Performance of American Schoolchildren on Raven's Colored Progressive Matrices Scale

Abstract

Raven's Colored Progressive Matrices Scale is considered a relatively non-biased test of cognition even though higher socioeconomic status students consistently out-perform less advantaged age peers. Because it correlates highest with WISC Block Design and lowest with Information scores suggests that it is a culturally-reduced measure of nonverbalized abstract thinking. Major studies reviewed show a linear increase in score or proficiency by age, with boys performing slightly better. This analysis of combined means from four previous massive American studies which had diverse sampling clusters ( N = 4841) used weighted averaging techniques to produce “hypothetical national mean raw scores” for ages 6 through 12 yr. since this type of information did not exist. Weaknesses of the technique are lack of control over the samplings and the fact that they are spread out over time, and use of local median scores which is what Raven used at the 50 %ile level for all ages in his manual are recommended. All of Raven's other percentile scores, based upon 608 Scottish children from 1949, were interpolated for smooth working.