Scaled score versus IQ

A clinical intelligence battery (for example a Wechsler scale) is a timed appointment: many subtests, breaks, an examiner, often one to three hours depending on the edition and the person. Screening tools can be shorter. A 12- or 28-minute original quiz on this site is a practice session. It does not include a full subtest set, does not use a normative sample in the psychometric sense, and does not print a clinical IQ. Digit-symbol lite and sequence-completion lite on this site last minutes on purpose. Treat the clock as a toy timer, not a WAIS stopwatch.

Older “ratio IQ” compared mental age to chronological age. Most modern scales use a deviation IQ: your standing relative to a recent normative sample, often scaled to mean 100 and standard deviation 15 (sometimes 16). That number is only as good as the sample, the items, and the administration. Matrix-odd lite and pattern-matrices lite on this site do not estimate a deviation IQ. They do not have a normative sample in the publisher sense. A percentage correct on eight original items is a game score.

If a test is scaled so that the normative sample has mean 100, then 100 means “average in that sample, at that time, under those rules.” It is not a moral grade, a salary, or a diagnosis. It is also not what this site’s practice quiz reports as a clinical finding. Matrix-reasoning lite, digit-ordering lite and serial-subtract lite can print a hit count. They do not place you at 100 on a Wechsler metric. If an online page always gives you 100, ask whether any real norming happened.

Proctoring is not a nicety. An examiner verifies identity, presents items as the manual requires, records behaviour, and stops coaching. High-stakes uses (disability, court, some admissions) assume those conditions. Unproctored websites cannot. Maze-path lite, delayed-match lite and running-digit lite on this site are unsupervised by design. A roommate can shout answers. A second tab can search. That is why a screenshot is not a clinical IQ.

Publishers spend years showing that a tablet edition matches a booklet — or they publish separate norms. Input (pencil versus mouse), scrolling, time-outs, and screen glare change behaviour. Trail-making B lite, paper-folding lite and cube-nets lite on this site are browser originals. They are not “the computer version of WAIS.” Mode effects are one more reason a lite score is not a clinical IQ.