A Detailed Method For Destination Image Analysis Using

Inf Technol Tourism (2016) 15:287–290
DOI 10.1007/s40558-015-0050-z
EDITORIAL
Editorial: Special issue on ENTER2015
Iis Tussyadiah1 • Alessandro Inversini2
Received: 17 December 2015 / Revised: 21 December 2015 / Accepted: 23 December 2015 /

Published online: 8 January 2016
Ó Springer-Verlag Berlin Heidelberg 2016
As information and communication technology (ICT) has become integral to key

processes supporting travel and tourism, it is important for research and development
to continue providing innovative approaches and critical reflections for the effective
use of ICT in the tourism industry. The papers published in this volume are extended
versions of highly selected research papers among the many contribution presented at
ENTER2015 eTourism Conference, which took place in Lugano, Switzerland, on
February 3–6, 2015, organized by the International Federation for IT and Travel &
Tourism (IFITT). While far from being inclusive of all research areas in ICT and
tourism, the collection of five articles in this special issue provides an excellent
selection of the latest research on applications of ICT in travel and tourism.
Social media has been one of the prominent areas of research in tourism and ICT
since the last decade (Dwivedi et al. 2011; Hvass and Munar 2012; Leung et al.
2013). In tourism literature, the topic is highly associated with the different aspects
of tourism marketing (Hays et al. 2013), including consumer behaviour (Munar and
Jacobsen 2014; Tham et al. 2013; Xiang and Gretzel 2010) and destination branding
(Lim et al. 2012; Munar 2011). The first three articles in this volume discuss how
tourism destinations and hospitality businesses make strategic use of social media
and user-generated content to communicate and assess destination brands among
their consumers. The first article (Huertas and Marine-Roig 2015) explores the
effectiveness of destination brand communication via social media by analysing
user reactions (i.e., likes, comments, shares, mentions, etc.) to destination brand
& Alessandro Inversini

ainversini@bournemouth.ac.uk
1
School of Hospitality Business Management, Carson College of Business, Washington State
University, Vancouver, WA, USA
2
Faculty of Management, Bournemouth University, Poole, Dorset, UK
123
288 I. Tussyadiah, A. Inversini
content (i.e., brand image and emotional values) posted on social media channels. In
particular, analysing selected destinations in Spain, this research identified specific
destination attributes and/or themes that generated most reactions from audience,
which reflect the value of brand engagement and interactivity. In this way, the
article provides Destination Marketing Organisations (DMOs) with suggestions to
identify social media content strategies that are effective in maintaining user
engagement and, at the same time, increasing destination popularity.
The second article (Lalicic and Dickinger 2015) demonstrates how DMOs can
assess destination brands from the consumer perspectives using two different
methods: an analysis of user-generated content from social media and a survey. In
particular, utilizing Aaker’s (1997) brand personality model, this article extracts the
dimensions of brand personality and emotional values from TripAdvisor reviews
across different aspects of tourism services (e.g., restaurants, hotels, attractions)
using content analysis and sentiment classification. Comparing the results of text
analysis with consumer responses to a survey, the researchers found different
dimensions from user-generated content that are not represented in survey
responses, suggesting the potentials of user-generated content to provide DMOs
with valuable knowledge to inform the development of proactive brand positioning
strategies. With the abundance of user-generated content afforded by the many
platforms of social media, this article signifies the importance of listening to
consumer conversation to be able to extract valuable intelligence for the
development of destination branding strategies.
In this vein, the third article (Marine-Roig and Clavé 2015) presents a detailed
method to analyse massive user-generated content from social media. Specifically, the
paper provides a guideline for semi-automatic downloading, arrangement, cleaning,
debugging, and analysis of large-scale travel blogs and online travel review data. To
illustrate the usefulness of the proposed method, the researchers extracted and
analysed more than 130,000 trip diaries from visitors to Catalonia, Spain between
2004 and 2014. The content analysis on the dataset revealed patterns concerning the
distribution of destination brands across different travel reviews that are important for
National Tourism Organizations (NTOs) and DMOs to improve their marketing and
branding policies. The results from this research confirm the value of extracting
knowledge from user-generated content with a greater level of details for the
development of marketing intelligence and destination performance metrics.
The fourth article (Oses et al. 2015) analyses the practice of dynamic pricing
among hoteliers in Basque Country, Spain by collecting hotel room price data
periodically from an Internet distribution channel in 2013–2014. Different
visualization and analysis techniques assisted in interpreting patterns of room rate
changes from a total of 3,924,016 observations, which include information
regarding when and why price changes occur in addition to the prices themselves.
As a result, two patterns of dynamic pricing practices were identified: hotels are
changing a number of prices of contiguous, future target dates on the same date or
changing prices a set number of days in advance. The evidence of dynamic pricing
practices on online booking engine can be useful for hoteliers to improve their
pricing strategies by taking into consideration their competitors’ practices as well as
for online travel agencies and infomediaries to provide more accurate booking
123
Editorial: Special issue on ENTER2015 289
recommendation for consumers based on the identified patterns of dynamic pricing

practices.
While the aforementioned articles discuss the advantages of harnessing
technology to assist strategic decisions among service providers and tourism
destinations, the last article (Dinhopl and Gretzel 2015) discusses the behavioural
implications of technological development in videography on various tourist
practices. In particular, the paper aims at developing a theoretical foundation of
tourist videography as a distinct practice from photography. While both practices of
tourist photography and videography signify the role of tourists as consumers and
producers of tourism-related media, this article demonstrates that videography and
photography are different in the practices of representation, seen from both
technological and social perspectives, as well as in the ways in which they mediate
tourism experiences. As a consequence of this new practice, destination and
attraction managers need to carefully integrate the facilitation of videography into
the design of tourism sites and experiences.
The five articles in this collection enrich the literature on ICT and tourism in
many ways. Firstly, while most of these papers confirm the application of existing
concepts in new research contexts (e.g., destination image and brand personality
models; Aaker 1997), they contribute to the important discussion and conceptu-
alization of emotion-based theory in destination branding (e.g., Hosany et al. 2014),
tourism and visual culture (Haldrup and Larsen, 2010), and mediation of tourism
experiences (Jansson 2002, 2007; Jennings and Weiler 2006; Tussyadiah and
Fesenmaier 2009). Secondly, these articles demonstrate the use of different methods
for collecting, managing, and analysing massive data sourced online from social
media, booking engines, and other publicly available data sources for tourism
research. As tourists and service providers increasingly use ICT and digital media
for various tourism experiences, the challenge for tourism destinations is to extract
valuable knowledge about consumer behaviour and experiences from a massive
amount of data collected online (Fuchs et al. 2014). The articles in this special issue
present the state-of-the-art of data analysis and visualization, which are useful for
research involving large-scale user-generated content. Lastly, these articles provide
managerial implications for NTOs, DMOs, hoteliers, and other business entities in
tourism, especially in areas of marketing (i.e., including branding and pricing) and
experience design.
References
Aaker JL (1997) Dimensions of brand personality. J Mark Res 34(3):347–356
Dinhopl A, Gretzel U (2015) Conceptualizing tourist videography. Inf Technol Tour, this issue
Dwivedi M, Yadav A, Venkatesh U (2011) Use of social media by National Tourism Organizations: a
preliminary analysis. Inf Technol Tour 13(2):93–103
Fuchs M, Höpken W, Lexhagen M (2014) Big data analytics for knowledge generation in tourism
destinations—a case from Sweden. J Destin Mark Manag 3(4):198–209
Haldrup M, Larsen J (2010) Tourism, performance, and the everyday. Routledge, London
123
290 I. Tussyadiah, A. Inversini
Hays S, Page SJ, Buhalis D (2013) Social media as a destination marketing tool: its use by national
tourism organisations. Curr Issues Tour 16:211–239
Hosany S, Prayag G, Deesilatham S, Cauševic S, Odeh K (2014) Measuring tourists’ emotional
experiences: further validation of the destination emotion scale. J Travel Res 59(5):483–495
Huertas A, Marine-Roig E (2015) User reactions to destination brand contents in social media. Inf
Technol Tour, this issue
Hvass KA, Munar AM (2012) The takeoff of social media in tourism. J Vacat Mark 18(2):93–103
Jansson A (2002) Spatial phantasmagoria: the mediatisation of tourism experience. Eur J Commun
17(4):429–443
Jansson A (2007) A sense of tourism: new media and the dialectic of encapsulation/decapsulation. Tour
Stud 7(1):5–24
Jennings G, Weiler B (2006) Mediating meaning: perspectives on brokering quality tourist experiences.
In: Jennings G, Nickerson N (eds) Quality tourism experiences. Elsevier Butterworth-Heinemann,
Oxford
Lalicic L, Dickinger A (2015) An analysis of destination brand personality and emotions: a comparison
study. Inf Technol Tour, this issue
Leung D, van Law R, Hoof H, Buhalis D (2013) Social media in tourism and hospitality: a literature
review. J Travel Tour Mark 30:3–22
Lim Y, Chung Y, Weaver PA (2012) The impact of social media on destination branding: consumer-
generated videos versus destination marketer-generated videos. J Vacat Mark 18(3):197–206
Marine-Roig E, Clavé SA (2015) A detailed method for destination image analysing using user-generated
content. Inf Technol Tour, this issue
Munar AM (2011) Tourist-created content: rethinking destination branding. Int J Cult Tour Hosp Res
5:291–305
Munar AM, Jacobsen JKS (2014) Motivations for sharing tourism experiences through social media. Tour
Manag 43:46–54
Oses N, Gerrikagoitia JK, Alzua A (2015) Evidence of hotels’ dynamic pricing patterns on an internet
distribution channel: the case study of the Basque Country’s hotels in 2013–2014. Inf Technol Tour,
this issue
Tham A, Croy G, Mair J (2013) Social media in destination choice: distinctive electronic word-of-mouth
dimensions. J Travel Tour Mark 30(1–2):144–155
Tussyadiah IP, Fesenmaier DR (2009) Mediating tourist experiences: access to places via shared videos.
Ann Tour Res 36:24–40
Xiang Z, Gretzel U (2010) Role of social media in online travel information search. Tour Manag
31:179–188
123
DOI 10.1007/s40558-015-0045-9
ORIGINAL RESEARCH
User reactions to destination brand contents in social

media
Assumpcio Huertas1 • Estela Marine-Roig2
Received: 12 June 2015 / Revised: 30 November 2015 / Accepted: 8 December 2015 /

Published online: 22 December 2015
Springer-Verlag Berlin Heidelberg 2015
Abstract With the aim of communicating their identities and brands, tourist
destinations have started to use social media. However, many do not know very well
how to manage them in order to improve their online communication. Several
studies have shown that a high level of interactivity generated with users in social
media leads to greater engagement and a better brand image. The aim of this study
is to analyse the brand contents that generate the most reactions among users and to
ascertain if the communication of the brand image and its emotional values also
generates reactions. Results show that the most identifying or destination-specific
themes/attributes are the ones that trigger the most reactions (interactivity), although
very often generic attributes are also communicated. Moreover, this study shows
that the communication of emotional brand values also generates reactions,
although emotional values are rarely communicated by tourist destinations. The
brand values and attributes that trigger the most user reactions do not usually
coincide with the most mentioned ones by destinations in their posts.
Keywords Online communication Social media Facebook Interactivity

Destination brand Content analysis
This paper is an extended and updated version of a conference paper previously published in ENTER
Proceedings of the International Conference in Lugano, Switzerland, February 3–6, 2015.
& Estela Marine-Roig

estela.marine@aegern.udl.cat
Assumpcio Huertas
sunsi.huertas@urv.cat
1
Communication Research Group (ASTERISC), Rovira i Virgili University, Catalonia, Spain
2
Department of Business Administration and Economic Management of Natural Resources
(AEGERN), University of Lleida, Catalonia, Spain
123
292 A. Huertas, E. Marine-Roig
1 Introduction
Social media are ‘‘a group of Internet-based applications that build on the
ideological and technological foundations of Web 2.0 and allow the creation and
exchange of User Generated Content’’ (Kaplan and Haenlein 2010, p. 61). Social
media are usually classified into social networks (e.g., Facebook), professional
networks (e.g., LinkedIn), video-sharing websites (e.g., YouTube), picture-sharing
platforms (e.g., Flickr), social bookmarks (e.g., Delicious, Digg), review websites
(e.g., TripAdvisor), wikis (e.g., Wikipedia), microblogging (e.g., Twitter), blogs
(e.g., Blogger), and user forums (Marine-Roig 2014; Tsimonis and Dimitriadis
2014). These social media allow users and organizations to engage in social
interactions in a way and on a scale that were not previously possible (Tsimonis and
Dimitriadis 2014).
In this context, social media have become important tools for the communication
of tourist destinations and their brands. They have changed communication in all
areas (Agarwal et al. 2011), but especially in the field of tourism and destinations
(Xiang and Gretzel 2010). Given that in general the places that people want to visit
they do not previously know, the opinions and estimations of other users, who have
no economic or other interests in the territory, can become highly credible (Litvin
et al. 2008), can be perceived as more trustworthy than the information from sources
such as official destination websites (Fotis et al. 2012; Leung et al. 2013), and can
influence tourists’ decision-making (Schmallegger and Carson 2010; Yoo and
Gretzel 2011; Zhang et al. 2010). However, some researchers (Jacobsen and Munar
2012) have found that social media are complementary information sources for
tourists, who continue to use traditional sources for information purposes.
‘‘Using social media, organizations can forge relationships with existing as well
as new customers and form communities that interactively collaborate to identify
problems and develop solutions for them’’ (Tsimonis and Dimitriadis 2014). In this
respect, social media have great potentialities to establish relations with users, to
know their images and necessities, to allow comments and participation/interaction,
and to communicate destination brands effectively. Within social media, Facebook
is the most-used social network both by users worldwide, according to Alexa.com,
and by Spanish tourist destinations (Huertas and Marine-Roig 2014).
Given the importance of social media as communication and marketing tools for
Destination Management Organizations (DMOs) and National Tourism Organiza-
tions (NTOs) (Leung et al. 2013; Hays et al. 2013), most destinations have started to
use them (mainly Facebook) in their communications; however, many do not know
very well how to manage them in order to reach their target publics and to know
what communications strategies to follow or what information to transmit (Huertas
and Marine-Roig 2014). In fact, social media usage among top DMOs is still largely
experimental, and strategies vary significantly (Hays et al. 2013). Because of this
usage, destinations need to assess the results of their social media communications
(Huertas and Marine-Roig 2014; Hays et al. 2013), and for that purpose, there has
been an increase in the number of professional tools created for data collection,
classification, and analysis (Marchiori and Cantoni 2012). These tools are especially
123
User reactions to destination brand contents in social media 293
useful to analyse social media data, which are often voluminous, rapid, and diverse
(Jang et al. 2013). Researchers have found that aspects of interactivity, especially
user reactions (e.g., likes, comments or shares), are the main items measured by
these online tools, but that, conversely, they have great limitations in the content
analysis of posts (Huertas and Marine-Roig 2014). They basically measure the most
frequent words or keywords, but they still have a long way to go in a more
qualitative analysis of the content of posts (Mandelli and Cantoni 2010) or in the
communication of brand values. In addition, content results are not crossed with
interactivity items such as reactions.
On the other hand, there are numerous studies about tourist destination brand
image (Echtner and Ritchie 1991; Keller 2003; Govers and Go 2003; Govers et al.
2007). Traditionally, the authors of these studies have used quantitative method-
ologies (Echtner and Ritchie 1991; Jenkins 1999; Pike 2002) based on measuring a
list of predetermined attributes. Other authors have used qualitative methodologies
(Tasci et al. 2007; Pan and Li 2011; Li and Stepchenkova 2012) based on
unstructured approaches, focus groups, and content analysis, which result in richer
data and deeper analysis. But, we should note that there are very few studies about
how tourist destinations use social media to communicate their identities and
brands. There is a lack of understanding about how and why organizations are
actually using them (Tsimonis and Dimitriadis 2014). Destination management
organizations should evaluate their communication in order to ascertain whether
they communicate their brands correctly and how users react to their
communications.
For all of the reasons above, this study seeks to contribute by conducting a
content analysis of Facebook posts with the aim of seeing which contents generate
comparatively more reactions and, consequently, more interactivity between
destinations and users and testing if the communication of emotional values of
destination brands also generates more user reactions. This aim is carried out by
taking into account different types of destination and with the purpose of seeing if
this information might prove useful to them and provide them with practical
suggestions. This methodology, of an exploratory nature, will be applied to the
official Facebook Fanpages of a selection of Spanish destinations. Special emphasis
will be given to the differences in user reactions to various destination contents
according to these different destination types, both in general and also by taking into
account the volume of ‘fans’ they have.
2 Background
This section first introduces the concept of destination brands and then focuses on
destination brand communication through social media, emphasising the role of
what contents are transmitted. Finally, the interactivity generated in these
communication channels will be discussed.
123
2.1 Destination brands
Destination brands have two important functions: their identification and their
differentiation, or distinction, from those of other destinations (Qu et al. 2011;
Huertas 2014). Similar to products, destination brands include a name, logo, or
symbol to identify and differentiate them (Koltringer and Dickinger 2015), but
unlike products, places include a complex variety of tangible and intangible entities
that represent them, from historical sites or beaches to culture, customs, and history
(Qu et al. 2011).
These entities or elements are encompassed in the destination brand identity,
which is the way in which DMOs and other agents project destination images into
the minds of target tourists, versus a destination image, which is considered the
tourists’ perception and interpretation of such destination’s image identity (Marine-
Roig 2015). Brand identity is created by the destination. It is a generalization of the
desired characteristics of the place to be promoted, which should enable tourists to
develop a relationship with a particular brand through a proposition that has value
for them (Qu et al. 2011). We note here that all tourist destinations have an identity
and should create a brand to identify them. Although not all destinations have
created their brands consciously, these brands, more widely understood as identities
(Lawson and Band-Bovy 1977), always exist and must be communicated.
Most academic papers about destination brands, since the beginning until today
(Kapferer 1997; Baloglu and McCleary 1999; Echtner and Ritchie 1991; Hosany
et al. 2006; Fernandez-Cavia et al. 2014; Huertas 2014; Marine-Roig 2015), affirm
that brands are made up of two principal components: tangible attributes, or
cognitive elements, and emotional values, or affective elements. Destination brands,
understood as identity (Lawson and Band-Bovy 1977), convey the duality of
cognitive and affective aspects: a functional or tangible aspect and an emotional
one, or the aspects relating to attraction factors and the values relating to emotion.
Attraction factors are interest assets that destinations have, and emotional values are
the values that represent their personalities.
The key to destination branding is that consumers perceive the difference
between brands at a product level because a brand that is seen as distinctive and
unique is hard to replace (Qu et al. 2011). The prospect that tourists will have a
positive brand image can be realized through an emphasis on strong, positive, and
unique brand associations (Qu et al. 2011). In fact, in an increasingly competitive
marketplace, destination branding represents a way of communicating the unique
identity of a destination so that it can set itself apart from its competitors (Qu et al.
2011; Marine-Roig 2015).
2.2 Destination brand communication in social media
Social media are key tools for communication that enable the effective commu-
nication of tourist destination identities and brands (Munar 2011). Thus, they
become important tools for destination branding. With social media, destinations
face, on the one hand, the challenge of mastering a powerful information source
123
and, on the other hand, the task of comprehending the rich data repository provided
by travellers (Koltringer and Dickinger 2015).
Tourists’ experiences and narratives in these social media spaces, where they also
express their feelings, help destinations to construct an online reputation (Inversini
and Buhalis 2009; Xiang and Gretzel 2010; Marchiori and Cantoni 2012) and create
an interpretative framework through which users establish relationships with their
brands (Tussyadiah and Fesenmaier 2009). Through social media, new conversa-
tions and destination identities are created (Govers and Go 2003; Marine-Roig
2013). Furthermore, active participation by users in social media or online
communities increases the emotional attachment users have to the brand of the
online space or territory (Algesheimer et al. 2005) and increases brand loyalty.
In this context, the destination brand identity must be a key aspect in destination
communication through social media as it usually enables the creation of a positive
image among the publics, the association and identification of users with certain
emotional brand values, and the generation of a favourable predisposition for a visit.
Some studies analyse how social media influence the formation of a destination
brand image among users and the relationships users create with brands (Govers
et al. 2007; Laroche et al. 2013; Stepchenkova and Zhan 2013). An empirical study
based on surveys of users (Laroche et al. 2013) demonstrated that social media have
positive effects on the user-brand relationship, which at the same time generates
positive effects on credibility and brand loyalty. Other studies analyse how tourist
destinations communicate their brands through these channels (De Moya and Jain
2013).
For all of the above, the identity and the brand of the destination, with their two
principal components (attributes or tourist attractions and emotional values), should
be key in the communication of tourist destinations through social media.
However, we should note that generally studies show a dominance in the
communication of tangible attributes to the detriment of the communication of
emotional values (Bigné et al. 2009; Michaelidou et al. 2013). Recent research by
Kim and Stepchenkova (2015) about the use of photographs in destination brand
communication through social media showed that photographs have great potential
for the communication of emotional values, but curiously, destinations use
photographs more to communicate attractions than the emotional values of the
brand.
Not only should these functional and emotional values be communicated, but
destination managers should assess or evaluate their branding strategies in relation
to a constructed identity, to their idea of what they want to be (Marine-Roig 2015).
Comparing projected and perceived images of a destination offers one way to
evaluate brand communication in social media (Koltringer and Dickinger 2015;
Marine-Roig 2015); the more similar the perceived brand image is to the projected
brand identity, the better. Additionally, an essential part in the evaluation of any
destination brand communication is the content of the communication itself. This
content evaluation is essential in online environments (Fernandez-Cavia et al.
2014). DMOs should assess whether or not the values and attributes of the brand are
being transmitted, what the most recurrent themes are, what types of messages are
being sent, and how they are being sent in their communications through social
123
media. Renowned authors in the field of destination brands state that to achieve
successful brand communication, the destination must transmit a single identity and
communicate the brand’s emotional values to appeal to users’ emotions in a way
that users feel attracted to the place (Morgan et al. 2003). Other authors deal with
how the interactivity and participation of the publics generate better brand image
(Munar 2011), but not the other way around—whether brand communication, with
its tangible attraction factors and emotional values, generates more reactions and
interactivity. Lastly, studies based on the content analysis of destinations’ social
media have analysed what users perceive from the online contents published by
destinations or what information they wish to find when they search for tourist
information (Marchiori and Cantoni 2012), but not what information or contents
effectively generate most of their reactions and interactivity.
2.3 Reactions and interactivity
Social media are eminent relational tools for destinations. Social media, and among
them Facebook, are characterized by their inherent interactivity (Lovejoy and
Saxton 2012) as they enable the reactions of users and the establishment of dialogue
with them (Saffer, Sommerfeldt, and Taylor 2013). ‘‘Interactivity is the process by
which customers are integrated into the organization, made part of the product
planning and development process, and dealt with as individuals’’ (Niemann-
Struweg 2014:6). On the one hand, the interactive nature of social media allows
users to share and exchange information with one another and, on the other hand, it
allows sellers to share information and interact with their customers, establishing
dialogue or trialogue, in which users engage with one another and with sellers in
meaningful relationships (Tsimonis and Dimitriadis 2014). Interactivity indicates
the extent to which the user is actively engaged with the content or promotion of the
site (Jung and Butler 2000). Therefore, users’ reactions to contents are fundamental
items to measure in the interactivity of social media. These reactions are measured
in likes, comments, and shares on Facebook.
In tourism, interactivity involves conveying the message that a destination is
interesting, offering something relevant to a specific target market, and telling
(potential) tourists that their points of view and experiences are truly welcome (Arlt
2006). Interactivity is also said to have positive effects on tourist satisfaction,
engagement, and decision-making (Walther and Jang 2012). Interactivity is a key
aspect in the success of a website (Jung and Butler 2000) and should, therefore, be
one of the key aspects of analysis for a destination’s online communication
(Fernandez-Cavia et al. 2014). Moreover, in social media, interactivity is
fundamental in developing a long-lasting relationship and gaining the commitment
of users on travel-related social network websites (Nusair et al. 2013). It is a key
tool to ensure a stream of target group-relevant information (Arlt 2006).
Consequently, interactivity in communication through social media has multiple
advantages: greater surfing time and involvement of users (Huertas and Fernández
Cavia 2006), greater processing and impact of information (Sicilia et al. 2005; Ko
et al. 2005), and greater capacity for the creation of relationships with users (Nusair
et al. 2013). Moreover, it generates an enhanced destination brand image (Munar
123
2011), maximizes added value (Buhalis and Law 2008), produces positive effects in
satisfaction and tourist decision-making (Walther and Jang 2012), and focuses on
specific target groups to create major opportunities in product distribution (Buhalis
and Law 2008). In this context, the contents of communication in social media are
the main elements for establishing dialogue and generating reactions from the public
around a specific brand, as well as for getting audiences to join, interact, or engage
in the different social media platforms (Valentini 2015).
In terms of the analysis of interactivity in social media, some authors look at the
engagement that social media generates among users (Del Chiappa 2011; Nusair
et al. 2013), while others consider the influence that interactivity has on the public’s
thought and behaviour (Fischer and Reuber 2011). Still others observe how
interactivity influences the creation of relationships users develop with organiza-
tions and their brands (Saffer et al. 2013). Recently, Munar and Jacobsen (2014)
analysed the motivations leading people to share tourism experiences in social
media.
In terms of the contents that generate the most interactivity or user reactions,
some studies have focused on an analysis of the types of media contents
communicated through different social media channels (video, photographs, links,
text, etc.) (Munar and Jacobsen 2014) and how these different content types
generate more engagement with and reactions from the public (Sabate et al. 2014).
Specifically, Sabate et al. (2014) found that posts with images and videos are more
prone to be liked and commented on in Facebook (more reactions) and to lead to
greater popularity.
Although most relevant studies have focused on an analysis of the contents of
communication and on brand content curation strategies (Miralbell et al. 2014), no
studies have been found that analyse which contents in tourist destination social
media generate the most reactions and the most user interactivity, and none has been
found that analyses different types of destination brand elements (tangible/func-
tional and emotional).
Moreover, in most online social media analysis tools, measurements of user
reactions are the main indicators of user interactivity with the posted contents
(Huertas and Marine-Roig 2014). Reactions to post contents usually consist of likes/
dislikes, comments, and shares. Hence, measuring user reactions to different
contents becomes a major item in the investigation of interactivity and the
assessment of destination communication in social media.
3 Methodology
The proposed methodology consists of a content analysis of attraction factors and

brand values in several of the destinations’ top Facebook Fanpage posts. It then
assesses which attraction factors and brand values trigger the most reactions
(operationalized through the sum of likes, comments, and shares) in general and
according to different destination types. This methodology is based on the tools of
communicative and relational analysis because it combines an analysis of contents
with the interactivity they generate. Our study can thus be useful to DMOs in
123
directing and adjusting their communication and branding strategies. We must

emphasize that this methodology is of an exploratory nature, and aims to give an
approximation and gain familiarity with a specific reality, in a relatively unexplored
field.
3.1 Destination and social media selection
The selected destinations for analysis are Spain (as a whole) and a total of 33
Spanish destinations located in five autonomous communities or regions (Andalusia,
Canary Islands, Catalonia, Galicia, and Madrid). These five regions were selected as
being representative of Spain according to the Nielsen areas in market research. The
34 destinations correspond to different types of destinations: National destination
(ND), Autonomous community (AC), Large municipality (LM), corresponding to
capital cities of Autonomous communities, Large littoral destination (LLD),
Heritage city (HC), Littoral destination (LD), Inland destination (ID), Medium-sized
city (MC), and Mountain destination (MD). These types of destinations are
proposed in the Manual of Local Tourism Management Models of the Spanish
Ministry of Industry, Tourism, and Trade (FEMP and TourSpain 2008) as being
representative of the different types of Spanish tourist destinations. The destinations
are grouped according to their common characteristics and the type of tourism they
attract. One destination of each type was chosen for each of the five communities in
our study, if its geographic features so permitted. In the analysis, ND and AC were
merged into one category as they represent umbrella brands that promote different
types of destinations, territories, and tourist modes in a relatively equilibrate and
politically correct way.
The social media selected for analysis were the official Facebook tourist pages of
these destinations. Facebook is by far the most widely used social media platform
among the selected destinations. Moreover, Facebook is also the most widely used
social media platform by users in general and is ranked the second most visited
website in Spain and in the world by Alexa (http://www.alexa.com/topsites). It
therefore represents a very important part of the communications in the social media
domain.
To better understand and to further contextualize results, Table 1 shows the
average Facebook Fanpage fans of each type of destination. In the case of the
studied destinations, these numbers of fans are in general proportional to the
territorial size and touristic importance of the destination and to the typology or
main assets of the destination.
3.2 Dataset
To retrieve data, different online tools for social media analysis were considered and
explored. The methods to measure interactivity usually involve ‘‘the extent to which
the visitor actively engages with the Web content or advertisement’’ (Jung and
Butler 2000, p. 168). Although different tools retrieve similar content and
interactivity data, FanPage Karma (http://www.fanpagekarma.com) was chosen
because it provides both information about the top posts (those with the most user
123
Table 1 Average Facebook Fanpage fans per destination type

Destination type ND AC LM LLD LD HC MC ID MD
Average number of fans 843,000 87,776 41,115 19,251 3754 6173 3916 3006 562
Data retrieved on 01/05/2014
reactions) during a period of time and the reactions of users to these posts. FanPage
Karma collects and measures public information available on the destinations’
Facebook Fanpages that would be very time-costly to retrieve and count manually.
On 1 May 2014, we downloaded a total of 34 Facebook Fanpage Reports from
FanPage Karma corresponding to each of the destinations chosen for the case study.
These reports included a list of the top posts during the whole month of April 2014,
with their associated user reactions. This period of time corresponded to mid-high
tourist season in Spain as it includes the Easter holiday period. We defined top posts
as those that present the most user reactions in total (sum of likes, comments, and
shares) and generate the most interactivity through user reactions (likes, comments,
and shares) during a period of 1 month on each of these Facebook Fanpages.
Analysis of the top posts was needed for this study because their contents were the
ones that caused the greatest impact or reached the broadest audience. The retrieved
lists of top posts usually consisted of 25 posts. However, for some destinations
whose Fanpages are less active, the list of top posts may have included fewer posts.
Thus, the dataset consisted of a matrix with a list of a total of 714 top posts of
destinations, each with the information about generated reactions (likes, comments,
and shares).
It is important to emphasize that the 714 analysed posts were already the top
posts triggering the most reactions over a period of time. In this respect, this sample
supports previous works (Sabate et al. 2014) in that the posts included multimedia,
especially photographs that trigger more reactions than posts with only text. Of
these 714 top posts, 509 contained photographs (71 %), 29 contained videos (4 %),
142 contained links (20 %), and 34 consisted only of textual/status messages (5 %).
These numbers indicate that people react more to posts with multimedia, especially
photographs.
3.3 Content analysis
Content analysis is ‘‘a method of gathering and analysing a text or content of a piece
of writing into various groups or categories depending on selected criteria and by
systematically identifying specified characteristics or patterns.’’ It also makes
predictions and draws inferences from data (Opoku 2006, p. 24). For example,
Opoku (2006) used content analysis to analyse destinations’ official website
communications.
The methods most widely used to analyse content are keyword frequency,
sentiment and coverage analysis, semantic analysis, and the association of topics to
brand values (Marchiori and Cantoni 2012). However, the authors note that there are
no standard models or procedures to conduct content analysis, so there is a need to
123
build templates or models for the content analysis of destination communication

through social media. Our methodology intends to contribute to this project.
To conduct content analysis on online media, some authors identify different
themes in communication texts (Beerli and Martin 2004; Xiang et al. 2008; Wenger
2008; Marine-Roig 2013) such as natural resources, leisure and recreation, culture,
art, history, atmosphere, transportation, accommodation, activities, shopping,
climate, etc. However, to analyse destination brand communication, thematic
content analysis does not suffice. Accordingly, some studies build specific analysis
templates for brand attributes (Aaker 1997), that, among other attributes, have been
widely used in research (De Moya and Jain 2013), and Huertas (2014) specifically
for tourist destination brands.
As previously explained, destination brands comprise cognitive/functional and
affective/emotional dimensions, features that should be considered when assessing
destination brand content communication. Therefore, the items proposed and used
in our methodology to analyse content of identity or brand communication are the
following:
(A) Attraction factors, Activities, and Services (themes) Nature (Nature and
natural landscape, Rural landscape, Mountain, Ecotourism); Tangible Heritage
(Sites, History, Religion, Works of art, Museums); Cityscape (Architecture, Urban
planning/landscape); Intangible Heritage (Intangible heritage/popular culture/tradi-
tions, Anthem/flag/national symbols); Gastronomy (Food/cuisine, Wine tourism);
Leisure (Urban and cultural leisure/shows, Night life, Shopping); Sun and Beach
(Sea/Beach, Sun, Climate/weather); Business/trade; Sports (Hiking, Winter sports,
Water sports, Adventure sports, Elite sports, Other sports); Technology (Social
media/ICT, Technology, Innovation); Services (Hotel/accommodation, Transport,
Other services); Things to Do; Tourist information/agenda; Institutional and Non-
tourist information.
These categories were based on an extensive literature review of previously used
categories (Beerli and Martin 2004; Xiang et al. 2008; Choi et al. 2007, among
others) and on the contents of the communications themselves.
(B) Brand values (We used an adaptation of the ‘‘Brand Personality Scale’’ by
Aaker (1997), which has been extended with other attributes and values relevant for
the analysis of tourist destinations).
• Sincerity: Down-to-earth (Family-oriented, Down-to-earth, Sustainable); Honest

(Calm, Real, Traditional, Honest); Wholesome (Original, Wholesome; Quality
of life); Cheerful (Happiness, Sentimental, Friendly).
• Excitement: Daring (Trendy, Daring, Exciting, Exotic, Fashionable); Spirited
(Cool, Spirited, Dynamic, Vital, Fresh, Young, Sensorial); Imaginative (Unique/
different/diverse, Imaginative, Creative); Up-to-date (Up-to-date, Independent,
Contemporary, Modern); Cosmopolitan (Cosmopolitan, Tolerant, Hospitable).
• Competence: Reliable (Reliable, Hard-working, Secure/safe, Rigorous/respon-
sible/pragmatic); Intelligent (Intelligent, Technical, Corporate, Innovative);
Successful (Successful, Leader, Ambitious, Powerful).
• Sophistication: Luxurious (Glamorous, Luxurious); Charming (Charming/
seductive, Smooth, Romantic, Magical).
123
• Ruggedness: Outdoorsy (Outdoorsy, Get-away, Recreational); Tough (Tough,

Rugged, Nonconformist).
For the classification of post contents into the different categories, the team of
researchers first agreed on the criteria to code content in the same ways and set
examples for the different categories. These contents were to be explicit or clear in
the post communication. In this case, the team of researchers classified the content
of the posts manually after reading them. This coding of data was done separately
and then put together. It was then checked to be sure that the content was classified
and coded in the same way. In the case of disagreement, the post in question was not
classified in the category.
3.4 Data analysis
First of all, we measured the average reactions to top posts in absolute numbers and
as a percentage of fans per destination type. Then we calculated the most mentioned
themes and brand values per destination type and the themes and brand values
triggering comparatively more reactions as a percentage of fans in general and per
destination type. Calculations per themes and brand values were in general
presented in likes, comments, shares, and total reactions. Calculations per
destination type were presented in terms of total reactions (sum of likes, comments,
and shares). Afterwards, a comparison of the most mentioned themes and brand
values vs. the themes and values triggering the most reactions was conducted in
general and per destination type.
4 Results
Table 2 shows the different numbers of top posts analysed per type of destination,
from the most numerous posts, corresponding to ND ? AC Facebook Fanpages, to
MD with the fewest posts. The differences in post numbers are due to the fact that
some destination types did not always have activity on their Facebook Fanpages or
Table 2 Average reactions in absolute numbers and as a percentage of fans per destination type
Destination type ND ? AC LM LLD LD HC MC ID MD
Total posts 178 104 128 40 119 63 56 26

Avg. likes 1437.6 153.7 25.4 48.4 60.3 36.0 19.8 29.6
Avg. comments 35.4 4.9 0.7 1.3 3.0 1.2 0.4 0.7
Avg. shares 248.1 25.1 3.6 3.8 7.0 35.6 2.3 0.8
Avg. total reactions 1721.0 183.7 29.7 53.4 70.3 72.8 22.5 31.1
Avg, % of fan likes 0.572 0.520 0.254 0.992 0.983 1.034 0.684 2.751
Avg. % of fan comments 0.019 0.025 0.007 0.024 0.043 0.034 0.011 0.067
Avg. % of fan shares 0.128 0.123 0.050 0.063 0.106 0.592 0.088 0.074
Avg. % of fan total reactions 0.718 0.667 0.311 1.078 1.132 1.659 0.783 2.891
123
had personal Facebook pages (not public Fanpages). Therefore, their top posts could
not be analysed. In the case of average reactions in absolute numbers, we observed
that the top posts of ND ? AC have by far the highest number of reactions per post.
LM (all autonomous community capitals) have the second highest average reactions
per post, but about ten times less than ND ? AC. In absolute numbers, regions and
cities have higher average reactions to their top posts if compared to littoral
destinations or inland and mountain destinations, which have the fewest reactions.
This finding can be explained by their different fan numbers. However, if we take
the average reactions to top posts per destination type relative to the number of fans
(as percentages), we observe that this tendency reverses. In this case, it is Mountain
destinations, Medium-sized cities; Heritage cities and Littoral destinations that have
more reactions relative to their fans. This result could indicate that, in general,
Facebook Fanpages with fewer fans manage to engage more fans or trigger more
reactions from their fans relatively better than pages with higher numbers of fans.
These results lead to an initial implication for social media managers of tourist
destinations. The main aim of communication should be not only to achieve the
highest number of fans, but to get their reactions and a high degree of engagement.
Destinations should avoid using social media as a one-way channel of information
and should use all their potential to stimulate engagement and generate interactive
dialogue with users.
Moreover, it is worth mentioning that all types of reactions (likes, comments, and
shares), have strongly significant (at a level 0.01) Pearson correlations between
them, superior to 0.8, meaning that when one reaction grows, the rest also grow in a
similar proportion: likes and comments (0.889), likes and shares (0.884), and
comments and shares (0.802). Therefore, this finding supports our hypothesis that
these different types of user reactions (likes, comments, and shares) can be grouped
and assumed to behave in a proportional way.
It is also interesting to see the proportionality of likes, comments, and shares on
average, which is that for every 100 likes, users leave about 3 comments and share
the content 16 times. This proportionality is similarly maintained across different
destination types. It shows that getting users to share destination contents, and
especially to comment on them, is much harder than getting likes and may indicate
to destinations what to expect and where to place their efforts.
4.1 Mentions and reactions per themes
4.1.1 Most mentioned themes
The most mentioned theme in general (Table 3) is Tangible Heritage (298)

[especially History (149) and Sites (70)].
The next most mentioned themes are Agenda (255), Nature (247), and Sun and
beach (218). The rest of the themes are mentioned much less. Remarkably, if we
look specifically at the different types of destinations, we find that in the case of
ND ? AC, Nature and Tangible Heritage are the most mentioned themes with little
difference. Also, according to the type of destination, the brand image and the
activities available there (some of which are the most mentioned assets) are
123
Table 3 Number of mentions of themes in top posts

ND ? AC LM LLD LD HC MC ID MD Total
Tangible heritage 105 38 20 9 65 40 20 1 298

Agenda 24 25 60 30 44 38 29 5 255
Nature 106 13 53 13 25 9 5 23 247
Sun and beach 69 18 85 16 16 10 3 1 218
Leisure 24 30 15 14 35 46 35 4 203
Intangible heritage 41 18 28 8 22 29 13 0 159
Cityscape 51 31 6 3 30 24 8 0 153
Sports 9 13 21 4 12 5 3 6 73
Non-tourist information 4 15 9 3 24 0 4 13 72
Institutional 10 11 4 4 19 4 7 2 61
Technology 8 10 4 0 7 3 0 0 32
Services 2 3 9 3 5 0 0 0 22
Business/trade 4 0 5 0 2 1 0 0 12
expected: Nature (MD), Tangible Heritage (LM and HC), Leisure (MC and ID), and
Sun and beach (LLD). However, in the case of LD, Agenda is the most mentioned
theme, and the second most mentioned is LLD, with no direct relationship to the
littoral nature of these destinations.
4.1.2 Themes triggering the most reactions
As shown in Fig. 1 and Table 4, Cityscape, although not one of the most mentioned
themes by destination, is the general theme that triggers the most average reactions
from the public, especially the subcategory Architecture (1.63). It is followed by
Nature, Tangible Heritage and Sun and beach, which, in this case, are among the
most mentioned themes.
1.4 1.31
1.24
1.2
0.96 0.95
1 0.88 0.82 0.81 0.78 0.74 0.72 0.71
0.8 0.6
0.6
0.37
0.4
0.2
0
Fig. 1 Reactions per theme as a percentage of the number of fans
123
As observed the themes that have the most reactions from the public are varied,
but in all cases highly imageable, themes corresponding to unique or characteristic
attributes of the Spanish destinations studied (Nature, Tangible Heritage, Cityscape,
or Sun and beach) and are highly related to tourist activities. These themes follow a
similar order of prominence if we just focus on likes, comments, or shares. In terms
of likes, Nature comes first, followed by Cityscape, Sun and beach, and Tangible
Heritage. In terms of comments and shares, Cityscape comes first.
In contrast, the themes that provoke the fewest reactions from the public are the
most generic and least related to the tourist characteristics of the destination:
Business/trade, Institutional messages, Services, and Agenda. However, other
unique subjects that are representative of the identity of the destinations such as
Intangible Heritage, do not trigger as many reactions from the public, which may
indicate a lack of awareness.
Moreover, non-tourist information comes in the middle of the classification. This
result can be explained by posts related to some specific tragic events that, although
not related to tourism, triggered remarkably high reactions and in some cases the
sharing of information related to some non-touristic issues such as football match
results, which have generally had relatively high reactions.
These results have implications for social media managers, who in many cases
use social media to inform and give more importance to the Agenda that they
publish than a picture with a comment. However, this last is usually what triggers
most reactions and, therefore, greater user engagement.
4.2 Mentions and reactions per brand values
4.2.1 Most mentioned brand values
The most mentioned brand value groups by destinations (Table 5) are Honest
[especially Traditional (172)]; Imaginative, with the outstanding attribute of
Unique/different/diverse (104); Outdoorsy; Down-to-earth, especially the Family-
oriented subcategory (78); and Wholesome, emphasizing Original (54) values. If we
look specifically at the different destination types, the most prominently mentioned
destinations are the Honest and Imaginative values. Remarkably, however, the case
of ND ? AC, if compared to other types of destinations, has the attributes Spirited
and Charming in a relatively prominent position, which emphasises the beauty and
exciting nature of the destination brands. Also remarkable are ID and MD, which
have as the most mentioned brand value Outdoorsy, related to nature and outdoor
activities. However, we observed that in the transmitted brand values in posts and
the different types of destinations are quite indistinct and convey similar values in a
similar order of prominence.
Therefore, these results show that tourist destinations generally do not have a
communication strategy to communicate their identifying brand values. It is a
shame that, in many cases, destinations do not use social media to communicate
their distinctive brands to differentiate themselves from other places.
Some of the most mentioned brand values emphasize the uniqueness, originality
and traditional values of the destinations. They are also related to Leisure values
123
Table 4 Average reactions to themes as a percentage of fans in general and per destination type
% Like/fans % Com/fans % Shar/fans % Reac/fans ND ? AC LM LLD LD HC MC ID MD
Cityscape 0.99 0.04 0.28 1.31 0.68 0.93 0.34 2.53 1.97 3.07 0.34 –
Nature 1.06 0.03 0.14 1.24 0.92 0.89 0.4 1.69 1.71 1.7 0.3 4.24
Tangible heritage 0.77 0.03 0.17 0.96 0.65 0.85 0.53 1.6 1.37 1.57 0.32 1.81
Sun and beach 0.79 0.03 0.13 0.95 0.96 1.22 0.21 1.85 2.7 2.42 0.39 2.46
Sports 0.76 0.02 0.1 0.88 0.81 0.6 0.33 0.3 1.11 0.42 1.02 4.19
Technology 0.61 0.03 0.18 0.82 0.89 0.88 0.3 – 0.57 2.53 – –
Non-tourist info 0.7 0.03 0.08 0.81 0.51 0.47 0.79 0.53 0.66 – 2.33 1.18
Leisure 0.65 0.02 0.11 0.78 0.8 0.78 0.13 0.72 1.22 0.88 0.5 0.62
User reactions to destination brand contents in social media
Agenda 0.63 0.02 0.09 0.74 0.72 0.65 0.23 0.76 0.69 0.94 0.65 3.59
Services 0.6 0.02 0.09 0.72 0.18 0.22 0.27 0.47 1.51 – – –
Institutional 0.56 0.02 0.13 0.71 0.75 0.7 0.16 0.43 0.84 0.37 0.48 0.05
Intangible heritage 0.45 0.02 0.13 0.6 0.61 0.16 0.33 0.43 0.74 1.33 0.54 –
Business/trade 0.3 0.02 0.05 0.37 0.13 – 0.12 – 0.41 0.37 – –
305
123
(Outdoorsy and Get-away) and are oriented towards attracting and creating empathy
with their target publics, which are largely families (as seen by the subcategory
Family-oriented).
4.2.2 Brand values that trigger the most reactions
As seen in Fig. 2 and Table 6, the general values that trigger the most average
reactions in proportion to the number of fans of destinations are Successful,
Intelligent, Cheerful, Tough, and Down-to-earth. The subcategory Leader (within
Successful) is especially prominent in terms of reactions (3.96). This specific case is
explained by one destination that won a contest for having the ‘‘Best Sun in the
World.’’ Other prominent subcategories of reactions are Family-oriented (1.11)
(within Down-to-earth) and Happiness (2.09) (within Cheerful), among others.
However, apart from Down-to-earth, most of the brand values that trigger the
most reactions do not coincide with the attributes and values most mentioned by
destinations in their posts. And this is important for tourist destinations, which
should incorporate these values in their communication if they are consistent with
their identity and brand.
Of relevance, Cheerful and Down-to-earth are the values that trigger the most
likes, and Successful and Intelligent are the values that trigger the most comments
relative to the number of Fanpage fans, when they are mentioned.
4.3 Reactions to certain themes/brand values and destination types
• In National Destination and Autonomous Communities (ND ? AC), the theme

that triggers the most average reactions is Sun and beach, which is the fourth
most mentioned theme by this type of destination. Next comes Nature, which is
the most mentioned theme, showing both high reactions from the public and
many mentions by the destination. Conversely, Technology is the third element
1.4
1.19
1.2
1.00
1 0.92
0.82 0.81 0.79
0.76 0.75
0.8 0.70 0.67 0.66 0.65
0.64 0.61
0.6
0.42
0.36
0.4
0.2
0
Fig. 2 Reactions per brand value as a percentage of the number of fans
123
triggering the most reactions, but it is one of the least mentioned themes by
ND ? AC. Following Technology are Sports and Leisure, in terms of reactions.
Remarkably, one of the most mentioned themes in ND ? AC (Tangible
Heritage) is not among the themes with the most reactions. Some brand values
elicit a great number of reactions: Honest, in first place, is followed by
Luxurious and Daring. Notably, the most mentioned value and the value having
the most reactions relative to fans coincide. However, this effect cannot be seen
in the values Luxurious and Daring, which are some of the less mentioned
values in these destination posts.
• Unexpectedly, the case of Large Municipalities (LM) shows that the theme
triggering the most reactions is Sun and beach, which is not one of the most
mentioned. Then, the theme triggering the most reactions in capital cities and
economic centres is Cityscape. Nature and Technology follow, with some of the
least present themes in LM posts. In these cases, some brand values trigger the
highest reactions: Reliable, Tough, Spirited, and Charming. All are related to the
values associated with Strong, Responsible, and Attractive capital cities. In
terms of reactions, Reliable, in first place, is highly dissimilar to the general
ranking in which it is second from the last. This outcome shows that capital
cities need to reinforce these types of values in their communication strategies,
including those related to the environment and social responsibility.
• Concerning Large Littoral Destinations (LLD), the theme that generated the
most reactions, surprisingly, is not related to Sun and beach (the most
mentioned) but is Non-tourist information due to a specific tragic event that led
to high numbers of comments in some posts. Next come Tangible Heritage and
Nature. These themes are very dissimilar to the general rank of reactions and the
most mentioned themes, except for Nature, which is among the first themes.
Notably, Agenda is the second most mentioned category in LLD, but it is one of
the categories triggering fewer reactions. With regard to brand attributes,
Reliable, Down-to-earth, Outdoorsy, and Charming trigger the most reactions.
However, Reliable is one of the least mentioned in the posts. This result shows
that, again, the subject of social and environmental responsibility should be
communicated and be strategic for these destinations, along with their focus on a
Safe environment (Reliable), Outdoor activities, and Family-oriented tourism
(usually their main targets).
• In the case of Littoral Destinations (LD), the four themes triggering the most
reactions (Cityscape, Sun and beach, Nature, and Tangible Heritage) are
relatively coincident with the general ranking and with the most mentioned
themes in LD, except for Cityscape, which is barely mentioned. Significantly,
this finding indicates that the fans of these Fanpages react to very different
themes. These themes generally relate to both the Urban and heritage
environment and the Nature and sea environment thus encouraging littoral
destinations to diversify their Sun and beach image with Nature and Cultural/
urban elements. Concerning brand attributes, Cheerful, Tough, and Cosmopoli-
tan trigger more reactions but are not among the most mentioned brand values.
123
These attributes are especially related to Happiness and International atmo-

sphere in these destinations, aspects that are not emphasized by destinations.
• Regarding Heritage Cities (HC), Cityscape, as expected, is the theme triggering
the most reactions, and Tangible Heritage comes in third (related to the nature of
these towns). Nature is in second place in terms of reactions, which could show
the diverse interests of tourists visiting HC. These destinations also mention
Agenda issues often, which do not trigger many reactions. Interestingly, more
comments are posted by users when the subcategory Shopping appears. In terms
of brand values, Cheerful and Honest trigger the most reactions, which is mostly
coincident with the most mentioned values in HC. Both are related to Sincerity
values (especially to being Traditional, Real, and Friendly places).
• In the case of Medium-sized Cities (MC), Cityscape (clearly related to city
tourism) triggers the most reactions, as can be expected. Technology and Sun
and beach, which do not coincide at all with the most mentioned themes,
follow. Although these destinations mention the Leisure and Tangible Heritage
elements most prominently, these elements do not elicit the greatest number of
reactions. In the case of values, Successful, Intelligent, and Charming trigger
the highest number of reactions and are relatively coincident with the most
mentioned attributes. Therefore, technological advancement, being Successful
and showing Intelligent ideas or elements, seems to trigger important reaction
in these towns.
• Concerning Inland Destinations (ID), the themes triggering the most
reactions are Non-tourist information and Sports. These themes do not
coincide with the most mentioned themes (Leisure and Tangible Heritage), a
result that could be related to this type of city. Remarkably, the Wine tourism
subcategory, within the Intangible Heritage theme, triggers relatively strong
reactions. Intelligent, Cheerful, and Wholesome attributes come first in our
results; they are not coincident with the main attributes or values mentioned
by the destination.
• In the case of Mountain Destinations (MD), the themes that trigger the most
reactions are Nature and Sports, as expected. Nature also happens to be the most
mentioned theme. Then comes Agenda, which also triggers notable reactions.
The value that engenders the greatest number of reactions is Down-to-earth, and
Charming comes third, both results showing that these values are related to the
nature of these destinations and to the type of activities found there. However,
posts related to Cosmopolitan nature also trigger important reactions, which may
be attributed to a low number of posts.
The implications of these results for tourist destinations are that they can have an
approximation of which emotional values generate more reactions, understand why
this is so and direct their communication brand strategies accordingly.
123
Table 5 Brand values mentions

ND ? AC LM LLD LD HC MC ID MD Total
Honest 69 60 61 20 46 6 18 2 282
Imaginative 47 26 44 8 28 2 12 2 169
Outdoorsy 29 17 36 5 23 0 19 6 135
Down-to-earth 21 22 20 14 18 2 18 3 118
Wholesome 25 22 26 5 14 0 9 1 102
Spirited 30 15 13 6 21 2 4 5 96
Cheerful 26 19 18 9 19 0 3 1 95
Charming 28 6 22 4 15 1 1 1 78
Daring 19 10 11 3 18 0 2 3 66
Cosmopolitan 18 8 10 4 15 2 1 1 59
Successful 17 5 12 0 6 3 0 0 43
Intelligent 12 4 13 0 1 2 1 0 33
Luxurious 5 0 18 2 3 0 0 0 28
Reliable 4 3 2 3 3 3 2 0 20
Tough 2 5 2 0 1 0 0 0 10
Up-to-date 4 3 0 1 1 0 0 0 9
5 Conclusions
The themes that generate the most reactions from the public are diverse, but in
general they are highly imageable themes and correspond to the unique or
characteristic attributes of the studied destinations. Generic themes (such as
Agenda) cause fewer reactions by the public. These results can serve as practical
recommendations for DMOs and NTOs that they should always try to increase
interactivity with their publics. Destinations should mainly communicate themes
and attributes that characterize them and distinguish them from other destinations,
and, to a lesser extent, generic ones. If destinations do not communicate these
aspects, they may lead fans to disengage. However, some generic themes and
attributes such as Agenda can also be communicated if destination managers think
that they are necessary for informational purposes, but these attributes should be
emphasised to a lesser degree.
The most mentioned themes in destination posts are not usually the ones leading
to the greatest number of reactions from the public. In general, there is a lack of
congruency between the most mentioned contents and those that trigger the most
reactions, in all destination types. Therefore, tourist destinations should try to
promote a communication of themes that generates a greater number of reactions
among users and thus achieve more interactivity, engagement, navigation time, and
positive feelings towards the destination and its brand (Buhalis and Law 2008;
Munar 2011). Some destination types create more engagement in their posts when
contents identify or are specific to them (e.g., large municipalities with Cityscape or
Mountain destinations with Nature).
123
310
123
Table 6 Average reactions in absolute numbers and as a percentage of fans to brand attributes
% Like/fans % Com/fans % Shar/fans % Reac/fans ND ? AC LM LLD LD HC MC ID MD
Successful 0.59 0.03 0.57 1.19 0.52 0.89 0.31 – 0.39 14.1 – –
Intelligent 0.68 0.03 0.30 1.00 0.69 0.49 0.13 – 0.65 5.07 8.30 –
Cheerful 0.77 0.02 0.12 0.92 0.71 0.78 0.21 1.96 1.10 – 4.69 0.45
Tough 0.69 0.01 0.11 0.82 0.60 1.21 0.04 1.66 – – – –
Down-to-earth 0.70 0.02 0.09 0.81 0.66 0.57 0.47 1.07 0.86 0.93 0.81 4.22
Reliable 0.63 0.03 0.13 0.79 0.74 1.27 1.31 0.47 0.32 0.97 0.40 –
Spirited 0.61 0.03 0.13 0.76 0.27 1.06 0.33 1.43 0.80 0.82 0.87 0.14
Honest 0.63 0.02 0.10 0.75 0.86 0.66 0.32 1.34 1.05 1.68 0.91 0.91
Wholesome 0.58 0.02 0.11 0.70 0.60 0.75 0.35 1.38 0.61 – 1.79 0.45
Charming 0.55 0.02 0.10 0.67 0.69 0.99 0.45 0.20 0.80 2.17 0.68 1.37
Outdoorsy 0.56 0.02 0.08 0.66 0.60 0.82 0.77 0.60 0.61 – 0.73 0.42
Cosmopolitan 0.55 0.02 0.08 0.65 0.57 0.52 0.16 1.50 0.61 1.08 0.54 1.81
Daring 0.56 0.02 0.06 0.64 0.79 0.46 0.32 0.59 0.72 – 0.53 1.28
Imaginative 0.50 0.02 0.09 0.61 0.65 0.64 0.30 0.36 0.65 1.90 1.57 –
Luxurious 0.37 0.01 0.04 0.42 0.79 – 0.22 0.62 – 0.53 – –
Up-to-date 0.31 0.01 0.05 0.36 0.41 0.43 – 0.47 0.41 – – –
A. Huertas, E. Marine-Roig
Concerning brand values, the ones that generate the most reactions also
emphasize the uniqueness of the destinations and are especially related to the
Successful and Intelligent elements of the destinations and to having Cheerful or
Happy experiences. They are highly related to the personality traits for which
specific tourists feel empathy. They are also related to Leisure and Activities and are
in accordance with the public to which they are mostly directed: families.
Remarkably, the public reacts differently to the same values in different destination
types. However, in terms of the transmitted brand values in posts, the different types
of destinations are quite indistinct in their nature and transmit similar values.
Consequently, our results emphasize and corroborate, on the one hand, that the
communication of distinctive emotional values of destinations’ identities generates
a higher number of user reactions. Hence, it is worth communicating such values
through social media; On the other hand, the communication of distinguishing,
experiential, and unique values generates the greatest number of reactions.
Therefore, tourist destinations should take good care in communicating the
emotional values of their brands and especially try to associate unique values with
their territories, values that their target publics can identify with. If all destinations
associate themselves with the same value, social media could lose their utility as
differentiation tools.
Finally, brand values that trigger the greatest number of reactions from users do
not usually coincide with the values most mentioned by destinations in their posts.
Therefore, we detect an important inconsistency. However, with respect to these
results, we do not think that destinations should only communicate the values that
generate the most reactions, but they should also communicate those that identify
their place brand, despite the fact that these values account for the creation of their
brand. Only in this way can the proper communication of identities and place brands
be achieved through social media. Considering place branding as a wider process
contributes to place building (Govers and Go 2009).
This research shows that destinations with fewer fans have greater user
engagement (reactions) as a percentage of the number of fans, which should
encourage researchers and DMOs to search for the best strategies to maintain user
engagement in social media while growing in popularity at the same time.
Concerning the analysis method, this research has contributed with the proposal
of an analysis template including both attraction factors and emotional values and
the measurement of reactions relative to the total number of fans, which have been
useful to achieve the study’s initial objectives and can be helpful for tourist
destination managers of social media communication as they provide information
about the themes and emotional values that trigger the greatest number of reactions,
and they can be adapted to different types of destinations.
Now, apart from the utility this study can have for social media managers, its
main contribution has been to illustrate that the communication of emotional values
and destination-unique elements also generates user reactions and interactivity.
Research has previously shown that greater interactivity leads to a better brand
image for tourist destinations (Munar 2011). But this study’s results also detect an
inverse relationship: the communication of emotional brand values and brand
identity elements engender more reactions and greater interactivity. Thus, we
123
conclude that the communication of both tangible/functional elements and,

especially, emotional values is recommended and even necessary for effective
communication about tourist destinations through social media.
This study also has some limitations. Firstly, the methodology is of an
exploratory nature and the study was conducted for a specific and not very extended
period of time. Therefore, future studies should be conducted on a yearly basis and
for other specific periods of time. Secondly, we have observed that a specific event
can easily distort results and hence, a study of the events in each destination should
be conducted to assess results. Thirdly, the case sample of the study corresponds to a
specific country. This study should thus be expanded to a larger number of
destinations to replicate the results in other countries. Another limitation is that this
study does not assess whether the percentages of the reactions per fans are low or
high in absolute terms. Therefore, future studies should assess what the expected
threshold of user reactions to destination social media posts is to assess if content
communication engagement strategies are successful or not.
We believe that future research in this area should assess whether or not specific
brand communication, with its attraction factors and emotional values, generates
more reactions among users, that is, whether there is a positive relationship between
the proper communication of these attributes and unique/real attributes of places and
the interactivity they generate.
Acknowledgments This work was supported by Spain’s Ministry of Economy and Competitiveness
(Grant ID: CSO2012-34824 ‘‘Uso e influencia de los social media y la comunicación 2.0 en la toma de
decisiones turı́sticas y en la imagen de marca de los destinos’’).
References
Aaker JL (1997) Dimensions of brand personality. J Mark Res 34:347–356
Agarwal S, Mondal A, Nath A (2011) Social media—the new corporate playground. Int J Res Rev
Comput Sci 2:696–700
Algesheimer R, Dholakia UM, Herrmann A (2005) The social influence of brand community: evidence
from European car clubs. J Mark 69:19–34. doi:10.1509/jmkg.69.3.19.66363
Arlt WG (2006) Not very willkommen: the Internet as a marketing tool for attracting German-speaking
tourists to non-European destinations. Inf Technol Tour 8:227–238. doi:10.3727/
109830506778690803
Baloglu S, McCleary K (1999) A model of destination image formation. Ann Tour Res 26:868–897.
doi:10.1016/S0160-7383(99)00030-4
Beerli A, Martin JD (2004) Factors influencing destination image. Ann Tour Res 31:657–681. doi:10.
1016/j.annals.2004.01.010
Bigné E, Sánchez I, Sanz S (2009) The functional-psychological continuum in the cognitive image of a
destination. A confirmatory analysis. Tour Manag 30:715–723. doi:10.1016/j.tourman.2008.10.020
Buhalis D, Law R (2008) Progress in information technology and tourism management: 20 years on and
10 years after the Internet—the state of eTourism research. Tour Manag 29:609–623. doi:10.1016/j.
tourman.2008.01.005
Choi S, Lehto XY, Morrison AM (2007) Destination image representation on the web: content analysis of
Macau travel related websites. Tour Manag 28:118–129. doi:10.1016/j.tourman.2006.03.002
De Moya M, Jain R (2013) When tourists are your ‘‘friends’’: exploring the brand personality of Mexico
and Brazil on Facebook. Public Relat Rev 39:23–29. doi:10.1016/j.pubrev.2012.09.004
Del Chiappa G (2011) Trustworthiness of Travel 2.0 applications and their influence on tourist behaviour:
an empirical investigation in Italy. In: Law R, Fuchs M, Ricci F (eds) Information and
communication technologies in tourism 2011. Springer, New York, pp 331–342
123
Echtner CM, Ritchie JRB (1991) The meaning and measurement of destination image. J Tour Studies
2:2–12
FEMP, TourSpain (eds) (2008) Modelos de gestión turı́stica local: Principios y prácticas. Federación
Española de Municipios y Provincias, Secretarı́a General de Turismo, Madrid
Fernandez-Cavia J, Rovira C, Diaz-Luque P, Cavaller V (2014) Web quality index (WQI) for official
tourist destination websites: proposal for an assessment system. Tour Manag Perspect 9:5–13.
doi:10.1016/j.tmp.2013.10.003
Fischer E, Reuber AR (2011) Social interaction via new social media: (How) can interactions on Twitter
affect effectual thinking and behavior? J Bus Ventur 26:1–18. doi:10.1016/j.jbusvent.2010.09.002
Fotis J, Buhalis D, Rossides N (2012) Social media use and impact during the holiday travel planning
process. In: Fuchs M, Ricci F, Cantoni L (eds) Information and communication technologies in
tourism 2012. Springer, Vienna, pp 13–24. doi:10.1007/978-3-7091-1142-0_2
Govers R, Go FM (2003) Deconstructing destination image in the information age. Inf Technol Tour
6:13–29. doi:10.3727/109830503108751199
Govers R, Go FM (2009) Place branding: Glocal, virtual and physical identities, constructed, imagined
and experienced. Palgrave Macmillan, UK
Govers R, Go FM, Kumar K (2007) Promoting tourism destination image. J Travel Res 46:15–23. doi:10.
1177/0047287507302374
Hays S, Page SJ, Buhalis D (2013) Social media as a destination marketing tool: its use by national
tourism organisations. Curr Issues Tour 16:211–239. doi:10.1080/13683500.2012.662215
Hosany S, Ekinci Y, Uysal M (2006) Destination image and destination personality: an application of
branding theories to tourism places. J Bus Res 59:638–642. doi:10.1016/j.jbusres.2006.01.001
Huertas A (2014) La comunicación de los territorios, los destinos y sus marcas. Guı́a práctica de
aplicación desde las relaciones públicas. DIRCOM-UOC, Barcelona
Huertas A, Fernández Cavia J (2006) Ciudades en la web: usabilidad e interactividad de las páginas
oficiales de los destinos turı́sticos. In: TURITEC 2006. Universidad de Málaga, Spain, pp 403–422
Huertas A, Marine-Roig E (2014) Uso y utilidades de las herramientas de análisis online para la
evaluación de la comunicación de las marcas de destino a través de los social media (Use and
utilities of online analysis tools for the communication evaluation of destination tourist brands
through social media). Sphera Publica, pp 117–134
Inversini A, Buhalis D (2009) Information convergence in the long tail: The case of tourism destination
information. In: Hopken W, Gretzel U, Law R (eds) Information and communication technologies in
tourism 2009. Springer, Vienna, pp 381–392
Jacobsen JKS, Munar AM (2012) Tourist information search and destination choice in a digital age. Tour
Manag Perspect 1:39–47. doi:10.1016/j.tmp.2011.12.005
Jang HJ, Sim J, Lee Y, Kwon O (2013) Deep sentiment analysis: mining the causality between
personality-value-attitude for analyzing business ads in social media. Expert Syst Appl
40:7492–7503. doi:10.1016/j.eswa.2013.06.069
Jenkins OH (1999) Understanding and measuring tourist destination images. Int J Tour Res 1:1–15
Jung TH, Butler R (2000) Perceptions of marketing managers of the effectiveness of the Internet in
tourism and hospitality. Inf Technol Tour 3:167–176
Kapferer JN (1997) Strategic brand management. Kogan Page, London
Kaplan AM, Haenlein M (2010) Users of the world, unite! The challenges and opportunities of social
media. Bus Horiz 53:59–68. doi:10.1016/j.bushor.2009.09.003
Keller K (2003) Strategic brand management: building, measuring, and managing brand equity, 2nd edn.
Prentice-Hall, Upper Saddle River
Kim H, Stepchenkova S (2015) Effect of tourist photographs on attitudes towards destination: manifest
and latent content. Tour Manag 49:29–41. doi:10.1016/j.tourman.2015.02.004
Ko H, Cho CH, Roberts MS (2005) Internet uses and gratifications. A structural equation model of
interactive advertising. J Advert 34:57–70. doi:10.1080/00913367.2005.10639191
Koltringer C, Dickinger A (2015) Analyzing destination branding and image from online sources: a web
content mining approach. J Bus Res 68:1836–1843. doi:10.1016/j.jbusres.2015.01.011
Laroche M, Habibi MR, Richard MO (2013) To be or not to be in social media: how brand loyalty is
affected by social media? Int J Inf Manag 33:76–82. doi:10.1016/j.ijinfomgt.2012.07.003
Lawson F, Band-Bovy M (1977) Tourism and recreational development. Architectural Press, London
Leung D, Law R, van Hoof H, Buhalis D (2013) Social media in tourism and hospitality: a literature
review. J Travel Tour Mark 30:3–22. doi:10.1080/10548408.2013.750919
123
Li X, Stepchenkova S (2012) Chinese outbound tourists’ destination image of America. J Travel Res
51:250–266. doi:10.1177/0047287511410349
Litvin SW, Goldsmith RE, Pan B (2008) Electronic word-of-mouth in hospitality and tourism
management. Tour Manag 29:458–468. doi:10.1016/j.tourman.2007.05.011
Lovejoy K, Saxton GD (2012) Information, community, and action: how nonprofit organizations use
social media. J Comput Mediat Commun 17:337–353. doi:10.1111/j.1083-6101.2012.01576.x
Mandelli A, Cantoni L (2010) Social media impact on corporate reputation: proposing a new
methodological approach. Cuadernos de Informacion 27:61–74
Marchiori E, Cantoni L (2012) The online reputation construct: does it matter for the tourism domain? A
literature review on destinations’ online reputation. Inf Technol Tourism 13:139–159. doi:10.3727/
109830512X13283928066715
Marine-Roig E (2013) From the projected to the transmitted image: the 2.0 construction of tourist
destination image and identity in Catalonia, PhD dissertation. http://hdl.handle.net/10803/135006
(accessed 31 May 2015)
Marine-Roig E (2014) A webometric analysis of travel blogs and reviews hosting: the case of Catalonia.
J Travel Tour Mark 31:381–396. doi:10.1080/10548408.2013.877413
Marine-Roig E (2015) Identity and authenticity in destination image construction. Anatolia 26:574–587.
doi:10.1080/13032917.2015.1040814
Michaelidou N, Siamagka NT, Moraes C, Micevski M (2013) Do marketers use visual representations of
destinations that tourists value? Comparing visitors image of a destination with marketer-controlled
images online. J Travel Res 52:789–804. doi:10.1177/0047287513481272
Miralbell O, Alzua A, Kepa J (2014) Content curation and narrative tourism marketing. In: Xiang Z,
Tussyadiah I (eds) Information and communication technologies in tourism 2014. Springer, Austria,
pp 187–199. doi:10.1007/978-3-319-03973-2_14
Morgan NJ, Pritchard A, Piggott R (2003) Destination branding and the role of the stakeholders: the case
of New Zealand. J Vacat Mark 9:285–299. doi:10.1177/135676670300900307
Munar AM (2011) Tourist-created content: rethinking destination branding. Int J Cult Tour Hosp Res
5:291–305. doi:10.1108/17506181111156989
Munar AM, Jacobsen JKS (2014) Motivations for sharing tourism experiences through social media. Tour
Manag 43:46–54. doi:10.1016/j.tourman.2014.01.012
Niemann-Struweg I (2014) An integrated communication implementation model for the post-2000
business environment. Public Relat Rev 40:184–192. doi:10.1016/j.pubrev.2013.08.011
Nusair K, Bilgihan A, Okumus F, Cobanoglu C (2013) Generation Y travelers’ commitment to online
network websites. Tour Manag 35:13–22. doi:10.1016/j.tourman.2012.05.005
Opoku RA (2006) Towards a methodological design for evaluating online brand positioning, Doctoral
dissertation, p 61. http://pure.ltu.se/portal/en/ (accessed 31 May 2015)
Pan B, Li X (2011) The long tail of destination image and online marketing. Ann Tourism Res
38:132–152. doi:10.1016/j.annals.2010.06.004
Pike S (2002) Destination image analysis—a review of 142 papers from 1973 to 2000. Tour Manag
23:541–549. doi:10.1016/S0261-5177(02)00005-5
Qu H, Kim LH, Im HH (2011) A model of destination branding: integrating the concepts of the branding
and destination image. Tour Manag 32:465–476. doi:10.1016/j.tourman.2010.03.014
Sabate F, Berbegal-Mirabent J, Cañabate A, Lebherz PR (2014) Factors influencing popularity of branded
content in Facebook fan pages. Eur Manag J 32:1001–1011. doi:10.1016/j.emj.2014.05.001
Saffer AJ, Sommerfeldt EJ, Taylor M (2013) The effects of organizational Twitter interactivity on
organization-public relationships. Public Relat Rev 39:213–215. doi:10.1016/j.pubrev.2013.02.005
Schmallegger D, Carson D (2010) Destination image projection on consumer generated content websites
(CGC): a case study of the Flinders Ranges. Inf Technol Tour 11:111–127. doi:10.3727/
109830509789994838
Sicilia M, Ruiz S, Munuera JL (2005) Effects of interactivity in a web site: the moderating effect of need
for cognition. J Advert 34:31–44. doi:10.1080/00913367.2005.10639202
Stepchenkova S, Zhan F (2013) Visual destination images of Peru: comparative content analysis of DMO
and user-generated photography. Tour Manag 36:590–601. doi:10.1016/j.tourman.2012.08.006
Tasci ADA, Gartner WC, Cavusgil ST (2007) Conceptualization and operationalization of destination
image. J Hosp Tour Res 31:194–223. doi:10.1177/1096348006297290
Tsimonis G, Dimitriadis S (2014) Brand strategies in social media. Mark Intell Plan 32:328–344. doi:10.
1108/MIP-04-2013-0056
123
Tussyadiah IP, Fesenmaier DR (2009) Mediating tourist experiences: access to places via shared videos.
Ann Tourism Res 36:24–40. doi:10.1016/j.annals.2008.10.001
Valentini C (2015) Is using social media ‘‘good’’ for the public relations profession? A critical reflection.
Public Relat Rev 41:170–177. doi:10.1016/j.pubrev.2014.11.009
Walther JB, Jang J (2012) Communication processes in participatory websites. J Comput Mediat
Commun 18:2–15. doi:10.1111/j.1083-6101.2012.01592.x
Wenger A (2008) Analysis of travel bloggers’ characteristics and their communication about Austria as a
tourism destination. J Vacat Mark 14:169–176. doi:10.1177/1356766707087525
Xiang Z, Gretzel U (2010) Role of social media in online travel information search. Tour Manag
31:179–188. doi:10.1016/j.tourman.2009.02.016
Xiang Z, Wober K, Fesenmaier DR (2008) Representation of the online tourism domain in search
engines. J Travel Res 47:137–150. doi:10.1177/0047287508321193
Yoo KH, Gretzel U (2011) Antecedents and impacts of trust in travel-related consumer-generated media.
Inf Technol Tour 12:139–152. doi:10.3727/109830510X12887971002701
Zhang L, Pan B, Smith W, Li X (2010) An exploratory study of travelers’ use of online reviews and
recommendations. Inf Technol Tour 11:157–167. doi:10.3727/109830509789994775
123
DOI 10.1007/s40558-015-0040-1
ORIGINAL RESEARCH
A detailed method for destination image analysis using

user-generated content
Estela Marine-Roig1 • Salvador Anton Clavé2
Received: 12 June 2015 / Revised: 16 October 2015 / Accepted: 18 November 2015 /

Published online: 8 December 2015
Springer-Verlag Berlin Heidelberg 2015
Abstract Social media and user-generated content (UGC) have revolutionized

tourism and hospitality communication and are seen as rich sources of information
for destination image analysis. Many articles have been published about travel-
related UGC, in particular, quantitative and qualitative content analysis of travel
blogs and online travel reviews (OTR). Researchers have typically analysed small
samples of population-representative travel diaries (tens or hundreds of files), which
allow for manual processing. However, the enormous growth of OTRs requires
operationalization through computerized methods, and the aim of this article is to
propose a detailed method for semi-automatic downloading, arrangement, cleaning,
debugging, and analysis of large-scale travel blogs and OTR data. This enables the
classification of collected webpages by dates and destinations and offline content
analysis of the text as written by the tourist. More than 130,000 useful trip diaries of
tourists who visited Catalonia between 2004 and 2014 have been gathered, and
significant results have been obtained in terms of content analysis in relation to
destination image.
This article is an extended and updated version of a conference paper previously published in ENTER
Proceedings of the International Conference in Lugano, Switzerland, February 3–6, 2015.
& Estela Marine-Roig

estela.marine@aegern.udl.cat
Salvador Anton Clavé
salvador.anton@urv.cat
1
Department of Business Administration and Economic Management of Natural Resources
(AEGERN), University of Lleida, Catalonia, Spain
2
Research Group on Territorial Analysis and Tourism Studies (GRATET), Rovira i Virgili
University, Catalonia, Spain
123
342 E. Marine-Roig, S. A. Clavé
Keywords Travel blog Online travel review Web harvesting Web data
mining Massive content analysis Catalonia
1 Introduction
Destination image is a complex construct, and is usually considered to be the sum of

beliefs, ideas and impressions that people have of a place (Kotler et al. 1993). It is a
complex construct resulting from both the projected and the perceived images of a
destination (Marine-Roig 2015a) and is said to be composed of two primary
components, the cognitive component (mental responses to the stimuli in the
environment, related to physical attributes) and the affective components (how one
feels about this knowledge). A third component, the conative (related to behaviour),
was later introduced by several researchers (Marchiori and Onder 2015; Serna et al.
2015).
In recent years the image of tourism destinations has depended on multiple
sources of information and content generated by travellers, suppliers and residents,
especially online social media (Anton Clavé and Gonzalez 2008; Llodra-Riera et al.
2015; Munar and Jacobsen 2013). The rise of social media has facilitated the
exponential growth of user-generated content (UGC). Today online UGC strongly
and rapidly influences the formation of destination images, since on the one hand,
tourists increasingly and actively post the recounts of their personal experiences
online, in the form of both textual and visual contributions, and, on the other hand,
they increasingly rely on UGC to reduce their uncertainty when making decisions
concerning travel and tourism (Amaral et al. 2014; Llodra-Riera et al. 2015;
Marchiori and Onder 2015).
Due to the influence of UGC sources in destination image formation, it is
important for tourist destinations and businesses to assess the extent to which this
content contributes to the creation of and changes in destination image (Llodra-
Riera et al. 2015; Marchiori and Onder 2015) and in promoting single operators and
the destination as a whole (Albastroiu and Felea 2014). Moreover, UGC data in
tourism is considered a good source of information for national tourism
organizations (NTOs) and other policymakers, destination marketing organizations
(DMOs) and other stakeholders, as well as for future travellers, because it consists
of opinions freely expressed by tourists who have visited the destination. UGC also
offers huge possibilities for e-commerce, business intelligence, marketing, and
social studies; a growing number of commercial firms contribute to hosting,
developing, distributing, rating, and mining UGC (Moens et al. 2014); and
increasingly tourism firms exploit UGC for improving services that analyse
travellers’ post-trip experiences (Sigala et al. 2012).
According to Standing et al. (2014), Travel 2.0 approaches have spawned large
amounts of UGC in the form of travel blogs and online travel reviews (OTR). Travel
blogs are a cost-effective means to gather rich, authentic, and unsolicited visitor
feedback (Pan et al. 2007). Travel blogs and OTRs allow for experiencing firsthand
events that travellers narrate during their stay at the destination as well as viewing
their photos and video uploads. Moreover, review sites such as TripAdvisor were
123
A detailed method for destination image analysis using… 343
found to have the highest level of trust by online users among several online social
media sources (Munar and Jacobsen 2013); however, because of the ease of
publishing fake reviews, concerns have arisen (Minazzi 2015; Schuckert et al.
2015b).
Regarding UGC as a research source in tourism, Lu and Stepchenkova (2015)
state that, in general, the methods used to collect data are unclear and mostly rely on
manually selecting and gathering information. In addition to the articles surveyed by
these authors, other recent examples, which have not fully developed their
methodologies include Wang et al. (2013), Schmunk et al. (2014), and Guo et al.
(2015) who mention an ad hoc web crawler to collect data, without further details;
Koltringer and Dickinger (2015) indicate a commercial web crawler (WebLyzard.-
com); whereas Fang et al. (2016) do not show how they collected their data.
To fill this gap in UGC-related research, this article aims to propose a detailed
method for semi-automatic downloading, arrangement, cleaning, debugging, and
analysis of large-scale travel blog and OTR data. The methodology is applied in the
case of Catalonia to analyse more than 130,000 useful travel diaries written by
foreign tourists who visited Catalonia between 2004 and 2014.
2 State of the art
Social media is characterized by using ever-changing tools; new platforms are

continuously created, and many destinations have difficulties achieving a long-term
strategy, requiring innovative strategies and methodologies of analysis (Munar
2012). Dwyer et al. (2014) classified research methods in tourism data analysis into
quantitative, qualitative and mixed approaches, but it must be considered that in the
last two decades the application of quantitative techniques in both academic and
non-academic (NTOs, DMOs, and other stakeholders) research has intensified. The
most popular research methods for the analysis of travel blogs have been content
analysis, both qualitative and quantitative, and narrative analysis (Banyai and
Glover 2012); and with respect to OTRs, Schuckert et al. (2015a) found that most
studies are based on quantitative content analysis.
The topic areas of the studies on tourism and hospitality are very diverse. Zeng
and Gerritsen (2014) analysed 165 articles, published between 2007 and 2013, and
found that the most frequently occurring keywords were ‘‘marketing,’’ followed by
‘‘customer/consumer behaviour,’’ ‘‘UGC,’’ ‘‘information search,’’ ‘‘destination
management,’’ and ‘‘(electronic)word-of-mouth’’ (WOM and eWOM). Gursoy
et al. (2015) presented over 200 measurement scales gleaned from studies published
in selected top journals in the field, between 1992 and 2013, and grouped them into
seven main categories: motivation; residents’ perceptions and attitudes; destination
image; performance, evaluation, quality assessment, loyalty, and satisfaction; tourist
behaviour; human resources; and hospitality and tourism operations. Lu and
Stepchenkova (2015) surveyed 122 articles, published between 2001 and 2013, and
classified them, in order of frequency, into service quality, destination image and
reputation, UGC such as eWOM, experiences and behaviour, and mobility patterns.
Schuckert et al. (2015a) analysed 50 articles related to online reviews, published in
123
academic journals between 2004 and 2013, and grouped them into five topical
clusters: online buying, satisfaction and management, opinion mining/sentiment
analysis, motivation, and the role of reviewers.
Such user (consumer)-generated content data have grown exponentially in recent
years, especially in the case of hospitality OTRs. For instance, in May 2015,
TripAdvisor asserted that it had reached more than 225 million reviews and
opinions, Booking almost 49 million verified reviews from real guests, and Trivago
indicated that it had received 140 million integrated user hotel reviews. It is now
considered that their processing requires the use of Big Data technologies
(Krawczyk and Xiang 2015). However, Lu and Stepchenkova (2015), in an
exhaustive work about UGC as a research mode, have proven that, in most studies,
UGC data were collected manually, limiting the sample size. Small samples can
hardly represent the population, and Banyai and Glover (2012) contend that usually
samples are not selected by chance, which also questions their representativeness.
Moreover, the collection of UGC data via manual copying is extremely time-
consuming (Johnson et al. 2012; Lu and Stepchenkova 2015). As examples of
relatively small sample size (tens or hundreds of entries), Wang and Morais (2014)
examined the tourists’ identity in 69 travel blogs from 16 websites utilising critical
discourse analysis; these weblogs were selected after a preliminary reading of all
blogs on the surveyed destinations, located through the Google search engine;
Amaral et al. (2014) unveiled tourists’ profiles and preferences in 813 consumer-
generated reviews on 20 restaurants from TripAdvisor.com; Kladou and Mavragani
(2015) assessed the cognitive, affective and conative components of the image of
Istanbul, through 203 reviews posted on TripAdvisor; and Lai and To (2015) using
Leximancer analysed 68 keywords from 440 entries collected in the same websites
studied in this paper to show the destination image of Macao.
Many studies focus on user (consumer)-generated content on specialized
websites. For instance, commissioned by the Catalan NTO, Gonzalez (2010)
analysed a total of 28 sources of online information, in different formats, selected on
the basis of the criteria of importance of the webpage inside the parameters of Web
2.0, presence of UGC and of a specific section for Catalonia, to unveil the image of
the Catalan tourism brands on the Internet. Schmunk et al. (2014) collected 1441
reviews about hotels from TripAdvisor.com and Booking.com to extract decision-
relevant knowledge from UGC. To do so, these authors proposed a sentiment
analysis process. Johnson et al. (2012), using a web harvesting programme, claimed
to automatize the collection of 5730 OTRs from TripAdvisor.com about Nova
Scotia, but recognized that they had to manually eliminate reviews of many other
destinations because they first identified the initial pages by searching ‘‘Nova
Scotia’’ in TripAdvisor.com and all the destinations that residents in this Canadian
province had visited also appeared. These authors performed a basic classification of
OTRs (attractions, hotels, and restaurants) and grouped them according to
destinations. Koltringer and Dickinger (2015) collected 5719 relevant documents
from online sources such as TravelBlogs.com, by the WebLyzard crawling agent,
and extracted destination brand identity and image through web content mining and
natural language processing, including keyword analysis and automatic sentiment
detection. Li et al. (2015) gathered 1033 verified travel blog articles from
123
CTrip.com to analyse the destination image of Taiwan. As examples of massive

downloads using ad hoc crawling or scraping software, Serna et al. (2015) examined
4123 OTRs from MiNube.com to analyse the cognitive-affective and conative
destination image of a Spanish region applying categorization techniques; Wang
et al. (2013) with 834,304 OTRs from TripAdvisor.com computed two impact
indexes to measure reviewer credibility based on the number of reviews and
destinations on which a reviewer had posted reviews, and the number of helpful
votes received by the reviews; and Xiang et al. (2015) investigated 60,648 online
customer reviews from Expedia.com corresponding to 10,537 hotels from the 100
largest USA cities to analyse guest experiences.
Researchers have suggested various methods to extract knowledge from online
sources. Abburu and Babu (2013) proposed a framework for web data extraction
and analysis based on three basic steps: finding URLs (uniform resource locator) of
webpages, extracting information from webpages, and data analysis. This suggested
system architecture is divided into three modules: web crawling, information
extraction, and mining. Chau and Xu (2012) proposed a framework for collecting
and analysing business intelligence in blogs on a topic of interest divided into four
steps: identify the explicit communities (blogring URLs), collect information about
bloggers (blogring pages and blogger pages), analyse the content posted (blogring
information, blogger information, and blog entries: term and opinion analysis),
analyse the interaction networks and implicit communities formed by the bloggers.
To mine valuable information from the structured tourism blogs, Guo et al. (2015)
recommended a framework divided into blog extraction and data pre-processing,
mining of frequent departure cities, mining of frequent travel spots, and spot
associated service detection. Schmunk et al. (2014) proposed a five-stage process,
which consists of selecting and collecting review pages, document processing
(information extraction, removing reviews with no text, filtering English texts, and
generating sentences), mining, evaluation, and usage. Lai and To (2015) suggested a
four-phased methodology: definition of goal and scope, data collection (identifying
online information sources, determining the sample size, and downloading text
files), data transformation (creating a master file, using WordSmith computer-aided
lexical software to identify keywords and their frequency counts, and using
Leximancer computer-aided lexical concept mapping software to project a holistic
view of the study) and the interpretation of results. Koltringer and Dickinger (2015)
apply a method, which includes data gathering, keyword analysis, sentiment
detection, category building, and correspondence analysis. However, they converted
data into a machine-readable format and only plain text remains for analyses. Li
et al. (2015) utilised two approaches for analysing traveller-generated content: text
mining (coding and classifying blog entries, string matching processing, and
translation from simplified Chinese to English) and content analysis (keyword
cleaning, keywords analysis, correspondence analysis, and affective analysis).
According to Lu and Stepchenkova (2015), in general, the studies on UGC as a
research source in tourism and hospitality are vague with respect to how they collect
data. They further indicate that the technical details of data analyses are often
incomplete and that descriptions of the software functionalities are limited (p 142).
That is why the method proposed in the next section presents a series of
123
contributions to research on the collection and analysis of massive UGC data that
are not recorded or detailed in previous works. Some of these contributions are:
• Detailed methods Examples of significant problems that can arise and how to
solve them; detailed explained are given.
• Suitable websites The selection of hosting websites most suitable for the case
study is warranted.
• Web mining Web structure mining and the process of downloading webpages are
illustrated.
• Dataset arrangement The organization of downloaded trip diaries allowing for
multiple classifications of tens of thousands of entries via operating system
utilities.
• Data quality Analysed data is reduced to what has been written and posted by
the user (blogger or reviewer) saving anomalies (character encoding problems,
most common mistakes, and translations) of proper names (especially,
destination and attraction names).
• Weight of keywords Besides the frequency analysis, the proposed method
calculates the weight of keywords and key phrases based on their emphasis
(impact within the webpage).
3 Methodology
This section presents the analysis methodology that is proposed, developed and
described in the following subsections, consisting of the steps for destination and
hosting websites selection, data downloading, data arrangement, data cleaning, data
debugging and content analysis.
The main feature of the method we propose is in the web data extraction phase,
because instead of simply extracting the information, we add the cleaning and
debugging phases to eliminate the noise present in the webpage to be able to reach
the content analysis phase with quality information in the original HTML
(HyperText Markup Language) format (see Fig. 5). That is to say, the resulting
webpages only contain what the users wrote and preserve their semi-structured
format to assess the HTML emphasis of keywords and key phrases and thus their
potential impact on the Internet.
To demonstrate the effectiveness of the method to manage large-scale data
placed in time and space, first a destination with a large tourist inflow and territorial
division was selected. Consecutively, the most suitable websites for hosting travel
blogs and OTRs for the case study were selected. To illustrate this method in detail,
we used an example, which does not contain any personal information, of a foreign
tourist who visited the selected destination and wrote one travel blog, one
travelogue and three travel reviews on one of the selected webhosts. Finally, the
different phases (data downloading, data arrangement, data cleaning and data
debugging) are described until we reach content analysis. The results section sets
out the outcome of the application of the methodology to the case study.
123
3.1 Destination selection
Although this method can be applied to any destination, more accurate results are
obtained when working with large-scale data. To classify entries, the destination
needs to be divided territorially, as for example a continent into countries, or a state
into regions or provinces. But it could also be applied directly to a destination city
like Barcelona. Catalonia has been selected because it fulfils the following terms.
Catalonia is a Mediterranean destination with a millenary history, its own culture
and language, and a rich historical and natural heritage. Catalonia offers many
attractions for all sorts of visitors: culture, relaxation, nature, family-friendly
facilities, sports, business, etc. Its great capacities and excellent facilities place it
among Europe’s prime tourist areas (CTB 2015). Catalonia is the third European
region in number of overnight stays (Eurostat 2015). In 2014, it welcomed more
than 20 million tourists, more than three-quarters of whom came from abroad
(Table 1).
The Catalan territory is made up of nine tourist brands gathered under and
promoted by tourist boards, which facilitates the study of delimited spaces with a
relatively homogeneous tourist offering (Fig. 1). Catalonia is not an Anglophone
region, and therefore, the problems related to character codification beyond ASCII
127 (American Standard Code for Information Interchange) should be considered,
specifically, those related to existing accent marks in destination and tourist
attraction factor names.
3.2 Hosting websites selection
By means of a popular specialized search engine (BlogSearch.Google.com: ‘‘travel

blog’’ OR ‘‘travel review’’), on 2015-06-01, 14,700,000 indexed pages were initially
obtained. The problem is that blogs come from diverse sources and websites, and do
not have homogeneous structures, which makes it impossible to automatize the
process of downloading, classification and refinement, as intended in this study.
Therefore, based on previous works, a group of websites hosting travel blogs and
reviews (OTRs) with at least 100 entries about Catalonia during the studied period
were selected. We also verify that the entries have a creation or modification date
and that we can deduce the destination to which they refer. Entries focusing on
hotels (Booking.com, Expedia.com, etc.) and restaurants were discarded because of
their great volume and specialization. Finally, eleven websites remained: GetJeal-
ous.com (GJ), MyTripJournal.com (MT), StaTravel.com (ST), TravBuddy.com
(TY), TravelBlog.org (TB), TravelJournals.net (TJ), TravellersPoint.com (TS),
TravelPod.com (TP), TripAdvisor.com (TA), Venere.com (VN), and Virtu-
alTourist.com (VT).
Most authors consider that the hyperlink-based Google’s PageRank (PR)
algorithm is adequate to classify websites according to their levels of ‘‘prestige’’
or ‘‘authority’’ (Baggio and Corigliano 2009; Liu 2011; Ying et al. 2014) and
researchers often use it to select the webpages more suitable for a case study (Law
and Cheung 2010; Pan et al. 2007). However the PR is insufficient to arrange the
webs for its low granularity and because it does not take into account the existing
123
348
123
Table 1 Foreign tourists (thousands) with Catalonia as principal destination
Year Other
be ca de fr it jp nl ru uk us
2009 441 88 1052 3773 1110 118 637 216 1644 425 3201
2010 423 81 1018 3920 1175 114 652 344 1280 485 3663
2011 491 103 1010 3571 1255 111 868 496 1284 457 3497
2012 515 150 1162 3816 1168 162 691 741 1606 577 3852
2013 562 130 1280 4172 1117 207 695 979 1687 569 4234
2014 593 117 1430 4604 1346 273 815 834 1782 513 4507
Source Trends in the main tourism magnitudes (Catalan Tourism Observatory)

E. Marine-Roig, S. A. Clavé
Tourist brand Abbr.
Barcelona Barna
Costa Barcelona cBarc
Costa Brava cBrav
Costa Daurada cDaur
Paisatges Barcelona pBarc
Pirineus Pyren
Terres de l’Ebre tEbre
Terres de Lleida tLlei
Val d’Aran vAran
(unclassified) unCla
Fig. 1 Tourist brands of Catalonia (CTB 2015)
Table 2 Webometrics of the top four websites hosting travel diaries (2015-05-31)
TA TB TP VT
Indexed pages Google.com 144,000,000 453,000 334,000 554,000

Bing.com 37,900,000 157,000 157,000 1050,000
Link-based rank Google PR 7 6 6 7
Yandex CY 1800 80 325 350
Visit-based rank Compete.com 53 39,033 21,669 2162
Quantcast.com 161 25,044 11,663 2798
Alexa.com 192 44,789 30,595 5050
Size Entries 119,016 3148 2375 7963
TBRH Rank 1 3 4 2
traffic (visitors, visits, etc.) on the website. As seen in Table 2, TA and VT have the
same PR, while the other metrics point to a large difference between the two
webhosts. Moreover, it is crucial to consider the volume of data that contains the
website in relation with the period, region and/or topic studied.
Therefore, from the above websites, a ranking was built by applying the weighted
formula ‘‘TBRH = 1 9 B(V) ? 1 9 B(P) ? 2 9 B(S)’’ (Marine-Roig 2014), where
‘B’ corresponds to Borda’s ordering method; ‘V’ to the visibility of the website
(quantity and quality of inbound links); ‘P’, its popularity (received visits and traffic
in general); and ‘S’, the size (number of entries related to the case study). Next the
first four in the ranking were selected (Table 2).
Previously, partial rankings are calculated with the same method:
• Visibility (V) = 1 9 B(Google) ? 1 9 B(Bing) ? 1 9 B(PR) ? 1 9 (CY). This

is composed of four metrics without weighting: The indexed pages of the two
123
search engines (Google and Bing) of major traffic in the West (Alexa.com: top
sites on the web) and two well-known rankings based on the quantity and quality
of incoming links, Google PageRank and Yandex citation index rank.
• Popularity (P) = 1 9 B(Compete) ? 1 9 B(Quantcast) ? 2 9 B(Alexa). Alexa
has been weighted because its rank is based on world traffic, while the other two
are limited to the USA (Marine-Roig 2014), but the sum of the two first ranks is
equivalent to Alexa considering that much of the audience comes from the USA
(Table 3).
• Size (S) As stated previously, reviews concerning hotels and restaurants were not
counted, nor were empty reviews (Schmunk et al. 2014) about Catalonia, with
title, date and author, but without content. For example, in the case of TA, more
than 100,000 attraction reviews were empty and were removed. Also, more than
250,000 hotel reviews and more than 300,000 restaurant reviews were discarded.
Finally, 132,502 relevant entries about Catalonia remained for subsequent
processing. In Table 4, strong growth by TA, but a decline in TB and VT, can be
observed. Notably, TP and VT were acquired a few years ago by TripAdvisor,
Inc.
The four selected websites are best suited to the case study according to Marine-
Roig (2014). These hosting websites (TA, TB, TP, and VT) coincide with those
selected by Lai and To (2015) to identify the destination image of Macao (China).
Moreover, they represent a variety of trip diaries: TB and TP host travel blogs; TA
hosts OTRs about hotels, restaurants and attractions; and VT, which is a virtual
community, hosts travel pages, travelogues, and OTRs about hotels, restaurants,
things to do, favourites, nightlife, off the beaten path activities, tourist traps,
warnings or dangers, transportation, local customs, what to pack, shopping, and
sports and outdoors.
3.3 Data downloading
Liu (2011) considers that web mining, using data mining techniques intends to find
useful information or to extract knowledge from the hyperlink structure and content
of webpages. To automatize the process of extraction, first a Web crawler
programme is needed, capable of roaming the hyperlink structure and downloading
the linked webpages.
• Web structure mining The first step to download data is to navigate the selected
websites manually to identify the initial pages, that is to say, those containing
hyperlinks which lead to the individual blogs and OTR pages, and save their
complete URLs. According to Liu (2011), web structure mining discovers
practical knowledge from hyperlinks, which represent the structure of the web,
and from anchor text associated with hyperlinks. Figure 2 is constructed using a
sitemap generator and only the hyperlinks related to the example of VT member
81,578 and the Barcelona destination remain. However, the only links that are
used are member’s profile (../81578/), travel blog (430de/), travelogue (../tt/
72833), review topic (430de/6/), travel review (../p/m/145865), and travel guide
123
Table 3 Visitors by country (Alexa.com, 2015-05-31)
TA TB TP VT
Country
Percent Rank Percent Rank Percent Rank Percent Rank
us 58.6 59 14.9 50,165 23.8 20,700 27.8 3712
in 1.9 1267 32.5 12,335 21.2 11,782 14.6 2991

A detailed method for destination image analysis using…
uk ? ? 4.7 33,522 4.0 24,048 5.9 2542
ca ? ? ? ? 4.0 16,063 2.6 3659
? insignificant audience
351
123
Table 4 Trends in web hosting

2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
TA 38 81 117 204 608 1421 5933 28,387 36,045 46,142

TB 139 254 427 662 415 328 362 231 148 160
TP 100 236 276 258 226 238 218 189 346 259
VT 1498 1023 1031 762 413 398 635 306 251 172
Fig. 2 Simplified VT website-map diagram
of member’s hometown (see Fig. 4), because we have discarded the reviews on
restaurants (../p/m/145852/& 430de/2/) and hotels (../p/m/144114/& 430de/3/),
photos-backup (../pb/430de/81578/), and statistics (../stats/81578/).
• Filter setting Second, once the information derived from hyperlinks is obtained,
a study about which are the most efficient filters (level, file type, URL, and
content) to minimize download time and space used on the local disk should be
conducted (Marine-Roig and Anton Clave 2015). Filters can be combined and,
except for the level filter, they can be inclusive or exclusive: (1) a Level 0 filter
only downloads the page indicated by the initial URL, a Level 1 filter,
downloads that page and all the resources directly linked to it, etc. (2) The file
type filter allows for downloads; for example, only HTML files and the
remaining files (multimedia, PDF, etc.) will be visualized if an Internet
connection is available. This system is ideal for analysing the textual content of
diaries saving space in the local disk. (3) The URL filters allow for action on any
part of it (protocol, server, domain, subdirectories or folders, filename and file
type); and (4) the content filter is the least efficient, because it is necessary to
download the page to assess whether or not it contains the chain of key
characters, while with URL filters, only the pages of interest are downloaded
(Fig. 3).
– TA In the case of TA, all the files of interest contain the word Catalonia.
Those which have hyperlinks that lead to OTRs start with Attraction, and
those of the same OTRs start with ShowUserReview; therefore, a couple of
inclusive filename filters are enough: Attraction*Catalonia and
123
Fig. 3 Simplified flow diagram of the process of downloading travel diaries
ShowUserReview*Catalonia. To understand the importance of the filters in

this case, we should bear in mind that TA reached more than 225,000,000
reviews and opinions, and all its webpages are linked at different levels by
hyperlinks.
– TB In this case, it is sufficient to place an inclusive folder filter:/Catalonia/
with no level limit to load files only within the start directory and below,
because the server has a hierarchical territorial structure of folders to store
the files and some excluded filename filters (keywords: photo*, map-*,
hotel*, flight*…) to prevent downloading pictures and maps, or information
about hotels, flights, etc.
– TP This webhost does not have a classification that includes all the
destinations in a region, and, therefore, we have to manually locate the start
webpages of all Catalan destinations (Barcelona, Lloret de Mar, Salou, etc.).
– VT In the case of VT the process must be carried out in two phases. Firstly,
the start pages of destinations are downloaded with a ‘‘Level 1’’ filter. In this
way, the member’s homepage and the destination’s travel page (Fig. 2) are
obtained, which allows for extracting the travelogues and OTR URLs which
correspond to said member and destination to download them in a second
phase with an included content filter (keyword: [destination name\/a[).
This filter prevents travel pages, travelogues and OTRs from getting mixed
up with other destinations the user has visited, because all pages have the
destination in the anchor text of the navigation menu bar.
123
• Download process Figure 3 shows schematically how a web copier works. All
or some of the filters can be activated; if one of the filters in the diagram is not
present, the process advances to the following step. In all cases, HTML files
have been downloaded through a type filter (*.htm/*.html) because they contain
all the necessary information for content analysis, and the files associated to the
webpage, such as pictures (Fig. 5), can be accessed with Internet connection.
The flowchart is more complex when other types of files, such as images or
multimedia, are downloaded because the hyperlinks of their respective
downloaded webpages must be converted in order to point to their offline
locations, and multimedia files occupy a lot of space on the local disk.
Firstly, the starting pages of Catalan destinations must be downloaded. If one

downloaded page does not fulfil the content filter, the process downloads the
following page. If it does fulfil it, the file is stored on the local hard disk and is
processed. After checking the level filter, the hyperlinks on the stored page are read
and URLs that have already been processed are excluded. Finally, if hyperlinks
fulfil the URL, filename and file type filters, they are placed in a queue of files to be
downloaded. The process continues until there is no file left in the queue.
Finally, once the web copier programme has been configured, we proceed with
the massive download of the HTML pages of each website. In this research, we used
the Offline Explorer Enterprise (OEE) application that has the capacity to download
up to 100 million URLs per project, and offers the fastest possible multi-threaded
processing of downloaded files by using all central processing unit (CPU) cores
(MetaProducts.com).
3.4 Data arrangement
To facilitate multiple classifications in a region divided into several territorial

brands, we adopted the following structure for folders and files (Fig. 4):
root\website\brand\destination\date_lang_isfrom_pagename_[theme].htm
A Website can consist of two-letter initials (Table 2); Brand, an abbreviation of
five letters (Fig. 1); Destination, the name of the destination or if it is a composite
name joined by a hyphen (e.g., Lloret-de-Mar); Date with the format YYYYMMDD,
based on the ISO 8061 norm (International Standards Organization) to allow its
alphabetical or numerical ordering; Lang two-letter code (ISO 639-1; e.g., ca:
Catalan, en: English); IsFrom two-letter country code (ISO 3166-2; e.g., GB, United
Kingdom, US, United States); PageName can contain a combination of codes and/or
words; Theme, an abbreviation of four letters if the website has a thematic
classification (e.g., hote: hotel, rest: restaurant, ToDo: things to do).
• Geographic classification This requires elaborate preparation. It includes in a

comma-separated values (CSV) file all the territorial information about the
names and codes (if available) of the destinations. The CSV files are plain text
and can be manipulated with a simple text editor or a spreadsheet.
123
Fig. 4 Files and folders of VT before and after the arrangement process
• Temporal classification Based on heterogeneous originals (TA: January 13,

2012; TB: April 24th 2004; TP: Saturday, August 24, 2013; and VT: Dec 6,
2004), dates are converted to the format YYYYMMDD.
• Thematic classification In TA and VT, there is a general classification of OTRs
into hotels, restaurants, and things to do. VT has catalogued the tips: Path (off
the beaten path), Shop (shopping), ToDo (things to do), Warn (warnings or
dangers), etc. TA sub-classifies the things to do into types and/or categories, but
a general thematic classification cannot be made, because the same attraction
can belong to multiple categories or types.
• Language detection. This process is more complex: (1) it has to be based on
textual content without structure (plain text) after the cleaning phase, because
what the traveller has written only represents a minimal part of the page content
(Fig. 5). (2) Specialized software is needed. In this study a Java programme
(Marine-Roig 2013: Annex A3) based on the Language Detection Library
(LDL) of N. Shuyo detects 53 languages, which can be extended with an
included process. (3) In the case of Catalonia, the Catalan language Wikipedia
(more than a million pages) was downloaded, and through a frequency analysis
of one character and groups of two and three consecutive characters, the
programme deduces the probability that such items appear at the beginning, end
or within a word. (4) By means of this system, based on the Naive Bayes
classifier, LDL detects each language with a great degree of precision
(probability higher than 99 %). To extract textual content a free utility, HTML
As Text (NirSoft.net), was used. After that, the language detection programme
was run (Marine-Roig, 2013: Annex A3), which returns a CSV file with the code
ISO 639-1 and the probability of success. Those diaries with a probability lower
123
Before: 74.58 KB After: 2.94 KB (Both files only have text-based HTML code)
Fig. 5 Blurred VT travel page before and after the cleaning stage
than 85 % were considered unclassified, because they usually have an

insufficient quantity of text or are multilingual.
Once the CSV files are ready, a batch programme (Marine-Roig 2013: Annex
A3) is run for each website, which goes through all files, extracts internal data such
as the date of the diary and the name of the destination, eliminates entries without
narrative content (more than 100,000 OTRs in the case of TA), changes the format
of such dates to YYYYMMDD, creates new territorial directories, and transfers the
diary to the destination folder previously prepared with its articulated name to
facilitate future classifications. Finally, the two-character ISO 639-1 codes are
introduced in the name of the files, after the date (Fig. 4).
This organization of data allows obtaining any target subset by means of the
utilities in the operating system. For example (see Fig. 4), we can select all trip
diaries written in English by Russian tourists during 2005 using a simple expression
with wildcards: 2005????_en_RU_*.htm (‘*’: any character zero or more times, ‘?’:
any character once).
3.5 Data cleaning
The data cleaning phase consists of eliminating all the noise surrounding what is
written and posted by the author (Fig. 5). The original HTML format should be
preserved in order to weight keywords and key phrases according to their emphasis
or potential impact (Wahsheh et al. 2012). Programmes such as Site Content
Analyzer (SCA; CleverStat.com) take this format into account to calculate the
123
frequency, site-wide density and average weight of keywords (Yadav and Yadav
2011).
Considering that the webpages of each site have a homogeneous structure and
codification, elements that are not going to be used should be deleted. In general, we
can eliminate some tags with their content: meta (metadata), form (forms), iframe
(inline frames), comment (\!– comments in the source code –[), and script (client-
side scripts). With a web editor, such as Microsoft Expression Web 4 (free version),
all the superfluous HTML elements can be located manually, such as the header and
footer sections (Fig. 5), which do not have a relationship with what the user has
written, and list opening and closing HTML tags in a TBL file (generic Table file).
Both the opening and closing tags as well as the text delimited by them have been
recursively removed (i.e., tags and nested tags are removed) with an ad hoc
programme (Marine-Roig 2013: Annex A3). The programme itself can serve to
eliminate the additional OTRs of TA. This website adds some OTRs related with the
review contained in the page, which already have their respective pages. To avoid
this redundant information, the programme must delete them all except for the first.
3.6 Data debugging
Although the analysed travel blogs and OTRs are written in English, there are
proper nouns (destination and attraction names) with codified cedillas and accent
marks, mistakes and translations. To understand how this step, along with the
previous one, affects the quality of collected data, we can consider an example
related to the case study: The Basilica (formerly Expiatory Temple/Church) of La
Sagrada Famı́lia (i.e., ‘i’ with an acute accent in Catalan), so-called the ‘‘unfinished
cathedral’’, is an outstanding landmark of Catalonia (Marine-Roig 2015b), which is
written in multiple ways by encoding problems (due to the acute accent),
translations (Spanish: Sagrada Famı́lia without accent; and English: Sacred/Holy
Family), and misspellings (as a curiosity, we detected and amended more than 100
different ways of writing Sagrada Famı́lia incorrectly).
• Character encoding problems Problems appear with the previously mentioned

characters, higher codes than ASCII 127 standard, because the website can
codify the ASCII extended characters in three ways: with an HTML entity
(number or name) or with the webpage code (UTF-8 charset). For example, the
surname of the Catalan architect Antoni Gaudı́ ends with an ‘ı́’ (i.e., ‘i’ with an
acute accent), then ‘ı́’ (ASCII 237) can appear as an HTML number
(Gaudí), HTML name (Gaudí), and UTF-8 (GaudÃ–).
• Misspellings The most common mistakes (MCM) are found in destination and
attraction names (Cakmak and Isaac 2012), when there is no orthographic
agreement between the Catalan phonemes and the English graphemes that
represent them. Additionally, it is quite common for an English-speaking
blogger to omit accents, for example, the surname mentioned in the previous
paragraph would be written as Gaudi (without accent). By means of a
preliminary frequency analysis we can locate the MCM to correct them in the
123
webpage text or take them into account in the categories, e.g., Gaudi would have
two keywords, one with an accent and another without.
• Translations It is also common to find proper nouns translated or semi-translated
from Catalan to Spanish or English. For example, Parc Güell (‘u’ with dieresis)
is an intensely visited attraction due to its World Heritage Site status (UNESCO
2005), and can be found written in multiple forms due to the dieresis (character
encoding problem) and the translations (Spanish: Parque, English: Park). On the
one hand, the MCM are corrected and on the other, relatively correct keywords
are introduced (Parc Guell, Park Guell, Guell Park, etc.) along with the correct
expression (Parc Güell) in the composite-words list and in its category (Tangible
heritage).
Such codifications and MCMs distort content analysis and should be corrected.
The UTF codes and HTML entities can be related in a CSV file with their
corresponding Latin-1 character (ISO 8859-15), and the MCM and translations
together with the correct word, and transfer it as a parameter to a search-and-replace
utility to proceed with their replacement.
3.7 Content analysis
Once the previous phases have been performed, the travel blog and OTR dataset is
ready for any kind of content analysis, qualitative or quantitative. In this research,
travel diaries written in English have been selected and a first offline analysis of the
frequency, density and weight of keywords has been conducted, with the SCA
programme. This software generates a CSV file for each travel blog or OTR
conveying all the words appearing in that entry-file, their frequency, density and
weight. It parses online and offline for keywords, suggests the most relevant and
weighty phrases, and analyses link structure (CleverStat.com). SCA assigns a
weight to keywords and key phrases according to their position within the webpage
and the HTML tag that defines their format and features (Wahsheh et al. 2012;
Yadav and Yadav 2011). The HTML tags in weight order are: first, the ‘‘title’’
(required in all HTML documents), then the headings (‘‘h1’’, ‘‘h2’’ and ‘‘h3’’),
followed by ‘‘a’’ (defines a hyperlink), ‘‘img alt’’ (alternate text for an image), then,
with the same weight, come ‘‘h4’’ (heading), ‘‘b’’ (bold font), ‘‘i’’ (italic font), ‘‘u’’
(underline font), ‘‘strong’’ (defines important text), ‘‘em’’ (renders emphasized text),
etc. Results can serve, for example, to study different tourist modes by grouping
keywords into categories (sun, sea, and sand; urban environment; tangible heritage;
etc.).
The SCA parser should be configured with the preferences for the case study
before it is conducted. The most important ones are the black list and the composite
words. The first prevents meaningless keywords (stop words) from being analysed,
such as adverbs, articles, conjunctions, prepositions and pronouns (Cakmak and
Isaac 2012; Krawczyk and Xiang 2015; Marine-Roig and Anton Clave 2015);
moreover, fewer than three-letter words are dismissed. The second list indicates the
word groups that form a unit (Cakmak and Isaac 2012; Li et al. 2015) such as
Sagrada Familia.
123
4 Results
With the classification system seen in Sect. 3.4 it is easy to analyse the trends
followed by the 132,502 travel blogs and OTRs along the 11 years of study
(Table 5). In this table, 434 unclassified entries do not appear because the travellers
put Catalonia as the destination without specifying brand, city or town.
Concerning the distribution by Catalan brands, we observe that the Barcelona
brand (Barna) far outnumbers all the rest and the number of travel diaries has
increased 23 fold in the last 5 years. Additionally, we can observe that the Terres de
Lleida (tLlei), Terres de l’Ebre (tEbre) and Val d’Aran (vAran) brands do not have
enough entries to sustain reliable content analysis. Comparing both major coastal
destinations, it can be seen that Costa Daurada (cDaur) has grown considerably in
recent years ahead of Costa Brava (cBrav). Finally, a decrease was observed during
2014 in travel diaries on Barcelona Landscapes (pBarc).
With respect to the problem of the seasonality of the tourist industry in the
Spanish coastal regions, consisting of the concentration of sun, sea and sand tourists
in the months of July and August, Fig. 6 show that high season for the type of
tourism studied extends to almost 6 months.
As an example of the validity, usability and capacity of the method, a simple first
level of content analysis using SCA was performed with 131,223 travel diaries
written in English (99.03 % of the 132,502 gathered). As illustrated in Table 6, very
significant results were obtained. Coinciding with the results in Table 5 concerning
Barcelona brand (Barna), it can be observed in Table 6 that the keyword Barcelona
has a much higher density than the other ones and a considerable weight. Barcelona
is a leading smart city (Marine-Roig and Anton Clave 2015) and the sixth most
powerful city brand in the world (Michael 2014). The Catalan Tourist Board is
aware of the significance of the Barcelona brand and recently renamed the ‘‘Central
Catalonia’’ brand ‘‘Barcelona landscapes’’ (pBarc) (Datzira-Masip and Poluzzi
Table 5 Trends in Catalan tourist brands

Barna cBarc cBrav cDaur pBarc Pyren tLlei tEbre vAran
2004 1177 34 201 61 57 10 6 4 1

2005 1374 42 204 46 45 20 1 3 0
2006 1191 53 163 82 38 12 1 0 7
2007 1309 70 238 117 37 25 3 1 0
2008 1367 79 191 134 45 8 5 1 0
2009 1295 34 134 121 20 10 5 5 3
2010 1742 63 177 288 35 22 11 1 2
2011 5828 115 332 698 89 14 19 2 3
2012 24,211 325 1448 2599 412 62 16 3 9
2013 30,875 560 1707 2498 927 149 16 10 11
2014 40,232 745 1928 3070 517 144 37 24 2
123
Fig. 6 Monthly distribution of travel blogs and OTRs (132,502 from TA, TB, TP, & VT)
2014). Among the top 25 keywords, we can find eight good feelings, as well as the
architect Antoni Gaudi and two of his masterpieces (Sagrada Familia and Parc
Guell) with a highly significant weight. Gaudi’s work is registered in UNESCO’s
World Heritage List (UNESCO 2005). A new record of visitors (3260,880) to the
Sagrada Familia basilica was reached in 2014, of which 12 % were North-
American, 10 % French, 6 % Italian, and 6 % English (Marine-Roig 2015b).
These results obtained by SCA can serve for different studies such as grouping
keywords into thematic categories, how to ascertain the weight of a certain tourist
modality, attraction factors, feelings, dichotomies, etc. in the whole region and in
sub-regions. The same studies can be conducted on the Catalan brand or at a city
level (Marine-Roig and Anton Clave 2015) and circumscribed to a certain period of
time by applying SCA programme to the corresponding HTML subset.
Furthermore, these initial results give insights into the cognitive/functional and
affective components of destination image. It is especially noteworthy that the sum
of almost half a million words (492,780) related to good feelings (according to
standard lists of positive adjectives in American and British English) present in the
top 25 words those users mention. This gives a good indication of the positiveness in
general of the affective component, which may have strong implications for tourist
satisfaction and destination loyalty. In terms of the cognitive or functional
component of image we get insights into the main destinations in the region
(Barcelona), the most popular attractions (e.g., Sagrada Familia) and the main
activities (e.g., tour, city, walk, and trip) conducted.
Moreover, in terms of the relationship of cognitive and affective image
components, and their relative weight, this could provide an interesting overview
showing that cognitive elements account for about 58 % of the most mentioned
words (697,169 mentions), and elements related to the affective component account
for approximately 42 % of UGC text in this case.
123
Table 6 The 25 most frequent words of 93,765 unique words

Rank Keyword Count Sitewide density (%) Average weight Remarks
1 barcelona 285,614 2.67 59.66 Capital of Catalonia

2 tour 117,845 1.10 33.62
3 great 81,265 0.76 24.17 Good feeling
4 sagrada familia 56,133 0.52 65.77 Gaudi’s masterpiece
5 gaudi 45,420 0.42 19.41 Catalan Architect
6 visit 42,001 0.39 15.16
7 place 40,329 0.38 16.40
8 amazing 40,076 0.38 24.90 Good feeling
9 city 39,373 0.37 12.12
10 good 38,753 0.36 15.24 Good feeling
11 beautiful 36,588 0.34 23.84 Good feeling
12 park 36,277 0.34 28.56
13 way 33,614 0.31 15.82
14 nice 29,774 0.28 20.23 Good feeling
15 guell park 29,555 0.28 64.07 Gaudi’s work
16 best 27,722 0.26 24.98 Good feeling
17 museum 27,194 0.25 31.06
18 experience 25,554 0.24 22.69
19 guide 24,183 0.23 10.70
20 walking 23,573 0.22 36.23
21 people 23,502 0.22 4.26
22 walk 22,169 0.21 10.79
23 fun 21,940 0.21 24.51 Good feeling
24 trip 21,515 0.20 14.61
25 interesting 19,980 0.19 17.23 Good feeling
5 Concluding remarks
The proposed methodology facilitates the massive gathering of online UGC data
from the most suitable sources for a specific case study. The hierarchical territorial
structure of folders and the articulation of the name of the files which contain
individual diaries, enable multiple classifications using utilities to order and
manipulate the files of the same operating system. This structure also allows us to
focus the analysis on a specific place, period, language or subject (if this is
available), in particular or combined, selecting the corresponding subset or a
random sample thereof. The cleaning and debugging phases are essential to
obtaining quality information, limited to the web content as written and posted by
the diary author, and overcoming the most significant errors.
Quantitative analysis results, at the level of territorial brands, may be useful for
NTOs to improve their branding and positioning policies; e.g., for the managers of
the Costa Daurada brand, it is interesting to know that tourists mention a theme park
123
(PortAventura) ten times more than the Roman amphitheatre in spite of the fact that
it is a World Heritage Site (UNESCO 2000). Metrics (visibility, popularity,
audience, and size) of webs hosting UGC data can be useful for the DMOs that want
to promote products or services on such websites.
This study aims to contribute to the advancement of research in the field of travel,
tourism and hospitality, because it proposes a methodology to extract knowledge
from large amounts of online UGC data with a degree of detail that was not covered
in previous works. At a first level of analysis, this methodology gives significant
insights concerning both the cognitive and the affective components of destination
image and may be useful for further analyses in this respect. Most of the proposed
phases of this method are useful for the content analysis of other web data sources.
The HTML dataset is prepared for any offline content analysis in future work, for
instance, a qualitative content analysis using qualitative data analysis software or a
sentiment analysis using a computer-aided text-analysis programme.
The main limitation of the application of the method to this case study was the
analysis at the territorial boundary brands and, more specifically, a certain influence
of the Barcelona brand on the two brands surrounding it: Barcelona coast and
Barcelona landscapes; that is, in some trip diaries of cBarc and of pBarc they find
references to Barcelona’s attraction factors. Moreover, the implementation of some
pre-processing steps requires computer skills that are not available to some DMOs
and other stakeholders.
Acknowledgments This work was supported by the Spanish Ministry of Economy and Competitiveness
[Grant id.: MOVETUR CSO2014-51785-R].
References
Abburu S, Babu GS (2013) A framework for web information extraction and analysis. Int J Comp
Technol 7:574–579
Albastroiu I, Felea M (2014) The implications of user-generated content websites for tourism marketing.
Int J Eco Pract Theor 4:222–229
Amaral F, Tiago T, Tiago F (2014) User-generated content: tourists’ profiles on TripAdvisor. Int J
Strategic Innov Marketing 1:137–147. doi:10.15556/IJSIM.01.03.002
Anton Clavé S, Gonzalez F (2008) A proposito del turismo: La construccion social del espacio turistico
[About tourism: The social construction of tourist space]. Editorial UOC, Barcelona
Baggio R, Corigliano MA (2009) On the importance of hyperlinks: A network science approach. In:
Hopken W, Gretzel U, Law R (eds), Information and Communication Technologies in Tourism
2009, Springer, Austria, pp 309-318. doi: 10.1007/978-3-211-93971-0_26
Banyai M, Glover TD (2012) Evaluating research methods on travel blogs. J Travel Res 51:267–277.
doi:10.1177/0047287511410323
Cakmak E, Isaac RK (2012) What destination marketers can learn from their visitors’ blogs: an image
analysis of Bethlehem, Palestine. J Destination Marketing Manag 1:124–133. doi:10.1016/j.jdmm.
2012.09.004
Chau M, Xu J (2012) Business intelligence in blogs: understanding consumer interactions and
communities. MIS Q 36:1189–1216
CTB (2015) Press Pack’15. Catalan Tourist Board. http://www.act.cat/press-pack/?lang=en. Accessed 31
July 2015
Datzira-Masip J, Poluzzi A (2014) Brand architecture management: the case of four tourist destinations in
Catalonia. J Dest Market Manag 3:48–58. doi:10.1016/j.jdmm.2013.12.006
123
Dwyer L, Gill A, Seetaram N (eds) (2014) Handbook of research methods in tourism: quantitative and
qualitative approaches, Reprint edn. Edward Elgar Publishing, Cheltenham
Eurostat (2015) Tourism. Eurostat regional yearbook 2015. Publications Office of the European Union,
Luxembourg, pp 189–206
Fang B, Ye Q, Kucukusta D, Law R (2016) Analysis of the perceived value of online tourism reviews:
influence of readability and reviewer characteristics. Tour Manag 52:498–506. doi:10.1016/j.
tourman.2015.07.018
Gonzalez R (2010) Estudi exploratori de les marques turistiques de Catalunya a la Web 2.0 [Exploratory
study of tourist brands of Catalonia to Web 2.0]. Vivential Value, Barcelona, Catalonia
Guo L, Li Z, Sun W (2015) Understanding travel destinations from structured tourism blogs. WHICEB
2015 Proceedings, pp 144–151. http://aisel.aisnet.org/whiceb2015/80. Accessed 31 Mai 2015
Gursoy D, Uysal M, Sirakaya-Turk E, Ekinci Y, Baloglu S (2015) Handbook of scales in tourism and
hospitality research. CABi, Wallingford
Johnson PA, Sieber RE, Magnien N, Ariwi J (2012) Automated web harvesting to collect and analyse
user-generated content for tourism. Curr Issues Tourism 15:293–299. doi:10.1080/13683500.2011.
555528
Kladou S, Mavragani E (2015) Assessing destination image: an online marketing approach and the case
of TripAdvisor. J Destination Market Manag 4:187–193. doi:10.1016/j.jdmm.2015.04.003
Koltringer C, Dickinger A (2015) Analyzing destination branding and image from online sources: a web
content mining approach. J Bus Res 68:1836–1843. doi:10.1016/j.jbusres.2015.01.011
Kotler P, Haider DH, Rein I (1993) Marketing places: Attracting investment, industry and tourism to
cities, states and nations. The Free Press, New York
Krawczyk M, Xiang Z (2015) Perceptual mapping of hotel brands using online reviews: a text analytics
approach. Inf Technol Tourism. doi:10.1007/s40558-015-0033-0 (in press)
Lai LSL, To WM (2015) Content analysis of social media: a grounded theory approach. J Electronic
Commerce Res 16:138–152. http://www.jecr.org/node/466. Accessed 31 May 2015
Law R, Cheung S (2010) The perceived destination image of Hong Kong as revealed in the travel blogs of
mainland Chinese tourists. Int J Hospitality Tour Administration 11:303–327. doi:10.1080/
15256480.2010.518521
Li YR, Lin YC, Tsai PH, Wang YY (2015) Traveller-generated contents for destination image formation:
mainland China travellers to Taiwan as a case study. J Travel Tourism Marketing 32:518–533.
doi:10.1080/10548408.2014.918924
Liu B (2011) Web data mining: exploring hyperlinks, contents, and usage data. Springer, Berlin
Llodra-Riera I, Martinez-Ruiz MP, Jimenez-Zarco AI, Izquierdo-Yusta A (2015) A multidimensional
analysis of the information sources construct and its relevance for destination image formation. Tour
Manag 48:319–328. doi:10.1016/j.tourman.2014.11.012
Lu W, Stepchenkova S (2015) User-generated content as a research mode in tourism and hospitality
applications: topics, methods, and software. J Hospitality Marketing Manag 24:119–154. doi:10.
1080/19368623.2014.907758
Marchiori E, Onder I (2015) Reframing the image of a destination: A pre-post study on social media
exposure. In: Tussyadiah I, Inversini A (eds), Information and Communication Technologies in
Tourism 2015, Springer, Switzerland, pp 335–347. doi:10.1007/978-3-319-14343-9_25
Marine-Roig E (2013) From the projected to the transmitted image: the 2.0 construction of tourist
destination image and identity in Catalonia. PhD dissertation. http://hdl.handle.net/10803/135006.
Accessed 31 May 2015
Marine-Roig E (2014) A webometric analysis of travel blogs and reviews hosting: the case of Catalonia.
J Travel Tourism Marketing 31:381–396. doi:10.1080/10548408.2013.877413
Marine-Roig E (2015a) Identity and authenticity in destination image construction. Anatolia Int J
Tourism and Hospitality Res 26:574–587. doi:10.1080/13032917.2015.1040814
Marine-Roig E (2015b) Religious tourism versus secular pilgrimage: The basilica of La Sagrada Familia.
Int J Religious Tourism Pilgrimage 3:25–37. http://arrow.dit.ie/ijrtp/vol3/iss1/5/. Accessed 31 July
2015
Marine-Roig E, Anton Clave S (2015) Tourism analytics with massive user-generated content: a case
study of Barcelona. J Destination Marketing Manag 4:162–172. doi:10.1016/j.jdmm.2015.06.004
Michael C (2014, May 6) From Milan to Mecca: the world’s most powerful city brands revealed.
TheGuardian, News, Cities, City brand. http://www.theguardian.com/cities/gallery/2014/may/06/
from-milan-to-mecca-the-worlds-most-powerful-city-brands-revealed. Accessed 31 Mai 2015
Minazzi R (2015) Social media marketing in tourism and hospitality. Springer, Cham
123
Moens MF, Li J, Chua TS (eds) (2014) Mining user generated content. CRC Press, Boca Raton
Munar AM (2012) Social media strategies and destination management. Scand J Hospitality Tourism
12:101–120. doi:10.1080/15022250.2012.679047
Munar AM, Jacobsen JS (2013) Trust and involvement in tourism social media and web-based travel
information sources. Scand J Hospitality and Tourism 13:1–19. doi:10.1080/15022250.2013.764511
Pan B, MacLaurin T, Crotts JC (2007) Travel blogs and the implications for destination marketing.
J Travel Res 46:35–45. doi:10.1177/0047287507302378
Schmunk S, Hopken W, Fuchs M, Lexhagen M (2014) Sentiment analysis: Extracting decision-relevant
knowledge from UGC. In: Xiang Z, Tussyadiah L (eds), Information and communication
technologies in tourism 2014, Springer, Switzerland, pp 253–265. doi:10.1007/978-3-319-03973-
2_19
Schuckert M, Liu X, Law R (2015a) Hospitality and tourism online reviews: recent trends and future
directions. J Travel Tourism Marketing 32:608–621. doi:10.1080/10548408.2014.933154
Schuckert M, Liu X, Law R (2015b) Insights into suspicious online ratings: direct evidence from
TripAdvisor. Asia Pacific J Tourism Res. doi:10.1080/10941665.2015.1029954 (in press)
Serna A, Marchiori E, Gerrikagoitia JK, Alzua-Sorzabal A, Cantoni L (2015) An auto-coding process for
testing the cognitive-affective and conative model of destination image. In: Tussyadiah I, Inversini
A (eds), Information and Communication Technologies in Tourism 2015, Springer, Switzerland,
pp 111–122. doi: 10.1007/978-3-319-14343-9_9
Sigala M, Christou E, Gretzel U (eds) (2012) Social media in travel, tourism and hospitality: Theory,
practice and cases. Ashgate Publishing, Surrey
Standing C, Tang-Taye JP, Boyer M (2014) The impact of the Internet in travel and tourism: a research
review 2001–2010. J Travel Tourism Marketing 31:82–113. doi:10.1080/10548408.2014.861724
UNESCO (2000) Archaeological ensemble of tarraco. world heritage list. http://whc.unesco.org/en/list/
875. Accessed 31 May 2015
UNESCO (2005) Works of Antoni Gaudi. world heritage list. http://whc.unesco.org/en/list/320. Accessed
31 May 2015
Wahsheh HA, Alsmadi IM, Al-Kabi MN (2012) Analyzing the popular words to evaluate spam in Arabic
web pages. Res Bull Jordan ACM 2:22–26
Wang Y, Morais DB (2014) An examination of tourists’ identity in tourist weblogs. Inf Technol Tourism
14:239–260. doi:10.1007/s40558-014-0016-6
Wang Y, Chan SC, Ngai G, Leong HV (2013) Quantifying reviewer credibility in online tourism. In:
Decker H et al (eds) DEXCA 2013 Proceedings of 24th International Conference: Database and
Expert Systems Applications. Czech Republic, Prague, pp 381–395
Xiang Z, Schwartz Z, Uysal M (2015) What types of hotels make their guests (un)happy? Text analytics
of customer experiences in online reviews. In: Tussyadiah I, Inversini A (eds), Information and
Communication Technologies in Tourism 2015, Springer, Switzerland, pp 33–45. doi:10.1007/978-
3-319-14343-9_3
Yadav Y, Yadav PK (2011) Site Content Analyzer in context of keyword density and key phrase. Int J
Comp Technol Appl 2:860–872
Ying T, Norman WC, Zhou Y (2014) Online networking in the tourism industry: a webometrics and
hyperlink network analysis. J Travel Res. doi:10.1177/0047287514532371
Zeng B, Gerritsen R (2014) What do we know about social media in tourism? A review. Tourism Manag
Persp 10:27–36. doi:10.1016/j.tmp.2014.01.001
123

A Detailed Method For Destination Image Analysis Using

Încărcat de

Informații document

Titlu original

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

A Detailed Method For Destination Image Analysis Using

Încărcat de

Drepturi de autor:

Formate disponibile

Inf Technol Tourism (2016) 15:287–290

Editorial: Special issue on ENTER2015

Iis Tussyadiah1 • Alessandro Inversini2

Received: 17 December 2015 / Revised: 21 December 2015 / Accepted: 23 December 2015 /

As information and communication technology (ICT) has become integral to key

& Alessandro Inversini

recommendation for consumers based on the identified patterns of dynamic pricing

User reactions to destination brand contents in social

Assumpcio Huertas1 • Estela Marine-Roig2

Received: 12 June 2015 / Revised: 30 November 2015 / Accepted: 8 December 2015 /

Keywords Online communication Social media Facebook Interactivity

& Estela Marine-Roig

2.1 Destination brands

2.2 Destination brand communication in social media

2.3 Reactions and interactivity

The proposed methodology consists of a content analysis of attraction factors and

directing and adjusting their communication and branding strategies. We must

3.1 Destination and social media selection

Table 1 Average Facebook Fanpage fans per destination type

Data retrieved on 01/05/2014

3.3 Content analysis

build templates or models for the content analysis of destination communication

• Sincerity: Down-to-earth (Family-oriented, Down-to-earth, Sustainable); Honest

• Ruggedness: Outdoorsy (Outdoorsy, Get-away, Recreational); Tough (Tough,

3.4 Data analysis

Total posts 178 104 128 40 119 63 56 26

4.1 Mentions and reactions per themes

4.1.1 Most mentioned themes

The most mentioned theme in general (Table 3) is Tangible Heritage (298)

Table 3 Number of mentions of themes in top posts

Tangible heritage 105 38 20 9 65 40 20 1 298

4.1.2 Themes triggering the most reactions

Fig. 1 Reactions per theme as a percentage of the number of fans

4.2 Mentions and reactions per brand values

4.2.1 Most mentioned brand values

4.2.2 Brand values that trigger the most reactions

4.3 Reactions to certain themes/brand values and destination types

• In National Destination and Autonomous Communities (ND ? AC), the theme

Fig. 2 Reactions per brand value as a percentage of the number of fans

These attributes are especially related to Happiness and International atmo-

Table 5 Brand values mentions

conclude that the communication of both tangible/functional elements and,

A detailed method for destination image analysis using

Estela Marine-Roig1 • Salvador Anton Clavé2

Received: 12 June 2015 / Revised: 16 October 2015 / Accepted: 18 November 2015 /

Abstract Social media and user-generated content (UGC) have revolutionized

& Estela Marine-Roig

Destination image is a complex construct, and is usually considered to be the sum of

2 State of the art

Social media is characterized by using ever-changing tools; new platforms are

CTrip.com to analyse the destination image of Taiwan. As examples of massive

3.1 Destination selection

3.2 Hosting websites selection

By means of a popular specialized search engine (BlogSearch.Google.com: ‘‘travel

Source Trends in the main tourism magnitudes (Catalan Tourism Observatory)

Tourist brand Abbr.

Fig. 1 Tourist brands of Catalonia (CTB 2015)

Indexed pages Google.com 144,000,000 453,000 334,000 554,000

• Visibility (V) = 1 9 B(Google) ? 1 9 B(Bing) ? 1 9 B(PR) ? 1 9 (CY). This

3.3 Data downloading

us 58.6 59 14.9 50,165 23.8 20,700 27.8 3712

in 1.9 1267 32.5 12,335 21.2 11,782 14.6 2991

uk ? ? 4.7 33,522 4.0 24,048 5.9 2542