In actual practice, things aren't quite so simple. There are several individually administered IQ tests, most notably the individual form of the Stanford-Binet IQ Test, which actually reflect this type of proportional calculation. An individual Psychologist will administer the test (with its 16 individual scales) to a single child. The test will then be scored and child's age in months will be factored into the computation of their IQ score. This type of testing is not done in the public schools, however. It takes far too much time to administer and score this test. If you want the accuracy of this type of test, you'll have to go to a private psychologist and pay for this individual attention.

The standardized paper and pencil tests created for mass testing programs typically uses a single standard age of ten and a half for all forth grade students. If your child is significantly older or younger than that, their IQ score may be seriously flawed. In fairness, you should understand that the average forth grader is ten and a half (fourteen and a half for eighth graders), and it is relatively easy to use the same standard chronological age for scoring all test. Indeed, with millions of tests to score it would be a logistical nightmare to calibrate them all individually.

On the other hand, as far as your child is concerned, this isn't the major problem with IQ testing. The real problem is how these test results are used. Because these tests are used year after year to test students at the same grade levels they produce a pattern of results that is remarkably stable and consistent. Because these age level scores do not change much from year, too many people in the testing profession have hastily concluded that individual scores do not change either. That is wrong, and the consequences of this error can be substantial. Individual IQ scores can change from year to year, sometimes a lot, particularly among elementary school students, and the change is usually for the better.

When an IQ test is created or revised for use in one of these national public school testing programs it must be standardized. This means that a fairly large population of students will take this test in conjunction with another previously standardized test which is used to validate the overall results. The results of the new test will be subjected to a complex pretest-posttest and split half correlation analysis. A reliability coefficient is then derived from these correlations and a conversion table created which will be used to assign an IQ score equivalent to every number of correct responses possible on the test. A child's IQ score is actually determined by using these conversion tables.

What does this IQ score really mean? The answer is not so simple. The reliability coefficient of each IQ test is actually a predictive indicator of each student's likely performance on future administrations of this test.

Let's say that your child recently took an IQ test and received an IQ score of 115. This is a score that is one standard deviation above the mean and indicates a performance level at the 85th percentile. This means that your child performed among the top 15% of all children at this grade level. You would therefore have every reason to be proud of this score, and your child's accomplishment. Now, let's take the Test Examiner's Manual and look up the reliability coefficient for this test. All of the standardized IQ test used in these national testing programs have reliability coefficients which fall within the range of +.65 and +.90, and typically near the center of this range.

Suppose that the reliability coefficient for the test your child just took is reported to be +.75. Since your child's IQ score is a predictor of future performance, we now know that if your child took this test again within a year that, 75% of the time, the second score would fall between 100 and 130. That is to say, one out of every four times a child takes this retest they will score in excess of one standard deviation more or less than their first IQ results. We also know from follow-up research on these subsequent tests that about 85% of the time when scores exceed a standard deviation from the first test, the second results are higher, not lower scores. This phenomena has been noted by professional test administrators as contributing to the "Flynn Effect" (IQ scores tend to increase over time). In fact, one study which compared the first and fifth grade test results for a group of both black and whites students found that the average fifth grade IQ test scores had improved for both groups (Deutsch and Brown).

There is one other serious statistical flaw in these overall IQ test results. Every standardized test which reports results through a standard deviation scale also produces its own imprecision index. This index is call the "standard error of measurement." All standardized IQ tests have a fairly consistent standard error of measurement of between four to five IQ points. This means that IQ differences of less than five points are meaningless. Five IQ points is one third of a standard deviation. This is a very large margin of error, and indicates a large degree of test imprecision.

All of this means that IQ tests are subject to a lot of factors which can influence an individual's IQ scores. By way of analogy, you should understand that if you drove your car through a police speed trap and the radar clocked you as doing 55 in a 55mph zone, you would probably feel relieved to know that you would not get a ticket. On the other hand, if you had the opportunity to take your car onto a race track to open it up and see what you could really do, that radar would undoubtedly clock you at a much higher rate of performance. The same thing is true for IQ test results. The crowded testing room, and its inherent tensions and pressure, is far from the idea environment for maximizing your child's intellectual performance. Indeed, most kids in that type of situation feel as if they are being watched by cops.

What should you be concerned about if your child is about to take an IQ test at school? There are a number of things to look out for. Let's examine the most important issues. There are four major points to consider.

First, where are the tests going to be given? Most frequently, these test are administered in the school auditorium or cafeteria. These are not ideal testing environments. These rooms always produce too much pressure on the students and too many distractions. Students perform best in familiar, friendly environments. They do best when the test are given in their own classrooms. Ask that you child be administered the test in their classroom.

Second, the identity of the test administrator is important. Alfred Binet found that a little encouragement significantly improved performance. It is therefore important that all of the test administrators assume a facilitative countenance. Of course, it is best if these administrators are not strangers to the children. Ideally, the students' own teachers will be used because that will provide maximum continuity and support for the children. If outsiders must be brought in to help administer these tests it is important they match the students' ethnic and cultural background as much as possible. Studies have shown that the greater the difference between the background of the administrators and the student the lower student score will (Watson).

Third, is the question of the cultural bias of the test itself. This is a very complex issue. The question is not, "are IQ tests culturally biased?" That is indisputable. The real issue is, "HOW culturally biased are they?" Bear in mind, some of these tests are much more biased than others. The proof of this bias is found when IQ tests are statistically compared to each other (and to other standardized achievement tests). A correlation called "the goodness of fit" test is used to evaluate these comparisons. The standard public school IQ test consistently reports the goodness of fit correlations to be from +.10 to +.16 (that is only 10% to 16%). This is barely more than no correlation at all, and dangerously close to random chance. If there was no bias and the uniform measurement of the same dimensions of intelligence on all of these tests the "goodness of fit" correlation would consistently be reported in the +.80 to +.90 range (Gould).

While it is true that some professional test administrators dismiss the research findings on cultural bias (because their jobs are threatened), the test manufacturers are not so cavalier in their disregard. For instance, the "Examiner's Manual" for the Lorge-Thorndike Intelligence Test states:

"To evaluate the ration validity of an intelligence
test one can examine the items to see if they require a
pupil to make responses which one would call 'intelligent'.
The items for the Lorge-Thorndike Intelligence Tests were
selected so that, for the most part, they deal with
relationships. In answering most of the items a pupil is
required to find a principle and then apply it. The tests,
then, have been designed to measure reasoning ability"(14).

The question we must therefore ask is, are these questions of intelligence appropriate for the specific cultural background of the pupils in your child's school? In one research study, John Garcia analyzed the results of a public school district's administration of a typical IQ test and its usual reported IQ differences of 15 points between the white students and the black and hispanic students. After isolating the test items with strong anglo-biases, he re-scored the all of the tests for the entire district and found that the new average score for all three groups was within the range of a single point. He found that with no cultural bias on an IQ test, there is no difference in the average group scores for white, black and hispanic students.

The most specific examination of the forms of cultural bias found on these IQ tests was the content analysis conducted by Drew Hurley (1980). He identified items which: displayed implicit prejudice toward minorities, required the identification of uncommon symbols, made use of very obscure terms and concepts, required an ability to interpret poorly written statements, had a critical word misspelt, required a precise ability to distinguish between homonyms, presented information which was grossly out of date and unrealistic, and asked questions for which there was no correct answer.

So what is a parent to do? First of all, ask that a school advisory committee be allowed to screen all of the available tests that are available to be administered at your child's school, and then recommend the most appropriate test for use in your child's school. If there is no school advisory committee, get involved and form one. Conscientious school officials want parents to be directly involved in matters of school policy because it legitimizes the performance of their implementation responsibilities, and it helps the students perform better. When parents are directly involved in school policy matters their children do better in school.

What should an advisory committee, composed of lay people with no specific training in educational testing, look for? There is a simple short cut to screening these tests that can be used with remarkable success. These are written tests, and are therefore vocabulary dependent. The committee needs to slowly go over each test completely. Be sure to look at both the instructions and the content of each of the items. The instructions should be clearly understood, straight-forward, and offer at least a hint of positive encouragement. Among the items, look specifically at the level of vocabulary mastery that is necessary to successfully answer these question. Look carefully at the test's use of symbols, formula, and unusual terms. Also be on watch for culturally elitist references to "summer cottages at the beach", "ski vacations to Aspen", "taking the commuter shuttle", and similar expressions of social class bias. After comparing all of the available tests, your advisory committee is now in the position to recommend the one test which appears to be the most appropriate for the level and style of instruction, curriculum content and cultural composition of your child's school. After this recommendation is made, the committee should also write a single standard letter explaining why they recommended the test they selected and what features they disliked the most on each of the other tests. Mail a copy of this letter to all of the test manufacturers and keep several copies on file at the school office. The test makers will get the message.

The final issue of great importance that directly influences the IQ performances of children is how adults convey performance expectations to the children. Elementary school children are remarkably adept at picking up the expectations communicated to them by parents, teachers and other significant adults who they regularly come in contact with. Because children want to please these people who are so important to them they will fulfill these expectations. In social science terminology, this is the self- fulfilling prophecy, and it has a major direct bearing on IQ test performance.

The definitive research study of the role of the self- fulfilling prophecy on IQ test scores was conducted by Robert Rosenthal and Lenore Jacobson. They chose two school in the same California school district that were as closely matched as possible in terms of their number of students, ethnic composition, social class indicators and curriculum. One of these schools became the control group; the other was the experimental group. In the Spring of 1964, both groups were given the "Flanagan Test of General Ability," but the name on the test booklets had all been changed to the "Harvard Test of Inflected Acquisition." The tests were then scored. Meanwhile, the school year ended and everyone when on vacation.

At the experimental school, when the teachers arrived a week before the Fall classes began to plan their curriculum, they were casually informed that the test they had administered the previous spring was able to predict intellectual bloomers from among their students. In confidence, each teacher was given the names of those students in their class who were predicted to blossom forth and experience a great burst of intelligence during the coming year. The teachers were also cautioned that this was confidential information and that they were not at liberty to discuss it with anyone.

In point of fact, Rosenthal and Jacobson used a table of random numbers to select 20% of the students from each class to be designated as one of these "Bloomers." This selection process insured that every child at that school had an equal opportunity to be selected as a "bloomer." None the less, the teachers daily went to work with the expectation that this blossoming of intelligence within their classrooms was imminent, and they even knew who to expect it from. Obviously, this prophecy had a profound effect on the dynamics within the class. Teachers became the model of nurturing facilitators. Each classroom became a marvelous laboratory of discovery, understanding and development.

At the end of the school year, the tests were given again to both groups. Then Rosenthal and Jacobson spent the summer pouring over the results. The children in the control group performed exactly has children usually do from year to year on this kind of test. The first grade group improved their IQ scores an average of 12.0 points; the second grade gained 7.0; the third grade gained 5.0; and the forth grade students gained 2.2 points. These results are fairly typical among elementary school children.

At the experimental school, however, the results for the first two grades were phenomenal. The average increase in the IQ scores of these first graders was 27.4 points, and the second graders gained 16.5 points. It is true that the third and forth graders fell off to the more normal levels of 5.0 and 5.6 point gains, but there is an open question about how much better they might have done had they experienced a comparable style of instruction when they were in the first and second grades.

Among the first and second grade classes, Rosenthal and jacobson reported:
"Less than half of the control group children gained 10
or more total IQ points, but about four out of five
experimental group children did. Every fifth control group
child gained 20 or more total IQ points, but nearly every
second experimental group child did. While one out of
twenty control group children gained thirty or more total IQ
points, one out of five experimental group children did" (263).

Just how smart do you want your child to be? The chances are they will fulfill exactly the expectations you communicate to them. These one year gains among these first and second grade students should impress anyone. What's more, these children are likely to go right on continuing to steadily gain IQ points from year to year as long as they stay in school. They have learned how to learn and thereby become more intelligent.

What does this mean for you and your child? There are two essential things that you can do to help him or her become a much more successful student. First, you must make him or her develop their vocabulary. IQ tests are word tests. Words are what comprise your vocabulary. As you gain vocabulary mastery you will improve your performance on word based IQ tests. Ask your child to start a vocabulary notebook and learn 10 news words that they write, along with its definition, in their notebook everyday. Every afternoon or evening, ask them to show you their word list for the day. Praise and encourage them regularly. More importantly, restrict their play and TV time until they have completed this task.

The other thing that you can do is take an active positive role in your child's education both at school and at home. Make sure that your child knows that their educational success is important to you. As much as it is possible, every day ask how things went at school. Discuss these matters and go over their daily new word list. Then READ TOGETHER. In this area of research, the one family variable that has the highest correlation with improving IQ scores is joint reading. When a child and parent read aloud to each other the child shows the greatest gains in IQ scores (Allport). As children become more articulate their grasp of syntax and structure (because reading is linear and sequential) gives them a command of the language. This increasing skill at communications coupled with their ever expanding vocabulary puts them on the fast track to improving their intelligence.

One other thing. When the IQ test results finally come back to the school and you pry them out of the school counselor's tight-fisted hands, be sure to praise your child for the results, what ever they are. But remember this was just the first try. It was like the first time you drove through that police radar trap and you were actually glad that you weren't speeding. Now, take this opportunity to set goals for the next time your child is scheduled to take an IQ test. Except that the next time the test is administered, your child, the driver, is going to take this vehicle onto that racing oval to open it up and see what (s)he can really do. And you are going to be the pit crew chief to make sure they are really ready to give it their best run. Make this a positive experience of achievement and success for both of you.

Finally, there is one other issue related to this discussion of IQ testing that should be mentioned. College students are often subjected to an intermittent barrage of IQ and achievement tests. These tests are usually administered without providing a complete explanation to the students and you are almost never privy to the results. The reason for the vagueness of the explanation is because professors do not want to produce a "Hawthorne effect" which will contaminate the results of the study, and the results of the study may not be immediately known. It often takes a year or longer to completely analyze all of the variables involved in a complex study, and write the results up in a form suitable for publication. Under these circumstances, just remember that you'll be an anonymous participant. Therefore, relax, do the best you can, and try to enjoy the experience.

As a college student, if you are really serious about improving your IQ, rest assured, it can be done. The average college student gains nearly three quarters of an IQ point per year while they are in college (Hurley, 1970), but much larger gains are possible. You, too, can work to steadily improve your vocabulary. Many of your courses will supply you with "word lists" that will be of great benefit. Write these words and their definitions in your vocabulary journal. There is something else you can also do: outline your textbook readings. As long as you are still enrolled in your required English composition classes, you should even make it a point to use full sentence outlines. Outlines are really very good for you. Now only do they reinforce your developing use of this vocabulary, but they also provide you with a visual glimpse of the organization and structure of the material you are dealing with. The outline allows you to see how everything is connected and how component parts all fit together.

Some colleges and universities have a student testing center on campus where you can be tested, and monitor the progress of your IQ gains. Sadly, many colleges do not have these facilities. However, this is not a serious problem, it just makes the task a little more difficult. Every college has a few "test buffs" hidden in their faculty. Ask your professors if they know anyone on the faculty who periodically gives their students IQ, achievement, or personality tests. The changes are a few names will quickly emerge in the education, psychology and sociology departments. If all else fails, go to these department offices and ask the secretaries. When you find the person you want, introduce yourself and ask if they would consider administering a test to you.

If you are a serious college student with the ambition to aspired to an academic or profession career, then you need to learn who your friends on the faculty are. Within your academic major, examine the faculty. For which professors do you earn the best grades? Which professor has the personality you like best and are most comfortable with? Choose the one you prefer. Visit their office. Try to establish a personal relationship. Find out what their interests are outside of class. See if you can find a common ground for developing a continuing dialog with this person. As this relationship develops, let it be known that you would welcome this professor's efforts to "mentor" you in the department.

Having a faculty mentor will not only help you succeed at graduating with your major, it will also provide you with recommendations, and a possible fellowship or assistantship, for graduate school. If you are accepted in this role, you will assist the professor in a number of routine, mundane and boring ways (for which they will be very grateful). You will also gradually learn the politics with the department and the specific emphasis that various faculty members place on different courses and issues they believe to be important. All of this will help you select the career option with the greatest advantage for you.

Return To Beginning of "IQ and You"

Go To "The Racist Curve?"