Culture Fair Intelligence Test

Fluid intelligence (Gf) taps “the level of complexity of relationships which an individual can perceive and act upon when he doesn't have recourse to answers to such complex issues already stored in memory” (Cattell, 1971, p. 99). It is “concerned with basic processes of reasoning and other mental activities that depend only minimally on learning and acculturation” (Carroll, 1993, p. 624). Therefore, tests of Gf have little informational content and require the ability to see complex relationships between simple elements like number and letter series, figure classification, figure analogies, spatial visualization, block designs, matrices, and so forth.

The Culture Fair Intelligence Test (CFIT) was designed by R. B. Cattell expressly as a nonverbal test to measure Gf. The CFIT comprises three scales: scale 1 for ages 4 to 8 as well as for mentally retarded adults, scale 2 for ages 8 to 13 as well as for average adults from the general population, and scale 3 for people of above-average intelligence. Scales 2 and 3 each have two parallel forms (A and B), which can be alternately used for retesting. The majority of these tests can be administered collectively, except some subtests from scale 1. The CFIT is highly speeded and requires detailed verbal instructions for administration.

Scale 1 includes eight subtests, but only half of them are really culturally fair. This scale cannot be recommended because some subtests must be administered individually, requiring complex instructions, and effects of familiarity with language and the habits of listening and attending are not minimized.

Scales 2 and 3 are quite similar, differing only in their difficulty level. They comprise four subtests: figure series (the individual is asked which figure logically continues a series of three model figures), figure classification (the individual is asked which two figures in each series go together), matrices (the individual must determine which of five alternatives most logically completes a given matrix pattern), and figure generalization (the individual must figure out the general rule for where a dot has to be placed by inferring the rule and picking the figure to which it applies). Each form of scales 2 and 3 takes about 30 minutes to administer.

Internal consistency and alternate-form reliability estimates generally range from .70 to more than .80. The lowest estimates are justified by Cattell on the scales' inclusion of several diverse formats to measure the same underlying construct. This is a main difference between the widely known Progressive Matrices Test and the CFIT. Cattell and Jensen[Page 205] discredited the former because of specific variance due to using only the matrix-problem format. These researchers agree that the CFIT does not contaminate the measurement of the construct of interest (Gf) with variance specific to item type. A fine-grained measure of Gf must employ several different subtests to wash out any contamination with test specificity.

Validity has been researched by means of factor analysis and correlations with other tests. CFIT correlates around .80 with a latent g factor, and correlations with other reasoning tests (i.e., Raven's Progressive Matrices, the Wechsler Intelligence Scale for Children, and the Wechsler Adult Intelligence Scale) are generally above .50. Predictive validity studies show moderate correlations with several scholastic and occupational criteria.

...