An Analysis on Needs for National Parks by Age Group using Semantic Network Analysis
Article information
Abstract
Background and objective
National parks are places people from all walks of life can utilize. However, many existing studies are limited in suggesting management plans that fit the characteristics of different age groups of people visiting national parks. Therefore, this study aims to identify the requirements for national parks and set the direction for national park services by age group.
Methods
This study collected the requirements for national parks from 902 people of different ages through an online self-administered questionnaire and analyzed them through semantic link network analysis.
Results
As a result of the analysis, there were management requirements in two areas related to the environment (rubbish and ecological environment, facilities environment such as safety and forest fires) and one area related to the economy, and a monitoring system for rubbish and ecological environment and a private land management policy for the economy were identified as alternatives. The younger generation was interested in environmental issues, the older generation was interested in economic issues, and the elderly was interested in the facilities environment for safety and forest fires. Therefore, it was found that environmental campaigns should be conducted for the younger generation, admission fees and parking rates should be improved for the older generation, and a safe tour environment should be provided for the elderly.
Conclusion
The needs analysis found that it is necessary to establish a monitoring system for all national parks from an environmental perspective and improve the management policy about private land from an economic perspective. Furthermore, environmental campaigns and events were required for the younger generation, benefits related to entrance fees for the elderly, and policies to create a safe facilities environment for the elderly. Establishing such national park policies will improve the satisfaction of users.
Introduction
National parks are established and managed under a system that is implemented worldwide to promote biodiversity and conserve the environment, and national parks are also designated throughout South Korea. Moreover, national parks provide not only nature tourism services but also ecotourism and environmental education for people. Accordingly, national parks offer services for both nature and humans. These national parks are operated using taxes paid by citizens and thus must meet their needs and expectations, such as conserving the natural ecosystem and providing safe visitor services, which are the initial purposes of designation (Suckall et al., 2009). Therefore, it is necessary to summarize the actual requirements of citizens and provide suitable national park services.
There have been studies mostly summarizing user requirements regarding national parks through surveys. One study conducted a survey on residents living near national parks in downtown areas, asking about which ecological plans they prefer, and proposed plans that meet their requirements such as birds and fruit trees (Xie et al., 2022). Another study identified the safety requirements of users in national parks of mountainous areas through a survey and confirmed that providing information related to safety such as signs is a prerequisite for safety management (Gstaettner et al., 2022). There was also a study that surveyed preferences for national parks based on photographs collected from various social classes and set the design direction for national parks considering the characteristics of each social class (Suckall et al., 2009). Another study concluded through a nationwide survey on national parks that plans to build a golf course or a hotel in a national park are not appropriate for maintaining good landscape (Arslan and Kaymaz, 2020). However, studies identifying user requirements through surveys have limitations as they cover only the requirements for items selected by the researchers.
Accordingly, there are studies summarizing user requirements in national parks via social networking service (SNS). These studies set the direction for national parks preferred by the public through text mining analysis on language used on SNS. For example, there were preferences for aesthetic elements (Huai and Van de Voorde, 2022; Jianrong and Zhenbin, 2022), link between the waterfront and road, park size, and various other items (Kong et al., 2022). Pictures on SNS were also analyzed to confirm that local residents mostly respond to plants and animals, whereas tourists respond to visual elements such as sculptures and buildings. However, the approach using SNS has limitations in that research is conducted only on social classes that have access to SNS (Huai et al., 2022).
In fact, national parks are for everyone; they are for all generations from children to the elderly to meet and get along. Accordingly, Korea Forest Welfare Institute under Korea Forest Service is providing forest welfare services according to the lifecycle, but no such system is established for national parks. Thus, it is necessary to identify requirements of all kinds of people and generations according to the lifecycle (Deng et al., 2003; Suckall et al., 2009). To keep national parks that are operated by the taxes paid by citizens from becoming the preserve of a certain privileged class, it is necessary to identify the characteristics of each class having trouble using national parks and sort out their difficulties and requests. Moreover, since there are study results that national parks have greater effect when visited by the socially underprivileged (Romagosa, 2018), it is more important to identify whether the underprivileged or disadvantaged classes facing difficulties in activity have any trouble visiting national parks. Therefore, there must be research analyzing the requirements of each class based on the experience of using national parks among all classes. Accordingly, this study received free-form responses about the requirements for national parks that 902 citizens of all classes have in South Korea. Text mining was performed to identify the requirements for national parks in each age group, based on which we came up with customized national park service plans for each class.
Research Methods
Participants
This study conducted an open-ended survey on 902 participants that have visited national parks in March 2022. The participants were selected evenly among different local governments, age groups, and genders. The survey required the participants to provide free-form answers to the direction, “Write down freely what to change or supplement in national park management in the future.” This is an open-ended survey method in which the researchers do not specifically limit the survey items but ask participants to provide diverse opinions about what they want or their inconveniences in using national parks. An online survey was used to conduct a nationwide survey in a short period of time, and as a result, there were relatively fewer participants from the elderly population since they are not familiar with online surveys (see Tables 1 and 2).
Research Methods
This study used a semantic network analysis to analyze the open-ended survey on what the participants want and their inconveniences in using national parks. Semantic network analysis is an analytical technique that expresses the relationship between words used together in a sentence or a paragraph as a network. A network is built by calculating the frequency in which words are used together, and the degree to which one word is used with another is expressed as centrality, thereby showing the status and role of certain keywords within a message. This quantifies and visualizes the key message through keywords and the structure of the message through the relationship between words (Doerfel and Barnett, 1999; Drieger, 2013). Thus, it has the benefit of being able to quantify, objectify, and visualize qualitative data that had been relatively dependent on the researcher’s subjectivity, such as content analysis. Accordingly, semantic network analysis is frequently used in all kinds of media and social sciences, and recently in the environment and forestry as a method of analyzing resident opinions (Lee, 2019; Lee et al., 2022).
This study attempted a cluster analysis based on betweenness centrality. Betweenness centrality is a measure that represents the effect of a given node (meaning) on network information diffusion (Barthelemy, 2004). In other words, it is a method for analyzing influential keywords, and thus is good for checking the status of keywords within many messages (Newman, 2005) (see Fig. 1).

Example of a node with high betweenness centrality (v has only 2 links, but nodes from both regions are connected and have the highest betweenness centrality) (Barthelemy, 2004, p.163).
Moreover, we attempted a cluster analysis between words that appeared to be closely related based on this betweenness centrality. Cluster analysis based on semantic network analysis organizes words into key semantic clusters, thereby more concisely organizing the message structure (Newman, 2005). Finally, we comparatively analyzed how the semantic clusters that appeared are used differently depending on the generation. For analysis, we used Netminer 4.3 that provides Hangeul-based morpheme analysis and semantic network analysis, and preprocessing was performed by excluding stop words and integrating similar words. Overall, we used simple frequency and frequency of words that commonly occurred in the text provided by one person. The detailed research process is as follows.
First, we attempted a morpheme analysis. We divided the requests from all survey participants into morphemes, the smallest meaningful unit. Second, major keywords were selected. We extracted nouns with independent meanings. Moreover, we assumed that the words that were derived relatively frequently have strong information power (Luhn, 1958) and selected only the top 10% nouns as major keywords. Third, we identified the co-occurrence frequency between words used together in a single person’s request, based on which we created a network of relationships between keywords and analyzed betweenness centrality. Fourth, we conducted a betweenness centrality-based cluster analysis of words based on the network analysis between words. This was to create clusters of words that are frequently used together and identify the clusters of major discussions. Fifth, we interpreted the results of analysis by comparing with actual requirements (see Fig. 2).

Research process (Revised to incorporate Lee, 2019 and Lee et al., 2022).
Results and Discussion
Betweenness centrality-based cluster analysis
As a result of conducting betweenness centrality-based cluster analysis, we came up with three clusters as the most appropriate clusters (Best Cut 9.758). In the first cluster, there were many words related to environmental management and ecological experience such as ‘rubbish’, ‘damage’, ‘ecology’, ‘reservation’, and ‘experience’. Regarding rubbish, participants mostly wanted waste sorting and recycling as well as a monitoring system to manage rubbish and smoking. There were also requests for measures to prevent ecological damage and increase fairness in reservation of ecological experience programs. In the second cluster, there were many words requesting management in the economic or financial aspect such as ‘ticket price’, ‘temple’, ‘collection’, ‘parking’, and ‘tourist’. There were arguments that it is unfair to impose temple ticket prices on those not visiting the temple, and that the parking rates are too expensive. In the third cluster, there were words related to facilities environment such as ‘safety’, ‘mountain climbing’, ‘forest fires’, ‘prevention’, and ‘danger’, requesting a safe landscape for the elderly and safety measures for forest fires (see Fig. 3 and Table 3).

Betweenness Cluster analysis (Circle size is the frequency of occurrence, line thickness is common frequency of occurrence).
Comparatively analysis of frequency and betweenness centrality
We created a matrix to compare simple frequency and betweenness centrality of keywords. Words with high betweenness centrality have substantial meaning as they are placed between different words and play a mediating role in network formation (Watts and Strogatz, 1998). This is because, although they do not appear frequently, they connect nodes in different areas and are likely to become an integrated solution. Therefore, in this study, we created a matrix of frequency and betweenness centrality and examined words (frequency 40 or lower, betweenness centrality 0.2 or higher) in the first quadrant (low frequency and high betweenness centrality) based on means. As a result, four keywords such as speculation (betweenness centrality 0.271), monitoring (betweenness centrality 0.252), private land (betweenness centrality 0.293), and tourist (betweenness centrality 0.301) showed high centrality compared to frequency (see Fig. 3 and Table 3). ‘Speculation’ and ‘monitoring’ were between ‘rubbish’ and ‘ecological damage’, showing that the problems of rubbish and ecological damage can be managed comprehensively through monitoring (monitoring and fines). In addition, ‘private land’ and ‘tourist’ were between ‘ticket price’ and ‘rubbish’, showing that improving tourist policies about private land (temple ticket price and rubbish management policies) can be a solution to these two problems. On the other hand, ‘temple (frequency 38)’ showed relatively higher frequency, which showed that although temple ticket price is a huge problem, it is an independent problem that is not related to other problems (see Fig. 4 and Table 4).
Ratio analysis by age group
We checked the distribution of the clusters by age group. The results showed that the 20s had many requests for management of rubbish and ecological environment (20s: rubbish and ecological environment 66%), followed by requests for economic management and environmental management regarding safety and forest fires. This distribution showed that older participants had relatively more requests for economic management and environmental management regarding safety and forest fires (60s: economy 41%, environmental management regarding safety and forest fires 41%). This trend clearly reveals the difference between the young generation interested in resource circulation and ecological environment, the older generation sensitive to economy, and the elderly aware of the need for safety and cautious about forest fires. Based on the above, it is necessary to promote campaigns related to environmental management for the young generation, provide clear solutions for private land management for the older generation, and offer safe national park facilities for the elderly (see Fig. 5 and Table 5).
Result and Discussion
Interpretation and implications
Previous studies on the demands for national parks have already identified the public needs for environmental conservation such as rubbish, ecology, and forest fires (Deng et al., 2003) and safe facilities for the elderly and the infirm (Gstaettner et al., 2022). However, the emphasis on economic elements related to temple ticket prices was a new discovery made by this study, which is unique to South Korea. This shows that coming up with an alternative to this issue is a prerequisite for national parks. Previous studies revealed that there are severe conflicts over collection of entrance fees to cultural heritage for temples since the tickets for national parks became free of charge (Chang, 2008; Park, 2008; Park, 2011). The results of this study confirmed that adequate alternatives to such entrance fees to cultural heritage have not been provided until 2022, implying that this is an urgent matter to consider for national park policies in South Korea.
Moreover, while many international studies have attached importance to the needs for leisure in national parks (Arslan and Kaymaz, 2020; Suckall et al., 2009), this was not found as much in this study. Of course, there were needs for events and programs in terms of management of the ecological environment, but the fact that there were not many requests for diverse programs and education (Arslan and Kaymaz, 2020), landscape (Huai and Van de Voorde, 2022), and health effect (Romagosa, 2018) suggests that people in South Korea still perceive national parks merely as a place for mountain climbing, rather than for leisure. Accordingly, it is necessary to implement policies for users as well in order for national parks to become a place that offers various landscape resources, programs, training facilities, and health effect beyond just a place for mountain climbing.
Furthermore, previous studies mostly classified the users of national parks into neighborhood residents and tourists (Huai et al, 2022). Of course, some studies claimed that it is necessary to enable various age groups (Arslan and Kaymaz, 2020), income brackets, and races (Suckall et al., 2009) to use national parks, but they failed to provide which policies are necessary for them. Accordingly, this study has significance in that it classified the participants into different age groups and identified their requirements, thereby confirming that the young generation is interested in the environment and thus needs campaigns and education on environmental protection, and that the elderly is more interested in safety and thus must be provided with safer forest tourism services. Of course, this study has limitations as it failed to include infants and children who could not participate in the survey, and also could not survey more diverse generations including persons with disabilities or low-income groups that are the underprivileged and disadvantaged people of society. Continuous research is needed to listen to the voices of various classes, since national parks are for everyone.
Conclusion
This study used text mining to identify the requirements for national parks. As a result, management was requested at two levels: environment (‘ecological environment’ and ‘facilities environment’) and ‘economy (temple ticket price)’. There were issues about rubbish and environmental damage regarding management of ecological environment, and strengthening of the surveillance system such as monitoring was discussed as a solution. For facilities environment, there were issues such as safety for the elderly, contagious diseases, and fires. There were issues about temple ticket prices and parking rates regarding economic management, which required improvement of private land policies. Moreover, the young generation had many requests for management of the ecological environment, while older participants had more requests for economic management and facilities management. In other words, it seems that policy alternatives to issues regarding the ecological environment would be more effective for the young generation, whereas policy alternatives about economic management and facilities measures would be more effective for the elderly. This study has limitations as it failed to include teenagers and younger participants in conducting the open-ended survey. Further research can be conducted on ways to provide systematized national park services for each age group by gathering the opinions of children and teenagers and comparatively analyzing the requirements of more diverse classes.