PISA seeks to report outcomes in terms of proficiency scales that are based on scientific theory and that are interpretable in policy terms. There are two further considerations for the development of the scales and levels: 1. PISA must provide a single score for each country for each of the three domains. It is also recognised that multiple scales might be useful for certain purposes, and the development of these has been considered alongside the need for a single scale. 2. The proficiency descriptions must shed light on trends over time. The amount of data available to support the detailed development of proficiency descriptions varies, depending on whether a particular domain is a ‘major’ or ‘minor’ domain for any particular survey cycle. Decisions about scale development need to recognise this variation, and must facilitate the description of any changes in proficiency levels achieved by countries from one survey cycle to the next. Development of a method for describing proficiency in PISA reading, mathematical and scientific literacy took place over more than a year, in order to prepare for the reporting of outcomes of the PISA 2000 surveys. The three Functional Expert Groups (FEGs) (for reading, mathematics and science) worked with the Consortium to develop sets of described proficiency scales for the three PISA 2000 test domains. Consultations of the Board of Participating Countries (BPC), National Project Managers (NPMs) and the PISA Technical Advisory Group (TAG) took place over several stages. The sets of described scales were presented to the BPC for approval in April 2001. This chapter presents the outcomes of this process.