During the last decade, outcome assessment in orthopaedic surgery has increasingly focused on patient self-report questionnaires. The Oxford Hip and Knee Scores (OHS and OKS) were developed for the self-assessment of pain and function in patients undergoing joint replacement surgery. These scores proved to be reliable, valid, and responsive to clinical change, however, no German version of these useful measures exists. We therefore cross-culturally adapted the OHS and OKS according to the recommended forward/backward translation protocol and assessed the following metric properties of the questionnaires in 105 (OHS) and 100 (OKS) consecutive patients undergoing total hip or knee replacement in our clinic: feasibility (percentage of fully completed questionnaires), reliability (intraclass correlation coefficients (ICC) and Bland and Altman’s limits of agreement), construct validity (correlation with the Western Ontario and McMaster Universities Index (WOMAC), Harris Hip Score (HHS), Knee Society Score (KSS), Activities of Daily Living Scale (ADLS), and Short Form (SF-)12), floor and ceiling effects, and internal consistency (Cronbach’s alpha, CA). We received 96.6% (OHS) and 91.9% (OKS) fully completed questionnaires. Reliability of both questionnaires was excellent (ICC > 0.90). Bland and Altman’s limits of agreement revealed no significant bias. Correlation coefficients with the other questionnaires ranged from −0.30 (SF-12 Mental Component Scale) to 0.82 (WOMAC) for the OHS, and from −0.22 (SF-12 Mental Component Scale) to −0.77 (ADLS) for the OKS. For both questionnaires, we observed no floor or ceiling effects. The internal consistency was good with a CA of 0.87 for the OHS and 0.83 for the OKS. In conclusion, the German versions of the OHS and OKS are reliable and valid questionnaires for the self-assessment of pain and function in German-speaking patients with hip or knee osteoarthritis. Considering the present results and the brevity of the measures, we recommend their use in the clinical routine.