2002). Full metric invariance is difficult to satisfy in practice, and some researchers (e.g., Byrne, 1989) propose partial 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 difference test, which is an excessively stringent test of invariance (Van de Schoot et al., 2012). Cheung and Rensvold (2002) proposed a practical criterion for evidence of invariance, namely ΔCFI < .01. Chen (2007) also recommended that ΔCFI < .01 and ΔRMSEA < .015 for tests of invariance. This paper presents testing results using both χ² test and change in CFI.

Lastly, with measurement invariance established, 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 statistically lower fit than the model with no constraints, a set of increasingly constrained SEM models were tested to locate the paths that were significantly different between residents and visitors. Demographic variables were included as controls. All descriptive statistics were calculated 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 normality in the data for further CFA and SEM. All five scales also showed high internal consistency with Cronbach’s α 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 displayed 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 models), 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 (Δχ² = 40.231, Δdf = 16, p = .001; ΔCFI = .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 (Δχ² = 17.499, Δdf = 12, p = .132; ΔCFI = .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.

Table 3.

Fit indices and tests for multi-group confirmatory factor analysis model and structural equation models.

Model	Multi-Group Specification	χ²/df	RMSEA	90% UL	CFI	TLI	PNFI	AIC	Δχ²	Δdf	p	ΔCFI
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 invariance	2.140	.034	.038	.978	.970	.702	795.117	17.499	12	.132	.000
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 paths	2.327	.037	.040	.956	.942	.693	1558.955	13.463	12	.336	.001

χ²/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: Tucker-Lewis Index; PNFI: parsimony normed fit index; AIC: Akaike information criterion; Δχ²: difference in χ² values compared with the unconstrained CFA or SEM model; Δdf: difference in number of degrees of freedom compared with the unconstrained CFA or SEM model; p: significance of the Chi-square likelihood ratio test; ΔCFI: difference in CFI values compared with the unconstrained CFA or SEM model.

The construct reliability and validity of all first- and second-order factors were assessed based on the results of the partial metric invariance 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 measurement error. The square root of each AVE was greater than the inter-construct correlations, suggesting that the constructs differed from each other. In summary, the results suggest that the constructs are reliable and valid in each group.

Table 4.

Construct reliability and validity for first-order and second-order factors in local sample (N = 295) and visitor sample (N = 680).

Factor	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.

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 significantly lower goodness of fit (Δχ² = 32.162, Δdf = 15, p = .006; ΔCFI = .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 (Δχ² = 13.463, Δdf = 12, p = .336; ΔCFI = .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 intention to visit; and, the effect of overall image onto intention to recommend the destination.

Figure 2.

Comparison of the hypothesised effects between residents and non-residents.

Discussion

This research empirically tests the relationship between destination familiarity, destination image and the behavioural intentions of visiting and recommending the destination and juxtaposes 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 statistically significant thereby confirming the literature that cognitive image predicts affective image (Elliot 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 Qu (2008) suggesting a positive relationship between destination familiarity, image, satisfaction and loyalty. Zhang et al. (2017) also found a complicated relationship between the characteristics of a destination’s slogan and a visitor’s destination familiarity, attitudes and intentions to visit. This is important as it suggest that information exposure, 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.

Ryan and Aicken (2010) and Moyle 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 affective image for residents. The most loyal of all people in a destination are residents and being a resident reduces the novelty of a destination, which may reduce emotional images. Thus, residents 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 Almeida-Santana and Moreno-Gil (2018), who determined that loyalty is negatively related to affective image. However, these findings contrast to those of Manyiwa et al. (2018) who conclude that affective image has a stronger significant positive effect on emotional attachment 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 marketing 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 Brinberg, 1997), but that both cognitive and affective are appropriate for residents (Stylidis et al., 2017). This research, however, finds that only cognitive marketing will be effective for residents.

The group comparisons found that H6 varied significantly between residents and visitors, suggesting that residents’ overall image of the destination 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 represents 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 influential force for delivering destination recommendations (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 residents’ 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 (Bianchi 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 relationship between destination familiarity and the behavioural 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 plausible 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 emotional images as they need to form a closer connection 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 minimise 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 destination (Sharifpour et al., 2014; Zhang et al., 2017).

Lastly, prior research has found that behavioural intentions are influenced by gender, age and income (Correia et al., 2015; Petrick, 2005; Petrick and Backman, 2001; Valle et al., 2008), but we found no significant effect for age, gender and income on intention to recommend 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 travellers are more likely to repeat visit (Correia 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 differences is important as residents can determine the path of destination development as well as influence visitors perceptions and behaviour. Theoretically, 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 visitors’ affective image and intention to visit associated with their familiarity with the destination, as well as differences between the groups overall image and intention to recommend the destination. Notably, we find that familiarity is negatively related to affective image in the case of Molise, which contrasts to that of Manyiwa et al. (2018) in the context of the emerging destination of Bratislava, Slovakia.

Measuring the destination image perceived by tourists and understanding how this relates to residents’ perceptions of the destination is essential for the proper strategic management of destinations (San Martín 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 (Echtner and Ritchie, 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 perceptions (Chen 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 residents’ perceptions of place image is therefore more likely to be effective through community consultation that enables community identified targeted infrastructure development, inclusive planning processes and involvement in the imaging campaign, for example through competitions to select logos (Avraham, 2004). Hence, our findings have important strategic implications for destination 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 understanding 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 cultural 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 vertical 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 perceptions of policymakers and destination marketers. 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 promoting to various types of people (residents, domestic/international visitors and non-visitors).

Footnotes

Declaration of conflicting interests

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

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Gian Luca Casali

Yulin Liu

Appendix 1. Construct scales and extended CFA results

Table 1A.

Items of construct scales and standardised factor loadings of partial metric invariance CFA.

ScaleItem statement (Name)	Standardised factor loading
ScaleItem 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.

^a Item with loading fixed as 1.

^b Unconstrained loadings between two group.

References

Albayrak

Caber

González-Rodríguez

, et al. (2018) Analysis of destination competitiveness by IPA and IPCA methods: the case of Costa Brava, Spain against Antalya, Turkey. Tourism Management Perspectives 28: 53–61.

Almeida-Santana

Moreno-Gil

(2018) Understanding tourism loyalty: horizontal vs. destination loyalty. Tourism Management 65: 245–255.

Avraham

(2004) Media strategies for improving an unfavorable city image. Cities 21(6): 471–479.

Baloglu

(2001) Image variations of Turkey by familiarity index: informational and experiential dimensions. Tourism Management 22: 127–133.

Baloglu

Brinberg

(1997) Affective images of tourism destinations. Journal of Travel Research 35(4): 11–15.

Baloglu

Henthorne

Sahin

(2014) Destination image and brand personality of Jamaica: a model of tourist behavior. Journal of Travel & Tourism Marketing 31(8): 1057–1070.

Baloglu

McCleary

(1999) A model of destination image formation. Annals of Tourism Research 26(4): 868–897.

Baltar

Brunet

(2012) Social research 2.0: virtual snowball sampling method using Facebook. Internet Research 22(1): 57–74.

Beerli

Martín

(2004) Factors influencing destination image. Annals of Tourism Research 31(3): 657–681.

10.

Bianchi

Milberg

Cúneo

(2017) Understanding travellers’ intentions to visit a short versus long-haul emerging vacation destination: the case of Chile. Tourism Management 59: 312–324.

11.

Bigne

Sanchez

(2001) Tourism image, evaluation variables and after-purchase behaviour inter-relationships. Tourism Management 22(6): 607–616.

12.

Braun

Kavaratzis

Zenker

(2013) My city–my brand: the different roles of residents in place branding. Journal of Place Management and Development 6(1): 18–28.

13.

Byrne

(1989) Suppressing valid inferences with conditionals. Cognition 31: 61–83.

14.

Byrne

(2016) Structural Equation Modeling with AMOS: Basic Concepts, Applications, and Programming, 3rd ed. London: Routledge.

15.

Campo

Alvarez

(2014) Can tourism promotions influence a country’s negative image? An experimental study on Israel’s image. Current Issues in Tourism 17(3): 201–219.

16.

Carballo

Araña

León

, et al. (2015) Economic valuation of tourism destination image. Tourism Economics 21(4): 741–759.

17.

Chen

Lin

(2012) Segmenting mainland Chinese tourists to Taiwan by destination familiarity: a factor-cluster approach. International Journal of Tourism Research 14(4): 339–352.

18.

Chen

Chung

Gao

, et al. (2017) Destination familiarity and favorability in a country-image context: examining Taiwanese travelers’ perceptions of China. Journal of Travel & Tourism Marketing 34(9): 1211–1223.

19.

Chen

C-C

Petrick

(2016) The roles of perceived travel benefits, importance, and constraints in predicting travel behavior. Journal of Travel Research 55(4): 509–522.

20.

Chen

Phou

(2013) A closer look at destination: image, personality, relationship and loyalty. Tourism Management 36: 269–278.

21.

Chen

(2007) Sensitivity of goodness of fit indexes to lack of measurement invariance. Structural Equation Modeling 14(3): 464–504.

22.

Cheung

Rensvold

(2002) Evaluating goodness-of-fit indexes for testing measurement invariance. Structural Equation Modeling 9: 233–255.

23.

Chi

(2008) Examining the structural relationships of destination image, tourist satisfaction and destination loyalty: an integrated approach. Tourism Management 29(4): 624–636.

24.

Choi

Tkachenko

Sil

(2011) On the destination image of Korea by Russian tourists. Tourism Management 32: 193–194.

25.

Cohen

(1972) Toward a sociology of international tourism. Social Research 39(1): 164–182.

26.

Correia

Zins

Silva

(2015) Why do tourists persist in visiting the same destination? Tourism Economics 21(1): 205–221.

27.

De Nisco

Mainolfi

Marino

, et al. (2015) Tourism satisfaction effect on general country image, destination image, and post-visit intentions. Journal of Vacation Marketing 21(4): 305–317.

28.

Echtner

Ritchie

JRB

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

29.

Elliot

Papadopoulos

Kim

(2011) An integrative model of place image: exploring relationships between destination, product, and country images. Journal of Travel Research 50(5): 520–534.

30.

Faullant

Matzler

Füller

(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.

31.

Fornell

Larcker

(1981) Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research 18(1): 39–50.

32.

Goldenberg

Han

Lehmann

, et al. (2009) The role of hubs in the adoption process. Journal of Marketing 73(2): 1–13.

33.

Gross

Brown

(2008) An empirical structural model of tourists and places: progressing involvement and place attachment into tourism. Tourism Management 29(6): 1141–1151.

34.

Ryan

(2008) Place attachment, identity and community impacts of tourism – the case of a Beijing hutong. Tourism Management 29(4): 637–647.

35.

Gursoy

(2011) Modeling Tourist Information Search Behavior: A Structural Modeling Approach. Saarbrucken: Lambert Academic Publishing.

36.

Gursoy

Del Chiappa

Zhang

(2018) Impact of destination familiarity on external information source selection process. Journal of Destination Marketing & Management 8: 137–146.

37.

Gursoy

McCleary

(2004) An integrative model of tourists’ information search behavior. Annals of Tourism Research 31(2): 353–373.

38.

Hair

Black

Babin

, et al. (2013) Multivariate Data Analysis, 7th ed. Upper Saddle River, NJ: Pearson Education Limited.

39.

Hammitt

Backlund

Bixler

(2006) Place bonding for recreation places: conceptual and empirical development. Leisure Studies 25(1): 17–41.

40.

Hammitt

Kyle

(2009) Comparison of place bonding models in recreation resource management. Journal of Leisure Research 41(1): 57–72.

41.

Henkel

Agrusa

, 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.

42.

Hoyle

(2012) Path analysis and structural equation modelling with latent variables. In: Cooper

(ed) APA Handbook of Research Methods in Psychology: Vol. 2. Research Designs. Washington, DC: American Psychological Association, pp. 333–367.

43.

Ritchie

JRB

(1993) Measuring destination attractiveness: a contextual approach. Journal of Travel Research 32(2): 25–34.

44.

Imada

Ellsworth

(2011) Proud Americans and lucky Japanese: cultural differences in appraisal and corresponding emotion. Emotion 11(2): 329–345.

45.

ISTAT (2018) Tourist movement in Italy. Available at: https://istat.it/it/files/./2017/10/movimento_2016_rettifica-1.pdf (accessed 3 November 2018).

46.

Kim

Richardson

(2003) Motion picture impacts on destination images. Annals of Tourism Research 30(1): 216–237.

47.

Kim

J-H

(2018) The impact of memorable tourism experiences on loyalty behaviors: the mediating effects of destination image and satisfaction. Journal of Travel Research 57(7): 856–870.

48.

Lai

(2016) Tourism destination image: conceptual problems and definitional solutions. Journal of Travel Research 55(8): 1065–1080.

49.

Lee

Busser

Yang

(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.

50.

Lin

Morais

Kerstetter

, et al. (2007) Examining the role of cognitive and affective image in predicting choice across natural, developed, and theme-park destinations. Journal of Travel Research 46(2): 183–194.

51.

Lin

Chen

Park

(2012) The salient and organic images of Taiwan as perceived by Mainland Chinese tourists. Asia Pacific Journal of Tourism Research 17(4): 381–393.

52.

McKercher

Denizci-Guillet

(2012) Rethinking loyalty. Annals of Tourism Research 39(2): 708–734.

53.

McLennan

Ritchie

Ruhanen

, et al. (2014) An institutional assessment of three local government-level tourism destinations at different stages of the transformation process. Tourism Management 41: 107–118.

54.

Mann

Stewart

(2000) Internet Communication and Qualitative Research. London: SAGE.

55.

Manyiwa

Priporas

Wang

(2018) Influence of perceived city brand image on emotional attachment to the city. Journal of Place Management and Development 11(1): 60–77.

56.

Michael

James

Michael

(2018) Australia’s cognitive, affective and conative destination image: an Emirati tourist perspective. Journal of Islamic Marketing 9(1): 36–59.

57.

Milman

Pizam

(1995) The role of awareness and familiarity with a destination: the Central Florida case. Journal of Travel Research 33(3): 21–27.

58.

Molinillo

Liébana-Cabanillas

Anaya-Sánchez

, et al. (2018) DMO online platforms: image and intention to visit. Tourism Management 65: 116–130.

59.

Moreno-Gil

Martín-Santana

(2015) Understanding the image of self-contained and serviced apartments: the case of sun and beach destinations. Journal of Hospitality & Tourism Research 39(3): 373–400.

60.

Moyle

Glen Croy

Weiler

(2010) Community perceptions of tourism: Bruny and Magnetic islands, Australia. Asia Pacific Journal of Tourism Research 15(3): 353–366.

61.

Oliver

(1999) Whence consumer loyalty? The Journal of Marketing 63: 33e44.

62.

Ozdemir

Aksu

Ehtiyar

, et al. (2012) Relationships among tourist profile, satisfaction and destination loyalty: examining empirical evidences in Antalya region of Turkey. Journal of Hospitality Marketing & Management 21(5): 506–540.

63.

Papadimitriou

Kaplanidou

Apostolopoulou

(2018) Destination image components and word-of-mouth intentions in urban tourism: a multigroup approach. Journal of Hospitality & Tourism Research 42(4): 503–527.

64.

Petrick

(2005) Segmenting cruise passengers with price sensitivity. Tourism Management 26(5): 753–762.

65.

Petrick

Backman

(2001) An examination of golf travelers’ satisfaction, perceived value, loyalty, and intentions to revisit. Tourism Analysis 6(3/4): 223–237.

66.

Phan

Airoldi

(2015) A natural experiment of social network formation and dynamics. Proceedings of the National Academy of Sciences 112(21): 6595–6600.

67.

Phillips

Jang

(2010) Destination image differences between visitors and non-visitors: a case of New York City. International Journal of Tourism Research 12(5): 642–645.

68.

Pike

Page

(2014) Destination marketing organizations and destination marketing: a narrative analysis of the literature. Tourism Management 41: 202–227.

69.

Pike

Ryan

(2004) Destination positioning analysis through a comparison of cognitive, affective, and conative perceptions. Journal of Travel Research 42(4): 333–342.

70.

Prayag

(2009) Tourists’ evaluations of destination image, satisfaction, and future behavioural intentions – the case of Mauritius. Journal of Travel & Tourism Marketing 26(8): 836–853.

71.

Prayag

Ryan

(2012) Antecedents of tourists’ loyalty to Mauritius: the role and influence of destination image, place attachment, personal involvement, and satisfaction. Journal of Travel Research 51(3): 342–356.

72.

Prentice

(2004) Tourist familiarity and imagery. Annals of Tourism Research 31(4): 923–945.

73.

Prentice

Andersen

(2000) Evoking Ireland: modeling tourism propensity. Annals of Tourism Research 27(2): 490–516.

74.

Qian

Zhu

(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.

75.

Kim

(2011) A model of destination branding: integrating the concepts of the branding and destination image. Tourism Management 32(3): 465–476.

76.

Ramkissoon

Nunkoo

(2011) City image and perceived tourism impact: evidence from Port Louis, Mauritius. International Journal of Hospitality & Tourism Administration 12(2): 123–143.

77.

Riscinto-Kozub

Childs

(2012) Conversion of local winery awareness: an exploratory study in visitor vs non-visitor attitude and perception. International Journal of Wine Business Research 24(4): 287–301.

78.

Rodríguez Molina

MÁ

Frías-Jamilena

Castañeda-García

(2013) The moderating role of past experience in the formation of a tourist destination’s image and in tourists’ behavioural intentions. Current Issues in Tourism 16(2): 107–127.

79.

Russel

Ward

Pratt

(1981) Affective quality attributed to environments: a factor analytic study. Environment and Behavior 13(3): 259–288.

80.

Ryan

Aicken

(2010) The destination image gap – visitors’ and residents’ perceptions of place: evidence from Waiheke Island, New Zealand. Current Issues in Tourism 13(6): 541–561.

81.

San Martín

Collado

Del Bosque

IAR

(2013) An exploration of the effects of past experience and tourist involvement on destination loyalty formation. Current Issues in Tourism 16(4): 327–342.

82.

San Martín

Del Bosque

IAR

(2008) Exploring the cognitive–affective nature of destination image and the role of psychological factors in its formation. Tourism Management 29(2): 263–277.

83.

Sharifpour

Walters

Ritchie

, 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.

84.

Snepenger

(1987) Segmenting the vacation market by novelty-seeking role. Journal of Travel Research 26(2): 8–14.

85.

Stylidis

Shani

Belhassen

(2017) Testing an integrated destination image model across residents and tourists. Tourism Management 58: 184–195.

86.

Wall

(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.

87.

Sun

Chi

(2013) Developing destination loyalty: The case of Hainan Island. Annals of Tourism Research 43: 547–577.

88.

Tan

(2016) An investigation of the relationships among destination familiarity, destination image and future visit intention. Journal of Destination Marketing & Management 5(3): 214–226.

89.

Taylor

(1998) Better loyalty measurement leads to business solutions. Marketing News 32(22): 41–42.

90.

Tsai

(2012) Place attachment and tourism marketing: investigating international tourists in Singapore. International Journal of Tourism Research 14(2): 139–152.

91.

Valle

POD

Correia

Rebelo

(2008) Determinants of tourism return behaviour. Tourism and Hospitality Research 8(3): 205–219.

92.

Van de Schoot

Lugtig

Hox

(2012) A checklist for testing measurement invariance. European Journal of Developmental Psychology 9(4): 486–492.

93.

Veasna

Huang

(2013) The impact of destination source credibility on destination satisfaction: the mediating effects of destination attachment and destination image. Tourism Management 36: 511–526.

94.

Walls

Okumus

Wang

, et al. (2011) An epistemological view of consumer experiences. International Journal of Hospitality Management 30(1): 10–21.

95.

Wang

Hsu

(2010) The relationships of destination image, satisfaction, and behavioral intentions: an integrated model. Journal of Travel & Tourism Marketing 27(8): 829–843.

96.

Weaver

Lawton

(2011) Visitor loyalty at a private South Carolina protected area. Journal of Travel Research 50(3): 335–346.

97.

Wrenn

Stevens

Loudon

(2007) Marketing Research: Text and Cases. Philadelphia, PA: The Haworth Press.

98.

Xiang

Magnini

Fesenmaier

(2015) Information technology and consumer behavior in travel and tourism: insights from travel planning using the Internet. Journal of Retailing and Consumer Services 22: 244–249.

99.

Xue

Zhang

(2020) The effect of distance on tourist behavior: a study based on social media data. Annals of Tourism Research 82: 10291b.

100.

Zenker

Beckmann

(2013) My place is not your place – different place brand knowledge by different target groups. Journal of Place Management and Development 6(1): 6–17.

101.

Zhang

Cai

, et al. (2014) Destination image and tourist loyalty: a meta-analysis. Tourism Management 40: 213–223.

102.

Zhang

Gursoy

(2017) The effects of associative slogans on tourists’ attitudes and travel intention: the moderating effects of need for cognition and familiarity. Journal of Travel Research 56(2): 206–220.