In the 841-person Japanese-version analysis, BrainTypeIQ's overall IQ had reliability of .929 and a standard error of measurement (SEM) of about 4.0 points. Combining all nine tasks makes overall IQ more stable than any of the five domain scores.
Overall IQ is more stable than the individual domain scores
More than 2,000 Japanese-version completion records were checked for response conditions, repeat attempts, and the items and scoring conditions used. The analysis retained 841 people who had completed the current nine tasks for the first time under the same conditions.
| Statistic | Result | What it means when reading a score |
|---|---|---|
| Composite reliability of overall IQ | .929 | Score precision when all nine tasks are combined into overall IQ |
| Standard error of measurement (SEM) | About 4.0 points | The amount of measurement error in a single overall-IQ result |
Reliability of .929 shows that combining the nine tasks produces a highly consistent overall-IQ score. The model reflects how the tasks are actually administered and accounts for each task's reliability and the current composite calculation method.
SEM expresses the size of measurement error on the same scale as IQ. Overall IQ can therefore be read as an estimate that includes a range of several points, rather than as an exact value fixed to a single point.
Overall IQ shows the general level; the five domains show the broad pattern
| Result | Primary role |
|---|---|
| Overall IQ | Combines all nine tasks to show the general level of cognitive performance |
| Five domains | Show the broad pattern across Gc, Gf, Gv, Gwm, and Gs |
Because each domain is based on one or two tasks, use the five domain scores to identify broad relative highs and lows rather than treating one-point differences as a meaningful rank order.
How to read the BrainTypeIQ report explains how to combine overall IQ with the five domain scores. IQ test measurement quality explains reliability, SEM, and the other evidence needed to evaluate a test.
Reliability across the five domains ranged from .726 to .852
Reliability for Gc, Gf, Gv, Gwm, and Gs was also estimated from the current task structure and composite method.
| Domain | Contributing tasks | Reliability |
|---|---|---|
| Gc | Vocabulary (Antonyms) and Analogies | .852 |
| Gf | Matrix Reasoning and Figure Weights | .834 |
| Gv | Paper Folding and Visual Puzzles | .789 |
| Gwm | Computation Span and Visual Reversal | .726 |
| Gs | Symbol Search | .734 |
All five domain estimates exceed .70 and provide enough precision to identify broad patterns across cognitive abilities. Overall IQ combines information from all nine tasks and reaches .929, making it the most stable score reported by BrainTypeIQ.
The nine-task and five-domain design shows how tasks map to domains, while the BrainTypeIQ scoring method explains how task scores become the five domains and overall IQ.
Nine-task consistency was nearly identical in the earlier 420-person and later 421-person subsamples
The 841 participants were divided by testing period into an earlier group of 420 and a later group of 421, and the consistency across all nine tasks was calculated in each group.
| Data set | Standardized Cronbach's alpha across nine tasks |
|---|---|
| Earlier 420 participants | .844 |
| Later 421 participants | .846 |
| All 841 participants | .845 |
The earlier and later values were almost identical. The nine tasks showed the same overall consistency in participants tested at different times, so the result was not confined to one period of data collection.
The Japanese-version norming analysis uses the same 841 participants included in both analyses. The comparison norms and reliability evidence are therefore based on the same analysis sample.
The English version uses provisional scoring while its own data continue to accumulate.