The purpose of this study was to examine further the factor structure, reliability, and validity of the Beck Anxiety Inventory (BAI) in samples of adolescents, ages 14 to 18 years. The BAI is a 21-item self-report measure of anxiety severity. The BAI total score differentiated between the inpatient and high-school adolescents. In the psychiatric inpatient sample, girls obtained higher anxiety-severity scores than boys; no gender differences were obtained for the BAI total score in the high-school sample. Confirmatory factor analyses did not provide adequate fit for the two- factor oblique BAI models to the separate male and female psychiatric inpatient data. Principal axes with varimax and promax rotations initially identified a four-factor solution in the separate male and female inpatient participants. However, second-order analyses of the primary factors provided stronger support for a single-factor structure in each sample. Estimates of reliability for the BAI were adequate in samples of psychiatric inpatient and high-school adolescents. Evidence for convergent and discriminant validity of the BAI was investigated separately in the male and female inpatient samples. Overall, the BAI showed acceptable psychometric properties in these populations. © 2002 Wiley Periodicals, Inc. J Clin Psychol 58: 443–456, 2002.