Raven'S Progressive Matrices

Raven's Progressive Matrices, published by Lewis, is a test that was originally introduced in 1938. Its purpose is to measure Spearman's g factor or general intelligence. For this reason, Raven chose a spatial format for the test, which required the education of relations among abstract items, that is, the visualizing of relationships that are based on perceptual similarities between stimuli.

It is a nonverbal test of reasoning ability that comes in three forms; Coloured Progressive Matrices, Standard Progressive Matrices, and Advanced Progressive Matrices (Sets I and II). The test can be administered in groups or individually, and each form takes from 15 to 30 minutes to complete.

The Coloured Progressive Matrices is a 36-item test that can be used mainly with children from 5 to 11 years of age. In this form, the figures are in color in order to appeal to the children and sustain their attention. The Standard Progressive Matrices consists of 60 items that are presented in five sets, with 12 items per set. The test is employed mostly with persons from 6 to 14 years of age. The Advanced Progressive Matrices is appropriate for older adolescents and adults, primarily for individuals with higher than average intellectual ability. It contains 12 items in Set I and 36 in Set II.

In each form of the test, the examinee is presented with a design or matrix from which a part has been removed. The testee has to look through the matrix and decide which of six or eight pieces given below the matrix is the right one to complete the matrix. The first item in a set is intended to be self-evident, and it is followed by another 11 items of increasing difficulty.

The items referred to are (a) consecutive patterns, (b) analogies between pairs of figures, (c) progressive modifications of patterns, (d) transformations of figures, and (e) disentangling of figures into component parts. Thus, the test estimates the ability to make comparisons, to [Page 827]deduce by analogy, and to organize spatial patterns into wholes.

The test items can be solved by employing problem-solving strategies. The principle that has to be used to solve each item is either to put forward orally or to draw information from visual perceptual recognition of the stimulus. In the first occasion, an analytical method is used, whereas in the second, a Gestalt procedure involving visual perception is used to solve the items.

The reliability and validity of the test is excellent, but the construct validity of the test is not supported by research because the very easy and very difficult items of the test measure different operations.

Although the test does not respond to the initial Raven's intentions of measuring Spearman's g factor, it is a useful indicator of nonverbal reasoning ability. More specifically, it is easy to administer and useful for screening the intelligence of children and adults with language difficulties and auditory or physical disabilities. Furthermore, it is useful for testees who do not speak English or have a restricted command of English.