Turkish Version of Acceptance and Action Questionnaire-II (AAQ-II): A reliability and Validity Analysis in Clinical and Non-Clinical Samples

Objective: Acceptance and Action Questionnaire-II (AAQ-II) is a self-evaluating scale that has been developed for assessing psychological inflexibility levels. The aim of the present study is to examine reliability and validity of the Turkish version of Acceptance and Action Questionnaire-II (Turkish AAQ-II) using clinical and non-clinical sample.Methods: The study group consisted of 207 patients who have at least one diagnosis of anxiety disorders, anti-social personality disorder, unipolar depression or bipolar disorder, and 267 healthy controls. A socio-demographic form, Turkish AAQ-II, Panic Disorder Severity Scale (PDSS), Ruminative Thinking Style Questionnaire (RTSQ), Beck Depression Inventory (BDI), Padua Inventory Washington State University Revision (PI-WSUR), Short Form-36 (SF-36), STAI (State-Trait Anxiety Inventory) I-II were all administered. Internal consistency and temporal stability analyses were performed to evaluate the reliability of Turkish AAQ-II. Exploratory factor analysis and confirmatory factor analysis (CFA) were also conducted to evaluate the construct validity of this instrument. Convergent, concurrent and predictive validity analyses were also performed.Results: From 474 participants across clinical and non-clinical samples our results indicated satisfactory reliability and validity of the Turkish AAQ-II. The Turkish AAQ-II showed good internal consistency with Cronbach's α coefficient of 0.84. 60 days test-retest reliability analysis also showed good temporal stability (Pearson's correlation coefficient, r = 0.85). For structural validity; principal component analysis was conducted and Kaiser-Meyer-Olkin index (r=0.83) showed suitability for factor analysis (Bartlett chi-square=1151.20; p<0.0001). One-factor solution (Eigenvalue of 3.62) accounted for 51.76% of the total variance. Confirmatory factor analysis demonstrates that a revised model of scale fits well with 7 items and a one-factor structure [RMSEA (0.079), SRMR (0.0210), CFI (0.971), GFI (0.972), NFI (0.961)]. Pearson's correlation analysis was used for evaluating convergent validity of Turkish AAQ-II and resulted in moderate correlations with RTSQ and STAI-II total scores (coefficients r = 0.566, r = 0.669, respectively). Concurrent validity analysis was performed to examine the predictive power of Turkish AAQ-II. Statistically significant correlations were found between total scores of Turkish AAQ-II and BDI (r=0.632), STAI-I (r=0.535), PI-WSUR (r=0.668) and PDSS (r=0.670). Predictive validity examined by comparing mean total Turkish AAQ-II scores of clinical and non-clinical groups and found statistically significantly higher scores in clinical (M=26.17, SD=8.81) group compared to to non-clinical [M=19.05, SD=7.76; t(443)=9.05, p<.0001] study group.Conclusions: It was found that Turkish AAQ-II has an one-factor structure with 7-item version. Higher levels of Turkish AAQ-II were found correlated with higher depressive, obsessive-compulsive, and anxiety related symptoms, and lower quality of life scores. Finally, it can be proposed that psychological inflexibility, assessed by Turkish AAQ-II, is a valid unidimensional measure in a variety of clinical sample as good as in non-clinical sample to measure the level of psychological distress.