Cross-cultural validation and psychometric properties of the Greek version of the Caring Behaviors Inventory: a methodological study

Abstract

Translation, adaptation and cross-cultural validation of instruments or scale to measure concepts of interest among different populations can facilitate sharing of knowledge and advance cross-cultural research globally, especially in countries with diverse populations. This is a two-phase study aiming to: (1) translate, adapt and cross-validate the 24-item English version of Caring Behaviors Inventory into Greek; and (2) evaluate the psychometric properties of the Greek version (CBI-GR). A descriptive correlational design was used to conduct the study. The procedure involved professional translators, committee approach, and a sample of nurses and patients. The psychometric evaluation of the CBI-GR used a sample of 245 Greek and Cypriot medical-surgical nurses. Data analyses comprised internal consistency reliability (Cronbach's alpha coefficient), stability (test-retest) reliability, item analysis and confirmatory factor analysis. The Cronbach's alpha for the overall CBI-GR was 0.92 and ranged from 0.76 to 0.87 for the subscales. Correlation coefficients between test and retest for the overall CBI-GR was r = 0.83 and ranged from r = 0.69 to 0.78 for its subscales. Confirmatory factor analysis of the scale supports its four-factor structure and the model indexes were adequate: X(2) (239) = 515.54, P < 0.001; comparative fit index = 0.91; incremental fit index = 0.91; and Tucker Lewis index = 0.90; parsimony normed fit index = 0.73; parsimony comparative fit index = 0.79; and root mean square of approximation = 0.07. The CBI-GR is comparable with the original 24-item English version of the CBI and it is suitable to measure nurse caring among Greek-speaking nurses. Further psychometric testing of the instrument among nurses and patients is warranted.