The Gerontologist 44:408-417 (2004)
© 2004 The Gerontological Society of America
Examination of the Philadelphia Geriatric Morale Scale as a Subjective Quality-of-Life Measure in Elderly Hong Kong Chinese
Eric Wong, MA1,
Jean Woo, MD2,3,,
Elsie Hui, MB2 and
Suzanne C. Ho, PhD3
Correspondence: Address correspondence to Professor Jean Woo, Department of Medicine & Therapeutics, Prince of Wales Hospital, Shatin, N.T., Hong Kong. E-mail: jeanwoowong{at}cuhk.edu.hk
 |
Abstract
|
|---|
Purpose: We examine the psychometric properties of the Philadelphia Geriatric Morale Scale (PGMS) in an elderly Chinese population in Hong Kong. Design and Methods: The study consisted of two cohorts: (a) 759 participants aged 70 years and older living in the community who were recruited as part of a territory-wide health survey and interviewed in 1993–1994; and (b) 388 participants living in long-term-care institutions in 1995–1996. Participants who were cognitively impaired (Abbreviated Mental Test score less than 7) or who could not answer questions for other reasons were excluded. The 15-item PGMS was administered. The Rasch dichotomous model was used to assess the validity of the PGMS, and the 15-item Geriatric Depression Scale, previously validated in Chinese, was also administered to examine the concurrent validity. Results: A confirmatory factor analysis identified three underlying factors similar to findings among Caucasians; however, not all the goodness-of-fit indices were acceptable. An exploratory factor analysis using principal axis factoring and promax rotation revealed two underlying factors that explained 35% of the total variance: reconciled aging and unstrained affect. Some items were redundant when applied to the institutional sample. Concurrent validity was demonstrated by the good correlation between the reconciled aging and unstrained affect domains and the Geriatric Depression Scale (r = –0.72 and r = –0.56 respectively). Female gender, older age, and residence in institutions were associated with lower morale. Implications: The PGMS is a valid quality-of-life measure in elderly Hong Kong Chinese persons, but its psychometric properties are slightly different from those for Caucasians.
Key Words: Chinese • Philadelphia Geriatric Morale Scale • Validity • Quality of life
Among elderly populations for whom there is a high prevalence of chronic diseases and disability, psychosocial outcome measures become increasingly important in evaluating the effectiveness of care programs or intervention because diseases and disability may be difficult to reverse. Indeed, although physical outcome measures may be unchanged, psychosocial measures may show improvement for certain intervention programs (Lee, Lee, Mackenzie, & Ho, 2002). Therefore it would be important to have available tools for the measurement of psychosocial aspects in the elderly population. A wide variety of scales has been used to capture these aspects, broadly grouped under the term "quality of life." In essence, they try to evaluate how people view the impact of life experiences on their well-being. Quality of life has come to be an important component in the comprehensive assessment of health and well-being of elderly populations. It constitutes a tool for assessment of health or social interventions and, in long-term-care settings, may be used as an indicator of quality of care (Hopkins, 1992). Scales measuring quality of life vary in their emphasis on different components perceived to contribute to well-being. At one extreme, there are health-related quality-of-life scales that are disease specific, such as the Western Ontario and McMaster Universities (WOMAC) Osteoarthritis Index (Bellamy, Buchanan, Goldsmith, Campbell, & Stirr, 1998). Some scales are generic, covering multiple domains where a number of questions are objective, such as the absence or presence of pain and its severity, such as the 36-Item Short-Form Health Survey, known as the SF-36 (Lam & Lauder, 2000). At the other end of the spectrum there are scales with subjective questions, consisting of items representing positive and negative feelings or morale (Ranzijn & Luszcz, 2000). Examples of these are the Life Satisfaction Index (Neugarten, Havighurst, & Tobin, 1961) and the Philadelphia Geriatric Morale Scale (PGMS), which has been studied extensively.
The original PGMS scale consisted of 22 items (Lawton, 1975), which have been subsequently reduced to 15 items (Liang & Bollen, 1983). Three consistently reproducible factors have been identified: agitation, attitude toward aging, and lonely dissatisfaction (Liang, Lawrence, & Bollen, 1986, 1987). The scale has been recommended as one of the standardized assessment scales for the elderly population by the United Kingdom's Royal College of Physicians and the British Geriatric Society (Royal College of Physicians, 1992). However, there have been relatively few studies examining ethnic variations in the psychometric properties of the scale or its validity as a quality-of-life measurement tool, and its applicability in the elderly Chinese population has not been examined. The choice of tools developed in other countries for different ethnic and cultural groups would enable researchers to make comparisons. However, the validity and psychometric properties of these tools would have to be examined for individual populations.
In this study, we examined the psychometric properties and construct validity of the PGMS translated into Chinese, in two cohorts of elderly persons living in long-term residential care as well as in the community, to determine whether it may be used as a tool for measurement of quality of life in older Chinese populations.
|
Methods
|
|---|
Participants in this study were cohorts from two surveys: a longitudinal study initiated in 1990–1991 of persons living in the community, and a survey carried out in 1995–1996 of residents living in different types of long-term-care facilities. For the first survey, we recruited 2,032 Hong Kong Chinese adults aged 70 and older from participants of the non-means-tested Old Age Allowance Scheme, using a stratified disproportional random-sampling method to yield 300 participants for each of four strata (men aged 70–74 and 75+; women aged 70–74 and 75+), and 150 participants for each of six strata (men aged 80–84, 85–89, and 90+; and women aged 80–84, 85–89, and 90+). At 36 months follow-up, 519 participants had died and 342 were lost to follow-up, leaving 1,171 participants available for interviews. Of these, 782 persons were residing in the community. We recruited individuals for interviews from this group of participants during 1993–1994. The second survey consists of 590 participants aged 65 and older living in four residential-care facilities with varying staffing levels. We used an abbreviated mental test validated in Hong Kong Chinese consisting of 10 items as an initial screening test to exclude persons who were cognitively impaired (Chu, Pei, Ho, & Chan, 1995). We excluded individuals with a score of less than 7. We also excluded those who relied on proxies to answer questions or who were younger than 65 years of age. There were 759 participants living in the community and 388 participants living in institutions who responded to the PGMS.
We used the 15-item version of the PGMS (Liang & Bollen, 1983). This was first translated into Chinese by a bilingual medical staff, and then back-translated into English by another staff member. The two English versions were then compared to ensure that the contents were unchanged. We administered the Chinese version to a group of 10 persons in a community center for elderly persons to check that they could answer the questions. In general, a higher score indicates a higher level of morale. We also administered the 15-item Geriatric Depression Scale, which had been validated against a psychiatrist's diagnosis with a cutoff point of greater than 8 indicating depression (Chiu et al., 1994), in order to assess the concurrent validity of the PGMS.
Statistical Method
We used both confirmatory and explanatory factor analyses to examine structural validity. First, we used confirmatory factor analysis (CFA) to evaluate the relative fit of the two models, that is, the 15-item, three-factor model for American noninstitutionalized elders and the 11-item, three-factor model for Japanese aged persons previously proposed by Liang and colleagues (Liang, Asano, Bollen, Kahana, & Maeda, 1987; Liang & Bollen, 1983), to our present sample data. We also applied explanatory factor analysis (EFA) to identify any factors peculiar to this Chinese elderly population, and we further examined this EFA model by CFA to evaluate model appropriateness.
We performed separate analyses for the community and institution samples, and for the sample as a whole. We entered polychoric correlation matrices in the CFA analyses, and we used the weighted least squares estimation procedure in LISREL 8.50 for Windows (Scientific Software International Inc., Lincolnwood, IL) to conduct each analysis. The goodness-of-fit indices that we used to assess the fit of each model in this study included (a) the comparative fit index (CFI), (b) the goodness-of-fit index (GFI), (c) the root mean square error of approximation (RMSEA; Steiger, 1990), (d) the normed fit index (NFI), (e) the nonnormed fit index (NNFI), and (f) the relative fit index (RFI; Tucker & Lewis, 1973; also see Bentler & Bonett, 1980). The RMSEA has the advantage of being robust with respect to increase in degree of freedom, and the latter three indices have the advantage of being robust with respect to sample size increases (MacCallum & Hong, 1997). The goodness-of-fit criteria for each index are as follows: CFI, GFI, NFI, NNFI, and RFI > 0.9 (Marsh, Balla, & Hau, 1996); and RMSEA < 0.05 (Browne & Cudeck, 1993).
We used a Rasch dichotomous model (Wright & Mok, 2000) specifying unidimensionality and additivity to assess the internal validity of the best fitted PGMS model. In the Rasch model, the parameters are estimated for each person (Bn) and item (Di), and their difference (Bn – Di) is compared against unidimensional measurement model predictions; this is indicated by the formula
In this model, Pni denotes the probability that person n with trait level Bn will endorse an item i (score of 1). Calibrations for each item are expressed in logits, which is the natural log of an odds ratio. Logits typically range from –3 to +3, with logits of greater positive magnitude representing increasing item endorsement.
We assess the adequacy of the fit of each item to the Rasch model by using the information-weighted mean square residual goodness-of-fit statistic (INFIT; Wright & Masters 1982), which is a measure providing information about the responses given to items around the same difficulty endorsement level as the person's agreeability. An acceptable range for an INFIT value is between 0.8 and 1.2 (Wright & Linacre, 1994). An INFIT value less than 0.8 indicates that the item does not provide additional information beyond the rest of items on the scale. In contrast, an INFIT value greater than 1.2 indicates that the item (a) does not define the same construct as the rest of the items in the instrument, (b) is poorly constructed or misunderstood, or (c) is ambiguously defined. We used Winsteps software, Version 3.04 (MESA Press, Chicago, IL), which implements an unconditional maximum likelihood procedure (Wright & Panchapakesan, 1969), to perform the analysis. We used Pearson's correlation coefficient to examine the association between PGMS and Geriatric Depression Scale score.
|
Results
|
|---|
The characteristics of the participants are shown in Table 1. Overall there were more women than men, and those living in institutions tended to be widowed or never married, older, and uneducated.
Structural Validity
Neither the 15-item nor the 11-item three-factor models previously proposed by Liang and colleagues (Liang, Asano et al., 1987; Liang & Bollen, 1983) provide an adequate fit to the data for the community, institution, and combined samples (Tables 2 and 3, respectively). Although some of the goodness-of-fit indices supported the 11-item model (i.e., the CFI was 0.91 and the RMSEA and its 95% confidence interval were less than 0.05), the corresponding NFI, NNFI, and RFI were all below the goodness-of-fit criterion of 0.90. The latter three indices were recommended by McDonald and Marsh (1990) and Marsh and colleagues (1996) for their robustness to sample-size increases. Table 3 shows the item-by-item comparison with the findings by Liang and Bollen (15- and 11-item questionnaires). Differences in factor loadings for items 6, 9, 11, 15, and 16 between the Hong Kong, American (15-item questionnaire), and Japanese American (11-item questionnaire) versions were noted.
View this table:
[in this window]
[in a new window]
|
Table 3. Item-by-Item Comparisons on the Factor Loadings by CFA of the Chinese PGMS With Models of Liang and Colleagues.
|
|
The EFA using principal axis factoring and promax rotation (Table 4) revealed two underlying factors: unstrained affect (items 5, 7, 11, 12, 13, 15, 16, and 17) and reconciled aging (items 1, 2, 6, 8, 9, 10, and 14) for the community, institution, and combined samples. The CFA also proved that this two-factor model adequately fit all the samples (Tables 2, 3, and 4).
View this table:
[in this window]
[in a new window]
|
Table 4. Factor Loadings for the Explanatory Factor Analyses (Promax Rotation) of the Philadelphia Geriatric Morale Scale.
|
|
The interfactor correlations for all of the models are shown in Table 5. The interfactor correlations were high in all the models proposed by Liang and colleagues (Liang, Asano, et al., 1987; Liang & Bollen, 1983), and some of the factors seemed to be identical to each other.
Internal Validity
Table 6 shows that potential ceiling effects were present in both the community and institutionalized elders in the unstrained affect scores of 43.9% and 33.6%, respectively. Construct validity is demonstrated by the good overall correlation between the reconciled aging and unstrained affect domains and the Geriatric Depression Scale total scores for the community sample (Pearson's correlation coefficients of –0.766 and –0.564, respectively), the institution sample (Pearson's correlation coefficients of –0.604 and –0.559, respectively), and the combined sample (Pearson's correlation coefficients of –0.720 and –0.563, respectively).
Unstrained Affect Subscale
For the noninstitution sample, the mean logit score was 1.00 (SD = 1.22), indicating that the eight items (mean logit = 0.00; SD = 0.68), on average, were easily endorsed with positive responses by the sample. All items reasonably fit the Rasch model, except the item "Do you get upset easily?" (INFIT = 1.22; Table 7). For the institution sample, the mean logit score was 0.83 (SD = 1.38), indicating that the eight items (mean logit = 0.00; SD = 1.01), on average, were easily endorsed with positive responses by the sample. The two items "Do you have a lot to be sad about?" and "Do you take things hard?" both had an INFIT below 0.8 (Table 7). In fact, these two muted items were at a similar logit level to the items "Do you get upset easily?" and "Are you afraid of a lot of things?" respectively (Table 7), and hence they were redundant.
Reconciled Aging Subscale
For the noninstitution sample, the mean logit score was 0.02 (SD =1.20), indicating that the seven items (mean logit = 0.00; SD = 0.61), on average, were well matched with the measured morale of the person. One item—"Is life hard for you most of the time?"—was misfit (INFIT = 1.26; Table 7). For the institution sample, the mean logit score was –0.61 (SD = 1.24), indicating that the eight items (mean logit = 0.00; SD = 0.86), on average, were relatively difficult for the sample to endorse. Two items—"As you get older, are things better than expected?" and "Is life hard for you most of the time?"—were misfit (INFIT > 1.2). Two items—"As you get older, do you feel less useful?" and "Do you sometimes feel that life isn't worth living?"—had an INFIT below 0.8 (Table 6).
Table 8 examined the effect of age, gender, and place of residence on PGMS scores. There were significant differences in the mean logits among the four groups in both subscales. In general, female gender, older age, and residence in institutions were associated with lower morale.
View this table:
[in this window]
[in a new window]
|
Table 8. Differences Between Sex, Age, and Place of Residence on the Two Subscales of the Philadelphia Geriatric Morale Scale.
|
|
|
Discussion
|
|---|
Currently the literature on the well-being of elderly people across cultures is sparse. Torres (1999) pointed out the need for studies into the understanding of how culture shapes different constructs of aging. An anthropological study carried out in Hong Kong, Pennsylvania, Illinois, Ireland, and Botswana found that how older people regard successful aging was contingent upon their cultural origins (Keith, 1994). Cultural differences have also been noted with regard to more negative perceptions of aging among Asians (Giles et al., 2000) and filial obligations (Ng et al., 2000). With regard to quantitative assessment tools, the PGMS had been used with Black, White, and Japanese American populations (Liang, Asano et al., 1987; Liang, Lawrence et al., 1987), as well as in the Spanish (Stock, Okun, & Benito, 1994) and Israeli (Shmotkin & Hadarï, 1996) elderly populations, and it was found to have cross-cultural comparability, with three components being consistently identified. There were minor modifications in that some items could be deleted (Liang, Asano et al., 1987). This tool has been applied in various situations, such as in patients with stroke (Löfgren, Gustafson, & Nyberg, 1999) or hip fracture (Shepherd & Prescott, 1996), in elderly persons living in geriatric wards (Coleman, Philp, & Mullee, 1995) and nursing homes (Given & Range, 1990; Nagamoto, Iwagawa, & Takigawa, 1997; Reinke, Holmes, & Denney, 1981; Ryden, 1984), and as predictor of mortality (Maier & Smith, 1999). The advantage of the scale is that it is relatively short and therefore easily administered.
When we apply it to the Hong Kong elderly Chinese populations, the two-factor structure appears to provide a better model than the three-factor structure. It is possible that a better fitting two-factor model, instead of a three-factor model, is a reflection of cultural differences in the perception of aging that have been observed in recent studies (Keith, 1994; Giles et al., 2000; Ng et al., 2000). Some redundant items in both subscales were identified when applied to the institutional group (questions 11 and 16 in the Reconciled Aging subscale and questions 6 and 9 in the Unstrained Affect subscale). Therefore, when used exclusively with institutionalized elderly people, these questions could be omitted. The difference in scores between age groups, gender, and place of residence shows that the scale is sensitive to differences between groups with different characteristics, and therefore it is suitable as a tool to monitor change in psychosocial well-being. The lower morale associated with increasing age perhaps is not surprising and is compatible with a generally negative perception of aging in this society. It highlights the fact that in the promotion of "positive aging," a current Chinese government policy, an important area to address is how to improve this negative perception. In Hong Kong, policies regarding elderly people are formulated by the government on advice from the Elderly Commission, a body consisting of heads of government departments, representatives from nongovernment organizations, and academics. A recent initiative was to strengthen preventive measures in promoting positive aging from psychosocial as well as physical perspectives. Much had been done to promote a healthy lifestyle, with an emphasis on diet, physical activity, and avoidance of smoking. However, there had been comparatively little activity in the promotion of psychosocial well-being, although it has been documented that Chinese people in the People's Republic of China and Hong Kong are notably more negative about increasing age than populations in other nations of the Pacific Rim (Harwood et al., 2001).
The generally lower morale among women compared with men is worthy of comment. Possible contributing factors include a higher prevalence of chronic disabling conditions in women compared with men, as well as lower self-esteem and lower educational level in this cohort of women. Further studies would include an examination of shortened versions of the PGMS, comparisons with single questions such as "Are you satisfied with life?" (Ho et al., 1995), and an analysis of the most important factors contributing to individual subscales and the overall PGMS score.
There are limitations to this study. The exclusion criteria for cognitive impairment were arbitrary, and it is uncertain how milder degrees of cognitive impairment may affect the response. The education level of this cohort was generally low. Currently everyone in Hong Kong receives at least 9 years of compulsory education, so that the response to the PGMS may be different for future elderly populations with a higher level of education. In spite of these limitations, we may conclude that the PGMS is a valid quality-of-life measure in elderly Hong Kong Chinese, but that its psychometric properties differ slightly from that observed in Caucasians. This finding emphasizes the usefulness of the PGMS across cultures, thus providing a useful tool for cross-cultural comparisons using quantitative methods.
|
Footnotes
|
|---|
1 Centre for Clinical Trials & Epidemiological Research, The Chinese University of Hong Kong. 
2 Department of Medicine & Therapeutics, The Chinese University of Hong Kong.
3 Department of Community & Family Medicine, The Chinese University of Hong Kong.
Decision Editor: Linda S. Noelker, PhD
Received for publication January 19, 2003.
Accepted for publication May 22, 2003.
|
References
|
|---|
- Bellamy, N., Buchanan, W. W., Goldsmith, C. H., Campbell, J., & Stirr, L. W. (1998). Validation study of WOMAC: A health status instrument for measuring clinically important patient relevant outcomes to anti-rheumatic drug therapy in patients with osteoarthritis of hip or knee. Journal of Rheumatology, 15, 1833-1840.
- Bentler, P. M., & Bonett D. G. (1980). Significance tests and goodness of fit in the analysis of covariance structures. Psychological Bulletin, 88, 588-606.
- Browne, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit. In K. A. Bollen & J. S. Long (Eds.), Testing structural equation models (pp. 136–162). Newbury Park: Sage.
- Chiu, H. K., Lee, H. C. B., Wing, Y. K., Kwong, P. K., Leung, C. M., & Chung, D. W. S. (1994). Reliability, validity, and structure of the Chinese Geriatric Depression Scale in a Hong Kong context: A preliminary report. Singapore Medical Journal, 35, 477-480.
- Chu, L. W., Pei, C. K. W., Ho, M. H., & Chan, P. T. (1995). Validation of the Abbreviates Mental Test (Hong Kong version) in the elderly medical patient. Hong Kong Medical Journal, 1, 207-211.
- Coleman, P. G., Philp, I., & Mullee, M. A. (1995). Does the use of the Geriatric Depression Scale make redundant the need for separate measures of well-being on geriatric wards? Age and Ageing, 24, 416-420.
- Giles, H., Noels, K., Ota, H., Ng, S. H., Gallois, C., & Ryan, E. B., et al (2000). Age vitality across eleven nations. Journal of Multilingual Multicultural Development, 21, 308-323.
- Given, J. E., & Range, L. M. (1990). Life satisfaction and death anxiety in elderly nursing home and public housing residents. Journal of Applied Gerontology, 9, 224-229.
- Harwood, J., Giles, H., McCann, R. M., Cai, D., Somera, L. P., & Ng, S. H., et al (2001). Older adults' trait ratings of three age-groups around the Pacific Rim. Journal of Cross Cultural Psychiatry, 16, 157-171.
- Ho, S. C., Woo, J., Lau, J., Chan, S. G., Yuen, Y. K., & Chan, Y. K., et al (1995). Life satisfaction and associated factors in elderly Hong Kong Chinese. Journal of the American Geriatrics Society, 43, 252-255.
- Hopkins, A., (Ed.) (1992). Measures of the quality of life and the uses to which such measures may be put. London: Royal College of Physicians.
- Keith, J., Fry, C. L., & Glascock, A. P. (1994). The ageing experience: Diversity and communality across cultures. London: Sage Publications.
- Lam, C. L., & Lauder, I. J. (2000). The impact of chronic diseases on the health-related quality of life (HRQOL) of Chinese patients in primary care. Family Practitioner, 17, 159-166.
- Lawton, M. P. (1975). The Philadelphia Geriatric Center Morale Scale: A revision. Gerontology, 30, 85-89.
- Lee, D. T. F., Lee, I. F., Mackenzie, A. E., & Ho, R. N. (2002). Effects of a care protocol on care outcomes in older nursing home patients with chronic obstructive pulmonary disease. Journal of the American Geriatrics Society, 50, 870-876.
- Liang, J., Asano, H., Bollen, K. A., Kahana, E. F., & Maeda, D. (1987a). Cross-cultural comparability of the Philadelphia Geriatric Center Morale Scale: An American-Japanese comparison. Gerontology, 42, 37-43.
- Liang, J., & Bollen, K. A. (1983). The Structure of the Philadelphia Geriatric Center Morale Scale: A re-interpretation. Gerontology, 38, 181-189.
- Liang, J., Lawrence, R. H., & Bollen, K. A. (1986). Age differences in the structure of the Philadelphia Geriatric Morale Scale. Psychology and Aging, 1, 27-33.
- Liang, J., Lawrence, R. H., & Bollen, K. A. (1987b). Race differences in factorial structures of two measures of subjective well-being. Gerontology, 42, 426-428.
- Löfgren, B., Gustafson, Y., & Nyberg, L. (1999). Psychological well-being 3 years after severe stroke. Stroke, 30, 567-572.
- MacCallum R. C., & Hong, S. (1997). Power analysis in covariance structure modeling using GFI and AGFI. Multivariate Behavioral Research, 32, 193-210.
- Maier, H., & Smith, J. (1999). Psychological predictors of mortality in old age. Journal of Gerontology: Psychological Sciences, 54B, P44-P54.
- Marsh, H. W., Balla, J. R., & Hau, K. T. (1996). An evaluation of incremental fit indices: A clarification of mathematical and empirical properties. In G. A. Marcoulides & R. E. Schumacker (Eds.), Advanced structural equation modelling: Issues and techniques (pp. 315–535). Mahwah, NJ: Erlbaum.
- McDonald, R. P., & Marsh, H. (1990). Choosing a multivariate model: Noncentrality and goodness of fit. Psychological Bulletin, 107, 247-255.
- Nagamoto, I., Iwagawa, S., & Takigawa, M. (1997). Abnormal behaviour of elderly people residing institutions of medical and social welfare systems. Psychiatry and Clinical Neurosciences, 51, 53-56.
- Neugarten, B. L., Havighurst, R. J., & Tobin, S. S. (1961). The measurement of life satisfaction. Journal of Gerontology, 16, 134-143.
- Ng, S. H., Loong, C. S. F., Liu, J. H., & Weatherall, A. (2000). Will the young support the old? An individual- and family-level study of filial obligations in two New Zealand cultures. Asian Journal of Social Psychology, 3, 163-182.
- Ranzijn, R., & Luszcz, M. (2000). Measurement of subjective quality of life of elders. International Journal of Aging and Human Development, 50, 263-278.
- Reinke, B. J., Holmes, D. S., & Denney, N. W. (1981). Influence of a "friendly visitor" program on the cognitive functioning and morale of elderly persons. American Journal of Community Psychology, 9, 491-504.
- Royal College of Physicians (Research Unit) and the British Geriatric Society. (1992). Standardised assessment scales for elderly people (pp. 16–17). London: Author.
- Ryden, M. B. (1984). Morale and perceived control in institutionalized elderly. Nursing Research, 33, 130-136.
- Shepherd, S. M., & Prescott, R. J. (1996). Use of standardised assessment scales in elderly hip fracture patients. Journal of the Royal College of Physicians London, 30, 335-343.
- Shmotkin, D., & Hadarï, G. (1996). An outlook on subjective well-being in older Israeli adults: A unified formulation. International Journal of Aging and Human Development, 42, 271-289.
- Steiger, J. H. (1990). Changing causal relationships without changing the fit: Some rules for generating equivalent LISREL models. Multivariate Behavioral Research, 21, 309-331.
- Stock, W. A., Okun, M. A., & Benito, J. G. (1994). Subjective well-being measures: Reliability and validity among Spanish elders. International Journal of Aging and Human Development, 38, 221-235.
- Torres, S. (1999). A culturally relevant theoretical framework for the study of successful ageing. Ageing and Society, 19, 33-51.
- Tucker, L., & Lewis, C. (1973). A reliability coefficient for maximum likelihood factor analysis. Psychometrika, 38, 1-10.
- Wright, B. D., & Linacre, J. M. (1994). Reasonable mean-square fit values. Rasch Measurement Transactions, 8, 370.
- Wright, B. D., & Masters, G. N. (1982). Rating scale analysis. Chicago: MESA Press.
- Wright, B. D., & Mok, M. (2000). Rasch models overview. Journal of Applied Measurement, 1, 83-106.
- Wright, B. D., & Panchapakesan, N. (1969). A procedure for sample-free item analysis. Educational and Psychological Measurement, 29, 23-48.
This article has been cited by other articles:
|
|
|
|
 
K. J. Anstey, R. Burns, C. von Sanden, and M. A. Luszcz
Psychological Well-Being Is an Independent Predictor of Falling in an 8-Year Follow-Up of Older Adults
J. Gerontol. B. Psychol. Sci. Soc. Sci.,
July 1, 2008;
63(4):
P249 - P257.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
Abstract
Methods
