Experiential Avoidance as a Common Psychological Process in European Cultures

Full reference: Monestès, J. L., Karekla, M., Jacobs, N., Michaelides, M. P., Hooper, N., Kleen, M., Ruiz, F. J., Miselli, G., Presti, G., Luciano, C., Villatte, M., Bond, F. W., Kishita, N., & Hayes, S. C. (2018). Experiential avoidance as a common psychological process in European cultures. European Journal of Psychological Assessment, 34(3), 176-184. https://doi.org/10.1027/1015-5759/a000327

Study type: Psychometric validation study with measurement invariance analysis across multiple intercultural samples.

Background and objectives

Experiential avoidance, defined as the tendency to rigidly escape from or avoid private psychological experiences (thoughts, emotions, sensations), has been identified as a central transdiagnostic process involved in the onset and maintenance of emotional disorders. The Acceptance and Action Questionnaire-II (AAQ-II) is a brief 7-item self-report measure designed to assess this fundamental construct in Acceptance and Commitment Therapy (ACT) theory and Relational Frame Theory (RFT).

Although multiple translations of the AAQ-II have been made into different languages, an important gap exists in the literature: it has not been systematically examined whether the factor structure, psychometric properties, and the extent to which the AAQ-II functions equivalently across different European languages and cultures. This is a critical step to ensure that researchers and clinicians can make legitimate comparisons of experiential avoidance scores between groups and cultures.

The primary objective of this study was to examine the cross-cultural validity of the AAQ-II by conducting measurement invariance analyses (configural, metric, and scalar) on AAQ-II versions in six languages (Dutch, English, French, Italian, Spanish, and Cypriot Greek) administered to participants from seven European countries. The authors predicted that the AAQ-II would maintain a consistent single-factor structure, demonstrate good psychometric properties, and show measurement invariance across language samples, thereby allowing valid comparisons of experiential avoidance scores between groups from different linguistic communities and European cultures.

Method

Participants

The study included data from six independent validation studies of the AAQ-II conducted in different European countries. The total sample consisted of 2,170 nonclinical participants from seven countries: Belgium/Netherlands (N = 370), United Kingdom (N = 423), France (N = 207), Cyprus (N = 177), Italy (N = 365), Spain (N = 628), and an additional sample of 160 participants. Participants were primarily university students recruited through classroom announcements, flyers, or email invitations. Participants younger than 18 years were excluded. Overall, 487 participants (22.4%) were identified as male, 1,526 (70.3%) as female, and 157 (7.2%) did not report their gender. Mean age was 24.84 years (SD = 9.35). All participants completed questionnaires voluntarily without compensation.

Measures

Four primary measures were used in the present study:

Acceptance and Action Questionnaire-II (AAQ-II): The AAQ-II is a 7-item self-report measure answered on a 7-point Likert scale (from 1 = never true to 7 = always true). Items measure willingness to experience uncomfortable emotions, thoughts, and sensations and the ability to behave according to values even when experiencing psychological distress. Total scores range from 7 to 49, with higher scores indicating greater experiential avoidance. The AAQ-II was translated into each language by a team of clinicians from each country, with back-translation into English by a native English speaker unaware of the scale's purpose. The original AAQ showed good psychometric properties (Cronbach's alpha α = .88).

Beck Depression Inventory-II (BDI-II): Used to assess depressive symptoms as a concurrent validity measure. Experiential avoidance was expected to correlate positively with depressive symptoms (hypothesis: r ≈ .57).

White Bear Suppression Inventory (WBSI): Used to assess thought suppression as an indicator of concurrent validity. Experiential avoidance was expected to correlate moderately with thought suppression.

Mindfulness Attention Awareness Scale (MAAS): Used as a discriminant validity measure. Experiential avoidance was expected to correlate negatively with mindfulness (hypothesis: r ≈ -.47).

Procedure

Research was approved by institutional review boards (IRBs) in each country. Participants completed questionnaires anonymously online (for most samples) or in paper-and-pencil format. Informed consent was obtained. In case of retest, participants were asked to provide a code during the first administration to indicate when they would take the second administration one month later.

Statistical analyses performed

Confirmatory Factor Analysis (CFA): A CFA with maximum likelihood estimation (MLE) was conducted to examine the fit of the single-factor model in each language sample. Fit indices were reported: Comparative Fit Index (CFI), Non-Normed Fit Index (NNFI), Root Mean Square Error of Approximation (RMSEA), and Satorra-Bentler chi-square.

Measurement Invariance Analysis: Configural invariance (is the number and pattern of factors equal across languages?), metric invariance (are factor loadings equal across languages?), and scalar invariance (are item intercepts equal across languages?) were examined following standard multigroup CFA procedures.

Internal Consistency: Cronbach's alpha was calculated for each language sample as a measure of internal consistency.

Concurrent Validity: Pearson correlations were calculated between AAQ-II scores and BDI-II, WBSI, and MAAS scores.

Results

Descriptive characteristics by country

The descriptive data table showed that mean total AAQ-II scores ranged between 18.80 (Belgium/Netherlands) and 23.29 (Italy) across samples. No significant or barely significant differences were reported between countries. Internal consistency scores (Cronbach's alpha) were generally excellent across all language samples, ranging between .84 (Spain) and .89 (Belgium/Netherlands), with minor variation in Italy (.85).

Factorial structure analysis

CFA conducted in each language sample separately showed that the single-factor model presented acceptable fit to the data. CFI values ranged between .87 and .99, NNFI values between .84 and .98. RMSEA values ranged between approximately .08 and .20. Overall, these fit indices suggest that the single-factor structure is sustained in each language sample.

Two pairs of error covariances were allowed in all samples: (a) between items 1 and 4, and (b) between items 2 and 3. These error correlations were identified because these pairs of items tend to share similar content related to personal memories and the ability to control worries.

Measurement invariance analysis

Configural Invariance: The configural model (baseline) was found to show acceptable fit, confirming that the number and pattern of factors were equivalent across all languages.

Metric Invariance: Imposing equality constraints on factor loadings (metric invariance) did not produce a significant deterioration in model fit compared to the configural model. The criteria of Cheung and Rensvold (2002) for acceptable changes in CFI (ΔCFI ≥ -.01) were met, indicating that factor loadings can be considered equivalent across languages.

Scalar Invariance: Imposing equality constraints on item intercepts (scalar invariance) was also maintained. Changes in CFI were acceptable (ΔCFI ≥ -.01), indicating that intercepts can be considered equivalent across languages.

The demonstration of invariance at configural, metric, and scalar levels provides strong evidence that the AAQ-II measures the same construct in comparable ways across the European languages and cultural samples studied.

Concurrent Validity Data: Correlations between AAQ-II and other measures were:

• BDI-II: r = .57 (consistent with prediction)
• WBSI: r = .57 (positive correlation with thought suppression)
• MAAS: r = -.47 (negative correlation with mindfulness)

These correlations were consistent with concurrent validity predictions, providing evidence that the AAQ-II correlates as expected with related constructs across language samples.

Discussion and conclusions

The results of this study provide compelling evidence that the AAQ-II is a psychometrically valid measure of experiential avoidance that functions equivalently across multiple European languages and cultural contexts. The demonstration of configural, metric, and scalar invariance is particularly important because it justifies comparison of experiential avoidance scores between groups from different linguistic communities and cultures.

The consistency of single-factor structure, excellent internal consistency, and appropriate correlations with concurrent and discriminant validity measures suggest that the AAQ-II is a robust instrument for measuring experiential avoidance in European contexts. The findings also support the notion that experiential avoidance is a transdiagnostic psychological process that is present and measurable in comparable ways across different European cultural groups.

Limitations include the use of convenience samples primarily from university populations, the predominance of female participants (70.3%), and the lack of data from clinical samples. Although the authors acknowledge that participants were primarily from Northern Europe, the successful cross-cultural validation of the AAQ-II across these six languages and seven countries provides solid support for its utility in research and clinical practice in European contexts.

Significance and contribution

This study makes a significant contribution to cross-cultural psychometric research by providing rigorous evidence that a central measure of ACT/RFT theory (experiential avoidance) can be measured equivalently across multiple European languages and cultures. The findings support the use of the AAQ-II in comparative intercultural research studies and in clinical practice with multilingual populations. Furthermore, the results support the universality of experiential avoidance as a fundamental transdiagnostic psychological process, a central hypothesis of relational frame theory.