Abstract

Previous research has suggested that the nature of cognitive style was incorrectly specified by Allinson and Hayes (1996) when they presented their Cognitive Style Index (CSI) as a unitary conception of the construct. Hodgkinson and Sadler-Smith (2003) presented empirical evidence suggesting that a two-factor model with correlated factors provides a better approximation of responses to the CSI. In their rebuttal, Hayes, Allinson, Hudson, and Keasey (2003) concluded that these authors had failed to present a robust challenge to the construct validity of the CSI. However, their arguments were purely theoretical with no empirical evidence to support this assertion. In this study, we report the findings of a series of confirmatory factor analyses (N = 593) in an attempt to either replicate or refute Hodgkinson and Sadler-Smith's earlier findings. Results suggest that research using the CSI should continue on the basis of its unifactorial structure.