Brand Personality of Tourist Destinations - An Application of Self-congruity Theory

lable at ScienceDirect

Tourism Management 32 (2011) 114e127

Contents lists avai

Tourism Management

journal homepage: www.elsevier .com/locate/ tourman

Brand personality of tourist destinations: An application of self-congruity theory

Ahmet Usakli a,*, Seyhmus Baloglu b,1

aDepartment of Tourism Management, Faculty of Tourism, Nevsehir University, 2000 Evler Mah., Zubeyde Hanim Cad., 50300 Nevsehir, TurkeybDepartment of Tourism and Convention Administration, William F. Harrah College of Hotel Administration, University of Nevada, Las Vegas, 4505 Maryland Parkway, Box 456023,Las Vegas, NV 89154-6023, United States

a r t i c l e i n f o

Article history:Received 6 March 2010Accepted 8 June 2010

Keywords:Destination brandingBrand personalityDestination personalitySelf-congruity theoryLas Vegas

* Corresponding author. Tel.: þ90 384 228 1130; faE-mail addresses: [email protected] (A

unlv.edu (S. Baloglu).1 Tel.: þ1 702 895 3932; fax: þ1 702 895 4870.

0261-5177/$ e see front matter � 2010 Elsevier Ltd.doi:10.1016/j.tourman.2010.06.006

a b s t r a c t

The purpose of this research was to investigate the perceived destination personality of Las Vegas and toexamine the relationships among destination personality, self-congruity, and tourist’s behavioralintentions. A convenience sample of 382 visitors to Las Vegas was surveyed, and 368 usable question-naires were analyzed. The findings of the study indicate that tourists ascribe personality characteristicsto destinations and that the perceived destination personality of Las Vegas is five dimensional: vibrancy,sophistication, competence, contemporary, and sincerity. These dimensions have a positive influence ontourists’ intention to return and intention to recommend. The study also supports the self-congruitytheory within the context of tourism destinations, indicating that both actual congruity and idealcongruity have a positive impact on behavioral intentions. The study concludes that self-congruity isa partial mediator on the relationship between destination personality and tourist’s behavioral inten-tions. The practical and theoretical implications are discussed within the contexts of destinationbranding and the self-congruity theory.

� 2010 Elsevier Ltd. All rights reserved.

1. Introduction

In today’s highly competitive market, consumers are not onlysurrounded by numerous brands but are also exposed to manydifferent marketing practices that are developed to differentiatethese brands from their competitors. The concept of branding hasbeen extensively applied to products and services in the genericmarketing field (Blain, Levy, & Ritchie, 2005), nevertheless brandsare also found inmany categories of tourismproducts and permeatealmost all facets of tourism activities (Cai, 2002; Gnoth, Baloglu,Ekinci, & Sirakaya-Turk, 2007). A tourism destination can also beseen as a product or perceived as a brand since it consists of a bundleof tangible and intangible attributes (Ekinci & Hosany, 2006; Ekinci,Sirakaya-Turk, & Baloglu, 2007; Hosany, Ekinci, & Uysal, 2007).

Although the idea of branding tourism destinations is relativelynew (Blain et al., 2005; Cai, 2002; Gnoth, 1998), many destinationsaround the world have been trying to adopt branding strategiessimilar to those used by Coca Cola, Nike, and Sony, in an attempt todifferentiate their identities and to emphasize the uniqueness oftheir products (Morgan, Pritchard, & Pride, 2004). In today’s world,

x: þ90 384 215 2006.. Usakli), seyhmus.baloglu@

All rights reserved.

companies/organizations are engaged in a battle of not onlyproducts or services but also perceptions in the consumer mind.Destination marketing organizations (DMOs) are also involved inthis constant battle to attract travelers because destinations arebecoming highly substitutable due to the growing global compe-tition (Pike & Ryan, 2004).

Destination branding has become a popular and powerfulmarketing tool because of increasing competition, product simi-larity, and substitutability in tourism markets. Many destinationsstill promote the similar attributes such as beautiful scenery, goldenbeaches, blue seas, or friendly places in their advertisements(Ekinci et al., 2007; Murphy, Benckendorff, & Moscardo, 2007a).However, travelers can find many destinations with beautifulscenery, blue seas or golden beaches to travel. The use of suchattributes in destination marketing no longer helps differentiatedestinations from their competitors. Positioning destinations basedon their functional attributes makes them easily substitutable.Therefore, Ekinci and Hosany (2006) suggest that destinationpersonality can be used as a viable metaphor for building desti-nation brands, understanding visitors’ perceptions of destinations,and crafting a unique identity for tourism places. Thus, DMOsshould focus on developing marketing campaigns emphasizing thedistinctive personality of their destinations.

Furthermore, according to the self-congruity theory, thereshould be congruence between destination personality and visitors’

mailto:[email protected]



www.sciencedirect.com/science/journal/02615177

http://www.elsevier.com/locate/tourman

http://dx.doi.org/10.1016/j.tourman.2010.06.006



A. Usakli, S. Baloglu / Tourism Management 32 (2011) 114e127 115

self-concept. Aaker (1995) mentions that the basic notion of theself-congruity theory is the drive in consumers to prefer brandswith personalities that are congruent with their own personality.Applying the basic notion of the self-congruity theory within thecontext of tourism destinations, it can be proposed that the greaterthe match between the destination personality and the visitor’sself-concept, the more likely it is that the visitor will have a favor-able attitude toward that destination (Sirgy & Su, 2000). This atti-tude might result in a visit or word of mouth. Thus, understandingthe congruity between destination personality and visitor’s self-concept is important to gain insight into the complex nature oftravel behavior.

2. Objectives of the study

Several researchers have studied the congruence betweenconsumer self-concept and product/brand image to predictconsumer behavior variables, such as product/brand attitude,intention, behavior, and loyalty (Sirgy, 1982, 1985a; Sirgy, Johar,Samli, & Claiborne, 1991). However, previous studies almostexclusively focused on the match between consumer self-conceptand product/brand image. Although Aaker (1999) found anotherself-congruity effect with reference to brand personality, there hasbeen sparse investigation on the congruence between consumerself-concept and brand personality (e.g., Aaker, 1999; Azevedo &Pessoa, 2005; Helgeson & Suphellen, 2004).

The purpose of this study is to investigate the perceived desti-nation personality of Las Vegas and to empirically examine therelationships among destination personality, self-congruity, andtourist’s behavioral intentions (return intention and recommen-dation). First, the perceived destination personality of Las Vegasand its underlying dimensions are investigated. Second, the studyexamines the effects of the destination personality on tourist’sbehavior in terms of intention to return and intention to recom-mend. Third, the effects of self-congruity on the behavioral inten-tions are examined. Finally, the relationship between destinationpersonality, self-congruity, and behavioral intentions is examinedto assess the mediating role of self-congruity between destinationpersonality and behavioral intentions and to identify any indirecteffects of the destination personality on behavioral intentionsthrough self-congruity.

3. Literature review

3.1. Brand personality

Brand personality can be defined as “the set of human charac-teristics associated with a brand” (Aaker, 1997, p. 347). Althoughbrands are inanimate objects, consumers often view them ashaving human characteristics (Aaker, 1997, 1999; Plummer, 1985).These characteristics could be traits such as youthful, outdoorsy,sporty, energetic, or sophisticated. For instance, one may use thefollowing words to describe some popular brands: “cool” for thesoft drink Coca Cola, whereas “young” for Pepsi (Aaker, 1997),“masculine” for Marlboro cigarettes (Ekinci & Hosany, 2006),“sophisticated” for a BMW car (Phau & Lau, 2000), and “unique” inthe case of Dr. Pepper (Plummer, 1985). Understanding brandpersonality is important because consumers select brands withpersonalities that are acceptable to them (Aaker, 1999). Manyresearchers have attempted to understand brand personality, butresearch on brand personality has remained limited due to the lackof both a conceptual framework and a reliable, valid, and general-izable scale to measure brand personality (Aaker, 1997). Accord-ingly, Aaker (1997) has developed a valid, reliable, andgeneralizable scale to measure brand personality, called The Brand

Personality Scale (BPS), based on a representative sample anda comprehensive list of personality traits. Aaker (1997) developedher scale on the basis of three sources: personality scales frompsychology, personality scales used by marketers, and originalqualitative research of personality traits associated with a numberof well-known brands. Thus, Aaker (1997) not only developed a 42-item BPS, but also developed a theoretical brand personalityframework that consists of five personality dimensions: sincerity,excitement, competence, sophistication, and ruggedness. Thesedimensions are derived from 15 personality facets of popularbrands. Aaker (1997) suggests that five dimensions of BPS aregeneric and can be used across product categories. However, shestates that the BPS may not be a perfect fit across cultures. There-fore, she has called for further research to determine the extent towhich personality dimensions are stable across cultures. Sincethen, literature and research on brand personality have flourished(Azoulay & Kapferer, 2003), and many researchers applied thebrand personality framework to various product groups and acrossdifferent cultures.

Although it has been widely adopted by researchers, Aaker’s(1997) BPS received some critics as well. Azoulay and Kapferer(2003) argue that the BPS does not in fact measure brand person-ality, but merge a number of dimensions of brand identity,personality being only one of them. According to Azoulay andKapferer (2003), this is because of Aaker’s (1997) loose definitionof brand personality, and thus a stricter definition is needed.Another criticism relates to the generalizability of the BPS. In theirstudy of re-examining the generalizability of the BPS, Austin,Siguaw, and Mattila (2003) found that Aaker’s (1997) brandpersonality framework does not generalize to individual brandswithin one product category. Austin et al. (2003) explains this issuewith the argument that Aaker (1997) tested the reliability andvalidity of her scale by aggregating data across diverse productcategories, rather than measuring the personality of individualbrands or aggregating data within a specific product category.Despite its criticisms, Aaker’s (1997) BPS is still the most stable,reliable and comprehensive scale developed for measuring brandpersonality (Ekinci & Hosany, 2006).

3.2. Destination personality: the application of brand personality totourism destinations

Destination personality refers to brand personality in thecontext of tourism literature. Ekinci and Hosany (2006) definedestination personality as “the set of personality traits associatedwith a destination” (p. 127), adapting Aaker’s (1997) brandpersonality terminology. Although product/brand personalityresearch in the consumer goods domain began in the early 1960s(Ekinci & Hosany, 2006), the investigation and application of brandpersonality to tourism destinations are relatively new (Ekinci &Hosany, 2006; Gnoth et al., 2007; Hosany, Ekinci, & Uysal, 2006;Hosany et al., 2007; Pitt, Opoku, Hultman, Abratt, &Spyropoulou, 2007; Tasci & Kozak, 2006). A well-establishedbrand personality facilitates differentiation of a brand from that ofits competitors (Aaker, 1996), enhances the brand equity (Keller,1993), increases brand preference and usage (Aaker, 1999;Malhotra, 1988; Sirgy, 1982), and develops strong emotional tiesbetween consumers and brands, thereby resulting in greater trustand loyalty (Fournier, 1998). Similarly, a distinctive and attractivedestination personality can effectively leverage the perceivedimage of a destination, and thereby, influence tourist choicebehavior (Ekinci & Hosany, 2006).

Although brand personality has been acknowledged as animportant component of branding, very few academics and prac-titioners have attempted to identify the salient personality

A. Usakli, S. Baloglu / Tourism Management 32 (2011) 114e127116

characteristics of tourism destinations. For example, the WesternAustralian Tourism Commission created BrandWestern Australia inthe late 1990s and positioned Western Australia as a premiernature-based tourism destination. The core personality elements ofBrand Western Australia emphasized the qualities “fresh”,“natural”, “free”, and “spirited” (Crockett & Wood, 2004).Henderson (2000) surveyed a sample of both local residents andinternational tourists in Singapore and found that the brandpersonality of New AsiaeSingapore Brand comprised characteris-tics such as “cosmopolitan”, “youthful”, “vibrant”, “modern Asia”,“reliable”, and “comfortable”. Santos (2004) conducted a framinganalysis of the travel sections in select US newspapers (New YorkTimes, Washington Post, Los Angeles Times, and USA Today)regarding featured articles, between 1996 and 2002, about tourismin Portugal. Santos (2004) found that Portugal was represented inUS newspapers’ travel sections with personality traits such as“traditional”, “contemporary”, “modern”, and “sophisticated”.

Ekinci and Hosany (2006) are the first to examine the applica-bility and validity of Aaker’s (1997) brand personality framework inthe context of tourism destinations. The authors found that touristsascribe personality characteristics to destinations, and hence, theconcept of BPS can be applied to tourism destinations. Theyconcluded that destination personality consists of three salientdimensions, rather than the original five dimensions: sincerity,excitement, and conviviality. Sincerity and excitement were found tobe the two main factors. Conviviality was new and specific todestinations (Ekinci & Hosany, 2006). Since then, empirical studieson destination personality began to emerge in the tourism litera-ture. A summary of previous studies on destination personality ispresented in Table 1.

3.3. Self-concept and the self-congruity theory

Self-concept has been advanced as a useful construct forunderstanding and explaining consumer choice behavior. It hasbeen suggested that consumers prefer products or brands that aresimilar to how they see or would like to see themselves (Landon,1974; Malhotra, 1988; Sirgy, 1982). In consumer behavior litera-ture, several researchers have concentrated on how the personalityof a brand enables consumers to express their own self (Belk, 1988;Birdwell, 1968; Dolich, 1969; Malhotra, 1988). Self-concept (alsoreferred to as self-image) has been defined as “the totality ofindividual’s thoughts and feelings having reference to himself as anobject” (Rosenberg, 1979, p. 7).

The earlier studies conceptualized self-concept as a unidimen-sional construct and treated it as the actual self-concept, whereaslater studies conceptualized it as having two components: actual andideal self-concepts (Malhotra, 1988). Sirgy (1982) has progressedbeyond this duality dimension and developed a multidimensionalconstructof self-concept thatnotonlyconsists of actual and ideal self-concepts, but also includes social self-concepts. According to Sirgy(1982), self-concept is a multidimensional construct having fourmajor components, namely, actual self-concept, ideal self-concept,social self-concept and ideal social self-concept. Within this multi-dimensional framework, the actual self-concept refers to howa person actually perceives himself or herself, whereas the ideal self-concept refers to how a person would like to perceive himself orherself. Social self-concept refers to how an individual thinks othersperceive him or her, whereas ideal social self-concept represents theway the individual desires to be perceived by others (Sirgy, 1982).

Self-congruity can be considered a natural extension of self-concept. The basic hypothesis in the self-congruity theory is thata consumer tends to select products or brands that correspond toone’s self-concept. This idea suggests that the greater the degree ofcongruence, the higher the probability of intention to purchase.

According to Sirgy et al. (1997), the degree of consistency betweenconsumer’s self-concept and that of brand is referred to as self-image/product image congruity, self-image congruence, or self-congruity for short. The self-congruity theory proposes thatconsumer behavior is determined, in part, by a cognitive matchingbetween value-expressive attributes of a product or brand andconsumer self-concept (Sirgy et al., 1991).

Because self-concept has been treated as a multidimensionalconstruct reflecting four major types of self-concept, self-congruityin turn, has been treated multidimensionally. Four major types ofself-congruity are defined in literature: actual self-congruity, idealself-congruity, social self-congruity, and ideal social self-congruity(Sirgy, 1982). The congruity between the actual self-concept andthe product/brand image has been referred to as actual self-congruity, between the ideal self-concept and product/brandimagedas ideal self-congruity, between the social self-concept andproduct/brand imagedas social self-congruity, and between theideal social self-concept and the product/brand imagedas idealsocial self-congruity (Sirgy, 1985b).

4. Proposed model

Fig. 1 outlines the model hypothesized in this research. It hasbeen argued that a distinctive brand personality has positive effectson brand attitudes (Helgeson & Suphellen, 2004). More specifically,it contributes to the differentiation of a specific brand from those ofits competitors (Aaker, 1996), increases brand preference and usage(Aaker, 1999; Malhotra, 1988; Sirgy, 1982), enhances the brandequity (Keller, 1993), builds strong emotional relationshipsbetween consumers and brands, and thus results in greater trustand loyalty (Fournier, 1998). Applying the concept of brandpersonality to tourism destinations, previous studies have investi-gated the effects of destination personality on tourist’s behavioralintentions. For instance, Ekinci and Hosany (2006) have found thatone of the dimensions of destination personality has a significanteffect on intention to recommend. Ekinci et al. (2007) have foundthat destination personality has a positive influence both onintention to return and word of mouth. Following these studies, thepresent study proposes that destination personality will havea direct positive influence on tourist’s behavioral intentions.Accordingly, the following hypotheses were derived:

H1. Destination personality will have a positive impact on inten-tion to return.

H2. Destination personality will have a positive impact onintention to recommend.

In the consumer behavior literature, self-congruity is defined asthe match between a product/brand image and consumer’s self-concept (Sirgy, 1985b; Sirgy et al., 1991, 1997). It consists of twocomponents, namely, self-concept and product/brand image.Several researchers have studied the congruence betweenconsumer self-concept and product/brand image to predictconsumer behavior variables, such as product/brand attitude,intention, behavior, loyalty, and so on (Sirgy, 1982, 1985a; Sirgyet al., 1991). However, these previous studies have extensivelyfocused on the match between the consumer self-concept andproduct/brand image, not on brand personality. Because of the poorconceptualization and a lack of empirical studies, there is muchambiguity in the relationship between brand image and brandpersonality (Hosany et al., 2006). Hosany et al. (2006) have testedthe relationship between brand image and brand personality in thecontext of tourism destinations and found that destination imageand destination personality are two different, but related, concepts.Destination image is an encompassing concept, whereas

Table 1Summary of destination personality studies.

Reference Destination(s) studied Sample Method Dimensions found Major findings

Ekinci and Hosany (2006) - A number of destinationsby recalling the lastdestination visited- A popular European city

A total of 250British travelers

- Structured: Aaker’s (1997)BPS, content validity, 27items of BPS, 5-pointLikert-type scale

3 Dimensions: Sincerity,excitement, and conviviality.

- Tourists attribute personalitycharacteristics totourism destinations.

- BPS is applicable totourism destinations.

- DP has a positive impacton intention to recommend.

Hosany et al. (2006) A number of destinations byrecalling the last destination visited

148 British travelers - Structured: Aaker’s (1997)BPS, content validity, 27items of BPS, 5-pointLikert-type scale

3 Dimensions: sincerity,excitement, and conviviality.

- DI and DP are two different,but related concepts. While DIis an encompassing concept,DP is more related to theaffective components of DI.

Ekinci et al. (2007) Mediterranean region of Turkey 365 German travelers - Structured: 20-itemDP adjectives recommendedby Ekinci and Hosany (2006),5-point Likert-type scale

3 Dimensions: conviviality,sincerity, and excitement.

- Host image has a positiveimpact on DP.

- DP has a positive impacton intention to returnand word of mouth.

Murphy, Moscardo,and Benckendorff (2007)

Two destinations inQueensland, Australia:

- Cairns- Whitsunday Islands

480 Visitors toQueensland

- Structured: 20 items ofAaker’s (1997) BPS,5-point Likert-type scale- Unstructured:Open-ended questions

Cairns (3 dimensions):sincere, sophisticated,and outdoorsy.Whitsunday Islands(4 dimensions): Upperclass, honest, exciting,and tough.

- The open-ended responsesof personality descriptorswere not as commonas Aaker’s (1997)personality traits.

- The findings providesome evidence that BPcan be used to differentiatetourism destinations.

Murphy et al. (2007a) Whitsunday Islands,Queensland, Australia


- Structured: 20 itemsof Aaker’s (1997)BPS, 5-point Likert-type scale

4 Dimensions: sophistication andcompetence, sincerity,excitement, and ruggedness.

- Provided evidence of arelationship between travelmotivation and DP.

- Provided evidence of a linkbetween DP and SC.

- Found no relationship betweenDP and actual andintended visitation.

Murphy, Benckendorff,and Moscardo (2007b)

Whitsunday Islands,Queensland, Australia


- Structured: 20 items ofAaker’s (1997) BPS,5-point Likert-type scale

4 Dimensions: sophisticationand competence, sincerity,excitement, and ruggedness.

- Tourist needs and DP perceptionsare associated with higher SC levels.Higher SC levels are relatedto satisfaction with destination,but not related to intention to visit.

Pitt et al. (2007) 10 African countries Official tourismwebsites of 10African countries

- Content analysis: a list of 922synonyms to Aaker’s (1997)42 personality traits werecollected, and then, categorizedaccording to Aaker’s (1997)BP dimensions.

Each country was evaluatedbased on Aaker’s (1997)BP dimensions.

- Demonstrated a researchmethod that shows howbrands communicate theirbrand personalities online.

- Found that some countriesfocus on specific dimensionsof Aaker’s (1997) BPframework while others arefailing to communicate anyof the Aaker’s (1997) BPdimensions at all.

Prayag (2007) Two destinations:- South Africa- Cape Town

85 Internationaltourists visitingCape Town

- Unstructured: Projectivetechniques, in-depth interviews

No dimensions - Projective techniques werefound effective to elicit thedestination-specificpersonality traits.

(continued on next page)

A.U

sakli,S.Baloglu/Tourism

Managem

ent32

(2011)114

e127

117

Table

1(con

tinu

ed)

Referen

ceDestination(s)studied

Sample

Method

Dim

ension

sfound

Major

findings

D’Astou

san

dBou

jbel

(200

7)Anumbe

rof

countries

representing

five

continen

ts

Fren

chsp

eaking

Can

adians

-Structured:previou

spersonalityscales,

5-pointbipolar

scales

-Unstructured:interviews

6Dim

ension

s:ag

reea

blen

ess,

wicke

dness,sn

obbism

,assiduou

sness,

conform

ity,

andunob

trusive

ness

-Dev

elop

edaco

untry

personalityscale.

-Th

escalewas

foundto

beless

usefulfor

predicting

peo

ple’sperception

sof

countriesas

trav

eldestinations.

Sahin andBalog

lu(200

9)Istanbu

l,Tu

rkey

272Intern

ational

visitors

toIstanbu

l-Structured:23

item

sfrom

Aak

er’s(199

7)BPS

,5item

sba

sedon

content

analysis

oftrav

elbroc

huresan

dintern

etsites

abou

tIstanbu

l,5-pointLike

rt-typ

escale

-Unstructured:

open

-ended

question

s

5Dim

ension

s:co

mpeten

cean

dmod

ernity,

originalityan

dvibran

cy,

sincerity,c

oolan

dtren

dy,

andco

nviviality

-Th

estudyfoundsome

perception

differencesacross

differentnationalities

forDPdim

ension

s.

BP:

bran

dpersonality,

BPS

:bran

dpersonalityscale,

DI:destinationim

age,

DP:

destinationpersonality,

SC:self-con

gruity.


destination personality is more related to the affective componentsof the destination image (Hosany et al., 2006).

Brand image refers to both functional and symbolic benefitsof a brand (Low & Lamb, 2000). On the other hand, brandpersonality only refers to the symbolic function of a brand(Keller, 1993). Thus, brand personality may have a closer link toconsumer self-concept than the functional benefits or attributesof a brand because it solely focuses on the personality traitsassociated with a brand. Therefore, in this study, the product/brand image concept is replaced by the brand personalityconcept and self-congruity is conceptualized as the matchbetween destination personality and tourist’s self-concept. Thefindings of Aaker (1999) also support the brand personalitycongruence effect. According to Aaker (1999), the main point ofself-congruity is that consumers prefer brands with a set ofpersonality traits congruent with their own. Thus, the hypoth-esized model posits that self-congruity will have a positiveimpact on tourist’s behavioral intentions. In other words, thegreater the match between the destination personality and thetourist’s self-concept, the more likely it is that the tourist willhave a favorable attitude toward that destination, resulting in anintention to return and intention to recommend.

H3. Self-congruity will have a positive impact on intention toreturn.

H3a. Actual self-congruity will have a positive impact on inten-tion to return.

H3b. Ideal self-congruity will have a positive impact intention toreturn.

H4. Self-congruity will have a positive impact on intention torecommend.

H4a. Actual self-congruity will have a positive impact intention torecommend.

H4b. Ideal self-congruity will have a positive impact on intentionto recommend.

It is important to note that there are four major types of self-congruity in the literature, namely, actual, ideal, social, and idealsocial (Sirgy et al., 1997). This study only concentrates on actual andideal self-congruity, because these two types of self-congruity havereceived the strongest empirical support and are most commonlyused (Sirgy, 1982; Sirgy et al., 1997).

Finally, the proposed model investigates the mediating impactof self-congruity on the relationship between destination person-ality and tourist’s behavioral intentions. According to the proposedmodel, destination personality also indirectly influences intentionto return and intention to recommend through self-congruity.Therefore, the following hypotheses were derived:

H5. Self-congruity will mediate the relationship between desti-nation personality and intention to return.

H5a. Actual self-congruity will mediate the relationship betweendestination personality and intention to return.

H5b. Ideal self-congruity will mediate the relationship betweendestination personality and intention to return.

H6. Self-congruity will mediate the relationship between desti-nation personality and intention to recommend.

H6a. Actual self-congruity will mediate the relationship betweendestination personality and intention to recommend.

H6b. Ideal self-congruity will mediate the relationship betweendestination personality and intention to recommend.

noitanitseD

ytilanosreP

ytiurgnoC-fleS

• ytiurgnoClautcA

• ytiurgnoClaedI

snoitnetnIlaroivaheB

• nruteR

• dnemmoceR

+ +

+

Fig. 1. The hypothesized model.


5. Methodology

This study applied a survey research design. A self-administeredquestionnaire was developed based on the literature review, scaledevelopment procedure, and interviews with the marketingmanagers of the Las Vegas Convention and Visitors Authority. Bothclose and open-ended questions were used in the questionnaire tocollect the required quantitative and qualitative data for the study.

5.1. BPS development for destinations

Although Aaker’s (1997) BPS is the most comprehensiveinstrument for measuring brand personality and is widely usedwithin different product categories and across different cultures, itis not specifically designed for tourism destinations. Hosany et al.(2006) suggest that some items of Aaker’s (1997) BPS are redun-dant in the context of tourism destinations, because they are notsuitable to define a tourism destination. Therefore, to ensure thatthe personality traits used in this study are pertinent to Las Vegas,a two-stage scale development procedure was conducted. In thefirst stage, a free-elicitation task was conducted to identify theunique traits that describe Las Vegas. In the second stage, the 42personality traits in the BPS were tested for content validity.

5.1.1. Stage 1: unique personality trait generationIn this stage, to identify the traits unique to Las Vegas, a free-

elicitation task was conducted among a group of 28 tourists visitingLas Vegas. Subjects were recruited using a convenience samplingtechnique on the Las Vegas Strip. Subjects (n¼ 28; 54% female, 46%male) were asked to think of Las Vegas as if it were a person and towrite down the personality traits that first came to mind. It waspredetermined that if a trait was mentioned by at least 25% of thesubjects, it would be included in the pool of personality traits. Nineunique traits resulting from this task met this criterion and wereadded to the pool of personality traits. The personality traits, theirfrequencies, andpercentageswereas follows: exciting (15; 54%), sexy(14; 50%), energetic (10; 36%), vibrant (9; 32%), independent (9; 32%),unique (8; 29%), alive (8; 29%), showy (8; 29%), and naughty (8; 29%).

5.1.2. Stage 2: the BPS content validityThe same sample of visitors was used in the second stage. In this

stage, the 42 personality traits in the BPS were tested for contentvalidity (Churchill, 1979). The items were measured using a 5-pointLikert-type scale, with anchors ranging from (1) not descriptive atall to (5) extremely descriptive, consistent with Aaker’s (1997)study. Subjects were asked to rate the degree to which theyperceived that each of the 42 personality traits accurately describedLas Vegas. To isolate the most relevant traits, the cutoff for the 42BPS personality traits was a scale with a mean rating of 3.00 orabove. A set of 23 items, split across 4 dimensions, was retainedfrom the BPS and were as follows: sincerity (original, cheerful,friendly); excitement (daring, trendy, exciting, spirited, cool, young,imaginative, unique, up-to-date, independent, contemporary);

competence (intelligent, successful, leader, confident); sophistica-tion (upper class, glamorous, good looking, charming, feminine).

The three personality traits, exciting, independent, and unique,which were generated in the first stage, were also among the 23items elicited in the content validity stage. Thus, the two stagesyielded a total of 29 personality traits for the final study.

5.2. The measurement

The destination personality was captured using personalitytraits derived from unique personality trait generation stage andmostly, Aaker’s (1997) BPS content validity stage. A total of 29destination personality items was measured on a 5-point Likert-type scale ranging from (1) strongly disagree to (5) strongly agree.

The ongoing debate in measuring self-congruity is whether touse the gap score formula (traditional method) or direct scoreformula (global measurement, new method) (Sirgy & Su, 2000).Sirgy et al. (1997) argue that measuring self-congruity with thedirect score formula is more predictive than that obtained using thegap score formula because it captures self-congruity more directlyand globally, contains less measurement errors, and is more holisticin capturing self-congruity. Thus, self-congruity was measuredwith an adaptation of the global measurement method developedby Sirgy et al. (Sirgy et al., 1997; Sirgy & Su, 2000). According to thismodel, the subject first describes the typical user of the brand (fordestinations, the typical visitor) and then states directly theconsistency between the typical user of the brand and his or herself-concept. The image of the typical user of the brand is believedto be reflective of the brand image. However, brand personality canbe formed and influenced by any direct or indirect contact that theconsumer has with the brand (Aaker, 1997). Brand’s user imagery,which is defined as “the set of human characteristics associatedwith the typical user of a brand” (Aaker, 1997, p. 348), is just one ofthe several approaches to brand personality formation. Thus, brandpersonality is a broader and more inclusive concept than the imageof the typical user of a brand (Helgeson & Suphellen, 2004). Addi-tionally, Keller (1998) has cautioned that user imagery and brandpersonality may not always be in agreement (as cited in Phau & Lau,2000). Aaker (1996) states that there are occasions where the userprofile is inconsistent with the personality that the brand projects.Furthermore, there might be more than one type of user imageryfor a brand. Thus, respondents were first asked to think of thedestination as if it were person. Next, they were asked to considerthe personality characteristics of the destination and their self-concept. Then, the respondents were asked to compare both thepersonality of the destination and their self-concept in their minds.Finally, they were instructed to express the consistency betweenthe destination personality and their self-concept by indicatingtheir levels of agreement with the statements regarding actual andideal self-congruity. Actual self-congruity statements were asfollows: “Las Vegas is consistent with how I see myself”; “I am quitesimilar to the personality of Las Vegas”; “The personality of LasVegas is congruent with how I see myself”. On the contrary, ideal


self-congruity was measured using the following statements: “LasVegas is consistent with how I would like to see myself”; “I wouldlike to be perceived as similar to the personality of Las Vegas”; “Thepersonality of Las Vegas is congruent with how I would like to seemyself”. Both actual and ideal self-congruity statements werederived from the previous research conducted by Sirgy et al. (1997),Sirgy and Su (2000), and Helgeson and Suphellen (2004). Thesecongruity statements were measured using a 5-point Likert-typescale ranging from (1) strongly disagree to (5) strongly agree.

Behavioral intentions (intention to return and intention torecommend) were measured using numerical scales. Intention torecommend was operationalized using a 10-point numerical scale,with (1) representing not recommend at all and (10) definitelyrecommend. Intention to return was also measured using a 10-point numerical scale, where (1) means do not intend to visit and(10) means very likely to visit. Overall destination brand person-ality evaluation was captured using a 10-point numerical scale,with (1) representing very negative and (10) very positive.

Four open-ended questions in the survey allowed respondentsto think freely about the destination and to express their originaland unique views. First, respondents were asked to describe theirgeneral images of Las Vegas using three words or phrases. Second,they were asked to list three personality traits associated with LasVegas. Third, they were asked to visualize and describe the typicalvisitor to Las Vegas to identify the typical user imagery, consistentwith Sirgy and Su’s (2000) suggested approach. Finally, respon-dents were asked towrite down a tourism slogan or a tagline for LasVegas in their own words.

Trip-related questions, including how long the visitor had beenin Las Vegas, their previous visits to Las Vegas, the main purpose oftheir trip, the party size, trip companions, tourist activities pursuedby visitors, and information sources used in holiday decisionmaking were all asked. Demographic questions, including age,gender, country/state of residence, marital status, level of educa-tion, and household income were also asked to provide additionalbackground on the respondents.

5.3. Pilot study

To discover any potential deficiencies and to test the feasibilityof the proposed questionnaire, a pilot study was conducted witha sample of 20 visitors to Las Vegas. There were no concernsregarding the wording of the questions or the format. Therefore, allthe questions were retained. The average response time tocomplete a questionnaire was 15 min.

5.4. Sample and data collection

The target population of this study comprised visitors to LasVegas. A convenience sampling method, which refers to thesampling procedure used to obtain units (or people) that are mostconveniently available (Zikmund, 2003), was used in the researchprocess. The self-administered questionnaires were distributed tothe visitors in front of the “Fountain Show” at Bellagio, one of thelargest casino hotels on the Las Vegas Strip. This place was deemedappropriate for data collection for two reasons. First, it wasa popular place on the Las Vegas Strip and, therefore, was mostlypopulated by visitors. Second, the pilot study showed that theaverage response time to complete a questionnaire was 15 min andvisitors who were waiting for the “Fountain Show” were bothinterested in the study and had sufficient time to complete thequestionnaire.

The datawere collected from January 19, 2009 toMarch 8, 2009.There were seven weeks in this data collection period. Two days ineach week were randomly selected, making a total of 14 days. The

surveys were conducted by three trained graduate students.Respondents were approached randomly rather than approachingall available visitors in a systematic way. The students approachedthe visitors who were waiting for the Fountain Show, identifiedthemselves, informed them about the study, and indicated thatparticipationwas confidential and voluntary. In general, the visitorswere participatory and showed a high level of interest. The refusalrates were relatively low (around 15%). Out of 382 questionnairescollected, 14 were not usable. Thus, a total of 368 questionnaireswere coded for data analysis.

5.5. Analysis of data

The software SPSS 16.0 was used to analyze the data. Dataanalysis included several stages. First, the data were explored forpossible errors in the entries and outliers; and descriptive statisticswere reported. Next, a factor analysis was performed to identify theunderlying personality dimensions of Las Vegas. After assessing thereliabilities using Cronbach’s alpha, factor scores for the identifieddimensions were calculated using the Anderson and Rubinmethod.Then, the reliability of the self-congruity measures was assessed byexamining the Cronbach’s alpha coefficients; the six self-congruitymeasures were subsequently reduced to two measures (actual andideal congruity) by calculating their mean scores. Finally, multipleregression analyses were conducted to examine the relationshipsamong destination personality, self-congruity, and behavioralintentions.

6. Results and discussion

6.1. Demographic profile of the respondents

There were slightly more female respondents (51.4%) thanmales. The age of the respondents ranged from 18 to 69, witha mean score of 36.6 years. Most of the respondents (73%) werefrom the United States (US), with themajority of them coming fromCalifornia (29.3%). This was followed by respondents from Arizona(10.3%) and Texas (7.6%). The rest of the US respondents were fromstates such as Utah, New York, Iowa, and Washington, amongothers. Of the 27% of the respondents from outside the US, 38.3%were from the United Kingdom and 25.2% from Canada. Among therespondents, 50.5% were married and approximately 38% helda university degree. The major portion of the respondents (27.5%)belonged to the income group of $60,000e$89,999.

6.2. Trip characteristics

More than half of the respondents (62.5%) were first-time visi-tors to Las Vegas. The remaining 37.5% had visited Las Vegaspreviously, and the mean number of previous visits within the pastthree years was less than three (mean¼ 2.9; median¼ 2.0). Slightlymore than half of the respondents (50.8%) traveled with friends,and 38.3% traveled with family and/or relatives. The average partysize was 3.5 persons (median¼ 3.0). Respondents stayed anaverage of 4 days (median¼ 4.0) in Las Vegas.

6.3. Qualitative perceptions

Respondents were asked four open-ended questions to learntheir original and unique perceptions regarding the image andpersonality of Las Vegas. These questions were focused on thefollowing areas: (1) general image or characteristics of Las Vegas,(2) personality characteristics associated with Las Vegas, (3) typicalvisitor to Las Vegas, (4) a tourism slogan or tagline for Las Vegas. Forthe first three areas, respondents were asked to list three words or

Table 3Respondents’ perceptions regarding brand personality characteristics of Las Vegas(N¼ 360).

Personality traits n %

Fun/fun-loving 89 24.7Exciting 59 16.4Outgoing 48 13.3Sexy 42 11.7Energetic 30 8.3Adventurous 29 8.0Friendly 21 5.8Alive 18 5.0Flamboyant 17 4.7Rich/wealthy 14 3.9


phrases that first came to their mind when they read the question.For the tagline part, respondents were given free space to writedown a tourism slogan or tagline for Las Vegas in their own words.The findings of these open-ended questions are content analyzed,and the most-frequent responses are presented in tables.

The top ten responses for the general image or characteristics ofLas Vegas are presented in Table 2. As shown in Table 2, Las Vegas ismostly associated with gambling (33.1%) by respondents. This is nota surprising result because Las Vegas is known as a popular gamingdestination in the world. Respondents reported shows (21.5%) mostfrequently after gambling. The third most-frequent response wasfun and entertainment (18.7%). The second (shows) and third (fun/entertainment) most-frequent images associated with Las Vegasshow that Las Vegas is not only perceived as a gaming destination,but also viewed as an entertainment destination in visitors’ minds.Other images associated with Las Vegas by respondents were asfollows: drinking/alcohol (17.9%), casinos/slot machines (17.6), lights/bright lights (11.7%), sex (9.2%), night life/night clubs (7.9%), The Strip(6.2%) and Sin City (4.6%). Although Sin City is at the end of the list,one can say that the Sin City image of Las Vegas is still dominant invisitors’ minds because other images associated with Sin City, suchas gambling, sex, and drinking, are placed in the top of the list.

Of the 368 respondents, eight did not answer the open-endedquestion that was designed to elicit the personality characteristicsassociatedwith Las Vegas. In this question, respondents were askedto think Las Vegas as if it were a person and then list threepersonality traits that reflect Las Vegas. The ten most-frequentpersonality characteristics are shown in Table 3. The most-frequentpersonality characteristic was fun/fun-loving (24.7%), followed byexciting (16.4%) and outgoing (13.3%). Sexy (11.7%), energetic (8.3%),adventurous (8.0%), friendly (5.8%), alive (5.0%), flamboyant (4.7%),and rich/wealthy (3.9%) were the other most-common personalitycharacteristics, respectively. As seen from the table, except excitingand friendly, the open-ended responses for personality character-istics were not as common as Aaker’s (1997) personality traits.Additionally, this open-ended question generated a greater varietyof personality characteristics (191 unique traits). On the basis ofthese findings, one can say that a free-elicitation approach is aneffective way to elicit destination-specific personality traits.

In the third open-ended question, respondents were asked todescribe the typical visitor to Las Vegas using personal adjectives,because Aaker (1997) argues that one of the several direct sourcesof brand personality formation is through the typical user ofa brand. To identify the imagery of a typical visitor to Las Vegas,Sirgy and Su’s (2000) suggested approach was utilized. Accordingto this approach, respondents were asked to visualize and describethe typical visitor to a destination. The most-common open-endedresponses for a typical visitor to Las Vegas are presented in Table 4.The ten most-common descriptors are young (22.0%), followed byfun/fun-loving (17.4), sexy (15.5%), rich/wealthy (14.9), old (14.1%),

Table 2Respondents’ images of Las Vegas (N¼ 368).

Images n %

Gambling 122 33.1Shows 79 21.5Fun/entertainment 69 18.7Drinking/alcohol 66 17.9Casinos/slot machines 65 17.6Lights/bright lights/neon lights 43 11.7Sex 34 9.2Night life/night clubs 29 7.9Las Vegas Strip 23 6.2Sin City 17 4.6

gambler (13.0), stylish (12.8%), adventurous (12.5), classy (11.7%), andpartier (8.7%).

Aaker (1997) proposes that brand personality can be formed intwo ways: directly (through people) and indirectly. In an indirectmanner, personality traits can be associated with a brand throughproduct-related attributes, such as packaging, brand name, symbol,logo, or advertising. Within the tourism destinations, it has beenproposed that brand personality can be formed in an indirectmanner through marketing programs such as cooperative adver-tising or media-construction of destinations (Cai, 2002; Ekinci &Hosany, 2006). Accordingly, this study argues that one of theindirect ways of destination personality formation is throughdestination slogans or taglines. Furthermore, these slogans or tag-lines can be used to understand the brand personality of destina-tions. More than one-fourth of the respondents (99 respondents,26.90%) answered to the last open-ended question and wrote downa tagline for Las Vegas. These taglines were content analyzed tounderstand the respondents’ perceptions regarding the personalityof Las Vegas. The words, especially the personal adjectives, used inthe taglines were analyzed. Content analysis of the taglinesrevealed that the respondents mostly associate Las Vegas with thepersonality trait exciting (14). Other personality characteristics thatwere emphasized in the taglines were: free (9), fun (9), showy (7),unique (6), sexy (5), and alive (3).

6.4. Exploratory factor analysis of destination personality items

An exploratory factor analysis was performed on the 29personality items to reduce data and to identify the underlyingdimensions. Principal component analysis, with varimax rotationand latent root criterion (eigenvalues> 1), was used in the factoranalysis. As recommended by Hair, Black, Babin, Anderson, andTatham (2005), factor loadings greater than .50 are considerednecessary for practical significance; therefore, a cutoff point of .50was established to include items in the interpretation of a factor. Inthe initial analysis, one of the five factors included single item

Table 4Respondents’ perceptions of a typical visitor to Las Vegas (N¼ 368).

Personality traits n %

Young 81 22.0Fun/fun-loving 64 17.4Sexy 57 15.5Rich/wealthy 55 14.9Old 52 14.1Gambler 48 13.0Stylish 47 12.8Adventurous 46 12.5Classy 43 11.7Partier 32 8.7


(naughty). The analysis was conducted again after excluding it. Inthe second factor analysis, 4 items exhibited both cross-loadingsand low factor loadings (<.50) and were removed. The itemseliminated from the second analysis were original, spirited, cool, andcontemporary. After removing these items, the analysis wasrepeated. All items exhibited factor loadings greater than .50, andno items were cross-loaded.

The results of Bartlett’s Test of Sphericity (p value .000, chi-square 5631.535, df 276) showed that sufficient correlations existamong the variables to run a factor analysis. The Kai-sereMeyereOlkin Measure of Sampling Adequacy (.920) wasexcellent, indicating that the principal component analysis wasvery appropriate to use on the data. The latent root criterion(eigenvalues> 1) revealed five-factor solution and explained 69.6%of the variance. Cronbach’s alpha was used to test the reliabilities.The reliability of the items was very satisfactory, ranging from .812to .915. Table 5 displays the factors, factor loadings, eigenvalues, thepercentage of variance explained by the factors, and the corre-sponding Cronbach’s alpha reliability coefficients.

A name was assigned for each factor based on the followingcriteria: factor loadings, the nature of the items in each factor, andcomparison of the items and factor names extracted in this studywith those in Aaker’s (1997) study. According to Hair et al. (2005),items with higher loadings are considered more important andhave greater influence on the factor labeling. Destination person-ality factor one consisted of “energetic, alive, vibrant, showy, exciting,sexy, and daring”. Thus, factor one was labeled “vibrancy”.

Table 5Exploratory factor analysis of destination personality items.a

Factors Factorloading

Eigenvalue Explainedvariance (%)

Reliabilityb

Vibrancy 10.835 45.147 .915Energetic .823Alive .821Vibrant .809Showy .738Exciting .585Sexy .584Daring .555

Sophistication 2.142 8.924 .867Feminine .771Charming .735Upper class .675Good looking .649Glamorous .562

Competence 1.516 6.317 .844Leader .823Successful .808Confident .769Independent .564Intelligent .514

Contemporary 1.133 4.721 .812Unique .676Up-to-date .667Imaginative .630Young .554Trendy .525

Sincerity 1.079 4.498 .833Friendly .832Cheerful .682

Total varianceexplained

69.607

a Extraction method: principal component analysis, rotation method: varimaxwith Kaiser normalization, KaisereMeyereOlkin Measure of Sampling Adequacy:.920. Bartlett’s Test of Sphericity p value .000 (chi-square: 5631.535, df: 276).

b Reliabilities were assessed using Cronbach’s alpha coefficients.

Destination personality factor two consisted of “feminine, charming,upper class, good looking, and glamorous”. Factor two was named“sophistication” because the items in factor two were similar tothose in Aaker’s (1997). Destination personality factor threeincluded “leader, successful, confident, independent, and intelligent”.Factor three was labeled “competence” given that four of the fiveitems were similar to those in Aaker’s (1997) study. Destinationpersonality factor four consisted of “unique, up-to-date, imaginative,young, and trendy”. Although the items in factor four were includedin the “excitement” dimension in Aaker’s (1997) study, it wasnamed as “contemporary” rather than “excitement” in this study,given that the items (e.g., exciting, daring) which had greaterinfluence on the factor name “excitement” were not included infactor four. Destination personality factor five consisted of “friendlyand cheerful”. For factor five, the name “sincerity” was chosenbecause these items were also loaded on the “sincerity” dimensionin Aaker’s (1997) study.

As seen in Table 5, Aaker’s (1997) original five personalitydimensions were replicated to a great extent. In three of the fivefactors (sophistication, contemporary, and sincerity), the personalityitems were loaded under the original dimensions of Aaker’s (1997)study (the factor “contemporary” corresponds to Aaker’s (1997)“excitement” dimension). In one factor (competence), the person-ality items were also similar to those in Aaker’s (1997) study, butthe item “independent”dwhich was loaded on “excitement” inAaker’s (1997) studydwas loaded on the “competence” dimensionrather than “contemporary”. Because Aaker’s (1997) “ruggedness”dimension failed in the content validity stage, it was not used in thestudy. Instead, destination-specific personality traits were used,such as energetic, sexy, alive, vibrant, showy, and naughty. Thesedestination-specific personality items were loaded on one factor,except the item “naughty” (failed in the initial factor analysis).However, “exciting” and “daring” were also loaded on the vibrancyfactor, which includes the destination-specific items.

In Table 6, the personality dimensions extracted in this studywere compared with those in Aaker’s (1997) study. Similar toAaker’s (1997) findings, five dimensions of brand personalityemerged for Las Vegas as a tourist destination. In general, four ofthe five factors appear to replicate the original dimensions ofAaker’s (1997) brand personality framework. Thus, the findings ofthis study indicate that Aaker’s (1997) brand personality frame-work is applicable to tourism destinations. However, a fifth factorappears to be relatively specific to Las Vegas. This specific factorexplained the majority of variance (45.1%). Thus, this studyproposes that a specific BPS is needed for tourism destinations,consistent with the findings of Hosany et al. (2006), who suggestthat some items of Aaker’s (1997) BPS are redundant and notsuitable for tourism destinations.

The factor scores were calculated using the Anderson and Rubinmethod with SPSS 16.0, because this method was found to be

Table 6A comparison of destination personality (DP) and brand personality (BP)dimensions.

The presentstudy (DPdimensions)

Comparison Aaker (1997)(BP dimensions)

Vibrancy Vibrancy is destination specific.Sophistication )Corresponds/ SophisticationCompetence )Corresponds/ CompetenceContemporary )Corresponds/ ExcitementSincerity )Corresponds/ Sincerity

Aaker’s (1997) ruggedness dimensionwas not used in this study becauseit failed in the content validity stage.

Ruggedness

Table 8Regression analysis: relationship between destination personality and behavioralintentions.

Dimensions Model 1 Intention to return Model 2 Intentionto recommend

Beta t-Value Sig. t Beta t-Value Sig. t

Vibrancy .208 4.332 .000 .280 6.038 .000Sophistication .317 6.605 .000 .322 6.942 .000Competence .108 2.242 .026 .171 3.689 .000Contemporary .084 1.746 .082 .105 2.269 .024Sincerity .283 5.895 .000 .273 5.902 .000Constant 48.867 .000 65.375 .000Multiple R .493 .545R2 .243 .297F test statistics/

significanceF¼ 21.046,p¼ .000

F¼ 27.649,p¼ .000

Table 9Regression analysis: relationship between self-congruity and behavioral intentions.

IVs Model 1 Intentionto return

Model 2 Intentionto recommend


unbiased and superior to the other two methods (the regressionand Bartlett’s method) used to estimate factor scores (Lastovicka &Thamodaran,1991; Sirakaya, Uysal, & Yoshioka, 2003). Additionally,the Anderson and Rubin method is the standard practice whenfactors are used as an input for subsequent analyses (multipleregression analyses, in this case) (Sirakaya et al., 2003).

6.5. Analyses of self-congruity measures

Respondents were asked to indicate their levels of agreement tothe six self-congruity statements. Because the factor structure ofthese statements was known, they were not submitted to factoranalysis; only the reliability scores were computed for thesemeasures. Both actual and ideal self-congruity statements demon-strated a strong internal consistency, shown by a¼ .984 anda¼ .985, respectively (Table 7). Following the reliability assessment,the six measures of self-congruity were reduced into two variablesas actual and ideal congruity, by computing their mean scores.

6.6. The model and hypothesis testing

The proposed model and the relationships between destinationpersonality, self-congruity, and behavioral intentions were testedusing several multiple regression analyses.

6.6.1. Destination personality perceptions and behavioral intentionsMultiple regression analyses were run using the five factors of

the destination personality as independent variables and thebehavioral intentions as dependent variables. The results are pre-sented in Table 8. As seen in Table 8, destination personalitydimensions were statistically significant in estimating intention toreturn and intention to recommend (p values¼ .000).

ThemultipleR coefficients showed that the correlations betweenthe destination personality and the two behavioral intentions arestrong to moderate (R values> .30) (Cohen, 1988). In Model 1, thecoefficient of determination (R2) was .243, indicating that approxi-mately 24% of the total variation in intention to returnwas explainedby the destination personality factors. On the other hand, the R2 was.297 inModel 2, whichmeans that 29.7% of the total variance for theestimation of intention to recommend is explained by the destina-tion personality factors. These results show that destinationpersonality is more predictive in the estimation of intention torecommend than intention to return since the R2 was higher inModel 2 than in Model 1. There was no effect of multicollinearity inboth models based on the fact that all VIF scores were 1.000.

The four personality dimensions, vibrancy (b¼ .208, p¼ .000),sophistication (b¼ .317, p¼ .000), competence (b¼ .108, p¼ .026),and sincerity (b¼ .283, p¼ .000), had a significant and positiveimpact on intention to return. Only the contemporary dimensionwas not statistically significant (b¼ .084, p¼ .082). However, all thefive dimensions of the destination personality were statisticallysignificant at .050 or better probability level and had a positiveimpact on intention to return. Based on the multiple regressionanalyses, the Hypotheses 1 and 2, the destination personality willhave a positive impact on intention to return and intention torecommend, were supported, respectively.

Table 7Reliability estimates for self-congruity statements (N¼ 368).

Self-congruity Reliabilitya Number of items

Actual congruity .984 3Ideal congruity .985 3

a Reliabilities were assessed using Cronbach’s alpha coefficients.

6.6.2. Self-congruity and behavioral intentionsIn this stage, both intention to return and intention to recom-

mend were regressed on the two measures of self-congruity. Theresults are presented in Table 9. The multiple R coefficients showedthat the correlations between the two measures of self-congruityand the twobehavioral intentions are strong (Rvalues> .50) (Cohen,1988). The two measures of self-congruity explained 53.4% and52.5% of the variation in intention to return and intention torecommend, respectively. The two regression models predictingintention to return and intention to recommend were significant atthe .001 or better probability level, indicating that there wasa significant relationship between self-congruity and behavioralintentions. Multicollinearity between the independent variableswas examined by calculating the VIFs. No effect of multicollinearityproblemwasdetected inbothmodels as all theVIF scoreswere1.199.

Both actual and ideal self-congruity were found to be significantpredictors of intention to return (bactual¼ .502, p¼ .000;bideal¼ .364, p¼ .000) and intention to recommend (bactual¼ .382,p¼ .000; bideal¼ .479, p¼ .000). These findings not only support theresults of many previous research studies in the consumer behaviorliterature, which have found similar effects of self-congruity onconsumer attitudes (see Sirgy, 1982 for a review), but also providesupport for Sirgy and Su’s (2000) arguments regarding the effects ofself-congruity in the context of tourism literature. Sirgy and Su(2000) propose that the greater the match between destinationimage and tourist’s self-concept, themore likely it is that the touristwill be motivated to visit that destination.

In Hypotheses 3 and 4, respectively, self-congruity washypothesized to have a positive impact on intention to return andintention to recommend. Based on the results of the multipleregression analyses, Hypotheses 3 and 4 were supported. Accord-ingly, the greater the match between destination personality andtourist’s self-concept, the more likely it is that the tourist will havea favorable attitude toward that destination, resulting in intentionto return and intention to recommend.


Actual congruity .502 12.829 .000 .382 9.676 .000Ideal congruity .364 9.299 .000 .479 12.114 .000Constant 5.284 .000 12.474 .000Multiple R .731 .724R2 .534 .525F test statistics/


F¼ 201.436,p¼ .000


In particular, the standardized beta coefficients were reviewedto find out whichmeasure of self-congruity has relative importanceon behavioral intentions. As seen in Table 9, actual self-congruityhas a relatively greater impact on intention to return (b¼ .502),whereas ideal self-congruity has relatively greater impact onintention to recommend (b¼ .479).

6.6.3. Destination personality perceptions, self-congruity, andbehavioral intentions

Hypotheses 5 and 6 dealt with the mediating impact of self-congruity on the relationship between destination personality andtourist’s behavioral intentions. To test these hypotheses, Baron andKenny’s (1986) approach for determining mediation was used,which is summarized as follows:

Step 1: Regressing the dependent variable on the independentvariable. In this step, the independent variable must affect thedependent variable. If not, mediation is not possible, and there isno need for further analyses.Step 2: Regressing the mediator on the independent variable. Inthis second step, the independent variable must affect themediator.Step 3: Regressing the dependent variable on both the inde-pendent variable and the mediator. The mediator must affectthe dependent variable.Step 4: If all the above conditions are met in the predicteddirection, then the effect of the independent variable on thedependent variablemust be less in the third step than in the firststep. If not, mediation is not supported. If the independentvariable is no longer significant in the third step, then perfectmediation is supported.

It should also be noted that not only the significance of thecoefficients, but also their absolute sizes should be examined in thefinal step to evaluate the mediation (Baron & Kenny, 1986; Kenny,2008). According to Hair et al. (2005), if the effect of the indepen-dent variable on the dependent variable is reduced in absolute sizebut remains significant when the mediator is included in themodel, partial mediation is supported. If the effect of independentvariable on the dependent variable is reduced to a point where it isnot significantly different from zero when mediator is included inthe model, full mediation is supported (Hair et al., 2005).

In Hypotheses 1 and 2, behavioral intentionswere regressed on alldimensionsofdestinationpersonalityanddestinationpersonalitywasfound to have a statistically significant effect on behavioral intentions(p values¼ .000). This denotes the first step in Baron and Kenny’s(1986) suggested approach, in which we have sufficient evidencethat this relationship might be mediated by another variable.

Table 10Regression analysis: relationship between destination personality and self-congruity

Dimensions Model 1 Actual congruity Model 2 Ideal congruity


Vibrancy .132 2.567 .011 .204 4.163 .000Sophistication .181 3.515 .001 .277 5.651 .000Competence .073 1.419 .157 .144 2.942 .003Contemporary .102 1.971 .050 .044 .896 .371Sincerity .253 4.903 .000 .264 5.387 .000Constant 40.391 .000 53.131 .000Multiple R .360 .459R2 .130 .211F test statistics/


F¼ 17.584,p¼ .000

In the second step of Baron and Kenny’s (1986) suggestedapproach, both actual and ideal self-congruity were regressed on allthe dimensions of destination personality. The results are reportedin Table 10. The two models investigated were found to be signif-icant at the .001 or better probability level, indicating that therewas a significant relationship between destination personality andself-congruity, which is consistent with the findings of Murphyet al. (2007a). The multiple R coefficients showed that the corre-lation between the dimensions of destination personality and thetwo measures of self-congruity are moderate (.30< R values< .50)(Cohen, 1988). The five dimensions of destination personalityexplained 13.0% and 21.1% of the total variation in actual and idealself-congruity, respectively. Although the amount of varianceexplained by the regressionmodel for actual self-congruity was low(R2¼ .130), the F value was highly significant (p¼ .000). Only theCompetence dimension was not statistically significant in predict-ing actual self-congruity (p¼ .157). On the other hand, theContemporary dimension was not statistically significant in pre-dicting ideal self-congruity (p¼ .371). However, four of the fivedestination personality dimensions were statistically significant inboth models. Because the overall regression models were signifi-cant at .001 or better probability level, the second step in Baron andKenny’s (1986) approach was completed, and the required condi-tion (the effect of independent variable on the mediator) was met.

In the third step, both intention to return and intention torecommend were regressed on both the five dimensions of desti-nation personality and the two measures of self-congruity. Theresults of the multiple regression analyses are shown in Table 11.The regression models that included both the independent variable(dimensions of destination personality) and the mediator (actualand ideal self-congruity) were statistically significant in estimatingintention to return and intention to recommend (p values¼ .000).

The multiple R coefficients showed that the correlationsbetween the destination personality, self-congruity, and thebehavioral intentions are strong (R values> .50) (Cohen, 1988). InModel 1, the coefficient of determination (R2) was .711, indicatingthat 71.1% of the total variation in intention to returnwas explainedby both the dimensions of destination personality and the twomeasures of self-congruity. On the other hand, the R2 was .664 inModel 2, which means that 66.4% of the total variance in the esti-mation of intention to recommend was explained by both thedimensions of destination personality and the two measures ofself-congruity.

It should be also noted that the explanatory power (R2) of themodels is increasedwhen actual and ideal self-congruity are added tothe equation (R2¼ .243 in Step 1, R2¼ .711 in Step 3 for intention toreturn; R2¼ .297 in Step 1, R2¼ .664 in Step 3 for intention to

Table 11Regression analysis: relationship between destination personality, self-congruity,and behavioral intentions.

IVs and mediator Model 1 Intentionto return

Model 2 Intentionto recommend


Vibrancy .052 1.705 .089 .140 4.246 .000Sophistication .105 3.370 .001 .132 3.910 .000Competence .003 .094 .925 .076 2.332 .020Contemporary .031 1.031 .303 .062 1.907 .057Sincerity .060 1.906 .058 .078 2.295 .022Actual congruity .265 7.208 .000 .184 4.633 .000Ideal congruity .592 15.312 .000 .565 13.539 .000Constant 2.628 .009 11.333 .000Multiple R .843 .815R2 .711 .664F test statistics/


F¼ 92.032,p¼ .000


recommend). The potential multicollinearity problemwas examinedby using VIFs. The VIFs ranged from 1.012 to 1.690, indicating thatthere were no concerns regarding the multicollinearity problem.

The third step of testing mediation required that self-congruity(mediator) affect the tourist’s behavioral intentions (dependentvariables). In Table 11, the regression coefficients representing theeffect of actual and ideal self-congruity on intention to return andintention to recommend were statistically significant (p val-ues¼ .000). These findings satisfied the third step of Baron andKenny’s (1986) test of mediation.

The final step for mediation in Baron and Kenny’s (1986)approach required that the regression coefficients representingthe independent variable’s effects on the dependent variable belower in magnitude in the regression equations that included themediator (regression results shown in Table 11) than the regressioncoefficients obtained from the regression equations that excludedthe mediator (regression results shown in Table 8). A comparison ofthe regression coefficients and the p values (results in Table 8 vs.Table 11) for the regression analyses in Step 1 and Step 3 are pre-sented in Table 12.

The three dimensions of destination personality (vibrancy,competence, and sincerity) were no longer significant in predictingintention to return when the two measures of self-congruity werecontrolled (Table 12). The contemporary dimension was notsignificant in Step 1, nor in Step 3, for intention to return. Only thesophistication dimension was still significant when the twomeasures of self-congruity were controlled.

On the contrary, four of the five destination personality dimen-sions (vibrancy, sophistication, competence, and sincerity) were stillsignificant in predicting intention to recommend when the twomeasures of self-congruity were controlled. Only the contemporarydimension was no longer significant in predicting intention torecommend when actual and ideal self-congruity were controlled.

A comparison of the regression coefficients showed that theeffects of all the dimensions of destination personality on intentionto return and intention to recommend were all lower in magnitudewhen the two measures of self-congruity were controlled. Thus,Hypotheses 5 and 6, proposing that self-congruity mediates therelationship between destination personality and behavioralintentions, were supported.

Because all the regression coefficients of the destinationpersonality dimensions are reduced, but some of them still remainsignificant when self-congruity is included as a mediator, it isconcluded that self-congruity is a partial mediator between desti-nation personality and behavioral intentions.

Table 12A comparison of regression coefficients and p values for destination personalitydimensions between Step 1 and Step 3.

IVs Model 1 Intention to return Model 2 Intentionto recommend

Step 1 Step 3 Step 1 Step 3

Beta Sig. t Beta Sig. t Beta Sig. t Beta Sig. t

Vibrancy .208 .000 .052 .089 .280 .000 .140 .000Sophistication .317 .000 .105 .001 .322 .000 .132 .000Competence .108 .026 .003 .925 .171 .000 .076 .020Contemporary .084 .082 .031 .303 .105 .024 .062 .057Sincerity .283 .000 .060 .058 .273 .000 .078 .022

Step 1: regression analyses predicting the behavioral intentions (dependent vari-able) with destination personality (independent variable).Step 3: regression analyses predicting the behavioral intentions (dependent vari-able) with destination personality (independent variable) and self-congruity(mediator).

7. Conclusions

The purpose of this study was to investigate the perceiveddestination personality of Las Vegas and to empirically examine therelationships among destination personality, self-congruity, andtourist’s behavioral intentions. The results of the present studymake important theoretical and practical contributions to theunderstanding of brand personality, self-congruity, and behavioralintentions in the context of tourism destinations.

7.1. Implications

Froma theoretical standpoint, the study indicates that tourists doattribute personality characteristics to tourism destinations. This isin linewith previous research ondestinationpersonality (e.g., Ekinci& Hosany, 2006; Ekinci et al., 2007;Murphy et al., 2007a). Similar toAaker’s (1997) brand personality framework, five dimensions ofdestination personality emerged for Las Vegas in this study.However, this study partially replicates Aaker’s (1997) original fivepersonality dimensions, considering that four of thefive dimensionsappear to replicate those in Aaker’s (1997) study. In three of the fivefactors (sophistication, contemporary, and sincerity), the personalityitems were located under the original dimensions of Aaker’s (1997)study (the factor “contemporary” corresponds to Aaker’s (1997)“excitement” dimension). In one dimension (competence), four ofthe five personality traits were the same as those in Aaker’s (1997)study. However, one trait (independent), which was located underthe “excitement” dimension in Aaker’s (1997) study, was loaded onthe competence dimension in this study. Another dimension,vibrancy, included five destination-specific personality traits andtwo traits from Aaker’s (1997) BPS. These two traits, exciting anddaring, shifted from the “excitement” dimension of Aaker’s (1997)study and became a part of the vibrancy dimension in this study.The shifting of the personality traits from one dimension to anotherwas also observed in previous research (e.g., Ekinci & Hosany, 2006;Murphy et al., 2007a). Ekinci and Hosany (2006) explain this issuewith the argument that the personality traits designed for consumergoods tend to shift when applied to tourism destinations. It shouldalso be noted that Aaker’s (1997) “ruggedness” dimension was notused in this study because it failed in the content validity stage usedfor identifying the personality traits to be included.

Furthermore, the results support Ekinci and Hosany’s (2006)argument that Aaker’s (1997) BPS may not fully represent allpersonality traits associated with tourism destinations. Indeed,destination-specific personality traits were loaded on one dimen-sion and explained the majority of the variance. Additionally, theopen-ended responses revealed personality traits that are quitedifferent from those in Aaker’s (1997) study.

The findings of this study indicate that destination personalityhas a positive impact on intention to return and intention torecommend, consistent with previous research (Ekinci & Hosany,2006; Ekinci et al., 2007). Another significant theoretical contribu-tion of this study is that the findings support the self-congruitytheory (the match between destination personality and tourist’sself-concept) in the context of tourism destinations. Although self-congruity has been studied widely in the consumer behavior liter-ature, there is a lack of research on it in the tourism literature. Thestudy finds evidence that self-congruity has a positive impact ontourist’s behavioral intentions and thus supports the self-congruitytheory. Thus, the greater thematch betweendestinationpersonalityand tourist’s self-concept, themore likely is that the touristwill havea favorable attitude toward that destination, resulting in intention toreturn and word of mouth. Additionally, the study shows evidencethat self-congruity is a partial mediator between destinationpersonality and behavioral intentions. In other words, destination


personality has a positive indirect effect on intention to return andintention to recommend through self-congruity.

Another theoretical implication of this study is methodological.The study shows that both qualitative and quantitative approachesshould be used in the measurement of destination personality,which is similar with the findings of the study of Baloglu and Love(2005). Baloglu and Love (2005) investigated the associationmeeting planners’ images of five convention cities and found thatopen-ended (unstructured) questions revealed unique perceptionsthat could not be captured by closed-ended (structured) questions.In this study, the open-ended questions generated a greater varietyof personality traits that are quite different from those in Aaker’s(1997) study. In addition, they converged in part with the quanti-tative responses (scale evaluations) and thus provided additionalsupport for the validity of the study.

From a practical standpoint, the findings of the study provideimportant implications fordestinationmarketers. Today, destinationsare faced with increasingly tough competition than they ever facedbefore. Promoting the functional attributes of destinations no longerhelps destinations to attract travelers because of the high productsimilarity and growing substitutability. The findings of this studyprovide evidence that the symbolic functions or benefits of a desti-nation brand are crucial in understanding the complex nature oftravel behavior. Indeed, the results indicate that destination person-ality has a positive impact on tourist’s behavioral intentions. Thus,destination marketers should focus on developing marketing strate-gies that emphasize the distinctive personality of their destinations.

The study has also specific practical implications for the desti-nationmarketers of LasVegas. Theperceiveddestinationpersonalityof Las Vegas has five dimensions: vibrancy, sophistication, compe-tence, contemporary, and sincerity. Although all the five personalitydimensions of Las Vegas were found to be significant predictors ofintention to recommend, only the contemporary dimensionwas notsignificant in predicting intention to return. Destination marketersof Las Vegas could either differentiate Las Vegas based on thesepersonality dimensions or these dimensions can be used in thepositioning efforts of Las Vegas. In particular, three of the fivepersonality dimensions were found to have relatively more influ-ence on tourists’ behavioral intentions. The sophisticationwas foundto be the most influential personality dimension affecting bothintention to return and intention to recommend. The second andthird most influential dimensions on intention to return weresincerity and vibrancy, respectively. However, the second and thirdmost influential dimensions on intention to recommend werevibrancy and sincerity, respectively. The destinationmarketers of LasVegas should concentrate more on these three dimensions (sophis-tication, vibrancy, and sincerity) in their marketing efforts.

An additional practical implication is that tourists who experi-ence a match between how they see the destinations and them-selves or how they would like to see themselves are more likely tohave favorable attitudes toward those destinations, resulting inintention to return and intention to recommend. Therefore, desti-nation marketers should place greater emphasis on buildinga connection between destination personality and tourist’s self-concept and develop marketing campaigns emphasizing thismatch. The combination of destination personality and self-congruity might provide a more comprehensive understanding ofhow visitors choose their destinations.

7.2. Limitations and future research

As any other study, the present study has some limitations thathave to be taken into account when considering the findings. Thefirst and the most significant limitation of this study is the lack ofrandom sampling. The data were collected via convenience

sampling method. Therefore, the sample did not reflect the entirepopulation of visitors to Las Vegas from which the respondentswere chosen. Second, the findings of this study are specific to onetourism destination (Las Vegas) and cannot be generalized to othertourism destinations. Therefore, a future research replicating thisstudy with a larger sample size, applying the random samplingmethod, and for other destinations will increase our understandingof this important research area. Third, the present study focused onthe visitors to Las Vegas, and the results are limited to the timeperiod of data collection. The samplewas surveyed in themonths ofJanuary, February, and March 2009. Thus, for more generalizableresults, it is advisable that the sample is surveyed throughout thewhole year to prevent any possible seasonal bias.

Fourth, this study measured self-congruity directly, using theglobal measurement method developed by Sirgy et al. (1997).However, there has been a considerable debate about whether touse the direct score formula or gap score formula in measuring self-congruity. Future research could measure self-congruity using thegap score formula (measuring self-concept and brand personalityseparately) or could employ both the direct score and gap scoreformulas and compare the results.

Fifth, the current study employed only the two dimensions ofself-congruity, namely, actual self-congruity and ideal self-congruity. The other two dimensions, social self-congruity andideal social self-congruity, were not included in this study. It shouldbe noted that social self-congruity has been increasingly drawingattention, especially in the tourism area, because the destinationchoice behavior is not only affected by personal factors, but alsoinfluenced by social factors. Thus, future research could investigatethe effects of social self-congruity within the context of tourismdestinations. Also, a future research agenda could investigate theimpact of trip-related factors (e.g. purpose of the trip, tripcompanions) or demographics (e.g. age, nationality) on the rela-tionship between self-congruity and tourist’s behavioral intentions.For instance, Sirgy and Su (2000) propose that tourists who travelwith significant others are likely to engage in impressionmanagement and, therefore, choose to visit those destinations thatwould make good impression on significant others.

Although the study found that Las Vegas has a five-dimensionaldestination personality, both the qualitative and quantitativeresponses found support for the argument that Aaker’s (1997) BPSmay not fully represent the personality traits associated withtourism destinations. Indeed, open-ended questions elicitedpersonality traits that are quite different from those in Aaker’s(1997) BPS, and exploratory factor analysis showed that themajority of variance was captured by the vibrancy factor, whichmostly consists of personality traits that are different from Aaker’s(1997) BPS. Thus, a BPS that is specifically designed for tourismdestinations is essential. Future research could fill this importantgap in the tourism literature by developing a valid, reliable, andgeneralizable destination personality scale.

Appendix. Supplementary material

Supplementary data associated with this article can be found inthe online version, at doi:10.1016/j.tourman.2010.06.006.

References

Aaker, D. A. (1996). Building strong brands. New York: Free Press.Aaker, J. L. (1995). Brand personality: Conceptualization, measurement and underlying

psychological mechanism (doctoral dissertation). Available from ProQuestDissertations and Theses database. (UMI No. 9602828).

Aaker, J. L. (1997). Dimensions of brand personality. Journal of Marketing Research,34, 347e356.



Aaker, J. L. (1999). The malleable self: the role of self-expression in persuasion.Journal of Marketing Research, 36, 45e57.

Austin, J. R., Siguaw, J. A., & Mattila, A. S. (2003). A re-examination of the general-izability of the Aaker brand personality measurement framework. Journal ofStrategic Marketing, 11, 77e92.

Azevedo, A., & Pessoa, F. (2005). Clothing branding strategies. Journal of Textile andApparel, Technology and Management, 4(3), 1e13.

Azoulay, A., & Kapferer, J. (2003). Do brand personality scales really measure brandpersonality? Journal of Brand Management, 11(2), 143e155.

Baloglu, S., & Love, C. (2005). Association meeting planners’ perceptions andintentions for five major US convention cities: the structured and unstructuredimages. Tourism Management, 26(5), 743e752.

Baron, R. M., & Kenny, D. A. (1986). The moderatoremediator variable distinction insocial psychological research: conceptual, strategic, and statistical consider-ations. Journal of Personality and Social Psychology, 51(6), 1173e1182.

Belk, R. W. (1988). Possessions and the extended self. Journal of Consumer Research,2, 139e168.

Birdwell, A. E. (1968). A study of the influence of image congruence on consumerchoice. Journal of Business, 41(1), 76e88.

Blain, C., Levy, S. E., & Ritchie, B. (2005). Destination branding: insights and prac-tices from destination management organizations. Journal of Travel Research, 43,328e338.

Cai, L. A. (2002). Cooperative branding for rural destinations. Annals of TourismResearch, 29(3), 720e742.

Churchill, G. A. (1979). A paradigm for developing better measures of marketingconstructs. Journal of Marketing Research, 16, 64e73.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.).New Jersey: Lawrence Erlbaum Publishing.

Crockett, S. R., & Wood, L. J. (2004). Western Australia: building a state brand. InN. Morgan, A. Pritchard, & R. Pride (Eds.), Destination branding creating theunique destination proposition (2nd ed.). (pp. 185e206) Oxford, England:Elsevier.

D’Astous, A., & Boujbel, L. (2007). Positioning countries on personality dimensions:scale development and implications for country marketing. Journal of BusinessResearch, 60, 231e239.

Dolich, I. J. (1969). Congruence relationships between self-images and productbrands. Journal of Marketing Research, 6(1), 80e84.

Ekinci, Y., & Hosany, S. (2006). Destination personality: an application of brandpersonality to tourism destinations. Journal of Travel Research, 45, 127e139.

Ekinci, Y., Sirakaya-Turk, E., & Baloglu, S. (2007). Host image and destinationpersonality. Tourism Analysis, 12, 433e446.

Fournier, S. (1998). Consumers and their brands: developing relationship theory inconsumer research. Journal of Consumer Research, 24, 343e373.

Gnoth, J. (1998). Branding tourism destinations. Annals of Tourism Research, 25,758e760.

Gnoth, J., Baloglu, S., Ekinci, Y., & Sirakaya-Turk, E. (2007). Introduction: buildingdestination brands. Tourism Analysis, 12(5/6), 339e343.

Hair, J. F., Black, W. C., Babin, B. J., Anderson, R. E., & Tatham, R. L. (2005).Multivariatedata analysis (6th ed.). New Jersey: Prentice Hall.

Helgeson, J. G., & Suphellen, M. (2004). A conceptual and measurement comparisonof self-congruity and brand personality. International Journal of Market Research,46(2), 205e233.

Henderson, J. C. (2000). Selling places: the new Asia-Singapore brand. The Journal ofTourism Studies, 11(1), 36e44.

Hosany, S., Ekinci, Y., & Uysal, M. (2006). Destination image and destinationpersonality: an application of branding theories to tourism places. Journal ofBusiness Research, 59, 638e642.

Hosany, S., Ekinci, Y., & Uysal, M. (2007). Destination image and destinationpersonality. International Journal of Culture, Tourism and Hospitality Research, 1(1), 62e81.

Keller, K. L. (1993). Conceptualising, measuring and managing customer-basedbrand equity. Journal of Marketing, 57, 1e22.

Keller, K. L. (1998). Building, measuring and managing brand equity. New Jersey:Prentice Hall.

Kenny, D. A. (2008). Mediation. Retrieved 20.03.09, from. http://davidakenny.net/cm/mediate.htm#ST.

Landon, E. L. (1974). Self-concept, ideal self-concept and consumer purchaseintentions. Journal of Consumer Research, 1, 44e51.

Lastovicka, J. L., & Thamodaran, K. (1991). Common factor score estimates inmultiple regression problems. Journal of Marketing Research, 28, 105e112.

Low, G. S., & Lamb, C. W. (2000). The measurement and dimensionality of brandassociations. Journal of Product and Brand Management, 9(6), 350e368.

Malhotra, N. K. (1988). Self-concept and product choice: an integrated perspective.Journal of Economic Psychology, 9, 1e28.

Morgan, N., Pritchard, A., & Pride, R. (Eds.). (2004). Destination branding: Creating theunique destination proposition (2nd ed.). Oxford: Elsevier.

Murphy, L., Benckendorff, P., & Moscardo, G. (2007a). Destination brand personality:visitor perceptions of a regional tourism destination. Tourism Analysis, 12,419e432.

Murphy, L., Benckendorff, P., & Moscardo, G. (2007b). Linking travel motivation,tourist self-image and destination brand personality. Journal of Travel & TourismMarketing, 22(2), 45e59.

Murphy, L., Moscardo, G., & Benckendorff, P. (2007). Using brand personality todifferentiate regional tourism destinations. Journal of Travel Research, 46, 5e14.

Phau, I., & Lau, K. C. (2000). Conceptualising brand personality: a review andresearch propositions. Journal of Targeting, Measurement and Analysis forMarketing, 9(1), 52e69.

Pike, S., & Ryan, C. (2004). Destination positioning analysis through a comparison ofcognitive, affective, and conative perceptions. Journal of Travel Research, 42,333e342.

Pitt, L. F., Opoku, R., Hultman, M., Abratt, R., & Spyropoulou, S. (2007). What I sayabout myself: communication of brand personality by African countries.Tourism Management, 28(3), 835e844.

Plummer, J. T. (1985). How personality makes a difference. Journal of AdvertisingResearch, 24(6), 27e31.

Prayag, G. (2007). Exploring the relationship between destination image and brandpersonality of a tourist destination: an application of projective techniques.Journal of Travel and Tourism Research, Fall, 111e130.

Rosenberg, M. (1979). Conceiving the self. New York: Basic Books, Inc.Sahin, S., & Baloglu, S. (2009, January). Brand personality and destination image of

Istanbul: A comparison across nationalities. Paper presented at the 14th annualgraduate student research conference in hospitality and tourism, Las Vegas, NV.

Santos, C. A. (2004). Framing Portugal: representational dynamics. Annals of TourismResearch, 31(1), 122e138.

Sirakaya, E., Uysal, M., & Yoshioka, C. F. (2003). Segmenting the Japanese tourmarket to Turkey. Journal of Travel Research, 41, 293e304.

Sirgy, M. J. (1982). Self-concept in consumer behavior: a critical review. Journal ofConsumer Research, 9, 287e300.

Sirgy, M. J. (1985a). Self-image/product-image congruity and consumer decisionmaking. International Journal of Management, 2, 49e63.

Sirgy, M. J. (1985b). Using self-congruity and ideal congruity to predict purchasemotivation. Journal of Business Research, 13, 195e206.

Sirgy, M. J., Grewal, D., Mangleburg, T. F., Park, J., Chon, K., Claiborne, C. B., et al.(1997). Assessing the predictive validity of two methods of measuring self-image congruence. Journal of the Academy of Marketing Science, 25(3), 229e241.

Sirgy, M. J., Johar, J. S., Samli, A. C., & Claiborne, C. B. (1991). Self-congruity versusfunctional congruity: predictors of consumer behavior. Journal of the Academy ofMarketing Science, 19(4), 363e375.

Sirgy, M. J., & Su, C. (2000). Destination image, self-congruity, and travel behavior:toward an integrative model. Journal of Travel Research, 38, 340e352.

Tasci, A. D. A., & Kozak, M. (2006). Destination brands vs destination images: do weknow what we mean? Journal of Vacation Marketing, 12(4), 299e317.

Zikmund, W. G. (2003). Business research methods (7th ed.). Mason, OH: SouthWestern Pub.

http://davidakenny.net/cm/mediate.htm%23ST

http://davidakenny.net/cm/mediate.htm%23ST

Brand Personality of Tourist Destinations - An Application of Self-congruity Theory

Documents

Transcript of Brand Personality of Tourist Destinations - An Application of Self-congruity Theory