IQ Testing Without the Mythology

The AGCT was an effective wartime classifier, but its published metrics do not establish a reproducible pure-g measure across populations and eras.

IQ research has a stronger cumulative record than many areas of psychology, but reliability cannot substitute for reproducible methods, accessible evidence, and limits on interpretation.

IQ scores estimate selected cognitive performance within a specific instrument, session, and norm group. Six recurring limitations define what those scores can support.

IQ publishers can protect operational items while still exposing norming protocols, analytic decisions, uncertainty, and controlled-access data to independent auditors.

Institutions made intelligence testing valuable as a sorting system. Publishers and online platforms turned repeated scoring, interpretation, and status comparison into products.

IQ scores measure performance on selected cognitive tasks. Treating them as moral rank ignores character, competence, judgment, and useful work.

IQ tests predict some educational and employment outcomes. Those findings support specific uses, not sweeping claims about genetics, human worth, or social policy.

Twentieth-century cohort gains in IQ scores occurred too quickly for genetic change to explain. They show how norms, environment, and historical conditions shape measured performance.

Heritability describes variation within a population under particular conditions. It cannot determine an individual's potential, the cause of a group difference, or the effect of a policy.

Nonverbal items can reduce language demands. Testing instructions, schooling, puzzle familiarity, timing, norms, and score use remain culturally situated.

The general factor of intelligence, g, summarizes shared variance across cognitive tasks. It can support prediction, but it cannot by itself establish a causal mechanism, a complete cognitive profile, or a hierarchy of human value.

Norm tables embed consequential decisions about comparison groups, exclusions, demographic targets, updates, smoothing, and uncertainty. Those decisions determine what a reported IQ score means.

When one seller controls a test, its norms, scoring platform, manuals, and data access, commercial conflicts require independent audit, controlled data access, and traceable revisions.

Most online IQ tests lack the representative norms, controlled administration, test security, error estimates, and validation required for formal assessment.

IQ scores correlate with educational outcomes and some measures of job and training performance. Those predictions are probabilistic, context-dependent, and narrower than many claims made about them.

A defensible cognitive test states its purpose, discloses its norms and uncertainty, limits claims to validated uses, tracks revisions, and permits independent review.

Online IQ tests can provide useful estimates, but accuracy depends on item coverage, administration, norms, error, and retest controls.

The best online IQ test depends on the purpose. Compare broad, adaptive, and public-domain forms by what they administer and report.