Date Presented Accepted for AOTA INSPIRE 2021 but unable to be presented due to online event limitations.
The purpose of this study was to analyze the Life Satisfaction Index–Z (LSI–Z) and provide its item-level psychometric properties. Although the LSI–Z consists of two separate unidimensional and valid subscales, it demonstrated poor precision. Therefore, OT should be aware of this critical measurement issue when using this instrument in their clinical settings and must be interpreted the test scores with some caution.
Primary Author and Speaker: Heesu Choi
Additional Authors and Speakers: Nam Sanghun, Ickpyo Hong
PURPOSE: Life satisfaction is a desired subjective feeling closely related to moral and adjustment and a key component of psychological well-being (Diener, Suh, Lucas, & Smith, 1999; Wallace & Wheeler, 2002). Occupational therapists utilize various outcome measures to assess an individual's life satisfaction as a part of preventive health programs. The Life Satisfaction Index-Z (LSI-Z) consists of 13 items that is a shorter version of the original Life Satisfaction Index A (LSI-A) (Wood & Sheafor, 1969). The LSI-Z has been widely used in studies of life satisfaction in older people (Yamada, Kawamata, Kobayashi, Kielhofner, & Taylor, 2010). While previous study has examined factor structure of the LSI-Z (Lyyra, Törmäkangas, Read, Rantanen, & Berg, 2006), the item-level psychometric properties of the LSI-Z has not been reported yet. Considering the widespread use of the LSI-Z, it is essential to examine the psychometric properties of the instrument to support its clinical use. Therefore, the purpose of this study was to analyze the LSI-Z using the Rasch model and provide its item-level psychometric information.
DESIGN: The study data were extracted from a secondary database of the Well Elderly 2 Study conducted between 2004 and 2008 (ICPSR 33641). A random sample of 200 subjects were selected out of a total of 1,600 community-dwelling older adults for the data analyses. The majority of the sample was female (n = 132, 66%), white (n = 76, 38%) participants, and the average age was 74.8 years old (SD = 8.0, range 60-90 years).
METHOD: The unidimensionality assumption of the LSI-Z was examined using exploratory factor analysis (EFA) and principal component analysis (PCA) of Rasch residuals. The eigenvalue value less than 2.0 on the first contrast from the PCA of Rasch residuals is considered as a negligible noise level. The rating scale model was used to examine the fit statistics [0.7 < Infit MnSq < 1.3] and the precision of the test items [person strata > 3.0, person reliability > 0.80]. Mplus version 8.4 was used for EFA and Winsteps version 4.4.5 used for Rasch analysis.
RESULTS: The EFA revealed two separate measurement constructs among the test items (Positively Worded Satisfaction n = 8, negatively Worded Satisfaction n = 5; RMSEA = 0.078, CFI = 0.910, TLI = 0.868, SRMR = 0.079). The PCA of Rasch residuals also confirmed that each construct was unidimensional (Eigenvalue in the Positively Worded and Negatively Worded Satisfaction = 1.88, 1.53, respectively. Both measurement constructs demonstrated no misfit items. The two measurement constructs demonstrated poor precision (person strata = 1.75, 1.11; person reliability = 0.53, 0.25, respectively)
CONCLUSION: The item-level psychometrics of the LSI-Z demonstrated that the two separate unidimensional and valid subscales. However, the instrument demonstrated poor precision, meaning that it is not a reliable outcome measure. The small number of the test items in each construct could result in the poor precision.
IMPACT STATEMENT: A wide range of item sets that can measure the life satisfaction latent trait would improve the psychometrics of the LSI-Z. Clinicians should be aware of this measurement issue when utilizing this instrument in their clinical settings and must be interpreted with some caution.
References
Lyyra, T. M., Törmäkangas, T. M., Read, S., Rantanen, T., & Berg, S. (2006). Satisfaction with present life predicts survival in octogenarians. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 61(6), P319-P326. https://doi.org/10.1093/geronb/61.6.P319
Wallace, K. A., & Wheeler, A. J. (2002). Reliability generalization of the life satisfaction index. Educational and Psychological Measurement, 62(4), 674-684. https://doi.org/10.1177/001316402128775085
Wood, V., Wylie, M. L., & Sheafor, B. (1969). An analysis of a short self-report measure of life satisfaction: Correlation with rater judgments. Journal of Gerontology, 24(4), 465-469. https://doi.org/10.1093/geronj/24.4.465
Yamada, T., Kawamata, H., Kobayashi, N., Kielhofner, G., & Taylor, R. R. (2010). A Randomised Clinical Trial of a Wellness Programme for Healthy Older People. British Journal of Occupational Therapy, 73(11), 540–548. https://doi.org/10.4276/030802210x12892992239314
