Anchoring health-related quality of life (HRQOL) measures in population norms makes clinical interpretations more meaningful and is in accordance with practice in other fields of medicine. In this paper norms for the Short Form 36 (SF-36) are presented in a random sample, representative of the general Norwegian population. In addition, sociodemographic variables affecting the scale scores are explored and discussed. The response rate was 67%, being lowest among subjects aged 70 years or over. Data-completeness strongly declined with increasing age. Physical health scales were also strongly affected by age. In all scales, with the exception of general health perceptions, women reported having poorer health than men. Marital status affected the four mental health scales. Educational status affected all the scales, but the effect was smallest in the mental scales. These norms can be employed for comparison in case-control studies, or to interpret HRQOL changes in prospective studies. Differences in social status should be given special attention. Caution should be exercised when assessing subjective health or employing the norms among subjects aged 70 years or over.