5. Exploring tourists’ experiences using digital resources
Analyzing Chinese tourists visiting the Geiranger fjord
using travel blogs
- Side: 88-102
- DOI: https://doi.org/10.18261/9788215037394-2020-05
- Publisert på Idunn: 2020-07-20
- Publisert: 2020-07-20
- Creative Commons (CC BY 4.0)
Denne artikkelen ser på kinesiske turisters opplevelser av besøk til Geirangerfjorden ved å analysere 196 blogginnlegg hvor de beskriver sine opplevelser. Analysen viser at 92 prosent av innleggene beskriver positive opplevelser. Ved bruk av ‘Grounded theory’ identifiserer vi fem konstruksjoner. Av disse er den mest interessante kinesiske turisters interesse og idealisering av levemåte og kultur i Norge og Skandinavia.NØKKELORD: turisme, naturbaserte opplevelser, Skandinavia, bloggdata
This paper explores the experiences of Chinese tourists who visit the Geiranger fjord by using 196 blog posts of tourist’s description of their experience. Our analysis shows that 92 percent of the blog posts portray positive experiences. Using grounded theory, we identify five constructs representing core motivational ideas. The most interesting construct we identify is the Chinese tourists’ interest and idealization of the Norwegian and Scandinavian way of living and culture.
Norway is a country with famous nature-based tourist destinations, and tourism is an important factor in the Norwegian economy. Every year, millions of foreign tourists visit Norway to experience these destinations (Haukeland and Rideng, 2007; Farstad et al., 2011). From 2000 to 2011, the number of foreign tourist arrivals increased from 3 million to nearly 5 million, while the number of guest nights increased from 22 to 33 million (Dybedal and Farstad, 2013).
In recent years the number of tourists from China has increased significantly. While the number of non-Chinese foreign overnight stays at Norwegian hotels has increased by 23.6 percent in the period 2010–2018, the increase from China is 562 percent in the same period, according to figures from Statistics Norway. The main factors behind this growth are Norway’s promotion of the tourist sector (e.g. www.visitnorway.no) and the recent economic development in China (Shi, 2015).
Understanding the experiences of these Chinese tourists is essential. The reason for this importance is that the Chinese culture is very different from the European. Although fragmentary interviews provide some knowledge of Chinese tourists coming to Norway, systematic analysis of Chinese tourists’ experiences in the Norwegian fjords is limited.
The recent digitalization of travel reviews offers new possibilities for studying these tourists. Web blogs and new methods of analyzing large and complex datasets (‘big data’ analytics) have also been applied to tourism research in recent years (Li et al., 2018). The development of China’s mainstream tourism websites provides the possibility of obtaining first-hand narrative information about tourists’ travel experiences. The purpose of this study is to use these travel blogs to gain a deeper understanding of Chinese tourists’ experience. As a case study, the analysis focuses on the Geiranger Fjord – one of Norway’s’ most popular tourist attractions. The paper uses a mixed-method approach by combining quantitative (sentiment scoring) and qualitative (grounded theory analysis) methods.
Previous research on tourism focuses on motivation, involvement, novelty, satisfaction, and the relationship between those concepts. Basic tourism theory suggests that people travel and participate in activities because they are pushed by their own internal motivations and/or pulled by the external forces of destination attributes (Kim & Lee, 2001; Mohammad & Som, 2010). Although not being completely separable from each other, push factors generally refer to the specific forces that influence a person’s decision to take a vacation (e.g. escape, novelty), while pull factors focus more on the forces that influence the person’s decision to select a specific destination (e.g. mountains and beautiful scenery, or historic and cultural resources) (Kim & Lee, 2001).
The main factors in tourist motivation identified in the literature are knowledge-seeking, cleanliness & safety, escaping from the daily routine, fitness in nature, relaxation, appreciating natural resources, and cultural attractiveness (Jang and Wu, 2006; Mohammad and Som, 2010; Dimitrovski and Todorovic, 2015; Landauera, 2011; Kim et al., 2003; van der Ark et al., 2006).
The findings in the literature highlight satisfaction as a key determinant of destination loyalty and revisits (Jang and Feng, 2007; Meleddu et al., 2015; Kozak, M, 2001; Denstadli and Jacobsen, 2011). Tourists’ mental representation of knowledge, feelings and overall perception of a destination (destination image) is a key factor in tourist satisfaction. A positive destination image results in a higher overall satisfaction (Fakeye and Crompton, 1991; Assaker et al., 2011; Albaity and Melhemb, 2017). Other experiencing factors that influence satisfaction are scenery, meals, road facilities, transport services, climate, hospitality, and customer care (Hasegawa, 2010; Denstadli & Jacobsen, 2011; Kozak, 2001; Denstadli et al., 2011).
Several studies have examined the satisfaction of tourists visiting Norway. In an econometric study, Engstrom and Kipperberg (2015) identified elasticity measures concerning the length of stay, travel party size, and income, and uncover several other statistically significant and economically important determinants of expenditure. Denstadli and Jacobsen (2011) indicate that it is important to take into consideration the motorists’ desire to experience attractive sceneries to increase overall route satisfaction. Denstadli et al., (2011) studies the importance of weather conditions for tourism in the north of Norway and found, somewhat surprisingly, that weather condition had minor behavioral impacts.
Two studies have examined the experience of Chinese tourists’ in Norway. Using narrative analysis of the online travel community, Shi (2015) explored how Chinese-speaking visitors shared their experience and found that their main motivation for traveling is the experience of the unique Norwegian nature. On the other hand, Zhang (2016) focuses on cultural difference. This study shows that understanding the preference of international tourists is important in promoting Norwegian tourism and can improve their satisfaction with service. Although these studies provide useful insights, to date no studies have specifically investigated Chinese tourist’s experience of the Norwegian fjords. Considering that destination development requires thorough understanding of tourists’ motives, behavior, and experiences (Manhas et al. 2016), the findings presented in this paper provide useful insights for development of nature-based tourist destinations in Norway, particularly the destination of Geiranger.
Recent advances in computer science offer new approaches to investigating tourists’ perceptions and experiences. Social media platforms are one of the promising data sources since tourists use them to share tourism-related information, such as travel reviews and experiences (Li et al., 2018). Compared with the traditional survey interview, blogs enable us to target specific questions at a low cost and provide rich data regarding tourists’ travel stories and emotions.
The collection of blog posts
The dataset used in our analysis is a collection of web blogs from one of the biggest Chinese-speaking travel communities, www.mafengwo.cn. In addition to blogs and presentations of destinations all over the world, the website provides hotel reservation and flight ticket ordering services. The blog data collection was conducted using web scraping in R (Munzert et al., 2015), which was performed from November to December in 2018.
The collection of the blog data followed three steps: First, blog posts about traveling to Norway were obtained by searching for 25 keywords (based on the information using the search engines www.baidu.com, travel website www.mafengwo.cn, and www.tripadvisor.com), in both Chinese and Norwegian (see Appendix A), representing the names of the most famous fjords, the largest cities in Norway, and the most popular attractions and roads near Geiranger. This search resulted in around 2000 blog posts written in Chinese text (one of the authors of this article has Chinese as mother tongue). Second, the blogs were cleaned up by removing unnecessary symbols and correcting spelling, then around 400 blogs were selected based on the relevant keywords directly connected to Geiranger or nearby cities and attractions. In the third stage, after manually identifying the content describing a trip to the Geiranger fjord, we were left with 196 usable blog posts. From these blog posts, texts about trips to Geiranger were selected and the relevant text paragraphs were extracted as the data source for analysis.
In the quantitative analysis, we calculated rough measures of tourist satisfaction regarding their trips to Geiranger. First, we translated destinations and attractions into Chinese. Using these words, the R package ‘chinese.misc’ (Wu, 2019) mines the HTML documents into text. From these documents, all sentiment words are identified using a subjective lexicon consisting of 27,000 Chinese sentiment words (Dalian University of Technology Chinese emotional words, 2017). Using this lexicon, each word is classified by sentiment type (good, joy, surprise, disgust, sadness, fear, anger), polarity (1/0/–1 for positive/neutral/negative) and strength (0–9). A simple approach to sentiment scoring (Kwartler, 2017) was adopted to evaluate the polarity of the blog text by adding up the positive score in a passage and subtracting the negative ones, giving a net polarity score (sentiment) for each blog.
The concept of net polarity can be described with an example sentence:
“I like the beauty of Geiranger fjord, but it often rains here.”
In this sentence, there are two sentiment words: beauty and rains. The calculation of the net polarity is shown in table 5.1. According to the lexicon, the word ‘beauty’ has a positive polarity (score=1), strength=5, and sentiment type ‘good’. The word ‘rain’ has negative polarity (score= –1), strength=1, and sentiment type ‘sad’. The net polarity of the sentence is found by multiplying the polarity with strength for each word and calculate the sum of all words. The weighted polarity of ‘beauty’ is 5 (5 × 1) and –1 for ‘rain’ (–1 × 1). The sum of the weighted polarity is, therefore, 5 + (–1) = 4.
The representative sentiment is calculated by choosing the sentiment type with the highest absolute value of the weighted polarity. In the example sentence, the representative sentiment is ‘beauty’ because of the absolute value of |5|>|–1|.
After directly extracting the text including sentiment words from the last step, grounded theory method (Auerbach & Silverstein, 2004) was employed to code and categorize useful repeating ideas in order to aggregate theoretical constructs and themes representing the tourists’ general satisfaction with the trip to the Geiranger fjord.
According to Auerbach and Silverstein (2004), grounded theory is a qualitative research method that allows researchers to develop hypotheses using theoretical coding based on text. In the context of consumers (e.g., international visitors) self‐reporting of lived experiences, employing grounded theory provides researchers with a rich method of capturing the complexities and nuances of leisure travel experiences and generates insights into both motivations and behavior (Martin &Woodside, 2007). With the proliferation and availability various type of user-generated data (photos, textual descriptions) in recent years, grounded theory has been also used in online review data analysis in holiday and tourism research (Papathanassis & Knolle, 2011).
The process of the grounded theory method in our example is represented in figure 5.2. The process started by extracting the passages relating to sentiment words from each blog post. These passages formed the basis of the grounded theory analysis. In this analysis, we identified the distinct ideas in each blog post. We investigated the repeating ideas both within and between blogs. The repeating ideas, which formed the building blocks of a theoretical narrative, were then organized into larger groups expressing common themes. Finally, we organized the themes into groups that we called theoretical constructs. An example of extracted construct, theme and repeating ideas is shown in Appendix B.
On the basis of the repeating ideas extracted from blog texts and of insights drawn from the literature above on travel (e.g., motivation, satisfaction, and knowledge-seeking, cultural tourism), we identified theoretical constructs found in blog texts that can represent tourists’ subjective experience of their Geiranger trips. We continued this process until no new ideas appeared and all constructs were well developed; this is often referred to as theoretical saturation (figure 5.2). All the 196 blogs were read by the researchers, although after the first 80 blogs there were no significantly new repeating ideas and constructs being generated. Finally, we created a theoretical narrative by retelling tourists’ experience in terms of theoretical constructs.
Quantitative analysis of sentiment words from blogs
The words from the blog posts with descriptions of trips to Geiranger is presented in table 5.2. The average length of the text describing the trip was 550 words. The shortest had only nine words, while the longest had almost 4000. The average text included 14 sentiment words, with a minimum of zero and a maximum of 107. Hence, the total number of sentiment words used in the analysis was 2,744 words.
Sentiment analysis indicates that the majority of Chinese tourists had a positive experience of Geiranger. Figure 5.3 shows the sentiment category distributions in all of the 196 blogs. On average, each blog contained 550 words, with 14 identified as sentiment words (table 5.2). Almost 98 percent of the sentiment words were positive: ‘good’ or ‘joy’ and ‘surprise.’ Less than 2 percent of the words were associated with negative type sentiments, such as disgust, sadness, fear, or anger.
Looking at each blog separately, we calculate the net positive polarity. A net positive polarity is a blog post with more positive sentiment strength than negative ones. Of the 196 blogs, 92 percent have a net positive polarity, 8 percent are neutral, while no individual blog has a net negative polarity.
The net positive polarity seems very high, but it is consistent with earlier findings. Haukeland and Rideng (2007) show that almost 98 percent of 1478 foreign ski tourists (over 95 percent from Europe) were either satisfied or very satisfied with their experience in Norway. Although not perfectly comparable, this represents a very similar finding. The high degree of satisfaction may also be due to a self-selection bias where the most satisfied tourists are overrepresented in the sample. However, the substantial growth of Chinese tourists to Norway supports the notion of high satisfaction.
Theoretical constructs from grounded theory analysis
The analysis identifies five constructs, which are presented in table 5.3. together with their themes. The table categorizes each construct into push/pull factor and shows different themes of each construct.
|Construct||The beauty of nature||Culture and tradition||The ideal living in Norway/Scandinavia||Escape from everyday life||Environmental concerns|
|Pull factor||Push factor||Push and pull factor|
|Theme 1||The beauty of the Geiranger fjord||Folk museum and Trolls||The idealization of the Norwegian society||Houses and gardens near the fjord||Untouched and unpolluted nature|
|Theme 2||The scenery views||Historical churches||Popular sports in Norway||Visit towns in Geiranger||Concern regarding China's comparable tourism|
|Theme 3||The challenging roads||Communicating with local people|
|Theme 4||Geographical formation of Geiranger||Economy and fish export|
Below we present these constructs in more detail, together with representative quotations of each construct.
The beauty of nature:
For Chinese tourists, the main motivation behind their visit to Norway and Northern Europe is the nature. Beautiful natural resources, such as Geiranger fjord, act as a main pull factor in this respect. They expressed a taste for the beauty of the fjord’s water and waterfalls. Visitors are also amazed by the natural attractions and mountain roads near the Geiranger fjord – not only the nature, but also the work of manpower and engineering, as indicated by the following quotes:
“Fjord combines the most spectacular and unique natural landscape”
“Geiranger, which is presented by Geography Magazine as the best untouched natural sanctuary, is mysterious and beautiful”
“The Ørnevegen is known for its successive hairpin bends on the road between Geiranger and Eidsdal”
Similar to what has been found in other tourism studies (Kim, et al. 2003), historic and cultural resources in Geiranger and Norway can be regarded as a type of factor that ‘pulls’ Chinese tourists and brings most of them into contact with the particular culture and tradition of Norway, which they have only read about or seen in the media before. The town tour has also satisfied their desire for knowledge of Scandinavian culture and history. While people “feel sad about the fire that occurred in Geiranger church before”, they are also impressed by the harmony between mankind and the nature reflected in the Geiranger Folk Museum. The following quote illustrates the role of cultural tourism.
“The childish painting on the wall looks like graffiti – people admire the simplicity of nature”
Another tourist added:
“The troll is a symbol for civilians and the freedom of their emotions, it’s the cutest thing about them”
The ideal living in Norway and Scandinavia:
A significant factor attracting Chinese tourists is that Norway is ranked as the country in the world with the highest standard of living. Many Chinese people are now aware that Norway has been ranked as the country with the highest Human Development Index for most years since 2001. Symbols of “a most livable country and the most highly civilized in the world” are a source of much novelty to Chinese tourists verifying what they have heard about typical lifestyles in Norway. Nearly all blogs about Geiranger discuss summer vacations; participation in, for example, cycling and hiking reminds tourists of the host city for the 1994 Winter Olympics, Lillehammer. One of the tourists stated the following about Norway’s wealth:
“Norway has oil and is the richest country in Europe.”
While another noted Norway’s high score on Human Development Index:
“Norway has also ranked # 1 in the Human Development Index for the majority of years since 2001.”
Escape from everyday routine:
A typical push factor and a construct behind tourist motivation that is found is Escape from everyday routine – that is, a tourist may want to make a trip to escape from his/her personal or interpersonal environment and to seek out psychological (intrinsic) rewards in the personal or interpersonal dimension (Kim et al. 2003). For most Chinese urban residents, most places in Norway look like towns and villages and represent the leisure and quiet life they are dream of and the ideal travel destinations they discuss with families and friends on “WeChat” app daily. It is for this reason that they visit several towns in Geiranger and take pleasure in residential houses and gardens near the fjord. This is exemplified by the following quotes:
“We spend time on the Geiranger shopping street and are attracted by the souvenirs.”
“We travel through Valldal and tasted one of the most famous and best strawberries in Norway.”
Tourism and environmental concern:
The construct of environment is a pull-and-push characteristic. The beautiful, clean Geiranger fjord will gain a good deal of admiration and attract more tourists than an environmentally unsustainable destination. Simultaneously, the nature and well-managed tourism industry in Norway also raise much concern about the protection of the environment and the conservation of tourism attractions among Chinese tourists. For example, in some blogs they express themselves in sarcastic and somewhat angry terms when they imagine how the Chinese local government would act towards their profit targets if Geiranger were in China. Encountering the simple-looking, yet excellent management of the Norwegian scenic regions, tourists are also reminded of the environmental issues and over-development problems in China’s scenic spots. The following quote portrays such a sarcastic comparison:
This study set out to investigate the experience of Chinese tourists visiting the Norwegian tourist destination of Geiranger. Using travel blogs collected from a Chinese website, we have analyzed the tourists’ self-reported experiences and the motivation behind their trips. The analysis used a mixed-method approach by combining quantitative and qualitative techniques.
The quantitative analysis used key sentiment words collected from 196 blog posts. The analysis shows that the vast majority of Chinese tourists used positive keywords when describing their visit to Geiranger. This result is in line with previous evidence about Chinese tourists to Norway (Shi, 2015; Zhang, 2016).
The qualitative analysis – applying the grounded theory method – identifies five constructs that represent the tourists’ motivation for travel to Geiranger. Three constructs are pull factors, one is a push factor, and one represents both a push and a pull factor. The most novel factor identified as a main pull factor behind their visits is Chinese tourists’ interest in the Norwegian/Scandinavian way of living – a way of living and culture regarded as almost the ideal way of living. The main push factor is the possibilities for relaxation and escaping from the high tempo of everyday life.
This study is the first study focusing on Chinese tourists’ experience of visiting Norwegian fjords using data from web blogs (with the authors’ knowledge) and has gone some way towards enhancing our understanding of the travel experiences and motivations of Chinese-speaking visitors, which might be regarded as one of the most booming international tourist groups to Norway in respect of the statistics of the last decade. Gaining more insight into these visitors’ preferences and satisfaction could help Norwegian tourism planners in offering marketing and services that better match the tourists’ particular preferences.
While negative emotions regarding weather, meals, and transport seem negligibly small, the emphasis potentially placed on such areas, such as more useful information for meals and preparation for the weather as well as travel and transportation during the marketing process (e.g. www.visitnorway.no) may bring about additional positive effects for the tourism industry in Norway.
Admittedly, there are also limitations to our study. Spelling errors exist in some of the blogs, especially for Norwegian location names, although this appears to a relatively low extent. It is possible that only the positive experiences were documented in the blog posts, which may have resulted in a self-selection bias in our data source. In the sentiment analysis, we have used a relatively simple approach, while other more advanced alternative approaches are possible for analyzing blog posts. For example, sentiment analysis based on machine learning (Li, et al. 2018) could be applied to analyze Natural Language Processing (NLP). This approach is more technical, but could save the time and effort of the detailed reading of the text. However, the treatment of Chinese texts seems rather more difficult than English with respect to its segmentation characteristics. Finally, the blogs include numerous pictures, which could be used to analyze the tourists’ experience further.