The Premorbid Adjustment Scale (PAS) was developed by Cannon-Spoor et al. 1982 for research use and has gained importance internationally. This scale is designed to measure the extent of attaining developmental goals premorbidly. The German version is presented here, with first data on the reliability and validity of the scale. In a sample of schizophrenic and schizoaffective patients (n = 86) and healthy parents of the patients (n = 38), DSM-IV diagnosis was made and PAS and Positive and Negative Syndrome Scale (PANSS) data were taken along with information on the course of the disorder. Using Cronbachs alpha, the estimated reliability for the scale and subscales lay between 0.809 and 0.931. High PAS scores, representing poor premorbid adjustment, correlated significantly with low age of onset, high PANSS scores, insidious onset, long hospitalisation, and serious course of the disorder. The threshold of PAS scores between healthy and sick probands was at 0.23. Patients with scores > 0.53 appeared to have an unfavourable course. With test results > 0.23, an odds ratio of 27.9 was ascertained (95% CI 9.39-82.89). The findings presented correspond with those from previous reports in literature.