Background This study assesses the psychometric properties of the German version of the Padua Inventory-Washington State University Revision for measuring pediatric OCD. Methods The parent-rating and self-rating inventory is assessed in a clinical sample (CLIN: n = 342, age range = 6–18 years) comprising an OCD subsample (OCDS: n = 181) and a non-OCD clinical subsample (non-OCD: n = 161), and in a community sample (COS: n = 367, age range = 11–18 years). Results An exploratory factor analysis yielded a four-factor solution: (1) Contamination & Washing, (2) Catastrophes & Injuries, (3) Checking, and (4) Ordering & Repeating. Internal consistencies of the respective scales were acceptable to excellent across all samples, with the exception of the self-report subscale Ordering and Repeating in the community sample. The subscales correlated highly with the total score. Intercorrelations between the subscales were mainly r ≤ .70, indicating that the subscales were sufficiently independent of each other. Convergent and divergent validity was supported. Participants in the OCD subsample scored significantly higher than those in the non-OCD clinical subsample and the COS on all scales. In the COS, self-rating scores were significantly higher than parent-rating scores on all scales, while significant mean differences between informants were only found on two subscales in the OCD subsample. Conclusion The German version of the Padua Inventory-Washington State University Revision for measuring pediatric OCD is a promising, valid and reliable instrument to assess self-rated and parent-rated pediatric OCD symptoms in clinical and non-clinical (community) populations. Electronic supplementary material The online version of this article (10.1186/s13034-019-0286-z) contains supplementary material, which is available to authorized users.