Online IQ Tests vs. Clinical Assessments
Online IQ tests — including the one on this site — are screening tools, not clinical assessments. They are useful for self-curiosity, for tracking your own performance over time on a consistent format, and for getting a rough estimate of where you sit in the population. They are not appropriate substitutes for a clinical evaluation if you need a score for school placement, disability accommodation, or workplace selection.
Online IQ tests — including the one on this site — are screening tools, not clinical assessments. They are useful for self-curiosity, for tracking your own performance over time on a consistent format, and for getting a rough estimate of where you sit in the population. They are not appropriate substitutes for a clinical evaluation if you need a score for school placement, disability accommodation, or workplace selection.
Frequently asked questions about IQ testing fall into a few clusters: what does the score mean, how reliable is it, can it change, how does it compare to other constructs, and how should it inform major decisions. The answers to these questions are well-established in the research literature but poorly communicated in popular media.
On Online IQ Tests vs. Clinical Assessments, the empirically supported answer differs from the most popular folk answer in important ways. The folk answer is usually based on a few salient examples, anecdotes, or oversimplifications of one study; the research answer integrates effect sizes across hundreds of studies and identifies the conditions under which each finding holds.
One of the persistent communication problems is that the relevant effect sizes are typically modest but reliably non-zero. This is hard to summarize. 'IQ predicts X' is true if the correlation is reliably non-zero, but the unaided reader will often interpret it as deterministic prediction. 'IQ does not predict X' is true if the correlation is small, but the unaided reader will often interpret it as no relationship at all.
The honest answer for almost every IQ-related question is: 'the relationship exists, the effect size is moderate, and individual variation around the trend is large enough that the trend is not predictive at the individual level'. This is unsatisfying but accurate.
If you have a specific question about this topic that is not covered here or in the linked articles, the American Psychological Association's task force reports and the open-access journal Intelligence are reliable starting points for further reading. Avoid popular-press books that frame the topic in deterministic or partisan terms.