Spanish Translation, Adaptation and Psychometric Analysis of a Short Version of the Intercultural Sensitivity Scale in Chilean Public Workers.

Abstract

Chile has been facing socio-cognitive, affective and cultural tensions and changes due to the constant increase of the migrant population in Chile over the last two decades. This means that local public policy workers must interact and achieve effective communication with migrant users living in their territories. Intercultural sensitivity is understood as the affective dimension of intercultural communication. This dimension facilitates the awareness that people differ in their behaviors, perceptions and feelings during intercultural communication processes, promoting acceptance and respect towards these differences. The aim of this paper was to analyze the factor structure, psychometric properties and invariance of the Intercultural Sensitivity scale (Chen & Starosta, 2000) in a sample of Chilean workers. The sample consists of 337 workers from five municipalities in two regions of Chile. The sample has a mean age of 42.8 years (SD = 10.4), and 62.9% report being female. To determine the factor structure, an unweighted least squares (ULS) parameter estimation method was used, which is recommended for relatively small sample sizes, as is the case in this study. The overall structure of the original scale, although showing adequate indicators in a confirmatory factor analysis, presented serious conflicts in the covariance matrix used. Therefore, the 5-factor structure seen in other studies is disconfirmed. Because of this, it was decided to construct an abbreviated version, but reducing the limitations encountered during the validation process. The final result is a 4-dimensional structure, corresponding to four of the original dimensions, with a total of 13 items. The goodness-of-fit indices (CFI, TLI, RMSEA and SRMR) had optimal levels. The Manifest and Subtle Prejudice scale was used for its divergent validity, presenting all the dimensions of the instrument, and its total score, inverse and significant correlations with manifest prejudice, subtle prejudice, and total prejudice, as expected.