J. People Plants Environ Search


J. People Plants Environ > Volume 26(6); 2023 > Article
Kim, Lee, Park, and Lee: An Analysis of Visual Preference and Characteristics through a Woodland Garden Perception Survey


Background and objective: Woodland gardens, or gardens designed to emulate the aesthetic and environmental characteristics of forest vegetation, can assist in resolving various environmental and societal problems. However, research regarding woodland gardens in a Korean context has been insufficient, as the concept has only recently been introduced to Korea. This study was conducted to investigate the Korean public’s perceptions of two types of woodland gardens and to suggest future directions for the popularization of woodland gardens in Korea.
Methods: This study was carried out based on a survey. The respondents were provided with a questionnaire on three sets of images representing two types of woodland garden (naturalistic and urban) and a typical formal garden. The questionnaire was designed based on the semantic differential method to gather information on the general public’s recognition of the concept of woodland gardens, and their preference and perception of each type of garden. The collected data were analyzed using SPSS Statistics 29.
Results: The results suggest that the majority of the respondents were familiar with the term “woodland garden.” Of the types of gardens, preference for naturalistic woodland gardens (NWGs) was the highest. Landscape adjectives associated with each type of garden were natural, harmonious, comfortable and simple for NWGs, and organized, artificial, simple, poor, and comfortable for urban woodland gardens (UWGs). A factor analysis on the adjectives identified four factors for each type. Of the identified factors, comfort and stability were found to have particularly greater effects on preference.
Conclusion: Preference for NWGs was greater than UWGs, due to the perception that they provided greater comfort and stability. It is expected that designing woodland gardens based on the finding of this study would aid in the popularization of woodland gardens, and the advancement of garden culture in Korea.


As social problems caused by the climate crisis and the pandemic worsen, the importance of green spaces is increasing. As a result, interest in gardens, one of the most accessible green spaces in our daily life, continues to grow. Since gardens can accommodate not only social and cultural functions that can improve people’s quality of life (e.g., provision of relaxation and job creation), but also ecological functions (e.g., biodiversity conservation and carbon absorption), they can contribute to solving a range of social and environmental issues that occur around the world (Park et al., 2017). To respond to the growing interest in gardens and utilize the functionality of gardens to address various social and environmental problems, the Korea Forest Service has established a “Garden Promotion Basic Plan” to spread a garden culture and has promoted a range of policies in this area, and many local governments have sought to activate a garden culture by creating gardens. In South Korea, social interest in gardens grew after the creation of the Suncheonman National Garden, and so research on gardens began to rapidly increase in the 2010s (Korea Forest Service, 2021; Lee, 2021; Park, 2021). According to Park (2021), there were 394 research papers on gardens from 1971–2020, most of which focused on topics such as analyses, planning and design of gardens, and garden theory. In comparison, it seems that research on garden styles and types, and perceptions of and satisfaction with gardens is still lacking.
Meanwhile, woodland gardens are a type of garden that appeared in the British Empire in the late 19th century when shade tolerant shrubs and small trees with colorful flowers introduced from India and China, including Rhododendron spp., Magnolia spp., and Camellia spp., began to be planted in the understory of tree plantations, arboretums, and other wooded areas(Elliott, 2007; Cox, 2018). In the early parts of its history, planting designs of woodland gardens were focused on providing color with shade-tolerant bulbs and flowering woody plants that bloom in spring and early summer. These days, however, the planting design has evolved to emphasize textures and forms, providing year-round interests which is achieved by the greater use of shade-tolerant perennials and pteridophytes, and utilization of native species (Darke, 2002; Wiley, 2014). This is caused in part by the newly arisen need to maintain ornamental value of the gardens throughout the year due to changes in usage patterns that have resulted from the publicization of gardens (i.e., donation of former private gardens to National Trust; Wiley, 2014), and in part by the influence of naturalism and environmentalism. Since woodland gardens typically possess a forest-like, stratified vegetation structure where woody plants take a prominent role, they can provide ecological and environmental benefits akin to those provided by forests, including cooling effects through transpiration and shading, and sequestration of carbon through the growth of woody plants and accumulation of soil organic matter. As woodland gardens can simultaneously provide both the aforementioned forest-like environmental services and the aesthetic and social services of the gardens, they can better contribute to solving global social and environmental issues, such as responding to climate change, than other types of gardens (Cameron and Hitchmough, 2016; Credit Valley Conservation, 2012; Hong and Lee, 2020). In addition, with increased labor and material costs, curtailed green space management budgets, and increasing regulations on crop protection agents and carbon-emitting equipment caused by changes in public perception of the environment, the means available for green space management are increasingly becoming limited worldwide (Cameron and Hitchmough, 2016). Garden types that adopt planting designs based on natural vegetation, such as woodland gardens, can provide an answer for these issues, as by applying ecological processes, including succession, the manpower and material input required for the maintenance of the gardens can be decreased (Cameron and Hitchmough, 2016; Hitchmough and Dunnett, 2008). Furthermore, as 62.6% of South Korea’s land area consisted of forests as of 2020 (Statistics Korea, 2020), the economic use of forests is important. The creation of woodland gardens will contribute to this aspect.
In South Korea, research related to woodland gardens began in the late 2010s, when research on gardens rapidly increased. The Korean Institute of Garden Design (2019) and Hong and Lee (2020) reviewed concepts related to forest gardens and used the Delphi technique with 12 experts to derive the values of forests and gardens: formal, scenic, ecological, historical/cultural, and useful values. Based on this, they established the definition and concept of woodland gardens in Korean context. Using the five values, woodland gardens were also categorized into four types: landscape/recreation, ecological preservation, cultural tourism, and research/learning (Korean Institute of Garden Design, 2019). Cho (2020) suggested that a brand and sense of place can be established by creating woodland gardens in a resort that stimulate various senses focused on wellness. CNU R&BD Foundation (2020) categorized the requirements for meeting woodland garden standards, and presented guidelines for creating woodland gardens according to environmental characteristics through an analysis of woodland garden examples in South Korea. Based on the five values of woodland gardens presented by Hong and Lee (2020), Hong et al. (2021) developed an evaluation index to distinguish whether a specific garden is a woodland garden, allowing us to evaluate various types of woodland gardens. To lay a foundation for creating woodland gardens with ecological stability, Nature and Forest Research Institute (2021) analyzed the composition and structure of the plant community dominated by each species, and the environmental characteristics of the sites where the communities of each species developed, targeting 35 representative and native woody forest species. Lee et al. (2021) summarized and organized the physiological and ecological characteristics of sciophytes in the context of their use in woodland gardens, and based on this, presented methods for planting and cultivating sciophytes. Industry-University Joint Research Institute of Gachon University (2022) categorized forest communities in South Korea, derived core landscape elements of forests, and proposed methods to apply them to woodland gardens. The types of woodland gardens presented by the Korean Institute of Garden Design (2019) were simplified and divided into two types: a natural type that emphasizes the attributes of ecological conservation, and research/learning; and an urban type that emphasizes those of landscape/recreation and cultural tourism. Naturalistic woodland gardens are a type that is created and managed indirectly based on ecological and dendrological methodologies, while Urban woodland gardens are focused more on meeting users’ aesthetic and landscape needs, created and managed directly based on landscape architecture and horticultural methodologies. Naturalistic woodland gardens are modified existing forests, while Urban woodland gardens can be created from either existing forests or bare ground where forests did not exist (Industry-University Joint Research Institute of Gachon University, 2022).
As such, it could be said that previous research on woodland gardens conducted in Korea has been focused on the creation of woodland gardens, including concept establishment, guideline development, typification/classification, and planting characteristics. It seems that the aim of such research was to establish a theoretical foundation for introducing a new concept of gardens called woodland gardens to the country. To promote a garden culture based on woodland gardens, it is necessary to understand the public’s demand for them, but since their concept was only recently introduced to South Korea, relevant research is still lacking. If woodland gardens are developed and promoted to meet public needs by determining the public’s perception of them, it is expected that they will contribute to the spread of garden culture, as well as to solving social and environmental issues occurring at home and abroad. Therefore, this study aimed to understand the public’s perception of woodland gardens, and investigate the public preference for each type of woodland garden through survey method.

Research Methods

Research scope

This study was conducted in the form of a survey based on example images of woodland gardens to examine the general public’s perception of different types of woodland gardens. Example images of woodland gardens were presented to those being surveyed by being divided into two types of natural and urban woodland gardens, based on research by the Industry-University Joint Research Institute of Gachon University (2022). Images of formal gardens were also presented to those being surveyed to determine whether they could cognitively distinguish woodland gardens from general formal gardens. As such, preference and image analyses were conducted for three types of natural and urban woodland gardens, and typical formal gardens. Moreover, to derive trait factors of woodland garden images, we derived factors for each type of woodland garden, excluding formal gardens, and examined whether the derived factors affected preference.


This survey was conducted as a face-to-face survey over 5 days from September 19–24, 2023, targeting the general public who visited the “30th Native Flower Exhibition” at the Korea National Arboretum. The questionnaire was distributed by setting up a separate booth at an experience event space in the exhibition, and respondents were allowed to choose between two forms: a printed-out paper-and-pencil type and an online QR code type. Of a total of 700 responses collected, 670 were analyzed after excluding 30 that were answered insincerely, such as by repeating the same answers. In the survey, separately printed-out example images of each type of woodland garden were also presented to the respondents together with the questionnaire. As examples of naturalistic woodland gardens, images were presented of Gwangneung Forest Trail Garden, and Rhododendron Garden and Moss Garden (sections of Jade Garden), which were improved by adopting relatively indirect methods such as thinning and supplemental planting while maintaining pre-existing forests. As examples of urban woodland gardens, images were presented of the courtyard gardens of Monoha Hannam and Amore Seongsu, where trees and sciophytes were planted on a site where no forest previously existed, using direct methods of landscaping and horticultural techniques including soil replacement and introduction of trees. Moreover, images of The Butchart Gardens, Longwood Gardens, Seomi Garden, and a garden model of the 2018 International Flower Expo, which have formal planting designs based on heliophytes as the main plant materials, were presented as examples of typical formal gardens (Table 1).
The survey items consisted of recognition of woodland gardens (Recognition), preference for each type of woodland garden (Preference), perception of woodland gardens (Perception), and demographic characteristics of respondents (DCR; Table 2). For the Recognition items, a nominal scale was used, and for the Preferences items, a 5-point Likert scale was used. For the Recognition items, the semantic differential scale (SDS) was used, which is a rating scale designed by Osgood in 1959 that measures respondents’ opinions with implied meaning towards an object by presenting a set of bipolar adjectives at opposite ends of a continuum (Shin et al., 2014). SDS, a scale commonly used in image analysis, is utilized to evaluate preference, and derives various characteristics by analyzing feelings about images based on adjectives (Joo and Im, 2003; Lee, 2016; Roh and Son, 2022). In research on gardens or arboretums, adjectives for landscapes have been used to be analyzed, mainly focusing on topics such as preference, perception, distinction of garden types, and adjective classification, but they haven’t been used in research on woodland gardens. Accordingly, in this study, 24 pairs of landscape adjectives were derived by referring to previous studies related to gardens or arboretums that used such adjectives (Roh and Son, 2022; Yum and Joo, 2017; Kim and Cho, 2017; Ham, 2020). Adjectives that were overlapping or had similar meanings in the derived adjective pairs were modified. Based on a preliminary survey of 25 majors in forestry and science of landscape architecture, a total of 14 pairs were finally selected after pairs that did not represent landscape characteristics for each type of garden were excluded.

Analysis method

The data collected through the survey were analyzed using SPSS Statistics version 29. Analysis techniques included a frequency analysis for DCR and Recognition, and descriptive statistics for Preference. Moreover, chi-square and independent samples t tests were used to examine differences in Recognition and Preference according to gender and age. For Perception, descriptive statistics were first gathered to understand the extent of respondents’ feelings about each type of woodland gardens. To determine the trait factors of woodland garden images, a reliability test was also conducted on the survey items, and then their validity was verified and trait factors derived through an exploratory factor analysis (EFA). Lastly, using factor scores derived through the EFA as independent variables, and Preference as dependent variables, a multiple regression analysis was conducted to determine the effect of woodland garden trait factors on Preference.

Results and Discussion

Demographic characteristics of respondents (DCR)

A frequency analysis was conducted to determine the demographic characteristics of respondents (DCR) to this survey. The total of 670 respondents broke down by gender into 473 females (70.6%) and 197 males (29.4%). By age, respondents in their 60s or older were the largest group with 202 people (30.1%), followed by respondents in their 40s with 167 (24.9%), respondents in their 50s with 162 (24.2%), respondents in their 30s with 91 (13.6%), respondents in their 20s with 36 (5.4%), and 12 teenagers (1.8%). For residence, the majority or 430 people (64.2%) lived in Gyeonggido, followed by Seoul with 191 people (28.5%) and Incheon with 28 people (4.2%). For occupation, ‘housewife’ was the most common with 187 people (27.9%), followed by ‘office worker’ with 156 (23.3%) and ‘professional’ with 149 (22.2%; Table 3). To sum up, the respondents were mostly female, in their 60s or older, and housewives, with the highest proportion of residents in the Seoul and Gyeonggido regions. The DCR was very similar to the characteristics of arboretum visitors derived in the study by Yoo et al. (2020).

Recognition of and preference for woodland gardens

For items on recognition of woodland gardens (Recognition), 640 responses were analyzed, excluding 30 with no response to the items. When respondents were asked if they had ever heard of woodland gardens, most responded that they “had heard of them (Yes; 80.0%).” For Recognition, by gender, 21.4% of males and 58.6% of females answered “Yes” (Table 4). By age, among those who answered “Yes,” the largest group was in their 60s or older at 24.4%, followed by those in their 50s (20.5%), and those in their 40s (19.7%; Table 5).
Looking at type preferences for woodland gardens (Preference), Preference was highest for type A (4.33 points), which is an example of a naturalistic woodland garden (NWG). This was followed by type C (4.14 points), which is an example of a typical formal garden, and type B (3.65 points), which is an example of an urban woodland garden (UWG). Furthermore, there was a statistically significant difference in preference for UWGs depending on gender (Table 6).
In terms of Recognition, the survey results found that the public’s recognition of woodland gardens was high. However, since no survey on the level of Recognition was conducted, it seems that there may be a difference in the level even if Recognition was high. In addition, of the subtypes of woodland gardens, the preference for NWGs was higher than that for UWGs, suggesting that the public’s preference for a type of garden that maintains the structure of a natural forest is higher.

Image analysis for each type of woodland garden

By presenting respondents with example images of each type of woodland garden and surveying them with 14 pairs of landscape adjectives on a semantic differential scale (SDS), the results were as follows. For Type A, the adjective pair artificial-natural was rated highest (4.15 points), followed by inharmonious-harmonious (3.90), whereas the pair comfortable-uncomfortable (2.34) was rated lowest, followed by plain-fancy (2.68). For Type B, traditional-modernistic (3.66 points) was rated highest, followed by untidy-tidy (3.63), whereas artificial-natural (2.68) was rated lowest, followed by simple-complex (2.80), poor-rich (2.80), and comfortable-uncomfortable (2.80). For type C, Dark-Light (4.48 points) was rated highest, followed by plain-fancy (4.22), while comfortable-uncomfortable (2.64) was rated lowest, followed by artificial-natural (2.95; Fig. 1).
Based on this, it was found that respondents perceived NWGs (Type A) as natural, harmonious, comfortable, and plain gardens. They perceived UWGs (Type B) as tidy, but artificial, simple, poor, and comfortable ones, while perceiving formal gardens (Type C) as light and fancy, but comfortable and artificial ones.

Factor analysis of woodland garden images

An exploratory factor analysis was conducted to determine what factors woodland gardens are composed of using 14 pairs of landscape adjectives. To this end, above all, by estimating the reliability of survey items, the Cronbach’s alpha of Types A and B was found to be 0.620 and 0.759, respectively, indicating acceptable reliability. Based on this, by conducting a principal component analysis of 14 pairs of landscape adjectives, trait factors for each type were extracted by rotating them using the varimax method.
For Type A, by examining the commonality of each variable through a factor analysis, one question with a commonality value of less than 0.4 was discarded. Kaiser-Meyer-Olkin (KMO) and Bartlett’s test of sphericity resulted in 0.803 and =1649.890 (p < .001), respectively, indicating good validity. Through this analysis, four factors were extracted from Type A. Factor 1 was named the comfort factor as it appeared as untidy-tidy, messy-clean, inharmonious-harmonious, cold-warm, dark-light, and closed-open. Factor 2 was named the orderliness factor as it was composed of informal-formal and plain-fancy. Factor 3 was named the stability factor, consisting of static-dynamic and comfortable-uncomfortable. Factor 4 was named the spatiality factor, consisting of simple-complex, poor-rich, and artificial-natural (Table 7).
For Type B, by examining the commonality of each variable through a factor analysis, it was found that there were no questions with a commonality value of less than 0.4 or less. KMO and Bartlett’s test of sphericity resulted in 0.841 and =2445.390 (p < .001), respectively, indicating good validity. Through this analysis, four factors were extracted from Type B. Factor 1 was named the comfort factor, consisting of closed-open, cold-warm, inharmonious-harmonious, artificial-natural, untidy-tidy, messy-clean, and dark-light. Factor 2 was named the modernity factor, consisting of traditional-modernistic and informal-formal. Factor 3 was named the diversity factor, consisting of simple-complex, plain-fancy, and poor-rich. Factor 4 was named the stability factor, consisting of static-dynamic and comfortable-uncomfortable (Table 8).
In the factor analysis of each type of woodland garden, NWGs were found to have factors of comfort, orderliness, stability, and spatiality, whereas UWGs were found to have factors of comfort, modernity, diversity, and stability. For woodland gardens, nature’s ecological orderliness and stability should be cultivated by planting various plants including trees, shrubs, and herbs in a stratified structure like forests. Since for their purpose, they should also provide a comfort space for people by reproducing the beauty of forests in a refined form, they seemed to be analyzed to have the above factors.

Preference analysis of each factors of woodland gardens

A multiple regression analysis was conducted to determine whether the traits factors of woodland gardens, which were rated by respondents and derived through a factor analysis, affected their preferences for each type of woodland garden (Preference).
By performing a multiple regression analysis of the factors of and preferences for NWGs, they were found to be significant at F=17.652 (p < .001), but adj.R2 = .102, showing a low explanatory power of 10.2%. Toya and Skidmore (2007) suggested that when considering social and economic variables, the coefficient of determination (R2) indicating explanatory power becomes 0.30 or less, and Falk and Miller (1992) proposed that if it is 0.10 or more, it has explanatory power. In addition, in the field of social science, Cohen’s guidelines (1988) are used, and an adjusted coefficient of determination (adj.R2) of 0.02 is considered as a small effect size, while one of 0.13 is a medium effect size, and that of 0.26 or more as a large effect size. On this basis, since the multiple regression analysis of factors of and preferences for NWGs resulted in adj.R2= 0.10, according to Cohen’s guidelines (1988), it is greater than 0.02 and close to 0.13, indicating more than a small effect size. As it fits the guidelines suggested by Falk and Miller (1992), it is considered to be significant, although it has low explanatory power. Accordingly, the regression equation derived through multiple regression analysis of NWGs is as follows.
Y(Preference for naturalisitc woodland garden)=4.33826+0.27245   (Factor 1)+0.01255   (Factor 2)-0.06653   (Factor 3)+0.03537   (Factor 4)
In the analysis, the items corresponding to Factor 1 (comfort) were found to have a positive (+) effect on preference with a B-value of 0.272 (p < .001): untidy - tidy, messy - clean, unharmonious - harmonious, cold - warm, dark - light, and closed - open. As such, it seems that the preference for NWGs will increase as each item approaches tidy, clean, harmonious, warm, light, and open ones. As the items corresponding to Factor 3 (stability) had a B-value of −0.067 (p < .05), they appeared to have a negative (−) effect on preference: static-dynamic and comfortable-uncomfortable ones. It seems that the preference for NWGs will increase as each item of Factor 3 approaches static and comfortable ones (Table 9).
Through a multiple regression analysis of the factors of and preferences for UWGs, it was found to be significant at F = 34.682 (p < .001), and had an explanatory power of 18.7% at adj.R2 = .187. Based on Cohen’s guidelines (1988), it is considered to be significant as it has a medium effect size or higher. Accordingly, the regression equation derived through multiple regression analysis of UWGs is as follows.
Y(Preference for urban woodland garden)=3.63651+0.41036   (Factor 1)+0.06259   (Factor 2)-0.15605   (Factor 3)-0.08718   (Factor 4)
Through the analysis, the items corresponding to Factor 1 (comfort) and Factor 3 (diversity) were found to have positive (+) effects, with B values of 0.410 (p < .001) and 0.156 (p < .001), respectively. The former consisted of closed-open, cold-warm, unharmonious-harmonious, artificial-natural, untidy-tidy, messy-clean and dark-light items; the latter consisted of simple-complex, plain-fancy, and poor-rich ones. Accordingly, as each item approaches open, natural, tidy, clean, and light ones for Factor 1 and complex, fancy, and rich ones for Factor 3, it seems that the preference for UWGs will increase. Moreover, since the items corresponding to Factor 4 (stability), including static-dynamic and comfortable-uncomfortable ones, had a B value of −0.087 (p < .05), it was found to have a negative (−) effect on preference. It seems that the preference for UWGs will increase as each item of Factor 4 becomes more static and comfortable (Table 10).
To sum up these results, it was found that comfort and stability affected preference for woodland gardens, regardless of their type. This seems to suggest that securing comfort and stability is important when creating woodland gardens.


This study aimed to determine the general public’s perceptions of each type of woodland gardens through a survey. The findings were as follows.
First, the survey found that most respondents recognized woodland gardens. They also appeared to prefer natural woodland gardens (NWGs). This suggests that the more faithfully the landscape preserved or emulated the forest structure, the higher their preference will be. However, since there may be differences depending on the garden area, as well as the season in which the presented images were taken and the resulting vegetation cover, it should be taken into account that these factors may have affected preference. Females tended to have a slightly higher preference for formal gardens over urban woodland gardens (UWGs) compared to males, but this is likely due to the color of the plant materials used in the presented images (Hurlbert and Ling, 2007). In addition, considering that the contents written by respondents in the free opinion section of the questionnaire included subjects such as complaints about difficulties in visiting the Korea National Arboretum (KNA) and comments on the UNESCO Gwangneung Forest Biosphere Reserve, it is possible that the term woodland garden was mistakenly perceived by some of the respondents as to referring the entire KNA, resulting in an illusion of high level of recognition of woodland gardens.
Second, based on an image analysis of such garden landscape using a semantic differential scale, both NWGs and UWGs tended to be perceived as having the attribute of comfort. However, there was a difference in that NWGs were perceived as having more natural elements, while UWGs were perceived as having more artificial elements. It seemed that the initial conditions of garden creation, i.e., whether forest vegetation was present, had a significant impact on visitors’ perception. However, since all the gardens presented as examples of UWGs were created less than 5 years ago and has just finished the establishment phase in the context of post-transplant management (Watson and Himelick, 2013), the perception may change in the future depending on the growth and development of the planted plants.
Third, the trait factors of woodland garden images were derived as comfort, orderliness, stability, and spatiality for NWGs, and comfort, modernity, diversity, and stability for UWGs. By analyzing the effects of the derived factors on preference for each type of woodland garden, it was found that comfort and stability affected preference for both NWGs and UWGs. Considering this, it seems that the key to creating effective woodland gardens is to provide a comforting environment for visitors and enable them to feel a sense of stability in terms of structure and planting. To create a space of this character, methods such as securing a sense of place through enclosure planting of woody plants (Jo et al., 2009) may be used.
Summarizing these results, the general public in South Korea had a high preference for NWGs that utilize existing forests and preserve their structures. This appeared to be due to the comfort provided by the scenery. However, UWGs may also receive the same level of preference as NWGs if the plantings mature sufficiently over time. Moreover, to secure green spaces in urban areas, the creation of UWGs will be more efficient than that of NWGs that utilize existing natural forests. Therefore, if woodland gardens with the trait factors derived in this study are created in the future, overall preference for woodland gardens can be increased, which may further lead to a quantitative expansion of such gardens. This will allow us to provide various environmental services in urban areas in a highly aesthetic form (Cameron and Hitchmough, 2016; Credit Valley Conservation, 2012), and can be expected to contribute to the spread of garden culture.
Since the concept of woodland gardens was introduced in South Korea not long ago, most research on the topic has been rudimentary studies including concept establishment and type classification. This study has significance in that consumers’ perception of woodland gardens was determined. However, as this study was conducted based on a survey using example images of woodland gardens, it had limitations in regards to the location of the survey and the choice of images presented to represent each type of woodland gardens. As such, there is a need to be more rigorous in controlling variables in future research, and it seems that follow-up research should be conducted targeting users and experts in related fields who have viewed actual woodland gardens, not just images. Furthermore, it is necessary to explore specific landscaping techniques and verify their effectiveness to ensure comfort and stability, which were found to affect the preference for woodland gardens.

Fig. 1
Result of woodland garden image analysis
Table 1
Images provided to respondents as reference for each type of garden (superscripted number refers to image credit)
Types Landscape images
Naturalistic woodland garden Gwangneung Forest Trail Garden Rhododendron and Moss Gardens of Jade Garden
ksppe-2023-26-6-735f2.jpg ksppe-2023-26-6-735f3.jpg ksppe-2023-26-6-735f4.jpg ksppe-2023-26-6-735f5.jpg
Urban woodland garden Courtyard garden of Monoha Hannam Courtyard garden of Amore Seongsu
ksppe-2023-26-6-735f6.jpg ksppe-2023-26-6-735f7.jpg ksppe-2023-26-6-735f8.jpg ksppe-2023-26-6-735f9.jpg
Typical formal garden 2018 Goyang Flower Exh. Seomi Garden Longwood Gardens1) Butchard Gardens2)
ksppe-2023-26-6-735f10.jpg ksppe-2023-26-6-735f11.jpg ksppe-2023-26-6-735f12.jpg ksppe-2023-26-6-735f13.jpg
Table 2
Composition of survey items and scales
Category Questionnaire Scale References
Recognition of woodland gardens Heard/not heard of woodland garden Nominal scale -
Preference for woodland gardens Preference for woodland garden types 5-point Likert scale -
Perception of woodland gardens 14 pairs of landscape adjectives Semantic Differential scale (S.D. scale) Roh and Son (2022), Yum and Joo (2017), Kim and Cho (2017), Ham (2020)
Characteristics of respondents Gender, Age, Occupation, Residence Nominal scale -
Table 3
Demographic characteristics of respondents
Variable Frequency Percent (%)
Gender Male 197 29.4
Female 473 70.6

Age 10s 12 1.8
20s 36 5.4
30s 91 13.6
40s 167 24.9
50s 162 24.2
60s over 202 30.1

Residence Seoul 191 28.5
Daegu 3 0.4
Daejeon 3 0.4
Incheon 28 4.2
Gwangju 1 0.1
Sejong 3 0.4
Gyeonggido 430 64.2
Gangwondo 2 0.3
Chungbukdo 5 0.7
Chungnamdo 2 0.3
etc. 2 0.3

Occupation Student 12 1.8
Office worker 156 23.3
Professional 149 22.2
Service industry 44 6.6
Sole proprietorship 43 6.4
Housewife 187 27.9
Unemployed 38 5.7
etc. 41 6.1

Total 670 100.0
Table 4
Recognition of woodland gardens by gender (N = 640)
Division Gender Total χ2(p)

Male Female
Yes 137(21.4) 375(58.6) 512(80.0) 5.750 (.015)*
No 48(7.5) 80(12.5) 128(20.0)

Total 185(28.9) 455(71.1) 640(100.0)

* p < .05

Table 5
Recognition of woodland gardens by age (N = 640)
Division Age Total χ2(p)

10s 20s 30s 40s 50s 60s over
Yes 7(1.1) 27(4.2) 65(10.1) 126(19.7) 131(20.5) 156(24.4) 512(80.0) 12.149 (.033)*
No 5(0.8) 9(1.4) 23(3.6) 39(6.1) 23(3.6) 29(4.5) 128(20.0)

Total 12(1.9) 36(5.6) 88(13.7) 165(25.8) 154(24.1) 185(28.9) 640(100.0)

* p < .05

Table 6
Type preference for woodland garden by gender (N = 670)
Division Male Female Total t(p)

Mean SD Mean SD
Preference for type A 4.33 .894 4.33 .855 4.33 −.005 (.996)
Preference for type B 3.51 1.086 3.71 1.022 3.65 −2.269 (.024)*
Preference for type C 4.04 1.081 4.18 1.014 4.14 −1.557 (.120)

* p < .05

Table 7
Result of factor analysis in type A
Division Factor Communality Cronbach’s alpha

Factor 1 Factor 2 Factor 3 Factor 4
Untidy-Tidy .754 .191 −.144 −.059 .629 .620
Messy-Clean .753 .072 −.178 .041 .606
Inharmonious-Harmonious .679 −.183 −.093 .259 .571
Cold-Warm .648 −.164 .180 .107 .491
Dark-Light .643 .270 .061 −.001 .490
Closed-Open .613 −.094 −.096 .171 .423
Informal-Formal .072 .838 .025 −.071 .713
Plain-Fancy −.015 .751 .160 .184 .624
Static-Dynamic .068 .004 .864 .056 .754
Comfortable-Uncomfortable −.228 .217 .732 −.103 .646
Simple-Complex −.139 .200 .235 .739 .661
Poor-Rich .308 .029 −.149 .684 .586
Artificial-Natural .391 −.235 −.246 .513 .531

Eigen value 3.136 1.593 1.552 1.442 - -
% of variance 24.125 12.257 11.938 11.095 - -
Cumulative % 24.125 36.382 48.321 59.416 - -

KMO (Kaiser-Meyer-Olkin) = .803, Bartlett’s χ2 = 1649.890 (p < .001)
Table 8
Result of factor analysis in type B
Division Factor Communality Cronbach’s alpha

Factor 1 Factor 2 Factor 3 Factor 4
Closed-Open .773 .098 .056 .060 .614 .759
Cold-Warm .750 −.070 .075 .194 .611
Inharmonious-Harmonious .735 .151 .216 −.185 .644
Artificial-Natural .656 −.289 .286 .073 .601
Untidy-Tidy .654 .457. 025 −.132 .654
Messy-Clean .582 .500 .047 −.255 .657
Dark-Light .478 .449 .267 −.207 .545
Traditional-Modernistic −.017 .751 −.004 .183 .598
Informal-Formal .027 .733 .106 −.069 .554
Simple-Complex .044 −.032 .784 .043 .619
Plain-Fancy .051 .207 .725 .200 .611
Poor-Rich .379 .040 .634 −.046 .550
Static-Dynamic .303 −.051 .123 .815 .773
Comfortable-Uncomfortable −.425 .079 .100 .673 .650

Eigen value 3.549 1.936 1.789 1.405 - -
% of variance 25.351 13.829 12.782 10.038 - -
Cumulative % 25.351 39.181 51.962 62.001 - -

KMO (Kaiser-Meyer-Olkin)=.841, Bartlett’s χ2=2445.390 (p<.001)
Table 9
Result of Multiple regression analysis of the factors and preference in naturalistic woodland garden
Model U.Coeff. S.Coeff. t p TOL VIF

B SE β
(constant) 4.33826 .03365 128.90435
Factor 1 .27245 .03363 .31717 8.10199 <.001*** .99990 1.00010
Factor 2 .01255 .03355 .01465 .37416 .70842 .99988 1.00012
Factor 3 −.06653 .03368 −.07734 −1.97572 .04866* .99989 1.00011
Factor 4 .03537 .03362 .04118 1.05191 .29328 .99974 1.00026

F(p) 17.65163***
adj.R2 .10206
Durbin-Watson 1.98095

* p < .05,

*** p < .001

Table 10
Result of Multiple regression analysis of the factors and preference in urban woodland garden
Model U.Coeff. S.Coeff. t p TOL VIF

B SE β
(constant) 3.63651 .03860 94.22180
Factor 1 .41036 .03834 .39923 10.70213 < .001*** .99984 1.00016
Factor 2 .06259 .03867. 06037 1.61833 .10613 .99983 1.00017
Factor 3 .15605 .03827 .15209 4.07723 < .001*** .99996 1.00004
Factor 4 −.08718 .03877 −.08389 −2.24892 .02489* .99996 1.00004

F(p) 34.68248***
adj.R2 0.18746
Durbin-Watson 1.86041

* p < .05,

*** p < .001


Cho, H.J. 2020. A study on development of resort wellness forest garden model considering the theme and environment. Master’s thesis. Hongik University, Seoul, Korea.

CNU R&BD Foundation. 2020. A study on the development of guidelines for creating a Korean style woodland garden Pocheon, Korea: Korea National Arboretum.

Cameron, R., J. Hitchmough. 2016. Environmental Horticulture: Science and Management of Green Landscapes Wallingford: CABI.

Cohen, J. 1988. Statistical Power Anaysis for the Behavioral Sciences (Second Edition.). Mahwah, NJ: U.S.A: Lawrence Erlbaum Associates.

Cox, K. 2018. Woodland Gardening: Landscaping with Rhododendrons, Magnolias, and Camellias Perth: Glendoick Publishing.

Credit Valley Conservation. 2012. Native Woodland and Forest Plantings Mississauga: Credit Valley Conservation.

Darke, R. 2002. American Woodland Garden Portland: Timber Press.

Elliott, B. 2007. From the arboretum to the woodland garden. In: Elliott P., Watkins C., Daniels S., (Eds), Cultural and Historical Geographies of the Arboretum (pp. 71-83). London: The Gardens Trust.

Falk, R.F., N.B. Miller. 1992. A primer for soft modeling Ohio, U.S.A: University of Akron Press.

Ham, E.J. 2020. Analysis of flowerscape typology images and preferences for the arboretum : focused on Garden of Morning Calm Arboretum. Master’s thesis. Hanyang University, Seoul, Korea.

Hitchmough, J., N. Dunnett. 2008. Introduction to naturalistic planting in urban landscapes. In: Dunnett N., Hitchmough J., (Eds), The Dynamic Landscape (pp. 1-22). New York: Routledge.
Hong, K.P., H.J. Lee. 2020. Defining a concept of forest garden through understanding values of forest and garden. The Journal of Convergence on Culture Technology. 6(1):171-178. http://doi.org/10.17703/JCCT.2020.6.1.171
Hong, K.P., H.J. Lee, H.Y. Jin, M.K. Lee. 2021. Development of forest garden value index. Journal of the Korean Institute of Garden Design. 7(2):79-92. http://doi.org/10.22849/jkigd.2021.7.2.001
Hurlbert, A.C., Y. Ling. 2007. Biological components of sex differences in color preference. Current Biology. 17(16):623-625. http://doi.org/10.1016/j.cub.2007.06.022

Industry-University Joint Research Institute of Gachon University. 2022. Identification of core landscape elements from references of forest community for application of forest garden model Pocheon, Korea: Korea National Arboretum.

Jo, H.G., T.W. An, H.S. Kim. 2009;A study on planning strategies for the promotion of green areas function. In: Proceedings of the Korean Institute of Landscape Architecture Conference; pp 14-17. Busan, Korea. Pusan National University;

Joo, S.H., S.B. Im. 2003. A study on the landscape adjectives for urban landscape analysis. 31(1):1-10. G704-000407.2003.31.1.004
Kim, S.H., S.H. Cho. 2014. Analysis on the users’ preference images on the types of arboretums: Case studies on Mulhyanggi Arboretum, Garden of Morning Calm and Hantaek Botanical Garden. Journal of the Korean Institute of Garden Design. 3(1):51-59. http://doi.org/10.22849/jkigd.2017.3.1.007

Korea Forest Service. 2021 The 2nd basic plan for garden promotion Daejeon, Korea. Korea Forest Service; Retrieved from https://www.forest.go.kr/.

Korean Institute of Garden Design. 2019. Typification of woodland gardens and development of site appraisal method for woodland gardens Pocheon, Korea: Korea National Arboretum.
Lee, G.H. 2021. Transition of garden culture and establishment of a strategy to promote the creation of a garden city. Korean Jounal of Forest Economics. 28(2):1-13. http://doi.org/10.31541/KJFE.28.2.1
Lee, M.K., D.R. Lee, H.Y. Jin, J.K. Bae, S.J. Song. 2021. Ecological characteristics of sciophyte for woodland garden. Journal of the Korean Institute of Garden Design. 7(4):277-285. http://doi.org/10.22849/jkigd.2021.7.4.001

Lee, S.B. 2016. A semiotic analysis on the structure of meaning and the emotion of landscape image - Focus on the case of slow city Joan. Doctoral dissertation. Kyunghee University, Yongin, Korea.

Nature and Forest Research Institute. 2021. A study on the development of various models of the Korean forest garden Pocheon, Korea: Korea National Arboretum.

Park, J.Y. 2017. A study on the recognition of urban gardener program participants on the role of public garden. Master’s thesis. Chonnam National University, Gwangju, Korea.
Park, M.O., J.H. Choi, B.H. Koo. 2017. A basic study on the setting of the functions of gardens. Journal of the Korean Institute of Garden Design. 3(2):96-102. http://doi.org/10.22849/jkigd.2017.3.2.006

Park, S.Y. 2021. A study on the evaluation of garden awareness and garden promotion plan for the revitalization of garden culture. Doctoral dissertation. Chonnam National University, Gwangju, Korea.
Roh, H.Y., Y.J. Son. 2022. A study on the selection and classification of representative adjectives of domestic garden images - Focused on the visual characteristics of plants. Journal of the Korean Institute of Garden Design. 8(1):15-26. http://doi.org/10.22849/jkigd.2022.8.1.002

Shin, J.H., M.J. Shin, D.H. Kim. 2014. Analysis of landscape image of old-town in Cheon-an - Focus on the commercial street landscape. Journal of the Korea Landscape Council. 6(1):26-38.
Toya, H., M. Skidmore. 2007. Economic development and the impacts of natural disasters. Economic Letter. 94(1):20-25. http://doi.org/10.1016/j.econlet.2006.06.020

Watson, G.W., E.B. Himelick. 2013. The Practical Science of Planting Trees Atlanta: International Society of Arboriculture.

Yoo, R.H., Y.S. Youn, J.H. Lee. 2020. Comparison of Visitor Characteristics in Recreational Forests and Arboretums. 109(4):532-543. http://doi.org/10.14578/jkfs.2020.109.4.532

Yum, H.L., S.H. Joo. 2017. A study on classification of garden styles by analyzing garden expo entries. Korea Society of Floral Art and Design. 36:39-58. UCI:G704-001419.2017.36.36.014

Daderot, 2013 View of flower beds - Longwood Gardens Retrieved from https://commons.wikimedia.org/wiki/File:View_of_flower_beds_Longwood_Gardens_-_DSC00824.JPG.

MasterShake. 2005 Butchardgardens Retrieved from https://commons.wikimedia.org/wiki/File:Butchardgardens.jpg.

Editorial Office
100, Nongsaengmyeong-ro, Iseo-myeon, Wanju_Gun, Jeollabuk-do 55365, Republic of Korea
Tel: +82-63-238-6951    E-mail: jppe@ppe.or.kr                

Copyright © 2024 by The Society of People, Plants, and Environment.

Developed in M2PI

Close layer
prev next