The aim of this cross-sectional community-based study was to examine the sensitivity and specificity of the Screen for Child Anxiety Related Emotional Disorders (SCARED) to the diagnosis of anxiety disorders (AD). Participants were 119 students aged 9-18. Psychiatric diagnoses were assessed by a psychiatrist throughout a structural clinical interview (K-SADS-PL). Forty-four participants had positive diagnosis for at least one AD. The total score of the SCARED significantly differentiated anxious from non-anxious children with an optimal cutoff point of 22 (sensitivity = 81.8 %; specificity = 52.0 %). SCARED subscales of social phobia and separation anxiety disorder, but not generalized anxiety disorder, revealed better discrimination proprieties than total scores to screen for that specific disorder (p less than 0.05). Both total and specific SCARED scores presented moderate sensitivity and specificity for detecting AD in a community sample. Investigators interested in screening for specific AD, rather than the group of AD, may benefit from using the specific subscales.