Psychometric properties of the expanded cultural intelligence scale in a South African context

Abstract

With organisations operating in a globalising world, intercultural contact within South African organisations is increasing more now than ever before. Developing cultural understanding amongst employees within multinational organisations may prove to be a completive advantage to be leveraged. Van Dyne et al. (2012) argue that Cultural Intelligence is a capability that can be measured and developed over time. The construct of Cultural Intelligence have evolved from a four factor model into an eleven factor construct which can be measured using the Expanded Cultural Intelligence Scale (E-CQS) in the scholarly literature. The purpose of this study was to evaluate the psychometric properties of the E-CQS as a proxy for Cultural Intelligence in a South African context, within a large multi-national organisation with a considerable cultural diversity. A diverse sample of 601 employees from a South African multinational organisation was used in the study. Confirmatory Factor Analysis (CFA) was used to confirm whether the eleven factor model proposed by Van Dyne et al. (2012) fitted the sampled data. The CFA indicated that the data did not fit the data well resulting in poor model fit in an eleven factor model. Subsequently, the four factor model was evaluated, and this model also resulted in poor fit. Problem items with statistical significant residuals in the estimated covariance matrix was identified, and removed from the model resulting in a 20 item four factor model based on the original Cultural Intelligence model proposed by Ang et al. (2008). The model with a reduced number of 20 items was evaluated using confirmatory factor analysis (CFA) within a Structural Equation Modelling (SEM) framework, which provided evidence for good model fit. The model was proven to be a reliable measure of Cultural Intelligence within the South African context, although the model lacked discriminant validity with high correlations between Metacognitive CQ and Behavioural CQ. Invariance testing of the four factor model provided evidence that the model is not invariant across race and gender groups and thus further research is required should South African organisations wish to capitalise on Cultural Intelligence within their organisations. An Exploratory Factor Analyses (EFAs) also confirmed the four factor model and provided insight into a possible three factor model that should be explored in future research.