The purpose of this paper is to use user-generated content (UGC) on social media platforms to infer the possible place brand identities of two famous metropolitan areas in Bangkok, Thailand, namely, Khaosan Road and Yaowarat (Bangkok’s Chinatown), both of which are famous for their street vendors and nightlife. These two places are interesting study sites because of recent identity conflicts among their stakeholders. The method developed in this research can help other places to better understand place brand identities and, as such, effectively plan for and manage those places.
The author used content analysis to study 782 user-generated images on Flickr and 9,633 user-generated textual reviews of Khaosan Road and Yaowarat from TripAdvisor and Google Maps’ Local Guide. MAXQDA was used to code all the images. User-generated textual reviews were studied using Leximancer. The author also introduced a positivity of concept analysis to identify positive and negative components of place brand identity.
The author developed a place brand identity framework that includes three pillars, namely, place physics, place practices and place personality. Content analysis of the images generated 105 codes and a count of the frequency of the codes that represent place brand identity. Content analysis of textual reviews created the concepts in the three pillars and identified the positive and negative concepts for both places. The results of both image and text analyses showed that street food vending is one of the most salient components of place brand identity for both Khaosan Road and Yaowarat.
The author suggested several place branding strategies for the Bangkok Metropolitan Administration such as turning Khaosan Road into a music scene for both visitors and locals, controlling excessive and aggressive commercialism, sponsoring the production of creative and authentic content, initiating a compelling online campaign that focusses on the items sold in Yaowarat, hosting a spotlight event such as a seafood festival and improving hygiene and walkability.
Both the advancement of digital technologies and the complexity of stakeholders create a need for empirical studies on place branding involving the participation of the widest possible range of stakeholders and studies on the influence of social media. This research is the first to use both image and text analyses to study place brand identity from UGC. The use of both analyses allows the two methods to complement one another while mitigating the weaknesses of each.
Taecharungroj, V. (2019), "User-generated place brand identity: harnessing the power of content on social media platforms", Journal of Place Management and Development, Vol. 12 No. 1, pp. 39-70. https://doi.org/10.1108/JPMD-11-2017-0117
Emerald Publishing Limited
Copyright © 2018, Emerald Publishing Limited
A place brand is a network of associations in the minds of consumers based on the visual, verbal and behavioural expressions of a place (Zenker and Martin, 2011). As such, place brands are socially and culturally embedded and are co-created by social actors (Pryor and Grossbart, 2007). Consequently, the variety and dynamism of stakeholders and social actors result in complex and ever-changing place brands (Moilanen, 2015). Advances in digital and online technologies also make place brand identities more complex by opening new venues for place branding (Kavaratzis, 2012), and accordingly, places all over the world have begun to incorporate social media into their place branding efforts (Andéhn et al., 2014; Zhou and Wang, 2014). Compared with their traditional counterparts, communications via social media are more integrative and inclusive (Ketter and Avraham, 2012). Places can use social media to communicate with their audiences, and social media platforms enable tertiary communication activities among place customers via user-generated content (UGC) (Andéhn et al., 2014). As a result, social media platforms play a unique role in formulating place identity, as users bring their subjective opinions and personal perspectives to the online space, affecting how others view the place (Fazel and Rajendran, 2015). Both the advancement of digital technologies and the complexity of stakeholders create a need for empirical studies on place branding that involve the participation of the widest possible range of stakeholders (Kavaratzis, 2012; Kavaratzis and Hatch, 2013) and studies regarding the influence of social media (Govers et al., 2017).
The objective of this research is to use UGC on social media platforms to infer the possible place brand identities of two famous metropolitan areas in Bangkok, Thailand, namely, Khaosan Road and Yaowarat (Bangkok’s Chinatown). This process includes using image and text analyses to identify the components that constitute the following three pillars of place brand identity: physics, practices and personality. Both Khaosan Road and Yaowarat are famous for their street vendors and nightlife. Khaosan Road is in Phra Nakhon District, whereas Yaowarat is in Samphamthawong District of Bangkok (Figures 1 and 2). These two places are interesting study sites because of recent identity conflicts among their stakeholders. In April 2017, the Bangkok Metropolitan Administration (BMA) announced that it would remove street vendors, including those on Khaosan Road and Yaowarat, from Bangkok in an attempt to return the pavement to pedestrians (Rujivanarom, 2017). Since the 1972 founding of the BMA, local and national governments in Thailand have adopted inconsistent policies on street vending. At times, the government tolerated street vending because it was regarded as an income-generating activity that alleviated poverty; at other times, street vending was perceived as a threat to the city’s safety and sense of order (Boonjubun, 2017). There are several reasons to forbid street food vendors, including encroachment on public pavement, counterfeit and poor-quality products and a lack of hygiene (Maneepong and Walsh, 2013). Nevertheless, the April 2017 announcement caused an uproar among both local residents and the domestic and international press, who revere Bangkok’s distinctive street food scene (Coffey, 2017). Later, officials backtracked and stated that street vendors would be strictly regulated (Coffey, 2017; Fredrickson, 2017). The public outcry following the BMA’s announcements is an example of the dissonance between the conceived place identity of the public and the desired place identity (Trueman et al., 2004) of officials. Accordingly, a successful management of metropolitan place brands requires a continuous monitoring of a place’s subtextual atmosphere (Mahnken, 2011). The method developed in this research can help other places to better understand place brand identities and, as such, effectively plan for and manage those places.
Place brand identity
In the commercial world, brand identity, which refers to a holistic concept that articulates a corporation’s ethos, aims and values to differentiate it, is a fundamental concept of corporate branding (Kavaratzis, 2004; Van Riel and Balmer, 1997). Aaker (2012) defined brand identity as “a unique set of brand associations that the brand strategist aspires to create or maintain”. However, places are different from products, services and organisations because nobody owns a country, a region or a city (Boisen et al., 2011), and therefore, place brands are not under the control of marketers (Iversen and Hem, 2008). Thus, the conceptualisation of place brand identity is different from that of commercial entities. A concept that is closely related to place brand identity is “place identity” (Skinner and Kubacki, 2007) which refers to the objective reality of the place, as opposed to place image (Kalandides, 2011). Often corresponding to the territorial–administrative levels, the term “place identity” has numerous variations, such as spatial, territorial, local and regional identity (Boisen et al., 2011; Kalandides, 2011). Place identity covers a wide range of characteristics, including physical, cultural and historical characteristics (Hofstede, 2014). Apart from portraying an objective reality of a place, another common conceptualisation of place identity is the degree to which people identify with the place or feel a sense of belonging (Hofstede, 2014; Kalandides, 2011; Mueller and Schade, 2012). Contrarily, “place brand identity” is defined as the component associated with the creation of the essence of the place brand (Hanna and Rowley, 2011). Unlike place identity, which has a wider scope, place brand identity is used primarily in the place branding process; therefore, it refers to a place brand component that is identified, extracted and orchestrated from a larger concept of place identity to mark the place with distinctive and positive characteristics in the minds of various target groups (Boisen et al., 2011; Hanna and Rowley, 2011; Skinner, 2008).
Another concept closely related to place brand identity is place brand image; therefore, many scholars have studied the similarities, differences and interactions between these two concepts. Lucarelli (2012) used the two terms interchangeably in a systematic literature review, and Hanna and Rowley (2011) posited that place brand identity is part of an image-building process. Kavaratzis and Hatch (2013) stated that place brand identity and place brand image continually and iteratively interact through mirroring and influencing processes. Similarly, in the context of destination branding, Cai (2002) states that brand identity and brand image reinforce each other. Another common explanation for the difference between the two concepts is that the place brand identity comes from the sender’s perspective, that is, what the organisation or place wants to present to its customers, whereas the place brand image is the sum of the receivers’ beliefs, ideas and impressions (Florek et al., 2006; Peighambari et al., 2016). This research distinguishes the two terms based on a statement by Barke and Harrop (1994) cited by Kalandides (2011):
[T]he ‘identity’ may be regarded as an objective thing: it is what the place is actually like. Identity though is not the same as ‘image’ which defines how an organisation or a place is perceived externally.
According to Pryor and Grossbart (2007), place brand identity is affected by shared and contested economic, social, political and cultural views of the place’s stakeholders. This conceptualisation is reinforced by scholars who view place brand identity as a unique combination of either socio-cultural meanings (Botschen et al., 2017; Levy and Lee, 2011) or socio-spatial interactions among local institutions, actors and residents (Mahnken, 2011). The brand identity of a place is co-created by multiple stakeholders including – but not limited to – businesses, residents, visitors, public administrators, homeless populations, special interest groups and other external actors (Levy and Lee, 2011; Pryor and Grossbart, 2007). Kavaratzis and Hatch (2013) further argued that scholars should not adopt the static view of identity, “which largely describes [identity] as something to be tapped, defined, and manipulated” and not assume that identities are internal and disconnected from outsiders. Place brand identity is erratic and fluid, and as such, it should be understood as interwoven and interactive processes rather than as a definitive single core identity (Kavaratzis and Hatch, 2013). Similarly, Kalandides (2011) conceptualised place identity as a sense of place that is created by various elements over time.
Place brand identity serves several purposes. First, it attracts customers such as residents, visitors, talents and investors (Anholt, 2007). A strong place brand identity transmits a consistent image to internal and external stakeholders, creating a valuable asset for the place brand (Ashworth and Kavaratzis, 2009; Kavaratzis, 2004; Stock, 2009). Upon its projection, place brand identity generates the place’s value propositions, establishes relationships with its internal and external stakeholders and eventually guides their (the stakeholders’) behaviours (Bregoli, 2013; Konecnik and Go, 2008; Sarabia-Sanchez and Cerda-Bertomeu, 2016). Second, place brand identity creates order out of chaos (Anholt, 2007). By using its distinctive characteristics, brand identity allows the place to communicate its position in relation to other competing place brands (Hankinson, 2004; Hanna and Rowley, 2011), and place brand identity helps specify what a place represents and what it aspires to be (Konecnik and Go, 2008). Nevertheless, place brand identity is a contentious and challenging subject for many places, some that do not have a clear identity upon which to build (Gould and Skinner, 2007) and others that may have multiple but conflicting identities (Bouchon, 2014). The sheer number of stakeholders in a place may make identifying the identity of a place brand even more complex (Gould and Skinner, 2007). Without proper processes, some stakeholders with superior power may direct the development of the place brand identity to favour their specific needs (Konecnik and Go, 2008; Shoaib and Keivani, 2015). Thus, it is essential for place brand managers to have a good framework from which to understand and identify the place brand identity. The next section explores such frameworks from the previous literature.
Three pillars of place brand identity
The existing literature discussed, formulated and conceptualised place brand identity frameworks that are used as a basis of this research. Cai (2002) developed a model of destination branding that centres on brand identity. The model features three brand associations, attributes and affective and attitude components that bind brand identity to brand image. Hankinson (2004) described the concept of core brand in the relational network brand model as a representation of brand identity. The core brand has three elements: personality, positioning and reality. Florek et al., (2006) studied cities’ websites by adopting the three-pronged decision tools for brand identity communications. The three tools are place visual presentation, place behaviour and place communication. Kalandides (2011) wrote that the elements of place are materiality, practices, institutions and representation. In a similar manner, Lindstedt (2011) stated that identity includes physical settings, human activities and meanings. Ruzzier and De Chernatony (2013) studied Slovenia and explained place brand identity using a model comprising six elements: mission, vision, values, personality, distinguishing preferences and benefits. Recently, Martínez (2016) developed a comprehensive network place branding model in which place identities are formed by three components: geography, institutions and sense of place. Based on the various models, this research groups elements and components from existing frameworks into three main pillars of place brand identity: physics, practices and personality.
In the existing models, many scholars have pointed to tangible and physical features as one of the important elements of identity. For example, Cai (2002) wrote about attribute components, which refer to perceptual tangible features characterising a place. Likewise, Hankinson (2004) called those tangible features the functional, utilitarian and environmental attributes of a place. The current research designates this first pillar of place brand identity as place physics. Place physics is akin to materiality and physical settings described by Kalandides (2011) and Lindstedt (2011), respectively, in their frameworks. Hankinson (2004) gave examples of tangible attributes such as museums, art galleries, theatres, sport facilities, exhibition facilities, public spaces, hotels, restaurants, night clubs and transport infrastructure. These attributes and symbols, which include landmarks, street indicators, urban design, historic plaques, decorations and sculptures, are both the visual presentations and the essence of place brand identity (Florek et al., 2006; Hanna and Rowley, 2011). Place physics also includes the temporary features noted by Kalandides et al., (2016), who described how pop-up shops and flexible uses of retail space stimulate urban renewal. In addition to artefacts, a place’s natural features and geography are important place physics that shape place brand identity (Martínez, 2016; Ruzzier and De Chernatony, 2013).
Another common component of a place brand identity is the actions, activities, behaviours, decisions and rituals of people within the place or place practices. Kalandides (2011) wrote that practices include the repetition of people’s actions and the breaking of their routines. The actions and decisions that formulate place products have been described by Florek et al., (2006) as place behaviour. These place practices are comparable to human activities (Lindstedt, 2011) and social interactions (Martínez, 2016), which are central to identity development. Place practices are the manifestation of how people dress, drive, and live, i.e. their lifestyles (Levy and Lee, 2011). They are the events and rituals–such as street parties, parades, ceremonies, and artistic performances–that shape the experiences for which place is known (Hankinson, 2004; Hanna and Rowley, 2011). Place practices also include social relations, including competition and cooperation among stakeholders (Martínez, 2016).
The last pillar of place personality is the affective, emotional and symbolic pillar of place brand identity. This pillar encompasses personal values, which is the affective component of the branding model by Cai (2002) and meaning (Lindstedt, 2011), which is the enduring psychological bonds between stakeholders and places. Hankinson (2004) referred to these qualities as the experiential and symbolic attributes of a place brand. Place personality is defined as the set of human characteristics associated with place brand (Aaker, 1997; Hanna and Rowley, 2016). The attribution of human characteristics to place brand distinguishes those characteristics from the competition, thus creating a unique position or identity (Glińska and Gorbaniuk, 2016). Hankinson (2004) wrote that place personality includes the character of the local residents, the profile of the typical visitors, the quality of service provided by the service personnel, the descriptor of place feelings and the character of the environment.
Figure 3 depicts the three pillars of place brand identity and their interactions. The interaction between place physics and place practices constitute institutions. Institutions refer to normative systems that “consists of forms of property, power and control relations, legal regulations, planning guidelines, social and aesthetic norms” (Kalandides, 2011). Kalandides (2011) further explained that institutions are the results of continuous social relations and practices, and they, in turn, reproduce the materiality or physical features of a place. Additionally, Martínez (2016) stated that the geographical area “provides a framework and milieu for social interactions, and then its institutional structure, which shapes formal and informal institutional social relations”. Markets are an example of important institutions shaped by the interactions between place physics and place practices whereby stakeholders trade, communicate, network and form relationships (Warnaby, 2009).
The second interaction, experience, occurs between a place’s practices and personality. People may consume a place’s products or engage with a place for reasons beyond functional or economic reasons, such as for experiential reasons (Warnaby, 2009). Experiences can be defined as something learned or felt by personal contacts (Beeho and Prentice, 1997). The human characteristics of place personality such as activity, emotion and responsibility are woven into the behaviours and rituals of people in a place, thereby giving the place unique experiences. For example, visiting a theme park might encompass a series of emotions including excitement, thrill and fear, which constitute a flow of experiences (Beeho and Prentice, 1997). Such experiences from practices ascribe the place with a unique personality, which, in turn, motivates people to engage more with it.
The interactions between place physics and personality form the ongoing representations of a place. Representations are narratives of a place in the media and other communication channels (Kalandides, 2011). Place personality infuses affective characteristics to the physical environment such as green, historic, modern and spacious environments (Hankinson, 2004). These representations of physical and tangible features are continually imprinted on the minds of the audience. Furthermore, they are often communicated through media and other communication channels by tools such as advertising, public relations and publicity (Florek et al., 2006).
Place brand identity is a concept that is continually produced and reproduced by interactions (Kavaratzis and Hatch, 2013), and it becomes more complex because of the advancement of digital technologies and social media platforms. The next section reviews the literature on online place branding and relevant social media platforms.
Place branding on social media platforms
Social media is a platform that combines Web 2.0 technology and UGC capabilities (Kaplan and Haenlein, 2010) to allow users to create identities, engage in conversations, share content, find other people, foster relationships, build reputations and join communities (Kietzmann et al., 2011; Cvijikj et al., 2013). Compared with traditional counterparts, social media platforms are more interactive, open, transparent and participative, and as such, they become new and effective channels for place branding and marketing (Zhou and Wang, 2014). Social media platforms give rise to new ways to communicate information regarding place brands. Unlike other traditional media, social media platforms have the capability to transform passive audiences into active participants (Ketter and Avraham, 2012). Kavaratzis (2004) referred to this type of word-of-mouth communications by audiences that are uncontrolled by place marketers as tertiary communications. Accordingly, place brand managers need to find new, innovative ways to harness the power of online audiences in brand building (Andéhn et al., 2014). To do so, Hollander et al., (2016) conceptualised a new and powerful research method called urban social listening which refers to the study of people’s perceptions of places on social media. This method typically used technologies to gain insights into the attitudes and opinions of people at places.
In recent years, albeit in a limited fashion, several scholars have studied place branding on social media platforms. Sevin (2013), for example, studied how American destination marketing projects use Twitter. Similarly, Andéhn et al., (2014) analysed tertiary communications about Stockholm on Twitter, and Zhou and Wang (2014) discussed the use of Weibo, a microblogging platform, by Chinese cities. Fazel and Rajendran (2015) explored the perceptions and behaviours of Foursquare users in Sheffield. Hollander et al., (2016) studied the sentiment of twitter users in New Bedford, Massachusetts, USA. Thus, it can be concluded that exploring the UGC on social media platforms offers a new way to observe a place. In this research, the author explores UGC on three popular social media platforms, namely, Flickr, TripAdvisor and Google Maps’ Local Guide to understand and identify the place brand identity components of Khaosan Road and Yaowarat. A limitation of using such social media platforms is the inclination towards content generated by visitors or tourists, although the platforms are open to content created by anyone. Nevertheless, these social media platforms are widely used, and their UGC is a valuable source of information to understand place brand identity.
Flickr is a social media platform that specialises in photography repository and sharing. As one of the largest photo-sharing platforms, Flickr allows users to upload photos, comment on postings and join groups (Van Dijck, 2011). Users express themselves and explain how they watch the world through their photos (Van Dijck, 2011; Zeng and Wei, 2013). Flickr also allows users to tag the place where the photo was taken, which then allows the photos to serve as volunteered geographical information of that place (Hollenstein and Purves, 2010). When users continually and collectively share photos of a particular place, a collective memory, a view and an interpretation of the place are naturally created (Van Dijck, 2011). The collective memory of a particular place manifested by user-generated photos makes Flickr an excellent tool for studying place brand identity:
Based on the user-generated images on Flickr, what are the place brand identities of Khaosan Road and Yaowarat?
TripAdvisor is a travel community that attracts tens of millions of visitors annually (Jeacle and Carter, 2011; Lee et al., 2011), with travellers visiting the TripAdvisor website as they search for information about potential destinations (Ayeh et al., 2013). Specifically, they are seeking comments and information posted by other reviewers. One study found that UGC posted by other users is the most frequently used source of information (Cox et al., 2009). Accordingly, the narrative content posted by TripAdvisor users serves as another form of tertiary communication about a place (Jeacle and Carter, 2011). Similarly, Google Maps’ Local Guide is a new feature available on Google Maps that allows users to share reviews and knowledge about specific places on Google Maps. Google encourages users to write reviews and rate places on a scale of one to five stars, and in return for doing so, the users are awarded points that lead to benefits, such as early access to Google features and special perks from partners (Google, 2017):
What are the place brand identities of Khaosan Road and Yaowarat according to user-generated textual reviews on TripAdvisor and Google Maps?
TripAdvisor and Google Maps’ Local Guide ask users to rate their experiences at a location on a five-point scale. This feature allows researchers to investigate the difference between high-rating reviews (a reviewer’s good experience) and low-rating reviews (a reviewer’s bad experience). Knowing which concepts and issues are more frequently found in high-rating reviews and which ones are more common in low-rating reviews can benefit place brand managers. These managers can come up with strategies that reinforce positive concepts and mitigate negative ones:
Based on user-generated textual reviews, which concepts are more frequently found in high-rating reviews (positive concepts) and low-rating reviews (negative concepts)?
Although some studies have explored place brand identity on social media platforms, no study has compared a content analysis of user-generated images (image analysis) with a content analysis of textual reviews (text analysis) to analyse place brand identity. Therefore, the final research question is as follows:
What are the strengths and weaknesses of using image and text analyses to identify place brand identity components?
Content analysis of user-generated images on Flickr using MAXQDA
The images from Flickr were collected by searching images that listed Khaosan or Yaowarat as identification information. Flickr returned results that mentioned Khaosan or Yaowarat in the description, the album or the tags of the image. The author searched images taken within a five-year period from 1 July 2012 to 30 June 2017. The five-year period gives a comprehensive perspective of the two places. Furthermore, the author searched for images with a Creative Commons licence that allows anyone to download and share under certain terms. Overall, 579 images of Khaosan and 482 images of Yaowarat were returned and collected. Some images of Khaosan were removed from the analysis because:
they were images submitted by the Khaosan Pathet Lao news agency, which is the official news agency of the Laos government;
they were images of the Khaosan Beppu Spa Hostel in Japan;
they were not the images of Khaosan Road but were part of a larger user-generated album that includes the Khaosan tag; or
the term Khaosan in the images was used for other purposes. After the removal of these images, 399 images of Khaosan were retained for content analysis.
In the case of Yaowarat, the author removed duplicate images (colour versus black-and-white images), and 383 images of Yaowarat were used for content analysis. Figures 4 to 7 are examples of images of Khaosan Road and Yaowarat.
To analyse the place brand identities of Khaosan Road and Yaowarat, the author created categories that captured the three pillars of place brand identities of the two places. The author used inductive content analysis (Elo and Kyngäs, 2008) because there were no previous studies on the comprehensive characteristics of the two places. Furthermore, the previous list of generic characteristics developed by Sevin (2013) was not specific enough for the purpose of the current research. From the three pillars of place brand identity, the author used MAXQDA Analytics Pro 12 (release 12.3.2) to code place physics and place practices in images from Flickr. MAXQDA allows researchers to specify parts of the images according to codes. In the coding process using MAXQDA, 100 images were selected to test the intercoder reliability between the two coders. The intercoder reliability demonstrated an agreement of 98.2 per cent, revealing a good level of agreement (Lombard et al., 2002).
Regarding place personality, the author used both inductive and deductive content analysis techniques. There are many existing frameworks and studies on place brand personality (Hanna and Rowley, 2016). The author used the framework developed by Geuens et al., (2009) because it improved upon Aaker’s model by removing all non-personality items, and the measures were cross-culturally tested. Furthermore, the framework was also previously used in a study on place branding (Taecharungroj, 2017). There are five dimensions of brand personality including responsibility, activity, aggressiveness, simplicity and emotionality. Codes were deduced from this brand personality framework. Nevertheless, the author also used inductive content analysis by exploring the images and formulating image descriptions that fit the five dimensions of personality. Finally, the author summarised the codes to explain the three pillars of place brand identity from Khaosan Road and Yaowarat (RQ1).
Content analysis of user-generated reviews on TripAdvisor and Google Maps using Leximancer
User-generated, textual reviews of Khaosan Road and Yaowarat on TripAdvisor and Google Maps’ Local Guide written in English were retrieved in September 2017 using the script programmed in Python coding language. Subsequently, the reviews were studied using Leximancer, which is a powerful tool for developing an evidence-based global analysis that “would otherwise tax the cognitive abilities of an analyst” (Angus et al., 2013). Leximancer automatically generates lists of concepts based on the input text. It does so by using a modified algorithm from the field of computational linguistics to iteratively grow “a thesaurus of words around a set of initial seed words” (Leximancer, 2018). The advantage is that the list is statistically reliable and reproducible because it does not require manual intervention (Angus et al., 2013). This approach to content analysis requires no particular word strings. Instead, concepts are identified and coded automatically in a grounded fashion (Cretchley et al., 2010). Concepts from the analysis represent the components of the three pillars of place brand identity (RQ2).
Furthermore, to answer RQ3, another analysis was conducted to identify which concepts are more frequently found in high-rating reviews (positive concepts) and which ones are more common in low-rating reviews (negative concepts). The author considered four- and five-point reviews as “high-rating” reviews because they are regarded as above-average and excellent, respectively, (TripAdvisor, 2018). In contrast, “low-rating” reviews are those that receive three points (average) or below. Leximancer was used to count the frequencies of each concept in high- and low-rating reviews for both places. To evaluate each concept, the author formulated a value called “positivity of concept” (PCi) to determine how positive or negative the concept is. In the first calculation step, the frequency of concept i in high-rating reviews (CHi) is divided by the total frequency of all the concepts in high-rating reviews (TH). The same computation is performed for the low-rating reviews (CLi/TL). Finally, the formula computes the increase in the proportion of a concept from low- to high-rating reviews in a percentage term. The positivity of concept is the percentage of which the concept is more likely to be found in high-rating reviews than in low-rating ones. The formula is as follows:
Place brand managers can evaluate the results to develop appropriate strategies to reinforce strengths (positive concepts) and address shortcomings (negative concepts).
From the content analysis of the user-generated images on Flickr using MAXQDA, the author generated 105 codes for both places. These codes are divided into three levels, namely, groups, categories and subcategories which are the three pillars of place brand identity (Figure 8). The results of the content analysis of user-generated images were used to answer RQ1.
With respect to place physics, the author identified two main groups: businesses and objects. These two groups of codes constitute the place physics of Khaosan Road and Yaowarat. Inductive content analysis generated additional categories within each group. Regarding the businesses in the two areas, the author identified 11 categories (Figure 8). There are eight main categories and 38 subcategories in the object group. For example, the category of “business objects” has ten subcategories including products, signs, mobile advertisement, decorations, handheld advertisements, parts, cooking equipment and ingredients and miscellaneous business objects. Place practices were divided into people and activities groups. The categories in the people group are businesspersons, visitors, locals, festivalgoers, worshippers and crowds. Some categories, such as businesspersons and locals, have subcategories. For example, subcategories of “businesspersons” include merchants, cooks or food preparers, service workers, festival presenters, labourers, drivers, mechanics/repairers, massage therapists and miscellaneous businesspersons. Subsequently, the author analysed the roles of people in those images and identified activities accordingly. The author identified 14 categories of activities (Figure 8). Finally, Table I shows a list of personality dimensions and their descriptions according to Geuens et al., (2009). The author explored the images and formulated image descriptions that best match each personality dimension. For example, an image of a worker or a businessperson who focusses on the task at hand matches the description of responsibility, whereas an image of a street full of activity or traffic denotes activity. As the inductive approach is used, image descriptions depend on the locality. In the case of Khaosan Road and Yaowarat, images that fit with “activity” and “simplicity” are common; hence, there are more image descriptions that fall under those dimensions than those that fall under the emotionality and aggressiveness dimensions.
Table II displays the frequency of codes for Flickr images of Khaosan Road and Yaowarat. There are 1,541 coded items related to Khaosan Road and 1,399 coded items related to Yaowarat. Regarding place physics, the group with the highest number of coded items is objects, followed by businesses. In place practices, activities have the same number of codes as people because they are directly derived from the roles of the people in those images. The results indicate that although Khaosan Road and Yaowarat have numerous similarities, there are also some differences.
Businesses were the first group under the place physics category. Excluding miscellaneous and unidentifiable businesses, food stalls and restaurants were the most common businesses found in images of both Khaosan Road and Yaowarat (Table II). Aside from the two food-related business categories, fashion and beauty stalls were common on Khaosan Road (29) but less so in Yaowarat (7). Conversely, there were some business categories that were present in Yaowarat but not present on Khaosan Road, such as stores selling crops, medicines, herbs and dried foods (12), hardware and electronics stalls (8) and lottery stalls (6).
There were also similarities and differences in the object group. Categories of codes that were common both on Khaosan Road and in Yaowarat include vehicles, signage, business objects and food and beverage. Festival objects, such as water guns, ice buckets and goggles, were apparently only found on Khaosan Road, whereas temples and religious items, such as incense sticks, Buddha images and oil lamps, were more frequently identified in Yaowarat.
With respect to place practices, businesspersons have the highest number of codes in the people group for both Khaosan Road and Yaowarat. Visitors and locals were the two next most common categories related to both places. The difference is that there were substantially more visitors identified in the images of Khaosan Road (108 versus 47), whereas locals were identified more frequently in images of Yaowarat (85 versus 22). Crowds, meaning large groups of people whose roles are mixed and are difficult to identify, were also present in both places, but they were more common in the Khaosan Road images (23 versus 2). Moreover, it was found that each place had a unique category of people. Specifically, festivalgoers were only found in the Khaosan Road area, whereas worshippers were only found in the Yaowarat area. Of the 345 coded activities in Khaosan Road and 305 activities in Yaowarat, there are four categories of activities in Khaosan Road and three categories in Yaowarat that comprised more than 10 per cent of all the activities in the respective places (Table II). Attending festivals, walking down the street, promoting businesses and products and dining and drinking were the four most common categories of activities in Khaosan Road. However, the three most common categories of activities in Yaowarat were preparing food and beverages, selling and serving, commuting and worshipping.
“Activity” is the most common personality dimension identified in images of both Khaosan Road (70 per cent of images) and Yaowarat (41 per cent). Simplicity is the second most common dimension found in images of Khaosan Road, at 17 per cent, whereas responsibility is the second most common dimension found in Yaowarat’s images, at 31 per cent. Emotionality is the least common dimension, identified in three images for Yaowarat and none for Khaosan Road.
Text analysis – identifying concepts that constitute the three pillars of place brand identity
The author collected user-generated textual reviews of Khaosan Road and Yaowarat from TripAdvisor and Google Maps’ Local Guide by 30 September 2017. Overall, 4,794 reviews of Khaosan Road from TripAdvisor and 850 reviews from Google Maps’ Local Guide were analysed. In the case of Yaowarat, 2,846 reviews from TripAdvisor and 1,143 reviews from Google Maps’ Local Guide were analysed. Sets of reviews of each place on each platform were analysed separately by Leximancer. In the first stage, i.e. semantic extraction, Leximancer generated a list of concepts. The author examined those concepts and made necessary adjustments. Some concepts were merged, such as singular and plural nouns (e.g. tourist and tourists), different verb tenses (e.g. sell and selling) and spelling variances (e.g. Khao san, Kao San and Khaosan). Additionally, less notable terms were eliminated from the sets, such as “time”, “town” and “down”. The author also created compound concepts in Leximancer such as “nightlife”, “street food” and “shark fin”.
The final lists of concepts from the reviews of Khaosan Road are presented in Figure 9. There are four name concepts including Khaosan, Bangkok, Thai and Thailand; the rest are word concepts. The content analysis of the reviews on TripAdvisor generated more concepts because there are significantly more reviews on TripAdvisor than on Google Maps’ Local Guide. As a result, place, street, food, bars and night are the most common concepts in Khaosan Road’s reviews. In the case of Yaowarat (Figure 10), reviewers usually referred to the place as Chinatown. Hence, the name concepts that emerged from the analysis are Chinatown, Bangkok, Chinese, Thai and China. Furthermore, there are differences between the concept lists of the reviews from TripAdvisor and Google Maps. Thai, for example, is a concept that emerged from the reviews on TripAdvisor but not from those on Google Maps. Conversely, China is a concept that emerged from the reviews on Google Maps. Similar to Khaosan Road, there are fewer concepts from Google Maps’ Local Guide because there were fewer reviews in the analysis. Most of the concepts from Google Maps’ Local Guide are present in the concept list of the reviews from TripAdvisor, except for four concepts, namely, delicious, fin, shark and shark fin. Unlike Khaosan Road, food is a concept that tops both lists of reviews with respect to Yaowarat.
To illustrate the three pillars of place brand identity for Khaosan Road and Yaowarat from the user-generated reviews and to answer RQ2, the author categorised all the concepts into three categories according to parts of speech, namely, nouns (place physics), verbs and person nouns (place practices) and adjectives (place personality) (Table III). In comparison, place physics of Khaosan Road and Yaowarat share many similarities, such as streets, food, shops, restaurants, stalls and vendors. However, there are a few concepts that differ between the two places. As indicated by Table III, Khaosan Road’s unique place physics include music, tourists, backpackers and beer, whereas examples of Yaowarat’s unique place physics include seafood, gold and shark fin.
The place practices of the two places are also quite similar. For example, common practices by both visitors and locals in these two areas include visiting, walking, trying, shopping, buying, selling, experiencing and looking. However, Khaosan Road offers a few other unique activities such as partying, drinking and massages. Place personality is characterised by concepts that are represented by adjectives. Although these two places have similarities, they also have differences. Personality concepts that are common to both are Thai, busy, cheap, worthy, best, crowded and nice. Khaosan Road, however, offers unique personality concepts associated with partying, i.e. loud and drunk. Yaowarat also has some unique concepts used to describe food, i.e. amazing, fresh and delicious.
To identify positive and negative concepts for both places (RQ3), the author separated reviews on TripAdvisor into high- (4 and 5 points) and low-rating (3 points and below) reviews. TripAdvisor reviews were selected because it comprised more reviews than Google Maps. In total, there were 3,078 high-rating reviews of Khaosan Road and 1,716 low-rating reviews. For Yaowarat, there were 2,100 high-rating reviews and 746 low-rating reviews. The author used Leximancer to generate lists of concepts separately for high- and low-rating reviews of each place. From the calculations of the positivity of concept, the concept “everything” is the most positive concept for Yaowarat, at 153 per cent. The count of frequency of “everything” in the high-rating reviews of Yaowarat is 183; the total count of all concepts in the high-rating reviews for Yaowarat is 17,642. Therefore, the proportion of “everything” in the high-rating reviews is 1.04 per cent, or 183 divided by 17,642. Using a similar calculation, the proportion of “everything” in the low-rating reviews for Yaowarat is 0.41 per cent, or 22 divided by 5,371. The increase of the proportion of “everything” from low-rating reviews (0.41 per cent) to high-rating reviews (1.04 per cent), or its positivity, is 153 per cent. The full list of concepts and calculations is given in the Appendix (Tables AI and AII). Some concepts in this analysis were not shown in the previous analysis because of their increased prominence when the high- and low-rating reviews were separated. For example, the concept “pong”, which refers to a sexual entertainment show, became more prominent in the bad reviews for Khaosan Road, whereas it was less significant in the overall picture. Likewise, “dirty” became an apparent concept among the low-rating reviews for Yaowarat. After all the concepts were listed, the frequency of each concept in high- and low-rating reviews was counted. Positivity of concept (PCi) was calculated for each concept. In summary, Table IV shows the concepts that are in the top 15 and the bottom 15 in terms of positivity. The results can help place brand managers develop appropriate strategies.
Concepts that have high positivity are those that are commonly found in high-rating reviews. From numerous positive reviews, the five-point review by Suyash M on 12 May 2017 captured many positive concepts of Khaosan Road:
“Khao San Road is basically a giant block party. Later into the night - everyone is in tanks, shorts, flip flops, drinking out of buckets and dancing on tables. Think spring break, but urban Thai-style. There are a ton of open air bars most of which spill out on to the sidewalk/street, a couple of clubs and on the very same road there are massage chairs, street food vendors, clothing markets and even vendors selling laughing gas (try this once!). There is ZERO Thai-culture, and actually not that many Thais (the crowd is mostly young Caucasian backpackers). At night, this place is for drinking, partying and having fun. Come during the daytime if you want something else. Definitely a must-see when in Bangkok!” [emphasis added]
Positive concepts such as night, party, fun, clothing, street and food are woven together in this review, illuminating the exciting and active atmosphere of Khaosan Road. Another TripAdvisor user, Jo W, on 5 August 2017, described Khaosan Road as follows: “[i]t’s similar to Camden market but an insane Camden market!” For Yaowarat, many positive reviews revolve around the food scene. A five-point review by Drktel from the USA on 29 May 2017 is one of the examples of such a review:
“Eat, eat, eat everything on the street!! At night eat the street food, be fearless and have the best food of your life! We are foodies from Los Angeles with our own Chinatown, [and] it doesn’t even compare. Let yourself wander, your nose leading the way, be polite, and point, but many speak enough English for you to get delicious noodles, meats, soups, everything!” [emphasis added]
Another positive example was written by a user, gordonj258, from the UK on 7 April 2017.
“You cannot miss visiting Chinatown it really is a must see. Vibrant, bustling, hectic, it is all these things and more. There are many shrines to visit. Dark winding and mysterious back alleys. There are huge markets piled high with all sorts of goods, a good place to buy souvenirs to take home. At night the place springs into even more life with superb seafood and other street food. The neon signs on Yaowarat Road alone make a visit worthwhile. Altogether this place cannot be missed, you will need a full afternoon and night to see even a small part.” [emphasis added]
The positive reviews of Yaowarat portray the lively food experience of visitors. Place brand managers should reinforce and support those concepts in these two places. In contrast, highly negative concepts are more common in low-rating reviews. For example, a TripAdvisor user, Zof42, from Paris, France, wrote a negative one-point review of Khaosan Road on 26 January 2017. This review contained several highly negative concepts, including “drunk”, “tourists”, “selling”, “pong” and “loud”:
“We walked through this street several times and every time it was awful. It’s crowded with drunk tourists shouting, that think when they are in another country they can do everything (mostly crap), and Thai take advantage of it by selling “ping pong pus**”, “p**sy bottle” etc, “laughing gas”, insects to eat (when even them don’t eat that) [sic], music is much too loud […] It’s like the worst you can find gathered in one street. And the saddest thing is that many people reduce Bangkok to Khao San Road. Well, it’s not.” [emphasis added, asterisks from the original review]
In the case of Yaowarat, a TripAdvisor user, Rambler U, wrote a one-point review on 18 November 2015 that contained three highly negative concepts: “dirty”, “crowded” and “traffic”:
“It’s hell. Dirty, polluted, crowded, noisy, unsanitary, overrated, overpriced, full of rude sellers, full of cheap Chinese crap that you can easily buy elsewhere, horrible traffic jam. I really don’t understand how can a person enjoy being in this environment. I won’t even start with the lingering smells or “shark fin” selling restaurants! I think the existence of this entire neighbourhood should be illegal and they should shut it down. Life is short, don’t waste a minute here.” [emphasis added]
Place brand managers must determine appropriate strategies to mitigate these problems in both places. The text analysis results show the concepts that constitute the three pillars of place brand identity, their relationships and their positivity or negativity. The next section discusses research and practical implications of the findings.
Discussion and conclusion
Research implication - the novel approach of content analysis
In this research, the author developed a place brand identity framework that includes the three pillars of place physics, place practices, and place personality. UGC from social media platforms was then used to make sense of the place brand identities of Khaosan Road and Yaowarat. Content analysis of images on Flickr generated codes and a count of the frequency of the codes that represent those three pillars. The content analysis of textual reviews on TripAdvisor and Google Maps’ Local Guide created concepts that were also categorised in those three pillars. Furthermore, the author introduces a method that compares high- and low-rating reviews, allowing researchers to identify the positive and negative concepts of each place.
This research demonstrated how UGC from social media platforms can be used to study place brand identity by visually and semantically analysing its three pillars. This new dual approach of content analysis cultivates and infers a fluid and erratic place brand identity. The two methods similarly identified important place brand identity components of the two places. Both methods identified components – codes (image analysis) – and concepts (text analysis) – such as restaurants, stalls, shops (or stores), and vehicles (taxis and tuk tuks) – as place physics (Tables II and III). They both illustrate that food-related premises are the most prevalent place brand identity components of Khaosan Road and Yaowarat. Food-related businesses, including food stalls and restaurants, accounted for more than half of the businesses in the two places (image analysis, Table II), while “food”, “street”, and “places” are the three most frequently found concepts in Khaosan Road and Yaowarat (text analysis, Figures 9 and 10). Both methods also identified similar place practices such as visitors (or tourists), vendors (or businesspersons), and locals walking, drinking, selling and shopping.
The use of both image and text analyses allows the two methods to complement one another while mitigating the weaknesses of each. The image analysis using MAXQDA allows for a meticulous development of the codes, which permits researchers to immerse themselves in the images and thoroughly describe the items. As a result, it generates more detailed components of place brand identity; stalls and stores were categorised into food, fashion and beauty, crops/medicines/herbs/dried food, gold and jewellery, hardware and lottery. Another example is the more specific types of businesspersons, including merchants (or vendors), cooks or food preparers, service workers, festival presenters, labourers, drivers, mechanics and massage therapists. Image analysis also enables researchers to detect less salient items that may be neglected in textual reviews, such as signage and business objects including products, store signs, mobile advertisements, decorations, handheld advertisements, parts, cooking equipment and ingredients. There are certain weaknesses involved in using MAXQDA for image analysis. For example, content analysis of user-generated images can be subjective because it relies on the manual coding by researchers. Therefore, intercoder reliability is crucial for ensuring the reliability of the process. Image analysis is also time-consuming compared with automated text analysis using Leximancer. Finally, although it is possible to generate codes from items and behaviours of people in images to represent place physics and place practices, it is challenging to inductively interpret place personality from images. Therefore, the author used a combination of both inductive and deductive analyses, which led to a more limited scope of codes.
The most obvious strength of content analysis using Leximancer is the rapid generation of concepts from large data sets without manual intervention. Furthermore, the need for minimal intervention by researchers is beneficial because researcher bias is thereby reduced. Another important strength of text analysis is the detection of non-visual components. The text analysis (Table III) found many important non-visual components that could not be detected using image analysis, such as loud (music), delicious and amazing (food), hot (weather), fun (party) and cheap (things). These concepts are important for formulating a detailed description of place personality. There are a few drawbacks to using Leximancer for text analysis, however. Apart from the fact that it can only analyse texts, the analysis tends to omit less salient concepts that were not mentioned by the reviewers. Furthermore, user-generated reviews from platforms such as TripAdvisor and Google Maps’ Local Guide generally represent visitors’ perspectives, and thus, they may not include the perspectives of stakeholders such as the residents and local business people. To answer RQ4, Table V summarises the strengths and weaknesses of the two types of content analysis.
This research not only analyses the three pillars of place brand identity but it also illuminates their complex interactions. Complex interactions can be interpreted as institutions, experiences and representations (Figure 3). The results of image analyses show how images of visitors walking down the street while workers promote food and drinks in front of a restaurant illustrate a unique commercial “institution” on Khaosan Road. Some images show the cultural and religious institutions in Yaowarat, where the worshippers light incense and pray in the Chinese temple. Some images signify “experiences”, or interactions between place practices and personality, by pairing festivalgoers enjoying the event with the activity dimension of personality and cooks attentively preparing food with the responsibility dimension. These practices continually ascribe the places with the related dimension of personality, which, in turn, brings about more of these practices. Likewise, examples of “representations” include graffiti on Khaosan Road, which connotes aggressiveness and images of ordinary street food in Yaowarat, which imply simplicity. These interactions between place physics and personality imprint a mental image on the minds of the audience; such representations are replicated in place communications in the media or among stakeholders.
Whereas image analysis displays the interactions of the three pillars visually, text analysis reveals these interactions semantically. The results of an analysis of a large set of textual reviews demonstrate deep and complicated relationships among concepts that constitute place brand identity. Examples of reviews from TripAdvisor also demonstrate unique institutions, experiences and representations. Unique commercial institutions in both places were elaborated by interesting phrases such as a Thai vendor selling “insects to eat” on Khaosan Road or “huge markets piled high with all sorts of goods” for visitors to shop in Yaowarat. Reviewers also described vivid experiences, such as an awful walk with loud music on Khaosan Road or being urged to “eat, eat, eat everything” in Yaowarat while being “fearless” and “polite”. Representations were also richly explained, whether describing a dark winding and mysterious back alleys in Yaowarat or relating Khaosan Road to a famous Camden market in London, calling it “an insane Camden market”.
Practical implication –foundation for branding activities
This study’s findings will enable the BMA to understand the place brand identity of Khaosan Road and Yaowarat from the perspectives of users of social media platforms. The BMA could use this method to track changes after policy implementation, and it can benchmark places with others to compare their various strengths and weaknesses. From the results of both image and text analyses, street food vending is one of the most, if not the most, salient components of place brand identity of both Khaosan Road and Yaowarat. Hence, the April 2017 policy to eliminate street food vending backfired and created upheaval among the stakeholders. It is imperative that the BMA and the brand managers of other places investigate place brand identity to make the best branding decisions.
Based on the findings, the BMA could develop branding policies or activities that reinforce strengths (positive concepts) and mitigate issues (negative concepts). Avraham and Ketter (2017) offered several media strategies that a place can implement to counter negative images. For example, the notable negative concepts from the text analysis of Khaosan Road are “loud” and “drunk tourists”. To counter the current connections between music and drinking, the BMA could use the strategy of spinning liabilities into assets (Avraham and Ketter, 2017) by turning Khaosan Road into a music scene for both visitors and locals. The BMA could develop new ways to promote Khaosan Road as a music hub that promotes diversity and creativity. Text analysis also reveals several negative concepts that involve excessive commercialism in Khaosan Road, such as “money” and “selling”. Likewise, the image analysis identified many components of commercialism, such as the prevalence of signage and handheld advertisements. Although commerce is an indispensable component of a tourist destination, to improve the experiences of visitors and locals, the BMA should find a way to control excessive and aggressive commercialism, which has been notorious around tourist spots in Thailand (Taecharungroj and Tansitpong, 2017).
Although there were several shortcomings of Khaosan Road, its vibrant streets and activities are very strong, positive concepts. These features of Khaosan Road can be leveraged to communicate its brands in many effective ways. For example, the BMA can strategically use films, TV series and books (Avraham and Ketter, 2017), sponsoring young creative professionals or students to produce creative works that promote the authentic Thai side of Khaosan Road. The BMA could also launch its own content about benign activities on Khaosan Road, such as an online documentary about foot massages, the cultural aspects of tattoos or the stories of traditional accessories vendors.
Yaowarat has several prominent and positive components of place brand identity, such as shopping, and hence, the BMA could initiate a compelling online campaign that focusses on the things that are sold in Yaowarat. This type of online campaign could promote businesses to potential visitors and display the diversity of Yaowarat. Another set of notable and positive concepts includes seafood restaurants and stalls. To capitalise on these concepts, the BMA could use the strategy of hosting a spotlight event (Avraham and Ketter, 2017) such as a seafood festival that attracts both visitors and locals while supporting area businesses. Yaowarat, for instance, already has a strong reputation for hosting an annual spotlight event, specifically, a large vegetarian festival that famously blends the street food flair of local cooks with Chinese tradition (Bangkok.com, 2017). Although street food is its strength, the BMA needs to consider implementing policies that can improve the hygiene of Yaowarat because according to the findings, dirtiness is a negative issue that adversely affects certain visitors. However, improving hygiene could be difficult because a lack of hygiene typically related to street food vending (Maneepong and Walsh, 2013).
Another finding from both image and text analyses is that although walking and commuting, primarily by pedestrians, are prominent place practices, vehicles remain a salient identity of both places where “traffic”, “taxi” and “tuk (tuk)” are notable negative concepts. The images often show the co-presence of people and vehicles in the two areas. Thus, to improve the experience of both visitors and locals in the two places, the BMA could develop policies to reduce traffic and improve walkability both on Khaosan Road and in Yaowarat. Although the term “walkability” is not yet prevalent in the city marketing field, it is a renowned concept in urban design and planning (Ewing et al., 2006). A city’s walkability is the quality of walking and non-motorised travel conditions (Litman, 2003); it improves quality of life, physical health and mental health (Litman, 2003; Rogers et al., 2011).
Limitations and future research
Despite its contributions, this research is not without limitations. Regarding the research framework (Figure 3), the scope of this research concerns the place brand identity formulated from UGC. Therefore, some elements of the models in the existing literature, such as vision, mission and values (Ruzzier and De Chernatony, 2013), were not incorporated because they were primarily elements of the brand identity of corporations crafted by managers (De Chernatony and Dall’Olmo Riley, 1998). There are also some limitations about the methodology of this current research. The positivity of concept analysis is a new method to analyse concepts generated by Leximancer. Although the calculation is straightforward, additional studies should use this analysis across various cases to test its validity. Another limitation is that the UGC on Flickr, TripAdvisor, and Google Maps’ Local Guide is dominated by visitors’ viewpoints, even though the platforms are open to other stakeholders. Therefore, future research should explore the possibility of incorporating the viewpoints of residents and businesspersons from other social media platforms. Another interesting possibility would be to compare the results of this new method with the traditional method to identify place brand personality such as focus group interviews or workshops with key stakeholders. Finally, future research should methodically explore the interactions including institutions, experiences and representations of the three pillars of place brand identity.
Place personality’s content category list with descriptions
|Category||Personality description||Image description|
|5.1 Responsibility||Down-to-earth, responsible and stable||An image of a worker/businessperson who pays attention to the task at hand|
|5.2 Activity||Dynamic, innovative and active||An image of a street full of activities or traffic
An advertisement or signage
|An image of a fun activity|
|An image of commuting visitors or local people|
|An image of people during a festival|
|An image of people shopping|
|An image of people dining/drinking at a restaurant or a bar|
|An image of people worshipping at the temple|
|5.3 Aggressiveness||Bold and aggressive||Wall graffiti|
|An image of soldiers|
|5.4 Simplicity||Ordinary and simple||A simple portrait of a person who is doing nothing in particular, e.g., relaxing|
|An image of products such as bags, religious objects, spare parts, food or beverages|
|An image of an animal|
|An image of buildings, including commercial, residential and religious buildings|
|An image of a bar|
|An image of objects such as a boat, a decoration item, a stall, magazines or a potted plant|
|5.5 Emotionality||Romantic and sentimental||An image of a beggar expressing sadness|
Frequency of codes
|Codes||n (Khaosan)||(%) of group||n (Yaowarat)||(%) of group|
|1. Businesses (physics)||129||100||166||100|
|1.1 Stalls (food)||36||28||67||40|
|1.3 Stalls (fashion and beauty)||29||22||7||4|
|1.4 Stores (crops/medicines/herbs/dried foods)||0||0||12||7|
|1.5 Stores (fashion)||3||2||6||4|
|1.6 Stores (gold and jewellery)||2||2||6||4|
|1.7 Stalls (hardware)||0||0||8||5|
|1.8 Stalls (lottery)||0||0||6||4|
|1.9 Massage parlours||6||5||0||0|
|1.10 Commercial buildings (misc.)||10||8||27||16|
|1.11 Stalls (misc.)||6||5||10||6|
|2. Objects (physics)||323||100||240||100|
|2.3 Business objects||65||20||28||12|
|2.4 Festival objects||71||22||0||0|
|2.5 Food and beverages||35||11||13||5|
|2.6 Religious objects||3||1||31||13|
|2.8 Other objects||4||13||50||21|
|3. People (practices)||345||100||305||100|
|4. Activities (practices)||345||100||305||100|
|4.1 Preparing/selling/serving food and beverages||31||9||57||19|
|4.2 Attending festivals||79||23||0||0|
|4.3 Walking down the street||65||19||12||4|
|4.6 Promoting businesses/products||38||11||1||0|
|4.9 Selling of other items||5||1||15||5|
|4.11 Providing transport services||8||2||7||2|
|4.12 Selling of electronics/hardware||0||0||10||3|
|4.13 Selling of fashion items||10||3||3||1|
|4.14 Selling of crops/medicines/herbs/dried foods||0||0||8||3|
|4.15 Providing repair/mechanic services||0||0||7||2|
|4.16 Selling of festival items||6||2||0||0|
|4.17 Selling of lottery tickets||0||0||5||2|
|4.18 Providing foot massages||5||1||0||0|
|4.19 Practices (misc.)||42||12||43||14|
|5. Personality dimension (personality)||399||100||383||100|
Three pillars of place brand identity for Khaosan Road and Yaowarat from text analyses
|Part of speech||Khaosan road||Yaowarat|
|Place physics||Nouns||place; street; food; bars; night; road; street food; shops; music; restaurants; day; prices; stalls; markets; things; beer; world; life; clothes; taxis; hotels; tuk (tuk); money; nightlife; atmosphere||food; place; street; street food; shops; market; restaurants; stalls; night; price; day; things; road; soups; seafood; gold; taxis; traffic; city; goods; stuff; station; river; fin; shark; shark fin|
|Place practices||Verbs||visit; drink; party; sell; walk; massage; try; experience; shopping; stay; buy; take; look; enjoy; travel; pay||visit; walk; buy; sell; shopping; experience; take; try; look|
|Nouns (person)||people; tourists; backpackers; vendors; locals||people; vendors; locals|
|Place personality||Adjectives||Thai; cheap; fun; full; nice; loud; busy; worthy; best; better; young; fried; fake; expensive; crowded; drunk||Chinese, Thai; busy; cheap; crowded; best; worthy; interesting; different; nice; full; amazing; old; hot; fresh; delicious|
Unique concepts are italicised
Concept comparison between high- and low-rating reviews of Khaosan Road and Yaowarat
|High (4-5 points)||Clothes (1523%); Thai (59%); massage (24%); night (22%); fun (21%); street (12%); road (12%); life (11%); bars (10%); party (10%); Bangkok (10%); food (9%); drink; (9%); place (8%); shopping (7%)||Everything (153%); price (126%); fresh (97%); amazing (93%); best (88%); fish (83%); seafood (74%); restaurants (56%); try (45%); soup (37%); old (36%); river (35%); night (35%); experience (25%); food (23%)|
|Low (1-3 points)||Drunk (−56%); selling (−52%); better (−46%); try (−46%); money (−38%); fake (−37%); pong (−32%); taxi (−30%); loud (−24%); tourists (−19%); tuk (−18%); travel (−16%); market (−15%); Thailand (−15%); worth (−15%)||Dirty (−88%); crowded (−46%); stuff (−40%); traffic (−37%); hot (−35%); selling (−32%); tuk (−31%); goods (−30%); busy (−28%); trip (−27%); nice (−25%); people (−22%); Chinese (−21%); taxi (−21%); interesting (−21%)|
Strengths and weaknesses of image analysis and text analysis
|Image analysis using MAXQDA||Text analysis using Leximancer|
|Strengths||Meticulous development of codes
Detection of less salient items
|Rapid generation of concepts
Detection of non-visual components
Difficult to interpret place personality
|Omission of less salient concepts
Positivity of concepts for Khaosan Road from TripAdvisor
|Concept||Counts in low-rating reviews (CLi)||Proportion to total counts in low-rating reviews (TL)||Counts in high-rating reviews (CHi)||Proportion to total counts in high-rating reviews (TH)||Positivity of concept (PCi)|
Positivity of concepts for Yaowarat from TripAdvisor
|Concept||Counts in low-rating reviews (CLi)||Proportion to total counts in low-rating reviews (TL)||Counts in high-rating reviews (CHi)||Proportion to total counts in high-rating reviews (TH)||Positivity of concept (PCi)|
Aaker, D.A. (2012), Building Strong Brands, Simon and Schuster.
Aaker, J.L. (1997), “Dimensions of brand personality”, Journal of Marketing Research, Vol. 34 No. 3, pp. 347-356.
Andéhn, M., Kazeminia, A., Lucarelli, A. and Sevin, E. (2014), “User-generated place brand equity on twitter: the dynamics of brand associations in social media”, Place Branding and Public Diplomacy, Vol. 10 No. 2, pp. 132-144.
Angus, D., Rintel, S. and Wiles, J. (2013), “Making sense of big text: a visual-first approach for analysing text data using leximancer and discursis”, International Journal of Social Research Methodology, Vol. 16 No. 3, pp. 261-267.
Anholt, S. (2007), “What is competitive identity?”, In Competitive Identity, Springer, pp. 1-23.
Ashworth, G. and Kavaratzis, M. (2009), “Beyond the logo: brand management for cities”, Journal of Brand Management, Vol. 16 No. 8, pp. 520-531.
Avraham, E. and Ketter, E. (2017), “Destination marketing during and following crises: combating negative images in asia”, Journal of Travel and Tourism Marketing, Vol. 34 No. 6, pp. 709-718.
Ayeh, J.K., Au, N. and Law, R. (2013), “Do we believe in TripAdvisor?” examining credibility perceptions and online travelers’ attitude toward using user-generated content”, Journal of Travel Research, Vol. 52 No. 4, pp. 437-452.
Bangkok.com (2017), Vegetarian Festival in Bangkok 2017. available at: http://bangkok.com/whats-on-events/vegetarian-festival.htm
Barke, M. and Harrop, K. (1994), “Selling the industrial town: identity, image and illusion”, Place Promotion: The Use of Publicity and Marketing to Sell Towns and Regions, pp. 93-114.
Beeho, A.J. and Prentice, R.C. (1997), “Conceptualizing the experiences of heritage tourists: a case study of new lanark world heritage village”, Tourism Management, Vol. 18 No. 2, pp. 75-87.
Boisen, M., Terlouw, K. and van Gorp, B. (2011), “The selective nature of place branding and the layering of spatial identities”, Journal of Place Management and Development, Vol. 4 No. 2, pp. 135-147.
Boonjubun, C. (2017), “Conflicts over streets: the eviction of Bangkok street vendors”, Cities, Vol. 70, pp. 22-31.
Botschen, G., Promberger, K. and Bernhart, J. (2017), “Brand-driven identity development of places”, Journal of Place Management and Development, Vol. 10 No. 2, pp. 152-172.
Bouchon, F.A. (2014), “Truly Asia and global city? Branding strategies and contested identities in Kuala Lumpur”, Place Branding and Public Diplomacy, Vol. 10 No. 1, pp. 6-18.
Bregoli, I. (2013), “Effects of DMO coordination on destination brand identity: a mixed-method study on the city of Edinburgh”, Journal of Travel Research, Vol. 52 No. 2, pp. 212-224.
Cai, L.A. (2002), “Cooperative branding for rural destinations”, Annals of Tourism Research, Vol. 29 No. 3, pp. 720-742.
Coffey, H. (2017), “News that Bangkok’s thriving street food scene is under threat has spread far and wide – but there are some doubts about whether it’s actually happening”, available at: http://independent.co.uk/travel/news-and-advice/bangkok-bans-street-food-chinatown-yaowarat-khao-san-road-thailand-capital-city-a7690526.html
Cox, C., Burgess, S., Sellitto, C. and Buultjens, J. (2009), “The role of user-generated content in tourists’ travel planning behavior”, Journal of Hospitality Marketing and Management, Vol. 18 No. 8, pp. 743-764.
Cretchley, J., Rooney, D. and Gallois, C. (2010), “Mapping a 40-year history with leximancer: Themes and concepts in the journal of Cross-Cultural psychology”, Journal of Cross-Cultural Psychology, Vol. 41 No. 3, pp. 318-328.
De Chernatony, L. and Dall’Olmo Riley, F. (1998), “Modelling the components of the brand”, European Journal of Marketing, Vol. 32 Nos 11/12, pp. 1074-1090.
Elo, S. and Kyngäs, H. (2008), “The qualitative content analysis process”, Journal of Advanced Nursing, Vol. 62 No. 1, pp. 107-115.
Ewing, R., Handy, S., Brownson, R.C., Clemente, O. and Winston, E. (2006), “Identifying and measuring urban design qualities related to walkability”, Journal of Physical Activity and Health, Vol. 3 No. s1, pp. S223-S240.
Fazel, M. and Rajendran, L.P. (2015), “Image of place as a byproduct of medium: Understanding media and place through case study of foursquare”, City, Culture and Society, Vol. 6 No. 1, pp. 19-33.
Florek, M., Insch, A. and Gnoth, J. (2006), “City council websites as a means of place brand identity communication”, Place Branding and Branding, Vol. 2 No. 4, pp. 276-296.
Fredrickson, T. (2017), “Street food update: Partial reprieve, stricter rules”, available at: http://bangkokpost.com/learning/advanced/1235251/street-food-to-stay-in-parts-of-khao-san-yaowarat
Geuens, M., Weijters, B. and De Wulf, K. (2009), “A new measure of brand personality”, International Journal of Research in Marketing, Vol. 26 No. 2, pp. 97-107.
Glińska, E. and Gorbaniuk, O. (2016), “Restrictions on the use of the aaker model in the measurement of city brand personality: the sender’s perspective”, Place Branding and Public Diplomacy, Vol. 12 No. 1, pp. 46-58.
Google (2017), “Local guides: How to contribute on google maps”, available at: https://maps.google.com/localguides/howto
Gould, M. and Skinner, H. (2007), “Branding on ambiguity? Place branding without a national identity: Marketing Northern Ireland as a post-conflict society in the USA”, Place Branding and Public Diplomacy, Vol. 3 No. 1, pp. 100-113.
Govers, R., Kaefer, F. and Ferrer-Roca, N. (2017), “The state of academic place branding research according to practitioners”, Place Branding and Public Diplomacy, Vol. 13 No. 1, pp. 1-3.
Hankinson, G. (2004), “Relational network brands: towards a conceptual model of place brands”, Journal of Vacation Marketing, Vol. 10 No. 2, pp. 109-121.
Hanna, S. and Rowley, J. (2011), “Towards a strategic place brand-management model”, Journal of Marketing Management, Vol. 27 Nos 5/6, pp. 458-476.
Hanna, S. and Rowley, J. (2016), “Do places have a personality?”, Strategic Place Branding Methodologies and Theory for Tourist Attraction, Vol. 21.
Hofstede, H. (2014), “Balancing between thick and thin regional identities: the case of De achterhoek, The Netherlands”, Journal of Place Management and Development, Vol. 7 No. 2, pp. 126-140.
Hollander, J.B., Graves, E., Renski, H., Foster-Karim, C., Wiley, A. and Das, D. (2016), Urban Social Listening: potential and Pitfalls for Using Microblogging Data in Studying Cities, Springer.
Hollenstein, L. and Purves, R. (2010), “Exploring place through user-generated content: Using flickr tags to describe city cores”, Journal of Spatial Information Science, Vol. 2010 No. 1, pp. 21-48.
Iversen, N.M. and Hem, L.E. (2008), “Provenance associations as core values of place umbrella brands: a framework of characteristics”, European Journal of Marketing, Vol. 42 Nos 5/6, pp. 603-626.
Jeacle, I. and Carter, C. (2011), “In TripAdvisor we trust: Rankings, calculative regimes and abstract systems”, Accounting, Organizations and Society, Vol. 36 Nos 4/5, pp. 293-309.
Kalandides, A. (2011), “The problem with spatial identity: revisiting the ‘sense of place’”, Journal of Place Management and Development, Vol. 4 No. 1, pp. 28-39.
Kalandides, A., Millington, S., Parker, C. and Quin, S. (2016), “Shopping districts and centres, markets, neighbourhoods, public squares, and urban gardens: Reflecting upon place management practice in Berlin”, Journal of Place Management and Development, Vol. 9 No. 3, pp. 351-359.
Kaplan, A.M. and Haenlein, M. (2010), “Users of the world, unite! the challenges and opportunities of social media”, Business Horizons, Vol. 53 No. 1, pp. 59-68.
Kavaratzis, M. (2004), “From city marketing to city branding: towards a theoretical framework for developing city brands”, Place Branding and Branding, Vol. 1 No. 1, pp. 58-73.
Kavaratzis, M. (2012), “From ‘necessary evil’ to necessity: stakeholders’ involvement in place branding”, Journal of Place Management and Development, Vol. 5 No. 1, pp. 7-19.
Kavaratzis, M. and Hatch, M.J. (2013), “The dynamics of place brands: An identity-based approach to place branding theory”, Marketing Theory, Vol. 13 No. 1, pp. 69-86.
Ketter, E. and Avraham, E. (2012), “The social revolution of place marketing: the growing power of users in social media campaigns”, Place Branding and Public Diplomacy, Vol. 8 No. 4, pp. 285-294.
Kietzmann, J.H., Hermkens, K., McCarthy, I.P. and Silvestre, B.S. (2011), “Social media? Get serious! Understanding the functional building blocks of social media”, Business Horizons, Vol. 54 No. 3, pp. 241-251.
Konecnik, M. and Go, F. (2008), “Tourism destination brand identity: the case of Slovenia”, Journal of Brand Management, Vol. 15 No. 3, pp. 177-189.
Lee, H.A., Law, R. and Murphy, J. (2011), “Helpful reviewers in TripAdvisor, an online travel community”, Journal of Travel and Tourism Marketing, Vol. 28 No. 7, pp. 675-688.
Levy, D. and Lee, C.K. (2011), “Neighbourhood identities and household location choice: estate agents’ perspectives”, Journal of Place Management and Development, Vol. 4 No. 3, pp. 243-263.
Leximancer (2018), “Leximancer concept miner”, available at: https://www.leximancer.com/site-media/lm/science/leximancer-concept-miner.pdf
Lindstedt, J. (2011), “Place, identity and the socially responsible construction of place brands”, Place Branding and Public Diplomacy, Vol. 7 No. 1, pp. 42-49.
Litman, T. (2003), “Economic value of walkability”, Transportation Research Record: Journal of the Transportation Research Board, Vol. 1828, pp. 3-11.
Lombard, M., Snyder-Duch, J. and Bracken, C.C. (2002), “Content analysis in mass communication: Assessment and reporting of intercoder reliability”, Human Communication Research, Vol. 28 No. 4, pp. 587-604.
Lucarelli, A. (2012), “Unraveling the complexity of ‘city brand equity’: a three-dimensional framework”, Journal of Place Management and Development, Vol. 5 No. 3, pp. 231-252.
Mahnken, G. (2011), “Place identity beyond province and metropolis: paths and perspectives in Germany’s ‘Capital region’ Berlin-Brandenburg”, Journal of Place Management and Development, Vol. 4 No. 1, pp. 67-79.
Maneepong, C. and Walsh, J.C. (2013), “A new generation of Bangkok street vendors: Economic crisis as opportunity and threat”, Cities, Vol. 34, pp. 37-43.
Martínez, N.M. (2016), “Towards a network place branding through multiple stakeholders and based on cultural identities: the case of ‘the coffee cultural landscape’ in Colombia”, Journal of Place Management and Development, Vol. 9 No. 1, pp. 73-90.
Moilanen, T. (2015), “Challenges of city branding: a comparative study of 10 European cities”, Place Branding and Public Diplomacy, Vol. 11 No. 3, pp. 216-225.
Mueller, A. and Schade, M. (2012), “Symbols and place identity: a semiotic approach to internal place branding–case study Bremen (Germany)”, Journal of Place Management and Development, Vol. 5 No. 1, pp. 81-92.
Peighambari, K., Sattari, S., Foster, T. and Wallström, Å (2016), “Two tales of one city: Image versus identity”, Place Branding and Public Diplomacy, Vol. 12 No. 4, pp. 314-328.
Pryor, S. and Grossbart, S. (2007), “Creating meaning on main street: towards a model of place branding”, Place Branding and Public Diplomacy, Vol. 3 No. 4, pp. 291-304.
Rogers, S.H., Halstead, J.M., Gardner, K.H. and Carlson, C.H. (2011), “Examining walkability and social Capital as indicators of quality of life at the municipal and neighborhood scales”, Applied Research in Quality of Life, Vol. 6 No. 2, pp. 201-213.
Rujivanarom, P. (2017), “BMA bans all street food across Bangkok this year”, available at: http://nationmultimedia.com/news/national/30312543
Ruzzier, M.K. and De Chernatony, L. (2013), “Developing and applying a place brand identity model: the case of Slovenia”, Journal of Business Research, Vol. 66 No. 1, pp. 45-52.
Sarabia-Sanchez, F.J. and Cerda-Bertomeu, M.J. (2016), “Place brand developers’ perceptions of brand identity, brand architecture and neutrality in place brand development”, Place Branding and Public Diplomacy, Vol. 1-14.
Sevin, E. (2013), “Places going viral: Twitter usage patterns in destination marketing and place branding”, Journal of Place Management and Development, Vol. 6 No. 3, pp. 227-239.
Shoaib, T. and Keivani, R. (2015), “Branding the new city: exploring place branding in Saudi Arabia”, Journal of Place Management and Development, Vol. 8 No. 3, pp. 254-265.
Skinner, H. (2008), “The emergence and development of place marketing’s confused identity”, Journal of Marketing Management, Vol. 24 No. 9-10, pp. 915-928.
Skinner, H. and Kubacki, K. (2007), “Unravelling the complex relationship between nationhood, national and cultural identity, and place branding”, Place Branding and Public Diplomacy, Vol. 3 No. 4, pp. 305-316.
Stock, F. (2009), “Identity, image and brand: a conceptual framework”, Place Branding and Public Diplomacy, Vol. 5 No. 2, pp. 118-125.
Taecharungroj, V. (2017), “City-district divergence grid: a multi-level city brand positioning tool”, Place Branding and Public Diplomacy, Vol. 14 No. 2.
Taecharungroj, V. and Tansitpong, P. (2017), “Attractions, attitude, and activities: tourism attributes that drive destination loyalty for international tourists visiting Thailand”, International Journal of Tourism Policy, Vol. 7 No. 2, pp. 129-150.
TripAdvisor (2018), “All about your TripAdvisor bubble rating. How your bubble rating is determined, how you can improve it and more”, available at: https://tripadvisor.com/TripAdvisorInsights/w810
Trueman, M., Klemm, M. and Giroud, A. (2004), “Can a city communicate? Bradford as a corporate brand”, Corporate Communications: an International Journal, Vol. 9 No. 4, pp. 317-330.
Van Dijck, J. (2011), “Flickr and the culture of connectivity: Sharing views, experiences, memories”, Memory Studies, Vol. 4 No. 4, pp. 401-415.
Van Riel, C.B. and Balmer, J.M. (1997), “Corporate identity: the concept, its measurement and management”, European Journal of Marketing, Vol. 31 No. 5, pp. 340-355.
Warnaby, G. (2009), “Towards a service-dominant place marketing logic”, Marketing Theory, Vol. 9 No. 4, pp. 403-423.
Zeng, X. and Wei, L. (2013), “Social ties and user content generation: Evidence from flickr”, Information Systems Research, Vol. 24 No. 1, pp. 71-87.
Zenker, S. and Martin, N. (2011), “Measuring success in place marketing and branding”, Place Branding and Public Diplomacy, Vol. 7 No. 1, pp. 32-41.
Zhou, L. and Wang, T. (2014), “Social media: a new vehicle for city marketing in China”, Cities, Vol. 37, pp. 27-32.
The author thanks Cathy Parker, Dominic Medway, Gareth Roberts, Heather Skinner, the delegates to the 5th Corfu Symposium on Managing & Marketing Places, and the two anonymous reviewers whose valuable comments help improve the manuscript. The author also thanks a research assistant, Jeesun Kelley Park. This research is funded by Mahidol University International College (Research Contract No. 05/2018).