100 should be the statistical average.
IQ is calculated as ("mental age" / "physical/real age" ) * 100. So if you're a 10-year old who performs at a 15-year level of cognition, you have (15/10) * 100 = 150 IQ.
Not that this is wrong per se, but they haven't done it like that since before almost all people on this board were born. They started doing away with that in WWII, and the new tests superceded the old by the 1960's. Even the older tests were modified to remove the quotient system.
Wechsler is best known for his intelligence tests. The Wechsler Adult Intelligence Scale (WAIS) was developed first in 1939 and then called the Wechsler-Bellevue Intelligence Test. From these he derived the Wechsler Intelligence Scale for Children (WISC) in 1949 and the Wechsler Preschool and Primary Scale of Intelligence (WPPSI) in 1967. Wechsler originally created these tests to find out more about his patients at the Bellevue clinic and he found the then-current Binet IQ test unsatisfactory. The tests are still based on his philosophy that intelligence is "the global capacity to act purposefully, to think rationally, and to deal effectively with [one's] environment" (cited in Kaplan & Saccuzzo, p. 256).
The Wechsler scales introduced many novel concepts and breakthroughs to the intelligence testing movement. First, he did away with the quotient scores of older intelligence tests (the Q in "I.Q."). Instead, he assigned an arbitrary value of 100 to the mean intelligence and added or subtracted another 15 points for each standard deviation above or below the mean the subject was. While not rejecting the concept of global intelligence (as conceptualized by his teacher Charles Spearman), he divided the concept of intelligence into two main areas: verbal and performance (non-verbal) scales, each evaluated with different subtests.
The WAIS is today the most commonly administered psychological test (Kaplan & Sacuzzo, 2005). The tests are currently updated approximately every ten years to compensate for the Flynn effect.
Pretty much every IQ test since at least the 1960's has used Wechsler's method and not the quotient method.
What they do is test samples of every age group and both genders, assign an arbitrary mean of 100, a normalized bell curve with standard deviation of 15, and fit everyone's data into that curve no matter how they actually scored.
This has obvious advantages and disadvantages, in that you don't have to come up with some amazing test, since the data gets molded into shape afterwards, but you also don't gain any information from the "fixed" scores? Is IQ going up or down? Are women or men smarter? You can't say, because you forced your data to fit an identical curve due to ideological reasons.
IQ tests: the only "scientific" field where they force data to fit a specific curve for ideological reasons, then claim they can tell you things about reality because of the shape of the curve.