Psychometric Evaluation of the NKASRP Scale in Turkish Clinical Nurses.
TL;DR
The NKASRP scale demonstrated low internal consistency and substantial item heterogeneity within this general clinical sample in Türkiye, indicating that the sensitivity of the scale is highly vulnerable to sample heterogeneity.
Abstract
Aim
This study aims to methodologically evaluate the psychometric performance and contextual sensitivity of the Nurses' Knowledge and Attitudes Survey Regarding Pain (NKASRP) across non-oncology general clinical settings according to clinical specialization.
Materials And Methods
This methodological study was conducted between February and August 2021. To ensure statistical homogeneity, the operating room group (n = 5) was excluded from the analysis, and the study was completed with 308 nurses from internal medicine, surgery, intensive care, oncology, and emergency departments. Anti-Image Correlation was used for structural validity, Item Difficulty (P) and Item Discrimination indices (D) for item analysis, and Cronbach's Alpha and KR-20 coefficients for reliability. Differences between units were analyzed using One-Way ANOVA and post-hoc tests.
Findings
42.5% of the participants had 0-5 years of professional experience. The overall internal consistency coefficients of the scale were found to be far below acceptable limits (Cronbach's Alpha: .27; KR-20: .24). According to the item analysis results, more than 50% of the scale items (22 items) had discrimination indices (D) below .19 (very weak). Advanced tetrachoric correlation modeling revealed marked psychometric instability with an uninterpretable overall KMO value of 0.07. In particular, ANOVA results (F = 5.183, p = .002, ƞ² = .064) demonstrated significant differences in mean knowledge scores across clinical departments-with higher mean scores observed in oncology and intensive care units-reflecting unit-based score variations rather than demonstrated differences in unit-specific psychometric properties.
Conclusion
The NKASRP scale demonstrated low internal consistency and substantial item heterogeneity within this general clinical sample in Türkiye. These findings highlight the contextual sensitivity of the instrument across diverse clinical settings and underscore the need to evaluate measurement properties carefully prior to routine deployment in heterogeneous nursing populations. The findings indicate that the sensitivity of the scale is highly vulnerable to sample heterogeneity. The development of new, simpler, and more discriminative measurement tools that are compatible with the diversity and cognitive realities of clinical practice emerges as a critical need.