Social phobia (SP) has onset during early adolescence, and is associated with significant impairment in social and educational functioning of adolescents. Therefore, valid and easy-to-use tools for screening and identification of SP among adolescent community populations are needed. We investigated both construct and discriminative validity, and screening properties of the 17-item Social Phobia Inventory (SPIN) relative to SP diagnoses based on a semi-structured clinical interview (K-SADS-PL), in a sample of 752 12 to 17-year-old Finnish students from general population. The SPIN demonstrated good properties to differentiate adolescents with SP and those with sub-clinical SP symptoms (SSP), from adolescents without SP. The SPIN also differentiated adolescents with SP from those with depressive and disruptive disorders. In this sample 27% of participants scored above the previously suggested SPIN cut-off (15 points) for adolescent SP. We suggest using a somewhat higher cut-off score, 24 points, when using the SPIN as a screen for SP in general adolescent populations. This cut-off score resulted in a sensitivity of 81.2%, a specificity of 85.1%, a positive predictive value of 26.9%, and a negative predictive value of 98.6% in relation to the SP diagnosis in our sample. To screen for both SP and SSP, 19 points as a cut-off score produced satisfactory diagnostic efficiency statistics. The SPIN appears to have good properties for screening and identification of adolescent SP.