Method and system for constructing a model of the driving forces and barriers to healthy behavior in urban green spaces

Abstract

The present invention relates to the technical field of research on health benefits of social benefits, and is a method and system for constructing a driving force and barrier model for healthy behavior in urban green spaces. The method includes: analyzing objective barriers and subjective barriers in the process of visiting urban green spaces, and proposing a mediation model hypothesis between the perceived quality of urban green spaces and the health behavior stage of users in green spaces based on the driving force and barrier model for healthy behavior in urban green spaces; establishing a structural model based on the mediation model hypothesis, and combining the structural model and the measurement model to form a mediation model. When the mediation model is established, the applicability of the mediation model is tested, and the stability of the mediation model is evaluated. The present invention constructs a driving force and barrier model for healthy behavior in urban green spaces, which takes into account the role of users in overcoming subjective barriers when visiting green spaces, has good applicability and stability, and can assist in the development of public participation activities in stages such as urban green space planning and design decision-making and event planning.

Description

Technical Field

[0001] The present invention relates to the technical field of research on health benefits belonging to social benefits in a landscape performance evaluation system, and in particular to a method and system for constructing a driving force and barrier model for healthy behavior in urban green spaces. Background Art

[0002] In recent years, human health and well-being have received extensive and in-depth attention around the world, and visiting urban green spaces is one of the many healthy behaviors of urban residents. Previous studies have used the accessibility and functionality of urban green spaces to explore the objective obstacles faced by users when visiting green spaces, but the mechanism of how urban green spaces can help potential users overcome subjective obstacles and visit green spaces remains unclear. Therefore, how to construct a user-centered urban green space health behavior model to assist public participation activities in urban green space planning and design decision-making and event planning is an important issue worthy of study.

[0003] In response to the above-mentioned issues, many studies have emerged internationally to summarize the pathways and mechanisms of action of urban green space on residents' health, such as the research of Hartig T, Lachowycz K, Villanueva K, etc. These pathways include the level of provision of "ecological products and services" that are beneficial to health, the level of stress perception, the level of concentration, physical activity, social cohesion, and neighborhood satisfaction, and it is believed that the various factors are interrelated and synergistic to produce health effects. Based on the mediation analysis method in social psychology, Herzele A. and Vries S. regard the above pathways as mediators between green space and health, and propose a mediation model between neighborhood green space and health. This model explains "what kind of natural environment and activities have a positive impact on the health of its participants" and "how such natural environment and activities are provided and carried by the material environment of urban green space"; and some studies have deepened the mediating factors into planning and design strategies after empirical or case analysis, so as to provide corresponding natural environment and activity space or facilities in green space.

[0004] However, as people are the main body of green space visit behavior, this mechanism does not reflect people's willingness during the green space visit process. The connection between "planning and design provide space and facilities for healthy environment and activities" and "people are affected by natural environment and healthy activities in green space to improve physical and mental health" is not self-evident. In between the two is the healthy behavior and process of people going to green space and then contacting healthy environment and engaging in healthy activities. Healthy behavior involves the actions of individuals, groups and institutions, and the determinants, interrelationships and results of these actions. These results may be social changes, policy development and implementation, improvement of coping skills and improvement of quality of life.

[0005] As for the role of urban green space in residents' health behavior, if the objective obstacles encountered by people in accessing and using green space can be responded to by improving the accessibility and functionality of green space, then when the material and social environment required for healthy behavior are in place, people with free will and the right to choose may still face their own subjective obstacles. The belief held by users in their ability to overcome such obstacles is called self-efficacy. In the study of urban green space and health, the mechanism of action of urban green space in affecting health behavior is not clear, and visiting urban green space is an important part of health behavior. Previous studies focused on analyzing the objective obstacles faced by users when visiting green space, such as the accessibility and functionality of using urban green space, and ignored the subjective obstacles overcome by users when visiting green space. Summary of the invention

[0006] In order to solve the technical problems existing in the prior art, the present invention provides a method and system for constructing a driving force and barrier model for healthy behavior in urban green spaces. According to the driving force and barrier model for healthy behavior in urban green spaces, a mediation model hypothesis between the perceived quality of urban green spaces and the health behavior stage of users in green spaces is proposed. The mediation model hypothesis uses self-efficacy for overcoming barriers as a mediating variable. The mediation model is constructed according to the mediation model hypothesis, so that the construction of the driving force and barrier model for healthy behavior in urban green spaces takes into account the role of users in overcoming subjective barriers when visiting green spaces, reflects people's willingness during the green space visit process, and has good applicability and stability.

[0007] The first object of the present invention is to provide a method for constructing a driving force and obstacle model for healthy behavior in urban green spaces.

[0008] A second object of the present invention is to provide a system for constructing a model of motivation and obstacles for healthy behavior in urban green spaces.

[0009] The first object of the present invention can be achieved by adopting the following technical solutions:

[0010] Methods for modeling the drivers and barriers to healthy behaviors in urban green spaces, including:

[0011] S1. Analyze the objective and subjective barriers in the process of visiting urban green spaces and establish a model of motivation and barriers to healthy behavior in urban green spaces;

[0012] S2. Based on the motivation and barrier model of healthy behavior in urban green space, a mediation model hypothesis is proposed between the perceived quality of urban green space and the health behavior stage of users in green space. The mediation model hypothesis uses self-efficacy to overcome barriers as a mediating variable.

[0013] S3, respectively measure the perceived quality of urban green space, self-efficacy in overcoming obstacles, and health behavior status in urban green space and obtain corresponding measurement data, pre-process the measurement data, and establish a measurement model;

[0014] S4. Establish a structural model based on the hypothesis of the mediation model, combine the structural model and the measurement model into a mediation model, conduct a covariance structure analysis on the actual data and the mediation model, and determine whether the mediation model is established;

[0015] S5. When the mediation model is established, test the applicability of the mediation model; test the total effect, indirect effect and direct effect between variables in different groups separately to evaluate the stability of the mediation model.

[0016] The second object of the present invention can be achieved by adopting the following technical solutions:

[0017] The systems that drive and hinder healthy behaviors in urban green spaces are modeled, including:

[0018] The driving force and barrier model module is used to analyze the objective and subjective barriers in the process of visiting urban green spaces and establish a driving force and barrier model for healthy behavior in urban green spaces;

[0019] The mediation model hypothesis module is used to propose a mediation model hypothesis between the perceived quality of urban green space and the health behavior stage of users in green space based on the motivation and barrier model of urban green space health behavior. The mediation model hypothesis uses self-efficacy to overcome barriers as a mediating variable;

[0020] The measurement model module is used to measure the perceived quality of urban green space, self-efficacy in overcoming obstacles, and health behavior status in urban green space and obtain corresponding measurement data, pre-process the measurement data, and establish a measurement model;

[0021] The mediation model module is used to establish a structural model based on the mediation model hypothesis, combine the structural model and the measurement model to form a mediation model, conduct covariance structure analysis on the actual data and the mediation model, and determine whether the mediation model is established;

[0022] The mediation model testing and evaluation module is used to test the applicability of the mediation model when the mediation model is established; the total effect, indirect effect and direct effect between variables in different groups are tested separately to evaluate the stability of the mediation model.

[0023] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0024] The present invention proposes a mediation model hypothesis between the perceived quality of urban green space and the health behavior stage of users in green space based on the motivation and barrier model of healthy behavior in urban green space. The mediation model hypothesis uses self-efficacy to overcome barriers as a mediating variable. A structural model is established based on the mediation model hypothesis. The structural model and the measurement model constitute a mediation model. When the mediation model is established, the applicability of the mediation model is tested and the stability of the mediation model is evaluated. This allows the construction of the motivation and barrier model of healthy behavior in urban green space to take into account the role of users in overcoming subjective barriers when visiting green space, and reflects people's willingness during the green space visit process. The present invention has good applicability and stability. The present invention can assist in the development of public participation activities in stages such as urban green space planning and design decision-making and activity planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying creative work.

[0026] Figure 1 is a flow chart of a model building method in an embodiment of the present invention;

[0027] Figure 2 is a schematic diagram of a driving force and obstacle model for healthy behavior in urban green space in an embodiment of the present invention;

[0028] Figure 3 The intermediate model assumption of the embodiment of the present invention is a schematic diagram;

[0029] Figure 4 Schematic diagram of the structure of the mesomodel in the embodiment of the present invention. DETAILED DESCRIPTION

[0030] The technical solution of the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is obvious that the described embodiments are part of the embodiments of the present invention, not all of the embodiments, and the implementation of the present invention is not limited to this. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0031] Example 1

[0032] The present invention is user-centered and proposes a "mediation model between perceived urban green space quality and user health behavior stage". The model is composed of perceived urban green space quality, which is composed of perceived urban green space environmental aesthetic quality and perceived urban green space open space communication quality. In the process of the user's visit to the green space, which is a stage of health behavior, the self-efficacy of overcoming subjective obstacles is a mediating variable. The present invention can assist in the development of public participation activities in the stages of urban green space planning and design decision-making and activity planning. The model obtained by the present invention can be used in subsequent empirical research for further demonstration.

[0033] Through theoretical research, model hypotheses are proposed, and a mediation model between the perceived quality of urban green space and the health behavior stages of users in green space is constructed with self-efficacy to overcome obstacles as the mediating variable. A cross-sectional study is also conducted. The hypothesis of the mediation model can only be proved to be valid if it passes empirical testing.

[0034] like Figure 1 As shown in the flowchart of the model construction method, this embodiment proposes a method for constructing a driving force and obstacle model of healthy behavior in urban green space, including the following steps:

[0035] S1. Analyze the objective and subjective barriers in the process of visiting urban green spaces and establish a model of motivation and barriers to healthy behavior in urban green spaces.

[0036] By reviewing the literature, we summarize the mainstream research perspective, which is to focus on how urban green space planning and design can remove the objective obstacles in the user's visit process; combined with the deduction of the summarized content and the observation and interview of the use of urban green space, we propose this research perspective, that is, whether urban green space planning, design, management, and services can help users and potential users overcome the subjective obstacles in the visit process, and propose a model of motivation and obstacles for healthy behavior in urban green space. The motivation and obstacles for healthy behavior in urban green space are shown in Table 1.

[0037] Table 1

[0038]

[0039] like Figure 2As shown in the figure, the model diagram of the motivation and obstacles of healthy behavior in urban green space is as follows. In the process of actually carrying out activities in urban green space, users perceive or think about the quality of urban green space, thus forming relatively unconscious feelings and relatively conscious cognition of the activity environment and the activity itself; the content of feelings and thoughts, after being screened by users' "identification" or "rejection", forms their personal experience of the green space environment and the activity itself; this personal experience is the user's experience, which together with the directly received social experience information, acts on the motivation generation process of the next healthy behavior and becomes the basis for users to judge whether they intend to carry out the next behavior; and then enters the next healthy behavior process concerned by the health behavior programming model, forming a cycle. When this cycle is repeated, it can be considered that the healthy behavior is maintained by the user.

[0040] S2. Based on the motivation and barrier model of healthy behavior in urban green spaces, a mediation model hypothesis is proposed between the perceived quality of urban green spaces and the health behavior stages of users in green spaces. The mediation model hypothesis uses self-efficacy to overcome barriers as a mediating variable.

[0041] Guided by the research perspective, a mediation model hypothesis is proposed, that is, whether the level of perceived urban green space quality can affect the health behavior of users by affecting their self-efficacy to overcome subjective barriers to visit green space. The perceived environmental aesthetic quality in urban green space (PEAQ) and the perceived open space quality in urban green space (POSQ) are temporarily referred to as perceived urban green space quality (PUGSQ). The self-efficacy to overcome barriers to health behavior is abbreviated as OBSE, and the health behavior stage in urban green space of potential users is recorded as HB_UGS.

[0042] like Figure 3As shown in the figure, for the relationship between the above three variables (perceived urban green space quality PUGSQ, self-efficacy to overcome obstacles OBSE, and health behavior stage HB_UGS of potential users in urban green space), the study proposed five sets of null hypotheses (H0) and their corresponding mediation model hypotheses (H1, H2, H3, H4, H5), and provided two null hypotheses (H0(6), H0(7)) for the applicability of the mechanism model in different groups. The mediation model hypotheses specifically include:

[0043] 1. The direct effect of perceived urban green space quality PUGSQ on the health behavior stage HB_UGS of potential users of urban green space, and the effect of perceived urban green space quality PUGSQ on self-efficacy to overcome obstacles OBSE

[0044] H0(1): The path coefficient from PUGSQ to HB_UGS is not significant, and the perceived quality of urban green space has no direct relationship with the stage of health behavior in urban green space.

[0045] H1: The path coefficient from PUGSQ to HB_UGS is significant and positive, and the perceived quality of urban green space has a positive direct impact on the health behavior stage in urban green space.

[0046] H0(2): The path coefficient from PUGSQ to OBSE is not significant, and the perceived quality of urban green space is not related to health behavior self-efficacy.

[0047] H2: The path coefficient from PUGSQ to OBSE is significant and positive, and the perceived quality of urban green space has a positive impact on health behavior self-efficacy.

[0048] 2. The effect of self-efficacy to overcome obstacles OBSE on the health behavior stage HB_UGS of potential users in urban green space, and the total effect of perceived urban green space quality PUGSQ on the health behavior stage HB_UGS of potential users in urban green space;

[0049] H0(3): The path coefficient from OBSE to HB_UGS is not significant, and health behavior self-efficacy is not related to the health behavior stage in urban green space.

[0050] H3: The path coefficient from OBSE to HB_UGS is significant and positive, and health behavior self-efficacy has a positive impact on the health behavior stage in urban green spaces.

[0051] H0(4): The total effect H4 from PUGSQ to HB_UGS (numerically, H4=H1+H2*H3) is not significant, and the perceived quality of urban green space is not related to the stage of health behavior in urban green space.

[0052] H4: The total effect of PUGSQ to HB_UGS H4 is significant and positive, and the perceived quality of urban green space has a positive impact on the health behavior stage in urban green space.

[0053] 3. The indirect effect of perceived urban green space quality PUGSQ on the health behavior stage HB_UGS of potential users of urban green space, that is, the perceived urban green space quality PUGSQ to the health behavior stage HB_UGS of potential users of urban green space through the self-efficacy of overcoming obstacles OBSE;

[0054] H0(5): The product of H2 and H3, H5 (numerically, H5 = H2*H3), is not significant, the indirect effect from PUGSQ to HB_UGS does not exist, and the mediating effect of OBSE does not exist.

[0055] H5: The product of H2 and H3, H5, is significant and positive. PUGSQ has a positive impact on HB_UGS indirectly through the mediating variable OBSE.

[0056] 4. Comparison of multiple groups with different activity intensities and multiple groups with different social statuses.

[0057] H0(6): The model of the low activity intensity group and the high activity intensity group have equal measurement model factor loadings, equal structural model path coefficients, and equal structural model variances and covariances.

[0058] H0(7): For the models of the solitary activity group and the group that engages in companionship, the measurement model factor loadings, structural model path coefficients, structural model variances, and covariances of the two groups are all equal.

[0059] S3. Measure the perceived quality of urban green space, self-efficacy in overcoming obstacles, and health behavior status in urban green space and obtain corresponding measurement data, pre-process the measurement data, and establish a measurement model.

[0060] Preferably, the perceived quality of urban green space, self-efficacy in overcoming obstacles, and health behavior status in urban green space are measured separately and corresponding measurement data are obtained, including drawing on proven scales, works that have been widely discussed and recognized, and observations and interviews with urban green space users, formulating quantitative scales for the three variables of perceived quality of urban green space, self-efficacy in overcoming obstacles, and health behavior status in urban green space to form a questionnaire, distributing the questionnaires and collecting them, and screening the data from the collected questionnaires.

[0061] In this embodiment, the scales are the "Perceived Environmental Aesthetics Scale", "Self-efficacy Scale for Overcoming Physical Activity Obstacles", "Healthy Behavior Stage Scale" and the compilation of the "Perceived Urban Green Open Space Communication Quality Scale". With the assistance of volunteers, the online questionnaire was placed on social platforms such as WeChat, QQ, Weibo, Taobao, and Baidu Tieba from March 3, 2020 to March 20, 2020, inviting netizens to fill it out voluntarily. In the end, a total of 695 questionnaires were collected. In the end, after screening, a total of 256 questionnaires were deleted and removed, and the remaining 439 questionnaires participated in the formal data analysis, questionnaires were placed, and data were collected and screened.

[0062] Preferably, a measurement model is established, including: for the unequally spaced ordered categorical variable of health behavior status in urban green space, performing optimal scaling conversion in the software SPSS Statistics; for the two relatively equidistant variables of perceived quality of urban green space and self-efficacy in overcoming obstacles, performing exploratory factor analysis on the measurement data in the software SPSS Statistics, and then further performing confirmatory factor analysis in the software SPSS Amos; and finally establishing a measurement model for the three variables of perceived quality of urban green space, self-efficacy in overcoming obstacles, and health behavior status in urban green space for later use.

[0063] For the ordered categorical variable of health behavior status with unequal intervals, the optimal scale transformation was performed in SPSS Statistics 23.0. Specifically, the analysis-regression-optimal scaling was run in SPSS Statistics software, and the "health behavior variable" was selected into the dependent variable box, and its scale was defined as "ordered", and the 14 self-efficacy measurement indicators that had undergone confirmatory factor analysis were selected into the independent variable box, and their scale was defined as "number". The optimal scale regression results showed that the selected independent variables had a significant effect on the dependent variable (p = 0.000 < 0.001), indicating that the regression model was statistically significant.

[0064] For the two relatively equidistant variables of perceived urban green space quality and self-efficacy in overcoming obstacles, all measurement data were subjected to exploratory factor analysis in SPSS Statistics 23.0, and then further confirmatory factor analysis was performed in SPSS Amos 23.0. Exploratory factor analysis was performed with the help of SPSS Amos software. In SPSS Statistics, "Analysis-Scale-Reliability Analysis" was selected to calculate the overall reliability of all measurement indicators. The internal reliability coefficient cronbach's α was greater than 0.8, and it can be considered that these measurement indicators have good internal reliability overall. Preferably, SPSS Statistics "Analysis-Dimensionality Reduction-Factor" was used to explore the structure of measurement indicators. First, choose to output the KMO (Kaiser-Meyer-Olkin) statistic and Bartlett test results to prove whether all measurement indicators are suitable for factor analysis. The KMO test statistic is an indicator used to compare the simple correlation coefficient and partial correlation coefficient between variables. It is mainly used in factor analysis of multivariate statistics. The KMO statistic takes values ​​between 0 and 1; select "Principal axis factoring (sometimes also called principal axis factor method)" as the factor extraction method; select "covariance matrix analysis" as the analysis method; by default, extract factors with eigenvalues ​​greater than 1 multiplied by the average eigenvalue, that is, the number of extracted factors is determined according to the size of the eigenvalue root; in the factor rotation method, the "optimal oblique method (Promax)" in the oblique rotation method is selected. Choose not to display the values ​​of factor loadings less than 0.3. Confirm the settings and run the calculation. If the calculated KMO of the measurement indicator is greater than 0.9, it means that there is an obvious structural relationship between the measurement indicators, which is very suitable for factor analysis.

[0065] Confirmatory factor analysis was performed with the help of SPSS Amos software. The "factors" in the SPSS Statistics analysis process will be renamed "constructs" in SPSS Amos, and they all represent the distribution structure of measurement indicators. According to the needs of this study, the purpose of confirmatory factor analysis is to produce a good-quality second-order measurement model for the variables involved in the mediation model. For each variable, after a large number of measurement indicators have been reduced in dimension once to form multiple "factors / first-order constructs", if you want to reduce these first-order constructs again to form each variable with only one construct, that is, a higher-level "factor / second-order construct", then using SPSS Amos instead of SPSS Statistics can respect the original data more and the result presentation method is more intuitive.

[0066] Preferably, in SPSS Amos software, a path diagram is established for the two models, and the model fitness index is written directly above the path diagram. Confirmatory factor analysis includes four steps: dealing with collinearity problems, optimizing factor loadings, correcting model fitness, and checking the intrinsic quality of the model; when the correlation coefficient between the first-order constructs of a measurement model is close to or greater than 0.75, or the standardized factor loading of the first-order construct on the second-order construct is greater than 0.95 or even close to or exceeds 1.00, it is considered that there is a collinearity problem between the first-order constructs, that is, they are very similar, and it is necessary to merge the paired first-order constructs with too large correlation coefficients, or delete the first-order constructs with too large loadings; with the assistance of the model modification index (Modification Indices) provided by SPSS Amos, the two models obtained by the merging method and the deletion method are modified to improve the fit between the model structure and the actual data.

[0067] In summary, after several rounds of revisions and comparisons between the original model and the exploratory model, a measurement model of three variables was finally established for future use.

[0068] In this example, among the measured variables, there are 2 measurement questions for the health behavior stage in urban green space (HB_UGS), but only 1 question is actually used. Such a small amount of measurement data is due to the overall trade-off and consideration of this study, but it is indeed not suitable to be directly used to establish the measurement model of this variable, but requires some preprocessing. Preferably, the preprocessing of the measurement data includes: theoretically guided measurement data setting, questionnaire internal testing and revision, abnormal option elimination, SPSSStatistics software reliability analysis, SPSS Statistics software optimal scale conversion, and SPSS Amos software single indicator reliability assignment.

[0069] Preferably, in the SPSS Amos software, a measurement model is established separately for the measurement data and error term e of the health behavior stage HB_UGS in the urban green space of potential users and the health behavior stage HBG_CP in the urban green space of potential users. The calculated measurement error of 0.13 is filled in the variance of the error term e, and 1 is filled in the variance of the construct; the path coefficient from the error term to the measurement data defaults to 1, and the path coefficient from the measurement data to the construct is the value to be estimated and is not filled. Run SPSS Amos for estimation, and the factor loading of the measurement data on the construct is 0.93. At the same time, the reliability of the measurement data is verified to be 0.87, which is equal to the initial setting value, indicating that the above calculation is successful.

[0070] The other three variables, perceived aesthetic quality of urban green space environment (PEAQ), perceived social quality of urban green space open space (POSQ), and self-efficacy in overcoming obstacles to engage in healthy behaviors (OBSE), each have a scale with a large number of questions. From the selection or compilation of the scale to the final formation of a qualified measurement model, the operation process includes:

[0071] Theoretical guidance of measurement data setting and classification, questionnaire internal testing and revision, SPSS Statistics reliability analysis, SPSS Statistics factor analysis, SPSS Amos collinearity problem processing, SPSS Amos factor loading optimization, SPSS Amos model fitness correction, and model quality inspection.

[0072] Finally, the perceived aesthetic quality of urban green space environment (PEAQ) and the perceived open space communication quality of urban green space (POSQ) together constitute the perceived quality of urban green space, forming a qualified measurement model; the self-efficacy of overcoming obstacles to engage in healthy behaviors (OBSE) forms a qualified measurement model, and the health behavior status in urban green space finally forms a qualified measurement model.

[0073] For the perceived aesthetic quality of urban green space environment (PEAQ), a scale with 23 measurement data and 5 factors was obtained from the original text of the PEAQ scale developer. After revision, a second-order model with 9 measurement data and 3 first-order constructs was obtained. The measurement data and first-order constructs have undergone major deletions and revisions. The timing and reasons for the deletions are summarized in Table 2. The deletions and classifications of the original measurement data of the PEAQ scale are summarized as follows:

[0074] Table 2

[0075]

[0076] For the perceived communication quality of urban green open space (POSQ), the present invention proposes a hypothesis with 15 measurement data, 5 of which are deleted after processing. The reasons for deletion are shown in Table 3. The deletion and retention classification of the original measurement data of the POSQ scale are summarized as follows:

[0077] Table 3

[0078]

[0079] Regarding self-efficacy for overcoming obstacles to engaging in healthy behaviors (OBSE), the original text of the OBSE scale developer originally contained 24 measurement data and 5 first-order constructs. After processing in this study, 14 measurement data and 3 first-order constructs remained. The processing process of the 10 measurement data is shown in Table 4. The deletion and retention classification of the original measurement data of the OBSE scale are summarized as follows:

[0080] Table 4

[0081]

[0082]

[0083] S4. Establish a structural model based on the hypothesis of the mediation model, combine the structural model and the measurement model into a mediation model, conduct covariance structure analysis on the actual data and the mediation model, and determine whether the mediation model is valid.

[0084] Preferably, according to the hypothesis of the mediation model, a structural model is established between the three variables of perceived urban green space quality, self-efficacy to overcome obstacles, and health behavior status in urban green space, and the structural model and the measurement model constitute a complete structural equation model. Specifically, the established measurement model is put into the software SPSS Amos, and according to the hypothesis of the mediation model, a structural model is established between the three variables of perceived urban green space quality, self-efficacy to overcome obstacles, and health behavior status in urban green space, and the structural model and the measurement model together constitute a complete structural equation model. Due to the particularity of the variable relationship in the hypothesis, the structural equation model is in the form of a mediation model.

[0085] In this embodiment, in the software SPSS Amos, the measurement models of the three constructs of perceived urban green space quality PUGSQ, health behavior self-efficacy OBSE and health behavior stage HB_UGS in urban green space are drawn on the same canvas at the same time, and a structural model is established between the three constructs, so that a complete structural equation model is formed, and then the latent variable path analysis and mediation effect test are performed. Each of the above measurement data formation processes is a measurement model. In the mediation model, the ellipse represents each order of constructs, the rectangle represents the measurement data, and the circle represents the error term of the endogenous variable.

[0086] like Figure 4 As shown in the figure, the schematic diagram of the mediation model structure. A complete structural equation model consists of a structural model and a measurement model. Among them, the structural model refers to the relationship between variables, while the measurement model refers to the relationship between variables and their measured data. Four variables were obtained in step S2, namely the measurement model of the four constructs in the mediation model: perceived aesthetic quality of urban green space environment (PEAQ), perceived communication quality of urban green space open space (POSQ), self-efficacy to overcome obstacles to engage in healthy behaviors (OBSE), and health behavior stage in urban green space (HB_UGS). This model expresses that the perceived quality of urban green space (perceived aesthetic quality of urban green space environment and communication quality of open space) can indirectly affect the health behavior stage in urban green space through self-efficacy to overcome obstacles.

[0087] Preferably, the software SPSS Amos 23.0 is used to perform a covariance structure analysis on both the actual data and the mediation model to obtain the difference between the actual data and the mediation model, and to determine whether the total effect between the variables, the indirect effect between the variables, and the direct effect between the variables in the mediation model exist. When the total effect between the variables, the indirect effect between the variables, and the direct effect between the variables in the mediation model exist, the mediation model is established and the mediation model is accepted; when the total effect between the variables, the indirect effect between the variables, and the direct effect between the variables in the mediation model do not exist, the mediation model is not established and the mediation model is rejected.

[0088] In this embodiment, in the Output column of the Analysis Properties analysis property panel in the Amos software, the "indirect, direct & total effects" checkbox is additionally checked. The estimation method of the above effects is particularly noteworthy. The default estimation method of Amos is the maximum likelihood method, which requires that the distribution form of the estimated parameters is a normal distribution. Then, in the Bootstrap self-sampling column of the Analysis Properties analysis property panel, select "Performbootstrap" for self-sampling, and fill in the number of self-sampling samples as 5000 []; check the two test methods "Percentile confidence intervals" and "Bias-corrected confidence intervals", and set the confidence intervals of both to 95%; check the "Bootstrap ML" self-sampling correction index. Run "Calculate estimates" in Amos to calculate the estimated values. Extract the self-sampling method calculation results related to the mediation effect in the Amos output results. The 95% confidence interval obtained by the Bias-Corrected and Percentile methods is the main indicator for testing each effect. If the lower and upper limits of the confidence interval of an effect do not contain 0, the effect is considered to be significant and exists.

[0089] For the measurement model, each first-order construct has its own measurement data, which only appears in the measurement model but not in the structural model. There are 34 measurement data in total. According to the recommendation that the ratio of sample size to the number of measurement data should be at least 10:1, if the model wants to obtain an appropriate solution, at least 340 samples are required to participate in the calculation. The actual sample size participating in this study is 439, which meets the requirements.

[0090] For the structural model, in order to verify the hypothesis of the mediation model, it is necessary to build the structural model into a mediation model: HB_UGS is the dependent variable, PUGSQ and OBSE are independent variables, among which PUGSQ is an exogenous independent variable and OBSE is a mediating variable between PUGSQ and HB_UGS; the path from PUGSQ to HB_UGS represents the direct effect of the former on the latter H1, and the path from PUGSQ through OBSE to HB_UGS is the indirect effect of PUGSQ on HB_UGS H5=H2*H3, and the total effect of PUGSQ on HB_UGS is H4=H1+H2*H3.

[0091] Each endogenous variable in the measurement model and structural model has an error term.

[0092] S5. When the mediation model is established, test the applicability of the mediation model; test the total effect, indirect effect and direct effect between variables in different groups separately to evaluate the stability of the mediation model.

[0093] S51. When the mediation model hypothesis is established, test the applicability of the mediation model

[0094] Preferably, during data processing, data samples are grouped according to different activity types of the user's final health behavior. After obtaining the grouped data, the measurement models and structural models of different groups are sequentially assumed to be equal in SPSS Amos and chi-square tests are performed to compare whether there are differences between the measurement model, structural model and actual data.

[0095] During data processing, data samples were grouped according to the different activity types of the user's final health behavior. The activity types of the user's final health behavior can be divided into low activity intensity group and medium-high activity intensity group, as well as solitary activity group and companion activity group according to the activity intensity and social status during the activity. After obtaining the grouped data, in SPSS Amos 23.0, the measurement model and structural model of different groups were assumed to be congruent (no difference) in turn. On the basis of this assumption, a chi-square test was performed to compare whether there was a difference between the model assumed to be congruent and the actual data (or the model that has been tested as congruent and is considered to be correct). If the chi-square test results show no difference, it means that the mediation model (including the measurement model and the structural model) can be used universally among the groups participating in the study and is applicable. If the mediation model assumption is not established and the mediation model is rejected, there is no need to evaluate its applicability.

[0096] In this embodiment, in the Amos interface, based on the streamlined model structure, different groups are first distinguished: two groups, "low intensity group" and "medium-high intensity group", are established, and the data divided into "walking" and "physical activity" in the questionnaire are imported into these two groups respectively. Then, inter-group comparison items are set: multiple group comparison items are set in "Multiple-Group Analysis", and the following five aspects of different group models are compared progressively:

[0097] (1) Whether the factor loadings (Measurement weights, parameter labels start with a) of the measurement model are all equal; (2) Whether the path coefficients (Structural weights, labels start with b) of the structural model are all equal; (3) Whether the variances and covariances (Structural covariances, only the variance in this result model, labels start with vvv) of the structural model are all equal; (4) Whether the structural model residuals (Structural residuals, starting with vv) are all equal; (5) Whether the measurement model errors (Measurement residuals, starting with v) are all equal.

[0098] In actual operation, if the last two items, namely errors and residuals, are required to be equal, it will be too strict, so this study mainly analyzes the identity of the first three items of different groups of models. The Bootstrap self-service sampling method was selected. After running "Calculate estimates" to calculate the estimated values, all estimated parameters of the two groups and the model diagram of unrestricted parameters were obtained. The model quality of the two groups with different activity intensities is still qualified. The chi-square difference test results between the model when the factor loadings of the two groups were set to be equal and the model when no parameters were restricted between the groups were not significant. The null hypothesis was accepted, indicating that whether the factor loadings of the two groups were set to be equal did not lead to significant differences in the models. Therefore, it can be considered that the factor loadings of the measurement model of the two groups can be equal, and the measurement model is applicable in the two groups of activity intensities at the same time. Under the premise of equal factor loadings and equal structural paths, after setting the variance or covariance of the structural model (there is no covariance in the structure of this structural model, only the construct PUGSQ has a variance) to be equal, no significant difference was found in the chi-square test (χ 2 (1) = 0.884, p = 0.347 > 0.05). This indicates that the variance of the construct PUGSQ is acceptable and equal in the two groups.

[0099] In summary, the mediation model proposed in the study is applicable to both low-intensity and medium-high-intensity groups in urban green space visiting behavior; among them, the mediating role of self-efficacy in overcoming obstacles between the perceived quality of urban green space and the stage of health behavior played a part of the explanatory power in the low-intensity group, and there was still a part of the effect that could not be explained, while in the medium-high-intensity group, it played a greater explanatory role, and the remaining unexplained effect was almost non-existent.

[0100] S52. Test the total indirect effect and direct effect of the variables in different groups to evaluate the stability of the mediation model.

[0101] Preferably, evaluating the stability of the mediation model includes: evaluating the stability of the explanatory power of the mediating variable (self-efficacy to overcome obstacles), testing the indirect effect and direct effect of the total effect between the variables in different groups respectively, judging whether the direct effect that cannot be explained by the mediating variable in each group is not significant in most groups, and testing from the side whether the indirect effect of the mediating variable plays a major explanatory role in most groups and whether the explanatory power is stable.

[0102] The mediating variable of this study is self-efficacy in overcoming obstacles. In SPSS Amos 23.0, the total effect indirect effect and direct effect between variables in different groups were tested separately. By checking whether the direct effect that cannot be explained by the mediating variable in each group is not significant in most groups, the indirect effect of the mediating variable in most groups is tested from the side to see whether the explanatory role and explanatory power of the indirect effect are stable. The specific steps are as follows:

[0103] Similar to the grouping operation of the two activity intensities, groups were created for the two social activity states "alone" and "together" divided in the questionnaire, and items for inter-group progressive comparison were set. The software was run to obtain estimated values, and a qualified unrestricted parameter model of two groups was obtained. The comparison results of the two groups showed that for the group of individual activities and the group of companion activities, there was no reason to reject the model's factor load equality, structural path coefficient equality, and variance equality of the structural model, and the model identity test of groups with different social status. The model proposed by the present invention is applicable to groups with both social statuses.

[0104] The problem of insignificant path coefficient from perceived urban green space quality to health behavior stage also appeared in the analysis of different social status groups. Combining the structural path significance test (as shown in Table 7) and the mediation effect test (as shown in Table 8) of different social status groups without parameter restrictions, it can be seen that: the total effect and mediation effect of PUGSQ on HB_UGS exist in both groups; the path from PUGSQ to HB_UGS in the solitary activity group is not significant (b1_1=0.249, t=1.645, p=0.100>0.05), and the confidence intervals obtained by the two self-service sampling methods both include 0, so it is believed that in the solitary activity group, the direct effect of the perceived quality of urban green space on the health behavior stage does not exist; this path in the companion activity group is significant (b1_2=0.256, t=1.972, p=0.049), and this piece of evidence does not seem to be used as evidence to reject the direct effect, but in the more rigorous mediation effect test method, that is, the two self-service sampling methods, the confidence intervals obtained by the test include 0, so the self-service sampling method still provides evidence to reject the direct effect of perceived urban green space on the health behavior stage.

[0105] Table 7

[0106]

[0107]

[0108] Table 8

[0109]

[0110] In summary, if the two social states of solitary activities and companion activities are used to divide the health behaviors in green spaces, the mediation model proposed in the study is still applicable to the two groups. Moreover, under this division method, self-efficacy as a mediating variable can explain most of the effects of perceived urban green space quality on the health behavior stage, and the remaining unexplained effects are almost non-existent.

[0111] The mediation model can be used to analyze the contribution of each perceived urban green space quality in the mediation model and rank the importance of each perceived urban green space quality. The application value of the mediation model is mainly reflected in the guidance of urban green space design and management strategies. In a qualified, stable and applicable mediation model, according to the path coefficient of each perceived urban green space quality, their importance in the overall relationship of the three variables can be ranked. Based on such an importance ranking, combined with relevant literature involving these qualities, suggestions can be made for the focus of design and management under the topic of urban green space serving residents' health.

[0112] Embodiment 2:

[0113] This embodiment provides a system for constructing a driving force and barrier model for healthy behavior in urban green space. The system includes a driving force and barrier model module, a mediation model hypothesis module, a measurement model module, a mediation model module, and a mediation model test and evaluation module. The specific functions of each module are as follows:

[0114] The driving force and barrier model module is used to analyze the objective and subjective barriers in the process of visiting urban green spaces and establish a driving force and barrier model for healthy behavior in urban green spaces;

[0115] The mediation model hypothesis module is used to propose a mediation model hypothesis between the perceived quality of urban green space and the health behavior stage of users in green space based on the motivation and barrier model of urban green space health behavior. The mediation model hypothesis uses self-efficacy to overcome barriers as a mediating variable;

[0116] The measurement model module is used to measure the perceived quality of urban green space, self-efficacy in overcoming obstacles, and health behavior status in urban green space and obtain corresponding measurement data, pre-process the measurement data, and establish a measurement model;

[0117] The mediation model module is used to establish a structural model based on the mediation model hypothesis, combine the structural model and the measurement model to form a mediation model, conduct covariance structure analysis on the actual data and the mediation model, and determine whether the mediation model is established;

[0118] The mediation model testing and evaluation module is used to test the applicability of the mediation model when the mediation model is established; the total effect, indirect effect and direct effect between variables in different groups are tested separately to evaluate the stability of the mediation model.

[0119] The method of establishing a structural model based on the hypothesis of the mediation model, combining the structural model and the measurement model to form a mediation model, and performing covariance structure analysis on the actual data and the mediation model to determine whether the mediation model is established includes:

[0120] The established measurement model was put into the software SPSS Amos, and a structural model was established between the three variables of perceived urban green space quality, self-efficacy in overcoming obstacles, and health behavior status in urban green space according to the mediation model hypothesis. The structural model and the measurement model together constituted a complete structural equation model;

[0121] The covariance structure analysis of the actual data and the mediation model was performed using the software SPSS Amos to obtain the differences between the actual data and the mediation model, to determine whether the total effect, indirect effect and direct effect between variables in the mediation model exist, and to determine whether there is reason to reject the mediation model hypothesis.

[0122] The method of establishing a structural model based on the hypothesis of the mediation model, combining the structural model and the measurement model to form a mediation model, and performing covariance structure analysis on the actual data and the mediation model to determine whether the mediation model is established includes:

[0123] The established measurement model was put into the software SPSS Amos, and a structural model was established between the three variables of perceived urban green space quality, self-efficacy in overcoming obstacles, and health behavior status in urban green space according to the mediation model hypothesis. The structural model and the measurement model together constituted a complete structural equation model;

[0124] The covariance structure analysis of the actual data and the mediation model was performed using the software SPSS Amos to obtain the differences between the actual data and the mediation model, to determine whether the total effect between variables, the indirect effect between variables, and the direct effect between variables in the mediation model exist, and to determine whether there is reason to reject the mediation model. When the total effect between variables, the indirect effect between variables, and the direct effect between variables in the mediation model exist, the mediation model is established; when the total effect between variables, the indirect effect between variables, and the direct effect between variables in the mediation model do not exist, the mediation model is not established.

[0125] The above embodiments are preferred implementation modes of the present invention, but the implementation modes of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, and simplifications that do not deviate from the spirit and principles of the present invention should be equivalent replacement methods and are included in the protection scope of the present invention.

Claims

1. A method for constructing a model of the drivers and barriers to healthy behavior in urban green spaces, characterized by: The method comprises the steps of: S1. Analyze the objective and subjective barriers in the process of visiting urban green spaces and establish a model of motivation and barriers to healthy behavior in urban green spaces; S2. Based on the motivation and barrier model of healthy behavior in urban green space, a mediation model hypothesis is proposed between the perceived quality of urban green space and the health behavior stage of users in green space. The mediation model hypothesis uses self-efficacy to overcome barriers as a mediating variable. The proposed mediation model hypothesis between perceived urban green space quality and users' health behavior stages in green space includes: The direct effect of perceived urban green space quality PUGSQ on the health behavior stage HB_UGS of potential users in urban green space, and the effect of perceived urban green space quality PUGSQ on self-efficacy to overcome obstacles OBSE; The effect of self-efficacy to overcome obstacles OBSE on the health behavior stage HB_UGS of potential users in urban green space, and the total effect of perceived urban green space quality PUGSQ on the health behavior stage HB_UGS of potential users in urban green space; The indirect effect of perceived urban green space quality PUGSQ on the health behavior stage HB_UGS of potential users in urban green space; Comparison of multiple groups with different activity intensities and multiple groups with different social statuses; S3, respectively measure the perceived quality of urban green space, self-efficacy in overcoming obstacles, and health behavior status in urban green space and obtain corresponding measurement data, pre-process the measurement data, and establish a measurement model; The step S3 comprises: for the unequally spaced ordered categorical variable of health behavior status in urban green space, performing optimal scale conversion on it in software SPSS Statistics; for the two relatively equidistant variables of quantity-perceived urban green space quality and variable self-efficacy in overcoming obstacles, performing exploratory factor analysis on the measured data in software SPSS Statistics, and further performing confirmatory factor analysis in software SPSS Amos; establishing a measurement model for the three variables; The establishment of the measurement model of the three variables includes: in SPSS Amos software, establishing a measurement model for the health behavior stage of the potential user of urban green space, the measurement data of the health behavior stage of the potential user of urban green space and its error term e; S4. Establish a structural model based on the hypothesis of the mediation model, combine the structural model and the measurement model into a mediation model, conduct a covariance structure analysis on the actual data and the mediation model, and determine whether the mediation model is established; The step S4 comprises: The established measurement model was put into the software SPSS Amos, and a structural model was established between the three variables of perceived urban green space quality, self-efficacy in overcoming obstacles, and health behavior status in urban green space according to the mediation model hypothesis. The structural model and the measurement model together constituted a complete structural equation model; The covariance structure analysis of the actual data and the mediation model was conducted by using the software SPSS Amos to obtain the differences between the actual data and the mediation model, and to determine whether the total effect, indirect effect and direct effect between variables in the mediation model exist. When the total effect, indirect effect and direct effect between variables in the mediation model exist, the mediation model is established; when the total effect, indirect effect and direct effect between variables in the mediation model do not exist, the mediation model is not established. S5. When the mediation model is established, test the applicability of the mediation model; test the total effect, indirect effect and direct effect of the variables in different groups respectively to evaluate the stability of the mediation model; S6. Analysis of the contribution of each perceived urban green space quality in the mediation model.

2. The method for constructing a model of motivation and barriers to healthy behavior in urban green spaces according to claim 1, characterized in that: The exploratory factor analysis of the measurement data in the software SPSS Statistics includes: Select output KMO and Bartlett test results to prove whether all measurement data are suitable for factor analysis; principal axis factorization is the factor extraction method, the analysis method is covariance matrix analysis, the number of extracted factors is determined according to the size of the characteristic root, and the factor rotation method is the optimal oblique method among the oblique rotation methods; The further confirmatory factor analysis in the software SPSS Amos includes: dealing with collinearity issues, optimizing factor loadings, correcting model fitness, and checking the intrinsic quality of the model.

3. The method for constructing a model of motivation and barriers to healthy behavior in urban green spaces according to claim 1, characterized in that: The preprocessing of the measurement data includes: measurement data setting guided by theory, questionnaire internal testing and revision, elimination of abnormal options, reliability analysis using SPSS Statistics software, optimal scale conversion using SPSS Statistics software, and reliability assignment of a single indicator using SPSSAmos software.

4. The method for constructing a model of motivation and barriers to healthy behavior in urban green spaces according to claim 1, characterized in that: The test of the suitability of the mediation model in step S5 includes: grouping the data samples according to the different activity types of the user's final health behavior during data processing, and after obtaining the grouped data, assuming that the measurement model and the structural model of different groups are equal in SPSS Amos and performing a chi-square test to compare whether there are differences between the measurement model, the structural model and the actual data; The stability of the evaluation mediation model includes: evaluating the stability of the explanatory power of the mediating variable, testing the indirect effect and direct effect of the total effect between the variables in different groups respectively, judging whether the direct effect that cannot be explained by the mediating variable in each group is not significant in most groups, and indirectly testing whether the explanatory role and explanatory power of the indirect effect of the mediating variable in most groups are stable.

5. A system for constructing a model of the drivers and barriers to healthy behavior in urban green spaces, characterized by: The system comprises: The driving force and barrier model module is used to analyze the objective and subjective barriers in the process of visiting urban green spaces and establish a driving force and barrier model for healthy behavior in urban green spaces; The mediation model hypothesis module is used to propose a mediation model hypothesis between the perceived quality of urban green space and the health behavior stage of users in green space based on the motivation and barrier model of urban green space health behavior. The mediation model hypothesis uses self-efficacy to overcome barriers as a mediating variable; The proposed mediation model hypothesis between perceived urban green space quality and users' health behavior stages in green space includes: The direct effect of perceived urban green space quality PUGSQ on the health behavior stage HB_UGS of potential users in urban green space, and the effect of perceived urban green space quality PUGSQ on self-efficacy to overcome obstacles OBSE; The effect of self-efficacy to overcome obstacles OBSE on the health behavior stage HB_UGS of potential users in urban green space, and the total effect of perceived urban green space quality PUGSQ on the health behavior stage HB_UGS of potential users in urban green space; The indirect effect of perceived urban green space quality PUGSQ on the health behavior stage HB_UGS of potential users in urban green space; Comparison of multiple groups with different activity intensities and multiple groups with different social statuses; The measurement model module is used to measure the perceived quality of urban green space, self-efficacy in overcoming obstacles, and health behavior status in urban green space and obtain corresponding measurement data, pre-process the measurement data, and establish a measurement model; For the unequally spaced ordered categorical variable of health behavior status in urban green space, the optimal scale transformation was performed in the software SPSSStatistics; for the two relatively equidistant variables of perceived urban green space quality and variable self-efficacy in overcoming obstacles, the measured data were subjected to exploratory factor analysis in the software SPSS Statistics, and further confirmed factor analysis was performed in the software SPSS Amos; the measurement model of the three variables was established; The establishment of the measurement model of the three variables includes: in SPSS Amos software, establishing a measurement model for the health behavior stage of the potential user of urban green space, the measurement data of the health behavior stage of the potential user of urban green space and its error term e; The mediation model module is used to establish a structural model based on the mediation model hypothesis, combine the structural model and the measurement model to form a mediation model, conduct covariance structure analysis on the actual data and the mediation model, and determine whether the mediation model is established; The established measurement model was put into the software SPSS Amos, and a structural model was established between the three variables of perceived urban green space quality, self-efficacy in overcoming obstacles, and health behavior status in urban green space according to the mediation model hypothesis. The structural model and the measurement model together constituted a complete structural equation model; The covariance structure analysis of the actual data and the mediation model was conducted by using the software SPSS Amos to obtain the differences between the actual data and the mediation model, and to determine whether the total effect, indirect effect and direct effect between variables in the mediation model exist. When the total effect, indirect effect and direct effect between variables in the mediation model exist, the mediation model is established; when the total effect, indirect effect and direct effect between variables in the mediation model do not exist, the mediation model is not established. The mediation model testing and evaluation module is used to test the applicability of the mediation model when the mediation model is established; the total effect, indirect effect and direct effect between variables in different groups are tested separately to evaluate the stability of the mediation model.

6. The system for constructing a model of motivation and obstacles for healthy behavior in urban green spaces according to claim 5, characterized in that: The method of establishing a structural model based on the hypothesis of the mediation model, combining the structural model and the measurement model to form a mediation model, and performing covariance structure analysis on the actual data and the mediation model to determine whether the mediation model is established includes: The established measurement model was put into the software SPSS Amos, and a structural model was established between the three variables of perceived urban green space quality, self-efficacy in overcoming obstacles, and health behavior status in urban green space according to the mediation model hypothesis. The structural model and the measurement model together constituted a complete structural equation model; The covariance structure analysis of the actual data and the mediation model was performed using the software SPSS Amos to obtain the differences between the actual data and the mediation model, and to determine whether the total effect, indirect effect, and direct effect between variables in the mediation model exist. When the total effect, indirect effect, and direct effect between variables in the mediation model exist, the mediation model is established; when the total effect, indirect effect, and direct effect between variables in the mediation model do not exist, the mediation model is not established.

Images