Urban green land user social health benefit analysis method based on PLS-SEM model

Abstract

The invention relates to the technical field of social benefit analysis, in particular to an urban green land user social health benefit analysis method based on a PLS-SEM model, and the method comprises the steps: building an urban green land user social health benefit analysis model based on the PLS-SEM model, building a measurement model based on a latent variable and a corresponding explicit variable, and calculating the social health benefit of the urban green land user based on the measurement model. Constructing a structure model according to the relationship between the latent variables; verifying the reliability of the measurement model and the goodness of fit of the structure model; identifying key urban green land environment characteristics and key recreation activity type participation degree characteristics of social health benefits of the user according to the direct path coefficient; analyzing the intermediary effect types of the key urban green space environment characteristics and the key recreation activity type participation degree characteristics, and drawing to obtain a path mechanism schematic diagram influencing the social health benefits of the users. And optimization of urban green land planning design oriented to health is facilitated.

Description

Technical Field

[0001] This invention relates to the field of health benefit analysis technology for social benefits, specifically to a method for analyzing the social health benefits of urban green space users based on the PLS-SEM model. Background Technology

[0002] With accelerating urbanization, environmental problems and changing lifestyles have brought new health challenges, such as chronic non-communicable diseases and mental illnesses. The definition of health has transcended mere physical and mental well-being; the social health of users has also become a crucial component. Public health concerns are gradually shifting from treatment to prevention, and urban green spaces have significant advantages in promoting health. Historically, health crises have often driven social change and improvements in urban public spaces. Landscape architecture practitioners have contributed to improving health through planning and design. Urban green spaces are not only places for rest but also spaces for social interaction, improving social relationships, and enhancing social cohesion, making them essential for building healthy, inclusive, and sustainable urban ecosystems.

[0003] Currently, the conceptualization and measurement of user social health face challenges. It is both a component of health and a factor influencing it. User social health involves the interaction between individuals and others, social institutions, and the overall health level of society. However, there is a lack of unified and effective methods for assessing and analyzing the social health benefits of urban green space users. A comprehensive and accurate analytical method is needed to evaluate the social health benefits of urban green space users, which would contribute to optimizing health-oriented urban green space planning and design. Summary of the Invention

[0004] To address the technical problems existing in the prior art, this invention provides an analytical method for the social health benefits of urban green space users based on the PLS-SEM model. This method can analyze the contribution of various urban green space environmental characteristics and participation characteristics of various recreational activities to the social health benefits of users. It can comprehensively and accurately assess the social health benefits of urban green space users, provide a scientific basis for urban green space planning and design, and help optimize health-oriented urban green space planning and design.

[0005] The first objective of this invention can be achieved by adopting the following technical solution:

[0006] The method for analyzing the social health benefits of urban green space users based on the PLS-SEM model includes the following steps:

[0007] S1. Based on the PLS-SEM model, construct an analytical model of the social health benefits of urban green space users. The analytical model includes a measurement model and a structural model. The latent variables are various urban green space environmental characteristics, participation characteristics of various recreational activity types, and social health benefits of users. The measurement model is constructed based on the latent variables and their corresponding manifest variables, and the structural model is constructed based on the relationship between the latent variables.

[0008] S2. Based on the preset evaluation criteria, verify the reliability and validity of the measurement model and the goodness of fit of the structural model respectively, and obtain the verified measurement model and structural model.

[0009] S3. Calculate the direct path coefficients of various urban green space environmental characteristics on users' social health benefits, calculate the direct path coefficients of various recreational activity type participation characteristics on users' social health benefits, and identify the key urban green space environmental characteristics and key recreational activity type participation characteristics of users' social health benefits based on the direct path coefficients.

[0010] S4. Analyze the mediation effect types of key urban green space environmental characteristics and key recreational activity types participation characteristics, quantify the mediation strength based on the variance value (VAF), and draw a schematic diagram of the path mechanism affecting the social health benefits of users based on the mediation effect type and mediation strength of each path.

[0011] Specifically, step S1 includes:

[0012] S11. Taking various urban green space environmental characteristics, participation in various recreational activities, and user social health benefits as latent variables, various urban green space environmental characteristics include: accessibility, sense of security, aesthetics, attractiveness, ease of use, and maintenance status.

[0013] S12. Construct an analytical model of the social health benefits of urban green space users based on the PLS-SEM model, construct a measurement model based on latent variables and their corresponding manifest variables, define each latent variable and its corresponding manifest variable, and associate each latent variable with its corresponding manifest variable; construct a structural model based on the relationship between latent variables, and describe the relationship between latent variables through the structural model.

[0014] Specifically, the construction of a structural model based on the relationships between latent variables, and the description of these relationships through the structural model, includes:

[0015] The structural model for constructing the analytical model takes multiple urban green space environmental characteristics as exogenous variables, multiple recreational activity types participation characteristics as mediating variables, and user social health benefits as endogenous variables.

[0016] We establish hypothetical pathways, assuming a positive correlation between multiple urban green space environmental characteristics and users' social health benefits; a positive correlation between participation characteristics of multiple recreational activity types and users' social health benefits; and that participation characteristics of multiple recreational activity types play a mediating role in the impact of multiple urban green space environmental characteristics on users' social health benefits.

[0017] By setting up path relationships, we can establish multiple paths that directly affect the social health benefits of users through various urban green space environmental characteristics, multiple paths that directly affect the social health benefits of users through various recreational activity types, and multiple mediating paths that affect the social health benefits of users through various recreational activity types. This results in multiple hypothetical paths.

[0018] Specifically, step S2 includes:

[0019] S21. Using Smart PLS software, select the PLS algorithm to estimate the measurement model, calculate factor loadings, Cronbach's Alpha coefficient, combined reliability, average variance extracted value, and variance inflation factor, respectively, preset the reliability and validity assessment criteria, and verify the reliability and validity of the measurement model according to the reliability and validity assessment criteria.

[0020] S22. Using Smart PLS software, run the structural model using the PLS algorithm and calculate R. 2 Value, Q 2 The SRMR value is calculated, a goodness-of-fit evaluation criterion is preset, and the goodness-of-fit of the structural model is verified according to the goodness-of-fit evaluation criterion.

[0021] Specifically, step S3 includes:

[0022] S31. Import the data from the measurement model and structural model that have been verified for reliability and validity into Smart PLS software, calculate the direct path coefficients of various urban green space environmental features on the social health benefits of users, conduct repeated sampling tests using the Bootstrap sampling method, obtain the T-statistic and P-value for each path, determine the statistical significance of the path coefficients based on the T-statistic and P-values, and identify the key urban green space environmental features that have a direct and significant positive impact on the social health benefits of users based on the direct path coefficients and statistical significance.

[0023] S32. Establish a path model in Smart PLS software, setting participation characteristics of various recreational activity types as exogenous variables and user social health benefits as endogenous variables to construct direct path relationships; run the PLS algorithm to calculate path coefficients and output the standardized path coefficient values ​​of each recreational activity type on user social health benefits; perform repeated sampling tests using the Bootstrap sampling method to obtain the T-statistic and P-value for each path, determine the statistical significance of the path coefficients based on the T-statistic and P-value, and identify the key recreational activity type participation characteristics that have a direct and significant positive impact on user social health benefits based on the direct path coefficients and statistical significance.

[0024] Specifically, step S4 includes:

[0025] S41. Using Smart PLS software, calculate the specific indirect effect path coefficients of the social health benefits of users through the participation characteristics of key urban green space environmental characteristics and key recreational activity types. Use Bootstrap sampling method to perform repeated sampling tests, obtain the T-statistic and P-value of each specific indirect effect path, identify significant specific indirect effect paths, and obtain the path coefficient distribution of each specific indirect effect path.

[0026] S42. Calculate the path coefficient of the total indirect effect of key urban green space environmental features on the social health benefits of users, compare the significance and direction of direct effects, total indirect effects and specific indirect effects, classify them according to the criteria for determining the type of mediation effect, and obtain the distribution of the mediation effect types of each key urban green space environmental feature affecting the social health benefits of users through the participation characteristics of different types of recreational activities.

[0027] S43. Calculate the variance fraction (VAF) of each specific indirect effect path. Based on the magnitude of the VAF value, quantify the mediating strength of the participation characteristics of each key recreational activity type in the process of the influence of key urban green space environmental characteristics on the social health benefits of users. Rank the mediating importance of each specific indirect effect path. Finally, based on the significant direct effect path, the total indirect effect path, and the specific indirect effect path and their effect strength, create a schematic diagram of the mechanism of influence of urban green space users on social health benefits.

[0028] A computer device includes a processor and a memory for storing a processor-executable program, characterized in that, when the processor executes the program stored in the memory, it implements the aforementioned method for analyzing the social health benefits of urban green space users based on the PLS-SEM model.

[0029] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0030] This invention provides a method for analyzing the social health benefits of urban green space users based on the PLS-SEM model. This method constructs measurement and structural models using the PLS-SEM model, calculates the direct path coefficients of various urban green space environmental features on users' social health benefits, and calculates the direct path coefficients of various recreational activity participation characteristics on users' social health benefits. Based on the direct path coefficients, it identifies key urban green space environmental features and key recreational activity participation characteristics that contribute to users' social health benefits. It then analyzes the contribution of these key urban green space environmental features and key recreational activity participation characteristics to users' social health benefits, ranks them by their level, analyzes the mediation effect types of these key urban green space environmental features and key recreational activity participation characteristics, quantifies the mediation strength using the variance account (VAF), and draws a schematic diagram of the path mechanisms affecting users' social health benefits based on the mediation effect types and mediation strengths of each path. This schematic diagram of the path mechanisms affecting users' social health benefits allows for a comprehensive and accurate assessment of users' social health benefits of urban green spaces. It can help designers understand how to guide beneficial social activities through environmental design, providing a scientific basis for urban green space planning and design and contributing to the optimization of health-oriented urban green space planning and design. Attached Figure Description

[0031] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0032] Figure 1 This is a flowchart of the method for analyzing the social health benefits of urban green space users based on the PLS-SEM model in Embodiment 1 of the present invention.

[0033] Figure 2 This is a schematic diagram of the analytical model structure for the social health benefits of urban green space users in Embodiment 1 of the present invention;

[0034] Figure 3 This is a schematic diagram of the social health benefits impact mechanism of urban green space users, drawn based on the analysis results in Embodiment 1 of the present invention. Detailed Implementation

[0035] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments, and the implementation of the present invention is not limited thereto. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0036] Example 1:

[0037] like Figure 1 As shown in the figure, this embodiment proposes a method for analyzing the social health benefits of urban green space users based on the PLS-SEM model, including:

[0038] S1. Based on the PLS-SEM model, construct an analytical model of the social health benefits of urban green space users. The analytical model includes a measurement model and a structural model. The characteristics of urban green space environment, participation in recreational activity types, and social health benefits of users are used as latent variables. The measurement model is constructed based on the latent variables and their corresponding manifest variables, and the structural model is constructed based on the relationship between the latent variables.

[0039] S11. Using the characteristics of urban green space environment, participation in recreational activities, and social health benefits of users as latent variables, the characteristics of urban green space environment include: accessibility, sense of security, aesthetics, attractiveness, ease of use, and maintenance status, resulting in latent variables of accessibility, sense of security, aesthetics, attractiveness, ease of use, maintenance status, participation in recreational activities, and social health benefits of users.

[0040] In this example, the participation rate of recreational activity types includes the participation rate of 14 different leisure activities: sightseeing, playing, photography, water activities, animal activities, singing and playing, chess and cards, painting and calligraphy, exhibition activities, running and walking, dancing, qigong and martial arts, ball games, and fitness equipment activities.

[0041] S12. Construct an analytical model of the social health benefits of urban green space users based on the PLS-SEM model, construct a measurement model based on latent variables and their corresponding manifest variables, define each latent variable and its corresponding manifest variable, and associate each latent variable with its corresponding manifest variable; construct a structural model based on the relationship between latent variables, and describe the relationship between latent variables through the structural model.

[0042] PLS-SEM stands for Partial Least Squares Structural Equation Modeling. It's a multivariate statistical analysis method that uses variance-based estimation to seek the optimal interpretation of relationships between latent variables through iterative calculations. A PLS-SEM model comprises a measurement model and a structural model. The measurement model defines the relationship between latent variables and manifest indicators, while the structural model describes the causal relationships between latent variables, typically including exogenous variables (unaffected by other variables in the model) and endogenous variables (affected by other variables).

[0043] S121. Construct a measurement model based on latent variables and their corresponding manifest variables, define each latent variable and its corresponding manifest variable, and associate each latent variable with its corresponding manifest variable, including:

[0044] The measurement model for the analytical model is constructed, defining the manifest variables corresponding to the latent variables of accessibility as: distance of green space from residence, identifiability of green space entrance, accessibility of internal roads within the green space, clarity of internal directional signage, connectivity of functional areas within the green space, and connectivity between the green space and surrounding areas. The manifest variables corresponding to the latent variables of safety perception are: lighting conditions, safety management measures, spatial visibility, and configuration of emergency facilities within the green space. The manifest variables corresponding to the latent variables of aesthetics are: the aesthetic appeal of the landscape, the aesthetic appeal of the water features, the aesthetic appeal of architectural elements, and the overall harmony of the landscape. The manifest variables corresponding to the latent variables of attractiveness are: distinctive scenic spots, diversity of activity spaces, seasonal landscape changes, and cultural elements within the green space. The manifest variables corresponding to the latent variables of maintenance status in urban green space environmental characteristics are: plant maintenance level, facility maintenance status, environmental sanitation conditions, water body maintenance, and overall management level. The manifest variables corresponding to the latent variables of usability are: convenience of rest facilities, suitability of activity spaces, completeness of service facilities, and configuration of barrier-free facilities. The manifest variables corresponding to the latent variable of participation in recreational activities are defined as follows: sightseeing, play, photography, water activities, animal activities, singing and playing instruments, chess and cards, painting and calligraphy, exhibition activities, running and brisk walking, dancing, qigong and martial arts, ball games, and fitness equipment activities. The manifest variables corresponding to the latent variable of user social health benefits are defined as the average values ​​of social relationships, social support, and social interaction.

[0045] like Figure 2The diagram illustrates the structure of an analytical model for the social health benefits of urban green space users. This model constructs an analytical model that examines how the environmental characteristics of urban green spaces influence users' social health benefits through recreational behavior. In the diagram, ellipses represent "latent variables," rectangles represent "manifest variables," and single arrows represent individual effects. The latent variables of urban green space environmental characteristics include: accessibility, sense of security, aesthetics, attractiveness, usability, and maintenance. Accessibility is measured by six manifest variables: distance from residence, identifiability of the green space entrance, accessibility of internal roads, clarity of internal directional signage, connectivity between functional areas within the green space, and connectivity between the green space and surrounding areas. Sense of security is measured by four manifest variables: lighting conditions, safety management measures, spatial visibility, and the configuration of emergency facilities. Aesthetics are represented by four indicators: the aesthetic appeal of the landscape, the aesthetic appeal of the water features, the aesthetic appeal of architectural elements, and the overall harmony of the landscape. Four explicit variables measure attractiveness: distinctive scenic spots, diversity of activity areas, seasonal landscape changes, and cultural elements. Five variables measure maintenance status: plant maintenance level, facility maintenance status, environmental sanitation conditions, water body maintenance, and overall management level. Four explicit variables measure usability: convenience of rest facilities, suitability of activity spaces, completeness of service facilities, and accessibility of facilities. The mediating variable "Characteristics of Leisure Activities" includes participation in 14 different leisure activities (sightseeing, play, photography, water activities, animal activities, singing and playing instruments, chess and cards, painting and calligraphy, exhibition activities, running and brisk walking, dancing, qigong and martial arts, ball games, and fitness equipment activities). As the dependent variable, "User's Social Health Benefits" is measured by social relationships, social support, social interaction, and average values ​​using an item packaging strategy to reduce parameter estimation bias during model construction.

[0046] S122. Construct a structural model based on the relationships between latent variables, and describe the relationships between latent variables through the structural model, including:

[0047] The structural model for constructing the analytical model takes multiple urban green space environmental characteristics as exogenous variables, multiple recreational activity types participation characteristics as mediating variables, and user social health benefits as endogenous variables.

[0048] We establish hypothetical pathways, assuming a positive correlation between multiple urban green space environmental characteristics and users' social health benefits; a positive correlation between participation characteristics of multiple recreational activity types and users' social health benefits; and that participation characteristics of multiple recreational activity types play a mediating role in the impact of multiple urban green space environmental characteristics on users' social health benefits.

[0049] By setting up path relationships, we can establish multiple paths that directly affect the social health benefits of users through various urban green space environmental characteristics, multiple paths that directly affect the social health benefits of users through various recreational activity types, and multiple mediating paths that affect the social health benefits of users through various recreational activity types. This results in multiple hypothetical paths.

[0050] The mediating effect of participation in different types of recreational activities on the relationship between urban green space and users' social health varies. Therefore, this embodiment assumes that participation in each type of recreational activity has a mediating effect on each green space environmental characteristic and users' social health. Moreover, the mediating role of participation in different types of recreational activities on the relationship between urban green space environmental characteristics and users' social health benefits will vary depending on the type of activity and environmental characteristics.

[0051] In this embodiment, it is assumed that there is a positive correlation between the six constructs of the green space environment and the social health benefits of users; there is also a positive correlation between the participation in 14 types of recreational activities and the social health benefits of users; the participation in these 14 types of recreational activities plays a mediating role in the influence of the six constructs of the green space environment on the social health benefits of users. Path relationships are established, including paths where the six environmental features directly affect the social health benefits of users, paths where the 14 activity types directly affect the social health benefits of users, and 84 mediating paths where the six environmental features affect the social health benefits of users through the 14 activity types, totaling 104 hypothetical paths.

[0052] As shown in Table 1, the analytical model of how urban green space environmental characteristics affect users' social health benefits through recreational behavior is transformed into multiple hypothetical paths, including 13 main hypothetical paths and 6 related hypothetical paths based on green space environmental characteristics and 1 user social health benefit construct multiplied by 14 activity types, i.e., 6 + 7 * 14 = 104 hypothetical paths. According to the analytical model of the impact on users' social health benefits, there is a positive correlation between the 6 green space environmental characteristics and users' social health benefits; there is also a positive correlation between the participation rate in the 14 recreational activity types and users' social health benefits; furthermore, the participation rate in these 14 recreational activity types plays a mediating role in the influence of the 6 green space environmental characteristics on users' social health benefits.

[0053] Table 1

[0054]

[0055]

[0056] S2. Based on the preset evaluation criteria, verify the reliability and validity of the measurement model and the goodness of fit of the structural model respectively, and obtain the verified measurement model and structural model.

[0057] S21. Using Smart PLS software, select the PLS algorithm to estimate the measurement model, calculate factor loadings, Cronbach's Alpha coefficient, combined reliability, average variance extracted value, and variance inflation factor, respectively. Preset the reliability and validity evaluation criteria, and verify the reliability and validity of the measurement model according to the reliability and validity evaluation criteria.

[0058] SmartPLS is a structural equation modeling (SEM) software based on partial least squares (PLS), primarily used for analyzing complex variable relationships. In this embodiment, collected questionnaire data is imported into SmartPLS software in Excel or CSV format. The associations between latent variables of urban green space environmental characteristics, latent variables of recreational activity type participation, and latent variables of user social health benefits and their corresponding manifest variables are established according to a preset measurement model structure. The PLS algorithm is selected for measurement model estimation, with a maximum iteration count of 300 and a stopping criterion of 10. -7 The weighting scheme uses a path weighting scheme. The factor loadings of each manifest variable can be viewed in the Outer Loadings report; the system automatically outputs the standardized loading coefficients of each manifest variable and its corresponding latent variable. The Cronbach's Alpha coefficient and combined reliability (CR) value can be obtained from the Construct Reliability and Validity report; the software automatically calculates internal consistency reliability based on the factor loadings of each manifest variable. The system automatically calculates the mean variance extracted (AVE) to assess convergent validity, and the variance inflation factor (VIF) test for multicollinearity can be viewed in the Collinearity Statistics report. The SmartPLS software generates a comprehensive report containing all reliability and validity indicators; by comparing these reports with preset evaluation criteria, it can be determined whether the measurement model has passed reliability and validity validation.

[0059] The pre-defined evaluation criteria for reliability and validity include: factor loadings ≥ 0.7, Cronbach's Alpha coefficient ≥ 0.6, combined reliability (CR) ≥ 0.7, mean variance extracted (AVE) ≥ 0.5, and variance inflation factor (VIF) < 5. Factor loadings represent the correlation between the observed variable and the latent variable. A value ≥ 0.7 is ideal; values ​​below 0.4 should be deleted; values ​​between 0.4 and 0.7 require consideration of other evaluation criteria. A larger absolute value of the factor loading indicates a stronger relationship between the measurement indicator and the latent variable. Cronbach's Alpha coefficient assesses the correlation and consistency among internal terms of the latent variable. A value ≥ 0.6 indicates good internal consistency among grouped items. Combined reliability (CR) reflects whether all items in each latent variable consistently explain that latent variable. A value ≥ 0.7 indicates good construct reliability for the latent variable. Mean variance extracted (AVE) assesses convergent validity by calculating the mean variance extracted. The standard value is ≥0.5, aiming to ensure that the observed variables can adequately explain the variation of the latent variable. The variance inflation factor (VIF) is used to check for multicollinearity among latent variables. The standard value is: <5 is acceptable, <3 is ideal. A VIF value higher than 5 indicates collinearity in the prediction structure; a VIF value less than 5 indicates no serious multicollinearity in the measurement model, meaning that the different measurement indicators are independent of each other.

[0060] Table 2 shows the reliability and validity test results of the measurement model. Most indicators exhibited high factor loadings, all exceeding the recommended threshold of 0.6. Although the factor loadings of indicators att2 and SCm were slightly below this threshold, close to 0.6, these values ​​remained within an acceptable range. Furthermore, the CR and AVE values ​​of the six latent variables related to user social health benefits and urban green space environmental characteristics were all above the threshold of 0.5, indicating good convergent validity for these latent variables. In addition, the absolute values ​​of factor loadings provided a basis for judging the strength of the relationship between latent variables and their measurement indicators. Higher factor loading values ​​reflected a strong association between the measurement indicators and latent variables; Cronbach's alpha values ​​were all greater than 0.7, indicating good consistency among the measurement indicators. CR values ​​were all greater than 0.7, indicating high internal consistency among the latent variables in the measurement model, meaning a strong correlation between the measurement indicators corresponding to each latent variable. AVE values ​​were all greater than 0.5, indicating a high proportion of variance jointly explained by the measurement indicators corresponding to each latent variable. The VIF values ​​are all less than 5, indicating that there is no serious multicollinearity problem in the measurement model, which means that the different measurement indicators are independent of each other.

[0061] Table 2

[0062]

[0063] Therefore, the reliability, internal consistency, convergent validity, and discriminant validity of the measurement model all meet the requirements, indicating that the measurement model has high reliability and effectiveness in measuring latent variables and has passed the reliability and validity verification.

[0064] In Table 2, the explicit variables are coded as acc1 to acc6, corresponding to: the distance of the green space from the residence, the identifiability of the green space entrance, the accessibility of the internal roads of the green space, the clarity of the internal directional signs of the green space, the connectivity of the various functional areas within the green space, and the connectivity between the green space and the surrounding areas, respectively; the explicit variables are coded as saf1 to saf4, corresponding to: the lighting conditions of the green space, safety management measures, spatial visibility, and the configuration of emergency facilities (measuring the sense of security); the implicit variables are coded as aes3 to aes6, corresponding to: the aesthetic appeal of the landscape, the aesthetic appeal of the water features, the aesthetic appeal of the architectural features, and the overall harmony of the landscape; and the explicit variables are coded as att1 to att2 and att5. Codes A through ATT6 correspond to: distinctive attractions of green spaces, diversity of activity areas, seasonal landscape changes, and cultural elements; codes Mai1 through Mai5 correspond to: plant maintenance level, facility maintenance status, environmental sanitation conditions, water body maintenance, and overall management level (measuring maintenance status); codes USA1, USA3, USA4, and USA5 correspond to: convenience of rest facilities, suitability of activity spaces, completeness of service facilities, and accessibility of facilities; codes SRm, SSm, and SCm correspond to: social relations, social support, and social interaction (measured by average values).

[0065] S22. Using Smart PLS software, run the structural model using the PLS algorithm and calculate R. 2 Value, Q 2 The goodness-of-fit value and SRMR value are used to set the evaluation criteria for goodness-of-fit. The goodness-of-fit of the structural model is verified according to the evaluation criteria.

[0066] In this embodiment, the structural model is run using the PLS algorithm through Smart PLS software, and the software automatically calculates various goodness-of-fit indices; the R-squared values ​​of each endogenous latent variable can be viewed in the coefficient of determination (R-Square) report. 2 Value; calculate the predictive correlation Q using the Blindfolding procedure. 2 The value is set to omit distance as 7. The software automatically calculates the model's predictive ability index through cross-validation. The system automatically calculates the standardized root mean square residual (SRMR) value based on the difference between the observed variable correlation matrix and the model's predicted correlation matrix, and obtains this index value from the Model Fit report. Finally, the software generates a model containing R... 2 Value, Q 2The comprehensive report on the goodness of fit of the SRMR value and the SRMR value compares the calculation results with the preset standard to verify whether the goodness of fit of the structural model has reached an acceptable level.

[0067] Preset the evaluation criteria for goodness of fit, including: setting R... 2 Value ≥ 0.6, Q 2 If the value is >0 and the SRMR value is <0.08, R will be calculated. 2 Value, Q 2 The values ​​and SRMR values ​​are compared with the settings to verify whether the structural model has a good fit.

[0068] Among them, R 2 The value represents the variance explanatory power for each internal latent variable, and is the square of the correlation coefficient between the actual and predicted values ​​of the internal factor surface. Value criteria: ≥0.6 is good, 0.33 is moderate, and 0.19 is poor. Q 2 The correlation coefficient (SRMR) is an indicator used in model prediction correlation studies. Value standard: >0, indicating a correlation between the path model and the predicted surface. The SRMR value represents the standardized root mean square residual, a metric used to evaluate model fit. It is typically used to measure the difference between the correlation matrix between observed variables in the model and the correlation matrix of model predictions. Value standard: <0.08, a lower SRMR value is better; generally, an SRMR less than 0.08 is considered to indicate a good model fit.

[0069] Table 3 shows the evaluation results of the structural model. In the evaluation of the social health benefits to users, the model's R... 2 The value is 0.307, meaning the model can explain approximately 30.7% of the variance of the target variable, which is generally considered moderate explanatory power. While this isn't particularly high explanatory power, it's usually sufficient for research needs in the field of social surveys. The model's Q... 2 A value of 0.189 indicates that the model has certain predictive power, better than random guessing. The SRMR value is 0.067, below the standard threshold of 0.08, indicating that the model fits the observed data well and the covariance matrix of the model differs little from the observed data. Overall, the R-value of the structural model is... 2 Q 2 Both the SRMR index and the SRMR index met the acceptable test criteria, indicating that the fit was excellent.

[0070] Table 3

[0071]

[0072]

[0073] S3. Calculate the direct path coefficients of various urban green space environmental characteristics on the social health benefits of users, calculate the direct path coefficients of various recreational activity type participation characteristics on the social health benefits of users, and identify the key urban green space environmental characteristics and key recreational activity type participation characteristics that have a direct and significant positive impact on the social health benefits of users based on the direct path coefficients.

[0074] S31. Import the data from the reliable and valid measurement model and structural model into Smart PLS software to calculate the direct path coefficients of various urban green space environmental features on the social health benefits of users. Use the Bootstrap sampling method for repeated sampling test to obtain the T-statistic and P-value for each path. Determine the statistical significance of the path coefficients based on the T-statistic and P-value. Identify the key urban green space environmental features that have a direct and significant positive impact on the social health benefits of users based on the direct path coefficients and statistical significance.

[0075] In this embodiment, Smart PLS software is used to calculate the standardized path coefficients of six different urban green space environmental characteristics—accessibility, safety, aesthetics, attractiveness, maintenance status, and ease of use—on the social health benefits to users. The data from the reliability and validity-validated measurement model and structural model are imported into the Smart PLS software. The maximum number of iterations is set to 300, and the stopping criterion is 10. -7 The weighted scheme selects the path weighting scheme; the PLS algorithm is run, and the software automatically calculates the composite score and standardized path coefficient of each latent variable; the path coefficient table is extracted from the results report to obtain the standardized path coefficient values ​​from the six environmental feature constructs to the user's social health benefits.

[0076] Based on the direct path coefficient, key urban green space environmental features and participation characteristics of key recreational activity types that have a direct and significant positive impact on users' social health benefits were identified. The magnitude and direction of the path coefficients were analyzed; path coefficients range from -1 to +1, with values ​​close to +1 indicating a strong positive correlation and values ​​close to -1 indicating a strong negative correlation. The absolute value of the path coefficients was evaluated to analyze the influence of urban green space environmental features on users' social health benefits; a larger absolute value indicates a greater influence and stronger direct impact.

[0077] Specifically, the significance of path coefficients is tested using the T-statistic and p-value. A bootstrap sampling method is used to perform 500 repeated sampling tests to obtain the T-statistic and p-value for each path. The statistical significance of the path coefficients is then determined based on the T-statistic and p-value. Bootstrap is a nonparametric repeated sampling technique that generates a large number of simulated samples (typically 500-5000 times) by repeatedly drawing the original sample (n times) with replacement, thus constructing an empirical distribution of the statistic. The bootstrap sampling method includes:

[0078] Establish the path equation: Y = β1X + β2M + ε; where β1 and β2 are path coefficients to be tested, X is the independent variable, Y is the dependent variable, M is the mediating variable, and ε is the error term.

[0079] Sampling process: Randomly draw n observations with replacement from the original data (sample size n), and repeat to generate 500 Bootstrap samples.

[0080] Path coefficient estimation and distribution construction: Fit the model for each Bootstrap sample and record the path coefficient estimates

[0081]

[0082] Generate the empirical sampling distribution of the path coefficients and calculate the standard error (SE boot ):

[0083]

[0084] where is the mean of 500 estimates.

[0085] Statistical significance judgment:

[0086] T-statistic calculation:

[0087] P-value determination. If a two-tailed test is used, P = 2 × P(T > |T|) <>

[0088] Judge the significance level according to the absolute value of the T-statistic: When performing a two-tailed test, |T| ≥ 1.96 indicates significance at the 5% level, |T| ≥ 2.58 indicates significance at the 1% level, and |T| ≥ 1.65 indicates significance at the 10% level. Judge the significance according to the P-value: P ≤ 0.01 is highly significant, 0.01 < P ≤ <0.05 is significant, 0.05 < P ≤ 0.1 is weakly significant, and P > 0.1 is not significant.

[0089] Taking the impact of attractiveness on the social health benefits of users as an example, its path coefficient is 0.282, the T-statistic is 7.759, and the P-value is 0.000. Since |T| = 7.759 > 2.58, it indicates that this path is highly significant at the 1% level; the P-value = <0.000 < 0.01, further confirming the high significance. The path coefficient is positive and has a relatively large absolute value, indicating that attractiveness has a strong positive impact on the social health benefits of users. In contrast, the path coefficient of sense of security is 0.034, the T-statistic is 0.879, and the P-value is 0.380. Since |T| = 0.879 < 1.65 and the P-value > 0.1, it indicates that this path is not significant, and sense of security has no direct impact on the social health benefits of users.

[0090] The significance of path coefficients can be tested using the T-statistic and p-value to avoid misinterpreting random errors as true effects. The T-statistic reflects the degree to which the path coefficient deviates from zero, while the p-value quantifies the probability that the effect is caused by random factors. This confirms whether a causal relationship exists between recreational activities and social health benefits, while ruling out spurious associations caused by sample fluctuations, thus demonstrating the statistical reliability of the path coefficients.

[0091] Table 4 shows the direct impact of urban green space environmental characteristics on users' social health benefits. The path coefficient analysis (direct effect) of "environmental characteristics → user social health" reveals that accessibility, aesthetics, attractiveness, and maintenance have a direct and significant impact on users' social health benefits. Based on the absolute values ​​of the path coefficients, the four urban green space environmental characteristics have a direct and significant impact on users' social health benefits, in descending order of influence: attractiveness, aesthetics, accessibility, and maintenance. Among them:

[0092] Attractiveness and accessibility have a significant positive impact on users' social health benefits. The path coefficient of attractiveness (β>0.2) indicates that the attractiveness of green spaces is an important factor in promoting users' social health benefits. The path coefficient of accessibility (β>0.1) reveals the positive contribution of the easy accessibility of green spaces to users' social health benefits.

[0093] The aesthetic characteristics of green spaces exhibit a significant negative correlation (β>0.1) with the social health benefits of users. While a beautiful environment may enhance an individual's aesthetic experience, this is insufficient to directly promote social interaction or provide social support, thereby improving users' social health benefits. Individuals engaging in specific activities may be more sensitive to the aesthetics of their environment, but this does not necessarily mean their social health benefits will be enhanced as a result. However, in PLS-SEM, this direct effect needs to be comprehensively understood in conjunction with the pathways of mediating effects.

[0094] Regarding the impact of green space maintenance on users' social health benefits, the negative impact is relatively weak (β<0.1). This indicates that although the level of green space maintenance is crucial to the overall perceived quality of green spaces, it is not a key factor determining users' social health benefits.

[0095] Table 4

[0096]

[0097] S32. Establish a path model in Smart PLS software. Set the participation characteristics of multiple types of recreational activities as exogenous variables and the social health benefits of users as endogenous variables to construct a direct path relationship. Run the PLS algorithm to calculate the path coefficients and output the standardized path coefficient values of each type of recreational activity on the social health benefits of users. Conduct repeated sampling tests through the Bootstrap sampling method to obtain the T statistic and P value of each path. Judge the statistical significance of the path coefficients based on the T statistic and P value, and based on the direct path coefficients and statistical significance; identify the participation characteristics of the key recreational activity types that have a direct and significant positive impact on the social health benefits of users according to the direct path coefficients and statistical significance.

[0098] In this embodiment, import the collected questionnaire data into Smart PLS software to ensure that the data formats of the participation of 14 types of recreational activities (sightseeing, playing, photography, water-related, small animal-related, singing and playing musical instruments, chess and card games, painting and calligraphy, exhibition participation, running and walking, dancing, qigong and martial arts, small ball games, and equipment fitness activities) and the social health benefits of users are correct. Establish a path model in Smart PLS software, set the participation of 14 types of recreational activities as exogenous variables and the social health benefits of users as endogenous variables to construct a direct path relationship. Run the PLS algorithm to calculate the path coefficients, and the software automatically outputs the standardized path coefficient values of each type of recreational activity on the social health benefits of users. Set 500 repeated samplings through the Bootstrap resampling procedure to calculate the T statistic and P value of each path. Judge the influence direction according to the sign of the path coefficient, determine the influence intensity by sorting according to the absolute value size, and determine the activity types that have a significant direct impact on the social health benefits of users.

[0099] Combine the significance test results to identify the key recreational activity types that have a direct and significant impact on the social health benefits of users. Take dancing activities as an example. Its path coefficient is 0.302, the T statistic is 11.492, and the P value is 0.000. Since |T| = 11.492 ≥ 2.58 and P ≤ 0.01, it indicates that dancing activities have a highly significant positive impact on the social health benefits of users; the path coefficient of small ball games is 0.191, the T statistic is 6.783, and the P value is 0.000, which also reaches a highly significant level, indicating that it has a significant positive promoting effect on the social health benefits of users. In contrast, although the path coefficient of playing activities is a positive value of 0.064, the T statistic is only 2.085 and the P value is 0.038, at the significant level of 0.01 < P ≤ 0.05, and the influence degree is relatively weak. The path coefficient of sightseeing activities is -0.032, the T statistic is 1.162, and the P value is 0.246 > 0.1, indicating that there is no significant association between it and the social health benefits of users.

[0100] As shown in Table 5, the direct impact of participation in recreational activity types on users' social health benefits is analyzed through the path coefficient analysis (direct effect) of "recreational activity type → user's social health". According to the absolute value of the path coefficient, the participation in the eight activity types has a direct and significant impact on users' social health benefits. In order of the degree of impact, they are dance, ball games, chess and cards, play, photography, singing and playing instruments, painting and calligraphy, and water activities.

[0101] Among these, the social health benefits of users were significantly positively influenced by dance (path coefficient > 0.3) and ball games (path coefficient > 0.1), and minimally positively influenced by play, photography, singing and playing instruments, and board games (path coefficient < 0.1). The negative influence from water activities and painting / calligraphy activities (path coefficient < 0.1) was also minimal. However, there was no significant correlation between sightseeing, animal interaction, exhibitions, running / brisk walking, qigong / martial arts, and fitness equipment activities and the social health benefits of users. Specifically, this includes the following:

[0102] The social health benefits of users were significantly positively influenced by dance activities (path coefficient > 0.3) and ball games (path coefficient > 0.1). Dance activities (β > 0.3) provided strong social interaction and physical activity, both key factors in enhancing users' social health benefits. Dance activities promoted emotional expression and the establishment of social support networks. Ball games (β > 0.1), through teamwork and competition, enhanced social connections and group cohesion. These activities also provided opportunities for physical exercise and relaxation, contributing to improved users' social health benefits.

[0103] The positive impact of play, photography, singing, and board games (β < 0.1) on user social health benefits is extremely weak. While these activities provide opportunities for social interaction, they lack sufficient intensity or depth, resulting in a small overall impact on user social health. These activities may be conducted alone or in small groups and often do not involve a wide social network. When these activities are limited to specific groups, such as family members or close friends, their role in promoting broader user social health benefits is relatively limited.

[0104] The negative impact of water-based and painting / calligraphy activities on users' social health benefits was extremely weak. Although these activities provided participants with opportunities for relaxation and personal fulfillment, their contribution to users' social health benefits was relatively limited. These activities are more of an independent or meditative form of leisure, tending to attract individuals to engage in them alone and even reducing interaction with others.

[0105] Activities such as sightseeing, animal interaction, exhibitions, running / brisk walking, qigong / martial arts, and gym equipment workouts showed no significant correlation with users' social health benefits. These activities focus more on physical and mental health than on users' social health benefits. Sightseeing and animal interaction activities provide individuals with a way to interact with nature or animals. This interaction is beneficial to individual mental health but does not necessarily promote social interaction. Running / brisk walking, qigong / martial arts, and gym equipment workouts are beneficial to individual physical health. These activities sometimes take place in social forms within sports groups, but their primary purpose is not to build or strengthen social relationships.

[0106] Therefore, different types of recreational activities have varying impacts on users' social health benefits. Dance and ball games, due to their strong social interaction and physical activity, are significantly effective in improving users' social health. While play, photography, singing, and board games have positive effects, these are relatively weak. Water activities and painting / calligraphy activities also contribute relatively little to users' social health and exhibit a slight negative impact. Other activity types, such as sightseeing, interacting with small animals, participating in exhibitions, running / brisk walking, qigong / martial arts, and fitness equipment, show no significant association with users' social health benefits.

[0107] Table 5

[0108]

[0109] S4. Analyze the mediation effect types of key urban green space environmental characteristics and key recreational activity types participation characteristics, quantify the mediation strength based on the variance value (VAF), and draw a schematic diagram of the path mechanism affecting the social health benefits of users based on the mediation effect type and mediation strength of each path.

[0110] Specifically, the types of mediation effects include: complete mediation, complementary partial mediation, competing partial mediation, and direct-only no-mediation effect. A complete mediation effect is characterized by a significant mediating effect only, with a significant indirect effect but no direct effect. A complementary partial mediation effect is characterized by significant direct and indirect effects in the same direction. A competing partial mediation effect is characterized by significant direct and indirect effects in opposite directions. A direct-only no-mediation effect is characterized by a significant direct effect only; the influence of variable X on variable Y is entirely achieved through the direct path, and the mediating variable M plays no role.

[0111] S41. Using Smart PLS software, calculate the specific indirect effect path coefficients of the social health benefits of users through the participation characteristics of key urban green space environmental characteristics and key recreational activity types. Use Bootstrap sampling to perform repeated sampling tests, obtain the T-statistic and P-value of each specific indirect effect path, identify significant specific indirect effect paths, and obtain the path coefficient distribution of each specific indirect effect path.

[0112] As shown in Table 6, there are 18 hypothetical paths with significant specific indirect effects. In the process of the impact of urban green space environmental characteristics on the social health benefits of users, the participation of 8 types of activities played a mediating role, including dance, photography, play, water activities, singing and playing instruments, painting and calligraphy, chess and cards, and small ball activities.

[0113] Table 6

[0114]

[0115] S42. Calculate the path coefficient of the total indirect effect of key urban green space environmental features on the social health benefits of users, compare the significance and direction of direct effects, total indirect effects and specific indirect effects, classify them according to the criteria for determining the type of mediation effect, and obtain the distribution of the mediation effect types of each key urban green space environmental feature affecting the social health benefits of users through the participation characteristics of different types of recreational activities.

[0116] Specifically, the mediation effect is classified according to the criteria for determining the type of mediation effect, including: when only the mediation effect is significant, it is determined to be a complete mediation effect; when both the direct and indirect effects are significant and in the same direction, it is determined to be a complementary partial mediation effect; when both the direct and indirect effects are significant but in opposite directions, it is determined to be a competing partial mediation effect; when only the direct effect is significant, it is determined to be a direct effect with no mediation effect; and when neither the direct nor the mediation effect is significant, it is determined to be an effect with no effect.

[0117] Table 7 shows the assessment results of the mediation effects of different activity types. First, the overall indirect effect of accessibility on users' social health benefits is not significant. However, there are three significant specific indirect effect pathways in the process of accessibility influencing users' social health benefits. This indicates that there are activity types with complementary and competitive mediation effects, making the overall indirect effect between accessibility and users' social health benefits insignificant. Second, safety, aesthetics, attractiveness, and ease of use have significant overall indirect effects on users' social health benefits. Compared with direct effects, there are mediating variables with complete mediation effects among safety, ease of use, and users' social health benefits. There are mediating variables with partial mediation effects among aesthetics, attractiveness, and users' social health benefits. Finally, the overall indirect effect of maintenance status on users' social health benefits is not significant. Moreover, there are no significant specific indirect effect pathways. Therefore, it can be concluded that none of the 14 activity types have a mediating effect between maintenance and users' social health benefits. This suggests that there may be other potential mediating variables not covered in this embodiment.

[0118] Table 7

[0119]

[0120] S43. Calculate the variance fraction (VAF) of each specific indirect effect path. Based on the magnitude of the VAF value, quantify the mediating strength of the participation characteristics of each key recreational activity type in the process of the influence of key urban green space environmental characteristics on the social health benefits of users. Rank the mediating importance of each specific indirect effect path. Finally, based on the significant direct effect path, the total indirect effect path, and the specific indirect effect path and their effect strength, create a schematic diagram of the mechanism of influence of urban green space users on social health benefits.

[0121] Direct effect paths refer to the paths through which urban green space environmental characteristics directly influence users' social health benefits; total indirect effect paths refer to the sum of paths through which environmental characteristics indirectly influence users' social health benefits via participation characteristics of all recreational activity types; specific indirect effect paths refer to the paths through which environmental characteristics influence users' social health benefits via participation characteristics of a single recreational activity type. The larger the absolute value of the mediating effect coefficient, the more important the role of that path in the relationship between environmental characteristics and users' social health.

[0122] In this embodiment, the three types of recreational activities partially mediate the relationship between accessibility and users' social health benefits. Water-related activities are the mediating variable with a complementary mediating effect, while photography and play activities are the mediating variables with a competing mediating effect. Specifically, the complementary mediating path "accessibility → water-related activities → users' social health" indicates that more easily accessible green spaces enhance users' social health benefits by encouraging participation in water-related activities. The competing mediating paths "accessibility → photography → users' social health" and "accessibility → play activities → users' social health" indicate that accessibility indirectly reduces users' social health benefits by encouraging play and photography activities. This may be because these activities are overly concentrated in specific groups such as families or photography enthusiasts, reducing social opportunities for adults.

[0123] Dance activities have a fully mediating effect on the relationship between perceived safety and users' social health benefits. The mediating path "perceived safety → dance activities → user social health" indicates that perceived safety indirectly promotes users' social health benefits by increasing participation in dance activities. In this context, dance activities amplify the positive impact of perceived safety on users' social health benefits. That is, in a safe environment, dance activities are more attractive to participants, thereby enhancing users' social health benefits.

[0124] Seven types of recreational activities partially mediated the relationship between aesthetic experience and users' social health benefits. Painting and calligraphy, dance, water activities, and ball games were complementary mediators, while singing and playing instruments, photography, and play were competing mediators. This indicates that singing and playing instruments, photography, and play activities enhanced the positive impact of aesthetic experience on users' social health, while painting and calligraphy, water activities, dance, and ball games enhanced the negative impact of aesthetic experience on users' social health.

[0125] On the one hand, the complementary mediating pathways of "aesthetics → painting and calligraphy → user social health," "aesthetics → water-based activities → user social health," "aesthetics → dance → user social health," and "aesthetics → ball games → user social health" indicate that the aesthetic appeal of green spaces diminishes the benefits to user social health through these activities. Painting, calligraphy, and water-based activities often require a tranquil and beautiful environment, which encourages participants to focus on personal contemplation and expression rather than social interaction. While these activities may have a positive impact on individual mental health, they may not be optimal social activities and therefore may have limited effectiveness in promoting user social health benefits. Tranquil and beautiful environments are typically dominated by plant and water features, which, while providing visual appeal, may not be sufficient to support a positive experience with active dance and ball games. For example, a lack of sufficient exercise space, suitable sports facilities, or well-organized competitions may reduce participant engagement and the overall experience, thus diminishing the user social health benefits.

[0126] On the other hand, the competing mediation pathways of "aesthetics → singing and performing activities → user social health," "aesthetics → photography and videography activities → user social health," and "aesthetics → play activities → user social health" indicate that the aesthetic appeal of green spaces enhances the social health benefits for users through these activities. Singing and performing activities and play activities are ways to express emotions and share shared experiences, often promoting social interaction and relationships among participants. Groups gathering in beautiful environments for photography and videography activities often share a common interest in natural landscapes, which helps to establish and deepen social connections.

[0127] Six types of recreational activities partially mediated the relationship between attractiveness and users' social health benefits. Water-based activities, dancing, board games, and ball games were complementary mediators. Photography and play activities were competing mediators.

[0128] On the one hand, the complementary mediating effect pathways—"Attraction → Dance → User Social Health," "Attraction → Water Activities → User Social Health," "Attraction → Dance → User Social Health," "Attraction → Chess and Card Games → User Social Health," and "Attraction → Ball Games → User Social Health"—indicate that the attractiveness of green spaces enhances user social health benefits through these activities. Water activities and dance activities are more likely to promote participation in attractive environments, thereby promoting social interaction and user social health benefits. Chess and card games and ball games, due to their social and competitive nature, are more likely to attract participants in highly attractive environments, thus promoting teamwork and the establishment of social relationships.

[0129] On the other hand, the competing mediation path of "attractiveness → photography activities → user social health" and "attractiveness → play activities → user social health" suggests that the attractiveness of green spaces diminishes the social health benefits for users through these activities. Photography activities, such as photographing birds or plants, tend to focus more on individual skills and interests. Therefore, these activities may attract specific groups in attractive green space areas but may not enhance the broader social health benefits for users. Similarly, play activities, especially those for children and families, may be over-concentrated in attractive green space areas. This may prevent other users, especially adults, from fully participating in or enjoying these activities, thus diminishing the broader social health benefits for users.

[0130] Dance activities have a fully mediating effect on the relationship between ease of use and user social health benefits, and are competing mediating variables. Improved ease of use does indeed help increase participation in dance activities, which is a positive sign for the specific group that loves dance. However, if dance activities primarily attract this small group and do not encompass the more diverse users in the green space, their effectiveness in promoting overall user social health will be limited.

[0131] The larger the absolute value of the mediation effect coefficient, the more important the role of this path in the relationship between environmental characteristics and users' social health. For detailed analysis results, please refer to Table 8 below.

[0132] Table 8. Path coefficients and significance of direct effects; path coefficients and significance of specific indirect effects; what mediating role did the activity type play?

[0133]

[0134] like Figure 3 The diagram illustrates the impact mechanism of social health benefits on urban green space users, drawn based on the analysis results. Using path coefficients and significance test results calculated through the analytical model of social health benefits for urban green space users, this diagram shows that, without considering demographic characteristics, green space accessibility, safety, aesthetics, attractiveness, maintenance, and usability influence users' social health benefits through various recreational activity types. Specifically, it includes direct effect paths—paths where six environmental characteristics directly impact users' social health benefits; and mediating effect paths—paths where environmental characteristics indirectly influence users' social health benefits through 14 recreational activity types, including sightseeing, play, photography, water activities, animal activities, singing and playing instruments, chess and cards, painting and calligraphy, exhibitions, running and brisk walking, dancing, qigong and martial arts, ball games, and fitness equipment.

[0135] This example provides a method for analyzing the social health benefits of urban green space users based on the PLS-SEM model. By standardizing new data according to a predetermined measurement model structure, it ensures that the 27 manifest variables corresponding to the six constructs of urban green space environmental characteristics (accessibility, safety, aesthetics, attractiveness, maintenance, and usability), the participation variables for 14 types of recreational activities, and the three dimensions of user social health benefits (social relationships, social support, and social interaction) all meet the model input requirements. Secondly, the processed new data is imported into the pre-constructed PLS-SEM model in Smart PLS software. The system automatically calculates the factor loadings, path coefficients, and significance levels of each latent variable. Then, based on preset evaluation criteria (factor loadings ≥ 0.7, CR ≥ 0.7, AVE ≥ 0.5, VIF < 5), the reliability and validity of the measurement model for the new data are verified using R... 2 Value, Q 2The system uses SRMR values ​​to test the goodness of fit of the structural model on new data. Next, it outputs the ranking results of direct path coefficients of environmental features on users' social health benefits, identifying the environmental factors with the most significant impact on the social health benefits of new sample users. It also outputs the direct effect analysis results of 14 activity types, determining the key activity types that promote users' social health benefits. Finally, the mediation effect analysis module determines the mediation effect type (full mediation, partial mediation, or no mediation effect) of various key activity types that promote users' social health benefits, calculates the VAF value to quantify the mediation strength, analyzes the mediation effect type of each path, and draws a schematic diagram of the path mechanism affecting users' social health benefits based on the mediation effect type of each path. This schematic diagram clearly reveals how the six environmental features of urban green spaces (accessibility, safety, aesthetics, attractiveness, maintenance, and usability) affect users' social health benefits through 14 recreational activity types. First, it clarifies the priority of environmental characteristics, namely the degree to which attractiveness, aesthetics, accessibility, and maintenance affect the social health benefits of users, thus guiding planners to prioritize resources for improving high-impact factors such as attractiveness. Second, it identifies dance and small ball games as key activity types that promote the social health benefits of users, providing clear direction for functional zoning and facility configuration. Finally, through mediation effect analysis, it reveals the bridging role of different key activity types between environmental characteristics and the social health of users. The schematic diagram of the path mechanisms affecting the social health benefits of users helps designers understand how to guide beneficial social activities through environmental design, providing a scientific basis for urban green space planning and design. In practical applications, planners can use the schematic diagram of the path mechanisms affecting the social health benefits of users for differentiated design: for example, setting up an open dance plaza with sound facilities in the center of the green space to utilize the strong positive effect of dance activities; arranging small sports facilities such as ping-pong tables and chess tables around water features to leverage the social promotion role of small ball games and chess; and avoiding excessive pursuit of static aesthetics while neglecting the creation of social functional spaces.

[0136] Example 2:

[0137] This embodiment provides a computer device, which may be a server, computer, etc., including a processor, memory, input device, display, and network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. When the processor executes the computer programs stored in the memory, it implements the social health benefit analysis method for urban green space users based on the PLS-SEM model of Embodiment 1 described above, as follows:

[0138] S1. Based on the PLS-SEM model, construct an analytical model of the social health benefits of urban green space users. The analytical model includes a measurement model and a structural model. The latent variables are various urban green space environmental characteristics, participation characteristics of various recreational activity types, and social health benefits of users. The measurement model is constructed based on the latent variables and their corresponding manifest variables, and the structural model is constructed based on the relationship between the latent variables.

[0139] S2. Based on the preset evaluation criteria, verify the reliability and validity of the measurement model and the goodness of fit of the structural model respectively, and obtain the verified measurement model and structural model.

[0140] S3. Calculate the direct path coefficients of various urban green space environmental characteristics on users' social health benefits, calculate the direct path coefficients of various recreational activity type participation characteristics on users' social health benefits, and identify the key urban green space environmental characteristics and key recreational activity type participation characteristics of users' social health benefits based on the direct path coefficients.

[0141] S4. Analyze the mediation effect types of key urban green space environmental characteristics and key recreational activity types participation characteristics, quantify the mediation strength based on the variance value (VAF), and draw a schematic diagram of the path mechanism affecting the social health benefits of users based on the mediation effect type and mediation strength of each path.

[0142] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be considered equivalent substitutions and shall be included within the protection scope of the present invention.

Claims

1. A method for analyzing social health benefits of urban green space users based on a PLS-SEM model, characterized in that, The method comprises the following steps: S1, constructing an analysis model of social health benefits of urban green space users based on a PLS-SEM model, the analysis model comprising a measurement model and a structural model, taking various urban green space environment characteristics, various recreational activity type participation characteristics, and user social health benefits as latent variables, constructing the measurement model based on the latent variables and corresponding manifest variables, and constructing the structural model according to the relationships between the latent variables; S2, verifying the reliability and validity of the measurement model and the goodness of fit of the structural model based on preset evaluation criteria, and obtaining the verified measurement model and the verified structural model; S3, calculating direct path coefficients of the various urban green space environment characteristics on the user social health benefits, calculating direct path coefficients of the various recreational activity type participation characteristics on the user social health benefits, and identifying key urban green space environment characteristics and key recreational activity type participation characteristics of the user social health benefits according to the direct path coefficients; S4, analyzing the mediation effect types of the key urban green space environment characteristics and the key recreational activity type participation characteristics, quantifying the mediation intensity according to the variance contribution VAF, and drawing a path mechanism diagram for influencing the user social health benefits according to the mediation effect types and the mediation intensity of each path. 2.The PLS-SEM model based urban green space user social health benefit analysis method of claim 1, wherein, The step S1 comprises: S11, taking various urban green space environment characteristics, various recreational activity type participation characteristics, and user social health benefits as latent variables, the various urban green space environment characteristics comprising accessibility, safety, beauty, attractiveness, ease of use, and maintenance condition, obtaining accessibility latent variables, safety latent variables, beauty latent variables, attractiveness latent variables, ease of use latent variables, maintenance condition latent variables, various recreational activity type participation characteristic latent variables, and user social health benefit latent variables; S12, constructing an analysis model of social health benefits of urban green space users based on a PLS-SEM model, constructing a measurement model based on the latent variables and corresponding manifest variables, defining each latent variable and its corresponding manifest variable, associating each latent variable with its corresponding manifest variable, constructing a structural model according to the relationships between the latent variables, and describing the relationships between the latent variables through the structural model. 3.The PLS-SEM model based urban green space user social health benefit analysis method of claim 2, wherein, The step of constructing a structural model of the analysis model, taking the various urban green space environment characteristics as exogenous variables, the various recreational activity type participation characteristics as intermediary variables, and the user social health benefits as endogenous variables; establishing a hypothetical path, assuming that the various urban green space environment characteristics and the user social health benefits have a positive correlation relationship, the various recreational activity type participation characteristics and the user social health benefits have a positive correlation relationship, and the various recreational activity type participation characteristics play a mediating role in the influence of the various urban green space environment characteristics on the user social health benefits; ​ The path relationship is set, a plurality of city green space environment characteristics directly affecting a path of a user social health benefit, a plurality of recreational activity type participation characteristics directly affecting a path of the user social health benefit, and a plurality of city green space environment characteristics affecting the user social health benefit through a plurality of recreational activity type participation characteristics are set, and a plurality of hypothesis paths are obtained. 4.The PLS-SEM model based analysis method of social health benefits of urban green space users according to claim 1, wherein, The step S2 comprises: S21, measuring model estimation is performed by selecting a PLS algorithm through the Smart PLS software, factor loading, Cronbach's Alpha coefficient, combined reliability, average variance extraction value, and variance inflation factor are calculated, preset evaluation criteria of the reliability and validity are set, and the reliability and validity of the measuring model are verified according to the evaluation criteria of the reliability and validity; S22, running the structural model by using PLS algorithm through Smart PLS software, respectively calculating R 2 value, Q 2 value and algorithm SRMR value, presetting the evaluation standard of fitting degree, verifying the fitting degree of the structural model according to the evaluation standard of fitting degree. 5.The PLS-SEM model based urban green space user social health benefit analysis method of claim 4, wherein, The preset evaluation criteria of the reliability and validity include that the factor loading is greater than or equal to 0.7, the Cronbach's Alpha coefficient is greater than or equal to 0.6, the combined reliability CR is greater than or equal to 0.7, the average variance extraction value AVE is greater than or equal to 0.5, and the variance inflation factor VIF is less than 5; The preset goodness-of-fit evaluation criteria include: R 2 value > 0.6, Q 2 value > 0, SRMR value < 0.

08. 6.The PLS-SEM model based analysis method of social health benefits of urban green space users according to claim 4, wherein, The step S3 comprises: S31, the measuring model and the structural model data that pass through the reliability and validity verification are imported into the Smart PLS software, a direct path coefficient of a plurality of city green space environment characteristics on a user social health benefit is calculated, repeated sampling inspection is performed through a Bootstrap sampling method, T statistics and P values of each path are obtained, statistical significance of the path coefficient is judged according to the T statistics and the P values, and key city green space environment characteristics that have a direct and significant positive influence on the user social health benefit are identified according to the direct path coefficient and the statistical significance; S32, a path model is established in the Smart PLS software, a plurality of recreational activity type participation characteristics are set as exogenous variables, the user social health benefit is set as an endogenous variable, a direct path relationship is constructed, a PLS algorithm is run to calculate a path coefficient, standardized path coefficient values of each recreational activity type on the user social health benefit are output, repeated sampling inspection is performed through a Bootstrap sampling method, T statistics and P values of each path are obtained, statistical significance of the path coefficient is judged according to the T statistics and the P values, and key recreational activity type participation characteristics that have a direct and significant positive influence on the user social health benefit are identified according to the direct path coefficient and the statistical significance. 7.The PLS-SEM model based analysis method of social health benefits of urban green space users according to claim 6, wherein, The step S4 comprises: S41, a specific indirect effect path coefficient of key city green space environment characteristics on a user social health benefit through key recreational activity type participation characteristics is calculated by using the Smart PLS software, repeated sampling inspection is performed through a Bootstrap sampling method, T statistics and P values of each specific indirect effect path are obtained, significant specific indirect effect paths are identified, and path coefficient distributions of each specific indirect effect path are obtained. S42, calculate the total indirect effect path coefficient of the key urban green space environmental characteristics on the social health benefits of users, compare the significance and direction of the direct effect, total indirect effect and specific indirect effect, classify according to the mediation effect type judgment standard, and obtain the mediation effect type distribution of each key urban green space environmental characteristic through different recreation activity type participation characteristics to influence the social health benefits of users; S43, calculate the variance of each specific indirect effect path VAF, quantify the mediation intensity of each key recreation activity type participation characteristic in the process of the key urban green space environmental characteristics influencing the social health benefits of users according to the size of the VAF value, sort the mediation role of each specific indirect effect path, and finally make a city green space user social health benefit influence mechanism diagram according to the significant direct effect path, total indirect effect path and specific indirect effect path and their effect intensity. 8.The PLS-SEM model based analysis method of social health benefits of urban green space users according to claim 7, wherein, The classification according to the mediation effect type judgment standard comprises: determining as complete mediation effect when only the mediation effect is significant; determining as complementary partial mediation effect when the direct effect and the indirect effect are both significant and the directions are the same; determining as competitive partial mediation effect when the direct effect and the indirect effect are both significant but the directions are opposite; determining as only direct non-mediation effect when only the direct effect is significant; and determining as no effect when the direct effect and the mediation effect are both not significant.

9. A computer device comprising a processor and a memory for storing a processor executable program, characterized in that, The processor implements the PLS-SEM model-based city green space user social health benefit analysis method of any one of claims 1-8 when executing the program stored in the memory.

Images