Testing psychometric properties of a Chinese version of perception of aggression scale.

Abstract

This study was to evaluate the psychometric properties of the Chinese version of the 12-item Perception of Aggression Scale (POAS). It consists of three phases of testing, including (1) translation and back-translation and content validity; (2) semantic equivalence between translated Chinese and original English version; and (3) construct validity, internal consistency and test-retest reliability. The translated Chinese and back-translated English version showed excellent similarities and agreements between two independent translators. The Chinese version indicated high item- and scale-level content validity indexes (0.86-1.00) and satisfactory semantic equivalence with the original English language version (weighted kappa=0.48-0.90; intraclass correlation coefficient=0.91). Exploratory factor analysis in 249 nursing students resulted in three components (dysfunctional, functional and protective dimensions), explaining 64% of the total variance, with satisfactory internal consistency (Cronbach's alpha=0.76-0.83) and good 2-week test-retest reliability (Pearson's r=0.87). The Chinese version of POAS was found to be a valid and reliable tool to examine nurses' attitudes towards patient aggression. Chinese nurses in this study viewed patient aggression to be an undesirable negative behaviour suggestive of uncontrollability behaviour presented by the patient under their care.