Measuring state and trait anxiety in Turkish adolescents: development and validation of the six-item short forms of the state–trait anxiety inventory

Anxiety disorders currently represent the most prevalent psychiatric conditions affecting children and adolescents globally, resulting in substantial impairments across multiple domains of daily functioning, academic performance, and interpersonal relationships. Addressing the pressing need for rapid, efficient, and psychometrically robust psychological screening tools in clinical and educational environments, researchers have successfully developed and validated abbreviated, six-item versions of the State-Trait Anxiety Inventory for State (STAI-S) and Trait (STAI-T) anxiety tailored specifically for Turkish adolescents. This advancement slashes questionnaire length by 70 percent compared to the original 20-item instrument while preserving core measurement reliability and structural validity.

Background Context and Methodological Chronology

The Spielberger State-Trait Anxiety Inventory has remained a cornerstone of psychological assessment since its inception, utilized across diverse cultural and clinical settings to differentiate between transient emotional states of tension and apprehension (State anxiety) and stable long-term predispositions to perceive environmental situations as threatening (Trait anxiety). However, the original 20-item configuration of both subscales presents a heavy administrative burden. In high-volume clinical settings, longitudinal investigations, and time-constrained educational evaluations, lengthy questionnaires frequently induce respondent fatigue, which compromises data quality and lowers completion rates.

Prior efforts to shorten the instrument had largely focused on adult populations worldwide or utilized limited item pools that neglected reverse-scored items or lacked validation across developmental cohorts in Türkiye. To bridge this critical research gap, a comprehensive two-phase psychometric validation design was executed.

Phase one focused on item selection and exploratory factor analysis (EFA). This initial stage leveraged a clinical sample consisting of 420 adolescents aged 11 to 17 years who attended an outpatient child and adolescent psychiatry clinic in Adana. Within this cohort, 97.9 percent were receiving psychiatric follow-up for diagnosed conditions, predominantly attention-deficit/hyperactivity disorder, generalized anxiety disorder, oppositional defiant disorder, depressive disorders, and exam anxiety. Utilizing corrected item-total correlations alongside factor loadings and content domain coverage, researchers systematically winnowed down the inventory while maintaining a balanced three-to-three ratio of anxiety-present and anxiety-absent items.

Phase two involved structural confirmation and criterion-related validity testing via confirmatory factor analysis (CFA) performed on an independent, non-clinical sample of 277 public high school students aged 13 to 18 years. Data collection for the entire project adhered strictly to the Declaration of Helsinki, with formal approval secured from the Adana Training and Research Hospital Scientific Research Ethics Committee in September 2024. Written informed consent was mandated from parents or legal guardians alongside individual assent from every adolescent participant.

Psychometric Performance and Empirical Findings

The resulting six-item short forms—designated as STAI-S-6 and STAI-T-6—demonstrated robust psychometric properties across both cohorts. Exploratory factor analyses established that the six-item configurations accounted for 53.7 percent of the total variance for the state scale and 57.3 percent for the trait scale. High internal consistency was verified through both Cronbach’s alpha and McDonald’s omega coefficients, which reached 0.804 for both scales in the clinical sample and remained comfortably above acceptable thresholds (0.764 to 0.779) in the school-based validation cohort.

Confirmatory factor analyses utilizing the robust weighted least squares mean and variance adjusted (WLSMV) estimator confirmed that the abbreviated instruments successfully retained the two-factor structure of the parent scales. Standardized factor loadings ranged from 0.614 to 0.870 for the STAI-S-6 and from 0.716 to 0.924 for the STAI-T-6. Incremental fit indices yielded strong results, with comparative fit index (CFI) values exceeding 0.97 and standardized root mean square residual (SRMR) values remaining below 0.035.

To determine generalizability across demographic and clinical boundaries, multiple-group measurement invariance tests were conducted within an ordinal framework across a combined dataset of 697 participants. The analyses confirmed threshold-and-loading invariance across clinical and community status, sex, and age brackets. This critical finding indicates that items function equivalently and are interpreted identically by clinical patients versus community students, boys versus girls, and younger versus older adolescents, allowing researchers to compare latent means directly across diverse groups.

Criterion-Related Validity and Diagnostic Utility

External validity was established through correlation analyses with established mental health measures, specifically the Generalized Anxiety Disorder-7 (GAD-7) and the Satisfaction with Life Scale (SWLS). The STAI-T-6 demonstrated a strong positive correlation with the GAD-7 ($r = 0.645$) and a robust inverse relationship with life satisfaction ($r = -0.550$). Part-whole corrected correlations against the non-overlapping remainder of the parent scales confirmed that the abbreviated forms capture the vast majority of variance inherent in the full 20-item instruments without relying on item overlap.

Receiver operating characteristic (ROC) curve analyses underscored the practical utility of the short forms in identifying elevated anxiety symptoms. Against a GAD-7 clinical cut-off score of 10 or above, the STAI-T-6 yielded an area under the curve (AUC) of 0.846 in the high school sample and 0.824 in the clinical sample, matching the diagnostic discrimination of the 20-item parent scale. Youden-optimal thresholds identified a score of 16 or greater as the optimal provisional cut-off for the STAI-T-6 across cohorts, offering high specificity for screening applications.

Implications and Future Directions

The introduction of the validated Turkish six-item STAI short forms carries significant implications for psychological research, school guidance counseling, and pediatric mental health monitoring. By curtailing testing times by 70 percent, researchers and clinicians can seamlessly integrate anxiety assessments into large-scale epidemiological surveys, longitudinal mental health tracking, and busy outpatient workflows without sacrificing data integrity.

Nevertheless, study authors and independent mental health experts emphasize notable limitations. The abbreviated instruments are engineered primarily for research utility and broad screening purposes rather than standalone clinical diagnostic decisions. Furthermore, test-retest reliability was not evaluated, and diagnostic cut-off thresholds remain provisional pending independent replication against structured clinical interviews.

Future investigations must prioritize examining test-retest stability across various time intervals, establishing normative data across broader geographical regions in Türkiye, and deploying item response theory to refine item performance further. As adolescent mental health encounters mounting pressures from academic competition and societal shifts, these streamlined diagnostic tools provide a vital foundation for early detection, timely intervention, and comprehensive psychological support.