Total vocabulary has the highest correlation (0.8) with overall IQ of any individual measure of intelligence. This seems surprising in the light of the observations that:

• it measures only vocabulary, as opposed to logical/mathematical, spatial visualization, and other specialized aptitudes;
• it measures only crystallized intelligence as opposed to fluid intelligence;
• it is extremely culturally dependent; and
• it isn't clear whether one's vocabulary repertoire increases with age, and/or the extent to which one can increase one's vocabulary through study.

• Generally, either you know a word or you don't. Vocabulary Tests can be adminstered with generous time limits or even with virtually no time limits.. i. e., they're "power tests" rather than mental efficiency tests.
• We can present vocabulary items that can probably exceed the capabilities of even the most brilliant mentalities. We needn't bump up against a ceiling with a total vocabulary test.
• If a sufficiently large selection of words can be normed (e. g., all the more-challenging words in the language) , we may be able to generate cheat-proof tests that will never have to be retired. This would be particulary true if we let a computer randomly select words at specified levels of difficulty (or rarity).
• Vocabulary tests can be computer-administered-and-graded group tests (although they would have to be proctored to prevent cheating).