The Beck Anxiety Inventory in a non-clinical sample

Abstract

This study investigated the properties of the Beck Anxiety Inventory (BAI) in a sample of 326 undergraduate students. Scores on the BAI were compared with data from the State-Trait Anxiety Inventory and the Beck Depression Inventory. The BAI demonstrated good psychometric properties, with a high level of internal consistency. Relatively low test-retest correlations, in comparison with the STAI-Trait, suggested that the scale was functioning as a state measure. Factor analysis revealed a unifactorial solution on the first administration (a time of low stress), but a two factor solution similar to that proposed initially by Beck, Epstein, Brown and Steer (Journal of Consulting and Clinical Psychology, 56, 893–897, 1988) at the second administration (hypothesised to be a time of increased stress). Thus, the two factor structure of the BAI (characterised by physical and cognitive symptoms) may not be distinguishable in the normal population in the absence of an external stressor. An apparent strength of the BAI was its superior ability in differentiating anxiety from depression when compared with the STAI. A combined factor analysis of the BAI and STAI-State revealed two distinct factors, suggesting that the scales may actually be measuring separate, although not necessarily independent, constructs. It is suggested that the high discriminant validity demonstrated by the BAI may have been achieved at the expense of some construct validity.