Psychometric properties of the Mindful Attention Awareness Scale in Colombian undergraduates (MAAS)

Introduction and Theoretical Framework

This study presents the adaptation and psychometric validation of the Mindful Attention Awareness Scale (MAAS) in a large sample of Colombian undergraduate students. Mindfulness meditation has gained increasing popularity in clinical, health, and educational contexts, driving the development of multiple self-report instruments for its assessment. Diverse measurement approaches exist with varying characteristics: the MAAS (Brown & Ryan, 2003) is unidimensional and does not require prior meditation experience; the KIMS (Baer et al., 2004), FMI (Buchheld et al., 2001), PHLMS (Cardaciotto et al., 2008), TMS (Lau et al., 2006), SMQ (Chadwick et al., 2008), and FFMQ (Baer et al., 2006) are multidimensional instruments. The MAAS is the most widely used unidimensional scale and is designed for general population use without requiring prior meditation experience.

Mindfulness is defined as deliberately attending to present moment experience in a nonjudgmental manner (Kabat-Zinn, 1990). Multiple theoretical perspectives conceptualize awareness of the present moment as the core dimension of mindfulness. The MAAS has consistently demonstrated a unifactorial structure across diverse samples. Several Spanish language adaptations exist with adequate psychometric properties (Barajas & Garra, 2014; Soler et al., 2012), all replicating the unidimensional structure of the original instrument. However, no psychometric data or factor structure validation of the MAAS existed specifically for Colombian samples.

Objectives and Methodology

The primary aim of this study was to analyze the psychometric properties of the Spanish version of the MAAS (Soler et al., 2012) in a large sample of Colombian undergraduate students, examining the factor structure and convergent-discriminant validity of the instrument.

Participants: The sample comprised 762 undergraduate students (age range: 18–63 years; M=21.16, SD=3.76) from 7 universities in Bogotá (2 public and 5 private institutions). Forty-six percent were Psychology majors, with remaining participants studying Law, Engineering, Philosophy, Communication, Business, Medicine, and Theology. The sample was predominantly female (62%). Twenty-six percent had previously received psychological or psychiatric treatment, 4.3% were currently in treatment, and 2.9% were taking psychotropic medication.

Instruments: The MAAS (15 items, 6-point Likert scale with higher scores indicating greater mindfulness) was administered alongside a comprehensive set of validation measures. For convergent-divergent validity assessment, the following instruments were included: ATQ-8 (automatic thoughts), Believability Scale (cognitive entanglement with thoughts), AAQ-II (psychological acceptance/flexibility), GHQ-12 (general psychological distress), DASS-21 (depression, anxiety, stress subscales), DAS-R (dysfunctional attitudes), and SWLS (life satisfaction). Moderate negative correlations were expected between MAAS and measures of psychological distress, psychological inflexibility, and negative thinking, with moderate positive correlations expected with life satisfaction.

Procedure: A pilot study with 10 Colombian undergraduate students confirmed item comprehensibility without requiring modifications. Administration was collective in classroom settings with 6 administrators following identical instructions. Participation was voluntary with informed consent provided.

Data Analysis

Factor structure was analyzed via confirmatory factor analysis (CFA) using LISREL 8.71 with weighted least squares (WLS) estimation and polychoric correlations, the recommended approach for ordinal Likert data in large samples (Jöreskog & Sörbom, 1996). Only the unidimensional model was tested given its strong theoretical and empirical foundation. Fit indices included RMSEA, CFI, and NNFI. Missing values (9 total MAAS values distributed across items 1, 2, 3, 6, 9, 13, and three in item 7) were imputed via response pattern matching in LISREL. Remaining analyses were conducted in SPSS 19. Internal consistency was calculated using Cronbach's alpha with 95% confidence intervals following Duhacheck and Iacobucci (2004) procedures. Corrected item-total correlations were examined (items removed if r<.20). Descriptive statistics were computed. For discriminant validity, groups were compared using GHQ-12 cutoff scores via independent-samples t-tests. Convergent and divergent validity were evaluated using Pearson correlations.

Results

Factor Structure: Confirmatory factor analysis confirmed the proposed unidimensional model with adequate to excellent fit indices: RMSEA=.063 (90% CI [.056, .070]), CFI=.99, NNFI=.98. Factor loadings for the 15 items ranged from .51 (item 6) to .81 (item 8), with the majority in the moderate-to-high range (.70 to .81 for six items). The common factor explained substantial variance.

Internal Consistency: The scale demonstrated excellent internal consistency (α=.92; 95% CI [.91, .93]). Corrected item-total correlations ranged from .46 to .74, all exceeding the .20 criterion. Removal of no single item would have improved internal consistency, indicating all items adequately contributed to the construct.

Descriptive Statistics: Individual item means ranged from 3.63 (item 4) to 4.69 (item 15). Total scale scores were M=63.33 (SD=16.17; N=753 after listwise deletion). Individual item standard deviations ranged from 1.45 to 1.80, indicating adequate variability.

Discriminant Validity: Participants scoring at or above the GHQ-12 cutoff indicating significant psychological distress (GHQ≥12; M=59.33, SD=14.70, N=294) scored significantly lower on the MAAS than those below the cutoff (GHQ<12; M=65.95, SD=16.57, N=457); t=-5.59, p<.001. This difference confirmed the discriminant validity of the instrument.

Convergent-Divergent Validity: All correlations (see Table 4) were significant in expected directions (p<.001):

• Negative associations: GHQ-12 (r=-.24), DASS-Depression (r=-.36), DASS-Anxiety (r=-.34), DASS-Stress (r=-.31), ATQ-8 Frequency (r=-.29), ATQ-8 Entanglement (r=-.27), AAQ-II (r=-.31), DAS-R Total (r=-.24), DAS-R Perfectionism (r=-.25), DAS-R Dependence (r=-.19)
• Positive association: SWLS (r=.31)

These correlations were of moderate magnitude, consistent with previous validation studies (Long & Hayes, 2014; Soler et al., 2012).

Discussion and Interpretation

The results provide robust evidence that the MAAS possesses excellent psychometric properties in Colombian populations. Replication of the unidimensional model is consistent with both the original instrument (Brown & Ryan, 2003) and prior Spanish adaptation (Soler et al., 2012), suggesting transcultural stability of mindfulness as a unidimensional phenomenon centered on present-moment attention and awareness.

The pattern of correlations validates the underlying construct. Negative associations with depression, anxiety, stress, automatic thinking, and dysfunctional attitudes reflect the theoretical understanding that greater mindfulness is linked with reduced rumination and emotional reactivity. The positive association with life satisfaction is theoretically coherent. The magnitudes of correlations (typically ranging .19 to .36), while moderate, are expected and consistent with other validation studies, suggesting that mindfulness is a related but distinct construct from these psychological symptoms.

Discriminant validity was clearly demonstrated: individuals with indicators of general psychological distress (GHQ≥12) showed significantly lower mindful attention, confirming the clinical sensitivity of the instrument.

Limitations: The study focused on a nonclinical sample of undergraduate students, limiting generalization to clinical populations or older individuals with less formal education. Diagnostic information was unavailable for participants with prior psychological treatment. Validation instruments had not been formally validated in Colombia, although their internal consistencies replicated values from original validation studies.

Conclusions: The MAAS demonstrates reliability and validity as a measure of mindfulness in Colombian undergraduate students, operationalizing mindfulness as a unidimensional construct of nonjudgmental awareness and attention to present-moment experience. The instrument is appropriate for research and evaluation in Colombian educational contexts.

---