Friday, March 21, 2008

IQ Test Or IA Test?

By: Steve Gillman

An IQ test is supposed to measure one's "intelligence quotient," to provide a basic rating of how smart a person is. When we try to measure people in this way, we hope it will predict how well they will perform in either their profession or academic work. Of course we know that this doesn't work very well. We can all think of examples of highly intelligent people who are less successful than those with "weaker" minds.

In the world of academics, studies have shown that a student's habits of self discipline are far more likely to predict high grades than the score on an IQ test. Obviously the usefulness of the latter kind of testing is limited, but what is the alternative? On possibility is the "intelligence application," or "IA test.".

The IA Test

The idea here is not to measure how well you can use your brain on "paper problems." This measuring of "brain potential" is what the IQ test is about. An IA score would be a rating of how well you actually apply your intelligence in your life. As far as I know, nobody has yet developed a systematic test of this sort. What would it consist of, then?

Most likely it would start with a standard IQ test of some sort. That measure of "what you have" would then be used in conjunction with a measure of "what you do with it" to come up with an IA score. The first number (your IQ score) might be added to or subtracted from according to measures of "application." But what would we use for these? There are many possibilities.

If we used income as an "intelligence application measure," we might start with zero for an average income for that particular IQ level, and add or subtract points for being above or below that. This is based on the idea that being more intelligent should lead to a higher income if one is applying that intelligence well. We could divide IQ scores into ten levels, and for each, find the average income or create a scheme for what ones income "should be." Then we could add a point for each level a person is above their "presumed" income, or subtract one for each level below.

Of course, this measure assumes that higher income is either an objective value, or at least a goal of the person being tested. That brings up the first serious problem with this new kind of test. Do we want to use an objective set of values against which to measure how well people apply their intelligence, or should we measure according to the person's own stated goals? There are tough challenges with both approaches. The first requires some agreement on what is universally valuable. The second assumes that people really know what they value.

This is the problem with the all the possible measures. We can create measures of some sort to see how well a person does in relationships, health maintenance, creative output, and even happiness, but we cannot so easily say what the value of those things is, or how much the person being tested really desires those things. At first this problem of standards and measurements may make it seem that we cannot develop an intelligence application test. But we can choose a standard (or develop two tests). As for the measuring, we already do it.

If you've ever said about a person, "He doesn't use what he's got," or "She's so smart, but she doesn't do anything with it," you were measuring. You can't say such things without having some idea of what a person could be doing better. These kinds of comments are a measuring of performance against that idea or standard. Granted it isn't precise, but it acknowledges that such things are measured.

The IA test would simply make it more precise. A statistician will tell you that when things are difficult to measure, you don't give up: You just measure more. For example, if all you have is flawed rulers to measure a door with, you don't use one once. You can get closer to a true measurement by using them all several times and taking the average. If you are measuring the relative happiness of people in various groups, whether the groups are determined by country of residence or IQ level, you can to some extent overcome the flaws in measurement by making enough of them. When a group consistently scores higher after thousands are tested using many methods, the data becomes more accurate and useful.

At the level of the individual, then, we would also want to measure many things in many ways, because of the uncertainty of any one measurement. We might measure income, as noted, but then measure again according to what income the person thinks would be ideal. We could also adjust this for age. We might measure success in relationships in six different ways, and the application of intelligence to household chores in several ways. Using all of these various measures, we might arrive at a IA test which can assign a score that actually means something. Or maybe not.

Perhaps the idea of "intelligence application" is most useful not for developing a new test, but for pointing out how flawed some kinds of tests can be. I've seen the same man score 70 on one IQ test and 140 on another, and I know of a millionaire who can't read. If after generations of these tests being used they cannot be more accurate or predictive than that, there may not be too much hope for an IA test either. Life itself is the real "IA test," and perhaps for a long time to come the closest measurements we'll get will be the intuitive ones that lead to the comments, "He sure knows how to use what he's got," or "She just isn't living up to her potential."

Article Source:- Directory Submission & Sexy Deepika Padukone

Read More..
Blog Widget by LinkWithin

The News from Out There

Five Years Gold Price


gold price charts provided by goldprice.org