A review of destination image studies from 2008 to 2012

This paper, continuing from Pike (2002), and Stepchenkova and Mills (2010), reviews 177 articles published between 2008 and 2012 about destination image. The major characteristics of these articles are highlighted to show the development of destination image studies. Two broad categories of destination image studies (i.e. perceived image, and projected image) are discussed separately. It is hoped that researchers of the following period can find some suggestions to locate their studies in this very popular body of research. © 2014 International University College. All rights reserved

Many attempts have been made since then to capture the development of tourism destination image studies (e.g. Chon, 1990;Echtner & Ritchie, 1991;Gallarza, Saura & García, 2002;Pike, 2002;Tasci, Gartner & Cavusgil, 2007;Stepchenkova & Mills, 2010). Chronologically, two reviews conducted by Pike (2002), and Stepchenkova and Mills (2010) have done the work for the period from 1973 to 2007. Following Pike, and Stepchenkova and Mills, the current review intends to summarise the articles in the period of 2008-2012. It is necessary to conduct this review in order to understand the recent development of studies on tourism destination image, and to show fellow researchers where they can find their originality. Thus, the two major questions addressed in this study are: (1) what are the characteristics of destination image studies in 2008-2012?, and (2) what can fellow researchers learn from previous studies? However, before moving to the discussion of the methods and findings of this study, it is necessary to start with a note on the review methodology. Whereas Pike (2002) titled his paper "review" or "synthesis," Stepchenkova and Mills (2010) named their work "metaanalysis." Thus, a question arises: why the two previous literature reviews had been treated as if they used different review methods?
According to Littell, Corcoran and Pillai (2008), there are many types of literature review, for instance, critical review, mapping review, rapid review, scoping review, state-of-the-art review, systematic review, systematic research and review, systematized review, and umbrella review. Particularly, systematic review includes quantitative review, qualitative review, and mixed review. Quantitatively, Gene V. Glass published the first article using the term "metaanalysis" in 1976 (see Hunter & Schmidt, 2004). Meta-analysis is a technique for summarizing data (Dickersin & Berlin, 1992), which only applies to empirical research that produces quantitative findings (Lipsey & Wilson, 2001). On the other spectrum, Phyllis Noerager Stern and Chandice C. Harris coined the term "qualitative meta-synthesis" in 1985 to refer to the amalgamation of qualitative studies (see Walsh & Downe, 2005). In addition, a mixed review is a review which combines the findings of both quantitative and qualitative research. Several authors (e.g. Gretzel & Kennedy-Eden, 2012) regarded the two former methods as quantitative meta-analysis and qualitative meta-analysis; however, they maintained that these types of review deal with the "findings" of previous studies.
In this paper, the generic term of literature review, or review, is adopted. It is a summary of the literature regarding tourism destination image published in academic journals in a period of time. Its purpose is to describe the characteristics of the collected literature, and not necessarily include the research findings (after Grant & Booth, 2009).

An overview of previous reviews
Perhaps the first review of studies on destination image in tourism was conducted by Chon (1990). In the review, Chon collected 23 papers and presented a conceptual model of the relationship between destination image and travelers' behaviors (e.g. satisfaction, decision making). Later, Gallarza, Saura and García (2002) examined 65 studies from 1971 to 1999. The topics covered in their review included the conceptualization, dimensions, formation, assessment and measurement, and characteristics of destination image. Echtner and Ritchie (1991) conducted another review which focused on the meaning, formation, and measurement of destination image. Some important conclusions are: two main components of destination image are attribute-based component and holistic component; each component of destination image consists of functional/tangible and psychological/abstract characteristics, and common and distinctive/unique traits; and a combination of structured and unstructured methods should be applied to measure all the aspects of destination image (Echtner & Ritchie, 1991). Continuing from Echtner and Ritchie (1991), Tasci, Gartner and Cavusgil (2007) investigated the papers published from 1991 to 2004. Besides the discussion on the conceptualization and operationalization of destination image, Tasci, Gartner and Cavusgil (2007, p.209) also looked at several aspects that they claimed to be "ignored" by Echtner and Ritchie (1991) such as frame of reference, associational investigation, validity and reliability, and implications of destination image studies, among others.
Another review line which almost covers the history of studies on destination image was started by Pike (2002). In Pike (2002), the author collected 142 papers , and found that the most popular region for study was North America, the most popular type of destination was countries, the majority of papers used structured methods, and the range of interests was diverse. Later, Stepchenkova and Mills (2010) analysed 152 papers published between January 2000 and October 2007. They confirmed the popularity of this body of research, which resulted into valuable implications for the development, marketing, branding, and overall management of tourism destinations.

Procedure
The search of the papers Several steps were followed when conducting this study. Firstly, the topic was decided after reviewing Pike (2002), and Stepchenkova and Mills (2010). Secondly, the researcher chose the period of 2008-2012, in order to continue from Stepchenkova and Mills. In addition, a period of five years is long enough to summarise the development of studies in this field. Thirdly, to be consistent with Pike, and Stepchenkova and Mills, the articles written in English and published in academic journals were targeted as the subjects of the study. This review was also conducted as a contextfocused study; that means, it emphasized on the studies put into the travel, tourism and hospitality contexts. Fourth, "destination image" was used as the main keyword. The next step was to collect the papers through an online search engine. There are a variety of search engines on the Internet, for instance, Web of Science, Scopus, PubMed, and Google Scholar; each of them has its own strong and weak points. Google Scholar, in particular, is said to be lack of competence (Jacso, 2005), and lack of subject indexing and/or classification access (Noruzi, 2005). In addition, the search finding is a mixture of both scholarly and non-scholarly materials (Noruzi, 2005;Meho & Yang, 2007;Kulkarni, Aziz, Shams & Busse, 2009), is less often updated and inadequate (Falagas, Pitsouni, Malietzis & Pappas, 2008), or has constantly-changing content, algorithms and database structure (Giustini & Boulos, 2013). However, this search engine also has many advantages. For instance, it leads searchers to the latest works and related papers, provides an international coverage of sources, and allows researchers to work on broad-based and multidisciplinary searches (Noruzi, 2005). It is more comprehensive for social sciences (Kousha & Thelwall, 2007), conference proceedings and international, non-English language journals (Meho & Yang, 2007), and economics and business topics (Harzing & van der Wal, 2008). It can help retrieve even the most oblige information (Falagas, Pitsouni, Malietzis & Pappas, 2008). And above all, it is free and friendly. Considering the benefits of Google Scholar, as well as the ways to delimit its drawbacks (i.e. to search and re-search in different points of time, and to carefully screen through all the papers listed in all pages of the search results), it was selected as the only search engine for this study.
The searches were conducted five times during a period of four months (February -May, 2013). In the fourth and fifth searches, two additional keywords were included (i.e. "cognitive destination image," and "affective destination image") in order to collect more database for the study. All suitable and downloadable journal articles were kept for further review (preprints with only "doi" addresses were eliminated). This resulted in 190 articles, which focused on destination/place image in the period of 2008-2012.
The categorization of theoretical, mixed, and empirical studies was implemented following Stepchenkova and Mills (2010, p.582). Accordingly, a theoretical/conceptual study is a study which used the existing knowledge to "develop an in-depth understanding of concepts." In addition, if an article tested a theoretical model or a theory using a sample of human/non-human population, it will be treated as either an empirical or a mixed study. Among the 190 articles, four are theoretical, six are of non-tourism context, and three are not directly related to destination image. Considering the limited number of the theoretical studies and the pre-determined travel/tourism/hospitality context, this study focused on the remaining 177 empirical and mixed theoretical/empirical works published from 2008 to 2012. In terms of the number of papers, Tourism Management (17 articles), Journal of Travel Research (14 articles), Journal of Travel & Tourism Marketing (13 articles), International Journal of Tourism Research (9 articles), Journal of Vacation Marketing (7 articles), Annals of Tourism Research (6 articles), and Asia Pacific Journal of Tourism Research (6 articles) rank top of the journals enlisted. There are six journals have four papers each, seven have three, nine have two, and the remaining 42 journals have one paper each ( Table 1).

The analysis of the papers
During and after the search of the papers, the main review was conducted. Following Pike (2002), Mills (2010), andHart (1998), an analysis protocol was developed which consists of the major components of an academic report: type of study (theoretical, empirical, or mixed), objectives, background theory, constructs/variables under study, items/attributes generating methods, objects and subjects under study, study methods, analysis methods, and key findings. This protocol was used to briefly code the content of each paper. Next, the first major coding was conducted using Microsoft Excel as the managerial tool; the elements of the analysis protocol were used as the coding variables. In the second major coding (also in Microsoft Excel environment), other coding variables which were not covered in the first major coding were added (e.g. definition and importance of destination image, sampling techniques), and several initial coding variables were divided into smaller units. While doing so, the contents of the former coding were checked to guarantee the accuracy of the whole process. The presentation of the findings is illustrated in Figure 1. The presentation will focus on the characteristics which show the development of destination image studies in the period of 2008-2012. In addition, research of the former period and of other fields of study will also be consulted to form a more holistic picture for this review. In this paper, studies on destination image will be divided into two broad categories: perceived image, and projected image. On the one hand, those studies which investigated how the audiences of a destination (e.g. tourists, residents) perceive the images of that destination will be grouped into the perceived image category. On the other hand, those studies which focused more on the images created by the managers/marketers and the likes of a destination to communicate or promote to its audiences will be put into the projected image categories. This categorization reflects the development of studies on projected destination image, which was overlooked in the previous reviews. There are, however, several exceptions, as some studies in the perceived category will be discussed for the projected category due to their relevance and contribution to this study area. The details will follow in the next sections of this paper.

A. Perceived image studies
There are 164 papers related to the perceived images of destinations. Some highlights of these studies are as follows.

Definitions and characteristics of perceived destination image
The study on destination image started with the perceived images of a destination, and nowadays this study line is still dominating the field. As seen in other reviews and papers (e.g.  Tasci, Gartner and Cavusgil (2007, p.199), each definition only defines "a particular aspect of destination image". Table 2, which selected several definitions of destination image collected by the authors of the previous reviews and of this paper, will demonstrate this issue.

Study objectives and methodology Previous reviews review
However, as time passes and researchers include other components into the concept of destination image, a more holistic picture is gradually formed. Take the definitions selected in Table 2 as examples, perceived destination image can be regarded as the knowledge, impressions, prejudices, imaginations, emotional thoughts, beliefs, ideas, conceptions, attitudes, benefits, values, expectations, and interpretations that an individual holds about a destination. Destination image is formed over time from a variety of information sources, and consists of both the cognitive/reasoned and affective/emotional aspects.
Destination image is a mental/psychological construct; however, the discussion on this basic characteristic of destination image was not received much attention, or was implicitly stated in previous reviews and studies. Several researchers observed that destination image has a psychological tradition (e.g. Frías, Rodríguez & Castañeda, 2008;Stepchenkova & Morrison, 2008;del Bosque, San Martín, Collado & de los Salmones, 2009;Heslop, Nadeau & O'Reilly, 2010). Its psychological nature is also explicitly stated in the definitions for destination image, for instance, image is a mental picture (Gil & Ritchie, 2009;, a mental notion (Molina, Gómez & Martín-Consuegra, 2010), a mental expression (Kesić & Pavlić, 2011), or a mental perception (Jamaludin, Johari, Aziz, Kayat & Yusof, 2012). According to Dermetzopoulos, Bonarou and Christou (2009, p.130), destination image is "associated with psychological dimensions manifested by feelings and attitudes towards the destination." The sum of beliefs, ideas, and impressions that a person has of a destination Embacher and Buttle (1989) c Ideas or conceptions held individually or collectively of the destination Dadgostar and Isotalo (1992) b The overall impression or attitude that an individual acquires of a specific destination. This overall impression is considered to be composed of the tourists' perceptions concerning the relevant qualities of the destination. Milman and Pizam (1995) b A sum total of the images of the individual elements or attributes that make up the tourism experience Mackay and Fesenmaier (1997) b A compilation of beliefs and impressions based on information processing from a variety of sources over time, resulting in an internally accepted mental construct Tapachai and Waryszak (2000) c Perceptions or impressions of a destination held by tourists with respect to the expected benefit or consumption values Kim and Richardson (2003) Table 3 and Figure 2 below for more details). The measurement of perceived destination image was conducted through psychological experiments in several studies (e.g. Pan, 2011;Jeong & Holland, 2012;Pezenka & Buchta, 2012;Yang, He & Gu, 2012). In addition, complex, multiple, relativistic, and dynamic are the other characteristics of destination image that are frequently cited (see Gallarza, Saura & García, 2002).

Destination typology
A destination is not restricted to a country, a region, or a city. Among the articles published between 2008 and 2012, other entities could be considered as destinations: night market (Lee, Chang & Hou, 2008) Kim, Ko, Connaughton & Lee, 2011;Kaplanidou & Gibson, 2012), were also targeted as the destinations or the contexts of many studies.
Destination image was also measured in combination with the images of other entities. For instance, country image and people image were investigated in the cases of Nepal In a smaller context, Ryan and Ninov (2011) focused on the relationship between a destination's images (i.e. Dubai), and its site's images (i.e. Dubai Creek). It was found in Ryan and Ninov's study that only a small portion of correlation existed between the two groups of image besides the majority of homogeneity. In another context, Altinbaşak and Yalçin (2010) identified the images of Istanbul and its museums. The results revealed that museums formed an important aspect of the city's image.

Destination image and its relationships with other constructs
Few studies only investigated the images of a destination as perceived in one of its markets (e.g. Ozretić-Došen, Škare & Krupka, 2008;Stăncioiu, Teodorescu, Pârgaru, Vlădoi & Băltescu, 2011;Matlovičová & Kolesárová, 2012). The focus of the rest is the identification  (2008)   of the relationships between destination image and other constructs (Table 3).
The review has revealed that destination image is affected by information sources, tourist characteristics, past experience, and familiarity/novelty seeking behaviour; is the antecedent of satisfaction, behavioural intentions, and loyalty; and has interrelationships with place attachment, attitudes, beliefs, destination personality, motivations, perceived quality, or perceived values. These interrelationships are visualised in Figure 2.
In the context of a tourism destination, destination images are considered as "pull" factors, and tourist motivation as "push" factors (e.g. Crompton, 1979 Wong, 2011). In addition, both destination images and motivations were found to be the antecedents of tourist satisfaction and behavioural intentions. In some studies, the attributes of a destination were utilized to measure tourist satisfaction (e.g. . Behavioural intentions, also under the effect of destination image, were usually measured by the intention to visit/revisit, and the intention to recommend a destination. However, there were several exceptions. For instance, Hudson, Wang and Gil (2011) applied the AIDA model of marketing (Attention -Interest -Desire -Action), whereas Yüksel (2011) used the approachavoidance model to measure the future behaviours of their respondents.
Newer/less popular constructs were also under investigation in relation to destination image. In the context of the influence of Korean pop culture, Kim, Agrusa, Chon and Cho (2008) measured the interaction between destination image held by Hong Kong residents, and the components of Korean pop culture (i.e. drama/movie, music, food, and Hallyu products). In another study, Lee, Scott and Kim (2008) considered the effect that the involvement with Korean celebrities of Japanese tourists had on their perceptions of the country; yet, the proposed relationship was not empirically supported. However, incorporating the personal involvement in the context of international tourists with Mauritius, Prayag and Ryan (2012) found significant evidence that personal involvement was an antecedent of destination image.
The list of other constructs found in destination   (Tasci, 2009), perceived authenticity and information search behaviour , telepresence or "the feeling of being there" (Hyun & O'Keefe, 2012), time exposure to information sources (Jeong & Holland, 2012), or perceptions of tourism logos (Lee, Rodriguez & Sa, 2012).
Perceptions of destination image attributes were used as a segmentation base in several studies (e.g. Kamenidou, Mamalis & Priporas, 2009;Lin & Huang, 2009;Tkaczynski, Rundle-Thiele & Beaumont, 2010;Kim & Perdue, 2011;. However, the majority of the findings were discussed for marketing and managerial implications. 3.2. Destination image in the brand/branding context There is a significant group of studies put into the context of brand/branding. For instance, Konecnik and Ruzzier (2008) applied the concept of brand equity in measuring German and Croatian tourists' perceptions of Austria, in which destination image was treated as an element of Austrian brand equity, besides awareness, perceived quality, and loyalty. The similar pattern was observed in the study of Chen and Myagmarsuren (2010); however, the context chosen was Mongolia, and the targeted subjects were the international tourists to the country. Destination image as a member of brand associations (i.e. cognitive, affective, and unique image components) was found in the studies of Hanzaee and Saeedi (2011), and Qu, Kim and Im (2011).
Several authors measured the construct of destination personality (e.g. Yuksel & Bilim, 2009;Gertner, 2010;Sahin & Baloglu, 2011;Caruntu, Stancioiu & Teodorescu, 2012;, which originated from brand personality studies (e.g. Aaker, 2002). Yuksel andBilim (2009), and also adapted the self-image/selfcongruency phenomenon (ideal-self vs. actualself) in their studies. In another attempt, Ivanov, Illum and Liang (2010) applied the concept of brand molecule (i.e. tourists' predominant perceptions of a destination and its organization in their minds), and brand concept maps (i.e. tourists' perceptions towards and their associations with a destination) in the context of a gaming destination (i.e. Las Vegas). The effects of country-of-origin, and the beliefs of/familiarity with the country's products were targeted in Elliot, Papadopoulos and , and Lee andLockshin (2011, 2012).

Destination image as the representative of other constructs
Destination image was treated as brand image in many studies (Prayag, 2010;Saraniemi, 2010;Hanzaee & Saeedi, 2011;Qu, Kim & Im, 2011;Bruwer & Lesschaeve, 2012;Naidoo & Ramseook-Munhurrun, 2012). In other studies, destination image could represent destination attractiveness (Cracolici & Nijkamp, 2008;Krešić & Prebežac, 2011;Kim & Perdue, 2011;Bui, 2012), destination competitiveness (Omerzel, 2011), destination reflect (Maroofi & Dehghan, 2012), consumption value (Shanka & Phau, 2008), or choice criteria (Kamenidou, Mamalis & Priporas, 2009). Tourists' perception of the safety and security of a destination was also considered as a specific aspect of perceived destination image in several studies (Donaldson & Ferreira, 2009;Fuchs & Reichel, 2011;Boakye, 2012). Liou (2010), in a study of Taiwanese tourists' perceptions of Japanese TV series-induced tourism, didn't measure destination image itself but focused on the image change as affected by J-drama series. , from another perspective, investigated the images of China in comparison with its major competitors in Asia which the author called "relative destination image." 4. Structure of destination image 4.1. A review of the most influential models of destination image formation The review has revealed that three models mostly adopted in investigating image structure are the cognitive/affective/conative model (Gartner, 1993); the attribute-based/holistic, functional/psychological, common/unique model (Etchner & Ritchie, 1991, 1993, 2003; and the cognitive/affective/overall model (Baloglu & McCleary, 1999). Gartner (1993) considered the cognitive component of destination image as "the sum of beliefs and attitudes of [a destination] leading to some internally accepted picture of its attributes" (p.193), and the affective component as "the motives one has for destination selection" (p.196). The co-native image, or the action component, is "analogous to behaviour;" it depends on the images developed during the cognitive stage and evaluated during the affective stage of destination image formation process (Gartner, 1993, p.196). Fellow researchers who applied this model include, for instance, Agapito, Mendes and do Valle (2010), Moon, Kim, Ko, Connaughton andLee (2011), andHallmann (2012). Etchner and Ritchie (1991, 1993, 2003, however, proposed that destination image has two main components: attribute-based and holistic; each component consists of functional (or more tangible) and psychological (or more abstract), as well as common and unique, characteristics. To measure the attribute-based and more common components, Etchner and Ritchie developed a scale with 35 items, ranging from the more functional attributes (i.e. tourist sites/activities, national parks/wilderness activities, historic sites/museums, beaches, fairs/exhibits/festivals, scenery/natural attractions, nightlife/entertainment, shopping facilities, facilities for information and tours, sports facilities/activities, local infrastructure /transportation, cities, accommodation/ restaurants, architecture/buildings, costs/price levels, and climate), the mixed attributes (i.e. crowdedness, cleanliness, degree of urbanization, economic development/affluence, extent of commercialization, political stability, accessibility, personal safety, ease of communication, customs/culture, and different cuisine/food/ drink), to the more psychological attributes (i.e. hospitality/friendliness/ receptiveness, restful/relaxing, atmospherefamiliar vs. exotic, opportunity for adventure, opportunity to increase knowledge, family or adult oriented, quality of service, and fame/ reputation). Moreover, to capture the holistic and unique components, they created three open-ended questions: (1) (1999). Accordingly, destination image is formed by three components: cognitive, affective, and overall; in which the affective component is affected by the cognitive one, and the overall component is influenced by the two former elements. Since then, many researchers have applied and retested this model in their studies (e.g. Gil & Ritchie, 2009;Alaeddinoglu & Can, 2010;Alvarez & Campo, 2011;Hung, Lin, Yang & Lu, 2012). Besides the cognitive image attributes that they developed, Baloglu and McCleary (1999) adopted an affective image scale invented by Russell and Pratt (1980). The affective scale consists of four bipolar items (arousingsleepy, excitinggloomy, pleasantunpleasant, and distressingrelaxing), or two items (arousingsleepy, and pleasantunpleasant) in case a short questionnaire is necessary. After Baloglu and McCleary, this affective scale has also been widely applied in destination image studies, although longer scales can be found elsewhere (e.g. Jeong, Kim, Ko, Lee & Jeong, 2009;Yüksel, 2011).

Development in destination image
structure studies Several attempts have been made to put forward the studies on destination image structure. For instance, Prebežac and Mikulić (2008), applied Etchner and Ritchie, qualitatively measured the activity-based component of Croatia and Hawaii besides the stereotypical, affective, and unique components. This resulted in "sightseeing," "historical sightseeing," "hiking," or "cultural experience" for Croatia, and "surfing," "swimming," "diving/scuba-diving," or "relaxing/ sunbathing" for Hawaii. In another study, Qu, Kim and Im (2011), followed both Etchner and Ritchie, and Baloglu and McCleary, combined the unique element into a structured model of destination image formation together with the cognitive, affective, and overall components. Testing with the domestic tourists in Oklahoma (America), factor analysis of the unique attributes resulted into three factors: "native American/natural environment," "appealing destination," and "local attractions." From a sensory approach, Huang and Gross (2010), and Dițoiu, Stăncioiu and Teodorescu (2012) investigated the visual, auditory, olfactory, gustatory, and tactile elements of destination image when at the same time measured the cognitive and affective components (as seen in Huang & Gross, 2010). In Huang and Gross (2010), for instance, the sensory images of Australia as perceived by the mainland Chinese travellers included the picturesque landscape elements (visual), sounds of nature (auditory), smell of air/sea/food (olfactory), wine/juice/food (gustatory), or touching objects/feeling warmth under the sun (tactile). However, the activity, unique, or sensory elements of destination image are simply the other forms of the cognitive image if that component is seen from a multi-dimensional perspective as suggested by Etchner and Ritchie.
In another attempt, Pezenka and Buchta (2012) tested the verbal and pictorial scales of the affective images of twelve European cities with a sample of Austrian tourists. The results revealed that the pictorial scale was easy to use but its interpretations were broad, whereas the verbal scale was more concise but respondents might find it harder to assign them to the cities.
The structure of the cognitive image component was further examined in several studies. For instance, Chen, Chen and Lee (2010) investigated Chinese residents regarding their perceptions of Kinmen Island  Alcañiz, García and Blas (2009) surveyed tourists at Peniscola, Valencia (Spain), and in their study, the functional, mixed, and psychological elements made up the cognitive image of the destination. Later, Lai and Li (2012) proposed and tested a core-periphery model of the cognitive image of Beijing, while Yang, He and Gu (2012) explicitly and implicitly investigated the cognitive images of Japan and Hong Kong. From a qualitative approach, Peña, Jamilena and Molina (2012) categorized the images of a rural destination (i.e. Andalusia, Spain) into five groups: "characteristics of the rural tourist destination," "characteristics of the service offer," "cultural offer," "offer of activities in nature," and "local products and gastronomy offer." In summary, destination image is considered as a multi-dimensional structure which consists of two basic components: cognitive, and affective. Particularly, the cognitive component may be divided into three elements: functional, psychological, and mixed. Thus, following Etchner and Ritchie (1991,1993,2003), Baloglu and McCleary (1999), and other researchers (Alcañiz, García & Blas, 2009), an integrated model of destination image can be formed as seen in Figure 3. The popular qualitative methods include interview (e.g. Huang & Gross, 2010;Ing, Liew-Tsonis, Cheuk & Razli, 2010;Campo, Brea & Muñiz, 2011;Prayag & Ryan, 2011), focus group (e.g. , and free elicitation (e.g. Stepchenkova & Morrison, 2008;Iordanova-Krasteva, Wickens & Bakir, 2010;Prayag, 2010;Pan & Li, 2011;Ryan & Ninov, 2011;. In particular, Garrod (2008Garrod ( , 2009) introduced a "volunteer-employed photography" method in which the respondents were asked to take pictures of a place (i.e. Aberystwyth, Wales), and these photos were content analysed to figure out how the respondents perceived the place. In addition, many researchers datamined tourists' posts from travel blogs/ forums to generate the images of a destination: Austria , Bethlehem (Çakmak & Isaac, 2012), China (Li & Wang, 2011), Hong Kong (Leung, Law & Lee, 2011), India (Dwivedi, 2009), or Zanzibar Island of Tanzania (Jani & Hwang, 2011).
Researchers also compared the images of different phases of the formation process: pre-and post-visit (e.g. Wang & Davidson, 2008;Jani & Hwang, 2011;Lee, Kang, Reisinger & Kim, 2012). Residents (including temporary ones) were only targeted in ten studies.
Although probability sampling methods are better for the generalizability of research findings, many researchers in the 2008-2012 period utilized the non-probability techniques (e.g. convenience, purposive, quota, or snowball). Subjectively, the lack of study resources was cited as the main reason for not selecting random sampling methods. In addition, since there is not a complete sampling frame available for the population in many studies, the application of the random methods is not possible (Yuksel & Bilim, 2009). The theoretical models, in these cases, can be replicated in other contexts or with other sample frames in order to extend the current research (Shanka & Phau, 2008).

Analysis methods
In terms of data analysis methods, confirmation analysis and structural equation modelling (SEM) are getting more popularity among the quantitative analysis techniques besides the traditional ones (e.g. descriptive analysis, factor analysis, regression analysis, correlation analysis, and analysis of variance). The review of the papers utilized SEM revealed that many authors conducted the data analyses through three steps: (1) exploratory/principle component factor analysis, (2) confirmatory factor analysis, and (3) SEM.
Although widely applied, one of the problems when using SEM is the sample size (Hox & Bercher, 1998). Many authors recommended a minimum sample of 200 (e.g. Hox & Bercher, 1998;Weston & Gore, 2006). Anderson and Gerbing (1988) suggested a sample size of 150 or more, while Iacobucci (2010) set the starting point at 50. Generally, Tanaka (1987) stated that the sample size should be determined based on the numbers of parameters measured in the model. A threshold of 50 observations may be sufficient in case when there are one latent variable underlying four measured indicators, but is insufficient if there are four latent variables with twenty measured variables (Tanaka, 1987). Practically, several papers published between 2008 and 2012 showed that the testing of a theoretical model on a small sample size (n = 103-180; 5-7 constructs per model) could generate acceptable/good fit-indices (e.g. Mohamed, 2008;Gil & Ritchie, 2009;Chen & Myagmarsuren, 2010;Hooper, Coughlan & Mullen, 2008).
Stepchenkova and Li provided unique examples of using perceptual mappings , and of quantifying the data to calculate the Diversity Indicators Index  when examining Chinese tourists' perceptions of the U.S. On the other spectrum, content analysis and theme analysis were among the popular qualitative analysis methods.

B. Projected image studies 1. Conceptualization of projected image
There are nine studies directly investigated the projected images of places/destinations, and one measured both the projected and the perceived images; thus, compared to the mainstream studies on perceived destination image, this study area is overlooked. Moreover, many researchers even did not explicitly state that their studies focused on the projected image, or did not define the concept. Only a few studies mentioned the fact that there are two perspectives in studying the images of a destination: on the one hand, the images perceived by the audiences; on the other hand, the images created by the destination (e.g. Kwek & Lee, 2008;Hsu & Song, 2012;Hunter, 2012).
Projected image is "the combined product of government authority and private commercial efforts" (Hunter, 2012, p.422). Although created to communicate to the target audiences to fulfill marketing purposes, there may be incongruences between the images created and the images actually perceived (e.g. Saraniemi, 2010;. Bramwell and Rawding (1996, p.202) regarded projected images as "the ideas and impressions of a place that are available for people's consideration," and are transformed through communication channels. Specifically, Gartner (1993) defined eight information categories which differ from each other by their credibility, market penetration ability, and cost. There are overt induced information agents (I and II) which consist of the more traditional forms of advertising (e.g. brochures, television, radio, print materials, billboards) and information provided by tour operators and wholesalers; and covert induced information agents (I and II), which are the communication efforts made by second parties. News and popular culture, documentaries, movies, or television programs are the members of the autonomous information category. In addition, the information received from family and friends are the organic agents (solicited and unsolicited), besides the actual experiences. In sum, destination images are formed by induced information and organic/autonomous information. Under the influence of the former, the images are "intentionally" created; on the other hand, the images are "unintentionally" projected (Ji, 2011). The efforts to create destination images through the induced image formation agents must be focused and long term (Gartner, 1993). However, the power of the organic/autonomous image formation agents should not be overlooked. Positive words-of-mouth may be created by guaranteeing current customers' satisfaction; in addition, good relations with the public in general and with the press in particular may result in the autonomous transmission of positive images. Moreover, the size of a destination may affect the development and change of its images, and an assessment of the perceived images is necessary for an effective image development and change campaign (Gartner, 1993).
Kwek and Lee (2008) utilized an intra-cultural approach to compare the images projected by Queensland tourism towards the Mainland Chinese market, and the Singaporean Chinese market. From a cross-cultural perspective, Huang and Cai (2010) measured the verbal images of hotel chains in China which were displayed in English and Chinese websites. Also measuring Beijing's hotel images in the websites, however, Law and Chen (2012) only focused on the cultural aspects of destination image. From other approaches, Milman (2011,2012) investigated the projected images of Alanya (Turkey), and Berlin using the tourist postcards, whereas the images of six Asian countries/territories projected for the Chinese outbound market were measured using the articles appeared in the travel magazines as study subjects by Hsu and Song (2012).
Other printed materials such as novels and newspapers articles were used to track the images of Madrid (Busby, Korstanje & Mansfield, 2011), and Russia (Stepchenkova & Eales, 2011). Although these two papers can also be considered under the perceived image category, they are introduced here as the examples of creative methods for investigating projected images. Specifically, Busby, Korstanje and Mansfield (2011) selected two novels written in French and Spanish to excavate the images of an urban Madrid. In another approach, Stepchenkova and Eales (2011) developed a Dynamic Destination Image Index to trace the image changes of a country (i.e. Russia) using newspapers' articles (i.e. UK's).
Methodologically, literature review, content analysis, theme analysis, and semiotic analysis were the techniques utilized in projected image studies. Almost all of them are qualitative studies, except Stepchenkova and Eales' (2011) in which the authors quantified the data to form an index of destination image changes. In addition, the attempts to investigate the projected images of destinations are indirect, which only focused on the images kept by the means of information communication. Thus, only one aspect of the projected images may be captured (i.e. either verbal or pictorial); moreover, they are all static images. Destination marketers and tourism practitioners, targeted in several perceived image studies (e.g., Ing, Liew-Tsonis, Cheuk & Razli, 2010;Saraniemi, 2010), may also be approached to understand the other aspects of projected destination image. In addition, only the cognitive aspects of the projected image were focused in these studies; thus, the other componentthe affective imageis understudied in case of projected destination image.

Concluding remarks
More than forty years have gone since the first publication on tourism destination image, and this body of research is still one of the most popular topics in tourism study. Continuing from previous efforts, this study attempted to review the papers regarding the topic in a five year period (2008)(2009)(2010)(2011)(2012). Utilizing Google Scholar search engine, a collection of 177 academic articles was generated for the analysis. The discussion was divided into two broad categories: projected image, and perceived image. Before this paper, the focus of the other reviews and papers was on the perceived images of destinations. Thus, a discussion on the conceptualization and measurement of the projected destination image construct is necessary in order to better cover the development of this field of study. In addition, following the previous reviews, this paper made several efforts to understand how researchers in 2008-2012 defined and measured the perceived image. Findings include the entering of image studies into new destinations, the development in structuring the components of destination image, the more widely application of SEM in investigating the relationships among destination image and other psychological constructs, among others.
The discussions on the managerial implications of perceived destination image studies now have a more specific target: the projectors of destination image, given the marketing and/or branding activities of destinations. In addition, considering the psychological nature of perceived destination image, future efforts in building and/or changing the images of a destination should not only focus on the conscious perceptions of its target audiences, but also on the subconscious or unconscious notions. The unintentional projection of destination images and the use of organic/autonomic sources of information should also be put under thorough consideration.
However, two limitations can be observed in this study. Firstly, only one search engine was utilized; thus, the missing of related papers might occur. Moreover, only one main keyword (i.e. destination image), and two additional ones (i.e. cognitive destination image, and affective destination image) were used in the searches; hence, those papers do not contain these three keywords in their titles, list of keywords, and abstracts might be ignored. Secondly, the presentation of the findings was selected and structured subjectively by the researcher; consequently, some contents were more focused while the others were less emphasized.
Future literature may see the penetration of image studies into new or less "cultivated" areas (e.g. South America, Africa/Middle East, North Korea, Bhutan, or the space), new types of "destination" (i.e. other members of the hospitality industry such as airlines); the combination of destination image with other constructs; the development of new research and analysis methods; or the application of other disciplines. Proportionate reviews, focusing on particular aspects of destination image studies, may be the new approach in the future. For instance, Dolnicar and Grün (2013) only focused on the scales utilized in previous studies' measurements, and found that a binary scale (i.e. yes/no) performed better than the Likert-type scales. Peña, Jamilena and Molina (2012), cited in this review, introduced another path of study when developing and validating a cognition-based scale to measure the images of a particular type of destination (i.e. rural destination). Future studies may emphasize on other aspects, for instance, the sampling methods, or the development of a common scale to measure the images of a particular type of destination (e.g. coastal region, mountainous area).
Researchers should also put more efforts in investigating the projected images of destinations. Future studies should examine the verbal and pictorial, as well as the cognitive and affective, aspects of projected image. In addition, besides the finding of the static images in the means of information communication, a more dynamic approach which subjects the projectors of destination image should also be considered. Moreover, a periodical review of studies on tourism destination image is necessary to see the development of the field, and to locate the originality of each researcher. Thus, the scholarship community may know if this body of research has reached its peak, or is still moving ahead.  32, 193-194. Chon, K.-S. (1990) (3) (2012)