PRI Efficacy Report
Monte Carlo simulation with 10,000 synthetic respondents validates the Psychedelic Readiness Index across six dimensions of preparedness.
Classification Accuracy: Likert vs. Forced-Rank
Forced-rank calibration nearly doubles classification accuracy by eliminating acquiescence bias and social desirability effects.
Dimension Priority Distribution
Likert scoring inflates “Medical” by 19.6%. Forced-rank produces near-uniform distribution.
Test-Retest Reliability
Correlation between first and second administration (n=2,000 retested)
Distribution Entropy
Higher entropy = more uniform distribution across dimensions (max: 2.585)
Methodology
The PRI validation uses a Monte Carlo simulation with 10,000 synthetic respondents across 6 readiness dimensions. Each respondent's answers are generated using dimension-specific trait distributions that model real-world response patterns including acquiescence bias and social desirability effects.
Key validation metrics:
- Classification Accuracy: Percentage of respondents correctly assigned to their true primary readiness dimension.
- Test-Retest Reliability: Stability of rankings across repeated administrations (n=2,000).
- Faking Resistance: Tested with 2,000 simulated fakers attempting to inflate scores.
- Distribution Entropy: Measures how uniformly respondents are classified across dimensions (max = log2(6) = 2.585).
The forced-rank calibration module eliminates first-dimension inflation bias (36.3% → 17.2%) by requiring explicit trade-off comparisons between dimensions.