Family Stressors,Home Demands and Responsibilities,Coping Resources,Social Connectedness,and Thai Older Adult Health Problems

Abstract

Objectives: To examine gender variations in the linkages among family stressors, home demands and responsibilities, coping resources, social connectedness, and older adult health problems. Methods: Data were collected from 3,800 elderly participants (1,654 men and 2,146 women) residing in Kanchanaburi province, Thailand. Results: Findings indicated gender variations in the levels of these constructs and in the mediational pathways. Thai women indicated greater health problems than men. Emotional empathy was the central variable that linked financial strain, home demands and responsibilities, and older adult health problems through social connectedness. Financial strain (and negative life events for women) was associated with lowered coping self-efficacy and increased health problems. The model indicated greater strength in predicting female health problems. Discussion: Findings support gender variations in the relationships between ecological factors and older adult health problems.

Keywords

family life challenges Thai elderly health coping self-efficacy emotional empathy social connectedness

The older adult population in Thailand has been growing at a faster rate than many other South-east Asian countries, and by the year 2025, the population of adults 60 years and older is projected to constitute 19% of the Thai population. By the year 2050, older adults will represent 26% of the Thai population (United Nations Population Funds [UNFPA], 2011). According to the UNFPA report titled the “Impact of Demographic Change in Thailand,” the rapid decline in fertility rates, declining births, increases in life expectancy, and rapid population aging have also been accompanied by other demographic shifts such as declines in family size, higher survival rates among women, a falling productive support base, changing family structures, migration of adult children for work, and changes in the living arrangements of older adults (co-residence with adult children declined from 77% in 1986 to 59% in 2007). These factors have shaped the life experiences of older adults in Thai society.

Similar to other world societies, aging among Thai older adults is a gendered experience with women likely to experience greater disadvantages or inequities on a wide range of social, health, and economic outcomes compared to men (Soonthorndhada, Gray, Soonthorndhada, & Vishwanathan, 2008). The Gender Inequality Index which ranks countries on a composite index based on gender disparities in economic participation and opportunity, educational attainment, and political empowerment has indicated that Thailand ranks 65th out of 136 countries in terms of gender inequalities (World Economic Forum, 2013). On the other hand, Sobieszczyk, Knodel, and Chayovan (2002) have indicated that the relationship between gender and aging within the Thai ecological context is complex and that older Thai women may not be disadvantaged in every aspect of aging. Rather, in some areas, Thai older men may be more disadvantaged than women (e.g., higher mortality rates among men).

Surveys conducted over the last three decades with Thai older adults have indicated significant advances in self-assessed health status; they have also pointed to gender inequalities in self-reported health status (e.g., Haseen, Adhikari, & Soonthorndhada, 2010; Knodel & Chayovan, 2008). Across these studies, Thai older adult women were more likely to report poorer self-assessed health status compared with older adult men. The focus of the current investigation is to examine the ecological factors that are associated with gender variations in the health status of Thai older adults. This area of research has remained understudied within the Thai context. The main lines of research in the area of older adult health have focused on the sociodemographic and economic variations in health status, sociodemographic predictors of health, and the relationship between social support networks and health status (e.g., Haseen et al., 2010; Thanakwang & Soonthorndhada, 2011).

Few investigations have explored the role of ecological factors such as family stressors, home demands and responsibilities, and social psychological constructs that could account for the gender variations in older adult health problems. Hence, the goals of this investigation were twofold: (a) to examine gender variations in the experiences of family stressors (financial strain and negative family life events), home demands and responsibilities, coping resources (coping self-efficacy and emotional empathy), social connectedness, and older adult health problems and (b) to examine gender variations in the mediating role of coping resources and social connectedness in the relationship between family stressors, home demands, and older adult health problems.

Direct and Indirect Effects

Although increasing health concerns are considered part of the aging process, family stressors (financial strain and negative life events) and home demands and responsibilities could play an important role in determining older adult health problems. Chronic financial strain (e.g., Krause, Jay, & Liang, 1991), negative life events (e.g., Jopp & Schmitt, 2010), and life challenges (e.g., spouse in poor health, difficulty meeting expenses, loss of spouse, loss of a child) (Beckett, Goldman, Weinstein, Lin, & Chuang, 2002) have been linked to increases in blood pressure and cardiovascular disease (Carroll, Ring, Hunt, Ford, & Macintyre, 2003), psychological distress (e.g., depression, somatic complaints) ( Krause et al., 1991), and poor overall health and functional limitations (e.g., Artazcoz, Borrell, & Benach, 2001; Beckett et al., 2002). The association between stressful situations and poor health status has its basis in stress theory which indicates that exposure to chronic stressors (e.g., financial strain, negative life events) can bring about physiological or emotional arousal which over time can lead to poor physical and psychological outcomes (Kelly, Hertzman, & Daniels, 1997; Thoits, 1995). In the study by Beckett et al. (2002), the association between chronic financial strain and poor health status was found to be stronger for older adult men than women, whereas the association between negative life events (e.g., poor spousal health) and poor health status was found to be stronger for older adult women than men.

The association between home demands and responsibilities (e.g., housework, childcare, caring for the disabled) and older adults’ health status has not been explored within the Thai context. Studies conducted in other countries linking home demands and responsibilities and older adult health status have indicated inconsistent findings (e.g., Artazcoz et al., 2001). One view might be that work at home is devalued in many cultures and as a result the gender that does most of this work will by association also be devalued by society, which in turn could negatively affect health status. The other argument is that gender ideologies regarding which household tasks (e.g., childcare, housekeeping, caring for other older adults) are considered as male and which are considered as female are culturally determined and are internalized by individuals in society as appropriate and fair (Bianchi, Milkie, Sayer, & Robinson, 2000). In Thai households, women have generally borne the responsibility for childcare and housework (similar to the findings in Western countries; Coltrane, 2000) whereas home management and planning have generally been the responsibility of men. If the distribution of home demands and responsibilities are seen as a fair and an appropriate division of labor, then it is possible that Thai men and women may have a favorable attitude toward home demands and may also attach great importance toward performing these responsibilities which could have a favorable impact on health outcomes. The positive link between home demands and positive health status may also be supported by activity theory first proposed by Havighurst and Albrecht (1953) and later advanced by others (for a detailed review see Adams, Leibbrandt, & Moon, 2011), which suggests that active participation and engagement in a range of formal and informal activities are critical to successful aging.

Another important body of work has indicated that older adults who enjoy greater informal social support and connectedness with their social environments are likely to experience better health status (Berkman, Glass, Brissette, & Seeman, 2000; Thanakwang & Soonthorndhada, 2011). In an investigation conducted with older adults in Taiwan, social connections such as the number of friends/neighbors that the older adults saw or talked to weekly and their participation in social activities was significantly linked to better health outcomes for women (Beckett et al., 2002). Participation in friendship networks and social activities was linked to better health status among Thai older adults (Apinunmahakul, 2012; Thanakwang & Soonthorndhada, 2011). Lower level of participation in social activities was more strongly associated with poorer health status for women than men (Beckett et al., 2002). The link between social connectedness and health outcomes is also supported by activity theory.

Social connectedness could function as a mediator of the link between family stressors, home demands and responsibilities, and older adult health problems. The mediating link between family stressors and poor older adults health status through lowered social connectedness has been suggested in conceptual frameworks proposed by Fry (1989) and tested in a few investigations (e.g., Kwag, Martin, Russell, Franke, & Kohut, 2011). Although there are a limited number of studies linking home demands and social connectedness, two potential hypotheses can be speculated. Home demands and responsibilities could hinder the time that older adults have for social activities, which could undermine older adults’ health status. On the other hand, meeting home demands may be viewed as an expected duty and responsibility of family members (and not a stressor), and hence does not take time away from social activities and relationships and therefore the link could be positive and associated with better health status.

Beyond these two critical linkages (i.e., the direct link between exposure to family stressors, home demands, and older adult health problems andthe mediating role of social connectedness in the relationship between family stressors, home demands and responsibilities, and older adult health problems), an important factor that is relevant to understanding the vulnerabilities of older adults to stressful situations is the availability of coping resources. Coping resources such as coping self-efficacy beliefs which include the self-confidence and ability of individuals to effectively cope with stressful situations (Bandura, 1997) could help explain variations in the link between exposure to negative family stressors, social connectedness, and older adult health outcomes. The construct of coping self-efficacy has its basis in stress and coping (Lazarus & Folkman, 1984) and self-efficacy (Bandura, 1997) theory. It includes the ability of an individual to judge whether a situation is controllable and also whether the person believes in his or her ability to use appropriate coping strategies when dealing with stressful situations (Chesney, Neilands, Chambers, Taylor, & Folkman, 2006). Drawing on the work by Caplan (1981) on the stress process and the work by Krause et al. (1991) linking financial strain and diminished personal control (control over events and carry out plans as expected) among Japanese and American older adults, the link between family stressors and coping self-efficacy can be hypothesized. Diminished confidence and the inability of individuals to effectively cope with stressful situations will lower their interest and competence to form and maintain social relationships which could be associated with poor health status.

Another key coping resource that could be critical in explaining the link between exposure to stressful situations and social connectedness includes the expressive ability of individuals to respond to the needs and concerns of others. Emotional empathy has been linked to prosocial behaviors (engaging in active helping and problem-solving behaviors) among older adults (e.g., Sze, Gyurak, Goodkind, & Levenson, 2012) and also to social connectedness (e.g., Hoffman, 2000). The need to be emotionally connected and responsive to the needs of others does not decline with advancing age and remains relevant and important to aging adults (Sze et al., 2012). Women have indicated higher levels of empathetic concern than men (O’Brien, Konrath, Grühn, & Hagen, 2013; Sze et al., 2012). Few investigations have explored issues of empathic concern in Asian families; the online study by O’Brien et al. (2013) indicated that Asian Americans reported higher levels of empathic concerns than European Americans. It can be suggested that in Asian cultures where social relationships and maintenance of group cohesiveness are valued, portrayal of sympathy and concern for others may be displayed more than in Western cultures (e.g., Triandis, 1994). Dealing with chronic stressors such as financial strain and negative life events could be debilitating and associated with emotional exhaustion and a lowered ability to show empathic concern and caring for others. On the other hand, home demands and responsibilities if considered as duties that must be carried out by individuals in daily family life (e.g., caring for family members) could bring about greater closeness to kin and family. Hence, empathy and caring for family members could be associated positively with expressions of emotional empathy for others beyond that which is extended to family members.

In summary, although not systematically tested in prior investigations, there is some piece-meal evidence that there may be gender variations in the relationships among various ecological factors and Thai older adults’ health status. The proposed model (see Figure 1) was informed by stress theory (Thoits, 1995), gender role theory (Bianchi et al., 2000), activity theory (Havighurst & Albrecht, 1953), and other conceptual models (e.g., Berkman et al., 2000) and is set within an ecological framework. The plus/minus signs (+/−) against the pathways in Figure 1 indicate the hypothesized direction of the effects of the overall model. In addition, we also posited the direct effects of family stressors and home demands on older adult health problems. We also examined additional pathways such as the mediating role of social connectedness (without the introduction of the coping resources) as indicated in the hypotheses below.

Figure 1.

Proposed conceptual model linking family stressors, home demands and responsibilities, and older adult health problems.

Hypotheses

Four hypotheses were examined in this investigation:

Hypothesis 1: There are likely to be gender variations in the levels of family stressors, home demands and responsibilities, coping self-efficacy, emotional empathy, social connectedness, and older adults’ health problems. We hypothesized that older adult men were more likely to report greater financial strain and coping self-efficacy than women, whereas older adult women were more likely to report higher levels of negative life events, home demands and responsibilities, emotional empathy, and health problems than men.

Hypothesis 2a: It was hypothesized that the association between family stressors and older adult health problems would be mediated through lowered social connectedness (family, friends, and active community participation).

Hypothesis 2b: It was hypothesized that the association between home demands and responsibilities and older adults’ health problems would be mediated through improved social connectedness (family, friends, and active community participation).

Hypothesis 3a: It was hypothesized that the association between financial strain, negative life events, and older adults’ health problems would be mediated through lower coping self-efficacy and lower emotional empathy.

Hypothesis 3b: It was hypothesized that the association between home demands and responsibilities and older adults’ health problems would be mediated through increased emotional empathy.

Hypothesis 4a: It was hypothesized that the association between financial strain, negative life events, and older adults’ health problems would be mediated through lower coping self-efficacy, lower emotional empathy, and lower social connectedness (family, friends, and active community participation).

Hypothesis 4b: It was hypothesized that the association between home demands and responsibilities and older adults’ health problems would be mediated through increased emotional empathy and social connectedness (family, friends, and active community participation). No pathway was hypothesized through coping self-efficacy.

It should be noted that the conceptual model linking ecological factors and older adult health problems was developed using propositions drawn from various theoretical frameworks and findings from the research literature. In addition, there is some piece-meal evidence of possible gender variations in associations among the key constructs. The association between financial strain and older adult health problems is likely to be stronger for men than women whereas the association between negative life events and older adult health problems is likely to be stronger for women than men.

Participants

The data for this study were drawn from a secondary data set titled “Project on Population, Social, Cultural, and Long-Term Care Surveillance for Thai Elderly People’s Health Promotion” collected in 2011 from older adults residing in Kanchanaburi province, Thailand (for additional details see http://www.ipsr.mahidol.ac.th/IPSR/Research/KanchanaburiProject/default.htm). Kanchanaburi province, located 129 km west of Bangkok, is Thailand’s third largest province and is representative of the sociodemographic changes occurring in Thai society. The sampling frame consisted of 100 village blocks located in urban or semi-urban, rice plantations, and mixed economy strata. The older adult population in Kanchanaburi province grew from 8.7% in 2000 to 12% in 2010, and the older adult dependency ratio increased from 13.4% in 2000 to 17.9% in 2010 (National Statistical Office of Thailand [NSO], 2000, 2010). The community has also experienced an increasing pattern of migration (10.2 per 100 in 2000 to 14.7 per 100 in 2005; Punpuing & Guest, 2005), which has continued through the last decade.

Twelve thousand households in Kanchanaburi that had at least one person 50 years or older were considered the sampling frame for the study. These participants were personally contacted by the interviewers to ascertain their interest in volunteering to participate in the study, and 10,665 participants agreed to participate (response rate = 88.9%). Individuals who fit the inclusion criteria for our study (i.e., 60 years or older with children and of Thai ethnicity) included 3,800 participants. Data were collected using face-to-face interviews. Prior to administering the survey, consent was obtained from the participants.

The sample for this investigation consisted of 3,800 Thai older adults (1,654 men and 2,146 women) with children¹ (M = 3.53, SD = 1.76; range = 1-15). The sociodemographic characteristics of the sample² (as indicated in Table 1) are representative of the older adult population in Thailand as described in scientific reports (e.g., Knodel & Chayovan, 2008).

Table 1.

Descriptive Characteristics of Participants by Gender (N = 3,800).

Variables	Women (n = 2,146)	Men (n = 1,654)
Buddhism	2,120 (98.8%)	1,628 (98.4%)
Relationship status
Married, living with spouse	1,027 (47.9%)	1,341 (81.1%)
Widowed	982 (45.8%)	229 (13.8%)
Divorced	17 (0.8%)	14 (0.8%)
Separated	100 (4.7%)	49 (3%)
Married, living separately	20 (0.9%)	21 (1.3%)
Read Thai	1,464 (68.2%)	1,448 (87.5%)
Employment status
Currently working	828 (38.6%)	951 (57.5%)
Looking for a job	1 (0.0%)	1 (0.1%)
Homemaker	192 (8.9%)	59 (3.6%)
Not working	1,125 (52.4%)	643 (38.9%)
Children	4.39 (2.16)	3.95 (2.02)
Children living in the same household	1,352 (63%)	935 (56.5%)

Measures

Items included in the survey were drawn from several available questionnaires and adapted for use within the Thai context. As the surveys were administered to Thai older adults whose native language was Thai, Thai faculty with research expertise working with older adult populations and fluent in both Thai and English translated all measures from English to Thai. All translated questions were then back translated and the English and the Thai surveys were evaluated to examine whether they retained their original meanings and intent. Prior to the commencement of the study, the questionnaire was pilot tested and concerns and suggestions were addressed and incorporated.

Family Stressors

Two types of family stressors were examined in this study. They included financial strain and negative family life events.

Financial strain

Financial strain was assessed using five items adapted from measures used by Krause et al. (1991) and Krause, Liang, and Gu (1998) among others. They included (a) household income not enough to meet family expenses each month; (b) did not have savings in multiple forms, for example, bonds, stocks, shares etc.; (c) had not saved enough for life; (d) family currently in debt; and (e) family debt was a burden. Participants indicated whether they had currently experienced each of these financial strains (1 = yes, 0 = no). The five items were summed to create an overall index of financial strain (0 = 554 [14.6%], 1 = 862 [22.7%], 2 = 711[18.7%], 3 = 915 [24.1%], 4 = 678 [17.8%], 5 = 80 [2.1%]).

Negative family life events

Negative family life events were assessed using nine statements drawn from multiple life event measures (e.g., Daily Inventory of Stressful Events [Almeida, 2005], Structured Life Events Inventory [Wethington, 1997]). Respondents were asked if they had experienced any of the following problems for more than 2 to 3 weeks over the last year (1 = yes, 0 = no). Items included such statements as (a) You, your spouse, children, grandchildren, siblings, or parents got sick, injured, or disabled; (b) You had a quarrel or fight that led to a broken relationship with your spouse, children, grandchildren, siblings, parents, or close relatives; (c) You did not keep contact with your children or had very limited contact; (d) Your children or grandchildren had problems at work or school such as losing a job and other difficulties at school or workplace; (e) Other critical problems happened to your children, grandchildren; (f) You or your spouse had problems at work or about making a living (losing a job or land for farming); (g) You had problems about housing such as house decay or problems with the landlord or neighbors; (h) You had problems when taking your close ones (spouse, children, grandchildren, sibling, parents, and even yourself) out for medical treatments (to see doctor/public health personnel); and (i) You had burdens looking after your spouse and others not including your grandchildren or great grandchildren. The items were summed to create an index of negative family life events (0 = 3,035 [79.9%], 1 = 533 [14.0%], 2 = 159 [4.2%], 3 or more = 73 [1.9%]).

Home Demands and Responsibilities

Participants were asked whether they had engaged in meeting certain home demands and responsibilities over the past 6 months—cooking, washing or ironing clothes, sweeping or mopping up, looking after children (0-15 years of age), looking after disabled or ill people lower than 60 years, and looking after older adults aged 60 years and over (1 = yes, 0 = no). The items were summed to create an index of home demands (0 = 274 [7.2%], 1 = 311 [8.2%], 2 = 439 [11.6%], 3 = 1,411 [37.1%], 4 = 1,104 [29.1%], 5 = 230 [6.1%], 6 = 31 [.8%]).

Coping Resources

Two types of coping resources were examined in this study. They included coping self-efficacy and emotional empathy.

Coping self-efficacy

Coping self-efficacy was assessed using three items that included statements about a person’s ability to control their emotions as well as their life problems (adapted from the Pearlin Mastery Scale; Pearlin & Schooler, 1978). They included (a) You can accept that some problems are difficult to be solved (when having a problem), (b) You are sure that you can control your emotions in emergencies or in very bad situations, and (c) You are confident to face very serious situations in life. Participants responded as to whether they had experienced these feelings over the last month on a 4-point response format (0 = no, 1 = a little, 2 = very much, 3 = extremely). The items were averaged to create a scale (M = 1.93, SD = 0.72). Higher levels indicated greater coping self-efficacy. The reliability of the scale was .83.

Emotional empathy

Emotional empathy was assessed using three items (adapted from the Empathic-Concern subscale of the Interpersonal Reactivity Index [IRI]; Davis, 1994). They included (a) You feel sympathy when seeing people in misery, (b) You feel happy with helping people in trouble, and (c) You help people when you have a chance to do so. Participants responded as to whether they had experienced these feelings over the last month on a 4-point response format (0 = no, 1 = a little, 2 = very much, 3 = extremely). The responses across items were averaged to create a scale for emotional empathy (M = 2.24, SD = 0.56). Higher levels indicated greater emotional empathy. The reliability of the scale was .80.

Social Connectedness

Social connectedness was conceptualized as older adults’ participation in community activities and their connectedness with family, friends, and neighbors. Items were drawn from multiple measures (e.g., Beckett et al., 2002). Three aspects of social connectedness were included in this investigation: (a) connectedness with family, (b) active community participation, and (c) connectedness with friends.

Connectedness with family was assessed using two items: (a) How often do you talk to your children face to face or via phone? and (b) How often do you talk to your relatives by face to face or phone or letter? Participants responded to both items on a 5-point response format (0 = not at all in the past 1 year, 1 = 2-3 times a year, 2 = at least once a month, 3 = at least once a month, 4 = everyday). Responses to both items were averaged to create a scale (M = 3.55, SD = 0.71). Active community participation was assessed by five items: In the past 6 months, did you join any social activities in your community such as (a) serve as a village health volunteer, (b) volunteer to look after the elderly, (c) volunteer to babysit children/look after the disabled, (d) volunteer for community development, (e) participate in other community activities. Participants responded to these items on a 3-point scale (1 = not at all, 2 = sometimes, 3 = always; M = 1.03, SD = 1.38). The reliability of the scale was .61. Connectedness with friends was assessed using a single item: How often do you meet or phone or write to your friends/neighbors? Participants indicated their responses on a 4-point scale (0 = not at all in the past year, 1= 2-3 times a year, 2 = at least once a month, 3 = at least once a month, 4 = everyday; M = 3.58 SD = 0.79). Higher levels indicated greater social connectedness with family, active community participation, and social connectedness with friends.

Older Adults’ Health Problems

Older adults’ health problems were assessed using two constructs: (a) health self-evaluation and (b) health conditions. The first construct termed health self-evaluation consisted of asking participants about their level of heath when compared with others of the same age (over the past month). Participants reported their evaluations on a 5-point scale (0 = very good, 1 = good, 2 = moderate, 3 = bad, 4 = very bad; M = 1.38; SD = 0.90). Participant ratings of their health status have been found to be significantly correlated with the presence of chronic diseases (e.g., Rakowski & Cryan, 1990). Health conditions were assessed using participants’ responses to whether or not they had one or more chronic illness or disabilities. Participants were presented with a list of 24 chronic illnesses and disabilities (e.g., Dementia or Alzheimer’s disease, diabetes, heart disease, high blood pressure). The presence of these illnesses were summed to create an index of health conditions (0 = 712 [18.7%]; 1 = 1,163 [30.6%]; 2 = 879 [23.1%]; 3 = 536 [14.1%]; 4 = 256 [6.7%]; 5 or more = 254 [6.7%]). The correlation between the health evaluation and health conditions was .29, p < .001 (d = .61).

Results

Prior to conducting the analyses, we examined the data for missing values. A check of the data indicated that less than .01% of the data were missing and that the data were missing at random. Multiple imputations were conducted to replace the missing values (Graham, Hofer, Donaldson, MacKinnon, & Schafer, 1997).

Mean Comparisons

To test Hypothesis 1, we conducted four one-way MANCOVAs (controlling for participants’ age and their ability to read and write in Thai).

Family Stressors and Home Demands

Results indicated gender differences in the overall model, F(3, 3794) = 64.31; p < .001, $η_{p}^{2}$ = .05; Wilks’ Lambda = .95. Examination of the ANOVAs for each of the dependent variables using a Bonferroni-adjusted alpha level of .017 indicated that there were differences by gender for home demands and responsibilities, F(1, 3796) = 192.27; p < .001, $η_{p}^{2}$ = .05. An inspection of the mean scores indicated that women reported higher levels of home demands and responsibilities than men (M = 3.19, SD = 1.19; M = 2.62, SD = 1.37).

Coping Resources

Results indicated gender differences in the overall model, F(2, 3795) = 16.14; p < .001, $η_{p}^{2}$ = .01; Wilks’ Lambda = .99. Examination of the ANOVAs for each of the dependent variables using a Bonferroni-adjusted alpha level of .03 indicated that there were differences by gender for coping self-efficacy, F(1, 3796) = 20.06; p < .001, $η_{p}^{2}$ = .01. An inspection of the mean scores indicated that men reported higher levels of coping self-efficacy than women (M = 2.00, SD = 0.68; M = 1.87, SD = 0.75).

Social Connectedness

Significant gender differences were revealed in the overall model, F(2, 3794) = 17.46; p < .001, $η_{p}^{2}$ = .01; Wilks’ Lambda = .99. Examination of the ANOVAs for each of the dependent variables using a Bonferroni-adjusted alpha level of .017 indicated that there were differences by gender for active community participation, F(1, 3796) = 18.41; p < .001, $η_{p}^{2}$ = .01; connectedness with family, F(1, 3796) = 23.69; p < .001, $η_{p}^{2}$ = .01; and connectedness with friends, F(1, 3796) = 6.55; p < .01, $η_{p}^{2}$ = .002. An inspection of the mean scores indicated that men reported higher levels of active community participation (M = 1.04, SD = 0.16; M = 1.02, SD = 0.11) and connectedness with friends or neighbors (M = 3.62, SD = 0.75; M = 3.55, SD = 0.82) than women. Women indicated higher levels of connectedness with family than men (M = 3.60, SD = 0.67; M = 3.48, SD = 0.75).

Older Adults’ Health Problems

Results indicated gender differences in the overall model, F(2, 3795) = 67.02; p < .001, $η_{p}^{2}$ = .03; Wilks’ Lambda = .97. Examination of the ANOVAs for each of the dependent variables using a Bonferroni-adjusted alpha level of .03 indicated that there were differences by gender for health self-evaluation, F(1, 3796) = 45.42; p < .001, $η_{p}^{2}$ = .02, and health conditions, F(1, 3796) = 109.44; p < .001, $η_{p}^{2}$ = .03. An inspection of the mean scores indicated that women reported higher levels of health self-evaluation (M = 1.48, SD = 0.89; M = 1.24, SD = 0.88) and health conditions (M = 2.07, SD = 1.59; M = 1.55, SD = 1.49) than men.

Hypothesis 1 was partially supported.

Test of Conceptual Model by Gender

Prior to testing Hypotheses 2a, 2b, 3a, 3b, 4a, and 4b, the skewness and kurtosis of all constructs were examined. Some of the scales did not meet the assumptions for multivariate normality and hence Blom’s transformation was conducted to transform the raw data and to reduce the effect of extreme values (Blom, 1958). To test structural invariance by gender, we set up an unconstrained model where the structural parameters were left to freely vary across the two groups against another model where the structural parameters were constrained to be equal across the two groups. If the chi-square difference statistic between the constrained and the unconstrained models indicated a significant difference, then the structural model was judged to be different for men and women. The conceptual model in Figure 1 was run controlling for participants’ age and their ability to read and write in Thai.

To examine model fit, we used a variety of fit indices (in addition to the chi-square statistic and the associated p value) such as the Goodness of Fit Index (GFI), the Adjusted Goodness of Fit Index (AGFI), and the Root Mean Square Error of Approximation (RMSEA; Hu & Bentler, 1999). Values of .95 and above for the GFI and AGFI indicated good model fit (Miles & Shevlin, 1998); RMSEA of .05 and below indicated good model fit (Kline, 2005). Tests of mediation were conducted using strategies suggested by Frazier, Tix, and Barron (2004). If the analyses indicated the presence of a mediational pathway, we utilized bootstrapping methods (Mallinckrodt, Abraham, Wei, & Russell, 2006) to test the significance of that pathway. Five thousand bootstrap samples were drawn with replacement from the study sample and the significance of the indirect effects was examined using 95% bias-corrected intervals.

Pathways Through Social Connectedness Constructs

The findings indicated that the structural pathways through social connectedness constructs varied by gender: unconstrained model, χ²(32, N = 3,800) = 306.29, p < .001; constrained model, χ²(50, N = 3,800) = 350.72, p < .001; difference between unconstrained and constrained models, χ²Δ(18, N = 3,800) = 44.43, p < .001. Hence, the models were run separately by gender.

Men

The model was significant for men, χ²(14, N = 1,654) = 79.50, p < .001, GFI = .99, AGFI = .96, RMSEA = .05. The model explained 14% of the variance in older adults’ health problems.

Three direct effects were found.³ Home demands and responsibilities (β = −.20, p < .001), financial strain (β = .08, p < .05), and negative life events (β = .19, p < .001) were directly associated with older adult health problems. Only one indirect pathway was statistically significant: Home demands and responsibilities were positively associated with social connectedness with friends (β = .06, p < .05), and social connectedness with friends was negatively associated with older adults’ health problems (β = −.11, p < .001). This indirect pathway was significant (p < .01).

Women

The model was significant for women, χ²(14, N = 3,800) = 96.29, p < .001, GFI = .99, AGFI = .97, RMSEA = .05. The model explained 16% of the variance in older adults’ health status.

Three direct effects were found. Home demands and responsibilities (β = −.20, p < .001), financial strain (β = .12, p < .001), and negative life events (β = .30, p < .001) were associated with older adult health problems. Three indirect pathways through social connectedness with friends were significant. The associations between (a) negative life events (β = −.05, p < .01), (b) home demands and responsibilities (β = .12, p < .001), (c) financial strain (β = −.04, p < .05) and social connectedness with friends were significant. Social connectedness with friends was negatively associated with older adult health problems (β = −.10, p < .001). The three indirect pathways were significant (p < .01, p < .01, p < .05, respectively). In addition, the associations between (a) negative life events (β = .10, p < .001) and (b) financial strain (β = −.05, p < .01) with active community participation were also significant. Active community participation was negatively associated with older adult health problems (β = −.08, p < .01). The two pathways were significant (p < .001, p < .01, respectively).

Hypotheses 2a and 2b were partially supported.

Test of the Overall Model by Gender

To test Hypotheses 3a, 3b, 4a, and 4b, we introduced the constructs of coping self-efficacy and emotional empathy (as seen in Figure 1). The significant pathways between family stressors, home demands and responsibilities, and older adult health problems through social connectedness constructs (as evidenced from the tests of Hypotheses 2a and 2b for men and women) were also included in the conceptual model. To improve model fit, we covaried the error terms between the two coping resources (coping self-efficacy and emotional empathy). The results are indicated in Table 2 and discussed below.

Table 2.

Pathways of Association Linking Family Stressors, Home Demands, and Older Adult Health Problems by Gender.

	Men			Women
	Beta	SE	β	Beta	SE	β
Financial strain → Coping self-efficacy	−.10	.02	−.11***	−.13	.02	−.12***
Home → EE	.05	.02	.05*	.07	.02	.07***
Negative life events → EE	−.03	.03	−.02	.03	.03	.02
Financial → EE	−.09	.02	−.09***	−.07	.02	−.07***
Negative life event → Coping self-efficacy	−.04	.03	−.03	−.06	.03	−.05*
Coping self-efficacy → SC Friends	.01	.02	.02	.03	.02	.04
Coping self-efficacy → SC Family	−.01	.03	−.01	.01	.02	.02
EE → SC Friends	.10	.02	.12***	.09	.02	.11***
Coping self-efficacy → ACP	.02	.02	.03	−.02	.01	−.04
EE → ACP	.03	.02	.04	.05	.01	.11***
EE → SC Family	.03	.03	.04	.01	.02	.02
Negative life events → SC Friends				−.06	.02	−.05**
Financial strain → SC Friends				−.03	.02	−.03
Home → SC Friends	.04	.02	.06*	.10	.02	.12***
Negative life events → ACP				.06	.01	.10***
Financial strain → ACP				−.02	.01	−.04*
Negative life events → Health problems	.17	.03	.20***	.20	.02	.30***
ACP → Health problems	−.06	.03	−.06	−.09	.04	−.08*
SC Friends → Health problems	−.09	.03	−.11***	−.05	.02	−.07*
SC Family → Health problems	.04	.02	.06	.02	.02	.03
Home → Health problems	−.13	.02	−.23***	−.11	.02	−.21***
EE → Health problems	.02	.02	.04	−.04	.02	−.08*
Financial strain → Health problems	.05	.02	.08*	.05	.02	.10***
Coping self-efficacy → Health problems	−.11	.03	−.18***	−.07	.02	−.15***
Health self-evaluation → Health problems	.72	.06	.45***	.98	.11	.54***
Health conditions → Health problems	1.00		.60***	1.00		.53***

Note. n = 1,654 men; n = 2,146 women. Health = older adult health problems; EE = emotional empathy; SC Friends = social connectedness with friends; SC Family = social connectedness with family; ACP = active community participation; Home = home demands and responsibilities; Beta = unstandardized beta; SE = Standard error; β = Standardized beta.

p < .05. **p < .01. ***p < .001.

Men

The model was significant for men, χ²(29, N = 3,800) = 128.78, p < .001, GFI = .99, AGFI = .97, RMSEA = .05. The structural model predicted 19% of the variance in older adults’ health problems. Financial strain, negative life events, and home demands and responsibilities were directly linked to older adults’ health problems even after the introduction of the mediating variables (β = .08, p < .05; β = .20, p < .001; β = −.23, p < .001, respectively).

The mediating role of coping self-efficacy was indicated by one pathway:

Financial strain was negatively associated with coping self-efficacy (β = −.11, p < .001); coping self-efficacy was negatively associated with older adults’ health problems (β = −.18, p < .001). Bootstrapping analyses indicated that this partially mediated pathway was significant (p < .001).

The mediating role of emotional empathy was indicated by two indirect effects (through social connectedness with friends):

Financial strain was associated with lowered emotional empathy (β = −.09, p < .001); emotional empathy was positively associated with social connectedness with friends (β = .12, p < .001); and social connectedness with friends was associated with older adult health problems (β = −.11, p < .001). Bootstrapping analyses indicated that this partially mediated pathway was significant (p < .001).

Home demands and responsibilities were positively associated with emotional empathy (β = .05, p < .05); emotional empathy was positively associated with social connectedness with friends (β = .12, p < .001); and social connectedness with friends was negatively associated with older adults health problems (β = −.11, p < .001). Bootstrapping analyses indicated that this partially mediated pathway was significant (p < .001).

Women

The model was significant for women, χ²(26, N = 3,800) = 141.25, p < .001, GFI = .99, AGFI = .97, RMSEA = .05. The structural model predicted 27% of the variance in older adults’ health problems. Financial strain, negative life events, and home demands and responsibilities were directly linked to older adult health problems even after the introduction of the mediating variables (β = .10, p < .001; β = .30, p < .001; β = −.21, p < .001, respectively).

The mediating role of coping self-efficacy was indicated by two pathways:

Financial strain was negatively associated with coping self-efficacy (β = −.12, p < .001); coping self-efficacy was negatively associated with older adult health problems (β = −.15, p < .001). Bootstrapping analyses indicated that this partially mediated pathway was significant (p < .001).

Negative life events were associated with coping self-efficacy (β = −.05, p < .05); coping self-efficacy was negatively associated with older adult health problems (β = −.15, p < .001). Bootstrapping analyses indicated that this partially mediated pathway was significant (p < .001).

The mediating role of emotional empathy was indicated by three pathways (through social connectedness with friends and active community participation):

Financial strain was negatively associated with emotional empathy (β = −.07, p < .001); emotional empathy was positively associated with social connectedness with friends (β = .11, p < .001); and social connectedness with friends was associated with older adult health problems (β = −.07, p < .001). Bootstrapping analyses indicated that this partially mediated pathway was significant (p < .001).

Home demands and responsibilities were positively associated with emotional empathy (β = .07, p < .001); emotional empathy was positively associated with social connectedness with friends (β = .11, p < .001); and social connectedness with friends was associated with older adult health problems (β = −.07, p < .05). Bootstrapping analyses indicated that this partially mediated pathway was significant (p < .05).

Home demands and responsibilities were associated with emotional empathy (β = .07, p < .001); emotional empathy was associated with active community participation (β = .11, p < .001); and active participation in community was associated with older adult health problems (β = −.08, p < .05). Bootstrapping analyses indicated that this partially mediated pathway was significant (p < .01).

Two additional partially mediated pathways through emotional empathy were also found:

Financial strain and home demands and responsibilities were associated with emotional empathy (β = −.07, p < .001;β = .07, p < .001 respectively). Emotional empathy was positively associated with older adult health problems (β = −.08, p < .05). Bootstrapping analyses indicated that the two partially mediated pathways were significant (p < .01; p < .01).

Hypotheses 3a, 3b, 4a, and 4b were partially supported.

Discussion

In this study, we pulled together information from several strands of research to develop a comprehensive conceptual model linking family stressors, home demands and responsibilities, coping resources, social connectedness, and older adult health problems. Findings indicated gender variations in the levels of these constructs as well as in the mediational pathways of association.

Despite the introduction of various health care governmental initiatives including expanding health care coverage for the elderly (e.g., introduction of the health care card and other health care schemes), chronic health problems and disability present increasing challenges to the health and well-being of Thai elderly (Haseen et al., 2010). Thai women in this study reported poorer self-assessed health evaluations and poorer overall health conditions (presence of chronic diseases) than men which is similar to findings in other studies with Thai older adults (e.g., Haseen et al., 2010; Knodel & Chayovan, 2008).

Thai older adult men reported higher levels of perceived coping self-efficacy, active community participation, and greater connectedness with friends than women whereas older adult women reported greater connectedness with their children and relatives than men (similar to findings by Beckett et al., 2002; Knodel & Chayovan, 2008). Like is the case in many traditional societies, the private sphere of life (e.g., family and kin) appears to be the purview of Thai women whereas the public sphere (e.g., interactions with non-kin and community) appears to be the purview of men. Our findings support this cultural notion. In addition, in the Thai culture and in Buddhist teachings, there is an expectation that one must give back to the community through teaching and community development. There is also an expectation that one must engage in charitable giving to the needy. Although these principles guide the lives of both men and women, men with their greater role in the public realm are more likely to take part in community activities (e.g., village health volunteers, community development) than women. The conceptualization of this construct needs to be further advanced in future studies given that older men and women in Thai society may play different roles in their communities. Our findings on the three aspects of social connectedness closely reflect the cultural interpretations of social relationships held by Thai older men and women.

Our findings also indicated that older adult women reported higher levels of home demands and responsibilities than men supporting the gendered distribution of household tasks in Thai society. Similar to studies conducted in other countries (e.g., Coltrane, 2000), women are primarily responsible for meeting the demands and responsibilities of the household in Thai families. Providing care for sick and older family members is the responsibility of the immediate family and extended family members rather than of formal service organizations (Boontawee, 1998; Chayovan, 1992). Filial obligations to one’s older parents and disabled members are expected of adult children within Thai cultural norms.

One of the strengths of this investigation was the inclusion of both multiple family stressors and home demands and responsibilities (within a single model) when examining the mediating effects of coping self-efficacy, emotional empathy, social connectedness with friends, family, and active community participation on older adult health problems. The direct effect of family stressors on older adult health problems was positive and provided support for the stress theory hypothesis (Thoits, 1995). On the other hand, findings from this investigation indicated that Thai older adults viewed home demands more as an expected responsibility of family members rather than as a stressor (based on notions of filial responsibilities and obligations). This was indicated by the negative link between home demands and health problems. Beyond these cultural explanations, this finding is also supported by the tenets of activity theory proposed by Havighurst and Albrecht (1953), which suggest that active involvement in daily activities is important for successful aging. Hence, men and women who actively participate in “household tasks” and meet their family responsibilities have more positive health outcomes compared to those who delegate these responsibilities to others. However, we also note that older men and women who actively participate in home activities may have better health to begin with and hence future studies using longitudinal designs should explore the directionality of the relationship between home demands and older adult health problems within the Thai context.

Emotional empathy appears to be the central variable that links financial strain, home demands and responsibilities, and older adult health status through social connectedness. For both men and women, emotional empathy or the ability to understand and respond to the needs of others is a critical coping resource that is negatively affected in situations of stress and enhanced when engaged in meeting home responsibilities. Individuals who are able to understand and respond caringly to other persons in need are likely to maintain social bonds and also develop new social connections. However, older adults’ experiences of financial strain may create preoccupations with their stressful situations and hence challenge their capacity to be compassionate to the feeling and sufferings of others. Lower ability to engage with others in a supportive and caring manner can harm social relationships and interactions in community settings (Hoffman, 2000). Friendship and community networks afford Thai older adults with opportunities for companionship and engagement which when challenged are associated with negative health outcomes. Social interactions help older adults to be more physically active and share information about health-related issues such as how to deal with health problems and address questions and concerns about medical care (Mongkolprasoet, 2002; Thanakwang & Soonthorndhada, 2011).

For older adult women alone, there was a mediated pathway between home demands and improved health outcomes through improved emotional empathy and active community participation. In addition to these interrelationships, emotional empathy was also a mediator of the relationship between home demands and responsibilities, financial strain, and health problems for Thai women.

The role of coping self-efficacy in this model was not exactly as we had hypothesized. We had expected that beyond the mediating role of coping self-efficacy in the relationship between family stressors and older adult health problems, coping self-efficacy would also be associated with older adults’ health status through the social connectedness constructs. This hypothesis was not supported. For both Thai men and women, financial strain (and for women, negative life events as well) was associated with an inability to cope effectively with the stressful situation. This was directly associated with increased older adults’ health problems (and not linked through any of the social connectedness constructs). The link between financial strain and older adult health status through coping self-efficacy was similar to the findings by Krause et al.’s (1991) work linking diminished personal control (control over events and carry out plans as expected) and psychological well-being among Japanese and American older adults. The finding also provides support for the work by Caplan (1981) on the stress process. The finding that negative life events was associated with poor coping self-efficacy and older adult health problems was supported only for older adult women and may be related to how men and women experience stressful personal life situations differently in Thai society and the centrality of family relationships for Thai women. It can be speculated that perhaps Thai men are better equipped to not place blame for the stressful negative life events on themselves and hence these stressful negative life events do not affect their ability to effectively cope with the situation. In addition, according to Buddhist philosophy, stress occurs as a result of a person’s inability to accept the reality of the situation (Ratanakul, 2004). Hence, being realistic about the stressor allows individuals to cope effectively with the situation surrounding the stressor.

Our study is not without its limitations. The cross-sectional nature of the data limited our ability to test the possibility of other hypothesized ordering of the variables. Although the conceptual model used in this investigation was based on both theory and information from the substantive literature, the need for collecting longitudinal data on these constructs is vital. Assessments at multiple time points will help us examine the issue of causality among the key constructs. There is a possibility of reverse causality in that health status could influence social variables as much as social variables could influence health status during the older years. We were also limited in this investigation by the use of a single informant. However, it appears that there was limited informant bias as only one set of error terms were required to be correlated to improve the fit of the model. Future investigations should include information from multiple respondents such as adult children and friends. The use of a secondary data set hindered the comprehensiveness with which some of the constructs were assessed. In particular, future studies would benefit from a deeper understanding of the role of social connectedness with family which did not play an important role in this conceptual model. As well, this study used older adult respondents self-reports of their health as a proxy for health status. Although there is support for the use of self-rated measures of health as a good predictor of morbidity and mortality among older adults (e.g., Fayers & Sprangers, 2002), nevertheless, future studies would be strengthened by the additional use of an objective measure of older adults’ health status.

One of the strengths of this study was that whereas most investigations linking life challenges and older adults’ health problems have mostly focused on the role of a single family challenge, this study included the role of multiple family stressors. The fact that many of the pathways hypothesized were supported by our findings is suggestive of the overall merit of the model to guide future research. This is the first test of this conceptual model and it would need to be replicated with other populations. However should the model be validated in future studies, it could offer policy makers a broader array of points at which to intervene and address the impact of various family stressors on older adults’ health status. A central policy focus could be on advancing the social connections among older adults through programs and services as well as increasing their participation in community activities which appears to be central to advancing the health of Thai older adults. In addition, given the rates of projected increase in the older adult population in Thailand, governmental assistance alone may not be enough to address the financial challenges facing many Thai older adults. Strategies to help older adults better handle financial matters as well as financial support for gainful employment (e.g., small business opportunities) for those able to contribute to their families may address some of the financial challenges facing the older adults. Findings from this investigation also point to the need to consider gender variations in studies with older adult populations. Our findings also have implications for public policy in terms of risk assessment of older adults’ health problems that could better inform the development of health care programs and thus promote the good health and care of the elderly. Efforts to have Thai women participate more in community activities would have to be factored into health promotion programs.

In conclusion, findings from this investigation point to the importance of examining the role of coping strategies in studies examining the relationship between family stressors, home demands and responsibilities, and health outcomes. As well, the significance of social relationships and the factors that advance and impede social connectedness are important in furthering our understanding of how family stressors affect the health outcomes of older men and women.

Footnotes

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: Primary funding for the project “Population, Economic, Social, Cultural and Long-Term Surveillance for Thai Elderly People’s Health Promotion” was provided by the Office of the Higher Education Commission, Thailand, in collaboration with Mahidol University to the Institute for Population and Social Research (IPSR), Mahidol University. Additional support was provided by the Falk College Research Center at Syracuse University to Ambika Krishnakumar and Lutchmie Narine.

Notes

References

Adams

K. B.

Leibbrandt

Moon

(2011). A critical review of the literature on social and leisure activity and wellbeing in later life. Aging & Society, 31, 683-712.

Almeida

D. M.

(2005). Resilience and vulnerability to daily stressors assessed via diary methods. Current Directions in Psychological Science, 14, 64-68.

Apinunmahakul

(2012). Financial and social capitals of elderly people in Thailand. Asian Social Science, 8, 147-156.

Artazcoz

Borrell

Benach

(2001). Gender inequalities in health among workers: The relation with family demands. Journal of Epidemiology & Community Health, 55, 639-647.

Bandura

(1997). Self-efficacy: The exercise of control. New York, NY: W.H. Freeman.

Beckett

Goldman

Weinstein

Lin

Chuang

(2002). Social environment, life challenge, and health among the elderly in Taiwan. Social Science & Medicine, 55, 191-209.

Berkman

L. F.

Glass

Brissette

Seeman

T. E.

(2000). From social integration to health: Durkheim in the new millennium. Social Science & Medicine, 51, 843-857.

Bianchi

S. M.

Milkie

M. A.

Sayer

L. C.

Robinson

J. P.

(2000). Is anyone doing the housework? Trends in the gender division of household labor. Social Forces, 79, 191-228.

Blom

(1958). Statistical estimates and transformed beta-variables. New York, NY: John Wiley.

10.

Boontawee

(1998). A child of the Northeast ( Kepner

S. F.

, Trans.). Bangkok, Thailand: Duangamol Press.

11.

Caplan

(1981). Mastery of stress: Psychosocial aspects. American Journal of Psychiatry, 138, 413-420.

12.

Carroll

Ring

Hunt

Ford

Macintyre

(2003). Blood pressure reactions to stress and the prediction of future blood pressure: Effects of sex, age, and socioeconomic position. Psychosomatic Medicine, 65, 1058-1064.

13.

Chayovan

(1992). Patronage family and lifecycle of the elderly and children. Bangkok, Thailand: Institute of Population Studies, Chulalongkorn University. (In Thai)

14.

Chesney

M. A.

Neilands

T. B.

Chambers

D. B.

Taylor

J. M.

Folkman

(2006). A validity and reliability study of the coping self-efficacy scale. British Journal of Health Psychology, 11, 421-437.

15.

Coltrane

(2000). Research on household labor: Modeling and measuring the social embeddedness of routine family work. Journal of Marriage and the Family, 62, 1208-1233.

16.

Davis

M. H.

(1994). Empathy: A social psychological approach. Madison, WI: Westview Press.

17.

Fayers

P. M.

Sprangers

M. A. G.

(2002). Understanding self-assessed health. The Lancet, 359, 187-188.

18.

Frazier

P. A.

Tix

A. P.

Barron

K. E.

(2004). Testing moderator and mediator effects in counseling psychology research. Journal of Counseling Psychology, 51, 115-134.

19.

Fry

P. S.

(1989). Mediators of stress in older adults: Conceptual and integrative frameworks. Canadian Psychology, 30, 636-649.

20.

Graham

J. W.

Hofer

S. M.

Donaldson

S. I.

MacKinnon

D. P.

Schafer

J. L.

(1997). Analysis with missing data in prevention research. In Bryant

Windle

West

(Eds.), The science of prevention: Methodological advances from alcohol and substance abuse research (pp. 326-366). Washington, DC: American Psychological Association.

21.

Haseen

Adhikari

Soonthorndhada

(2010). Self-assessed health among Thai elderly. BMC Geriatrics, 10(30). Retrieved from http://www.biomedcentral.com/content/pdf/1471-2318-10-30.pdf

22.

Havighurst

R. J.

Albrecht

R. E.

(1953). Older people. London, England: Longman.

23.

Hoffman

M. L.

(2000). Empathy and moral development: Implications for caring and justice. Cambridge, UK: Cambridge University Press.

24.

Bentler

P. M.

(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1-55.

25.

Jopp

D. S.

Schmitt

(2010). Dealing with negative life events: Differential effects of personal resources, coping strategies, and control beliefs. European Journal of Ageing, 7, 167-180.

26.

Kelly

Hertzman

Daniels

(1997). Searching for the biological pathways between stress and health. Annual Review of Public Health, 18, 437-462.

27.

Kline

R. B.

(2005). Principles and practice of Structural Equation Modeling (2nd ed.). New York, NY: Guilford Press.

28.

Knodel

Chayovan

(2008). Population ageing and the well-being of older persons in Thailand: Past trends, current situation and future challenges. Bangkok: UNFPA Thailand and Asia and the Pacific Regional Office.

29.

Krause

Jay

Liang

(1991). Financial strain and psychological well-being among the American and Japanese elderly. Psychology and Aging, 6, 170-181.

30.

Krause

Liang

(1998). Financial strain, received support, anticipated support, and depressive symptoms in the People’s Republic of China. Psychology and Aging, 13, 58-68.

31.

Kwag

K. H.

Martin

Russell

Franke

Kohut

(2011). The impact of perceived stress, social support, and home-based physical activity on mental health among older adults. International Journal of Aging & Human Development, 72, 137-154.

32.

Lazarus

R. S.

Folkman

(1984). Stress, appraisal, and coping. New York, NY: Springer.

33.

Mallinckrodt

Abraham

W. T.

Wei

Russell

D. W.

(2006). Advances in testing the statistical significance of mediation effects. Journal of Counseling Psychology, 53, 372-378.

34.

Miles

Shevlin

(1998). Effects of sample size, model specification and factor loadings on the GFI in confirmatory factor analysis. Personality and Individual Differences, 25, 85-90.

35.

Mongkolprasoet

(2002). Network development on self-health care cooperation of Wat Khamruang, Namuen District, Nan Province. Bangkok: Thailand Research Fund.

36.

National Statistical Office of Thailand. (2000). Retrieved from http://web.nso.go.th/en/census/poph/prelim_e.htm

37.

National Statistical Office of Thailand. (2010). Retrieved from http://www.nso.gov.mt/statdoc/document_file.aspx?id=3173

38.

O’Brien

Konrath

S. H.

Grühn

Hagen

A. L.

(2013). Empathic concern and perspective taking: Linear and quadratic effects of age across the adult life span. Journal of Gerontology: Psychological Sciences, 68, 168-175.

39.

Pearlin

L. I.

Schooler

(1978). The structure of coping. Journal of Health and Social Behavior, 19, 2-21.

40.

Punpuing

Guest

(2005). Migration report. In Guest

Jampaklay

(Eds.), The Report of Round 4 Census (2003) (Publication No. 304, pp. 85-90). Nakhon Pathom, Thailand: Institute for Population and Social Research, Mahidol University.

41.

Rakowski

Cryan

C. D.

(1990). Associations among health perceptions and health status within three age groups. Journal of Aging and Health, 2, 58-80.

42.

Ratanakul

(2004). The Buddhist concept of life, suffering and death, and related bioethical issues. Eubios Journal of Asian and International Bioethics, 14, 141-146.

43.

Sobieszczyk

Knodel

Chayovan

(2002). Gender and well-being among the elderly: Evidence from Thailand (PSC Research Report No. 02-531). Ann Arbor: Population Studies Center, University of Michigan.

44.

Soonthorndhada

Gray

Soonthorndhada

Vishwanathan

P. K.

(2008). Elderly women in Thailand: Roles and position. In Troisi

Pawiliczko

A. L.

(Eds.), The elderly women in Asia: Her roles and position (pp. 285-310). Malta: United Nations International Institute on Ageing.

45.

Sze

J. A.

Gyurak

Goodkind

M. S.

Levenson

R. W.

(2012). Greater emotional empathy and prosocial behavior in late life. Emotion, 12, 1129-1140.

46.

Thanakwang

Soonthorndhada

(2011). Mechanisms by which social support networks influence healthy aging among Thai community-dwelling elderly. Journal of Aging and Health, 23, 1352-1378.

47.

Thoits

P. A.

(1995). Stress, coping, and social support processes: Where are we? What next? Journal of Health and Social Behavior, 35, 53-79.

48.

Triandis

H. C.

(1994). Culture and social behavior. New York, NY: McGraw-Hill.

49.

United Nations Population Funds. (2011). The impact of demographic change in Thailand: Full report. Retrieved from http://thailand.unfpa.org/documents/symposium/Impact%20Full%20Report%20Eng%20UNFPA%20Web.pdf

50.

Wethington

(1997). Structured Life Events Inventory. In Zalaquett

C. P.

Wood

R. J.

(Eds.), Evaluating stress: A book of resources (pp. 391-403). Boston, MA: Scarecrow Press.

51.

World Economic Forum. (2013). The global gender gap report 2013. Retrieved from http://www3.weforum.org/docs/WEF_GenderGap_Report_2013.pdf