Retirement resources and post-retirement well-being: the mediating effect of sense of coherence and the moderated mediation effect of leisure activity types

This study investigated the mediating effect of sense of coherence on the relationship between retirement resources and post-retirement well-being (i.e., retirement adjustment and retirement satisfaction) and the moderated mediation effect of leisure activity types (i.e., active vs. passive physical activity) on the relationship. An online survey was distributed to Korean baby boomer retirees, born between 1955 and 1963, and 442 participants completed the survey. Structural equation modeling and multigroup analyses were performed to examine the hypothesized model. The results showed that sense of coherence (SOC) mediated the relationship between retirement resources, retirement adjustment, and satisfaction. Additionally, the analyses demonstrated a moderated mediation effect from leisure activity types on this relationship. Specifically, the effect of retirement resources on retirement satisfaction through SOC was higher for Korean baby boomer retirees who participated in active leisure types, such as golfing and climbing. Therefore, considering active leisure types to better cope with the post-retirement lifestyle and increase post-retirement well-being is crucial.


Introduction
Retirement can be a stressful transition, often causing a wide range of noticeable changes, such as the loss of stable income, increased free time, and detachment from colleagues at work (Leung & Earl, 2012).Previous studies have found that Korean "baby boomers," born between 1953 and1963, are the first generation to go into the 20 -30 years following retirement unprepared owing to the increase in life expectancy (Kim, 2016;Lee, 2016).Given the influence of Confucianism, South Koreans are morally responsible for caring for their aging parents (Jeong, 2012).They are also culturally expected to provide financial support until or when their siblings marry (Lee et al., 2018).Thus, most Korean baby boomers face numerous emotional and financial difficulties during retirement because they are often sandwiched between supporting their parents/siblings and their own children (Jeong, 2012;Jeong et al., 2011;Lee et al., 2018;Park, 2017).Given these challenges, understanding the influences of baby boomers' successful post-retirement well-being (Lee et al., 2018) and quality of life is imperative.
To understand retirees' perceptions of well-being, researchers (e.g., Hansson, 2019;Pinquart & Schindler, 2007;Wang et al., 2011) have used the resource-based dynamic process theory, which emphasizes the role of resource availability in an individual's well-being (Antonovsky, 1987).Personal and social factors, such as economic status, health status, and educational level, represent important resources for coping with new challenges after retirement (Wang, 2007).The resources, means, or assets that can be used to cope with transition-related challenges (Doo & Chang, 2015) are assumed to be key to the successful adjustment to retirement because they influence retirees' financial status (Hansson, 2019).The level of adjustment after retirement depends on retirees' resources (Doo & Chang, 2015;Wang, 2007); those with abundant resources are more successful (Pinquart & Schindler, 2007), which has a positive effect on retirement satisfaction (Wang, 2007).
A key concept within the theory is sense of coherence (SOC), which refers to a person's coping capacity to deal with stressors in everyday life and was introduced to explain why some people with greater SOC manage stress well and stay positive under stressful conditions (e.g., retirement (Antonovsky, 1979).SOC comprises three elements: comprehensibility, manageability, and meaningfulness.It examines an individual's understanding of their situation, the appropriate use and allocation of available resources, and their ability to find meaning in life on their own (Antonovsky, 1987).Thus, if an individual has greater SOC, they can effectively cope with upcoming stressors (Antonovsky, 1993) and consequently adapt to stressful situations, maintaining a positive view of retirement (Lee et al., 2018), which can positively affect satisfaction.Previous studies have found that increased SOC significantly reduces negative emotions, such as depression, stress, anxiety, and loneliness, and results in a better quality of life among retirees (Antonovsky, 1993;Antonovsky et al., 1990;Karlsson et al., 2000;Lee et al., 2018).Although SOC explains how individuals manage stress during retirement, research directly examining the mediating role of SOC among retirement resources, retirement adjustment, and retirement satisfaction is limited.Therefore, this study tested whether SOC has a mediating effect on the relationship between retirement resources, retirement adjustment, and retirement satisfaction.
Leisure activity types can also affect retirement adjustment processes in different ways (Driver et al., 1991;Jeong & Kim, 2017).For example, active leisure (e.g., engaging in sports during free time) facilitates positive adjustment because it promotes a healthy lifestyle among retirees (Lee, 2010;Nam, 2017).Physical leisure activities can also positively influence retirees' satisfaction (Jeong & Kim, 2017).Furthermore, participation in various types of leisure activities is not only closely related to SOC but also contributes to physical and mental health (Lee et al., 2020;Nagata et al., 2020).Nagata et al. (2020) noted that leisure behavior contributes to lower depressive symptoms and predicts greater SOC.Therefore, greater SOC is related to lower perceived stress and depressive symptoms.Those with high SOC engaged in more physical activities in their spare time.(Myers et al., 2011).The relationship between SOC and leisure behavior has been revealed in previous studies.Thus, our hypothesized model will be tested by two different leisure groups: passive and active leisure participants.
To advance our understanding of the roles of SOC and leisure activity types in post-retirement well-being in South Korea, this study investigated their respective mediating and moderated mediation effects on the relationship among retirement resources, retirement adjustment, and retirement satisfaction.This study provides a theoretical review of the retirement process of Korean baby boomers who bear the financial burden of their children while supporting their parents and offers suggestions regarding possible policy decisions.The following section reviews the literature on resource-based dynamic process theory and the role of SOC.

Resource-based dynamic process theory
Resource-based dynamic process theory emphasizes the role of resource availability in individual well-being (Wang et al., 2011), offering an integrative theoretical framework to understand the dynamic, multidimensional aspects of retirement.The theory assumes that resource change is the driving mechanism of changes in well-being during the transition to retirement (Wang et al., 2011; see Fig. 1).Resources are key elements because they define retirement conditions and influence what people can do, physically and financially.Thus, more resources presumably lead to fewer adjustment problems and better wellbeing.
Multiple factors at the micro (personal and situational), meso (job and organizational), and macro (societal norms and government policies) levels influence the availability of and changes in resource capability.This, in turn, determines how individuals adjust to retirement (Wang et al., 2011).The physical, cognitive, motivational, financial, social, and emotional aspects of a person's total resource capabilities are important to this process (Wang, 2007).The resource-based dynamic model proposes that well-being can fluctuate owing to changes in an individual's resources over time, and that a loss in one resource may be compensated for by gains in others (Wang et al., 2011) (Fig. 2).

The role of SOC in the relationship between retirement resources and post-retirement well-being (retirement adjustment and retirement satisfaction)
Access to resources defines an individual's retirement condition and influences the opportunities and quality of their retirement experience (Kim & Moen, 2001).Among the numerous variables influencing retirement well-being (retirement adjustment and retirement satisfaction), resources stand out as a key variable that has generated research interest in recent years (Wang, 2007;Wang et al., 2011).Wang et al. (2011) describes retirement adjustment as a longitudinal process, during which retirees' adjustment levels fluctuate as a function of individual resources and changes to them.Retirement satisfaction is primarily related to access to key resources, such as health, finances, and marital relationships (van Solinge & Henkens, 2008).Most psychological research on satisfaction and adjustment to retirement has focused on the process of adjusting to life's changes (Carter & Cook, 1995;Folkman & Lazarus, 1980).Numerous studies have examined whether cognitive and behavioral coping (e.g., altering sources of distress and/ or eliminating stressful emotions) can be more effective in promoting well-being than others (Folkman & Lazarus, 1980;Schroevers et al., 2007).In this regard, Antonovsky (1987) asserts that SOC-that is, how confident an individual is in predicting and managing their environment using personal psychological resources-is associated with how an individual copes with life stresses, such as retirement (Lee et al., 2018).SOC emphasizes three components: the comprehensibility of one's world (cognitive aspect), meaning of one's life (motivational aspect), and manageability of one's outcomes (behavioral aspect) (Antonovsky, 1987;Lee et al., 2018).The level of SOC is shaped by an individual's cumulative experience utilizing their current resources (Antonovsky, 1987).
Based on the reviews of literature, this study investigated the mediating effect of SOC on the relationships between retirement resources, retirement adjustment, and retirement satisfaction among Korean baby boomer retirees using the resource-based dynamic process theory.The moderated mediation effect of leisure activity types on the 1.Retirement resources and post-retirement well-being (retirement adjustment and retirement satisfaction)

Retirement resources and retirement adjustment
Multiple factors at the micro (personal and situational), meso (job and organizational), and macro (societal norms and government policies) levels influence the availability of and changes in resource capability.This, in turn, determines how individuals adjust to retirement (Wang et al., 2011).Our study thus aimed to explore how retirement resources support relates to retirement adjustment.
H1-1: Retirement resources are positively related to retirement adjustment.

Retirement resources and retirement satisfaction
The Resource-based dynamic process theory assumes that resource change is the driving mechanism of changes in well-being during the transition to retirement (Wang et al., 2011).Our study was conducted within the theory, the study proposes this hypothesis to identify the relation between retirement resources and retirement satisfaction.H1-2: Retirement resources are positively related to retirement satisfaction.
2. Mediating effect of SOC Antonovsky (1987) asserts that SOC is associated with how an individual copes with life stresses, such as retirement (Lee et al., 2018).The level of SOC is shaped by an individual's cumulative experience utilizing their current resources (Antonovsky, 1987).Our study was conducted with how retirement resources relate to retirement adjustments and retirement satisfaction through the path SOC.Thus, our hypotheses are: H2-1: SOC mediates the relationship between retirement resources and retirement adjustment.H2-2: SOC mediates the relationship between retirement resources and retirement satisfaction.

Moderated mediation effect of leisure activity types
Compared with passive leisure, retirees have more opportunities to participate in social gatherings during active leisure activities (Kim, 2004;Na, 2004).Based on previous research, we proposed two hypotheses: H3-1: Leisure activity type moderates the mediating effect of the relationship among retirement resources, SOC, and retirement adjustment.This relationship is stronger (weaker) when retirees participate in active (passive) leisure.H3-2: Leisure activity type moderates the mediating effect of the relationship among retirement resources, SOC, and retirement satisfaction.This relationship is stronger (weaker) when retirees participate in active (passive) leisure.

Pilot study
A pilot survey with 150 participants was conducted for scale reduction and to validate the variables of retirement resources, post-retirement well-being adjustment and satisfaction, and SOC (see "Measures" section below for further detail).Using a significance sampling method, data were obtained from an online Google survey (response rate of 98.1%) conducted in September 2021.Based on confirmatory factor analysis (CFA), we deleted nine questions (factor loadings < 0.04) and six questions (crossloadings) from the retirement resources scale.Thus, the final survey included 20 items on retirement resources, 13 on retirement adaptation, 11 on retirement satisfaction, and 14 on SOC, in addition to eight demographic questions.

Main study
An a priori power analysis was conducted using pwrSEM (power analysis in Structural Equation Modeling), which is a power analysis tool for parameter estimation in structural equation modeling (SEM; Wang & Rhemtulla, 2021).The analysis showed that 200 participants were required for adequate power.Using purposive sampling, data were collected using the M-brain survey platform (n = 200), an online survey service, and Google Survey (n = 242) to prevent data bias.Data were collected in September and October 2021.Upon successful completion of the survey, participants were given a mobile coffee coupon as compensation.To screen for eligibility as retirees from the baby boomer generation, respondents were asked whether they were born between 1955 and 1963 and whether they had quit their previous main job or were engaged in small work but were not willing to work.Only those who met the criteria (i.e., year born and currently retired from a major job) proceeded to the next questions.In total, 442 retirees participated in the final survey.Four invalid responses were removed owing to missing data which completion rate was under 90%; thus, 438 valid responses were included in the analyses.This study was approved by the Institutional Review Board of Ewha Womans University (ewha-202109-0036-01).

Measures
Retirement resources were assessed using the Retirement Resources Inventory (Leung & Earl, 2012), which contains five dimensions: physical and financial (three items), social (four items), emotional (four items), motivational (four items), and cognitive (five items).In a previous study, all subscales showed very good internal consistency (Cronbach's α = 0.81-0.89,overall = 0.90) (Leung & Earl, 2012).SOC was measured using items adapted from Antonovsky (1987) and Kim (2013) and included the dimensions of comprehensibility (five items), manageability (four items), and meaning (four items).In a previous study, the overall Cronbach's α coefficients for the above measures were 0.88, showing acceptable levels of reliability, and all factor loadings were significant at the 0.001 level (Lee et al., 2018).Retirement satisfaction was measured using 11 items adapted from the Retirement Satisfaction Inventory (Floyd et al., 1992).Previous estimates of internal consistency of retirement satisfaction showed adequate internal consistency for this tool (Cronbach's α = 0.70) (Floyd et al., 1992).To determine retirement adjustment, items were adopted from Schultz and Schultz (1997), which includes two dimensions: financial adjustment (three items) and retirement adjustment (10 items).In a previous study, the Cronbach's α coefficient for overall retirement adjustment was 0.81, showing good internal consistency (Schultz & Schultz, 1997).Retirement adjustment and retirement resources were measured on 5-point Likert scale, ranging from 1 ("Strongly Disagree") to 5 ("Strongly Agree").Retirement satisfaction was measured on a 6-point Likert scale, ranging from 0 ("Not applicable"; 1 = "Strongly Disagree") to 5 ("Strongly Agree").SOC was measured on a 7-point Likert scale, ranging from 1 ("Never happened") to 7 ("Always happened").Finally, leisure activity types were classified based on Koh (2010), which divided the leisure activities of middle-aged people and retirees.Types of leisure were measured by the respondents' answers: 1 = "active leisure activities" (i.e., physical activities such as swimming, tennis, golf, gate ball, dance sports, badminton, climbing, gymnastics, etc.) and 2 = "passive leisure activities" (i.e., non-physical activities such as play go, reading, chess, calligraphy, listening to radio, watching TV, etc.).

Analysis
The data were analyzed using SPSS 22.0 and Mplus 7.0 to conduct CFA, mediating effect analysis, and multi-group analysis.First, CFA was performed using Mplus, where all factor loadings were above the 0.5 threshold, and the internal consistency (Cronbach's α) of all factors was above Van de Ven and Ferry's (1980) 0.70 threshold.Second, SEM was conducted to determine the relationships among retirement resources and SOC, retirement adaptation, and retirement satisfaction, and a bootstrap analysis was used to verify the mediating effects.The structural model of this study was judged to have a predictive comparative fit index (CFI) and root mean square error of approximation (RMSEA).The criteria for a good model are a CFI > 0.90 and RMSEA < 0.80 (Hu & Bentler, 1995).Thus, we set the bootstrap confidence interval (CI) to 95% and the number of bootstrap samples to 500.If zero was not included in the 95% CI, the mediating effect was considered significant (MacKinnon, 2012;MacKinnon et al., 2004).Item parceling was conducted for the mediating variable of SOC, which was treated as a manifest variable in the analyses.Mediation was tested using the summed scores of all 13 items.Finally, a multigroup analysis was conducted to determine whether the type of leisure activity moderated the hypothesized relationships (Fornell & Larcker, 1981).Based on this analysis, clustering was considered a complex option in Mplus (Muthen & Muthen, 2017).

Descriptive analysis
Approximately 55% of respondents were men, with a mean birth year of 1959.Over half of the respondents were university graduates (54.8%) and 82.9% were married.Approximately 62.6% had participated in active leisure activities (Table 1).

Reliability and validity tests
Using SPSS 18.0, Mplus 7.0, and standardized factor loading, the Cronbach's α value (internal consistency reliability coefficient) was used to measure the convergence validity of each potential variable in the model.The results are summarized in Table 2.The Cronbach's α value of each variable was above 0.713, and the standardized factor loading was above 0.520, which indicated that each potential variable had acceptable convergence validity and reliability (Qu et al., 2009;Zhang et al., 2021).

The mediating role of SOC in the relationship between retirement resources, retirement adjustment, and retirement satisfaction
We found significant direct effects of retirement resources on retirement adjustment and satisfaction (see Table 3).The direct effect of retirement adjustment was 0.149 (p < 0.01), and the total effect was 2.502.The direct effect on retirement satisfaction was 4.664 (p < 0.001), and the total effect was 6.178.Furthermore, we found significant indirect effects from retirement resources on retirement adjustment and retirement satisfaction through SOC.The relationship between retirement resources and retirement adjustment (indirect effect: 2.353 [95% CI: 1.577, 3.129]) did not contain zero and was significant.Another pathway of retirement resources and retirement satisfaction through SOC (indirect effect: 1.513 [95% CI:0.198, 2.971]), did not contain zero and was also significant.Thus, SOC was shown to be a significant mediating variable.

The moderated mediation effect of leisure type
According to the multi-group analysis, leisure type had a significant moderated mediation effect on the relationship between retirement resources, SOC, retirement adjustment, and retirement satisfaction (Table 4).Regarding retirement resources to SOC pathway, the active leisure type group (Group B) scored 1.411 higher than the passive leisure type group (Group A); that is, the positive effect of Korean baby boomers' retirement resources on SOC was stronger among active leisure participants than passive ones.Regarding the indirect effect of retirement resources on retirement adjustment through SOC, Group B scored 0.637 higher than Group A. Regarding the indirect effect of retirement resources on retirement satisfaction through SOC, Group B scored 1.133 higher than Group A. These results indicate that if respondents participate in active leisure activities, they are more likely to have higher retirement resources (Heintzman & Coleman, 2010;Lee et al., 2018;Yeung & Zhou, 2017) and a better perception of retirement well-being.Among leisure participants in later life, it seems important to engage in active types of leisure to better cope with post-retirement life.

Discussion
In this study, we examined whether SOC -one's ability to manage their resources to deal with stressors -played a mediating role in the relationship between retirement resources, retirement adjustment, and retirement satisfaction among Korean baby boomer retirees.We then examined how active versus passive types of leisure activities moderated the mediation of this relationship.As baby boomers contemplate retirement, it is essential to understand what, how, and where they plan to maximize their well-being (Emerson et al., 2022).
The findings demonstrated the following.First, SOC plays a significant role in the relationship between retirement resources, retirement adjustment, and satisfaction.Specifically, SOC appears to help solve problems by using various resources in transitional situations, such as retirement.This coping process positively affected retirement adjustment and retirement satisfaction.Previous studies (Leung & Earl, 2012;Wang et al., 2011) have also indicated that different resource types (i.e., physical, financial, social, and emotional) positively affect retirement adjustment.Additionally, retirement satisfaction is positively predicted by retirees' resources, such as health and financial conditions (van Solinge & Henkens, 2008).The SOC dimensions of comprehensibility, manageability, and meaning represent the cognitive, behavioral, and motivational aspects of retirement resources, respectively.The results of the current study show that cognitive (comprehensibility) and motivational (meaning) resources are important for well-being (Lee et al., 2018;Van der Colff & Rothmann, 2009).Similarly, a higher SOC contributes to promoting retirement satisfaction, implying that a stronger SOC helps individuals deal with challenging situations after retirement by adopting cognitive and emotional resources (Van der Colff & Rothmann, 2009).People with higher SOC tend to think positively about their life after retirement (Lee et al., 2018).Previous researchers have also found a consistent positive relationship between SOC and post-retirement well-being, suggesting that for those who reside in a midwestern state in the US, having higher SOC brought positive changes in the perception of well-being (Lee et al., 2018).Moreover, SOC has a positive effect on psychological well-being because it controls negative emotions, such as anxiety and stress, which are closely related to positive retirement well-being (Ying et al., 2007).
Overall, the results of this study demonstrate the mediating role of SOC between retirement resources and retirement well-being.This relationship is supported by the findings of previous studies (Lee et al., 2018).The higher Korean baby boomers' retirement resources, the higher the level of SOC, allowing them the flexibility to cope with changes in the external environment.This effect can positively influence retirement adjustment and satisfaction (Antonovsky et al., 1990;Ying et al., 2007).In addition, leisure type (i.e., active and passive leisure) showed a moderated mediation effect on the hypothesized relationships.Specifically, the strength of the paths from retirement resources to retirement adjustment and retirement satisfaction, which is mediated by SOC, depends on the type of leisure activity.In a previous study, active leisure participation, such as gateball and cycling, promoted social engagement among retirees (Na, 2004).Thus, some degree of physical activity remains important to promoting physical and mental well-being (Majeed & Ramkissoon, 2020, 2020;Ramkissoon, 2023).Compared with passive leisure, retirees have more opportunities to participate in social gatherings during active leisure activities (Kim, 2004;Na, 2004).Later in life, the social interactions brought about by leisurely activities have a positive influence on adjustment to a retirement lifestyle (Kim, 2014).Moreover, our results show that the social, emotional, and mental satisfaction obtained through physical activity contributes to the formation of positive emotions, which in turn play an important role in individual satisfaction and happiness (Park, 2008;Ramkissoon, 2020).Thus, this study showed that active leisure participation can have a positive effect on retirement well-being by helping retirees find meaningful ways to cope better with the stressors fueled by retirement.

Theoretical implications
Most importantly, this study was of theoretical significance in that it directly verified that the strength of the relationship between retirement resource, SOC and retirement well-being (i.e., retirement adjustment and retirement satisfaction) could vary depending on the type of leisure activity.Various researchers have explored the resource-based dynamic process theory since the 1990s (Hansson, 2019;Wang, 2007;Wang et al., 2011).Still, however, it is of academic significance because few studies have verified this theory based on the type of leisure activities.
Furthermore, this study was meaningful because it expanded retirement and leisure research related to baby boomers, mostly concentrated in Western cultures, to Eastern cultures.Baby boomers in Western societies -such as the USA, the UK, and Canada -emerged relatively earlier because of the First and Second World Wars.Thus, their well-documented personal interests in self-fulfilling leisure activities and experiences and their high demands for customized or specialized travel products and services have generated various travel and leisure products (Patterson, 2006).However, because of the late emergence of Baby Boomers in Korea and despite the apparent potential of the Baby Boomer market, very little research has been paid to these individuals (Han et al., 2011;Kim et al., 2012;Lee, 2011).

Practical implications
Here are some practical implications based on the results of the study findings.First, for Korean baby boomer retirees, this study showed that SOC could be a key factor that positively affects retirement lifestyle.Although they have sufficient retirement resources, having a better SOC (i.e., managing retirement resources well) was the most important coping process for maintaining better retirement satisfaction and adjustment.Specifically, regarding the Korean baby boomer generation, going through an appropriate resocialization process after retirement was found to have a significant effect on post-retirement health (Um et al., 2020).The importance of SOC was further emphasized because continuous social contact is crucial to solving the emotional problems experienced after retirement and to lead a happy retired life (Jang, 2010;Kim & Lee, 2009).Moreover, regardless of how much financial, time, and environmental resources retirees have, government-level education on how to manage them efficiently is arguably necessary as education plays a significant role in preparing better societies (Ramkissoon, 2022).People who worked at state-run companies or schools in Korea may be less worried about the economy after retirement because they can receive pension benefits after retirement.However, our study showed that people who suddenly/ involuntarily retired from private companies without any preparation or education needed help at the national level.Thus, practical education that can increase SOC is required, rather than the amount of retirement resources themselves.Second, as a practical implication of the study, leisure practitioners who target baby boomers born between 1955 and 1963 should consider active types of leisure when designing leisure programs as our study shows that the retirement resources of those participating in active leisure types had a stronger influence on retirement adjustment and satisfaction.In Korea, public sports facilities (e.g., swimming, badminton, gymnastics, table tennis, etc.) and programs for older adults and retirees are relatively well-equipped in each community, but their utilization rate of these facilities/programs is less than half (Jung, 2021).Active publicity and efforts are needed to ensure that these facilities and programs, which are equipped through public resources, can be known to and used by more people.

Limitations and future directions
Although this study provides insights into baby boomers' retirement processes, it has several limitations.First, the active and passive leisure activity types were measured using only a few examples, including swimming, tennis, and watching TV.Thus, future researchers should consider how diverse active leisure activities promote retirees' adjustment and satisfaction.Second, socioeconomic factors, such as gender, should be among the factors considered in future retirement studies.Previous studies (Kim & Choi, 2011;Um et al., 2020) have indicated that Korean women adapt more easily to socialization after retirement than men.Regarding this aspect, it will be necessary to closely examine the differences between Korean men and women in adapting to retirement.Third, retirement timing should be considered when investigating retirees' coping (adjustments) and satisfaction with retirement.Since the number of years spent after retirement can affect the diversity of retirees' experiences and coping processes, future studies should measure retirement periods.Despite these limitations, this study contributed important findings, demonstrating the mediating effect of SOC between retirement resources, retirement adjustment, and retirement satisfaction.Furthermore, these relationships were stronger for active leisure types compared with passive types.Therefore, active leisure can positively influence retirement well-being.
Authors contributions All authors contributed to the study conception and design.Material preparation, data collection, and analysis were performed by Jinyoung Joo.The first draft of the manuscript was written by Jinyoung Joo and Jee In Yoon and all authors commented on previous versions of the manuscript.All authors read and approved the final manuscript.

Table 1
Sociodemographic characteristics of respondents

Table 2
Results of confirmatory factor analysis and reliability analysis

Table 3
Bootstrap methods for mediation among SOC, retirement resources, retirement adjustment, and retirement satisfaction ** p < .01,*** p < .001RR Retirement resources; SOC Sense of coherence; RA Retirement adjustment; RS Retirement satisfaction *