Abstract
This study explored the effects of community environment on older adults’ attitudes toward retirement based on a sample survey of 1,979 retired workers in a suburban district of Beijing, China. After controlling for demographic characteristics, it is found that older people’s satisfaction with the community environment has a positive and direct effect on their retirement attitudes; community participation plays a mediating role in the relationship; and the mediating effect of community participation is moderated by a higher level of self-efficacy. Our results have practical significance for improving the quality of retirement life among older adults.
Introduction
China became an aging society in 2001 when the number of people older than 65 years reached 90.62 million or 7.1% of the population, which had grown to 176.03 million or 12.57% by 2019, making an annual increase of 3.28% (NBS, 2020). It is projected that China will move to an aged society by 2027 when 14% of the population will be aged 65 years and older. 1 According to the China National Committee on Aging, the population older than 65 years will reach over 400 million by 2035, accounting for over 30% of the population. 2 Rapid population aging has witnessed a surge in the number of retirees. Between 2001 and 2019, the number of retirees drawing benefits from the Urban Employees’ Basic Pension System (UEBPS) increased from 33.81 million to 123.10 million. 3
During the planned economy period, workunits (Danwei) functioned as a “mini-society,” providing social security, social services, and housing for their employees. Both employees and retirees used to live in the same residential community, in which housing was owned by the workunit and used by employees and retirees as “dormitories,” and their basic needs could usually be met within the community. The retirees could continue to maintain close contact with and retain their identity as a member of workunit (Yuan, 2006). In addition, being a highly homogenous community often with a closed and concentrated physical space, residents were familiar with each other, and they could easily form a comfortable, stable, reliable, and convenient relationship with their neighbors (Jiang & Wang, 2017).
Along with reforms of state-owned enterprises and the workunit-based social security system in the early 1990s, the government launched a series of community-building programs aimed to rebuild communities independent of both government and the workunits. A variety of new communities with different compositions of residents had emerged under the management of the Community Residents’ Committee, a self-governing grass-roots civil organization. Many social functions previously assumed by workunits were also shifted to the Committees, with the expectation that employees and retirees would identify more with and meet their needs in the communities where they live instead of the workunits. Meanwhile, the urban housing reforms that began in the late 1990s have led to fundamental changes in both the residential space and the composition of members in a community, and many of the original residents in the Danwei community moved to live in different communities (Li, 2018; Sun, 2012). Community heterogeneity was further increased along with accelerated urbanization since the early 1990s due to large numbers of rural migrants swarming into the cities (Du & Chen, 2019). As the price of housing in the new communities often differs substantially, people of different social and economic statuses may live in the same community. For the elderly, these changes have led to a reduction in social interaction, discontinuation of their past role, and loss of social resources they once possessed (Jiang & Wang, 2017). In this context, many retirees may need to make efforts to adjust themselves to the new community in which they would spend their postwork life.
A variety of measures have been suggested by researchers to help retirees to achieve a successful retirement in a changing environment. The resource-based dynamic perspective, which is used to determine the factors influencing retirement adjustment, emphasizes that retirement-related outcomes depend on the resources available in the adaptation process (Topa & Pra, 2018; Wang & Shultz, 2009). Empirical studies have focused on a narrow set of resources, typically individual resources, as the antecedents of retirement adjustment (Topa & Pra, 2018). However, it is also necessary to explore the distal antecedents of retirement adjustment and its influencing mechanism. As the main platform for the elderly to obtain social resources, the community environment is an important distal antecedent of retirement adjustment, which also affects the retirement life of the elderly (Mutran et al., 1997).
This study explores the effects of the community environment on the retirement attitude of the elderly. Drawing on the structural equation model (SEM), the influencing path and mechanism is also attempted.
Community Environment and Attitude Toward Retirement
Retirees’ retirement attitudes reflect the thoughts, feelings, and memories of their own experience of retirement (Chan et al., 2020). A positive or negative retirement attitude reflects the outcome of retirement adjustment, which is the process that retirees adapt themselves to new conditions. Wang and Shultz (2009) contended that retirement decision or arrangements are the result of the match between retirees’ personal and environmental characteristics. According to Chai and Liu (2002), both socioeconomic factors and the physical environment of the community both a role in influencing the daily life and spatial behavior of the elderly.
In the context of aging, the community environment refers to the natural, social, cultural, and economic conditions available for the elderly in their daily life and activities in a community (Ma, 2018). Along with a decline in their physiological functioning, the residential community becomes the main platform for the elderly to obtain resources and support, carry out the daily activities, rebuild social networks, and participate in society (Diez, 2002; Glass & Balfour, 2009). Existing empirical research suggests that the quality of the community environment, both physical and social, can positively predict the mood and the health status of the elderly (Parra et al., 2010; Schwirian & Schwirian, 1993; Yan, 2013; Zhang & Zhang, 2017). On the basis of these findings, we hypothesize that community environment influences the retirement attitudes of the elderly.
Potential Mediating Role of Community Participation
Community participation refers to engagement in activities that are complex in nature, social, and nondomestic (Vaughan et al., 2016). Unlike activities occurring in the home setting, community participation emphasizes connection with the society and generally requires the higher capacity of the participants (Chang et al., 2013). The physical environment may promote or hinder residents’ community participation. Studies have found that perceived neighborhood environment was associated with social and physical functioning (Bowling & Stafford, 2007), physical activity behavior (Morris et al., 2008), and voluntary engagement at older ages (Dury et al., 2016). One study of older adults in China showed that the availability of elderly care services and activity facilities in the community affects the elderly people's community participation (Xie & Wang, 2019). Yang et al. (2012) also found that community features strongly predict the frequency of the elderly using community facilities, especially for those with mobility limitations.
Negative retirement attitudes generally arise from the sudden reduction of social interaction caused by the transition from work to retirement (Zhan et al., 2015). The classical activity and continuity theory holds that elderly people try to maintain the status, roles, and activities of their earlier life stages and replace the lost roles with similar roles, and this helps them in coping with problems in the retirement or aging adaptation processes (Hooyman & Kiyak, 2008). Another study found that retirees’ social and community participation could provide the same psychological resources obtained from employment, which is conducive to mental health after retirement (Olesen & Berry, 2011). In this light, it is reasonable to assume that the community environment affects retirees’ attitudes toward retirement through community participation.
Potential Moderating Role of Self-Efficacy
According to Bandura's reciprocal determinism, individual behavior is the result of the interaction between individual cognition and environment. Specifically, the environment stimulates an individual behavior response, which is regulated by his/her self-cognition ability (Little, 2018). Self-efficacy refers to an individual's belief in their ability to achieve a specific achievement (Bandura et al., 2003), and it is generally regarded as the sole social cognitive determinant of physical activity (McAuley et al., 2009). Research has demonstrated that self-efficacy is significantly related to social participation (Perkins et al., 2008). For example, older adults with higher self-efficacy would walk longer each week (Gallagher et al., 2012).
Researchers have also explored the possible moderating effects of self-efficacy in the relationship between environment and individual behavior. Wang et al. (2017) found that the interaction between built-up environment characteristics and self-efficacy had a significant impact on adolescents’ active transportation. Kaczynski et al. (2012) proved that self-efficacy can modulate the effect of perceived neighborhood walkability on adults’ physical activity. On the basis of the aforementioned analysis, we assume that self-efficacy may modulate the impact of the community environment on community participation.
Research Gaps
The community environment has been recognized as a distal antecedent that affects the retirement adjustments of the elderly. However, few studies have explored the path and the mechanism of the influence, especially the interaction between the community environment and the individual. Our research would fill the gaps by constructing a moderated mediation model to explore the effect of community environment on retirement attitude and examining the mediating effect of community participation and the moderating effect of self-efficacy in the relationship between community environment and community participation.
Methods
Data
The data in this study were obtained from District A, Beijing, China, in January 2018. The survey was conducted using a systematic sampling method (fixed sampling intervals K = 6), and the respondents were urban residents aged 65 years and older in this district. All interviewers received questionnaire training, and after obtaining the informed consent of the elderly, they conducted face-to-face interviews.
Participants
A total of 2,161 valid questionnaires were collected. The response rate of the survey was 94.3%. The research object of this article was the urban retired elderly, so we only kept the cases with work experience. After data cleaning, the total sample size of this study was 1,979. The research was approved by the Human Research Ethics Committee at the authors’ institute.
Measurement
Attitudes Toward Retirement
Attitudes toward retirement were measured by the 14-item semantic differential scale, which was developed by Atchley (1974). Each respondent was asked, “Which of the following 14 sets of adjectives is more relevant to your retirement life?” The score of each item ranged from 1 to 7. The total scores for all items ranged from 14 to 98 are used as indicators of attitudes toward retirement. Higher scores indicate more positive attitudes toward retirement. In this study, Cronbach's alpha of the attitudes toward retirement was 0.922.
Community Environmental Satisfaction
In this study, community environmental satisfaction was the independent variable of this study, which was evaluated by nine items including security facilities, road traffic, community greening and cleaning, recreational and exercise facilities, and provisions of recreational, medical, and elderly care services. The options include “very satisfied” (5), “quite satisfied” (4), “fair” (3), “not very satisfied” (2), and “not at all satisfied” (1). The total scores for all items range from 9 to 45.
Self-Efficacy
Self-efficacy was measured by the Chinese version of the General Self-Efficacy Scale (GSES; Schwarzer & Born, 1997), which is the most widely used self-efficacy scale at present. The Chinese version of the scale was translated by Wang et al. (2001) and has been proved to meet the requirements of psychological measurement in China. The scale is a one-dimensional scale with 10 items. Each option ranges from 0 (least consistent with your idea) to 4 (closest to your idea). The total scores range from 0 to 40, and higher scores indicate a higher sense of self-efficacy. In this study, Cronbach's alpha of the GSES was 0.941.
Community Participation
The community participation was assessed by the self-reported participation frequency of each community activity. The respondents were asked the following question: “Did you perform the following activities in the past three months (multiple choices)?” The activities included lectures, psychological classes, manuals, community patrols, environmental sanitation protection, dispute mediation, voluntary escort chats, voluntary services requiring professional skills, and so on. Options were coded as 1 for “never,” 2 for “very little,” 3 for “little,” 4 for “quite a lot,” and 5 for “a lot.” The total scores range from 8 to 40.
Sociodemographic Variables
We included sociodemographic characteristics, health status, economic condition, and social support as control variables. Sociodemographic characteristics include age, gender (male = 0 and female = 1), education (middle school or above = 0 and primary school or below = 1), and marital status (married/cohabited = 0 and single/widowed/divorced = 1). Health status includes self-rated health and cognitive ability. Self-rated health was measured by the answer to the question “How would you rate your overall health?” The options range from 1 for “very poor” to 5 for “very good.” Cognitive ability was measured by the Chinese version of the Mini-Mental State Examination (MMSE). The total scores range from 0 to 30. The Cronbach's alpha of the MMSE was 0.825. Economic condition was represented by income adequacy (roughly enough/quite abundant = 0 and very difficult/difficult = 1). Social support was measured by the Lubben Social Network Scale (LNSS-6; Lubben et al., 2006). The scale includes two dimensions, family network, and friend network, with a total of six items. Both family network and friend network were rated on a scale of 0–15, with a total score of 0–30. The Cronbach's alpha of this scale was 0.710.
Statistical Analyses
All analyses were conducted by STATA15.0 and AMOS23.0. First, we used STATA to make a descriptive analysis of the demographic characteristics of the samples. Second, we used AMOS software to conduct confirmatory factor analysis (CFA) on the hypothesis model to test the data fitness of the model and the validity of the construction of latent variables. Then, a SEM was established. We used the Bootstrap method with 500 bootstrap replications to conduct a simple slope analysis on moderating effects and to obtain conditional indirect effects at values of the moderator. The 95% confidence interval was used in the analysis.
Results
Descriptive Analyses of Respondents’ Characteristics
As presented in Table 1, the age range of the respondents was 65–97 years. Approximately half of them were women (n = 1,006), and a third were living without a spouse. Overall, the retired elderly generally had high socioeconomic status and good health conditions. In terms of the source of social support, the retirees got more support from their families than from their friends. The average GSES score of the sample was 27.88 (SD = 8.68), which was slightly lower than the international adult norm (28.6 ± 5.2SD; Schwarzer & Born, 1997). The sample had relatively high scores on community environmental satisfaction (30.96 ± 5.57SD) and attitudes toward retirement (75.81 ± 15.27SD), but the score of community participation was low (9.77 ± 3.82SD).
Demographic Characteristics (N = 1,979).
Test of Measurement Model
Before testing the SEM, we performed CFA on the whole measurement model to verify its feasibility. The fit indexes showed that the measurement model fits well with the data although the p-value was significant: χ2/df = 2.705 (<3), CFI = 0.956 (>0.90), GFI = 0.938 (>0.90), and RMSEA = 0.029 (<0.05). The standard factor load values of all latent variables were higher than 0.363, which met the requirement of the minimum cut-off value of 0.30, indicating that the convergent validity of the measurement model in this study was good (see Table 2).
Standardized Factor Loadings of Observed Variables.
Test of Structural Model
We used SEM to perform the moderated mediation analysis. Since both independent variable and moderate variable were latent variables according to the strategy suggested by Ping (1996), we constructed standardized “high to high, low to low” matching product terms of indicators. Considering the simplicity of the model, only four product terms with high loads were retained. The goodness-of-fit estimates for the moderated mediation model were χ2 = 3,325.534, p < .001; χ2/df = 2.704; CFI = 0.959; GFI = 0.940; and RMSEA = 0.029. Therefore, the test results of the structural model fitted well with the data.
Figure 1 and Table 3 show the statistically significant path coefficients for the model estimates. Community environmental satisfaction had a significant positive effect on attitudes toward retirement (β = 0.057, p < .01). Those elderly who are more satisfied with the community environment they live in will also have a more positive attitude toward retirement. In addition, the mediating effect of community participation was significant, and its mediating effect value (0.013) accounted for 12.3% of the total effect value (0.109), indicating a partial mediating effect. The results confirmed that community environmental satisfaction, to a certain extent, enhances the elderly's positive attitudes toward retirement by increasing their community participation. Moreover, the interaction between community environmental satisfaction and self-efficacy was statistically significant for community participation (β = 0.102, p < .001), indicating that self-efficacy moderates the indirect effects of community environmental satisfaction on attitudes toward retirement.
Standardized solutions for the structural model of community environmental satisfaction, self-efficacy, community participation, and attitudes toward retirement of the elderly retirees. * p < .05, ** p < .01, and *** p < .001.
Model Path Coefficient.
Note. CR = critical ratio.
Among the covariates, marital status (β = 0.061, p < .001), self-rated health (β = 0.268, p < .001), and friend support (β = 0.124, p < .001) were significantly associated with attitudes toward retirement. Older people without a spouse are more likely to view retirement negatively. Health and friend support are favorable factors for the elderly to maintain more positive attitudes toward retirement.
Further simple slope analysis (see Figure 2) showed that medium and high self-efficacy had significant moderating effects on the relationship between the community environmental satisfaction and community participation (β = 0.144, 95% CI [0.059, 0.241], p < .01), while the moderating effect of low self-efficacy was not significant. The aforementioned results show that only a higher level of self-efficacy can play a moderating effect on the relationship between community environmental satisfaction and community participation, and with the improvement of individual self-efficacy, the predictive effect of community environmental satisfaction on community participation is gradually increasing.
Simple slope of the moderating effect. Note. The coefficients in the figure were standardized.
To further estimate how the size of the indirect effect explained by the mediating variable varies with the value of the moderating variable, we estimated the indirect effects at low (M − 1SD), moderate (M), and high (M + 1SD) levels of self-efficacy, respectively (Preacher et al., 2007). Results in Table 4 show that the mediating effect of community participation was significant only in the medium and high levels of self-efficacy (β = 0.013, 95% CI [0.004, 0.029], p < .01; β = 0.025, 95% CI [0.01, 0.044], p < .01). The mediating effect of community participation on the relationship between community environmental satisfaction and retirement attitudes also shows an upward trend with the improvement of self-efficacy, that is, with the improvement of individual self-efficacy, community environmental satisfaction is more likely to promote positive retirement attitudes through inspiring community participation.
Conditional Indirect Effect of Community Environmental Satisfaction on Attitudes Toward Retirement at Values of Self-Efficacy.
Note. Nonstandardized coefficients.
* p < .05, ** p < .01, and *** p < .001.
Discussion
Our study has found that the retirement attitudes of the elderly were directly affected by the community environment. This finding is consistent with the environmental gerontology perspective, which “provided a better understanding of the interrelations between older persons and their physical-social environments, and how these relations shape outcomes in later life” (Chaudhury & Oswald, 2019). Similar findings were also obtained by Smith and Cartlidge, which revealed that “feeling comfortable, secure and attached to the community are important to the quality of life of retired seniors” (Smith & Cartlidge, 2011). Generally speaking, older people who are work oriented tend to have a stronger sense of loss caused by retirement (Ji, 2016). As empty-nest families become more and more common in China, the emotional supply of the family is gradually weakening. The community becomes the main source for retirees to find emotional sustenance, establish a life center, and regain a sense of belonging. Therefore, the elderly's subjective evaluation of the community environment reflects the performance of the community in the aforementioned functions.
The study has also demonstrated that community participation plays a mediating role in the relationship of community environment and older adults’ retirement attitudes, which was also proved by other studies that showed it to be an important indicator of retirees’ well-being (Fadila & Alam, 2016; Kim & Moen, 2002). In several other studies, older adults’ perception of the neighboring physical and social environment was found to be one of the influencing factors on their physical activity and social participation (Chaudhury et al., 2016; Moran et al., 2014). Other studies have also found that social participation is associated with older adults’ subjective well-being (Zhang & Zhang, 2015) and their adaptation to retirement (Hooyman & Kiyak, 2008; Wang & Shultz, 2009). It was realized that retirement problems do not arise from the advantages of work, but from the situation that the abundance of free time was not properly managed in later life (Moody, 2006).
Self-efficacy, especially at the higher levels, is found in our study to play a moderating role in the relationship between older people's satisfaction with the community environment and community participation. Among older adults with high self-efficacy, the more satisfactory they are with the community environment, the more frequently they would engage in community activities. This finding is also consistent with the hypothesis of the General Ecological Model of Aging proposed by Lawton and Nahemow based on the Person-Environment Fit theory, which shows adaptive outcomes are determined by the degree of match between one's competence and the environment at a given stress level (Nahemow, 2000). In other words, people with higher psychological resources are more resilient to the level of environmental pressure. The same conclusion was also reached in the study by Van Holle et al. (2015), which indicated that self-efficacy significantly moderated the relationship between community walkability and leisure walking, and the moderating effect was significant only at a high level.
Finally, the demographic variables, such as marital status, health, and social support, were also associated with retirement attitudes. They are also important individual resources that have been widely confirmed in previous studies (Bai & Liu, 2020; Fadila & Alam, 2016; Hansson et al., 2019). Our study also shows that adequate individual resources are associated with better adjustment to retirement.
The study has several limitations. First, it did not explore the situation that retirement adjustment problems may decrease over time. More studies are needed to determine whether the influence of the community environment on the retirement attitudes of the elderly changes over time. Second, we did not compare the influence of physical and social environments on the retirement attitudes of older adults. To make the intervention more targeted, further research should be done to reach a more detailed classification of community environment types. Third, our study focused on the individual's perception of the community environment. It is likely that the objective design of the environment may also play a role. Qualitative research is needed for an in-depth understanding of how older people perceive and evaluate the physical environment. Finally, in this study, community participation is used as the path in the analysis of the effect of the community environment on the retirement attitudes of the elderly. Other paths need to be explored in future studies.
The findings of the present study have practical implications for the Chinese government to carry out urban community construction and deal with the aging crisis. First, the construction of an old-age friendly community is a necessary environmental preparation to cope with the retirement boom. The government should increase the support for planning and land in the construction of elderly-friendly communities and work with the market to promote the development of elderly-friendly housing and communities. Second, the Community Residents’ Committee should provide platforms and services for the elderly to participate in society in various forms and develop activities that are compatible with their age, intelligence, physical strength, skills, and time. Finally, we should pay more attention to the cultivation and accumulation of psychological resources, such as self-efficacy, for old adults. Mental health education should be carried out in the community through regular community health services or professional psychological support activities provided by social organizations.
Conclusion
This article studied the effect of community environment on retirement attitudes of the elderly. The results show that the improvement of the community environment is conducive to the positive retirement attitude of the elderly, and community participation plays a mediating role in that relationship. Moreover, an interaction between self-efficacy and satisfaction with the community environment was observed in older adults’ community participation. The policy implication of the finding is that improvement in the community environment alone is not enough to motivate older adults’ community participation. Other measures should be used to improve their self-efficacy. These findings suggest that the development of elderly-friendly communities in China should be a major goal in combating the aging of the population.
Footnotes
Acknowledgments
We thank the Research Center for Aging of Beijing Normal University for the data provided and the local civil affairs department for their support with the survey. We also thank all participants for their positive cooperation in completing this survey.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by The Elderly Care Planning Project in Yanshan District, Beijing (grant number 03000-227900009).
