Next Article in Journal
Biofuels Induced Land Use Change Emissions: The Role of Implemented Land Use Emission Factors
Next Article in Special Issue
Satisfaction with Life and Health Behaviours in Pre-Seniors and Seniors: A Cross-Sectional Cohort Study
Previous Article in Journal
From Nano Zero-Valent Iron to Nanocomposite Materials for Sustainable Water Treatment
Previous Article in Special Issue
Age-Inclusive Healthcare Sustainability: Romania’s Regulatory and Initiatives Landscape in the European Union Context
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Sustainability
Volume 16
Issue 7
10.3390/su16072730
sustainability-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Leisure and Happiness of the Elderly: A Machine Learning Approach

by
Eui-Jae Kim
1,
Hyun-Wook Kang
1,* and
Seong-Man Park
2,*
1
Department of Recreation and Leisure Sports, College of Sport Science, Dankook University, Cheonan-si 31116, Republic of Korea
2
Department of English Language, College of Foreign Languages, Dankook University, Cheonan-si 31116, Republic of Korea
*
Authors to whom correspondence should be addressed.
Sustainability 2024, 16(7), 2730; https://doi.org/10.3390/su16072730
Submission received: 17 January 2024 / Revised: 14 March 2024 / Accepted: 24 March 2024 / Published: 26 March 2024
(This article belongs to the Special Issue Healthy Aging and Sustainable Development Goals)
Browse Figures
Versions Notes

Abstract

:
Leisure activities play an important role in improving happiness levels for the elderly. The purpose of this study is to explore leisure-related factors that affect the happiness of the elderly using machine learning algorithms. For this research, the 2019 National Leisure Activity Survey released by the Ministry of Culture, Sports and Tourism, Republic of Korea, was used to analyze the data of 1769 elders over the age of 65 among 10,060 men and women aged 15 years and older in 17 cities and provinces nationwide, and it went through the process of data preprocessing, data segmentation, prediction model construction and evaluation, and model tuning. According to the findings of the study, the main factors predicting the happiness index of the elderly were leisure life satisfaction, leisure time, whether to use public leisure facilities, leisure policy satisfaction, and leisure activity companionship. The overall findings of this study imply that exploring sustainable policy towards the achievement of sustainable happiness for the elderly is important. Based on these results, policy measures to improve the happiness level of the elderly were discussed.

1. Introduction

According to the UN’s World Population Prospects 2022, the proportion of the world’s population aged 65 and over is expected to reach 10% in 2022, and even 16% by 2050. In particular, Republic of Korea’s aging population is even more prominent. In 2020, Korea’s elderly population was 15.7%, and it is expected to exceed 25.5% in 2030, 34.4% in 2040, and 40% in 2050 [1]. Korea’s aging population is progressing at a very rapid pace, and by 2050, the level of population aging is predicted to be the highest in the world [2]. In this regard, Republic of Korea needs to prepare appropriate countermeasures to the cases of super-aging.
As we enter an aging society, interests in how to help the elderly live healthier and happier lives are very high. How satisfied are the elderly living in modern times with their lives? According to the 2022 World Happiness Report [3] recently released by the United Nations, in most countries, such as the United States, the United Kingdom, and Germany, the happiness index shows a U-shaped curve that decreases towards middle age and then rises as one enters older age. However, South Korea was different from other countries. The happiness index declined in midlife and then did not rise again. This shows that the happiness level of the elderly in Korea is relatively low compared to that of the elderly in other countries. In this regard, efforts are urgently needed to improve the quality of life and raise the level of happiness of the elderly in Korea.
In order to increase the level of happiness of the elderly, it is first necessary to identify the factors that affect the happiness of the elderly. In a number of studies, it has been shown that demographic and sociological factors such as income, gender, age, and education level [4,5,6,7]; physical factors such as objective and subjective health status and physical activity [8,9]; social factors such as social relationships, social status, and social support [10,11], as well as depression, anxiety, and self-efficacy; and psycho-emotional factors such as perception of aging [12,13,14,15] affect the happiness of the elderly.
Another concept that often appears as a factor that influences the happiness of the elderly is leisure. First, leisure activities themselves have a positive effect on the well-being of the elderly [16,17,18]. In addition, happiness levels vary depending on what kind of leisure activities are performed and with whom [19,20,21]. In addition, the frequency of leisure activities [22,23,24], leisure spaces and facilities [25,26], smartphones and internet use [27,28], and so-called leisure resources such as leisure time and expenses also affect happiness [29,30].
In addition to objective factors, subjective factors are also reported to influence the happiness of the elderly. In particular, it can be seen that happiness levels vary depending on the extent to which participation in leisure activities is restricted [31,32], and it can be found that satisfaction as a result of leisure activities also influences happiness [33,34].
A comprehensive review of the studies presented above shows that the happiness of the elderly is influenced by a variety of leisure factors. However, existing studies have the limitation of analyzing the relationship between leisure factors and happiness from a one-dimensional perspective. Of course, some studies have a multidimensional approach, but this also has the limitation of not being able to approach and analyze the leisure factors that affect happiness in a diverse and comprehensive way. Therefore, for a more in-depth discussion of leisure factors that predict happiness in the elderly, multidimensional studies that comprehensively consider various predictors are needed.
Therefore, this study seeks to examine the relationship between leisure and happiness among the elderly by comprehensively considering various predictive factors. In particular, machine learning techniques are applied in this study to improve the accuracy of prediction results. Machine learning is regarded as the core of predictive analysis, and it produces highly reliable prediction results while minimizing prediction errors [35]. In addition, it has the advantage of relatively little intervention from researchers and more accurate prediction and decision-making compared to traditional statistical techniques in that it predicts the outcome of new data by learning rules and patterns from existing data [36]. In this respect, building a model to predict the happiness of the elderly through machine learning is expected to have great value and implications in academic and practical terms.
In this regard, this study aims to explore leisure-related factors that affect the happiness of the elderly using machine learning algorithms. First, we will use various machine learning algorithms to build a model to predict the happiness of the elderly, compare each model, and adopt the model with the best predictive performance as the final model. In addition, we want to identify the relative importance of predictors in the final model to derive the main factors that affect the happiness of the elderly. Based on this analysis, we would like to explore sustainable policy measures to improve the happiness level of the elderly.

2. Literature Review

2.1. Concept of Happiness and Factors Affecting Happiness

Happiness is a very abstract term, even among various psychological terms. The concept of happiness is defined differently in each country depending on cultural and historical factors, and is also defined somewhat differently among scholars [37]. Diener (1984) [38] defined happiness as subjective well-being and argued that happiness is determined by subjective judgment about one’s life. Subjective well-being can be said to be distinct from other happiness-related concepts in that it emphasizes momentary satisfaction rather than overall satisfaction with life. Meanwhile, some researchers explain happiness in relation to life satisfaction. In this regard, Veenhoven (2010) [39] defined happiness as overall satisfaction with one’s entire life. Kalmijn and Arends (2010) [40] explained happiness as subjective appreciation of one’s life as a whole. As such, happiness includes various concepts ranging from momentary satisfaction with life to overall satisfaction.
Happiness is related to various factors. In particular, an individual’s income has been the most actively discussed area in explaining happiness [41,42]. One study found that higher personal income was positively associated with happiness in Organization for Economic Co-operation and Development (OECD) countries [43]. Individual health and social relationships are also important [44,45,46,47]. Higher levels of self-rated health were associated with higher levels of happiness [48,49], and good relationships with family, friends, and neighbors had a positive effect on happiness [50,51]. In addition, education [52], labor [53], and politics [54] are related to happiness. Furthermore, leisure is an important factor that affects happiness [55,56].

2.2. Leisure and Happiness of the Elderly

Leisure activities play an important role in enhancing our quality of life. Particularly, for the elderly, leisure activities serve as a resource that helps them adapt well to their new life after retirement. Leisure activities have a positive impact on the physical, social, and psychological health of the elderly [57]. Leisure activities reduce depression and anxiety levels among the elderly [58], promote neighborhood and family relationships, and improve quality of life [59]. Participating in leisure activities, especially for the elderly who suffer from loneliness, significantly contributes to their psychological health, which is effective in preventing and treating individual diseases [60]. In addition, leisure activities have a positive effect on the happiness of the elderly [61]. In this way, leisure plays a key role in improving the quality of life of seniors and improving their level of happiness. However, the relationship between leisure and happiness in the elderly may vary depending on the type of leisure activity, companion, frequency, time, space, and so on.

2.3. Leisure Factors Affecting the Happiness of the Elderly

Leisure factors affecting the happiness of the elderly can be broadly divided into objective factors and subjective factors. Objective factors include leisure activity type and companionship, frequency, cost, time, space, whether public leisure facilities are used, and time spent on smart devices.
Regarding the types of leisure activities, the elderly are more likely to be happier when they engage in active leisure activities such as physical activity, tourism, and cultural and artistic activities, rather than passive leisure activities such as watching TV and surfing the Internet [17,20,21,62,63,64]. The elderly who enjoy social leisure activities such as volunteering, clubs, and religious activities are said to have higher levels of happiness [19,65,66,67,68,69,70]. In addition, there are also studies that have emphasized the importance of passive leisure activities such as napping, watching TV, and reading, along with active leisure activities [71].
The companion factor of leisure activities is also related to the happiness of the elderly. It has been shown that the elderly who have many companions to spend their leisure time with have higher levels of happiness [72]. In addition, it is analyzed that the elderly who engage in leisure activities with family and friends are relatively happier than those who engage in leisure activities alone [19,73,74].
The elderly who participate in leisure activities more frequently have lower levels of depression and higher levels of happiness than those who do not [22,23,24]. On the other hand, there are studies that show that the happiness level of the elderly decreases as the frequency of participation in leisure activities increase [75], and some studies have shown that the frequency of participation in leisure activities itself does not determine the happiness level of the elderly [76].
Leisure resource factors such as leisure time and expenses also affect happiness in older adults. Much leisure time tends to reduce happiness levels [29,77]. This may be linked to results showing that economically active older adults have higher levels of life satisfaction and happiness compared to those who do not do so [26,66,78,79,80]. Some studies have shown that participation in economic activities does not significantly affect the happiness levels of the elderly [81,82,83]. Along with leisure time, leisure costs are an important factor affecting happiness. A number of studies have shown that the elderly who spend more on leisure consumption have higher levels of happiness [29,30,77,84].
Leisure space and facilities are also important factors. The elderly with higher participation in outdoor leisure activities are more likely to have higher life satisfaction than those with higher participation in indoor leisure activities [85], and the elderly with low participation in outdoor leisure activities are more likely to experience a sharp decline in physical function [86]. Increased outdoor leisure facilities in the community are associated with a reduced risk of depression in the elderly [87]. In addition, elderly people who use public leisure facilities have higher life satisfaction than seniors who use private leisure facilities [25], and the number of elderly care facilities and the happiness of the elderly are closely related [88]. On the other hand, there are also studies that show that the type of leisure facility does not significantly affect the happiness level of the elderly [77,89].
The use of smart devices also affects the happiness of the elderly. The elderly who use and are accustomed to smartphone use have higher levels of happiness compared to those who do not [28,90], and social media use and satisfaction improve happiness levels in the elderly [27,75,91,92]. Some studies have shown that the use of smart devices is not necessarily positive for the lives of the elderly [93,94,95,96,97].
In addition to objective factors, subjective factors are also reported to influence the happiness of the elderly. In particular, leisure constraints that limit leisure activities are closely related to happiness levels [32]. Leisure constraints reduce productive and active leisure activities in the elderly, and thus it is analyzed that elderly people who are highly restricted in their leisure activities have higher levels of depression and a low sense of self-integration [31,98].
Leisure satisfaction, the result of leisure activities, is a major predictor of happiness [99,100,101], and, in particular, the influence of leisure satisfaction on happiness may be greater in older adults [34]. A number of studies have shown that leisure satisfaction lowers levels of depression and improves well-being in the elderly [33,88,101].

3. Research Methods

3.1. Research Model

The objective of this study is to investigate leisure-related factors affecting the happiness of the elderly based on machine learning. Through the review of prior research, a research model was constructed, as shown in Figure 1.

3.2. Data Analysis and Target Population

In this study, the National Leisure Activity Survey, a national statistic, was used as analysis data. The National Leisure Activity Survey is a state-approved statistic that ‘aims to analyze the actual state of leisure activities of the people to understand changes in lifestyle and quality of life, and to use them as basic data for the formulation of related policies’ [102,103]. Currently, the data of the 2021 National Leisure Activity Survey have also been released; however, considering that the survey period is during the COVID-19 period (it is believed that COVID-19 may have had a significant impact on the survey results), the 2019 National Leisure Activity Survey data were selected for analysis. The population of the 2019 National Leisure Activity Survey is 10,060 men and women aged 15 years and older in 17 cities and provinces nationwide. In this study, 1769 elderly people over the age of 65 were selected as the final study population. Specific information about the analysis data is as follows:
  • Survey structure
-
Survey agency: Ministry of Culture, Sports and Tourism
-
Survey subject: Population over 15 years old nationwide
-
Number of valid respondents: 10,060
-
Survey period: 9 September 2019–14 November 2019
-
Survey method: Household visit interviews in which a professional surveyor visits a sample of households and fills in the responses to the questionnaire
-
Legal basis: Statistics of approval designated by the National Statistics Agency (approval number: No. 113014)
  • Sampling design
-
Target population: Population over 15 years old nationwide
-
Survey population: Members of households over the age of 15 who reside in all households in Korea at the time of the survey
-
Sampling frame: Use of data from the National Statistics Office’s ‘2017 Census’ survey
-
Stratification: A total of 17 cities and provinces are stratified into urban cities and rural areas which divided into dong (i.e., towns) and eup/myeon (i.e., villages), reflecting the characteristics of urban and rural areas.
-
Sampling method: Stratified Multi-Stage Cluster Sampling
-
Sample allocation and extraction method: Random extraction after allocation by square root proportional distribution in consideration of the number of households in the trial by the precision and adequacy of the sample, etc.
  • Weighting
Final weight: Design weight × non-response adjustment factor × (1/in-household extraction rate) × population information adjustment factor.

3.3. Analytic Variables

3.3.1. Result Variable

The variable set as the result variable is the ‘happiness index’. The happiness index uses the following questionnaire: ‘How happy do you think you are right now?’ The questionnaire is structured to respond on a 10-point scale (1 = ‘very unhappy’, 10 = ‘very happy’), and higher response values indicate a higher level of happiness.

3.3.2. Predictor Variables

The variables set as predictors are ‘the type of leisure activity in which the most participated, companionship, frequency, cost (based on one time), leisure cost (monthly average), leisure time, most used leisure space, use of public leisure facilities, participation in club activities, volunteer activity, time spent on smart devices, leisure policy satisfaction, overall leisure life satisfaction, and leisure activity constraints’.
The type of leisure activity in which the most participated, companionship, frequency, space (place), use of public leisure facilities, participation in club activities, and volunteer activity were set as categorical variables, and leisure cost (one-time basis, monthly average), leisure time, time spent on smart devices, leisure policy satisfaction, overall leisure life satisfaction, and leisure activity constraints were set as continuous variables. Specific information about the predictors is shown in Table 1.

3.4. Analysis Method

The data analysis method went through the processes of data preprocessing, data partitioning, building and evaluating predictive models, and hyperparameter optimization. The specific analysis process is shown in Figure 2.

3.4.1. Data Preprocessing

One-Hot Encoding

As a preprocessing procedure for data analysis, categorical data were converted to one-hot encoding. Then, considering the multicollinearity problem among the transformed variables, one dummy variable was removed. The list of removed variables (baseline variables) is shown in Table 2.

Data Scaling

In order to match the range of different data values to a certain level, the data values were standardized to an average of 0 and a variance of 1.

Multicollinearity

As a method for identifying multicollinearity problems, the VIF (Variance Inflation Factor) was identified. The multicollinearity criterion was a VIF of 10 or higher. As a result of calculating the VIF, it was found that there was no multicollinearity problem, with a minimum of 1.04 to a maximum of 1.59.

Data Partitioning

In order to generalize the predictive model, the analysis was conducted by dividing the entire data into a training data set and a test data set. The data split ratio was applied as 70% for training data and 30% for evaluation data.

Building and Evaluating Predictive Models

In this study, based on eight supervised learning algorithms (Linear Regression, Ridge, Lasso, SVM, Decision Tree, Random Forest, XGBoost, and LightGBM), a model was built to predict the happiness index of the elderly, and the performance of each model was compared. The performance evaluation of the model used the typical performance indicators of the regression model: R2 (r-squared) and root mean square error (RMSE). R2 refers to a coefficient that measures the extent to which the regression line estimated in regression analysis explains the actual sample. In other words, it refers to an indicator that shows how well the predictive (independent) variable explains the resulting (dependent) variable. The range is between 0 and 1, and a value of 1 means that the regression line perfectly matches the data. Conversely, if the coefficient of determination is 0, it means that the regression line does not explain the distribution of the data at all. RMSE is a generalized measure of standard deviation and represents the average difference between the value predicted by the prediction model and the actual value. Therefore, the lower the value of root mean square error, the better the prediction model.

Hyperparameter Optimization

Hyperparameters were optimized to improve the performance of previously built predictive models. As a method for optimizing hyperparameters, grid search was used. Grid search refers to a method of specifying certain values for each hyperparameter and learning data for all combinations of specified values to search for a combination of hyperparameters that represent optimal performance indicators. In this study, the hyperparameters were optimized for each algorithm, and the model with the best performance was adopted as the final model. The final model was then used to determine the relative importance of the predictors.

Analysis Tools

Python 3.7 and jupyter notebooks were used to perform machine learning.

4. Research Results

4.1. Model Performance Evaluation

The performance of the machine learning model was evaluated by applying the predictive model built from the training data set to the test data set. The prediction results for each algorithm are presented in Table 3. In the case of RMSE, it ranged between 1.200–1.638, followed by SVM, Ridge, Linear Regression, Random Forest, LightGBM, XGBoost, Lasso, and Decision Tree. In the case of R2, it was distributed between −0.278–0.314, followed by Decision Tree, Lasso, XGBoost, LightGBM, Random Forest, Linear Regression, Ridge, and SVM. After applying eight algorithms, the SVM algorithm performed best (RMSE = 1.200, R2 = 0.314). Figure 3 and Figure 4 visualize the model performance evaluation.

4.2. Hyperparameter Optimization

In order to improve the accuracy of the machine learning model, the optimal hyperparameters for each algorithm were searched and applied. After hyperparameter optimization, the results for model performance are shown in Table 4 and Figure 5 and Figure 6. In the case of RMSE in Table 4, it was distributed between 1.195–1.225, followed by Lasso, XGBoost, LightGBM, SVM, Ridge, Linear Regression, Random Forest, and Decision Tree. R2 ranged between 0.284–0.319, followed by Decision Tree, Random Forest, Linear Regression, Ridge, SVM, LightGBM, XGBoost, and Lasso. After optimizing the hyperparameters, the Lasso algorithm performed best (RMSE = 1.195, R2 = 0.319). Therefore, in this study, the Lasso model with the best predictive performance was adopted as the final model.

4.3. Importance of Predictor Variables

The Lasso algorithm analyzed the relative importance of variables that predict the happiness index of the elderly. Out of a total of fourteen variables added to the model, the top five variables with relatively high predictive power were extracted, as shown in Figure 7. The important predictors were leisure life satisfaction (q30), leisure time (q13), whether to use public leisure facilities (q19), leisure policy satisfaction (q27), and leisure activity companions (q2_2_1_1). Looking specifically at these, it has been shown that the higher the satisfaction with leisure life, the less leisure time, the more public leisure facilities that are used, the higher the satisfaction with leisure policies, and the more leisure activities with others, the higher the happiness index is.

5. Discussion and Conclusions

The purpose of this study is to explore leisure-related factors that affect the happiness of the elderly using machine learning algorithms. Through this study, we wanted to propose sustainable policy measures to improve the happiness level of the elderly. The main findings of this study are summarized and discussed as follows.
The performance of each model in predicting happiness among the elderly was found to be relatively consistent. This indicates that the prediction model developed in this study demonstrates a degree of stability. However, it is regrettable that the model’s performance does not reach an exemplary level. Considering the challenge in interpreting RMSE [104], an evaluation of the model’s performance based on R-square reveals that the model constructed in this study accounts for approximately 32% of the total explained variance. The suboptimal performance of the model appears to be associated with the selection of predictor variables. Consequently, it is imperative to exercise greater care in selecting predictors and to endeavor to formulate a more robust model by leveraging the insights gleaned from prior research findings. Moreover, concerted efforts are required to enhance the model’s accuracy by continually exploring variables that were not considered in this study.
The top five leisure factors that predict the happiness of the elderly through machine learning were extracted, and they were leisure life satisfaction, leisure time, use of public leisure facilities, leisure policy satisfaction, and leisure activity companionship. Prior studies in all age groups have shown that leisure time, leisure purpose, leisure companionship, degree of work–leisure balance, and leisure cost are the main predictors of happiness [105]. This shows that the time and money required for leisure activities, so-called leisure resources, among relatively young people are determinants of happiness, while in older people, satisfaction with leisure life and leisure policy factors play an important role. In this regard, policy differentiation is required in the formulation and promotion of leisure policies.
On the other hand, leisure satisfaction has been shown to be the most important factor in determining the happiness of the elderly. Many studies highlight the importance of leisure satisfaction in the quality of life of the elderly [34,73,106,107]. Our research, including prior research, strongly supports that leisure satisfaction, the outcome of leisure activities, is the most important determinant of happiness in the elderly. Therefore, in order to increase the happiness level of the elderly, efforts must be made to increase leisure satisfaction. Leisure satisfaction is influenced by a combination of factors. Serious leisure participation of the elderly maximizes leisure satisfaction [107,108], and family capital plays an important role in leisure satisfaction [109]. In particular, leisure education contributes significantly to increasing leisure satisfaction among the elderly [110]. Therefore, education will be needed to enable the elderly to participate in leisure activities in a more active and serious manner, and furthermore, the development of various family leisure programs to strengthen family ties will be required. The results of this study emphasize the importance of education for the elderly to be able to participate in leisure activities in a more effective and active manner, and thus the need for sustainable policy support to foster it.
The results of an increase in happiness levels as leisure time decreases are noteworthy. These results are consistent with previous studies [29,77] that revealed an inverse relationship between leisure time and happiness index among the elderly. It is also in line with previous studies that have shown that economically active seniors have higher levels of happiness compared to those who do not do so [26,79,80]. This speaks to the importance of quality over quantity of leisure time. Having a lot of free time does not necessarily mean being happy, and how you greet and use that time is important. The results of studies showing that elderly people with relatively little leisure time enjoy being active and engaging in active leisure, such as sports activities and tourism activities [106], also support the findings of this study to some extent.
The availability of public leisure facilities has been identified as a major factor affecting the happiness of the elderly. The results show that the elderly who use public leisure facilities have higher levels of happiness compared to those who do not use them. These results support a prior study [25] that found that the elderly who use public leisure facilities have higher life satisfaction than older adults who use private leisure facilities. The results of this study, which proved the value of public leisure facilities, suggest that the government should pay a lot of attention and effort to leisure facilities for the elderly. Regarding leisure welfare facilities for the elderly, experts point to the lack of the number of facilities, the regional ubiquity of facilities, the uniformity of leisure programs, and the lack of professional personnel to systematically manage and supervise them [111,112]. Therefore, it is likely that policies will be needed that take into account the expansion of balanced leisure welfare facilities for the elderly among regions, diversification of leisure programs, and training leisure-related experts for the elderly.
Another interesting finding is that satisfaction with leisure policies has been shown to be a major factor in the happiness of older adults. In particular, it is noteworthy that, unlike prior studies of all age groups [105], leisure policy factors have been shown to be the main predictors of happiness. This shows that the role of the government, as the main body of national policy, is paramount for the happiness of the elderly. Therefore, the government needs to put more effort into formulating leisure welfare policies that can improve the quality of leisure for the elderly.
In addition, leisure activity companions were found to be a major factor affecting the happiness of the elderly, and it was confirmed that elderly people who engage in leisure activities with family, friends, neighbors, etc. have a higher level of happiness than those who do leisure activities alone. These results support prior research that emphasized the importance of accompanying leisure activities, such as with family, neighbors, and friends [19,73,113]. In terms of the life cycle, old age experiences social isolation through the loss of a spouse, the disconnection of social relationships with neighbors and friends, as well as the shrinking of social networks from a social structural perspective [114]. Therefore, leisure activities with family, friends, neighbors, and so on, during this period are likely to have had a significant impact on the happiness of the elderly. Recently, Korean society has seen a phenomenon of individualization of leisure that prefers leisure done alone rather than collective leisure. Since 2012, solo leisure has surpassed companion leisure time with others [115]. The phenomenon of personalization of leisure has both positive and negative aspects. However, considering the characteristics of old age, which experiences physical and social limitations, the phenomenon of the personalization of leisure by the elderly has more negative aspects than positive aspects. The results of this study, which show that elderly people who engage in leisure activities with others have higher levels of happiness than those who engage in leisure activities alone, highlight the importance of accompanying leisure for the elderly and the need for sustainable policy support and development to promote it.
This study was based on machine learning algorithms to explore leisure factors that are highly related to happiness in the elderly. Through this, it is possible to focus more on leisure factors and discuss what efforts should be made to improve the happiness of the elderly. However, this study leaves a few things to be desired from an academic perspective. First, this study has the limitation of not conducting an in-depth analysis of the study subjects. In the future, if an in-depth analysis is carried out according to the characteristics of the elderly, such as gender, income, and region of residence, it will be possible to have a more in-depth discussion on the leisure factors that affect the happiness of the elderly. Second, this study used cross-sectional data. If longitudinal studies analyze how factors affecting the happiness of the elderly appear over time, there will be a different meaning of academic significance. Lastly, it would be desirable to enhance the performance of the model constructed in this study to reach an exemplary standard in future studies. This suggests the possibility of the existence of leisure factors that can better predict the happiness of the elderly. Therefore, it is hoped that a research model that can provide a more comprehensive understanding of the happiness of the elderly can be constructed by observing the relationship between leisure and happiness among the elderly from diverse perspectives. Despite these limitations, it is significant that this study used highly reliable and nationally representative national statistical data to construct a model for predicting happiness among the elderly. In particular, it can be said that the use of machine learning, which is evaluated as the core of predictive analysis, was able to improve the accuracy of research results. Finally, we hope that this study will be used as a basis for formulating and promoting leisure policies.
Overall, the findings of this study will be able to contribute to the increase of awareness about the relationship between leisure factors and happiness, and eventually to both mental and physical health related features of sustainability among the elderly in the Republic of Korea.

Author Contributions

Conceptualization, E.-J.K. and H.-W.K.; methodology, E.-J.K.; validation, E.-J.K. and H.-W.K.; formal analysis, E.-J.K.; investigation, E.-J.K. and H.-W.K.; data curation, E.-J.K. and H.-W.K.; writing—original draft preparation, E.-J.K. and H.-W.K.; writing—review and editing, S.-M.P.; visualization, E.-J.K.; supervision, H.-W.K.; project administration, H.-W.K. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Conflicts of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

References

  1. Statistics Korea. Future Population Projections. 2021. Available online: https://kostat.go.kr/portal/korea/kor_nw/1/2/6/index.board (accessed on 1 September 2022).
  2. United Nations. World Population Ageing 2019; UN: New York, NY, USA, 2020.
  3. Helliwell, J.F.; Layard, R.; Sachs, J.D.; De Neve, J.-E.; Aknin, L.B.; Wang, S. (Eds.) World Happiness Report 2022; Sustainable Development Solutions Network: New York, NY, USA, 2022.
  4. Kim, J.Y. Path Analysis Amongs Education level, Employment Status and Self-Esteem Effecting on the Life Satisfaction of the Elderly. Korean J. Gerontol. Soc. Welf. 2017, 72, 167–190. [Google Scholar]
  5. Chirinda, W.; Phaswana-Mafuya, N. Happy life expectancy and correlates of happiness among older adults in South Africa. Aging Ment. Health 2019, 23, 1000–1007. [Google Scholar] [CrossRef] [PubMed]
  6. Didino, D.; Taran, E.A.; Gorodetski, K.; Melikyan, Z.A.; Nikitina, S.; Gumennikov, I.; Casati, F. Exploring predictors of life satisfaction and happiness among Siberian older adults living in Tomsk Region. Eur. J. Ageing 2018, 15, 175–187. [Google Scholar] [CrossRef] [PubMed]
  7. Zhao, C.; Guo, J. Are veterans happy? Long-term military service and the life satisfaction of elderly individuals in China. J. Happiness Stud. 2022, 23, 477–508. [Google Scholar] [CrossRef]
  8. Lobos, G.; Grunert, K.G.; Bustamante, M.; Schnettler, B. With health and good food, great life! Gender differences and happiness in Chilean rural older adults. Soc. Indic. Res. 2016, 127, 865–885. [Google Scholar] [CrossRef]
  9. Steptoe, A. Happiness and health. Annu Rev Public Health 2019, 40, 339–359. [Google Scholar] [CrossRef] [PubMed]
  10. Dorji, N. Older Adults’ Happiness and Quality of Life: Spiritualism, Social Connectedness, and Health Status. In Handbook of Health and Well-Being; Springer: Singapore, 2022; pp. 437–451. [Google Scholar]
  11. Saber, M.; Rashedi, V.; FadakarDavarani, M.M.; Borhaninejad, V. Social Support, Happiness, and Self-Rated Health among Older Adults: A Population-Based Study. Adv. Gerontol. 2021, 11, 22–28. [Google Scholar] [CrossRef]
  12. Kang, M.H. The Basic Pension System and Happiness of the Elderly Women in Poverty: Focusing on the Mediating Effect of Self-Efficacy. Korean J. Converging Humanit. 2019, 7, 7–29. [Google Scholar]
  13. Arias-Monsalve, A.M.; Arias-Valencia, S.; Rubio-León, D.C.; Aguirre-Acevedo, D.C.; Rengifo, L.R.; Cortes JA, E.; Paredes-Arturo, Y.V. Factors Associated with Happiness in Rural Older Adults: An Exploratory Study. Int. J. Psychol. Res. 2022, 15, 70–83. [Google Scholar] [CrossRef]
  14. Ingrand, I.; Paccalin, M.; Liuu, E.; Gil, R.; Ingrand, P. Positive perception of aging is a key predictor of quality-of-life in aging people. PLoS ONE 2018, 13, e0204044. [Google Scholar] [CrossRef]
  15. Luchesi, B.M.; de Oliveira, N.A.; de Morais, D.; de Paula Pessoa, R.M.; Pavarini SC, I.; Chagas MH, N. Factors associated with happiness in the elderly persons living in the community. Arch. Gerontol. Geriatr. 2018, 74, 83–87. [Google Scholar] [CrossRef] [PubMed]
  16. Kaliterna Lipovčan, L.; Brkljačić, T.; Prizmić Larsen, Z.; Brajša-Žganec, A.; Franc, R. Leisure activities and the subjective well-being of older adults in Croatia. GeroPsych J. Gerontopsychol. Geriatr. Psychiatry 2018, 31, 31. [Google Scholar] [CrossRef]
  17. Lee, H.; Heo, S. Benefits of leisure activities for the quality of life of older South Korean adults. Leis. Stud. 2021, 40, 199–211. [Google Scholar] [CrossRef]
  18. Yoshida, Y.; Iwasa, H.; Ishioka, Y.; Suzukamo, Y. Leisure activity moderates the relationship between living alone and mental health among J apanese older adults. Geriatr. Gerontol. Int. 2021, 21, 421–425. [Google Scholar] [CrossRef] [PubMed]
  19. Jung, S.M.; Lee, Y.S. The Effect of Leisure Time and Leisure Companions on Life Satisfaction for Retired Elderly Couples: Focusing on Gender Differences. Fam. Cult. 2022, 34, 101–138. [Google Scholar]
  20. Mukherjee, P.; Ghosh, A.; Dasgupta, S. Role of leisure activity in quality of life and meaning in life of male and female older adults. J. Geriatr. Ment. Health 2022, 9, 4. [Google Scholar]
  21. Wion, R.K.; Fazeli, P.L.; Vance, D.E. The association between leisure activity engagement and health-related quality of life in middle-aged and older people with hiv. Gerontologist 2022, 62, 1018–1028. [Google Scholar] [CrossRef]
  22. Mito, H.; Kinjo, H. A cross-sectional study on relationships between musical activities and quality of life in Japanese older adults. Res. Stud. Music Educ. 2022, 45, 37–55. [Google Scholar] [CrossRef]
  23. Moon, J.; Lee, W.S.; Shim, J. Exploring Korean Middle-and Old-Aged Citizens’ Subjective Health and Quality of Life. Behav. Sci. 2022, 12, 219. [Google Scholar] [CrossRef]
  24. Wang, J.; Xu, J.; Nie, Y.; Zhang, X.; Li, Y.; Wu, Q. Effects of Social Participation and Its Diversity, Frequency, and Type on Depression in Middle-Aged and Older Persons: Evidence From China. Front. Psychiatry 2022, 13, 825460. [Google Scholar] [CrossRef]
  25. Park, S.M.; Mun, S.Y. The Effect of Participation in Social Activities on the Subjective Health Satisfaction of the Older Adults with and without Chronic Illnesses. Korean Soc. Health Serv. Manag. 2018, 12, 113–123. [Google Scholar] [CrossRef]
  26. Cha, M. Factors Influencing Happiness of Elderly People with Cognitive Impairment in Korea. NVEO Nat. Volatiles Essent. Oils J. 2021, 8, 5083–5096. [Google Scholar]
  27. Gaia, A.; Sala, E.; Cerati, C. Social networking sites use and life satisfaction: A quantitative study on older people living in Europe. Eur. Soc. 2021, 23, 98–118. [Google Scholar] [CrossRef]
  28. Thaithatkul, P.; Chalermpong, S.; Laosinwattana, W.; Kato, H. Mobility, Activities, and Happiness in Old Age: Case of the Elderly in Bangkok. Case Stud. Transp. Policy 2022, 10, 1462–1471. [Google Scholar] [CrossRef]
  29. Min, K.S. Effects of Leisure Resources on Leisure Life Satisfaction and Happiness in the Elderly:Focusing on analysis of difference by household types. Korean J. Leis. Recreat. Park 2020, 44, 45–57. [Google Scholar] [CrossRef]
  30. Choung, Y.; Pak, T.Y.; Chatterjee, S. Consumption and life satisfaction: The Korean evidence. Int. J. Consum. Stud. 2021, 45, 1007–1019. [Google Scholar] [CrossRef]
  31. Lee, M.S. The relationship among dancesports participants’ restriction on exercise, depression and ego integrity of the elderly in COVID-19. Off. J. Korean Soc. Danc. Sci. 2022, 39, 29–41. [Google Scholar]
  32. Chick, G.; Hsu, Y.C.; Yeh, C.K.; Hsieh, C.M. Leisure constraints, leisure satisfaction, life satisfaction, and self-rated health in six cities in Taiwan. Leis. Sci. 2015, 37, 232–251. [Google Scholar] [CrossRef]
  33. Kim, I.H.; Kim, C.S. “Leisure Life Satisfaction”: Will It Have a Beneficial Impact on Depression Among Older Adults in Community Care Settings in Korea? J. Prev. Med. Public Health 2022, 55, 398–406. [Google Scholar] [CrossRef]
  34. Kuykendall, L.; Tay, L.; Ng, V. Leisure engagement and subjective well-being: A meta-analysis. Psychol. Bull. 2015, 141, 364. [Google Scholar] [CrossRef]
  35. Kwon, C.M. Python Machine Learning Perfect Guide; Wikibooks: Paju, Republic of Korea, 2019. [Google Scholar]
  36. Géron, A. Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems; O’Reilly Media, Inc.: Sebastopol, CA, USA, 2019. [Google Scholar]
  37. Oishi, S.; Graham, J.; Kesebir, S.; Galinha, I.C. Concepts of happiness across time and cultures. Personal. Soc. Psychol. Bull. 2013, 39, 559–577. [Google Scholar] [CrossRef] [PubMed]
  38. Diener, E. Subjective well-being. Psychol. Bull. 1984, 95, 542. [Google Scholar] [CrossRef]
  39. Veenhoven, R. Greater happiness for a greater number: Is that possible and desirable? J. Happiness Stud. 2010, 11, 605–629. [Google Scholar] [CrossRef]
  40. Kalmijn, W.M.; Arends, L.R. Measures of inequality: Application to happiness in nations. Soc. Indic. Res. 2010, 99, 147–162. [Google Scholar] [CrossRef] [PubMed]
  41. Wu, Y.; Cao, Y. Research on the influence of ecological environment satisfaction and income level on Chinese residents’ happiness: Empirical analysis based on CGSS data. Sustainability 2023, 15, 8175. [Google Scholar] [CrossRef]
  42. Zhu, J.; Liang, C.; Lucas, J.; Cheng, W.; Zhao, Z. The influence of income and social capital on the subjective well-being of elderly Chinese people, based on a panel survey. Sustainability 2020, 12, 4786. [Google Scholar] [CrossRef]
  43. Frank, R.H. Does money buy happiness. In The Science of Well-Being; Oxford University Press: Oxford, UK, 2005; pp. 461–473. [Google Scholar]
  44. Cheon, H. The structural relationship between exercise frequency, social health, and happiness in adolescents. Sustainability 2021, 13, 1050. [Google Scholar] [CrossRef]
  45. Koçak, O.; Ak, N.; Erdem, S.S.; Sinan, M.; Younis, M.Z.; Erdoğan, A. The role of family influence and academic satisfaction on career decision-making self-efficacy and happiness. Int. J. Environ. Res. Public Health 2021, 18, 5919. [Google Scholar] [CrossRef]
  46. Layard, R. Happiness: Lessons From a New Science; Penguin Press: London, UK, 2005. [Google Scholar]
  47. Zhu, S.; Li, M.; Zhong, R.; Coyte, P.C. The effects of co-residence on the subjective well-being of older Chinese parents. Sustainability 2019, 11, 2090. [Google Scholar] [CrossRef]
  48. Frey, B.S.; Stutzer, A. What can economists learn from happiness research? J. Econ. Lit. 2002, 40, 402–435. [Google Scholar] [CrossRef]
  49. Leyden, K.M.; Goldberg, A.; Michelbach, P. Understanding the pursuit of happiness in ten major cities. Urban Aff. Rev. 2011, 47, 861–888. [Google Scholar] [CrossRef]
  50. Helliwell, J.F.; Putnam, R.D. A well-being manifesto for a flourishing society. In The Science of Well-Being; Oxford University Press: Oxford, UK, 2005; pp. 435–460. [Google Scholar]
  51. Putnam, R.D. Bowling alone: The collapse and revival of American community. J. Cathol. Educ. 2000, 6, 12. [Google Scholar]
  52. Liang, C.; Sun, J. A study of the happiness of chinese university students and Its influencing factors—A case study of Beijing universities. Sustainability 2022, 14, 16057. [Google Scholar] [CrossRef]
  53. Haar, J.; Schmitz, A.; Di Fabio, A.; Daellenbach, U. The role of relationships at work and happiness: A moderated moderated mediation study of New Zealand managers. Sustainability 2019, 11, 3443. [Google Scholar] [CrossRef]
  54. Li, Q.; An, L. Corruption takes away happiness: Evidence from a cross-national study. J. Happiness Stud. 2020, 21, 485–504. [Google Scholar] [CrossRef]
  55. Chen, Q.; Chou, C.Y.; Chen, C.C.; Lin, J.W.; Hsu, C.H. The effect of leisure involvement and leisure satisfaction on the well-being of pickleball players. Sustainability 2021, 14, 152. [Google Scholar] [CrossRef]
  56. Sanagustín-Fons, M.V.; Tobar-Pesántez, L.B.; Ravina-Ripoll, R. Happiness and cultural tourism: The perspective of civil participation. Sustainability 2020, 12, 3465. [Google Scholar] [CrossRef]
  57. Kim, J.; Yamada, N.; Heo, J.; Han, A. Health benefits of serious involvement in leisure activities among older Korean adults. Int. J. Qual. Stud. Health Well-Being 2014, 9, 24616. [Google Scholar] [CrossRef]
  58. Urzeala, C.; Popescu, V.; Courteix, D.; Mitrache, G.; Roco, M.; Teodorescu, S. Barriers and facilitators for the Romanian older adults in enjoying physical activity health-related benefits. Sustainability 2021, 13, 12511. [Google Scholar] [CrossRef]
  59. Zhang, L.; Zhang, J. Impacts of leisure and tourism on the elderly’s quality of life in intimacy: A comparative study in Japan. Sustainability 2018, 10, 4861. [Google Scholar] [CrossRef]
  60. Kim, J.; Lee, S.; Chun, S.; Han, A.; Heo, J. The effects of leisure-time physical activity for optimism, life satisfaction, psychological well-being, and positive affect among older adults with loneliness. Ann. Leis. Res. 2017, 20, 406–415. [Google Scholar] [CrossRef]
  61. Lin, Y.T.; Chen, M.; Ho, C.C.; Lee, T.S. Relationships among leisure physical activity, sedentary lifestyle, physical fitness, and happiness in adults 65 years or older in Taiwan. Int. J. Environ. Res. Public Health 2020, 17, 5235. [Google Scholar] [CrossRef] [PubMed]
  62. Ryu, J.R. Effects of Relational Goods on Happiness of Elderly People in Korea. Korean Soc. Secur. Stud. 2017, 33, 29–57. [Google Scholar]
  63. Yamashita, T.; Bardo, A.R.; Liu, D. Experienced subjective well-being during physically active and passive leisure time activities among adults aged 65 years and older. Gerontologist 2018, 59, 718–726. [Google Scholar] [CrossRef] [PubMed]
  64. Lee, J.H.; Lee, J.H.; Park, S.H. Leisure activity participation as predictor of quality of life in Korean Urban-dwelling elderly. Occup. Ther. Int. 2014, 21, 124–132. [Google Scholar] [CrossRef] [PubMed]
  65. Kim, K.W.; Hwang, H.S. A Study on the Influence Factors of Happiness in single-person elderly households by Age Group of the Elderly. Korean J. Gerontol. Soc. Welf. 2021, 76, 45–70. [Google Scholar] [CrossRef]
  66. Kim, J.H. The Effect of Social Activities on Life Satisfaction in the Elderly Living in Jeju: A Comparison Household Types. J. Community Welf. 2021, 77, 1–31. [Google Scholar] [CrossRef]
  67. Lee, H.K. Volunteer activity changes after middle years old and the influences of volunteering on well-being. Innov. Aging 2019, 3, S178. [Google Scholar] [CrossRef]
  68. Cho, D.; Post, J.; Kim, S.K. Comparison of passive and active leisure activities and life satisfaction with aging. Geriatr. Gerontol. Int. 2018, 18, 380–386. [Google Scholar] [CrossRef]
  69. Gil IS GA, I.; Gil-Lacruz, L.M. Active aging and happiness: European senior citizen volunteers. Sociológia 2019, 51, 290–312. [Google Scholar]
  70. Ryu, J.; Heo, J. Relationships between leisure activity types and well-being in older adults. Leis. Stud. 2018, 37, 331–342. [Google Scholar] [CrossRef]
  71. Oerlemans, W.G.; Bakker, A.B.; Veenhoven, R. Finding the key to happy aging: A day reconstruction study of happiness. J. Gerontol. Ser. B Psychol. Sci. Soc. Sci. 2011, 66, 665–674. [Google Scholar] [CrossRef] [PubMed]
  72. Oh, Y.E.; Lee, J. Study on variables affecting rural elderly’s self-efficacy: Focused on mediating effect of social capital. Korean J. Community Living Sci. 2017, 28, 561–579. [Google Scholar] [CrossRef]
  73. Hodge, C.; Bocarro, J.N.; Henderson, K.A.; Zabriskie, R.; Parcel, T.L.; Kanters, M.A. Family leisure: An integrative review of research from select journals. J. Leis. Res. 2015, 47, 577–600. [Google Scholar] [CrossRef]
  74. Park, K.R. The effect of leisure activities and leisure satisfaction on subjective well-being of the pre-elderly. Korean J. Hum. Ecol. 2009, 18, 287–300. [Google Scholar] [CrossRef]
  75. Kim, S.H. A Study on the Difference between the Factors Affecting Happiness between the Baby Boom Generation and the Elderly Generation. J. Contents 2021, 21, 591–602. [Google Scholar] [CrossRef]
  76. Kim, E.K.; Ha, K.S. The Effects of Social Participation and Social Support on Quality of Life of the Elderly. J. Digit. Converg. 2013, 11, 811–820. [Google Scholar] [CrossRef]
  77. Park, C.J. The Factors of Leisure Affecting Happiness of the Elderly by Sex in Korea. J. Korean Gerontol. Soc. 2020, 40, 163–178. [Google Scholar] [CrossRef]
  78. Kwon, H.C.; Yeum, D.M.; Cho, H.C. Are the Working Elderly Happy?: Longitudinal Mediating Effect of Self-esteem. J. Humanit. Soc. Sci. 21 2020, 11, 67–76. [Google Scholar]
  79. Park, S.M.; Park, S.A. The Factors Associated with Depression in Pre-elderly and Young-elderly. J. Korea Acad.-Ind. Coop. Soc. 2022, 23, 519–529. [Google Scholar] [CrossRef]
  80. Kolosnitsyna, M.; Khorkina, N.; Dorzhiev, H. Determinants of life satisfaction in older Russians. Ageing Int. 2017, 42, 354–373. [Google Scholar] [CrossRef]
  81. Kim, H.Y. Happiness Inequality in Old Age. Korean J. Soc. Welf. 2020, 72, 97–123. [Google Scholar] [CrossRef]
  82. Park, C.J.; Shin, D.H. The Factors Affecting Happiness in Older Rural Residents. Health Soc. Welf. Rev. 2018, 38, 98–128. [Google Scholar] [CrossRef]
  83. Ergin, I.; Mandiracioglu, A. Demographic and socioeconomic inequalities for self-rated health and happiness in elderly: The situation for Turkey regarding World Values Survey between 1990 and 2013. Arch. Gerontol. Geriatr. 2015, 61, 224–230. [Google Scholar] [CrossRef] [PubMed]
  84. DeLeire, T.; Kalil, A. Does consumption buy happiness? Evidence from the United States. Int. Rev. Econ. 2010, 57, 163–176. [Google Scholar] [CrossRef]
  85. Cha, Y.J. Correlation between leisure activity time and life satisfaction: Based on KOSTAT time use survey data. Occup. Ther. Int. 2018, 2018, 5154819. [Google Scholar] [CrossRef] [PubMed]
  86. Kim, H.J.; Fredriksen-Goldsen, K.; Jung, H.H. Determinants of Physical Functioning and Health-Related Quality of Life among Sexual and Gender Minority Older Adults with Cognitive Impairment. J. Aging Health 2022, 35, 138–150. [Google Scholar] [CrossRef] [PubMed]
  87. Zhou, Y.; Wu, Q.; Li, C.; Ding, L. Association between community environment and depressive symptoms among Chinese middle-aged and older adults: Evidence from national longitudinal surveys from 2011 to 2018. J. Gerontol. Ser. A 2022, 77, 2265–2271. [Google Scholar] [CrossRef]
  88. Kim, D.; Jin, J. The impact of welfare facilities on happiness of the elderly: Evidence from Seoul, Korea. J. Plan. Educ. Res. 2019, 43, 332–345. [Google Scholar] [CrossRef]
  89. Park, C.J. The Factors of Leisure Affecting Happiness of the Middle-aged in Korea. Discourse Policy Soc. Sci. 2019, 12, 23–48. [Google Scholar] [CrossRef]
  90. Chai, X.; Kalyal, H. Cell phone use and happiness among Chinese older adults: Does rural/urban residence status matter? Res. Aging 2019, 41, 85–109. [Google Scholar] [CrossRef] [PubMed]
  91. Fang, Y.; Chau, A.K.; Wong, A.; Fung, H.H.; Woo, J. Information and communicative technology use enhances psychological well-being of older adults: The roles of age, social connectedness, and frailty status. Aging Ment. Health 2018, 22, 1516–1524. [Google Scholar] [CrossRef] [PubMed]
  92. Ractham, P.; Techatassanasoontorn, A.; Kaewkitipong, L. Old But Not Out: Social Media Use and Older Adults’ Life Satisfaction. Australas. J. Inf. Syst. 2022, 26, 1–28. [Google Scholar] [CrossRef]
  93. Bertocchi, F.M.; De Oliveira, A.C.; Lucchetti, G.; Lucchetti AL, G. Smartphone Use, Digital Addiction and Physical and Mental Health in Community-dwelling Older Adults: A Population-based Survey. J. Med. Syst. 2022, 46, 53. [Google Scholar] [CrossRef]
  94. Damant, J.; Knapp, M.; Freddolino, P.; Lombard, D. Effects of digital engagement on the quality of life of older people. Health Soc. Care Community 2017, 25, 1679–1703. [Google Scholar] [CrossRef] [PubMed]
  95. Lewin, K.M.; Meshi, D.; Schuster, A.M.; Cotten, S.R. Active and passive social media use are differentially related to depressive symptoms in older adults. Aging Ment. Health 2022, 27, 176–183. [Google Scholar] [CrossRef] [PubMed]
  96. Zhao, L. The effects of mobile social media use on older migrants’ social integration and life satisfaction: Use types and self-esteem perspective. Soc. Sci. Comput. Rev. 2021, 41, 249–264. [Google Scholar] [CrossRef]
  97. Nimrod, G.; Ivan, L. The dual roles technology plays in leisure: Insights from a study of grandmothers. Leis. Sci. 2022, 44, 715–732. [Google Scholar] [CrossRef]
  98. Kim, M.Y. The Influence of Serious Leisure on the Ego-Integrity among older Adults and the Moderating Effect of Leisure Constraints. Korean J. Gerontol. Soc. Welf. 2018, 73, 87–113. [Google Scholar]
  99. Argan, M.; Argan, M.T.; Dursun, M.T. Examining relationships among well-being, leisure satisfaction, life satisfaction, and happiness. Int. J. Med. Res. Health Sci. 2018, 7, 49–59. [Google Scholar]
  100. Liu, H.; Chen, X.; Zhang, H. Leisure satisfaction and happiness: The moderating role of religion. Leis. Stud. 2021, 40, 212–226. [Google Scholar] [CrossRef]
  101. Tokay Argan, M.; Mersin, S. Life satisfaction, life quality, and leisure satisfaction in health professionals. Perspect. Psychiatr. Care 2021, 57, 660–666. [Google Scholar] [CrossRef] [PubMed]
  102. Ministry of Culture, Sports and Tourism. 2019 National Leisure Activity Survey. Available online: https://www.mcst.go.kr/kor/s_policy/dept/deptView.jsp?pSeq=1770&pDataCD=0406000000&pType=08 (accessed on 1 September 2022).
  103. Ministry of Culture, Sports and Tourism. 2020 National Leisure Activity Survey. Available online: https://www.mcst.go.kr/kor/s_policy/dept/deptView.jsp?pSeq=1830&pDataCD=0406000000&pType= (accessed on 1 September 2022).
  104. Chicco, D.; Warrens, M.J.; Jurman, G. The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation. PeerJ Comput. Sci. 2021, 7, e623. [Google Scholar] [CrossRef] [PubMed]
  105. Kim, E.J.; Kang, H.W. Exploring the Factors Influencing Happiness using Machine Learning: Focusing on Factors of Leisure. J. Leis. Stud. 2022, 20, 167–181. [Google Scholar]
  106. Kim, M.H.; Kim, J.H. Trends in leisure activities among Korean older adults: 7-year changes based on the Korean National Leisure Condition Survey. J. Korean Gerontol. Soc. 2020, 40, 871–894. [Google Scholar] [CrossRef]
  107. Lee, M.S.; Kang, H.Y.; Lee, C.W. The Relationship among Serious Leisure, Leisure Satisfaction and Quality of Life for Older Korean Adults. Korean J. Phys. Educ. 2017, 56, 503–515. [Google Scholar] [CrossRef]
  108. Park, K.W.; Lee, C.W.; Kim, M.J. Effect of serious leisure experience on the leisure satisfaction among senior sport games participants. J. Phys. Educ. Sport 2016, 16, 584. [Google Scholar]
  109. Huang, C.; Li, Y. Understanding leisure satisfaction of Chinese seniors: Human capital, family capital, and community capital. J. Chin. Sociol. 2019, 6, 5. [Google Scholar] [CrossRef]
  110. Dattilo, J.; Lorek, A.; Sliwinski, M.; Chen, S.T.; Hill, N. An examination of video assisted leisure education in middle-aged and older adults. J. Leis. Res. 2021, 52, 227–246. [Google Scholar] [CrossRef]
  111. Seon, E.A. A study on the efficient operation of leisure and welfare facilities for the elderly in the senior welfare law. Public Land Law Rev. 2014, 65, 341–359. [Google Scholar]
  112. Hong, S.H.; Kim, S.E. Constraints to Participation in Senior Welfare Centers: Focusing on the Difference of Regional Types. Health Soc. Welf. Rev. 2016, 36, 125–156. [Google Scholar] [CrossRef]
  113. Seino, S.; Kitamura, A.; Nishi, M.; Tomine, Y.; Tanaka, I.; Taniguchi, Y.; Shinkai, S. Individual-and community-level neighbor relationships and physical activity among older Japanese adults living in a metropolitan area: A cross-sectional multilevel analysis. Int. J. Behav. Nutr. Phys. Act. 2018, 15, 46. [Google Scholar] [CrossRef]
  114. Steptoe, A.; Shankar, A.; Demakakos, P.; Wardle, J. Social isolation, loneliness, and all-cause mortality in older men and women. Proc. Natl. Acad. Sci. USA 2013, 110, 5797–5801. [Google Scholar] [CrossRef]
  115. Kim, G.E.; Choi, M.J. Characteristics of Solitary Leisure in Korean Society: Focusing on Teenage Male and Elderly Male Group. J. Korea Plan. Assoc. 2019, 54, 18–29. [Google Scholar] [CrossRef]
Sustainability 16 02730 g001
Figure 1. Research model.
Figure 1. Research model.
Sustainability 16 02730 g001
Sustainability 16 02730 g002
Figure 2. Analysis process.
Figure 2. Analysis process.
Sustainability 16 02730 g002
Sustainability 16 02730 g003
Figure 3. Model Performance Evaluation (RMSE).
Figure 3. Model Performance Evaluation (RMSE).
Sustainability 16 02730 g003
Sustainability 16 02730 g004
Figure 4. Model Performance Evaluation (R-Square).
Figure 4. Model Performance Evaluation (R-Square).
Sustainability 16 02730 g004
Sustainability 16 02730 g005
Figure 5. Model Evaluation After Optimizing the Hyperparameters (RMSE).
Figure 5. Model Evaluation After Optimizing the Hyperparameters (RMSE).
Sustainability 16 02730 g005
Sustainability 16 02730 g006
Figure 6. Model Evaluation After Optimizing the Hyperparameters (R-Square).
Figure 6. Model Evaluation After Optimizing the Hyperparameters (R-Square).
Sustainability 16 02730 g006
Sustainability 16 02730 g007
Figure 7. Importance of predictor variables (top five variables).
Figure 7. Importance of predictor variables (top five variables).
Sustainability 16 02730 g007
Table 1. List of predictor variables included in the analysis.
Table 1. List of predictor variables included in the analysis.
VariableVariable TypeItemValue
q2_1_n2_2categorical variableType of leisure activity in which the most participated in the past year1 = Cultural and artistic activity
2 = Sports activity
3 = Tourism
4 = Hobbies
5 = Rest activity
6 = Social and other activities
q2_2_1_1categorical variableCompanionship (leisure activity companion)1 = Alone (without a companion)
2 = Together (with a companion)
q2_3_1_1categorical variableFrequency in leisure activities1 = Everyday
2 = Not everyday
q2_5_1continuous variableLeisure cost (one-time basis)Response value (won)
q9continuous variableLeisure cost (monthly average)Response value (won)
q13continuous variableAverage leisure time per dayResponse value (hour)
q16_1categorical variableThe most used leisure space outside of the home1 = Indoor space
2 = Outdoor space
q19 categorical variableUse of public leisure facilities1 = Yes
2 = No
q22categorical variableParticipation in club activities1 = Yes
2 = No
q23categorical variableVolunteer activity1 = Yes
2 = No
q24continuous variableTime spent on smart devices during leisure timeResponse value (hour)
q27continuous variableLeisure policy satisfaction7-point scale
1 = Very dissatisfied
7 = Very satisfied
q30continuous variableOverall leisure life satisfaction7-point scale
1 = Very dissatisfied
7 = Very satisfied
q35continuous variableLeisure activity constraints (degree)7-point scale
1 = No constraints at all
7 = Very constrained
Table 2. List of baseline variables.
Table 2. List of baseline variables.
VariableBaseline Variables
q2_1_n2_2Rest activity
q2_2_1_1Alone
q2_3_1_1Everyday
q16_1Indoor space
q19Yes
q22Yes
q23Yes
Table 3. Model Performance Evaluation.
Table 3. Model Performance Evaluation.
ModelRMSER2
Linear Regression1.2020.311
Ridge1.2010.312
Lasso1.4490.000
SVM1.2000.314
Decision Tree1.638−0.278
Random Forest1.2170.294
XGBoost1.3510.130
LightGBM1.2780.222
Table 4. Hyperparameter optimization.
Table 4. Hyperparameter optimization.
ModelHyperparameter OptimzationRMSER2
Linear Regression-1.2020.311
Ridgealpha = 501.1980.316
Lassoalpha = 0.011.1950.319
SVMkernel = ‘linear’, C = 0.1, gamma = 0.0011.1980.316
Decision Treemax_depth = 3, min_samples_leaf = 451.2250.284
Random Forestn_estimators = 1701.2070.305
XGBoostcolsample_bylevel = 0.5, colsample_bytree = 0.6, gamma = 0.1, learning_rate = 0.01, max_depth = 3, min_child_weight = 5, n_estimators = 500, reg_lambda = 0.1, subsample = 0.81.1950.319
LightGBMcolsample_bytree = 0.3, learning_rate = 0.03, max_depth = 3, n_estimators = 300, num_leaves = 5, reg_alpha = 0.007, reg_lambda = 0.21.1970.317
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Kim, E.-J.; Kang, H.-W.; Park, S.-M. Leisure and Happiness of the Elderly: A Machine Learning Approach. Sustainability 2024, 16, 2730. https://doi.org/10.3390/su16072730

AMA Style

Kim E-J, Kang H-W, Park S-M. Leisure and Happiness of the Elderly: A Machine Learning Approach. Sustainability. 2024; 16(7):2730. https://doi.org/10.3390/su16072730

Chicago/Turabian Style

Kim, Eui-Jae, Hyun-Wook Kang, and Seong-Man Park. 2024. "Leisure and Happiness of the Elderly: A Machine Learning Approach" Sustainability 16, no. 7: 2730. https://doi.org/10.3390/su16072730

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop