Life Satisfaction of the Older Thai: Findings from the Pilot HART1
Dararatt Anantanasuwong and Udomsak Seenprachawong
Thailand has become an aging society. In 2007, 11% or approximately 7 million people of the total population was aged 60 and older. In 2030, one-fourth of the population (approximately 17.7 million people) will be aged 60 and older. In preparing for the transition to an aging society, the Thai government has developed public policy and strategies to create national long-term plans for the elderly2 since 1986, and included language on the issue in the 2007 Constitution.3 Internationally, the Ministry of Social Development and Human Security has been authorized to take responsibility for protecting and improving the quality of life of the Thai elderly in compliance with the Madrid International Plan of Action on Ageing (MIPAA) 2002-2010. The center of this public commitment is the development of an aging society that advances health and well-being into old age and ensures an enabling and supportive environment.
In formulating plans and strategies under this commitment, the government and the relevant agencies have relied mainly on a national database on aging provided by the central survey agency, the National Statistical Office (NSO), i.e., a bi-annual national cross-section survey
____________
1 The authors acknowledge the kind support from the Research Development Committee of the National Institute of Development Administration in writing this chapter.
2 The First National Long-term Plan for the Thai Elderly, 1986-2001, and the Second Long-term Plan for the Thai Elderly, 2002-2021
3 Sections 30, 40, 53, 80, and 84 in the Constitution of Thailand 2007 include statements to protect and improve the quality of life of the Thai elderly.
on aging.4 The cross-section data from the household surveys have contributed significantly in understanding the current status and situation of the older population in Thailand, both by researchers and policymakers. However, to understand the aging process of the population is a life-course study that involves specialists in multi- and interdisciplinary fields, such as demography, epidemiology, health, psychology, economics, sociology, and survey methodology. An intensive database from each round of the survey and from the same sample households and individuals in each dimension will enhance knowledge about the process of aging biologically, psychologically, sociologically, and economically.
A national longitudinal study using panel data to formulate scientific knowledge on aging and to inform public policy, such as the Health and Retirement Study (HRS) conducted by the University of Michigan, has not been carried out nor has it been an interest of Thai researchers. Such a survey would be complicated, time-consuming, and costly to collect, maintain, and disclose the data, which may be the main reason preventing NSO or other research organizations in Thailand from conducting a large-scale longitudinal panel survey on the older population. Yet, longitudinal studies using panel data like the HRS have contributed significantly to advances in knowledge about demography, economics, sociology, and epidemiology of aging. Currently, HRS has become the pathbreaker for longitudinal and panel studies on aging around the globe5 (Hauser and Willis, 2004, 2011).
During 2006-2007, a group of researchers at the National Institute of Development Administration saw the value of a large-scale longitudinal study of aging like the HRS and attempted to establish a similar study in Thailand. A proposal for a pilot project, titled the Panel Survey and Study on Health, Aging, and Retirement in Thailand (HART), was developed in 2008 and received a one-year research grant from the National
____________
4 The national cross-section survey on aging was conducted bi-annually with a sample size of 79,500 households by interviewing each member aged 50 and older. The questionnaire is composed of various dimensions: demography, living conditions, employment and income, health and healthcare, social activities, information access, transfer and visit, knowledge in elderly care, and household asset ownership.
5 HRS’ contribution to the scientific knowledge of aging has influenced the development of large-scale longitudinal studies in many countries. These include the Mexican Health and Ageing Study (MHAS), English Longitudinal Study of Ageing (ELSA), Korean Longitudinal Study of Ageing (KLoSA), Japanese Study of Aging and Retirement (JSTAR), China Aging and Retirement Longitudinal Study (CHARLS), and Longitudinal Aging Study for India (LASI) (Hauser and Weir, 2011).
Research Council of Thailand (NRCT) in 2009.6 The pilot baseline survey of 1,500 household samples7 was conducted during August–October 2009 by interviewing, face-to-face, one member aged 45 and older from each household.8 In 2011, a second pilot HART project received a one-year research grant from the National Higher Education Commission (NHEC). A national-scale panel survey and study on aging in Thailand is expected to follow after the learning experiences from the pilot HART projects are digested.9
This chapter is one of several research studies10 using the baseline data collected from the pilot survey. It focuses on the life satisfaction dimension of the older Thai.11 Life satisfaction has a long research history. A considerable amount of research has been devoted to explain life satisfaction of the older population (Elwell and Maltbie-Crannell, 1981; George, Okun, and Landerman, 1985; Liang et al., 1980; McClelland, 1982; Mutran and Reitzes, 1981). Among the socioeconomic characteristics, several factors have been pointed out as explanatory variables of life satisfaction: age, gender, income, educational level, physical status, emotional health, social support, and locus of control. Many studies have reported a negative relationship between happiness and age (Campbell, 1981; Ferring et al., 2004). However, a significant positive relationship was reported by several other studies (Jason and Mueller, 1983; Witt et al.,
____________
6 The pilot HART proposal was mentored by Dr. James P. Smith from RAND and received personal support from Dr. Richard M. Suzman from the U.S. National Institute on Aging (NIA). The pilot HART proposal was submitted for research funding from NIA and from NRCT. With the funding approval from NRCT, the NIA funding application has been withdrawn. However, Dr. Smith kindly agreed to mentor the pilot project as an international advisor. The survey instrument was an adoption of that of KLoSA with some adjustments to fit local conditions. Because of funding constraints, paper questionnaires, not computer-assisted personal interviewing, were used for conducting face-to-face interviews.
7 The household samples were drawn from the NSO sampling frame from the “2004 Household Surveys in Bangkok, Nonthaburi, Pathumthani, Samuthprakarn, and Khonkaen” by using a two-stage stratified sampling method.
8 The individual respondents were chosen by the following rule: first from the eligible head of the household; if the head of the household is not eligible, then the most eligible volunteer household members. The project acknowledges the possibility of sampling bias in the individual sampling process in the pilot survey, which can raise concerns about the representativeness of the data. However, the learning experience from the pilot survey will lead to the quality improvement of the future national-scale panel survey.
9 It is hopeful that NRCT will grant research funding for the national longitudinal study on aging with the baseline survey in 2012
10 The researchers from the pilot HART project are developing four papers for publication, including this one, using the baseline data from the pilot survey. The other papers will concern aging of the population in Thailand, family structure and intergenerational transfers, and financial and social capital of the Thai elderly.
11 As a leading study on life satisfaction of the aging in Thailand based on data from the pilot HART project, the authors include the data from respondents aged 45 and older.
1980). Delhey (2004) found only a slight effect of income on life satisfaction. Diener (1984, 2000) found no relationship between income and life satisfaction. Nonetheless, Veenhoven (2000) related income to life satisfaction at the national level and found a larger effect of income, and the same positive relationship was found by Inglehart (1997) at a country level. A study done by Fernandez-Ballesteros, Zammaron, and Ruiz (2001) found moderate positive effects of socioeconomic variables, including income and education, on life satisfaction. With regard to physical health, a substantial body of evidence positively relates health with life satisfaction (Fernandez-Ballesteros, Zammaron, and Ruiz, 2001; Lehr, 1982; Mannell and Dupuis, 1996).
In sum, life satisfaction has been considered the subjective expression of quality of life (Abu-Bader, Rogers, and Barusch, 2002; Fernandez-Ballesteros, Zammaron, and Ruiz, 2001). There is a link between advancing age and decreased subjective well-being in social, physical, and economic domains (Dolyer and Forehand, 1984; George et al., 1985; Jason and Mueller, 1983). In other words, the major life events experienced in the aging process are expected to have a profound impact on the aging population’s life satisfaction. As people get older, life satisfaction may decrease sharply year by year as they are affected by major life events such as retirement and physical health deterioration. Therefore, an examination of differences in the aging population’s life satisfaction becomes critical for social welfare programs. This is especially relevant for Thailand as an aging society. Subgroups of the population perceive satisfaction with life domains differently. Individuals have different lifestyles and behaviors, which determine their perceptions of their satisfaction with life domains. In order to design adequate welfare programs, policies, and regulations for the aging population, an understanding of the relationships between people’s individual characteristics and their perceptions of satisfaction with life is needed.
With the availability of data on life satisfaction from the pilot HART survey, this chapter is an attempt to lay the groundwork for a study on life satisfaction of the older Thai. Its main purpose is to explain life satisfaction of older Thai respondents in three respects: (1) whether differences in life satisfaction in general exist between those in the urban and rural areas in Bangkok and vicinity and in Khonkaen, (2) which factors are associated with life satisfaction in general, and (3) the degree of association of specific life satisfaction domains with life satisfaction in general.
The chapter is organized into five parts. First, the significance of the longitudinal study on aging and the life satisfaction of the older population are introduced. We then elaborate on the characteristics of older Thai respondents from the pilot HART baseline data and describe the differences in life satisfaction in different domains between respondents
in urban and rural areas. We next present the analytical models used, followed by an explanation of the empirical findings. The chapter closes with our conclusion.
BASELINE DATA AND LIFE SATISFACTION DOMAINS OF OLDER THAI RESPONDENTS
In the 2009 pilot baseline survey of the HART project, the collected data consisted of seven dimensions: population characteristics, family and transfer, health, employment, income and expenditures, assets and debts, and life expectancy and life satisfaction.12 Life satisfaction perception and other relevant data were selected from the rich database according to the scope of the study. This present analysis is, thus, necessary cross-sectional because only one series of the survey has been completed.
The socioeconomic factors selected include the respondent’s age, marital status, gender, educational level, membership in social clubs, employment status, and household characteristics such as income and assets. The data on perceived life satisfaction consist of five domains: physical health, economic status, relationship with children, relationship with spouse, and life in general. The sample size drawn from the database is 1,468 observations from Bangkok and its vicinity and Khonkaen province.13
The descriptive statistics for the 2009 survey are presented in Table 18-1 through Table 18-4. Of the total, 66.7% of the respondents were female and 33.3% were male. Sixty-three percent of the respondents reported having no income from work, while 25% of those who work reported incomes of less than 100,000 baht per year.14 The majority of respondents (70%) had a primary level of education. More than half of the respondents were between 45 and 59 years old.
Table 18-5 summarizes responses obtained from the urban and rural samples for the domains of health, income, and family life. The last panel presents responses to the question regarding life satisfaction in general. For all domains and for general satisfaction, the distributions in the urban and rural respondents are significantly different (see the Chi-squared tests reported at the bottom of each panel).
Before turning to between-location differences, it is useful to highlight some differences across the domains of life. Both urban and rural people appear to be more unsatisfied with their spousal relationship
____________
12 A homepage on the HART project with the data archive is under construction. The baseline data from the pilot project will be accessible by the public when construction is completed.
13 The complete data from the 1,500 respondents (observations) resulted in 1,468 after all the missing values were rejected.
14 Approximately equivalent to US$3,330 (US$1 = 30 Baht).
TABLE 18-1 Distribution of Respondents by Province and Gender (in percentage)
Province | Female | Male | Total |
Bangkok | 20.7 | 9.7 | 30.4 |
Vicinity | 18.1 | 9.4 | 27.5 |
Khonkaen | 27.9 | 14.2 | 42.1 |
Total | 66.7 | 33.3 | 100.0 |
TABLE 18-2 Distribution of Respondents by Province and Age Group (in percentage)
Age | |||||
Province | 45-59 | 60-69 | 70-79 | Over 80 | Total |
Bangkok | 15.3 | 8.8 | 5.0 | 1.4 | 30.4 |
Vicinity | 15.1 | 7.7 | 3.9 | 0.8 | 27.5 |
Khonkaen | 21.5 | 10.8 | 6.9 | 2.8 | 42.0 |
Total | 51.8 | 27.3 | 15.8 | 5.0 | 100.0 |
than with other domains. Approximately 30% of respondents are not happy with their relationship with their spouses. The economic status domain is next, followed by the physical health domain. Finally, a small proportion of both urban and rural respondents appear to be dissatisfied with their relationship with their children.
We first considered how satisfied respondents in the two locations are with their physical health. Rural respondents (65%) tend to be more satisfied with their physical health than urban respondents (57%). Similarly, a larger fraction of rural respondents (46%) said that they are satisfied with their economic status compared to 34% among urban respondents. A slightly larger fraction of rural respondents (63%) are happy with their relationship with their spouse, compared to 59% among urban respondents. Regarding relationship with their children, rural respondents (80%) appear to be more satisfied than urban respondents (76%). The majority of both urban and rural respondents appear to be satisfied with their lives when answering a question about life satisfaction in general. However, rural respondents (80%) are much more satisfied with their lives in general than urban respondents (69%).
TABLE 18-3 Distribution of Respondents by Province and Income from Work (in percentage)
Province | No Income from Work | <100,000 baht/year | 100,000-200,000 baht/year | 200,001-300,000 baht / year | 300,001-400,000 baht/year | 400,001-500,000 baht / year | More than 500,000 baht/year | Total |
Bangkok | 20.3 | 6.3 | 2.7 | 0.7 | 0.1 | 0.1 | 0.3 | 30.4 |
Vicinity | 16.8 | 6.7 | 2.7 | 0.5 | 0.3 | 0.4 | 0.1 | 27.5 |
Khonkaen | 26.4 | 12.1 | 2.2 | 0.7 | 0.4 | 0.2 | 0.1 | 42.0 |
Total | 63.4 | 25.1 | 7.5 | 2.0 | 0.9 | 0.7 | 0.5 | 100.0 |
TABLE 18-4 Distribution of Respondents by Province and Education (in percentage)
Province | No Schooling | Primary | Secondary | High School | Diploma | Undergraduate | Graduate | Doctoral | Other | Total |
Bangkok | 2.5 | 19.7 | 2.1 | 3.0 | 0.5 | 1.2 | 0.1 | 0.0 | 1.4 | 30.4 |
Vicinity | 1.3 | 18.0 | 2.5 | 2.2 | 0.6 | 1.3 | 0.0 | 0.0 | 1.6 | 27.5 |
Khonkaen | 1.7 | 32.2 | 2.2 | 2.3 | 0.8 | 2.1 | 0.2 | 0.1 | 0.4 | 42.0 |
Total | 5.4 | 69.9 | 6.9 | 7.5 | 1.9 | 4.6 | 0.3 | 0.1 | 3.5 | 100.0 |
TABLE 18-5 Perceived Satisfaction with Domain of Life (in percentage)
Rural | Urban | |
1. How satisfied are you with your physical health? | ||
Very dissatisfied | 5.89 | 6.26 |
Not satisfied | 9.86 | 10.92 |
Neither satisfied nor not satisfied | 18.31 | 24.89 |
Satisfied | 27.14 | 39.59 |
Very satisfied | 38.80 | 18.34 |
Pearson Chi-Square = | 77.3884 | p-value = 0.0000 |
2. How satisfied are you with your economic status? | ||
Very dissatisfied | 7.94 | 6.84 |
Not satisfied | 13.96 | 18.20 |
Neither satisfied nor not satisfied | 32.01 | 40.32 |
Satisfied | 23.43 | 24.75 |
Very satisfied | 22.66 | 9.90 |
Pearson Chi-Square = | 47.6906 | p-value = 0.0000 |
3. How satisfied are you with your spouse relationship? | ||
Very dissatisfied | 24.71 | 28.53 |
Not satisfied | 3.71 | 2.04 |
Neither satisfied nor not satisfied | 7.94 | 10.04 |
Satisfied | 16.39 | 27.80 |
Very satisfied | 47.25 | 31.59 |
Pearson Chi-Square = | 51.6912 | p-value = 0.0000 |
4. How satisfied are you with your children relationship? | ||
Very dissatisfied | 6.66 | 9.61 |
Not satisfied | 1.92 | 2.18 |
Neither satisfied nor not satisfied | 10.50 | 11.50 |
Satisfied | 20.23 | 35.66 |
Very satisfied | 60.69 | 41.05 |
Pearson Chi-Square = | 63.5018 | p-value = 0.0000 |
5. How satisfied are you with your life in general? | ||
Very dissatisfied | 2.18 | 2.33 |
Not satisfied | 3.33 | 4.08 |
Neither satisfied nor not satisfied | 13.70 | 24.45 |
Satisfied | 26.76 | 42.94 |
Very satisfied | 54.03 | 26.20 |
Pearson Chi-Square = | 120.0655 | p-value = 0.0000 |
EMPIRICAL FINDINGS
In order to understand how socioeconomic and demographic factors are associated with life satisfaction in general, we employed an ordered probit estimation method. In this analysis, the dependent variable is the ranking the respondent gave for his perceived life satisfaction in general. The respondent was asked to rate life satisfaction in general: very dissatisfied (y = 0); not satisfied (y = 1); neither satisfied nor dissatisfied (y = 2); satisfied (y = 3); and very satisfied (y = 4). The respondents rate the level of life satisfaction as a discrete ordinal number (0, 1, 2, 3, 4). In addition, standardized regression coefficients were computed from a multiple regression method in order to investigate the degree of influence each domain of life has on life satisfaction in general.
The results of the factors associated with life satisfaction in general of the older Thai respondents are summarized in Table 18-6. All statistically significant coefficients, based on a two-tailed test at the 0.05 significant levels, are marked. As indicated in Table 18-6, level of education (EDU), age (AGE), income from work (INCOME), ownership of house (OWN), and gender (MALE) are related to level of satisfaction with life in general. The higher a respondent’s level of education, the happier with quality of life he or she indicates. Older respondents tend to be happier with their life in general, and high-income respondents tend to be happier with their lives than the low-income respondents. Similarly, the respondents who are house owners indicate higher satisfaction with their life than the ones who are not. Regarding the gender variable, male respondents appear to be more satisfied with their lives than their female counterparts. Respondents who are actively involved in social clubs (MEMBER) and meet with friends (FRIEND) tend to be happier with their lives compared to those who are not. In addition, the respondents who are in urban areas (URBAN) appear to be happier with their life in general than those who live in rural areas.15
The investigation of the degree of association of each domain of life on life satisfaction in general is shown in Table 18-7.16 As expected, the results show that satisfaction with life in general is positively associated with satisfaction within each of the four domains. Life satisfaction in general is a function of physical health (standardized beta = 0.322), economic status (standardized beta = 0.239), relationship with spouse (standardized beta = 0.115), and relationship with children (standardized beta = 0.238). The results indicate that physical health has the strongest association with life satisfaction in general. Relationship with spouse is least associ-
____________
15 This result, which was tested statistically using an econometric model, is different from the descriptive result described earlier.
16 The respondents who have no children were excluded from the analysis.
TABLE 18-6 Satisfaction with Life in General
Coefficient | t-ratio | P-value | |
SINGLE | -0.1602 | -1.1047 | 0.2693 |
EDU | 0.1802 | 7.3310 | 0.0000 |
FRIEND | 0.3988 | 5.6000 | 0.0000 |
MEMBER | 0.4270 | 5.1650 | 0.0000 |
OWN | 0.3697 | 5.9970 | 0.0000 |
WORK | 0.8677 | 12.3837 | 0.0000 |
AGE | 0.0939 | 3.1335 | 0.0017 |
INCOME | 0.0825 | 2.5014 | 0.0124 |
MALE | 0.1260 | 2.0627 | 0.0391 |
URBAN | 0.1590 | 2.3419 | 0.0192 |
TABLE 18-7 Ranking of Four Domains of Life Affecting Life Satisfaction
Domains | Standardized Coefficient | t-ratio | P-value |
(Constant) | 15.401 | 0.0000 | |
Physical health | 0.322 | 13.430 | 0.0000 |
Economic status | 0.239 | 10.111 | 0.0000 |
Relationship with spouse | 0.115 | 5.089 | 0.0000 |
Relationship with children | 0.238 | 10.433 | 0.0000 |
Dependent variable: Life satisfaction | |||
ated with life satisfaction. A 1-point change in perceived physical health satisfaction is associated with increased life satisfaction by 0.322 point, whereas a 1-point change in perceived spouse relationship satisfaction is associated with increased life satisfaction by only 0.115 point, holding other things constant. Economic status and relationship with children are moderately related to life satisfaction.
CONCLUSIONS
Employing the baseline data from the pilot HART project, this study focused on three main questions about life satisfaction of Thai respon-
dents aged 45 and older in Bangkok and the vicinity and Khonkaen. We looked at whether perceived life satisfaction of respondents in the rural and urban areas is different, what factors are related to their life satisfaction in general, and the degree of association of specific domains of satisfaction with their life satisfaction in general. The results indicate that the perceived life satisfaction of respondents in rural and urban areas in five domains—physical health, economic status, spouse relationship, children relationship, and life in general—are significantly different. Level of education, age, income from work, ownership of house, and being male are positively and significantly associated with life satisfaction in general. Other factors that are significantly associated with life satisfaction in general include being involved in social activities, meeting friends, and living in an urban area. The degree of association of each domain with life satisfaction in general indicates that physical health has the strongest association with life satisfaction in general, while relationship with spouse is least associated.
The study indicates clearly that the aging process in Thailand is just beginning. Older Thai respondents perceived their life satisfaction positively or optimistically. However, the finding that the highest proportion of them, both in rural and urban areas, are not satisfied with their relationship with their spouse implies that the marriage dimension of the older Thai may become an important policy issue to address.
Related to the factors associated with life satisfaction, building a good environment with facilities where the Thai elderly can participate in social activities and meet friends, creating economic security (e.g., financial literacy, employment, and asset ownership), and gaining and maintaining good physical health are also important public policies for the aging society.
REFERENCES
Abu-Bader, S.H., A. Rogers, and A.S. Barusch. (2002). Predictors of life satisfaction in frail elderly. Journal of Gerontological Social Work 38(3):3-17.
Campbell, A. (1981). The Sense of Well-Being in America. New York: McGraw-Hill.
Delhey, J. (2004). Life Satisfaction in an Enlarged Europe. Luxembourg: Office for Official Publications of the European Communities.
Diener, E. (1984). Subjective well being. Psychology Bulletin 95:542-575.
Diener, E. (2000). Subjective well being. American Psychologist 55:34-43.
Dolyer, D., and M.J. Forehand. (1984). Life satisfaction and old age. Research on Aging 6:432-448.
Elwell, F., and A.D. Maltbie-Crannell. (1981). The impact of role loss upon coping resources and life satisfaction of the elderly. Journal of Gerontology 36:223-232.
Fernandez-Ballesteros, R., M.D. Zammaron, and M.A. Ruiz. (2001). The contribution of socio-demographic and psychological factors to life satisfaction. Aging Society 22:25-43.
Ferring, F.D., C. Balducci, V. Burholt, C. Wenger, F. Thissen, G. Weber, and I. Hallberg. (2004). Life satisfaction of older people in six European countries: Findings from the European Study on Adult Well-being. European Journal of Ageing 1:15-25.
George, L.K., M.O. Okun, and R. Landerman. (1985). Age as a moderator of the determinants of life satisfaction. Research on Aging 7:209-233.
Hauser, R.M., and D. Weir. (2011). Longitudinal studies of aging in the United States. EURAMERICA 41(1):87-179.
Hauser, R.M., and R.J. Willis. (2004). Survey design and methodology in the Health and Retirement Study and the Wisconsin Longitudinal Study. Pp. 209-235 in Aging, Health, and Public Policy: Demographic and Economic Perspectives, L.J. Waite (Ed.). New York: The Population Council.
Inglehart, R. (1997). Modernization and Postmodernization: Cultural, Economic, and Political Change in Societies. Princeton, NJ: Princeton University Press.
Jason, P., and K.F. Mueller. (1983). Age, ethnicity and well-being: A comparative study of Anglos, Blacks and Mexican Americans. Research on Aging 5:353-368.
Lehr, U. (1982). Socio-psychological correlates of longevity. Annual Review of Gerontology and Geriatrics 3:102-147.
Liang, J., L. Dvorkin, E. Kahana, and F. Mazian. (1980). Social integration and morale: A re-examination. Journal of Gerontology 35:746-757.
Mannell, R.C., and S. Dupuis. (1996). Life satisfaction. In Encyclopedia of Gerontology, Second Edition, J.E. Birren (Ed.). San Diego, CA: Pergamon Press.
McClelland, K.A. (1982). Self-conception and life satisfaction: Integrating and subculture and activity theory. Journal of Gerontology 37:723-732.
Mutran, E., and D.C. Reitzes. (1981). Retirement, identity and well-being: Realignment of role relationships. Journal of Gerontology 36:741-749.
Panel Survey and Study on Health, Aging, and Retirement in Thailand. (2009). Bangkok: National Research Council of Thailand.
Veenhoven, R. (2000). The four qualities of life: Ordering concepts and measures of the good life. Journal of Happiness Studies 1:1-39.
Witt, D.P., G.D. Lowe, C.W. Peek, and E. Curry. (1980). The changing association between age and happiness: Emerging trend or methodological artifact. Social Forces 58:1,302-1,307.
This page is intentionally left blank.