Factorial structure, validity, and gender invariance of the UCLA-R loneliness scale in ecuadorian adolescents

Abstract

Background Loneliness has a significant impact on mental and physical health across different stages of development, with particularly evident effects during adolescence. During this period, young individuals undergo substantial social and emotional transformations, making loneliness a global concern. Objective This study aimed to analyze the psychometric properties of the UCLA-R Loneliness Scale in Ecuadorian adolescents through exploratory and confirmatory factor analysis, assess its internal consistency, and examine factorial invariance across genders. Method The sample consisted of 718 school-aged adolescents (288 males and 430 females) aged 14–17 years (M = 15.72, SD = 0.747), selected through probabilistic sampling. A sociodemographic ad hoc questionnaire and the Spanish version of the 20-item UCLA-R Loneliness Scale were administered. The sample was randomly divided into two equivalent subsamples (n = 359) to separately perform an Exploratory Factor Analysis (EFA) and a Confirmatory Factor Analysis (CFA). The EFA was conducted using principal axis factoring, oblique rotation, and polychoric correlations. The CFA evaluated one-factor, two-factor, and three-factor models, employing fit indices such as CFI, TLI, RMSEA, and SRMR. Internal consistency was estimated using Cronbach’s alpha and McDonald’s omega coefficients. Additionally, factorial invariance analyses by gender were performed, along with a univariate ANOVA to examine potential gender differences. Results The exploratory factor analysis (EFA) results indicated the presence of two factors, whose structure explains 40.8% of the total variance, with 19 items. The confirmatory factor analysis (CFA) confirmed that the bifactor model with 19 items exhibited a superior fit compared to the unidimensional and three-factor models with 20 items, with fit indices falling within acceptable to excellent ranges (CFI = .936, TLI = .927, RMSEA = .050, SRMR = .048). Additionally, the bifactor model demonstrated measurement invariance across genders. Regarding internal consistency, the scale demonstrated a Cronbach’s alpha of.876 and an omega coefficient of.83, confirming its reliability. Conclusion It is concluded that the 19-item UCLA-R Loneliness Scale is a valid and reliable instrument for assessing perceived loneliness in Ecuadorian adolescents, regardless of gender. Future research could replicate these findings in other regions and cultural contexts to validate its use on a larger scale.