Research on SWB has used various orientations and measurement indicators. In terms of orientation, the marginal utility theory founded by Jevons initiated the research from an economic perspective (Sun, 2004). Since then, other scholars have conducted research from a psychological perspective, focusing more on individuals’ SWB, which comprises quality of life, mental health, and psychological development (Xing, 2002).

In terms of indicators for measuring SWB, psychological development indicators have been added beyond the initial indicators of quality of life and mental health (Li, 2006), including strength, sense of autonomy, pleasure, self-confidence, and others. Self-report scales and timely measurement methods have gradually been developed (Yu, 2018). The former mainly includes the Satisfaction with Life Scale, the General Well-Being Schedule, the Memorial University of Newfoundland Scale of Happiness (MUNSH), and the Multidimensional Psychological Well-being Scale. The latter mainly includes the day reconstruction method, ecological momentary assessment, and experience sampling method. Among these, the Philadelphia Geriatric Center Morale Scale and the MUNSH have most commonly been used to measure older adults’ SWB.

Research on SWB in China began in the mid-1980s, and the findings mainly involved economics, psychology, and sociology. The focus has since shifted from qualitative description to quantitative measurement. In the early research stage, domestic scholars mostly borrowed research ideas and measurement tools from foreign scholars; for example, Yang et al. (1988) used the MUNSH to study the happiness of older adults. With the depth of domestic research, the development of localized measurement tools has received increasing attention and many scholars have revised international SWB scales accordingly based on practical needs.

Xing and Liu (2008) proposed the idea of experiential SWB with both form and content, and his Subjective Well-Being Scale for Chinese Citizens (SWBS-CC) has been well accepted in China. The scale contains 10 dimensions of interpersonal adaptation experience, mental health experience, family atmosphere experience, and so on, which reveal the SWB of Chinese citizens (including older adults) in a more comprehensive way. In this study, the SWBS-CC was adjusted based on actual survey data to measure the SWB of older learners in three dimensions: adaptation satisfaction experience, physical and mental health experience, and self-development experience (Sun and Liu, 2020).

Studies have reached different conclusions regarding the relationship between learning activities and SWB. Some scholars believe that there is a positive correlation between learning and SWB; the higher the level of education, the higher the individual’s SWB (Yue et al., 2021). Researchers have noted that education level is an important factor influencing the overall life satisfaction of older adults (Cheng and Yan, 2021). Older adults’ motivation to participate in lifelong education plays an important role in learning satisfaction, which has a positive impact on quality of life (Kim and Lim, 2014; Hu, 2019), and teachers and fellow students have the greatest influence on the quality of life of older learners (Escuder-Mollon et al., 2014). Meanwhile, participation in learning activities can promote quality of life in older adults, effectively resolve their inner conflicts, and help improve their SWB (Bolton-Lewis and Li, 2017; Sun and Liu, 2020; Xiang and Li, 2020; Feng and Liu, 2022). However, some scholars have also found that the correlation between participation in learning activities and SWB is not necessarily significant or can even be negative (Powdthavee et al., 2015) because the premise that adults’ participation in lifelong learning activities positively affects their mental health and quality of life is consistent with their learning interests (Hammond, 2004).

At present, it is common to study SWB in entire populations, and few studies have specifically examined older adult learners. Some scholars believe that there is a significant difference between the SWB of urban and rural residents, and urban residents are happier than rural residents (Zhou, 2015). In contrast, Luo (2006) came to the opposite conclusion by using explicit variables to measure happiness. Su and Zhang (2021) concluded that rural residents’ social class and income had a significant effect on their SWB; however, although older adults with agricultural household registration were generally less satisfied with their lives than those with non-agricultural household registration, they scored higher in positive affect, positive experience, and overall happiness in rural areas than in urban areas (Jiang and Lin, 2008). Besides, the expansion of unequal opportunities will reduce rural adults’ SWB (Yu and Zhang, 2019; Zhang and Wan, 2020).

Most research on the factors influencing the SWB of older adult learners has involved demographic and psychological variables, such as age, gender, education, aging attitudes, and self-perceptions. Research from the perspective of learning participation has been conducted in terms of learning motivation, learning needs, learning intentions, and learning opportunities. Some researchers have identified the factors in terms of their engagement in learning activities, learning atmosphere, and learning experience (Sun and Liu, 2020; Sun and Zhang, 2021). Fan (2009) pointed out that older adults’ participation in learning activities includes six factors that can greatly enhance their SWB: learning atmosphere, learning interest, learning initiative, learning difficulties and challenges, and learning commitment, and participation (Bo, 2017). Several studies found that the creation of “learning circles” and “social circles” can realize the self-worth of older participants and promote interpersonal communication, thus affecting the quality of older adults’ happiness (Le and Xia, 2016; Gu and Wang, 2020).

Previous studies on older adult learning and SWB have mainly focused on the study of education level’s influence of on older adult learners’ SWB. Few scholars have explored the difference in learning participation’s influence on urban and rural older adult learners’ SWB (Xiang and Yao, 2016), which is the concern of this study. Based on the above research, the current study sought to investigate whether there are differences in older adult learning’s influence on urban and rural older adults’ SWB and the factors contributing to that difference. The specific hypotheses were as follows:

The modified SWBS-CC (See Figure 1; Xing and Liu, 2008) was used to obtain survey data from the middle and eastern regions of Zhejiang, Anhui, and Shandong Provinces, where senior education is better developed and the representation is more prominent. A random sampling method was adopted to randomly distribute electronic questionnaires to urban and rural residents over 50 years old (defined by the enrollment requirements of senior colleges) who participate in learning through educational institutions, such as senior colleges, community colleges, and adult schools (based on household registration location). At the beginning of the study, the inclusion criteria and the study goals were explained to all participants, and their consent to participate was obtained. For the participants who could not read or understand questionnaires independently, the investigators read questions one by one helped them fill in the questionnaires. Participants were assured that confidentiality and anonymity would be maintained. Ultimately, 2,007 valid questionnaires were obtained, with a valid questionnaire recovery rate of over 85%. The final samples consisted of 1,234 urban older adult learners (61.5%) and 773 rural older adult learners (38.5%). Table 1 provides the demographic characteristics of the participants in the current study.

We used SPSS 21.0 as the research tool for this study. First, descriptive statistical analysis and difference test were used to analyze the differences in the SWB of older adult learners and the dimensions of older adult learning in terms of demographic variables. Second, correlation analysis was used to verify whether older adult learning (learning investment, learning experience, and learning atmosphere) affects the SWB of urban and rural older adult learners. Independent t-test was used to verify whether there was a significant difference in SWB between urban and rural older learners. Finally, multiple linear regression was used to examine the mechanism of older adult learning on the SWB of urban and rural older learners and to discover the differences in the impact effect of older adult learning on enhancing the SWB of urban and rural older adults.

Means, SDs, and possible ranges were calculated for main study variables in the current study. According to the research data (Table 6), there were significant gender differences in SWB, with women significantly higher than men (5.55 ± 0.757 vs. 5.25 ± 0.955, t = −5.716, p < 0.01); there were significant age differences in participants’ SWB, with those aged 50–59 having the highest SWB and those aged over 80 having the lowest (F = 15.217, p < 0.01); there were also significant differences in SWB of participants with different incomes (F = 4.858, p < 0.01); SWB was highest for participants with disposable income between RMB 1,000 and 2,000. There were also significant differences in SWB related to education level; specifically, the lower the education level, the higher the SWB (F = 15.761, p < 0.01).

There were significant gender differences in learning atmosphere and learning experience in the dimensions of older adult learning, with females scoring higher than males; the learning investment, learning atmosphere, and learning experience of older seniors were significantly higher than those of younger seniors. There were also significant differences in learning investment, learning atmosphere, and learning experience among participants with different incomes and education levels; specifically, older adults with incomes above RMB 4,000 had the highest learning commitment, and those with lower education levels had the highest learning experience. The learning investment and learning atmosphere of older adults with lower education levels were better.

To identify the differences in the correlations of learning between urban and rural older adult groups,¹ the raw sample data were processed for case selection using SPSS, and correlation analysis was conducted for both separately (Table 7). Data analysis revealed a significant positive correlation (p < 0.01) between older adult learning (learning investment, learning atmosphere, and learning experience) and SWB (including physical and mental health experience, self-development experience, and adaptation and satisfaction experience) of urban and rural older adults. Learning investment had a low correlation with older adults’ SWB (r = 0.139), learning atmosphere had a moderate correlation with SWB (r = 0.774), learning experience had a strong correlation with SWB (r = 0.851), and the latter two were more closely related to older learners SWB. There was a positive correlation between senior learning and learners’ SWB. Accordingly, Hypothesis 1 was well verified.

To further clarify the differences between the SWB of urban and rural older adult learners, independent t-test was performed on the sample data (Table 8). Results showed significant differences between rural and urban older learners in terms of learning investment, learning atmosphere, and learning experience. Learning atmosphere and learning experience had higher scores in rural than in urban older learners, while the opposite was true for learning investment. In addition, there were significant differences in physical and mental health experience, self-development experience, and adaptation satisfaction experience. In terms of mean values, physical and mental health experience, self-development experience, and adaptation satisfaction experience were significantly stronger in rural older learners than in urban older learners. Overall, SWB differed between urban and rural older learners. Therefore, Hypothesis 2 was supported.

The results of interaction effect analysis (Table 9) show that there was a significant effect of older adult learning on SWB (F = 32.189, p = 0.000 < 0.05, η² = 0.745); there was a significant effect of type of household registration on older adults SWB (F = 8.036, p = 0.005 < 0.05, η² = 0.005); the interaction of senior learning and type of household registration had a significant effect on SWB (F = 1.513, p = 0.002 < 0.05, η² = 0.073). Therefore, regression analysis was conducted with different samples from rural and urban areas to discern the urban–rural variability of the effect of learning factors on the SWB of older learners.

Given the numerical characteristics of the study variables, multiple linear regression analysis with ordinary least squares was used to estimate the parameters of the factors influencing SWB of older learners. The study used age, gender, education, marital status, and monthly income as control variables, SWB as the dependent variable, and learning investment, learning atmosphere, and learning experience as independent variables. Y was the dependent variable (SWB), X₁,X₂...X_k were the independent variables (older learners and their factors), and the multiple linear regression model was set as follows (Table 10):

Regression analysis of the overall sample of older learners revealed that the standardized equations for the three multiple linear regressions for the full sample, the rural subsample, and the urban subsample were:

Combining Table 6 and the standard equation of the regression model, it can be seen that learning engagement, learning atmosphere, and learning experience have a positive predictive effect on the SWB of older learners (F = 1926.256, p < 0.001, R² = 0.743), which can explain 74.3% of the variance probability of SWB, among which learning engagement has a weak effect on the SWB of older learners. The regression results of SWB of rural older learners showed that learning engagement, learning atmosphere, and learning experience had significant positive effects on SWB (F = 774.831, p < 0.001, R² = 0.751), which could explain 75.1% of the variance in SWB. The regression results of SWB of urban older learners showed a significant positive effect of learning atmosphere and learning experience (F = 1138.128, p < 0.001, R² = 0.735), which could explain 73.5% of the variation in SWB, but learning engagement (p = 0.237 > 0.05) had no corresponding effect.

The comparison of urban and rural older adult learners shows that learning commitment, learning atmosphere, and learning experience all have a significant positive effect on the SWB of rural older adult learners and better explain their SWB, while only learning atmosphere and learning experience have a positive predictive effect on the SWB of urban older adult learners and more poorly explain SWB of this group. In other words, there were some urban–rural differences in senior learning’s effect on participants’ SWB. Therefore, Hypothesis 3 was well verified.

Some limitations of this study should be noted. First, the survey data came from the middle and eastern regions of Zhejiang, Anhui, and Shandong Provinces, where the development of senior education is enhanced, with increased representation; however, the sample may not be representative of the population nationally, which limits the scope of promotion. Second, due to the cross-sectional design, the causal relationship of main study variables should be further confirmed in the future studies. Third, the use of the Internet and institutional social media may have caused a bias against older adults who cannot use the Internet. Finally, although we pointed out that there are differences in the effects of older adult learning on urban and rural older learners’ SWB, there are many other educational factors, such as education gap, that can cause differences, which will be our focus in future studies.

Without disregarding these limitations, our results introduce diverse policy implications aimed at supporting older adults in their learning to contribute to their SWB. On the one hand, echoing China’s policy reports, our findings posit the urgency of generating strategies to assist older adult education agencies and older adult learners in strengthening SWB. On the other hand, our study explores the urban–rural differences in SWB and the influencing mechanisms of urban–rural older adults in terms of learning participation, which is unique to our study. While previous studies have focused on the influence of objective factors on older adults’ SWB, this study provides a new perspective to study the differences between older adults. Additionally, our study has a large sample size to be of practical guidance. Finally, through empirical research, this study truly reflects the current situation of older adults’ participation in senior education, objectively measures their SWB, comprehensively analyzes the relationship between senior education and older adults’ SWB, and enriches the relevant theoretical research.