IQ and reasoning explained

An IQ test samples cognitive abilities: reasoning with unfamiliar material, verbal comprehension, working memory and processing speed are the most common. It converts performance into a score relative to people of the same age, usually scaled so the average is 100. MedlinePlus describes IQ testing as standardized testing designed to measure intelligence relative to an age group. The word "sample" matters: a test covers some abilities, not everything a person can do.

Most tests set the average at 100 with a standard deviation of 15. About 68 percent of people score between 85 and 115, and about 95 percent between 70 and 130. Percentiles are often clearer: a score of 120 means performing above roughly 90 percent of the comparison group. Every score carries measurement error, which is why professional reports give a confidence interval rather than a bare number.

Some things a test measures respond to practice, and some of that practice transfers to real work: checking a rule against every case instead of the first one, writing intermediate steps instead of holding them in your head, estimating before calculating, separating what a text states from what you assume, and noticing when you have stopped reasoning and started defending a first guess. These habits show up in this guide's sections on pattern recognition, verbal and numerical skills, and test strategy.

Professional tests are generally reliable, and scores relate to educational outcomes at group level. Education itself raises measured scores, so causality runs in both directions. Associations with job performance are reported but their size is debated. Average scores rose across populations during the twentieth century, which is why tests are renormed. None of this supports predicting an individual's life from a number, and no test measures creativity, judgment built from experience, persistence or character.