Psychometric properties of the Cultural Intelligence Scale based on item response theory

Abstract

The aim of this study was to investigate the psychometric properties of Cultural Intelligence Scale (CQS), based on item response theory (IRT) using the graded response model (GRM). The study calibration sample included 400, while the study sample included 1000, male and female Saudi participants, aged between 18 and 62 years. IRT-GRM results supported the quality of the psychometric properties of CQS, and its appropriateness to measure cultural intelligence (CQ) for the majority of individuals. CQS well-distinguished people at different ability levels along the CQ latent trait, particularly with middle and low abilities. However, CQS full scale and subscales had less accurate measurement precision at high levels of CQ, and some subscales had more precision at low level abilities. CQS items had medium ability to differentiate among subjects, and they provided more information in evaluating individuals with medium CQ. Therefore, CQS might be more suitable for identification and development purposes, where low to med-levels of CQ are expected. Additional assessment procedures need to be added, for selection or promotion purposes to increase the measurement precision. Confirmatory factor analysis results confirmed the multidimensional construct of CQS with four specific-related factors at the first level, and an aggregate factor at the second level. This model provided better model fit using IRT-GRM approach, and it was supported by classical test theory analysis results. Therefore, it is important to rely on subscale scores, besides the total score to interpret CQ for individuals. The study stressed the importance of examining CQS item parameters and information based on the country it is adapted for, to investigate how they interact with country culture; and to take into account ability level, when selecting optimal measures.