Measurement invariance and differential item functioning of the positive and negative affect schedule: a psychometric...

This study aimed to validate the two-factor structure of the Spanish version of the Positive and Negative Affect Schedule (PANAS) in a sample of Ecuadorian young adults, examining its reliability, construct validity, and measurement invariance across gender. A total of nine hundred and eighteen participants completed the PANAS, along with measures of personality traits and negative life events. Confirmatory factor analysis (CFA), measurement invariance testing, and Differential Item Functioning (DIF) analyses were conducted. The two-factor model showed excellent fit after removing the item Alerta ("Alert"), which exhibited poor loading likely due to contextual reinterpretation. Both Positive Affect (PA) and Negative Affect (NA) scales demonstrated strong internal consistency (α >0.89). Discriminant validity was supported by near-zero latent correlations and compliance with the Fornell-Larcker criterion. Partial metric and scalar invariance across gender were observed, with DIF analyses revealing item-level differences, especially for fear- and hostility-related emotions. Criterion validity was confirmed via expected correlations with life events and personality traits. The PANAS shows robust psychometric properties in this population, although some items exhibit gender-based variability in interpretation. Cultural sensitivity and periodic item review are essential in emotional assessment tools to ensure conceptual and contextual validity.