Five categories
Every attempt includes three questions from logic, pattern, numerical, spatial, and analogy reasoning. The question bank is larger than a single test and avoids duplicate IDs within an attempt.
Randomization
Questions are sampled independently within each category. Answer choices are shuffled after selection, while the correct answer stays attached to its text.
Scoring
Accuracy is converted with 100 + (accuracy − 50) × 0.6, bounded to 70–130. Percentile bands and classifications are approximate references, not population norms.
Limitations
Three questions per category cannot provide precise measurement. Language, attention, fatigue, device, familiarity, and environment affect results.