EVALUATING THE SATISFACTION WITH LIFE SCALE ACROSS DIVERSE ROLES: A MULTI-GROUP INVARIANCE ANALYSIS OF STUDENTS, TEACHERS, MANAGERS, AND INDIVIDUALS WITH DISABILITIES FROM ONE COUNTRY
Author(s)
Amira Mohammed Ali, PhD1, Carlos Laranjeira, PhD2, Ayed Ahmad, PhD3, Imre Boncz, MSc, PhD, MD4, Atalla Amal Diab Ghanem, PhD5, Murat Yildirim, PhD6, Marwa Samir Sorour, PhD7, Annamaria Pakai, MSc, RN, PhD8.
1Department of Psychiatric Nursing and Mental Health, Faculty of Nursing, Alexandria University, Alexandria, Egypt, 2School of Health Sciences, Polytechnic University of Leiria, Leiria, Portugal, 3Arab American University, Jenin, Palestinian Territory, 4Institute of Health Insurance, University of Pécs Faculty of Health Sciences, Pécs, Hungary, 5Department of Nursing Adminstration, Faculty of Nursing, Alexandria University, Alexandria, Egypt, 6Department of Psychology, Agri Ibrahim Çeçen University, Agri, Turkey, 7Nursing Administration Department, Faculty of Nursing, Tanta University, Tanta, Egypt, 8University of Pécs Faculty of Health Sciences, Pécs, Hungary.
1Department of Psychiatric Nursing and Mental Health, Faculty of Nursing, Alexandria University, Alexandria, Egypt, 2School of Health Sciences, Polytechnic University of Leiria, Leiria, Portugal, 3Arab American University, Jenin, Palestinian Territory, 4Institute of Health Insurance, University of Pécs Faculty of Health Sciences, Pécs, Hungary, 5Department of Nursing Adminstration, Faculty of Nursing, Alexandria University, Alexandria, Egypt, 6Department of Psychology, Agri Ibrahim Çeçen University, Agri, Turkey, 7Nursing Administration Department, Faculty of Nursing, Tanta University, Tanta, Egypt, 8University of Pécs Faculty of Health Sciences, Pécs, Hungary.
OBJECTIVES: To examine Satisfaction with Life Scale (SWLS) structural stability across distinct populations from one European country.
METHODS: This cross-sectional study utilized multi-group confirmatory factor analysis to evaluate SWLS measurement invariance in a convenience sample (N = 1651, 57.8% females) comprising university students, teachers, managers, and individuals with acquired movement disabilities.
RESULTS: The unidimensional structure expressed good fit in the pooled sample (χ2(5) = 40.16, p = 0.001, CFI = 0.993, TLI = 0.985, RMSEA = 0.065, SRMR = 0.015). The baseline configural and metric invariance models demonstrated acceptable fit (CFI = 0.981, 0.971; TLI = 0.962, 0.963, respectively), with slightly inflated RMSEA (0.104, 0.102) while all fit indices indicated collapse of the scalar model (ΔCFI = 0.032, ΔTLI = 0.027, and ΔRMSEA = 0.024). Modification indices suggested correlating the residuals of items 1 and 2 to account for unique shared variance in the manager and disability groups. Based on group-specific χ2 contributions and intercept discrepancies, partial scalar invariance was established by freeing the intercepts of items 1, 2, and 4 across groups, while constraining items 3 and 5, which yielded acceptable fit (CFI = 0.966; TLI = 0.954, RMSEA = 0.115—minor fit changes relative to the metric model (ΔCFI = 0.005, ΔRMSEA = 0.013)). Kruskal Wallis and Mann Whitney tests revealed significantly lower scores of this two-item SWLS in student and disability groups relative to teachers and mangers (χ2(3) = 94.08; U = 14636.0, 11743.0, 85958.0, 108877.5; Z = -3.95, -3.89, -7.24, -7.86; p-values <0.001). Scalar invariance was maintained in a model comprising only teachers and students.
CONCLUSIONS: Despite overall robustness of the unidimensional SWLS, significant baseline response differences in items 1, 2, and 4 preclude full conceptual equivalence across social groups. Comparisons based on two-items urge policymakers to target psychological/financial vulnerability among students and individuals with disabilities.
METHODS: This cross-sectional study utilized multi-group confirmatory factor analysis to evaluate SWLS measurement invariance in a convenience sample (N = 1651, 57.8% females) comprising university students, teachers, managers, and individuals with acquired movement disabilities.
RESULTS: The unidimensional structure expressed good fit in the pooled sample (χ2(5) = 40.16, p = 0.001, CFI = 0.993, TLI = 0.985, RMSEA = 0.065, SRMR = 0.015). The baseline configural and metric invariance models demonstrated acceptable fit (CFI = 0.981, 0.971; TLI = 0.962, 0.963, respectively), with slightly inflated RMSEA (0.104, 0.102) while all fit indices indicated collapse of the scalar model (ΔCFI = 0.032, ΔTLI = 0.027, and ΔRMSEA = 0.024). Modification indices suggested correlating the residuals of items 1 and 2 to account for unique shared variance in the manager and disability groups. Based on group-specific χ2 contributions and intercept discrepancies, partial scalar invariance was established by freeing the intercepts of items 1, 2, and 4 across groups, while constraining items 3 and 5, which yielded acceptable fit (CFI = 0.966; TLI = 0.954, RMSEA = 0.115—minor fit changes relative to the metric model (ΔCFI = 0.005, ΔRMSEA = 0.013)). Kruskal Wallis and Mann Whitney tests revealed significantly lower scores of this two-item SWLS in student and disability groups relative to teachers and mangers (χ2(3) = 94.08; U = 14636.0, 11743.0, 85958.0, 108877.5; Z = -3.95, -3.89, -7.24, -7.86; p-values <0.001). Scalar invariance was maintained in a model comprising only teachers and students.
CONCLUSIONS: Despite overall robustness of the unidimensional SWLS, significant baseline response differences in items 1, 2, and 4 preclude full conceptual equivalence across social groups. Comparisons based on two-items urge policymakers to target psychological/financial vulnerability among students and individuals with disabilities.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
EPH51
Topic
Epidemiology & Public Health
Topic Subcategory
Public Health
Disease
No Additional Disease & Conditions/Specialized Treatment Areas