Is There Really a General Intelligence?

Considering the rise of public acknowledgement on multiple intelligence theories, many people are now questioning the validity of general IQ tests - the tests that rank one person better than another using general IQ scores. Research studies show that general IQ tests seem to predict academic achievement, work performance, and receiving higher-paying and more prestigious jobs. However, closer inspection reveals that in terms of work performance, there are better predictors - motivation, educational level, and length of time working in a specific job. Despite these factors that seem to limit the validity of general intelligence tests, these tests are still highly being used to predict school and work performance, not to identify who will succeed in school or in work, but to avoid getting or hiring people that will most likely not to excel. People who scored high in general IQ tests tend to differ greatly in how they perform in subtests, but people who scored generally low tend to perform poorly in all subtests. For example, suppose a general intelligence test consists of 30 items - 10 items for mathematics, 10 items for abstract reasoning, and 10 items for vocabulary. Suppose the threshold for high scores is answering 15 items correctly. One high scorer may answer mathematical items perfectly and only answer 3 items each on abstract reasoning and vocabulary. Another high scorer may answer 8 out of 10 items correctly each on mathematics and abstract reasoning, but then fail to answer even one correct answer in vocabulary. There's great variation between the two high scorers; however, all low scorers almost always never get a score higher than half of the items in all subtests. Because general intelligence tests seem to predict low scorers more than high scorers, it still remains to be a useful measure of intelligence. Consequently, the popular consensus nowadays is to acknowledge the existence and the limitations of "general intelligence."