The Standard Progressive Matrices: methodological problems associated with the administration of the 1992 adult standardisation sample

The Standard Progressive Matrices [SPM (Raven, Court & Raven, Manual for Ravenʹs Progressive Matrices and Vocabulary Scales, 1992) is commonly used for measuring non-verbal intellectual abilities. Periodically the normative scores have to be revised and new standardisation tests conducted. The most recent standardisation of the adult norms suffers from serious methodological problems associated with the ways in which the test was administered. The most fundamental flaw is that the administration of the test was not supervised and two or more persons could easily have worked jointly on the items (i.e. the subjects could have cheated). Other problems relate to the inordinate length of time (i.e. up to 1 week) that subjects could have theoretically spent on the test and the fact that the range of scores that fall between the 5th and 95th percentile rank is exceptionally narrow. There is also the additional problem that scores obtained during self-administration of the test are not directly comparable with those obtained during individually supervised administration. These factors may result in reasonably high scores obtained during an individually supervised performance being misinterpreted as impaired performance when compared with the normative data. It is recommended that the 1992 adult normative data for the SPM are used cautiously when interpreting the performance score of an individual case.