This study examined the criterion validity of the Child Behavior Checklist (CBCL) and Teacher's Report Form (TRF) problem scales and items in demographically-matched Singapore samples of referred and non-referred children (840 in each sample for the CBCL and 447 in each sample for the TRF). Internal consistency estimates for both the CBCL and TRF scales were good. Almost all CBCL and TRF problem scales and items significantly discriminated between referred and non-referred children, with referred children scoring higher, as expected. The largest referral status effects were on attention problems scales and their associated items, with the TRF having larger effects than the CBCL. Effect sizes for demographic variables such as age, gender, ethnicity and SES were much smaller than effect sizes for referral status, across both the CBCL and TRF forms and at both the scale and item levels. These findings suggest that teachers can be effective partners in identifying children who need mental health services and those who do not.