The Sense of Safety and Active Leisure in Gated Enclaves: Evidence from Fuzhou University Campus

Abstract

Scientific interest in how residential patterns affect both people’s subjective sense of safety and their behavior is increasing. The surge of gated communities in the world has changed the way we live to a great extent. Research on the gated development trend in postmodern cities is still limited; therefore, the purpose of this study was to analyze the relationship between residents’ attitudes toward gated enclaves and their sense of safety. At the same time, the relationship between a sense of security and active leisure behavior was also investigated. Using data collected from 350 college students in Fuzhou University Town, this study introduces a conceptual model to test the relationship between closed enclaves, campus security, and active leisure behavior while controlling population and community characteristics. The results of structural equation model analysis show that gated enclaves positively correlate with campus safety and positively correlate with active leisure behavior, and a safe campus positively correlates with active leisure behavior. The results of this study focus on the importance of gated enclaves as a living environment, and the discovery of functional characteristics of gated enclaves supports future interventions. In other words, when promoting active leisure behavior and increasing the sense of safety in the neighborhood environment, attention must be paid to the characteristics of these gated enclaves. In addition, the simultaneous measurement of these structures provides a dynamic observation of the existing environment, as well as information for future research and construction. Decision makers and urban planners can use these results to promote interaction and healthy behavior in the community under the multi-angle development of the existing access control, thereby improving residents’ sense of security, and increasing leisure participation.

1. Introduce

Urban evolution should focus on the development process of urban social and physical elements, such as the natural environment, infrastructure, social services, social integration, and architecture. Postmodern cities built using current practices are becoming more impenetrable than industrial cities, and urban development has become an enclosed and privatized field. Postmodern cities are facing an enclosed development trend, “challenging to shape the space, organization and institutional order of modern cities” [1,2,3]. In the past 20 years, gated communities have been widely promoted throughout the world as a mode of living. The impact of an enclosed environment on the welfare of its residents has always been an important research topic.

A gated community generally refers to a residential development project on a private road, which is closed to general traffic through a gateway across the main passage [4]. These developments may be gated by fences, walls, or other natural environments, further restricting public access. It turns out that the concept of community is often controversial (for example, [5,6]). Whether gated enclaves can be classified as communities is a question that demands further discussion. In this paper, we define gated enclaves as research objects, which can help us realize our research objectives. In modern cities, gated enclaves are defined as residential communities sharing facilities in space, and they are also a community of shared interests in the social network, whose scale and organization are varied to different degrees [4]. Gated enclaves in China were originally unit housing provided by large companies for their employees [7]. The wealthy groups that later emerged in the commercial residential areas were influenced by Western experiences and mainly through practices applied in Hong Kong. More recently, gated enclaves have become a common model in new urban residential areas. From 1991 to 2000, around 83% of residential areas in Shanghai closed their doors. During the same period, Guangdong Province, where the two developed cities of Guangzhou and Shenzhen are located, witnessed 54,000 residential areas adopting the closed-in management, which cover more than 70% of urban and rural residential areas and more than 80% of the population. In 2000, China had around 4.41 billion square meters of residential construction areas, and almost all newly-increased houses appeared in the form of gated enclaves [8]. These numbers generally reflect the extensive application of gates in China [9].

The importance of environmental safety to the urban environment has been much investigated. In a survey of the urban layout, it is verified that accessibility for elders by walking is restricted by the urban road network and environmental safety characteristics at the same time [10,11]. An examination of the residents’ sense of security is a core element of the gated enclave. Under normal circumstances, a sense of security can be defined as a need for security and certainty; its root is the fear of crime [12]. After a large number of research investigations [13,14,15,16,17], it has been found that, compared to the actual situation of security, the feeling of security may have a deeper impact on the decision making, actions, and well-being of residents [18]. It was also found that crime and fear of crime are related to crowds and other social activities [19].

Due to the interrelationship between residents’ sense of security and well-being, various types of interventions have been explored to improve their perceived security [16]. Studies of gated enclaves often point out that their origin is in response to a “fear of crime [20]”. The development of a security zone has a defensive function. In gated enclaves around the world, people use fences and walls to provide home safety. However, in the middle and late stages of this study, it was questionable whether a large number of security attributes prompted residents to deepen users’ needs for security. Research [21,22,23] has been observed that fear of crime is more common than the actual crime.

In the urbanization process of China, cultural traditions and the political environment have dramatically changed. Walls are not only a source of safety; they can keep privacy within walls. During the COVID-19 lockdowns, walls served as physical obstacles to effectively prevent the spread of COVID-19. This case suggests that walls can maintain the order both physically and socially, which is considered a source of safety. In China, the addition of defensive measures in the physical environments to resist the fear towards crimes is a widely accepted practice. People are afraid of crimes even if crimes have not happened actually. People have a strong need for security. In a survey concerning livability of city life in Fuzhou, the degree of satisfaction toward public security is just behind the degree of satisfaction of residential environment but higher than the degree of satisfaction of traffic and environmental status [24]. It is evident that the satisfaction toward security is the main factor in the urban livability survey of China. Chinese university towns are usually sited for construction at the border between urban and rural areas, and during the construction of university towns, there are disputes over demolition and relocation due to land replacement, and social security problems due to cultural discord. The security survey of the University Town after construction shows that security issues on campus, such as theft, fraud, and fighting, have been the top concerns of the University Town. Though enclosed walls and gates of the University Town have been adequate under unified construction, the large-scale enclaves have posed tremendous challenges to the campus security administration.

The application of gated enclaves on the university campuses in China is considered to have been formed in the mid-1950s and relied on the “Soviet model” university system [25]. They are spatially isolated from urban public spaces [26]. In the first decade of the 21st century, China has already set up more than 100 university towns, the majority of which are located in provinces with a dense population and its metropolises, particularly developed areas of coastal east China, such as Zhejiang, Guangdong, and Fujian [27]; Figure 1 shows site changes of Fuzhou College Town over the recent two decades. Under the protection of the concentrated use of land-related income in China, campus-enclosed fences have created a huge space. Through planning and development, universities of new urban areas are combined. The campus, built in a university town, remains enclosed. In Chinese universities, students are required to stay on campus [28]. The campus provides a large number of dormitories for them at a price much lower than the house rent for rural immigrants on the public market. The campus provides all kinds of service facilities and life resources, which can cover the necessities for basic life. According to the definition of enclaves, the university campus is an approved survey object.

Discussions about it in colleges or universities can promote the exploration of significance of enclaves towards adults, and also provide an environment for new autonomous behaviors. Research has substantiated that frequent leisure behaviors in college campuses might influence life-long physical activities and habits [29]. Leisure behavior is closely related to health research. Research on leisure behavior and various forms of personal welfare has accumulated a lot of evidence, including enhanced physical health [30] and mental health [31], stress reduction [32,33], and the improvement of life satisfaction and quality of life [34,35]. Research direction surrounding leisure activity, environmental perception, and physical environment characteristics, such as gated enclaves, have not received much attention. The relationship between leisure activities, psychological perception, and neighborhood environment is complex because it involves behavior conducted in an external environment. Research confirms that factors linked to the neighborhood environment, such as a sense of community [36], are related to aesthetics, transportation, and leisure activities [37,38]. Though previous work has proven the impact of leisure activities on personal welfare, and the correlation between leisure activities and the formation of neighboring environmental factors, how leisure behaviors are correlated with gated enclaves, and sense of safety as a dimension in the social ecological network has not yet been explored. Particularly, considering the little research attention paid to leisure activities in gated enclaves, and the research gap of its correlation with other neighboring environmental factors, this paper formed its conceptual framework.

The purpose of this research is to investigate how gated enclaves affect college students’ attitude toward security, and the relationship between security attitudes and leisure activity. Using techniques such as multiple regressions does not capture the complex interactions and correlations between various results. Therefore, this study employs structural equation modeling (SEM) technology to simultaneously analyze the prior relationship between the attitudes toward campus gated enclaves, campus safety, and leisure activity structure, with the aim of providing a richer, dynamic image of the neighborhood. From the perspective of urban sustainability, the value of the complex interaction between individuals and multiple cultural, social, and physical factors has been continuously confirmed in the literature [39]. This stems from the recognition that most health behaviors are too complex to be considered from a single-level analysis; hypothetical behavior is influenced by multiple factors at many levels [40]. A method based on the perspective of social ecology provides a wider range of intervention opportunities and studies the dynamic relationships that occur between multiple areas of interest.

The remainder of this paper is structured as follows. Section 2 discusses the theoretical background and conceptual framework. Next, the data collection and methods are described, and then the results of the study are discussed. The last section contains conclusions, a discussion of limitations, and recommendations for future work.

2. Conceptual Framework

2.1. Conceptual Model

The interaction between the living pattern and quality of life of residents creates the environmental system as a whole. Conversely, a study of environmental perception and behavior needs a comprehensive and interdisciplinary approach to obtain a thorough perspective [41]. The role of the enclave environment in the environmental system can be observed by studying how the functional characteristics of the gated enclaves influence people’s attitudes. The security perception of the campus environment is affected by circumstantial stimuli to different degrees, depending on the residents’ living and learning preferences, and these factors are also affected by social demographic characteristics. An array of demographic variables and sociocultural factors may act as barriers or facilitators with regard to participation in leisure activities [37]. The degrees of education, health status, gender, employment status, and income level all significantly influence behavioral models. This new conceptual framework assessment for the multi-level influence of the gated enclave on campus safety perception and active leisure activity.

2.2. Gated Theory

Blakely and Snyder [42] provided one of the most thorough investigations of gated enclaves. The gated property serves as a barrier, and development projects may use fences, walls, or other restrictions that further restrict public access; the property may also be surrounded by natural barriers [4]. The boundaries around the community have several functions: creating visual screens, allowing privacy, defining property, and restricting access. The authors have defined a series of physical, economic, social, and symbolic barriers. Some are penetrated easily, while some are very tall or opaque, others create personalities and identities, and others inspire fear and disgust. Some barriers are physical, while others may be psychological or symbolic [4]. Over time, the function of the fence may change. Now, at a time when gated enclaves are developing vigorously, the model of gated enclaves has gradually converged, social functions and economic functions have gradually converged, and many behavioral functions have appeared in gated environments.

However, it has been recognized that the study literature of gated enclaves in residential patterns has not been further developed after Blakely and Snyder’s typology of gated enclosures. Regarding the gated enclaves that are widely used in China today, only minimal research has been conducted. In this survey, we explored the types of phenomena related to gated enclaves that are rarely discussed [42,43].

The functions that provide security, privacy, and control are at the core of many gated communities today. In some high-end projects, armed guards patrol the grounds at all times and use closed-circuit television monitoring. Armed guards and CCTV (closed circuit television) serve a socio-economic function, and each family might also have a private alarm connected to the central security service. Although these attributes may reassure residents that their home is a shelter against the dangers of the world, they also expose the fears of community members [44,45]. Some neighborhoods have found strategies to deter tourists instead of adopting total closure; they may use speed bumps or relieve traffic pressure.

Adding signs that read “private roads” or “no through traffic” signs serves symbolic functions and makes strangers feel unwelcome. Take a university town as an example: when the road traffic enters the university area, the signs showing the entrance of the university town and other road signage are used as the boundaries when people enter. Additionally, the time or number of days the campus alternately closes the gates may hinder access. The parking lot at the entrance of the project or a road paved with non-standard materials also conveys the feeling of visitors entering a distinct space [4,42,46].

In contemporary gated enclaves, the use of personal data and the proliferation of security applications have allowed entrance card swiping, fingerprint recognition, facial recognition, and other features to be used in gated enclaves. At the same time, behavior function software screens the people that meet gated access control and regulates their behavior when they are behind the gates.

We, therefore, have learned the following: the enclosure of a gated enclave is composed of physical, socio-economic, behavioral, and symbolic functions. The physical function is defined as a physical barrier to protect the property and personal safety of the internal space. The socio-economic function is defined as the ability to control the interior and maintain internal order. The behavioral function is defined by the ability to guide the community and regulate the entry and exit of the space. The symbolic function is to control the exterior and to show the attributes of space, status, and power.

Hypothesis 1a (H1a).

Physical functions positively relate to gated enclaves.

Hypothesis 1b (H1b).

Socio-economic functions positively relate to gated enclaves.

Hypothesis 1c (H1c).

Behavioral functions positively relate to gated enclaves.

Hypothesis 1d (H1d).

Symbolic functions positively relate to gated enclaves.

2.3. Campus Sfety

Research on gated enclaves often points to their origin as a response to “criminal fear” [17]. However, research regarding the performance of enclosed environments in terms of security has yielded complicated findings. On the one hand, studies have shown that gates help to improve residents’ sense of security and actual safety. Newman [47] pointed out people’s perceptions of greater safety in gated and fenced areas. For example, gated enclaves with physical barriers to entry exhibited lower levels of burglary. There is likewise evidence that residents believe that a gated enclave is a safe place to live [48].

Results from other studies showed that there is a questionable relationship between barriers and security and even opposition to the above assertions. For example, it has been hypothesized that living in a gated community actually reduces the risk of people becoming victims [49]. Other studies, examining the degree of risk perception [50], state that environmental factors have the least impact. However, gated elements in gated communities can cause these communities to lose social diversity, leading to a trend of social isolation [51].

Feeling safe in school effectively promotes students’ learning and healthy development [52]. Feeling safe socially, emotionally, intellectually, and physically is a basic human need [53]. Campus safety not only refers to the physical safety of students but also requires emotional safety and internal order in the campus environment. It is more practical to measure the sense of security felt by students on campus. Campus safety is an integral part of the campus atmosphere. However, a large number of studies have shown that many students do not feel physically and emotionally safe in school, which affects the quality of their education and can have a long-term negative impact on the future of students. The results of these studies suggested that this is mainly due to the influence of the contextual variables that define the school atmosphere and the interpersonal atmosphere [54]. Under this premise, we need to examine the impact of campus security in a gated enclave.

Hypothesis 2 (H2).

Gated enclaves are positively perceived as relating to campus safety.

2.4. Active Leisure

Leisure is defined as an activity of free choice, which is meaningful and intrinsically motivated and produces pleasant experiences. The classification of leisure activities is not consistent in the literature [55]. Active leisure is referred to as a type of physical activity and social activity that involves some form of physical movement and social interaction during free time, which can include many different activities (for example, exercise, cycling, swimming, gardening, walking) [56]. Activities such as reading or social gatherings can also be classified as leisure activities [57].

The traditional view is that in addition to formal study and career preparation, students’ college life is a compelling time for activities. Leisure time activities influence people’s development [58] and happiness [59]. Compared to other adults, college students are likely to participate in a lot of leisure activities. Estimates from the U.S. Time Use Survey indicated that, on an average working day, full-time college students spend three to four hours in leisure and sports activities. This report further broke down the leisure pursuits of all people aged 15 and above, including social activities, sports activities, games, and reading activities. In previous surveys, campus leisure activities were an important indicator of college students’ life satisfaction, but we know very little about how they relate to the physical environment and psychological perception on campus, and this issue is worthy of attention.

Hypothesis 3 (H3).

Campus safety positively relates to active leisure.

Hypothesis 4 (H4).

Gated enclaves positively relate to active leisure.

As shown in Figure 2, the research model of this study is presented below.

3. Materials and Methods

3.1. Research Instrument

The questionnaire was divided into two parts. Respondents in the first part were asked to answer questions about their gender, school, location of accommodation by using choice questions. In addition, participants were asked to answer questions related to their age and residence period by using gap filling. In the second part, participants were asked about their attitudes regarding gated enclaves and campus safety, and the frequency of leisure behavior within the campus. To ensure content validity, these projects were developed on the basis of previous research. To verify the questionnaire, a pilot research was conducted among respondents from university towns. Though sentence structures of the final questionnaire are subtly adjusted, the overall effects of the pilot research helped establish the reliability and validity of the questionnaire for data collection. In order to verify the questionnaire, 350 respondents from university towns were involved.

To measure the attitude towards gated enclaves, respondents were asked to finish the seven-point Likert scale involving 12 questions. The responses ranged from strongly disagree to strongly agree. The scale examined four functional dimensions of enclosed enclaves, and the questions were adapted from definitions of Jill and Lindsey [4] about functions of enclosure in their research of enclosed enclaves. On the first dimension, respondents were asked about whether they agreed with the following attributes of physical functions of the enclave, including to block the influence of external events, to prevent dangerous traffic incidents, and to repel harmful people. Additionally, there are three items to measure the socio-economic functions, including the provision of resident assistance, the direction of resident behavior, and the maintenance of social order. The behavioral measures have gradually been playing a dominating enclosing role in the gated environment over the past ten years. The gated environment aims to direct people’s behavior, to standardize accessibility, and to give specific requirements for access. Finally, the following three items belong to symbolic functions: to identify the ownership of the property, to indicate the right to use its resources, and to guide correct behavior inside the enclave.

To analyze people’s attitude towards campus safety is different from the measurement of residents’ perception of safety. The former is based on a research of the campus atmosphere. Based on the research update of [60], the influencing factors of the campus security were composed of the following three aspects: emotional safety, behavioral safety, and regulatory measures. The following 7 questions were posed in the questionnaire: I am proud of this school; I feel a sense of identity in school; There is violence and conflict in the school (reverse); I do not want to leave school; I have a sense of belonging to the school; You feel safe when walking in the school at night; You feel safe when wandering around the campus during the day. These items were measured using a 7-point Likert scale, ranging from strongly agree to strongly disagree.

With regard to the active leisure activities performed on campus, the measure by which participants were asked to rate the frequency of leisure activities they usually perform on campus runs from 1 (never) to 7 (almost every day). This measurement method has found wide applications in survey of leisure activities [61]. For example, what is the frequency of your participation in the following leisure activities over the past one week, including attending a party, walking or jogging, and admiring the scenery. These indexes are not highly correlated, so they are dealt with as formative indexes in the structural equation [62].

3.2. Analytical Method

In order to assess the complex hierarchical model, we used the IBM SPSS 25.0 version 3.2.9 [63] to analyze the data collected. The structural equation model (SEM) is an iterative estimation approach that combines the PCA and the multiple regression. The SEM can be used to directly measure the variable observed, while the latent variable can be inferred from the variable observed. The measurement model and the structural model form the SEM together. The correlation between the latent variables is indicated in the structural model. The CB-SEM and the PLS-SEM are two forms of SEM. In this research, PLS-SEM outperformed CB-SEM in that the former could better cope with the non-normal data set and use the theoretical model which was not yet improved for an exploratory research [54,55]. Additionally, PLS-SEM can cope with two measurement models (reflective model and formative model) involved in the model put forward in this research. In PLS-SEM, composite reliability and average variable (AVE) loaded and extracted from the measurement item of the corresponding structure were used to measure the validity of the measurement model. The internal consistency is measured by the compound reliability. The difference between the given constructs and other constructs is defined as the discrimination validity [58]. In order to acquire adequate discrimination validity, the AVE square root of every structure should be higher than the structural relevance. Besides, the loading of measurement items on respective structure should exceed the cross loading [57]. At last, the collinearity (internal variance inflation factor), significance and relevance of the structural model relation, and coefficient of determination (R2) are used to assess the validity of the structural model.

3.3. Data Collection

The data used in this study came from a field survey conducted in Fuzhou, China in January 2021. Fuzhou is located in eastern Fujian in the east region of China, near the lower reaches of the Minjiang River and coastal areas. More than 8 million people live there. This study selected Chinese university campuses as the survey object of the gated enclave. Fuzhou University City is a high-tech development zone integrating education, culture, ecology, and life. As planned, the industrial park is going to hold a population of around 251,000, of which the residential area is planned to accommodate a population of 74,000 and the area for colleges and universities will carry around 177,000 students. The University Town discussed in this paper started construction since 2000. As planned, 12 universities of higher learning are located there. The boundary between these campuses is a typical enclosed gate. Every non-private region of the campus is equipped with the CCTV and adequate patrolling security guards. The campus also has simple material supply that is necessary for daily life to satisfy basic living needs.

The target research objects are college students from Fuzhou University Town. The Research samples are college students from nine gated college campuses presented in Figure 3, and these research samples are collected by the web-based data collection tool. Deciding on an appropriate sample size is critical to ensure the quality and accuracy of any research. For this reason, Hair recommended using the sample rule proposed by Barclay et al. [64] to determine the minimum sample size in PLS-SEM analysis. This rule stipulates that the minimum number of samples should be greater than “10 times the maximum number of structural paths to a specific component in the structural model”. The structural model of this study contains 7 structural paths. According to this 10-fold rule standard, our minimum sample size should be no less than 70 respondents. In order to improve the response rate, participants received 5–8 RMB rewards when participating in this study. Due to budget constraints, the number of participants in the first part was limited to 350 people.

4. Results

4.1. Demographics

As part of the preparation for data analysis, a thorough screening process was carried out. The data were tested for normality, outliers, and demographic characteristics in Statistical error tests using SPSS. Missing values were processed with the replacement method, which is widely recommended in studies [62]. This option is a built-in function of SmartPLS, which keeps our sample size and mean constant.

For data analysis and a discussion of the research results, the study must first briefly describe the demographic characteristics of the interviewees such as age and gender; their residential location was also part of the questionnaire. Of the 350 respondents, 35.1% were male and 64.9% were female. Most respondents (more than 87%) were between 18 and 22 years old. The vast majority of respondents (just under 94%) live in dormitories provided by the school; only 5% of the population rented off-campus, and 1% of the surveyed college students lived in their own homes.

4.2. Econometric Model Analysis

The conceptual model of this study includes two measurement models: the formative measurement model and the reflective measurement model. Among the three second-order variables, one variable (active leisure) has a formative measurement model, and two variables (campus security and gated enclave) have a reflective measurement model. The four first-order variables (physical function, socio-economic function, behavioral function, and symbolic function) belong to the reflective measurement model. The statistical evaluation criteria of the formative measurement model are different from the reflective measurement model [62]. In the formative measurement model, the concept of internal consistency is inappropriate [65], because the items on the formative measurement scale may represent an independent cause and do not necessarily highly correlate with each other [62]. For the purpose of this study, we evaluated the reflectivity and formative measurement models separately. The structural reliability and validity of all the reflex measurement models were analyzed, and the convergence and validity of formative measurement models (in this case, active leisure) were analyzed according to the guidelines of Hair Jr., Hult, Ringle, and Sarstedt [62].

4.3. Reflective Measurement Model Analysis

The structures of the first-order and second-order reflection measurement models were analyzed separately. The reliability and validity of all the models were evaluated according to the guidelines of Hair Jr. et al. [62] and Henseler et al. [66]. The results showed that almost all the structures have a fairly acceptable factor loading value between 0.70 and 0.90. In addition, as an exploratory study applicable to PLS, Cronbach’s α value of the physical function and socioeconomic function in the first-order model is at the critical level near 0.70 as proposed by Cohen [67]. The comprehensive reliability (CR) value of all the structures is higher than 0.70. In addition, this study also calculated the square root of the average variance extraction (AVE), which exceeded the correlation between this structure and other structures in the model to ensure the validity of the discrimination. The AVE values of all the structures are also higher than the critical value of 0.50 proposed by Hair Jr., Hult, Ringle, and Sarstedt [62]. The complete results of the validity and reliability of all the structures are shown in

Table 1. Validity and reliability of the latent structures.

First Order	Loadings	CR	AVE	Second Order	CR	AVE
Physical functions	0.67	0.819	0.602	gated enclaves	0.847	0.582
Socio-economic functions	0.688	0.828	0.617
Behavioral functions	0.701	0.834	0.627
Symbolic functions	0.798	0.882	0.713
				campus safety	0.874	0.505

.

The Fornell–Larcker criterion, as shown in

Table 2. Discrimination validity (Fornell–Larcker criterion). The Vnal-Lake Code.

	Campus Safety	Physical Functions	Socio-Economic Functions	Behavioral Functions	Symbolic Functions	Gated Enclaves
Campus safety	0.711
Physical functions	0.404	0.776
Socio-economic functions	0.546	0.477	0.785
Behavioral functions	0.404	0.448	0.525	0.792
Symbolic functions	0.419	0.324	0.448	0.445	0.844
Gated enclaves	0.584	0.709	0.812	0.786	0.746	0.61

The values in bold are the square root of the mean.

, was used to evaluate the discriminant validity. The bold values in Table 2 show the square root of AVE. The square root of the AVE of each structure is greater than its correlation with other structures. This proves the discriminative validity of the structure involved in the proposed measurement model [62,68].

By evaluating all the cross-load values of the reflection structure index, another test was carried out on the discriminative validity of the reflection measurement model. According to the survey results in

Table 3. Cross loadings among reflective measurement scale items.

	Active Leisure	Campus Safety	Physical Functions	Socio-Economic Functions	Behavioral Functions	Symbolic Functions	Gated Enclaves
I am proud of this school	0.367	0.82	0.29	0.457	0.291	0.308	0.445
I feel a sense of identity in school	0.404	0.858	0.357	0.461	0.334	0.373	0.503
There is violence and conflict in the school (reverse)	0.238	0.566	0.205	0.323	0.202	0.24	0.32
I do not want to leave school	0.283	0.647	0.194	0.33	0.214	0.208	0.313
I have a sense of belonging to the school	0.358	0.854	0.301	0.458	0.293	0.328	0.457
You feel safe when walking in the school at night	0.261	0.559	0.351	0.313	0.294	0.274	0.401
You feel safe when wandering around the campus during the day	0.187	0.588	0.286	0.33	0.366	0.325	0.43
To identify the ownership of the property	0.203	0.387	0.287	0.399	0.441	0.834	0.657
To guide correct behavior inside the enclave	0.28	0.352	0.278	0.385	0.366	0.827	0.625
To indicate the right to use its resources	0.271	0.319	0.254	0.348	0.314	0.871	0.603
To provide resident assistance	0.25	0.459	0.37	0.831	0.421	0.4	0.672
To maintain social order	0.278	0.481	0.354	0.754	0.456	0.3	0.621
To direct resident behavior	0.2	0.345	0.403	0.769	0.361	0.352	0.618
To standardize accessibility	0.173	0.352	0.366	0.427	0.833	0.386	0.656
To give specific requirements for access	0.153	0.332	0.345	0.419	0.762	0.329	0.604
To direct people’s behavior	0.171	0.272	0.353	0.402	0.779	0.339	0.606
To block the influence of external events	0.173	0.265	0.749	0.326	0.289	0.246	0.504
To prevent dangerous traffic incidents	0.225	0.346	0.808	0.408	0.402	0.28	0.6
To repel harmful people	0.194	0.326	0.77	0.372	0.344	0.227	0.54
Attending a party	0.685	0.303	0.159	0.2	0.123	0.206	0.227
Walking or jogging	0.823	0.357	0.214	0.219	0.201	0.24	0.287
Admire the scenery	0.682	0.282	0.19	0.292	0.12	0.207	0.269

Bold values are loadings for items, which are above the recommended value of 0.5.

, all the indicators (the measurement scale items) of the reflection measurement model have a higher load on their respective potential structures than any other structure involved in the model [62,68]. Therefore, these findings meet the cross-load evaluation criteria and provide satisfactory evidence for the discriminative validity of the reflection measurement model.

4.4. Formative Measurement Model Analysis

The evaluation process of formative constructs is different from the evaluation criteria of reflective constructs. All the formative measurement models may represent independent causes of the underlying structure. Therefore, formative indicators do not have a high degree of correlation among the measurement scale items [62,69]. As mentioned in the previous section, this research involves a formative measurement model, represented by active leisure.

The external weight (also called relative importance) of formative indicators was evaluated to determine the relative importance of the indicator to its underlying structure. Given Hair Jr. et al. [62] and Henseler et al. [70], these external weight values were evaluated as being significant. It can be seen that the three measurement indicators of active leisure are all significant at the 0.001 level.

The survey results in

Table 4. Outer weights of items involved in formative constructs.

	Outer Weights	T Statistics (|O/STDEV|)	p Values	VIF
Walking or jogging → active leisure	0.580	4.940	0.000	1.226
Attending a party → active leisure	0.445	4.004	0.000	1.118
Admiring the scenery → active leisure	0.320	2.561	0.010	1.250

showed that the various indicators of the formative measurement model have reached the standards for determining their relevance and significance. On the basis of the above discussion, the suitability of the formative structure has been established, the reflective measurement model and formative measurement of the model were evaluated overall, and acceptable results were obtained.

4.5. Structural Model Analysis

This study used SEM to test the hypothetical relationship between the seven variables (

Table 5. Hypothesis assessment.

Associations	Latent Constructs	Β-Value	t-Value	p-Value	Decision
Second order dimensions	Gated enclaves → Campus safety	0.584	13.793	0.000	Supported
	Campus safety → Active leisure	0.342	5.682	0.000	Supported
	Gated enclaves → Active leisure	0.154	2.451	0.014	Supported
First order to second order dimensions	Physical functions → Gated enclaves	0.284	14.52	0.000	Supported
	Socio-economic functions → Gated enclaves	0.35	20.703	0.000	Supported
	Behavioral functions → Gated enclaves	0.316	16.464	0.000	Supported
	Symbolic functions → Gated enclaves	0.356	15.779	0.000	Supported

). The R2 value was used to evaluate the overall explanatory power of the structural model, and the Q2 value and the path coefficient β value were used to evaluate the predictive correlation.

These results showed that physical function (β = 0.284, p < 0.001), socioeconomic function (β = 0.35, p < 0.001), behavior function (β = 0.316, p < 0.001), and symbolic function (β = 0.356, p < 0.001) all have a significant impact on the perceptions of gated enclaves and thus support H1a to H1d (Figure 4). In addition, the study found that the relationship between gate control and campus safety (β = 0.584; t-value = 13.793; p = 0.000) is positive and significant, providing support for H1. These results showed that the proposed model has 34.2% explanatory power for campus safety with R2 = 0.342. Similarly, H2 is also supported; this is the relationship between campus safety and active leisure (β = 0.342; t-value = 5.682; p = 0.000). Additionally, the hypothesized relationship between gate control and active leisure (β = 0.154; t-value = 2.451; p = 0.000) is very important, supporting H3. These variables explain the 20.2% difference in active leisure.

According to research in [62,71], it is not sufficient to support the model based on the R2 value alone. Therefore, the Stone–Geisser [72] Q2 test was used to evaluate the predictive correlation of structural models. According to the standard rule, if the Q2 value is greater than zero, then the potential exogenous structure involved in the structural model has a predictive correlation with the potential endogenous structure [62,69]. The Q2 values of the model are 0.368 for gate control, 0.164 for campus safety, and 0.102 for active leisure; this supports the basic hypothesis of this research—that is, the endogenous structure involved in this research has predictive relevance to the conceptual model.

5. Discussion

The main goal of this research was to consider the relationship between residents’ perception of and behavior in gated enclaves through empirical research. Previous studies did not consider the function of gated enclaves from a typological perspective or their role in the environment, and research on the multilayered effects of active leisure in this type of environment is still limited. The first-order model of this research was a confirmatory study on the typology of gated enclaves. A homogeneous sample was selected so as to verify the attitude toward gated enclaves. In addition, previous research on gated enclaves was mainly based on homogeneous samples.

Active leisure was measured according to the respondents’ frequency of campus activity (for example, walking, jogging, socializing, etc.), while other behavioral studies have not taken this into consideration. In addition, this research contributes to the existing theory by simultaneously analyzing all the conventional relationships between campus safety and active leisure in a single model, controlling individuals, residential locations, and length of residence, and using PLS structural equation analysis. Research that combines subjective and objective measurements is still limited, but it is very important for a better understanding of people’s safety perceptions and active leisure in gated enclaves [73].

5.1. Gated Enclaves and First-Order Dimensions (H1)

In the first-order model of this study, the typological attitude toward gated enclaves was verified. The results of this research showed that the greater the symbolic function (such as slogans), the greater the role of the gated enclaves. These findings are consistent with previous research in this field. For example, Ali et al. [74] and Nadiri et al. [75] also observed that the quality of symbolic function has a great influence on gated enclaves.

This study also showed that the behavioral function has a greater predictive ability than the socio-economic function. Under the general trend of the data environment, new gated enclaves’ behavior functions gradually occupy a dominant position in the perception of gated enclaves. However, it is worth mentioning that our research results showed that compared with the above three functions, the physical function has the lowest influence among the residents of the gated enclave. This evidence supports previous research about the complex influence of physical function on mental perception. Within the typological network of gated enclaves, physical functions are significantly related but do not play a dominant role.

What is worth discussing is that physical functions and socio-economic functions that are realized through gates, walls, and CCTV are more effective in preventing actual criminal behaviors. This research observed that, compared with the actual capability against a criminal defense, people’s degree of recognition of gated enclaves is subject to more complex factors. Symbolic functions are reflected as a deterrence of symbols and behavioral functions as an identity that discriminates one resident from the other as well as the measures adopted by these two types of functions for the slight transformation of the physical environment. Comparatively, symbolic functions of gated enclaves are more recognized by users.

5.2. Gated Enclaves and Campus Security (H2)

The awareness of gate control has significantly improved the understanding of security and verified an implicit assumption held by the New Urbanism movement [76]. These findings help to further clarify this relationship. The results of this study further confirmed that the gated enclave has a significant correlation to the sense of security, and the sense of security is further developed in such an environment.

Jane Clark Linder’s study [77] further explains that the idea of schooling is often benevolent, enticing parents and guardians to trust in an imaginary school state that offers more safety measures than other institutions. Reactions to safety failures have generated school policies and rules, and these policies and rules may exacerbate fear in the community. Based on this study, it can be further confirmed that whether enclaves may increase a population’s need for a sense of security, and more comprehensive evidence can be provided for the theory of campus safety.

5.3. Campus Safety and Active Leisure (H3)

The neighborhood is the main location of the most common forms of active leisure [78]. The main barriers to leisure participation are generally considered to be a lack of support, safety, and trust [73,79]. In this study, participation in active leisure was significantly associated with the perception of campus safety, denoting that the psychological need for the perception of safety exceeds personal rewards, contributes to the promotion of healthy behaviors, and encourages social interaction and contact [80,81]. In this study, there is a correlation between the sense of security and physical environment, and the sense of security also has a positive influence on positive behaviors. Sense of security has been a widely-discussed issue and a basic demand of social behaviors. Communities can be one of the largest sources of sustainable social and emotional support [82]. Studies have shown that active communities that promote walking can cultivate residents’ sense of security and prevent illegal activities [80,81].

5.4. Gated Enclaves and Active Leisure (H4)

The results showed that there is a significant relationship between gated enclaves and leisure participation, which is consistent with existing research focusing on the health effects of artificial environments. Many studies have examined these communities and artificial environments as they pertain to active leisure and found that leisure activities in gated enclaves include walking [83], exercise [84,85], and going to parks.

Though some researchers have investigated the similar correlation such as between the community entertainment and the community interaction, they should emphasize the specific space for active leisure activities [86,87,88]. A natural experiment carried out in the campus of Nanjing University(Nanjing, China), China provided solid evidence for the dose-response relationship between jogging and physical health [89]. It also suggested the importance of the enclosed enclave as an environment towards active leisure activities. In the follow-up research, relevant standards can be formulated to discuss how to improve the environment and how to promote and strengthen the health-promoting behaviors and feelings.

The evidence in the literature is mixed. Some studies have found that the environment is a powerful predictor of social interaction [36,90,91], while other studies have found that, although the physical properties of an environment may promote social interaction [92], they may have nothing to do with raising social consciousness. The conclusion of this study demonstrated that the physical environment does have an impact on active leisure, but the correlation between a sense of security and behavior is even stronger.

6. Implications and Conclusions

6.1. Implications

As a widely-adopted approach for urban planning, gated enclaves combine multiple means to improve a neighborhood for the purpose of crime prevention and alleviation of the fear of crimes. This strategic approach has found wide applications in residential communities and regulated enclaves.

The main purpose of this research was to analyze the relationship between the attributes and functions of the gated enclave on campus safety and active leisure. SEM shows that physical functions, socio-economic functions, behavioral functions, and symbolic functions can exert a significant impact on the gated enclave. This finding shows good agreement with previous research results. Research has also substantiated a significant correlation between the gated enclave and campus safety and active leisure. This confirms the importance of the physical environment as it pertains to people’s health behaviors and satisfaction in life and well-being.

Prior research findings mostly believed in a complicated relationship between the fear of crime and anti-crime measures and gated enclaves. The purpose of discussing gated enclaves is to enhance the resistance against crimes and reduce any fear of them. However, findings showed that if the external image is enhanced to reduce crimes, then it cannot enhance the public recognition of the gated enclave’s capability to keep people safe. Research held that the fear of crimes might not result from one or two singular attributes, but from multiple attributes and their interaction. Therefore, not only should attention be paid to attributes over the fear of crime, but the interaction among these attributes should also be noted. Research into measures taken by gated enclaves should be further detailed, with different consequences of measures identified, and a balance pursued for the correlation between enhancing the capability to resist crime and reducing the fear of crime.

From the perspective of a city, resolving security concerns should rely not just on the promotion of environmental security via the external gated enclave environment. A potential new research perspective is to examine these security concerns under the background of a socio-ecological network. Studies on cities in the post-industrialization era are no longer limited to barriers facing the urban environment transformation and challenges for urban dwellers and policymakers. They have been extended to various socio-ecological problems, such as economic decline, social segregation, cultural conflict, and declining quality of life. This study provides feasible research perspectives for the research topics in this field. Research findings can be promoted in more extensive urban fields.

To sum up, this research can contribute positive implications to measures taken by gated enclaves, physical support, planning of a positive life, and policy issues. Previous studies have shown that the quality of gated enclosures (including accessibility, safety, size, aesthetic appearance, and ease of social interaction) will indirectly affect walking frequency. Therefore, it is of great significance to researchers and practitioners in the field of health promotion and community development to improve gated enclaves to increase residents’ sense of security and leisure participation.

6.2. Shortcomings and Future Research

One of the main advantages of this research is to use the SEM method to simultaneously study the relationship between a gated enclave, campus safety, and the underlying structure of active leisure in a single model. First-generation analysis techniques, such as regressions, require separate or hierarchical analysis, and these analyses may not be able to capture all the variable associations from common causes. Even so, this study has limitations. The study on limitations of design factors on the effectiveness and generalization of the model, and the calculation of the potential for specification of errors implicit in computing parameter. In addition, as this survey collected cross-sectional data, it was impossible to test the longitudinal path and draw causal conclusions. The community attitude and behavior mode discussed in this study cannot be deemed as the common attitude and behavior of communities.

The active leisure participation data were collected by asking respondents to identify the activities they participated in during the week. Although the use of this form is a common method of obtaining information on active leisure [93], attention should be paid to the possibility of inaccurate recall and prejudices [94], which may influence the inference of correct conclusion about leisure activities in the study results.

Caution should further be exercised when comparing the results of this study. This research paid special attention to university campuses. The method to collect data in this study is to fill in the questionnaires answering the attitude towards personal perceptions and the number of leisure participations, but this may be flawed. When only one measurement type is used, it poses a threat to the effectiveness of the structure [95], because a single method deviation can cause an amplified correlation between the two structures.

Future research can study the characteristics of urban public space (such as air quality, aesthetic quality, atmosphere, smell, accessibility, adequate parking spaces, distance from facilities, traffic safety, natural elements, noise level, cleanliness, and space maintenance) in greater detail. Other studies have found that participating in social activities and leisure activities in the community can also increase residents’ sense of security and trust [96]. Therefore, further exploration of the promotion of active leisure in gated communities is of great significance for community-building and overcoming traditional barriers to participation.

Through the development of the attributed typology of gated enclaves, the future development of gated communities can be discussed. While promoting a sense of security, even symbolically, gated enclaves can reduce the restrictions of physical functions on community behavior and promote the coordinated development of the community’s physical and social environments.