Further comments on the nature and measurement of anxiety sensitivity: A reply to Taylor (1995b)

In a recent article in this journal, Taylor raised a number of important issues regarding our review of research on anxiety sensitivity (AS). Nonetheless, we contend that (a) Taylor's claim that the relationship between the ASI and self-report measures of fear-proneness is not attributable to method variance is unconvincing; (b) Taylor is incorrect that the expectancy model of anxiety predicts that only the main effect of AS, rather than the interaction between AS and anxiety expectancy, influences fear-proneness; (c) Taylor's analyses examining the interaction between AS and trait anxiety are questionable; (d) Taylor's assertion that the relationship between the ASI and panic disorder cannot be dismissed as tautological does not withstand close scrutiny; and (e) Taylor's claim that the ASI is unifactorial rather than multifactorial is not supported by available evidence. We discuss the implications of Taylor's analysis examining the hierarchical relation between AS and trait anxiety, and suggest research designs for elucidating the association between these two constructs.