Psychometric properties of the Avoidance and Fusion Questionnaire – Youth in Colombia

Context and Objectives

The Avoidance and Fusion Questionnaire – Youth (AFQ-Y) is a widely used measure of psychological inflexibility in children and adolescents. Originally developed by Greco, Lambert, and Baer (2008), the AFQ-Y comprises 17 items with a brief 8-item version (AFQ-Y-8). Although adapted into various languages including Spanish, debate remains about its factor structure. This study aimed to adapt the AFQ-Y for Colombian participants and analyze its psychometric properties in a large sample of children and adolescents aged 8-18 years, including measurement invariance evidence across gender and age groups.

Method

Participants. The sample consisted of 1127 participants (57% female) aged 8-18 years (M = 11.11, SD = 2.73), enrolled in third to eleventh grade in Colombia (44.1% private schools, 55.9% public schools), recruited from nine educational institutions in Bogotá and surrounding areas.

Instruments. In addition to the AFQ-Y: GPQ-C (8 items, α = 0.83), DASS-21 (for adolescents 13-18; αs: Depression = 0.90, Anxiety = 0.88, Stress = 0.87), DASS-C (24 items, for children 8-12; αs: Depression = 0.78, Anxiety = 0.79, Stress = 0.69), PSWQ-C (14 items, α = 0.89), and PTQ-C (15 items, α = 0.93).

Adaptation. Adapted from the Spanish version (Valdivia-Salas et al., 2017) with minor wording changes to four items (2, 5, 13, 14) by two Colombian psychologists.

Analysis. Missing data (0.80%) imputed via similar response pattern in LISREL. CFA using Robust DWLS with polychoric correlations. One-factor and two-factor models compared. Measurement invariance (configural, metric, scalar) across gender and age group. Two-way ANOVA for score differences. Pearson correlations for convergent validity.

Results

All 17 items showed good discrimination (corrected item-total correlations 0.39-0.60). Alpha = 0.88 (95% CI [0.87, 0.89]); AFQ-Y-8 alpha = 0.82 (95% CI [0.80, 0.83]). Both one-factor and two-factor models showed excellent fit: one-factor S-Bχ²(119) = 397.709, RMSEA = 0.046, CFI = 0.99; two-factor S-Bχ²(118) = 334.553, RMSEA = 0.041, CFI = 0.99. One-factor selected for parsimony (interfactor r = 0.92). Metric and scalar invariance confirmed across gender and age for both versions. Girls scored higher than boys (F = 24.41, p < .001, η² = 0.028). No significant age group effect. Convergent correlations: GPQ-C (r = 0.65), PSWQ-C (r = 0.70), PTQ-C (r = 0.75), DASS Depression (r = 0.58-0.63), Anxiety (r = 0.59-0.60), Stress (r = 0.58-0.60). AFQ-Y/AFQ-Y-8 correlation: r = 0.93.

Discussion and Conclusions

The Colombian AFQ-Y is a valid and reliable measure of psychological inflexibility in children and adolescents. The one-factor solution was preferred. Correlations with repetitive negative thinking (r = 0.70-0.75) are higher than in adults, suggesting stronger relationships between these constructs in youth. Scalar invariance enables valid gender and age comparisons. Limitations include exclusive reliance on self-reports, lack of formal validation of some instruments in Colombia, and no clinical sample.

Significance and contribution

This study contributes significantly to the field of child psychological assessment by providing the first comprehensive validation of the Avoidance and Fusion Questionnaire – Youth in a large Latin American sample. The AFQ-Y demonstrated robust psychometric properties including unidimensional structure, adequate internal consistency, measurement invariance across gender and age groups, and theoretically coherent correlations with related measures, enabling researchers and clinicians to analyze the role of psychological inflexibility in child development and psychopathology in Spanish-speaking contexts.

---