Dr Gudjonsson has used some of the information on the sampling and administration procedures adopted in the 1992 Dumfries norming of the Standard and Advanced Progressive Matrices (SPM and APM) completed at leisure to suggest that these norms might lead to seriously misleading conclusions if used as reference data when reporting the scores of individuals tested in the normal way. The material on which Dr Gudjonsson has drawn comes entirely from Raven, Court and Raven (1992a). The purpose of this Response is to summarise evidence from other sections of the Manual which indicates that the errors which would arise from the adoption of these norms are likely to be much less serious than those which would arise if any of the alternative sets of reference data were used.Before reviewing the evidence, however, it may be noted that the administration procedure adopted in the 1992 adult standardisation differs less from the recommended procedure than Dr Gudjonsson implies. The standard procedure requires only that those to be tested work on their own in a quiet room. Neither the presence of a tester nor isolation from others is necessarily involved. It is known that individual administration by a psychologist introduces emotional factors which affect scores. For many years, the Manual contained separate norms for the test administered in this way. Unfortunately, these are now so dated as to be worthless.