Acceso usuarios
Temuco, Chile
El objetivo de este estudio fue examinar las propiedades psicométricas del Inventario Multifacético de Soledad en el adulto mayor (IMSOL-AM) en personas mayores de 60 años residentes en la comuna de Temuco. Se utilizó un diseño no experimental, correlacional y de carácter transversal. Con un muestreo no probabilístico, se seleccionó una muestra de 229 personas mayores de 60 años. Los participantes respondieron cuatro instrumentos: IMSOL-AM, la Escala de Depresión Geriátrica (GDS-15); la Escala Multidimensional de Apoyo Social Percibido (MSPSS, sigla en inglés), junto con un cuestionario sociodemográfico y de estado general de salud. Los resultados obtenidos sugieren que la escala IMSOL-AM es esencialmente unidimensional con buenos índices de confiabilidad del factor general. Con respecto a la validez convergente, se aprecia una correlación significativa, positiva y fuerte entre el IMSOL-AM y la Interdisciplinaria, 2025, 42(2), 463-486 463 Escala de Soledad Social en personas mayores GDS-15 (Rho = .620, p < .001), mientras que la correlación con el apoyo de la familia y otros significativos de la MSPSS fue significativa, inversa y fuerte (Rho = -.54, p < .001). La asociación con la dimensión apoyo de los amigos fue significativa, inversa y pequeña (Rho = -.134, p =.043), aportando evidencias de validez discriminante de la escala en estudio. Por lo tanto, el IMSOL-AM muestra una validez convergente apropiada y una estructura factorial esencialmente unidimensional.
The experience of loneliness in the elderly has been found to relate to general health problems, mental health and lower levels of well-being, as well as to greater social isolation.
Therefore, it is important to assess levels of loneliness among the elderly at an early stage.
Having assessment instruments with appropriate psychometric properties would provide timely information to design public health strategies to address loneliness. Since there is no validated scale to assess loneliness in the Chilean elderly population, the aim of this study was to examine the psychometric properties of the Loneliness Scale in the Elderly (IMSOL-AM) in adults over 60 years residing in the commune of Temuco. The study design was non-experimental, cross-sectional and correlational. The non-probabilistic sample comprised 229 adults aged 60 years and over living in Temuco. The variables under study were measured on a single occasion using the following scales: Loneliness in the Elderly Scale (IMSOL-AM) Geriatric Depression Scale (15-GDS); Multidimensional Scale of Perceived Social Support (MSPSS), and Sociodemographic and General Health Status Questionnaire. Confirmatory Factor Analysis was conducted to examine the factor structure of the IMSOL-AM. A bifactor model showed better fit to data compared to two-factor model, one-factor model and second order model (CFI=.969;
TLI = .961; RMSEA = .056; SRMR = .054). All the items showed statistically significant factor loadings to the general factor in the bifactor model. Several statistical indices derived from the bifactor model were analyzed to better inform psychometric properties of the scale under study and score interpretation. Explained Common Variance (ECV) was .831 indicating that the general factor explained 83 % of the common variance and the remaining 17 % is attributed to the two specific factors. This ECV is higher than .70 which is the threshold to accept essential unidimensionality of the scale. The Percent Uncontaminated Correlations (PUC) was .521 and the Factor Determinacy (FD) of the general factor was .982 higher than the threshold .90, suggesting that the scores of the general factor can be interpreted.
The reliability of the general factor and the specific factors were examined through Omega Hierarchical (�H) and the Omega Hierarchical Subscale (�HS), respectively. Hierarchical Omega indicated that 90.3 % of the variance of the total raw scores might be attributed to individual differences in the general factor. While subscale hierarchical Omega (�HS) indicated that the reliability of both specific factors (Family alienation and Personal dissatisfaction) was very low after controlling for the general factor. Overall, the bifactor statistical indices criteria suggest that IMSOL-AM is essentially unidimensional because ECV was higher than .60; PUC was lower than .80 and Omega Hierarchical was higher than .70. Although some degree of multidimensionality was detected, it is insufficient to question the essentially unidimensional structure.
Regarding the convergent validity, the correlation between IMSOL-AM and 15-GDS was significant, positive and strong (Rho = .620, p < .001), while the correlation with the MSPSS was significant, inverse and moderate (Rho = -.433, p < .001).
Therefore, the IMSOL-AM shows an essentially unidimensional factor structure, appropriate reliability of the general factor and convergent validity. These results suggested that feeling alienated from family is closely linked to personal dissatisfaction and both components are part of experiences of loneliness among elderly people. Reliability indices indicated that raw total scores consistently measure the general factor. Results also highlighted the high correlation between depressive symptoms and levels of loneliness. In this study, the psychometric properties of the IMSOL-AM were rigorously tested allowing researchers and health practitioners to use this instrument to evaluate levels of loneliness experienced by older people with similar characteristics to the study sample. The IMSOL-AM is a brief and easy to administer scale which comprises items with adequate comprehensibility for people aged sixty and over.
