A Theory on IQ Test Point Deviations

I don't put much stock in an IQ. Perhaps part of that is that I'm not intelligent enough to personally recognize the significance. But it just doesn't do anything for me. That a rocket scientist is a rocket scientist isn't the result of a test suggesting the rocket scientist's IQ fell within the parameters of other rocket scientists. That a ditch digger is a ditch digger doesn't mean their IQ couldn't be that of a rocket scientist's. And I just wanna say rocket scientist one more 'gain to say I said it. It's very unimportant for most people, I imagine. I received a score somewhat above average when I was tested, and I have since taken tests again for shites and giggles. But I had more of a Karl Pilkington reaction to my score either way -- a blank stare and an "alright." It wasn't life altering; I didn't think I could or couldn't do something based on any final result. But if it makes someone feel great to get a high score, I suggest cheating!

I think that you have misunderstood the significance of the numbers 15 and 3 when given as measures of standard deviation. You probably need to know more about general probability theory and the way that IQ scores are adjusted to fit the normal curve (often called the 'bell curve' because of its shape. When a test is set up, it is first used on a representative sample of the population and the method of scoring is then adjusted to normalize the 'bell curve' so that the median of the results is equal to 100 and that 95% of the whole population's scores lie within plus or minus two SDs of the median. The SD is normalised (in most modern tests) to be 15 so that in any test of a representative sample of the population, 95% of the scores will be between 70 and 130. Any individual who takes the test can now be matched to the rest of the population and any particular score can be said to show that there is a likelihood (not an absolute certainty) that that individual's mental ability is the same as that of others with the same score. This probability that two scores being the same represent the same mental ability can also be mapped onto a normal ('bell') curve and the standard error of measurement is about 3 points, or in other words, the odds are about 2 out of 3 that a person's true IQ is in range from 3 points above to 3 points below the test IQ. If you were to take a standard test and your score was 115, therefore, the chances are about 66% that your actual level on the scale will be between 112 and 118 or, to put it another way, there is a 95% chance that your true IQ is in a range from 4 to 5 points above to 4 to 5 points below the test IQ, depending on the test in question. I have only given a brief outline above and you can read much more about it here: http://en.wikipedia.org/wiki/Iq There are plenty of pages on the web which will give you a basic introduction to probability theory and will help you understand why, in any particular test, your score isn't absolute but shows, with a specified level of confidence that you are within a range of other, similar, scores by other people. It is also worth stressing that you cannot rely on the IQ scores given by tests on the Internet. These are not taken under the controlled conditions required to obtain a valid score (the dog may jump up at you and distract you while you are taking the test or you may decide that you need a coffee!) and, although the questions and the scoring methods may be derived from actual tests, it is quite likely that the programmers have 'adjusted' the results so that everyone's scores are somewhat higher than they should be.

I have taken several IQ tests over the course of the last fifty odd years, and all of them, official or entertainment standard, have fallen in that 4-5 point variation you mentioned. When I went back to Uni in my late 50s it seemed to stimulate my brain, and I scored at the top end of "my" scale at the end of my PhD program. At the beginning, after many years of intellectual laziness, I scored at the lower end of that variation. As psychologist Abraham Maslow, himself tested at 182, said, IQ tests have definite limitations, but within those limitations are a useful tool. My study of Jung in recent years convinced me that the standard IQ test shows only one kind of intelligence, "instrumental" or logic based intelligence, and doesn't necessarily have much to do with either people skills (Emotional IQ), ethical standards (Ethical IQ), creativity, or common sense. As a guide to potential success in academic pursuits, however, it is highly accurate. It is biased toward verbal skills, abstract symbology, and numeric competence, so environmental factors play a large part in the result. Considering that my Jungian typology puts me firmly in the "emotional" (value-based) sector of human type rather than the logical, it always surprises me how well I actually perform in logic-based IQ tests. Lash