ABSTRACT
Any useful test of intelligence must provide consistent and replicable results so that the performances of individuals can be compared with each other. A test that is designed to measure verbal ability and which yields only one factor to which all the items contribute significantly would have good factorial validity. Criterion validity is measured by comparing the results of the test against some external criterion such as performance in an examination. If a test produces the results predicted by a theory then it is said to have good construct validity. The child's mental age was then divided by their chronological age and multiplied by 100 to give the intelligence quotient (IQ). A child who scored at the average level for their age would always have an IQ of 100. Binet's concept of IQ as the ratio of mental age to chronological age is no longer in technical use.
