Enhancing Forest Utilization and Quality of Life: An Integrated Approach to Promoting Forest Activities

Abstract

This study investigates the impact of forest-activity diversity on life satisfaction using data from the 2023 Forest Recreation and Welfare Activities Survey conducted by the Korea Forest Service. Forest-activity diversity was categorized into two dimensions: experience diversity (combining forest recreation and welfare-activity experiences) and utilization diversity (the use of forest recreation and welfare facilities and programs). Hierarchical regression analysis revealed that greater diversity in forest activities and utilization significantly enhances life satisfaction. However, participants who engaged in both forest recreation and welfare activities or utilized both facilities and programs reported comparatively lower life satisfaction. The findings indicate that while activity and utilization diversity are critical in improving life satisfaction, the benefits may plateau with excessive participation. The effects of leisure activities also vary based on individual factors such as age, gender, and life context, underscoring the importance of tailored forest welfare programs. Balanced and meaningful engagement in diverse activities is essential for maximizing well-being. While forest activities generally enhance life satisfaction, programs must be inclusive and flexible to address diverse user needs. Future research should incorporate objective measures and longitudinal designs to better evaluate the long-term effects of forest welfare activities.

1. Introduction

In modern society, rapidly changing environments and technological advancements have significantly increased the mental and physical burdens on individuals. These changes have led to rising levels of stress and anxiety, extending beyond individual challenges and becoming societal issues. As a countermeasure, the positive impacts of engaging in natural environments on health and well-being have garnered increasing attention. “Natural environments can alleviate physiological stress responses and provide emotional stability, thereby contributing significantly to enhancing individual quality of life” [1,2]. Numerous studies have demonstrated that exposure to nature and green spaces reduces anxiety and depression [3,4,5,6], alleviates stress [1,7,8,9,10,11,12,13,14], promotes physiological relaxation [15,16,17,18], enhances immune function [19,20,21,22], and improves overall quality of life [23,24]. These effects underscore the critical role of natural environments in improving psychological and physiological health [25,26,27].

The Korea Forest Service (KFS) has particularly emphasized utilizing forests to enhance the quality of life of its citizens. “The rapid industrialization and urbanization of the 1950s and 1960s severely degraded forests, negatively impacting the quality of life for the population”. In response, South Korea initiated large-scale reforestation projects in the 1970s to restore these degraded forests and transform them into resources for public health and well-being. Against this backdrop, policies and laws were enacted to promote forest-based welfare and public health. The “Act on the Promotion of Forest Welfare” aims to support the enhancement of citizens’ quality of life and happiness through forest welfare services. Similarly, the 5th Basic Forest Plan (2008–2017) designated the expansion of green spaces and services as one of its five strategic pillars for improving quality of life. The subsequent 6th Basic Forest Plan (2018–2037) articulated a vision for enhancing quality of life through forest welfare services. Moreover, the “Act on Forest Culture and Recreation” aims to improve citizens’ quality of life by providing forest culture and recreation services. “These policies highlight the critical role of forest environments in contributing to public well-being and life satisfaction.”

Recent studies have highlighted the diverse benefits provided by natural environments, emphasizing how forests contribute to daily well-being through ecological, social, and emotional benefits [28]. In Europe, demographic trends such as an aging population and increasing urbanization have further underscored the importance of forests in mitigating the mental health challenges associated with urban living, such as anxiety and depression [29,30]. Urban forests and green spaces play crucial roles in providing opportunities for stress relief and enhancing quality of life [31]. Urban residents participating in forest therapy programs reported improved mental health and life satisfaction [32], demonstrating the psychological benefits of nature-based interventions. Similarly, forest-based interventions significantly reduced stress levels and enhanced immune function [33]. These findings collectively underscore the critical role forest activities play in promoting health and well-being, particularly through forest welfare services. Additionally, they emphasize that forest welfare services serve as a crucial link between forest policy and management in the EU, highlighting the potential for these services to bridge societal demands and forest-based solutions. Forest welfare services, which encompass therapy, recreation, education, and cultural activities, offer solutions to demographic and social challenges by reducing stress, enhancing cognitive function, and strengthening social cohesion [11,34]. Tailored programs can address the needs of aging populations or urban residents seeking respite from high-density living conditions, further demonstrating forests’ multifaceted contributions to quality of life. Research on urban forestry also shows that individuals exposed to green spaces have lower cortisol levels and improved cardiovascular health [35].

Internationally, several studies have supported similar findings. In the United States, research demonstrated that urban parks significantly increase physical activity levels and overall well-being due to their accessibility and design [36]. Integrating forest welfare services into EU policies could effectively address demographic and social challenges, underscoring their role in improving public health and fostering sustainable forest management [37]. Additionally, a Taiwanese study showed that the environmental quality of urban parks positively influenced leisure activities and the well-being of middle-aged and elderly populations [38]. These findings align with the results of a Malaysian study that demonstrated how recreational activities in urban parks promote social connections and enhance the quality of life of local communities [39].

However, most existing studies are limited to specific groups, such as the elderly, or focus on the effects of individual forest programs or facilities without systematically analyzing how diverse forest activity experiences influence overall quality of life. For example, one study explored the impact of forest-based healing programs on happiness and positive psychology among middle-aged women in Seoul and its metropolitan areas but did not extend its findings to other age groups [33]. Another study assessed user satisfaction with specific forest facilities but did not examine how the diversity and frequency of forest activities contribute to variations in quality of life [40].

This study aims to address these limitations by utilizing data from the “Forest Recreation and Welfare Activity Survey”, conducted by the Korea Forest Service, to comprehensively analyze how the diversity of forest activity experiences and the use of facilities and programs impact life satisfaction. Specifically, this study systematically evaluates individuals’ multifaceted experiences of forest activities to demonstrate how forest welfare services contribute to enhancing citizens’ quality of life. This survey, designed to understand the purposes, demands, and selection criteria regarding citizens’ forest and recreational welfare activities, is conducted annually with regionally representative samples. By leveraging these data, this study categorizes forest activities based on experience diversity and the use of facilities and programs, providing empirical evidence on the educational, therapeutic, and cultural impacts of forest welfare services on quality of life.

This research provides an in-depth understanding of how forest welfare services contribute to public health and quality-of-life improvements. It proposes effective directions for designing and operating forest welfare programs. Furthermore, by comprehensively analyzing individuals’ multifaceted experiences of forest activities, this study demonstrates how forest welfare services contribute to sustainable environmental management and the formation of a healthy social structure. Ultimately, it highlights the value of forest utilization and lays the groundwork for realizing the future vision of forest welfare.

2. Methodology

2.1. Survey Design and Sampling

This study was conducted based on the Forest Welfare and Recreation Activity Survey (Korea Forest Service, Korea, Korea), a nationally approved statistical survey authorized by Statistics Korea. The purpose of this survey is to understand the public’s forest welfare and recreation activities, including with respect to their purposes, demands, and selection criteria, thereby providing a foundation for the supply of forest welfare services and the formulation of relevant policies.

The survey content is composed of five main sections, covering experiences with forest welfare and recreation activities and usage of related facilities and programs. The survey was conducted using a stratified probability-proportional-to-size systematic sampling method to ensure representativeness and reliability. In the first stage, survey districts were selected based on regional and housing-type strata, followed by systematic sampling within each stratum. In the second stage, households were randomly selected from the chosen districts, with all household members aged 15 years or older being included in the survey. This method ensured comprehensive coverage of diverse population groups while maintaining statistical validity.

The survey targeted 11,457 individuals from 5000 households, including household heads and members aged 15 years or older, residing in 17 provinces across South Korea. The survey was conducted over a 25-day period from 5 June to 30 June 2023. Data collection was carried out through face-to-face interviews by trained surveyors to ensure data accuracy and consistency.

2.2. Operational Definition

This study was conducted based on the survey items and definitions provided by the Forest Welfare and Recreation Activity Survey. The variables used in the study are as follows:

The term “Personal characteristics” refers to the sociodemographic characteristics of respondents. In this study, the basic respondent information collected from the survey was set as the personal characteristic variable. The personal characteristic variables included gender, age, monthly household income, occupation, education level, whether the respondents had children, housing type, and residential area size.

The term “Forest recreation and welfare activities” refers to various types of activities conducted in forest areas (e.g., forests, mountains, parks, and urban forests) or within facilities established in forest environments. These activities encompass leisure, hobbies, culture, arts, creative activities, physical exercise (sports and recreational activities), and social participation, all conducted for leisure purposes. Forest recreation and welfare activities are categorized based on the duration of participation, as follows:

Daily Activities: Activities conducted near one’s home or workplace for a relatively short duration (less than 4 h, including travel time).

Day-trip Activities: Activities that require a significant portion of the day (4 h or more, including travel time).

Overnight Activities: Activities lasting at least one night and two days, with a duration of 4 h or more, including travel time. This category does not include non-overnight activities.

“Forest recreation and welfare facilities” consist of infrastructure established within forests to provide services related to culture, recreation, healing, and education. Examples of forest recreation and welfare facilities include natural recreation forests, healing forests, forest parks, forest trails, and arboretums, which serve as accessible public spaces that allow people to interact with nature and enhance their psychological and physical well-being.

“Forest recreation and welfare programs” are structured activities conducted by certified forest welfare professionals to facilitate physical and emotional engagement in forest environments. These programs include forest-based healing programs, forest interpretation programs, and forest experience programs for children, aiming to deepen participants’ connections with forests and contribute to their psychological and physical health. Together, these facilities and programs form the foundation of forest-based welfare services, supporting personal growth and well-being through both structured opportunities and spontaneous recreational activities.

“Life satisfaction” was measured based on the guidelines for ‘subjective well-being’ defined by the OECD (2013), which include the sub-factors of life satisfaction, positive affective experiences, negative affective experiences, and eudaimonia [41]. These guidelines were selected because they provide a comprehensive framework for evaluating subjective well-being from multiple perspectives, capturing both positive and negative emotional experiences as well as an individual’s sense of purpose and fulfillment.

The variables used for measuring life satisfaction consisted of respondents’ scores on happiness, depression, anxiety, life satisfaction, and work value (

Table 1. Survey questions for forest recreation and welfare activities considered in this study.

Survey Questions
How happy do you think you were yesterday?
How depressed were you yesterday? (reverse question)
How worried were you yesterday? (reverse question)
How satisfied are you with your overall life these days?
How valuable to you is the work you are doing these days?

). Each of these components represents a critical aspect of subjective well-being: happiness indicates positive affect, while depression and anxiety reflect negative emotional states. Life satisfaction and work value represent eudaimonia, reflecting an individual’s sense of purpose, fulfillment, and overall evaluation of life.

The life satisfaction score was defined as the average of the respondents’ response scores (ranging from 0 to 10) for these five components. Items related to depression and anxiety were reverse-coded to ensure that higher scores consistently indicated greater well-being across all variables.

The Cronbach’s α value for these items was 0.772, indicating a desirable level of internal consistency. Typically, a Cronbach’s α value above 0.7 is considered to demonstrate acceptable reliability, suggesting that the items can reliably measure a single underlying construct [42]. As such, the average score of these items was used to represent the respondents’ overall life satisfaction, given that they exhibited satisfactory internal consistency.

2.3. Study Methods

The purpose of this study was to analyze the effects of diversity in forest recreational welfare experiences and in the use of facilities and programs on the life satisfaction of users. The analysis was performed as follows.

First, through a descriptive analysis, we recorded the respondents’ forest recreation and welfare experiences and determined facility and program usage and sociodemographic characteristics. We confirmed the basic statistics and group frequencies for experience diversity, utilization diversity, and personal characteristics.

Second, to verify the differences in life satisfaction between different user groups, corresponding to experience and utilization diversity, we conducted the Kruskal–Wallis test, a nonparametric test used for evaluating differences in the median values between groups. Subsequently, the differences between the groups were confirmed through a post hoc analysis (Dunn’s test).

Third, Crammer’s V coefficient and variance inflation factors (VIFs) were compared to evaluate the correlation and multicollinearity between the variables that constituted the personal characteristics of the respondents. The variables with high correlation coefficients and multicollinearity were excluded from the analysis. This minimized the problems arising from the correlations between the independent variables in the regression analysis. Before performing the regression analysis, the independent variables of activity diversity, utilization diversity, and personal characteristics were converted into dummy variables, and the reference variables were selected.

Finally, a hierarchical regression analysis was conducted to examine the impacts of personal characteristics, experience diversity, and utilization diversity on the life satisfaction of the respondents. The hierarchical multiple regression model consisted of three models; the equation for each model is described below [Equations (1)–(3)]. Model 1 included only the control variables that represented the personal characteristics of the respondents, Model 2 included the variables related to activity diversity, and Model 3 included the variables related to utilization diversity (Figure 1).

$$\mathrm{y}={\mathsf{\beta }}_0+{\mathsf{\beta }}_1{\mathrm{X}}_{Controlvariables\left(P\mathrm{e}\mathrm{r}\mathrm{s}\mathrm{o}\mathrm{n}\mathrm{a}\mathrm{l}\mathrm{c}\mathrm{h}\mathrm{a}\mathrm{r}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{e}\mathrm{r}\mathrm{i}\mathrm{s}\mathrm{t}\mathrm{i}\mathrm{c}\mathrm{s}\right)}+\in$$

(1)

$$\mathrm{y}={\mathsf{\beta }}_0+{\mathsf{\beta }}_1{\mathrm{X}}_{Controlvariables\left(P\mathrm{e}\mathrm{r}\mathrm{s}\mathrm{o}\mathrm{n}\mathrm{a}\mathrm{l}\mathrm{c}\mathrm{h}\mathrm{a}\mathrm{r}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{e}\mathrm{r}\mathrm{i}\mathrm{s}\mathrm{t}\mathrm{i}\mathrm{c}\mathrm{s}\right)}+{\mathsf{\beta }}_2{\mathrm{X}}_{Activitydiversity}+\in$$

(2)

$$\begin{array}{c}\mathrm{y}={\mathsf{\beta }}_0+{\mathsf{\beta }}_1{\mathrm{X}}_{Controlvariables\left(P\mathrm{e}\mathrm{r}\mathrm{s}\mathrm{o}\mathrm{n}\mathrm{a}\mathrm{l}\mathrm{c}\mathrm{h}\mathrm{a}\mathrm{r}\mathrm{a}\mathrm{c}\mathrm{t}\mathrm{e}\mathrm{r}\mathrm{i}\mathrm{s}\mathrm{t}\mathrm{i}\mathrm{c}\mathrm{s}\right)}+{\mathsf{\beta }}_2{\mathrm{X}}_{Activitydiversity}\\ +{\mathsf{\beta }}_3{\mathrm{X}}_{Utilizationdiversity}+\in \end{array}$$

(3)

Figure 1. Research flowchart.

Figure 1. Research flowchart.

3. Results

3.1. Descriptives Satistics

In this study, we aimed to determine the distribution of respondents who participated in the forest recreation and welfare activity survey conducted in South Korea. In addition, a frequency analysis was conducted to determine the status of participation in forest recreation and welfare activities, forest recreation, and welfare facilities and programs.

The personal characteristics of the respondents who participated in the forest recreation welfare activity survey are given below.

The gender distribution was as follows: male respondents accounted for 46.32% of the total respondents (5307), whereas female respondents accounted for 53.68% (6150). The age distribution showed that respondents aged 60 years or older comprised the largest group, with 3458 respondents (30.18% of the total respondents), followed by those aged 40 or older but younger than 49, with 1749 respondents (15.66% of the total respondents). The lowest number of respondents were in the age range of 15–19, with 862 respondents (7.52% of the total respondents).

Information on the average monthly household income was only provided by the head of each household; information pertaining to household income for household members was not provided. In this study, to replace the missing values, the data were corrected by unifying the income information of the household heads and members based on the household number assigned to each survey respondent. After processing the missing values, the respondents’ average monthly household income was KRW 4 million or more but less than KRW 5 million, with the highest proportion corresponding to 2137 people (18.65% of the total participants), followed by KRW 5 million or more but less than KRW 6 million, corresponding to 2062 people (18.00% of the total participants). The number of respondents with an average monthly income of less than KRW 1 million was the lowest, at 399 (3.48% of the total participants). The main occupation was office worker, corresponding to 2235 people (19.51% of the total participants), followed by housewife, corresponding to 1902 people (16.60% of the total participants), and service worker, corresponding to 1712 people (14.94% of the total participants).

The distribution of the respondents by marital status was as follows: 7074 married (61.74% of the total participants), 3090 unmarried (26.97% of the total participants), and 1293 “others” (11.29% of the total participants). Regarding whether the participants had children, 9052 people (79.01% of the total participants) did not have children, while 2405 (20.99% of the total participants) had children. The highest level of education was college graduate or higher for 5233 respondents (45.67% of the total participants), followed by high-school graduate or higher for 4176 respondents (36.45% of the total participants).

The distribution by residential area showed that 2263 people lived in Gyeonggi (19.75% of the total participants) and 2188 people lived in Seoul (19.10% of the total participants), indicating that a high proportion of the respondents were from metropolitan areas. The sizes of the residential areas were as follows: 7595 respondents (66.29%) lived in large cities, 2943 (25.69%) lived in medium-sized cities, and 919 (8.02%) inhabited small rural areas. Finally, the distribution of respondents by type of residence showed that 6848 (59.8%) lived in apartments and 2472 (21.6%) lived in single-family homes (

Table 2. Demographic characteristics of participants.

Variables		Number of Respondents (N)	Percentage (%)	Variables		Number of Respondents (N)	Percentage (%)
Gender	Male	5307	46.32	Children	With	9052	79.01
	Female	6150	53.68		Without	2405	20.99
Age	15~19	862	7.52	Type of Residence	Single-Family House	2472	21.58
	20~29	1339	11.69		Apartment	6848	59.77
	30~39	1625	14.18		Townhouse/Rowhouse	1964	17.14
	40~49	1794	15.66		Other	173	1.51
	50~59	2379	20.76	Regional Scale	Major City	7595	66.29
	Over 60	3458	30.18		Small-to-Medium City	2943	25.69
Monthly Household Income (million won)	Under 100	399	3.48		Rural Area	919	8.02
	100~200	736	6.43	Educational Level	<Elementary School	632	5.52
	200~300	1118	9.76		Middle-School Graduate	1416	12.36
	300~400	1664	14.52		High-School Graduate	4176	36.45
	400~500	2137	18.65		>University graduate	5233	45.67
	500~600	2062	18.00	Marital Status	Single	3090	26.97
	600~700	1638	14.30		Married	7074	61.74
	Over 700	1703	14.86		Other	1293	11.29
Occupation	Management	164	1.43	Residential Area	Seoul	2188	19.10
	Professional or Related Worker	319	2.78		Busan	611	5.33
	Clerk	2235	19.51		Daegu	542	4.73
	Service Worker	1712	14.94		Incheon	646	5.64
	Sales Worker	1240	10.82		Gwangju	438	3.82
	Skilled Agricultural Forestry and Fishery Worker	383	3.34		Daejeon	435	3.80
	Craft and Related Trade Worker	678	5.92		Ulsan	335	2.92
					Sejong	216	1.89
	Equipment Operator	267	2.33		Gyeonggi	2263	19.75
					Gangwon	403	3.52
	Elementary Worker	420	3.67		Chungbuk	483	4.21
	Soldier	13	0.11		Chungnam	552	4.82
	Student	1288	11.24		Jeonbuk	471	4.11
	Housemaker	1902	16.60		Jeonnam	459	4.01
	Preparing for Employment	73	0.64		Gyeongbuk	612	5.34
	Unemployed	755	6.60		Gyeongnam	598	5.22
	Others	8	0.07		Jeju	205	1.79

).

Based on the analysis of forest recreation frequency and welfare activities, 2020 people (17.63%) had not participated in any forest recreation or welfare activities. The highest proportion, 3399 respondents (29.68%), had only engaged in daily activities. The group that only experienced same-day activities included 1667 people (14.55%), and the group that only experienced accommodation activities contained 800 people (6.98%). The group that experienced only the daily and same-day types included 1915 people (16.71%), and the group that experienced only the daily and accommodation types had the lowest response rate, with 368 respondents (3.21%). The group that experienced only the same-day and accommodation types comprised 729 people (6.36%), and the group that experienced all of the daily, same-day, and accommodation types of forest recreation and welfare activities comprised 559 people (4.88%). A frequency analysis of the respondents’ participation in forest recreation and welfare activities confirmed that the majority of participants participated in daily or same-day forest recreation and welfare activities, and that the participation rate in accommodation-type forest recreation and welfare activities was lower than the rates for other activities (Figure 2 and

Table 3. Frequency analysis for activity diversity.

Variables	Number of Respondents (N)	Percentage (%)
None	2020	17.63
Only daily	3399	29.68
Only same-day	1667	14.55
Only accommodation type	800	6.98
Daily and same-day	1915	16.71
Daily and accommodation-type	368	3.21
Same-day and accommodation type	729	6.36
All	559	4.88
Total	11,457	100.00

).

For forest recreation and welfare facilities and programs, a frequency analysis was conducted based on the number of times they were used along with the presence or absence of experiences. First, the analysis of the frequency of facility use revealed that the largest number of respondents, 8006 (69.89%), never used a facility. Next, 2047 people (17.87%) used a facility once, 910 (7.95%) used a facility twice, and 674 (5.89%) used a facility three or more times. In the case of forest recreation and welfare programs, the largest number of respondents, 10,745 (93.80%), never participated in one; 687 (6.01%) had participated in such programs once; 164 (1.43%) had participated in such programs twice; and 41 (0.36%) had participated in such programs three or more times. For facilities and programs, we confirmed that as the frequency of use increased, the number of respondents who participated in them significantly decreased. Therefore, in this study, considering the difference in the size of the groups, one’s experience in engaging in forest recreation and welfare facilities and programs was classified by the presence or absence of experience rather than by frequency. The respondent groups were classified as “experience with both facilities and programs”, “experience with only facilities or programs”, and “no experience”. Notably, 860 respondents (7.51%) had experience with both facilities and programs, 2803 (24.46%) had experience with only facilities or programs, and 7794 (68.03%) had no such experience (

Table 4. Frequency analysis for program experience, facility experience, and utilization diversity.

Variables	Number of Respondents (N)	Percentage (%)
Program Participation
Never	10,745	93.80
1 time	687	6.01
2 times	164	1.43
$\ge$3 times	41	0.36
Total	9437	100.00
Facility Use
Never	8006	69.89
1 time	2047	17.87
2 times	910	7.95
$\ge$3 times	674	5.89
Total	9437	100.00
Utilization Diversity
All	860	7.51
Facility or Program	2803	24.46
None	7794	68.03
Total	11,457	100.0

).

The following are the descriptive statistics used for comparing the life satisfaction of the respondents by activity and utilization diversity (

Table 5. Descriptive statistics for forest recreation and welfare activity diversity.

Variable	Number of Respondents (N)	Average	Standard Deviation	Kurtosis	Skewness
None	2020	6.474	1.297	0.172	−0.253
Only daily	3399	6.744	1.182	−0.173	−0.072
Only same-day	1667	6.998	1.141	−0.417	−0.076
Only accommodation type	800	7.006	1.132	0.002	−0.321
Daily and same-day	1915	7.087	1.096	−0.243	−0.209
Daily and Accommodation-type	368	7.135	1.214	−0.819	0.025
Same-day and Accommodation-type	729	7.116	1.133	−0.315	−0.260
All	559	7.391	1.097	−0.118	−0.322
Total	11,457	6.877	1.200	−0.062	−0.220
Levene’s statistic = 10.289, p = 0.000

). When examining the statistics according to activity diversity, the kurtosis and skewness for each group were <|8| and <|3|, respectively, indicating that all the groups satisfied normality. Levene’s statistic, which was the result of the homogeneity test, was p < 0.001, confirming that the variances between the groups were not equal. In addition, when examining the average life satisfaction by group, the average of all the respondents was 6.877 points, and the standard deviation was 1.197. The average life satisfaction by group increased in the following order: no experience (6.474) < daily-type-only experience (6.744) < same-day-type-only experience (6.998) < accommodation-type-only experience (7.006) < daily-type- or same-day-type-only experience (7.087) < same-day-type- or accommodation-type-only experience (7.116) < daily-type- or- accommodation-type-only experience (7.135) < both experiences (7.391). The groups with activity experience exhibited higher life satisfaction than the group without activity experience, and the life satisfaction of the respondents tended to increase with the increase in the variety of forest recreational welfare activities.

The groups stratified according to utilization diversity also had kurtosis and skewness values within the range satisfying normality (

Table 6. Descriptive statistics results for facility and program utilization diversity.

Variable	Number of Respondents (N)	Average	Standard Deviation	Kurtosis	Skewness
None	6.8079	7794	1.22344	0.022	−0.203
Facility or Program	6.7777	860	1.17151	−0.389	0.249
All	7.1007	2803	1.11618	−0.208	−0.359
Total	6.8773	11,457	1.20087	−0.062	−0.220
Levene’s statistic = 10.289, p = 0.000

); Levene’s statistic was p < 0.001, violating the assumption of homoscedasticity between the groups. The average life satisfaction by group was as follows: only used facilities or programs (6.778) < no experience of use (6.808) < used all facilities (7.100). The results of the analysis indicated that contrary to H2, the group that used only one of the facilities and programs depicted the lowest satisfaction when compared to the group that had no experience with utilization diversity. Thus, additional analysis was necessary to determine whether there was a significant difference in life satisfaction between the groups.

3.2. Group Difference Analysis

To compare the life satisfaction of the different groups by activity and utilization diversity, we examined the differences between the groups through a post hoc analysis (after conducting a Kruskal–Wallis test) (

Table 7. Kruskal–Wallis test results.

Activity Diversity Life Satisfaction		Utilization Diversity Life Satisfaction
$x^2$	461.667	$x^2$	146.234
Degree of Freedom	7	Degree of Freedom	2
p-value	0.000	p-value	0.000

). The results of the descriptive statistical analysis of activity and utilization diversity showed that the homogeneity of variance between the groups was violated. In this study, the nonparametric Kruskal–Wallis test was used to test the differences between the groups. Both activity and utilization diversity exhibited a significance level of <0.001, confirming that there was a significant difference in the average life satisfaction among the activity and utilization diversity groups.

Figure 3 presents the results of the post hoc analysis of the Kruskal–Wallis test for activity diversity (

Table 8. Post hoc test (Dunn’s test) results for activity diversity.

Variables	Intercept Statistics	Standard Deviation	Standardized Statistics	p-Value
None—Daily	−620.597	92.802	−6.687	0.000 ***
None—Same-Day	−1303.933	109.306	−11.929	0.000 ***
None—Accommodation Type	−1386.521	137.992	−10.048	0.000 ***
None—Daily and Same-Day	−1581.568	105.357	−15.012	0.000 ***
None—Daily and Accommodation Type	−1520.176	187.227	−8.654	0.000 ***
None—Daily and Accommodation Type	−1669.088	142.725	−11.694	0.000 ***
None—All Types	−2428.614	157.868	−15.384	0.000 ***
Daily—Same-Day	−683.337	98.773	−6.918	0.000 ***
Daily—Accommodation Type	−765.925	129.809	−5.900	0.000 ***
Daily—Daily and Same-Day	−960.971	94.385	−10.181	0.000 ***
Daily—Daily and Accommodation Type	−999.580	181.280	−5.514	0.000 ***
Daily—Same-Day and Accommodation Type	−1048.491	134.828	−7.776	0.000 ***
Daily—All Types	−1808.018	150.767	−11.992	0.000 ***
Same-Day—Accommodation Type	−82.588	142.077	−0.581	1.000
Same-Day—Daily and Same-Day	−277.634	110.653	−2.509	0.339
Same-Day—Daily and Accommodation Type	−316.243	190.257	−1.662	1.000
Same-Day—Daily and Same-Day and Accommodation Type	−365.154	146.677	−2.490	0.358
Same-Day—All Types	−1124.681	161.451	−6.966	0.000 ***
Accommodation Type—Daily and Same-Day	−195.047	139.061	−1.403	1.000
Accommodation Type—Daily and Accommodation Type	−233.655	208.067	−1.123	1.000
Accommodation Type—Same-Day and Accommodation Type	−282.567	169.140	−1.671	1.000
Accommodation Type—All Types	−1042.093	182.100	−5.723	0.000 ***
Daily and Same-Day—Daily and Accommodation Type	−38.609	188.016	−0.205	1.000
Daily and Same-Day—Daily and Accommodation Type	−87.520	143.758	−0.609	1.000
Daily and Same-Day—All Types	−847.046	158.804	−5.334	0.000 ***
Daily and Accommodation Type—Same-Day and Accommodation Type	−48.912	211.235	−0.232	1.000
Daily and Accommodation Type—All Types	−808.438	221.749	−3.646	0.007 ***
Daily and Accommodation Type—All Types	−759.526	185.712	−4.090	0.001 ***

***: p < 0.001.

). The majority of groups depicted significant differences in activity diversity. We confirmed there was a statistical difference between the group with no experience in engaging in forest recreation welfare activities and the group with no experience at all. This result supports the statistical significance of the descriptive statistics, indicating that the groups with activity experience had higher life satisfaction than the group without activity experience. However, no significant differences were observed between the groups. In particular, the groups with only same-day-type or only accommodation-type participation did not show statistically significant differences in life satisfaction compared to the other groups, with the only exception being the group that engaged in all forest recreation welfare activities.

The results of the post hoc analysis of the utilization diversity group (see Figure 3 and

Table 9. Post hoc test (Dunn’s test) results of utilization diversity.

Variables	Intercept Statistics	Standard Deviation	Standardized Statistics	p-Value
No experience—Experience only in facility or program	201.240	188.694	1.695	0.270
Experience only in facility or program—Experience in all	−1040.807	128.768	−8.083	0.000 ***
No experience—Experience in all	−839.567	72.753	−11.540	0.000 ***

***: p < 0.001.

) showed that there was no statistically significant difference between the group with no utilization experience and the group that only experienced the facility or program at a significance level of >0.05. We noted significant differences in the life-satisfaction levels between these groups and compared the previously conducted technical statistics with the post-analysis results. There was a difference in life satisfaction between the group that only used one facility or program and the group that had no experience using it; however, the difference was not statistically significant. Therefore, life satisfaction according to utilization diversity resulted in higher life satisfaction in the group that had experience using both facilities and programs.

3.3. Correlation and Multi-Commonality Analysis

To identify the relationships between the variables that constituted personal characteristics, a correlation analysis was performed, and the VIF index was derived. This process was used to avoid multicollinearity problems that may occur during a regression analysis. The correlation between the variables was derived using Cramer’s V coefficient (

Table 10. Results of Cramer’s V coefficient analysis for personal characteristic variables.

		A	B	C	D	E	F	G	H	I	J
A	CC	1.000
	p-value
B	CC	0.002	1.000
	p-value	0.863
C	CC	0.173	0.241	1.000
	p-value	0.000	0.000
D	CC	−0.036	−0.376	−0.313	1.000
	p-value	0.000	0.000	0.000
E	CC	−0.108	−0.302	−0.590	0.347	1.000
	p-value	0.000	0.000	0.000	0.000
F	CC	0.084	0.730	0.153	−0.324	−0.160	1.000
	p-value	0.000	0.000	0.000	0.000	0.000
G	CC	−0.460	0.116	−0.255	0.183	0.207	0.221	1.000
	p-value	0.000	0.000	0.000	0.000	0.000	0.000
H	CC	0.000	−0.126	−0.085	0.130	0.139	−0.077	0.048	1.000
	p-value	0.988	0.000	0.000	0.000	0.000	0.000	0.000
I	CC	0.005	0.108	0.075	−0.197	−0.186	0.087	−0.048	−0.274	1.000
	p-value	0.566	0.000	0.000	0.000	0.000	0.000	0.000	0.000
J	CC	0.001	0.097	0.087	−0.233	−0.148	0.099	−0.043	−0.266	0.607	1.000
	p-value	0.944	0.000	0.000	0.000	0.000	0.000	0.000	0.000	0.000

CC: correlation coefficient A: gender, B: age, C: occupation, D: monthly household income, E: educational level, F: marital status, G: children, H: type of residence, I: regional scale, J: residential area.

). Based on the results, variables with a Cramer’s V coefficient greater than 0.6, such as residential area and marital status, were removed to reduce multicollinearity. Furthermore, during the VIF analysis, the dummy variable for age (15–19 years) and occupation (student) were deleted as they had VIF values close to 5, which indicated potential multicollinearity issues.

3.4. Hierarchical Regression Analyses

Before we conducted the hierarchical regression analysis, the data were converted into dummy variables, as the independent variables of activity diversity, utilization diversity, and personal characteristics were measured on nominal and ordinal scales. By conducting a regression analysis using dummy variables, the influence of the nominal and ordinal variables was minimized, allowing for a more accurate measurement of the relationship between the independent and dependent variables. By using dummy variables, the potential biases associated with nominal and ordinal variables were minimized, allowing for a more precise estimation of the relationship between the independent and dependent variables. The primary purpose of employing dummy variables was to appropriately incorporate nominal and ordinal variables into the regression model, thereby preserving analytical robustness. In regression analysis, a number of dummy variables with different levels are created, using one of the levels as a reference category. In this study, the reference level was chosen based on the highest frequency, constituting a standard and suitable method.

To evaluate the effect of the independent variables on the dependent variable, a hierarchical multiple regression analysis was conducted. Hierarchical regression models are particularly useful for assessing the incremental value of predictor variables added in a sequential manner, enabling researchers to evaluate the additional explanatory power and moderation effects at each stage. Hierarchical regression analysis allows for hypothesis verification by examining the extent to which the dependent variable is explained as new variables are added as well as assessing changes in explanatory power and statistical significance. The crucial aspect is determining an appropriate order for entering variables to adequately test the hypothesis. Generally, control variables are entered first, followed by other predictors added individually to assess their unique contributions.

In this study, we conducted a hierarchical regression analysis to examine the factors that influenced the life satisfaction of the respondents through three models. Model 1 was a basic model that included demographic characteristics, such as gender, age, monthly household income, occupation, education level, presence of children, housing type, and the size of the residential area. Model 2 accounted for experience diversity variables related to forest recreation and welfare activities, and Model 3 further included utilization-diversity variables.

In the first step, Model 1 included only control variables representing personal characteristics. The coefficient of determination (R²) was 0.079, indicating that the model explained approximately 7.9% of the variance in the dependent variable. The adjusted value was 0.077, suggesting that, after accounting for the number of predictors, the model’s explanatory power was approximately 7.7%. The F-change value was 28.931, indicating that Model 1’s results were statistically significant (R²), demonstrating that the control variables had a significant impact on explaining the variance in the dependent variable.

In the second step, the R² value increased to 0.102, and the adjusted R² was 0.099, showing that adding the activity diversity variable improved the explanatory power of the model. The F change value was 41.547, confirming that the improvement in Model 2 was significant (p < 0.001), suggesting that activity diversity contributes significantly to explaining life satisfaction. This indicates that participation in diverse activities is an important determinant of life satisfaction. In the third step, with additional explanatory variables included, the R² value slightly increased to 0.109, indicating that approximately 10.9% of the variability in the dependent variable was explained. The adjusted R² was 0.106, showing a slight improvement in explanatory power compared to Model 2. The F change value was 46.254, indicating that there was a statistically significant improvement in the third step as well (p < 0.001), suggesting that the added variables contribute to better explaining the dependent variable. Although the increase in explanatory power was modest, it indicates that utilization diversity also plays a role in enhancing life satisfaction.

The hierarchical regression analysis results showed that adding explanatory variables at each stage significantly increased the explanatory power of the model. The R² value and adjusted R² value increased stepwise, which suggests that the variability of the dependent variable could be explained better as the explanatory variables were added. Additionally, the change in F at each step was statistically significant (p < 0.001), confirming that the improvement of the model was significant. However, the Durbin–Watson value was 0.909, suggesting that further examination of autocorrelation is needed (

Table 11. Hierarchical regression analysis results.

		Model 1		Model 2		Model 3
		B	$\mathsf{\beta }$	B	$\mathsf{\beta }$	B	$\mathsf{\beta }$
	(Constant)	7.394		7.278		7.279
Gender	Female
	Male	−0.007	−0.003	0.010	0.004	0.008	0.003
Age	Over 60
	20~29	−0.032	−0.009	0.018	0.005	0.025	0.007
	30~39	−0.151	−0.044 ***	−0.134	−0.039 **	−0.129	−0.037 **
	40~49	−0.098	−0.030 *	−0.088	−0.027 *	−0.095	−0.029 *
	50~59	−0.096	−0.032 **	−0.098	−0.033 **	−0.096	−0.033 **
Household Income (KRW million)	400~500
	Under 100	−0.254	−0.0039 ***	−0.236	−0.036 **	−0.235	−0.036 **
	100~200	−0.197	−0.040 ***	−0.173	−0.035 **	−0.170	−0.035 **
	200~300	−0.316	−0.078 ***	−0.288	−0.071 ***	−0.296	−0.073 ***
	300~400	−0.161	−0.047 ***	−0.151	−0.044 ***	−0.149	−0.044 ***
	500~600	−0.017	−0.006	0.008	0.002	0.004	0.001
	600~700	−0.042	−0.012	−0.032	−0.009	−0.028	−0.008
	Over 700	−0.093	−0.027 *	−0.070	−0.021	−0.073	−0.022
Occupation	Clerk
	Management	−0.431	−0.043 ***	−0.435	−0.043 ***	−0.452	−0.045 ***
	professionals and related workers	−0.058	−0.008	−0.072	−0.010	−0.063	−0.009
	Service worker	−0.045	−0.013	−0.054	−0.016	−0.061	−0.018
	Sales worker	−0.062	−0.016	−0.064	−0.017	−0.068	−0.017
	Skilled agricultural forestry and fishery worker	−0.193	−0.029 **	−0.202	−0.030 **	−0.211	−0.032 **
	Craft and related trade workers	−0.121	−0.024 *	−0.116	−0.023 *	−0.127	−0.025 *
	Equipment operator	−0.213	−0.027 **	−0.194	−0.024 *	−0.188	−0.024 *
	Elementary worker	−0.193	−0.030 ***	−0.189	−0.030 **	−0.194	−0.030 **
	Soldier	0.214	0.006	0.250	0.007	0.219	0.006
	Housemaker	−0.236	−0.073 ***	−0.251	−0.078 ***	−0.252	−0.078 ***
	Job Seeker	−0.931	−0.062 ***	−0.846	−0.056 ***	−0.860	−0.057 ***
	Unemployed	−0.439	−0.091 ***	−0.414	−0.086 ***	−0.425	−0.088 ***
	Others	1.046	0.023 *	1.008	0.022 *	1.036	0.023 *
Educational Level	>University Graduate
	<Elementary school	−0.775	−0.147 ***	−0.611	−0.116 ***	−0.617	−0.117 ***
	Middle-school graduate	−0.489	−0.134 ***	−0.364	−0.100 ***	−0.369	−0.101 ***
	High-school graduate	−0.257	−0.103 ***	−0.207	−0.083 ***	−0.215	−0.086 ***
Children	With
	Without	0.054	0.018	0.064	0.022	0.061	0.021
Type of Residence	Apartment
	Single-family house	0.045	0.016	0.051	0.018	0.048	0.017
	Townhouse/ Rowhouse	−0.206	−0.065 ***	−0.193	−0.060 ***	−0.184	−0.058 ***
	Other	0.024	0.002	−0.001	0.000	−0.009	−0.001
Regional Scale	Major City
	Small-to-medium city	−0.170	−0.062 ***	−0.165	−0.060 ***	−0.158	−0.057 ***
	Rural area	−0.151	−0.034 ***	−0.171	−0.039 ***	−0.179	−0.041 ***
Activity Diversity	Only Daily
	None			0.447	0.080 ***	0.487	0.087 ***
	Only same-day			−0.306	−0.097 ***	−0.299	−0.095 ***
	Only accommodation type			0.074	0.022 *	0.081	0.024 *
	Daily and same-say			0.135	0.027 **	0.153	0.031 **
	Daily and accommodation type			0.200	0.062 ***	0.203	0.063 ***
	Same-say and accommodation type			0.231	0.034 ***	0.249	0.037 ***
	All			0.056	0.012	0.065	0.014
Utilization Diversity	None
	All					−0.330	−0.072 ***
	Facility or program					0.101	0.036 ***
Model	R2	Adjusted R2	∆	F	p-value		Durbin- Watson
1	0.079	0.077	0.079	28.931	0.000
2	0.102	0.099	0.023	41.547	0.000
3	0.109	0.106	0.007	46.254	0.000		0.909

*: p < 0.01, **: p < 0.005, ***: p < 0.001.

). While the presence of positive autocorrelation may indicate potential bias in the model’s residuals, this analysis remains valid for exploratory purposes. Nevertheless, to enhance the reliability of the findings, addressing this autocorrelation issue through additional model modifications or the application of robust standard errors will be considered in future research.

In this study, we examined the factors that influenced the life satisfaction of the respondents through three models. To clearly evaluate the relative influence of each variable on the life satisfaction of the respondents, the standardized coefficient (β) was used for comparison and interpretation. The results are as follows.

The results indicated that among personal characteristics, some sub-categories of age, monthly household income, occupation, education level, and residential area size significantly influenced the life satisfaction of the respondents. Specifically, when examining the influence of standardized coefficients by variable, no significant difference in gender was found across all models, indicating that gender did not have a substantial effect on the results. Compared to the reference group of those aged 60 years and above, the individuals in their 30s (β = −0.044, p < 0.001), 40s (β = −0.030, p < 0.01), and 50s (β = −0.032, p < 0.005) showed significantly lower life satisfaction, suggesting relatively lower life satisfaction among middle-aged adults. For household income, using the reference category of KRW 4–5 million per month, the groups with lower income levels had significantly lower life satisfaction. Regarding occupation, housemakers (β = −0.073, p < 0.001), job seekers (β = −0.062, p < 0.001), and unemployed individuals (β = −0.091, p < 0.001) showed lower life satisfaction than clerks. Higher education levels were associated with higher life satisfaction, and residents of metropolitan cities had higher life satisfaction than those living in smaller cities or rural areas.

In terms of activity diversity, compared with daily day experiences, those who engaged in a combination of daily and same-day (β = 0.027, p < 0.005), daily and accommodation-type (β = 0.062, p < 0.001), or same-day and accommodation-type (β = 0.034, p < 0.001) activities reported significantly higher life satisfaction. However, individuals who engaged in all types of activities did not report significantly higher life satisfaction (β = 0.012, p > 0.05), suggesting that increased activity diversity does not linearly enhance satisfaction beyond a certain point. This implies that a balanced approach to activity participation may be more effective than maximizing diversity.

In terms of utilization diversity, compared with the group that did not use facilities or programs, those that used either facilities or programs reported higher life satisfaction (β = 0.036, p < 0.001), whereas those that used all facilities and programs reported lower life satisfaction (β = −0.330, p < 0.001). This suggests that while the use of forest recreation and welfare facilities and programs has some impact on life satisfaction, a more tailored approach to utilization is needed rather than focusing on intensive usage. Lastly, in Models 2 and 3, which controlled for personal characteristics, both the activity diversity and utilization diversity variables still had significant effects on life satisfaction. Additionally, when the utilization diversity variable was added to Model 3, the standardized coefficient for activity diversity increased slightly, indicating that activity and utilization diversity interacted and thus positively influenced life satisfaction. This suggests that combining facilities and program usage with diverse activities could have a greater effect on the well-being of users.

The hierarchical regression analysis results indicate that adding activity diversity and utilization diversity to the model significantly improved the explanatory power at each stage. These findings imply that promoting diverse and balanced forest recreation and welfare activities, along with the strategic use of facilities and programs, may contribute to enhancing the quality of life of the users. These results also emphasize the importance of considering different user groups when developing forest-related recreational and welfare programs. Instead of offering standardized programs to all users, tailored programs that reflect diverse needs and preferences should be offered to provide greater benefits to users.

4. Discussion

In this study, we analyzed the impact of the diversity of forest-related recreational and welfare activities on individuals’ quality of life and satisfaction. The findings suggest that both activity diversity and utilization diversity positively influence life satisfaction, which is consistent with prior studies showing that engaging in diverse activities enhances individuals’ physical, mental, and social well-being [43]. However, our study also revealed some important nuances that distinguish it from existing research. When we analyzed the influence of personal characteristics, such as age, monthly household income, and occupation, these factors were also found to have a significant effect on life satisfaction. Specifically, individuals in their 30s, 40s, and 50s showed significantly lower life satisfaction compared to those aged 60 and above, suggesting that middle-aged adults have relatively lower levels of life satisfaction. Furthermore, the groups with lower household incomes reported significantly lower life satisfaction, highlighting the important role that economic stability plays in overall well-being.

This suggests that leisure activities contribute to subjective well-being, but not all activities have the same effects on everyone, and the effects vary by age and gender [44]. Additionally, although leisure activities increase life satisfaction, the effect may vary depending on an individual’s mood and current life context [43]. While it is important to provide a variety of activities and experiences in a forest, quality of life can be further improved by developing and providing customized programs tailored to the needs of the target group.

Regarding the diversity of forest activities, our study found that activity diversity and utilization diversity both significantly influenced life satisfaction. However, an interesting finding of our study is that the group that experienced all types of forest activities did not report significantly higher levels of life satisfaction. This non-linear effect suggests that, beyond a certain point, simply increasing the number of activities does not necessarily enhance life satisfaction. This aligns with previous research suggesting that excessive engagement in terms of quantity or duration does not contribute significantly to well-being [45,46,47]. It is likely that a balanced and meaningful engagement in diverse activities yields greater benefits than merely engaging in more of them.

In this context, activity diversity and utilization diversity interact and affect life satisfaction in a more complex way than previously thought. The results of this study emphasize that while the diversification of activities is essential for enhancing life satisfaction, a tailored and personalized approach to forest welfare programs is equally important. This suggests that policymakers should not only increase the number of activities available but also ensure that these activities meet the specific needs of diverse demographic groups.

Our findings have significant policy implications. While diversifying forest activities is essential for enhancing life satisfaction, it is equally important to offer personalized programs that consider both individual characteristics and the diversity of activities and facilities offered. Specifically, activity diversity and utilization diversity should be considered when designing forest welfare programs to maximize their impact on well-being.

For instance, middle-aged adults, who exhibited lower life satisfaction, could benefit from targeted programs addressing work–life balance and family responsibilities. Similarly, low-income groups may benefit from programs focused on stress reduction and economic stability, helping to improve their quality of life.

Policymakers should focus on developing flexible, accessible, and participatory forest welfare programs that cater to diverse demographic groups. Specifically, policymakers should aim to provide a balanced combination of activities and facilities to meet the varied needs of users. For example, forest programs should not only include recreational activities but also therapeutic services for stress relief, mental health support, and social interaction. Furthermore, special attention should be paid to urban environments, where stress levels are higher and access to natural spaces may be limited.

By incorporating both activity diversity and utilization diversity, forest welfare programs can more effectively improve life satisfaction and promote well-being across different user groups. This approach will also contribute to the development of sustainable and inclusive forest welfare policies, benefiting a wide range of populations and fostering a more holistic approach to health and well-being. Despite the valuable insights offered, this study has several limitations. First, our reliance on self-reported data introduced the possibility of recall bias, which may have affected the accuracy of the responses. Since the survey collected data on forest recreation and welfare activities conducted over the previous year (2022), the respondents may have faced difficulties in recalling and accurately reporting their activities. This time lag between the activities and the survey could have led to recall bias, as respondents may not have fully remembered the frequency, duration, or types of activities they participated in. Additionally, the presence of positive autocorrelation in the data suggests potential issues regarding residual bias, which could affect the robustness of the results.

Future studies should incorporate more objective measures of activity participation, such as actual visit records, and apply robust statistical techniques to correct for autocorrelation. Furthermore, the cross-sectional nature of this study limits our ability to draw causal conclusions. Longitudinal studies would be necessary to explore the long-term effects of forest welfare activities on life satisfaction and overall well-being.

5. Conclusions

This study demonstrates that both activity diversity and utilization diversity positively influence life satisfaction. However, the relationship between these factors and life satisfaction is non-linear. While increasing the number of activities and facilities initially enhances satisfaction, beyond a certain point, additional activities do not lead to significant improvements in well-being. A more meaningful and balanced engagement with a variety of activities is crucial for maximizing life satisfaction.

Personal characteristics, such as age and income, also play a significant role in life satisfaction. Specifically, middle-aged adults and individuals with lower household incomes reported lower life satisfaction, suggesting a need for tailored programs that address these groups’ unique challenges.

The findings have important policy implications. Policymakers should focus not only on expanding the quantity of forest activities but also on ensuring that these activities meet the diverse needs of different demographic groups. Tailored, flexible, and participatory programs that address specific needs are essential for improving life satisfaction, especially for urban residents, in whom stress levels are higher.

Finally, this study reaffirms the importance of forest welfare activities in promoting nature engagement. Future studies should aim to explore the long-term effects of forest welfare activities on life satisfaction, particularly by including a more diverse population reflecting various socio-economic and cultural backgrounds.

Despite the valuable insights obtained, there are limitations of this study, such as its reliance on self-reported data, which could have introduced recall bias. Future research should incorporate more objective measures, such as actual visit records, and examine the long-term effects of forest welfare activities.