In psychological research, an assessment is considered good if it is both valid and reliable. Validity refers to an assessment's ability to produce meaningful results. To be considered a valid, an IQ test must be a proven measure of intelligence. Without a consensus concerning the nature of intelligence, it is impossible to deem any assessment "valid."
There are different types of reliability. While many tests are accepted as reliable measures of mathematical and spatial reasoning, the correlation between test scores and intelligence is debated.
Until the definition and theory of intelligence has been standardized, no assessment may be universally deemed "best."
School districts and internationally recognized intellectual promotion organizations, such as Mensa, use mathematical and spatial reasoning assessment tests. There are hundreds of such tests, including the Stanford-Binet Intelligence Scales and the Cognitive Abilities Test.
No single intelligence test is universally considered to be best: Two or more tests are sometimes used in conjunction with each other to ensure reliability.
Starting with the research of Harvard University educator Howard Gardner in 1983, tests with a broader concept of IQ have been developed. Gardner's research gave birth to the concept of multiple intelligence, which is now a familiar part of discussions of human intelligence. The theory of multiple expressions of intelligence challenges the use of assessments that measure only mathematical or spatial reasoning.