One of the greatest criticisms (and completely justified) of the Stanford-Binet test was that it was culturally biased. In large measure the WAIS-R has overcome that, but it's impossible to test raw intelligence with the tools that exist at the present time. That's why it's so difficult to test creativity, for instance. The purpose of those tests is to predict academic success, and they rely upon, and are constructed considering the stage in life that the testee will be. Tests are different, for example, for children and adults.
When I took a testing course at the urging of my academic mentor, who felt that we should have at least one clinical course in our repertoire, we were required to administer tests to adult and child volunteers (the latter with parental permission!). We also were taught to use the tests not so much as predictors of intelligence, but as diagnostic tools: that is, to attend to the manner in which the person answered the questions, and not simply what they answered. It was an interesting experience, and taught me above all to be extremely skeptical about canned testing devices.