Imagine a sheet of paper with one line on it: 'IQ — 92'. A number like this looks exact and final, and many parents read it as a verdict on their child. In fact it is the outcome of one short session, one set of tasks and a comparison with other children. Knowing how it is made helps you read it calmly.
A small sample of behaviour
An intelligence test is a small sample of what a child does on a given day: explaining words, spotting a pattern in a row of shapes, solving a puzzle, repeating a string of digits. The instructions, time limits and scoring rules are the same for everyone. Before a test is used, it is tried on a large group of children of each age, whose results become the norms. Your child is compared with the norms for their age, so the result is always relative: intelligence has no zero point, and the number only shows where a child stands among peers.
Since David Wechsler's scale of 1939, most tests have calculated the IQ (intelligence quotient) so that the average for each age is 100 and the standard deviation is 15 points. Scores follow a bell-shaped normal distribution: about 68% of people score between 85 and 115, and about 95% between 70 and 130. So a score of 92 means only that, on this test, the child is within the broad average range for their age. Nothing more.
Reliability and validity in plain words
Reliability answers the question: would the child get roughly the same score if tested again? Good individual tests are highly reliable, yet no score is exact. Tiredness, nerves or a noisy room can shift it, so a conscientious specialist reports a range in which the true score probably lies, not a single point.
Validity asks whether a test measures what it claims and predicts what it should. IQ scores do predict school results, but only moderately: many children with the same score get very different marks. Much stays outside the test: creativity, practical intelligence (the ability to handle real-life situations), motivation, character, a particular talent for music, drawing, sport or crafts. Nor does a number show how the child reached an answer, which is often the most useful thing to know.
Culture, language and change
Every test carries the culture in which it was made. As early as 1929 a researcher named Shimberg put together a test on which, unlike the usual ones, rural children scored higher than city children: what a test treats as 'easy' depends on the experiences it takes for granted. A child tested in a second language, or meeting such tasks for the first time, may show far less than they can do.
Scores also change. In a well-known Berkeley study, more than half of the children's scores shifted by about 15 points or more over the school years, and almost one child in ten by 30 or more. Across generations the change is clearer still: James Flynn showed that average scores rose by about 3 points per decade during the twentieth century in many countries. Genes cannot change that fast, so schooling, nutrition, health and a more complex everyday life must play a part. Heredity matters too, but heritable does not mean fixed.
A number is not a diagnosis
Identifying an intellectual disability takes far more than one score. Qualified specialists — a psychologist, a doctor and, where needed, others — study the child's developmental history, give individual tests and, above all, assess everyday functioning: communication, self-care, play and learning at home and at school. Weak school results are a different matter: they often disappear once their causes are dealt with. A score itself can also be pulled down by a different home language, low motivation or later maturing.
Good practice goes back to Alfred Binet. In 1905 he and Théodore Simon created their scale to find pupils who needed extra help, and Binet warned that results are provisional and that intelligence can be developed. Sulaiman al-Khodari al-Sheikh's book stresses the same: a test should help a child, not decide their fate. That is why Zehnhoi Gunogun does not give IQ tests. Our questionnaires describe a child's strengths so that parents, teachers and the child can talk and plan together; they produce neither an intelligence number nor a label.
If your child receives a test result
Ask calmly and specifically:
- What exactly was measured? Which test, which abilities, and was it given one-to-one?
- Under what conditions? In which language, how was the child feeling, had they met such tasks before?
- What is the margin of error, and which norms were used — from which country and which year?
- What will be done to help? What will change at school or at home, and when will progress be reviewed?
Never let your child hear the number as a verdict. A child who hears a low score may decide 'I'm stupid' and stop trying; one who hears a high score may start to fear any task that could prove otherwise. Talk instead about what went well, what was hard and what will help next. A test result is information about one day; your child is much more than that.