Objectives: Anxiety Sensitivity (AS) is conceptualized as a cognitive vulnerability reflecting the fear of anxiety symptoms based on beliefs about their harmful consequences. There has been a debate as to whether AS has a dimensional or taxonic latent structure. Given the contradictory findings, our two research groups, Croatian and Serbian, re-examined the question of AS latent structure. Method: In the Croatian study, 1704 undergraduate and graduate students (31% males, age = 20.65 (2.30)) completed the Anxiety Sensitivity Index (alpha = .84). In the Serbian study, 1621 undergraduates (33% males, age = 19.75 (1.48)) filled out the Anxiety Sensitivity Index-3 (alpha = .86). Data were submitted to the MAMBAC, MAXEIG, and L-Mode procedures. Results: In the Croatian sample, the taxon base rates across the taxometric procedures ranged from .27 - .37, Cohen’s ds ranged from 1.26 - 1.36, whereas the CCFIs had values from .249 - .298. In the Serbian sample, the taxon base rates had values from .04 - .11, Cohen’s ds ranged from .87 - 2.27, with the CCFIs spanning from .326 - .366. Conclusions: In the Croatian sample, all indicators across the three procedures were valid. In the Serbian sample, the results of the MAMBAC and MAXEIG procedures were based on valid indicators, whereas the validity of the indicators when submitted to the L-Mode was questionable. Overall, across the three independent taxometric procedures and the samples, comparisons of empirical and simulated data supported consistently the dimensional nature of AS. Research and clinical implications will be discussed. |