Measuring affect at work based on the valence and arousal circumplex model.

Affective states have become a central topic of interest in research on organizational behavior. Recently, scholars have been paying more attention to the proposals of the Circumplex Model (Russell, 1980) in order to gain a finer grained understanding of job-related affect. However, the limited availability of well-validated measures to test this model in work settings, particularly in non English-speaking populations, is still a major drawback. Using three samples of English-speaking and Spanish-speaking workers, this article offers the cross-validation of the Multi-Affect Indicator (Warr, 2007) between the original English version and its corresponding translation into Spanish. Multi-group Structural Equation Modeling supported the instrument's structure and its invariance between the two languages (English: χ2 = 65.56, df = 48, p = .05; RMSEA = .06; CFI = .97; Spanish: χ2 = 68.68, df = 48, p = .03; RMSEA = .05; CFI = .97). Furthermore, Circular Stochastic Modeling supported the theoretically proposed circumplex representation (χ2 = 139.85, df = 51, p < .01; χ2/df = 2.74, RMSEA = .06). Thus, this study offers an instrument that provides a more accurate approximation to affect at work, both in English and in another of the major language communities in the world, the Spanish-speaking population.