The quantitative data analyzed in this dissertation and participants of qualita-tive interviews were a part of a Dr. Tony Szu-Hsien Lee’s project entitled ‘Gender differences and influence of walking exercise and a psychosocial intervention on psychological well-being and sense of coherence amongst retired senior citizens’.
The project was funded by National Science Council, ROC. (Grant No.
99-2410-H-003-127-MY2) and aim to investigate the effect of walking exercise and psychosocial intervention on physical activity, sense of coherence and happiness among retired elderly and its gender difference. It is a two-year true-experimental study. Constructing a reliable and valid instrument in the first year; and then discuss the influence of 3-month psychosocial and group walking intervention in the second year which applied on three group: (1) psychosocial intervention and group walking;
(2) group walking and (3) individual walking. The process and result of this project could be refer to the project report of National Science Council of Dr. Tony Szu-Hsien Lee.
Combining qualitative and quantitative research methods were used in this dissertation. First step was to use quantitative approaches to examine the construct of happiness. A face-to-face interview based on a structured questionnaire to investigate the dimensions of Chinese Happiness Inventory and examine the relationship be-tween happiness and related factors. Three steps were used to investigate the concept of happiness. First, a pilot was conducted to detect the dimensions of the Chinese
Happiness Inventory and its criterion related validity and reliability among retired el-derly. Secondly, preliminary survey was carried to examine the influencing factor of happiness and relationship among happiness and perceived health status, sense of co-herence, family relationship. Finally, a qualitative interview was used to detect the process of happiness broadening and its influencing factors among participants who received 3-month psychosocial intervention with significant increase of happiness scores.
Study 1: The participants were recruited from social service centers and public parks from September to November 2010. A face-to-face interview based on a struc-tured questionnaire was used to collect data. The questionnaires included demo-graphics, family relationships scale, Chinese Happiness Inventory and the Ryff PWB Scales.
Study 2: A preliminary face-to-face interview based on a structured question-naire was conducted. The participants were recruited from social service centers and public parks from Taipei city and New Taipei City, Taiwan. The questionnaire in-cluded demographics,13-sense of coherence, family relationship scale and Chinese Happiness Inventory.
Study 3: A retrospective in-depth interview was used to interview 6-8 partici-pants with significantly increase of happiness on post test of psychosocial intervention with group walking. The purpose of in-depth interview was to under-stand the process of broaden resources and build their happiness. Using qualitative interview to explain the sequence of broaden and build process among happiness on perceived health status.
3.1.1 Participants
The participants were recruited from social service centers. The original sample consisted of retired elders in Taipei city, Taiwan who met the following criteria: (1) retired from full-time work, (2) age 50 to 75 years, and (3) able to communicate ver-bally with the interviewer. Retirement age was defined as the age at which one stops working and earning money. The exclusion criteria including: (1) with any psycho-logical illness (2) with any infectious disease (3) without any physical problems who can not walk independently (4) has no regular habit of exercise at least 3 times a week, 30 minutes per time in the past half a year.
3.1.1.1 Sampling
Study 1: The participants were recruited from social service centers and public parks from September to November 2010 in Taipei city, Taiwan.
Study 2: A preliminary survey was the secondary data analysis from the study of Tony Szu-Hsien Lee funded by National Science Council R.O.C. (Grant No.
99-2410-H-003-127-MY2). The study '' Gender differences and influence of walking exercise and a psychosocial intervention on psychological well-being and sense of coherence amongst retired senior citizens'' was an true-experimental design with purposive sampling and random assigned. The retired elders were choose from Taipei city and New Taipei City, Taiwan with purposive sampling. Taipei city is divided up into 12 administrative distracts, which have apparent demographic-economical difference. To exclude the factor of social-economic difference within each administrative distract, we ranked 12 districts into four categories by ranking of annual household income from statistical analysis report of 2009. The highest category including Daan district (1,915,450 NT.), Songshan district (1,808,327 NT),
Neihu district (1,705,826 NT.), then the second higher category including Xinyi district (1,704,768 NT.), Zhongzheng district (1,699,904NT.), Wenshan district (1,635,584 NT), otherwise, the lower categories including Shilin district (1,587,197 NT.), Zhongshan district (1,566,328 NT.), Beitou district (1,552,111 NT.), and the lowest category including Nangang district (1,433,834 NT.), Datong district (1,413,700NT.), Wanhua district (1,276,492 NT.). We choose one district from each category as follows: Daan, Zhongzheng, Shilin and Wanhua districts. Otherwise, we choose Yungho district from New Taipei city because of high population density and similar location.
Study 3: A qualitative interview was conducted from the participants of ex-perimental group receiving psychosocial intervention with group walking (Appendix 3,4,5) which increase happiness status significantly. The participants from the study
of Dr. Lee were purposive sampling from social service center and random assigned into two groups: (1) experimental group: psychosocial intervention with walking ex-ercise, (2) control group: walking exercise in individuals. After examining the effectiveness of happiness state and cardiopulmonary fitness after three-month inter-ventions to support the idea that participants who received psychosocial intervention to broaden resources can increase happiness. A retrospective in-depth interview was used to interview 6-8 participants with significantly increase of happiness on post test after 3-month psychosocial intervention with group walking.
3.1.2 Site and setting
The pilot study was conducted in social service center or public parks of Taipei city. The preliminarily study and the final study was conducted in Taipei.
Taipei consists of Taipei city and New Taipei city, Taipei city is divided up into 12 administrative distracts, we choosed Daan, Zhongzheng, Shilin and Wanhua districts from 12 administrative distracts and Yungho distracts from New Taipei city accord-ing to the rankaccord-ing of annual household income and geographic location.
The explanation presentation and constructive survey were presented in con-vention center in local district. Both concon-vention center are quiet and comfortable for participants to fill out the questionnaire and interview. Moreover, booklet and pen are essential for participants to record the process of walking intervention. The question-naire test should be stopped if a person deny o refused to answer. A voice recorder is essential to record the process of in-depth interview, quiet and safety environmental, table and chair for interview, note writing tools to process the interview and survey are essential.
3.1.3 Instrumentation
The questionnaires included demographics, scale of family relationships, the Ryff PWB Scales, Chinese Happiness Inventory, and Scale of sense of coherence.
The demographics were gender, age, marital status, education level, perceived health status, and economic status. Responses for age, gender, marital status, and educa-tional level were coded by the interviewer. Perceived health status was assessed by asking, “How does your health affect your daily life?” Scores on this item ranged from 1 (often) to 3 (not at all). Economic status in the past 6 months was assessed by asking, “Do you have enough money for daily expenses?” Answers were enough or not enough. The Family Relationships measure was constructed from a three-item scale: (1) “How much are your significant family members involved in your daily
life?” (involvement), (2) “How are your emotional relationships with your close fam-ily members?” (emotional relation ), and (3) “What is the atmosphere of your family interactions?” (sociability). Responses were made on a 4-point Likert scale: none or bad (1), sometimes or fair (2), often or good (3), and always or great (4). Possible scores ranged from 3 to 12, with a higher score indicating more positive family rela-tionships.
Psychological well-being
We used the short version of the Ryff PWB Scales, which has six subscales with 14 items per subscale, seven worded positively and seven negatively. The Scale thus has 84 items. The short version Scale has been shown by Ryff (1995) to have good validity and reliability (α = 0.83 to 0.91), and correlations with the long form ranged from 0.97 to 0.99. Participants respond to each item using a 6-point format: strongly disagree (1), moderately disagree (2), slightly disagree (3), slightly agree (4), moder-ately agree (5), and strongly agree (6). Items with negative content were reverse scored and the 14 items summed into a total score for each subscale. It is this total score which is then used for further analysis. This study used the subscale scores, be-cause previous studies in different cultures (Abbott et al., 2006; Ryff & Singer, 2006;
Van Dierendonck, 2004) employing CFA with all the scale items led to the conclu-sion that the six original well-being dimenconclu-sions should be kept in tact. Higher scores indicated higher PWB.
Chinese translation of the Ryff PWB scales
Establishing and testing the equivalence of translated instruments across differ-ent cultures or languages requires both translation and back-translation. We followed the suggestions of Hyrkäs, Appelqvist-Schmidlechner, & Paunonen-Ilmonen ( 2003)
in this regard. The short form of the PWB Scales (Ryff, 1995) was translated from English to Mandarin with Ryff’s permission. The translator was fluent in both Eng-lish and Mandarin, had a Ph.D. in psychology, and had more than 10 years’ experi-ence in health promotion research. The Mandarin PWB was then back-translated by a different translator who had Ph.D.s in gerontology and psychiatry. The originally translated and back-translated PWB were then compared by these two translators, and all points of divergence were reconciled to accurately reflect the intent/accuracy of the item. Next, the English and final Mandarin scales were checked for equiva-lence by five experts who are fluent in English and Mandarin in the fields of nursing, human development, social psychology, public health, and geriatrics, respectively.
Specifically, they evaluated each item in the Mandarin version for linguistic appro-priateness, and they compared the words, phrases, and sentences in the original and back-translated English versions for proper meaning. The final scale had a high con-tent validity index (CVI) of 0.89, based on ratings by the five experts.
The Mandarin PWB Scales was then pilot tested with five elderly patients in a local health service center to determine whether they could understand the items.
They found the individual questions relevant to their daily life experience. They sug-gested rewording of some items to match their experience of well-being, and these changes were adopted without altering the context/meaning of the original PWB items.
Ch inese Happiness Inventory (CHI)
The CHI consists of 48 items, with 20 items derived from a qualitative study conducted in Taiwan (Lu & Shih, 1997) and 28 items taken from the Oxford Happi-ness Inventory (Lu & Shih, 1997a). The qualitative study show that harmony of
interpersonal relationships, praise and respect from others, satisfaction of material needs, achievement at work, downward social comparisons, and peace of mind are characteristics of happiness in Chinese society, and seven sub-scales were formed from the OHI: optimism, social commitment, positive affect, contentment, fitness, self-satisfaction, and mental alertness. CHI employs a 4-point scale; each statement represents a different level of subjective experience of happiness, which is then coded as 1, 2, 3, or 4 to measure the positive skewed nature of the happiness con-struct (Lu & Shih, 1997a) . The original CHI has been used extensively with elderly research subjects (Lu & Shih, 1997b; Lu & Lin, 1998). Reports on the reliability of the CHI Scale were good (α = .93 to .94)(Lu et al., 2001).
We used the shorten version of the Chinese Happiness Inventory (CHI), which has 20 items. The 20-item measurement was based on both Western questionnaire and qualitative research survey in Chinese population and testing of reliability and validity in several studies(Lu, 2006; Lu & Lin, 1998; Lu & Shih, 1997; Lu, Shih, Lin, & Ju, 1997). Scores on this item ranged from 1 (not at all) to 4 (often). Higher scores means higher happiness status. The item total correlation are very high with no item dropped, exploratory factor analysis showed that only one factor with inter-nal consistency was .95, test-retest reliability within 1 month interval was .66 (Lu &
Shih, 1997). The CHI Scale was then pilot tested with five elderly patients in a local health service center to determine whether they could understand the items and found them relevant to their daily life experience. (See Appendix 1).
Sen
se of coherence (SOC)
The original scale of sense of coherence consists of 29-item and 13- item ver-sion of the scale has also been used. The SOC questionnaire has been used in at least
33 languages in 32 countries with at least 15 different versions of the question-naire(Gana, 2001). Initially, the term ‘sense of coherence’ was used to refer to a way of seeing the world that facilitated successful coping with adversity. These factors are Comprehensibility, Manageability and Meaningfulness(Eriksson & Lindström, 2005).
A systematic review analyses the validity and reliability of sense of coherence scale which includes 458 scientific publications and 13 doctoral theses. In 124 stud-ies using SOC-29, the result showed that the Cronbach's alpha has ranged from 0.7 to 0.95. Moreover, the alphas of 127 studies using SOC-13 range from 0.7 to 0.92 and in 60 studies using a modified SOC scale indicted that the alphas range from 0.35 to 0.91. The relatively few test-retest correlations show considerable stability, e.g. 0.64 over a 3-year period (Eriksson & Lindström, 2005).
We used the shorten version of SOC-13. Scores on this item ranged from 1 (never) to 7 (always). Higher scores means higher sense of coherence status. The re-liability of SOC-13 version was found to be satisfactory and stability. For the SOC-13, alpha was 0.79 for the total score, and 0.6, 0.5, 0.63 for meaningfulness, manageability and comprehensibility, respectively. The construct validity was moder-ate to high to self-actualization (r=0.54, p<0.05) (Gana & Garnier, 2001).
3.1.2 Qualitative Interview
3.1.2.1 Interview guideline
Interview questions were crafted to relate to the research questions. The semi-structure interview guideline aims to explore the broaden and build effect of happiness and explore the personal resources such as intellectual, physical,
psycho-logical and social resources offered from the psychosocial intervention and walking exercise. There are six open-ended questions in the interview guideline.
1. What is your motivation to join the study?
2. How do you feel in the beginning of the intervention?
3. How do you feel after finishing the whole intervention? (four resources) 4. What is the effect of walking exercise and intervention course?(broaden effect) 5. What is the influence on your life after finishing the intervention? (build effect) 6. What is the difference on your life after finishing the intervention?
3.1.2.2 Validity and reliability of qualitative interview
According to Golafshani (2003) who suggest that the validity and reliability of qualitative research could be referred by the way of triangulation. Triangulation was used to assure reliability between investigator, data gathering and data transla-tion, data induction and reduction. Moreover, it also includes factors about the researcher's assumptions and the detailed choice of sampling method. Nevertheless, a thorough of relevant literature is also important to incorporated with interpretation in the result.
The qualitative research offers a rigorous treatment of internal validity, trian-gulation, member checks, and consideration of researcher bias. Triangulation results from the use of multiple data collection methods including quantitative result, recog-nize themselves. Member checks means double check with interviewee when finishing interview or in the end of interviews. Researcher also intent to continue to collect or observe data until the data were saturated while there are no new construct or interpretation.
To enhance the external validity, this study provides rich and thick descrip-tion so that transferability is possible by the interested reader. It can not be denied that a single experience is also valuable and conceivable. A single magnificently deep case could provide rich enough data for a study to be valid; however, failing this, additional cases could be included. The data included the point of saturation, in-terview, observation, and document collection, and analysis.