Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.1 6/24/83; site ttidcc.UUCP Path: utzoo!linus!philabs!ttidca!ttidcc!hollombe From: hollombe@ttidcc.UUCP (The Polymath) Newsgroups: net.singles Subject: Re: Re: Intelligence Message-ID: <487@ttidcc.UUCP> Date: Mon, 17-Jun-85 14:34:04 EDT Article-I.D.: ttidcc.487 Posted: Mon Jun 17 14:34:04 1985 Date-Received: Tue, 18-Jun-85 04:48:29 EDT References: <378@boulder.UUCP> <1174@sjuvax.UUCP> Reply-To: hollombe@ttidcc.UUCP (The Polymath) Distribution: net Organization: The Cat Factory Lines: 35 Summary: In article <1174@sjuvax.UUCP> jss@sjuvax.UUCP (J. Shapiro) writes: >Has anyone out there noted yet that the standardized IQ tests test only for >a particular kind of logic, and that such tests are invalid for many of >the testees - the fact that your logic and the test's logic do not agree >does not render you stupid. There are many different types of test, each looking at different aspects of performance, some looking at many. The Miller's Analogies test is certainly very different from the SAT, for example. Many people do well on one test and poorly on another. That's why Mensa accepts adequate scores on _any_ standardized test of intelligence as criteria for membership. Mensa will also accept the word of a qualified psychologist if test scores are unavailable or inappropriate for any reason. >Also, IQ tests are only as good as the people who make them up. They are >probably not valid much over the median IQ of the testwriters. Not true. The people making up the questions and answers don't have to do so under test conditions. They have reference works available. Also, these tests are normed on the general population and questions that don't distinguish are weeded out. Some tests do have an upper limit above which they don't discriminate and this is noted in their scoring instructions. The process of creating and norming a new test of this type is very complex and _very_ expensive (millions of dollars to do it right). That's why new ones are seldom published. The market is too limited to make it worth the test developers' trouble. -- -_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_- The Polymath (aka: Jerry Hollombe) Citicorp TTI "How goes the rat race?" 3100 Ocean Park Blvd. "The rats are winning." Santa Monica, CA 90405 -- Paul Lynde (213) 450-9111, ext. 2483 {philabs,randvax,trwrb,vortex}!ttidca!ttidcc!hollombe