Assess App/More/Insights
Insights
Classical item analysis, reliability, fairness and AI interpretation
Demo Simulation
Ask natural-language questions grounded in your charts and records:
Tooltip: facility is the proportion correct (0.3 to 0.8 ideal); discrimination at or above 0.30 is acceptable. Items bottom-left need review.
- Cronbach α
- SEM
Q-2073First-languageDIF 0.18
Q-2076First-languageDIF 0.22
Q-2052GenderDIF 0.05
A DIF flag above 0.15 suggests construct-irrelevant variance; review item wording.
CV
78%
Resp
82%
GI
69%
Neuro
74%
Endo
66%
Renal
58%
Q-2041, Q-2067, Q-2075 show discrimination shifts above 0.10 vs prior sittings — possible cueing or cohort effect. Confidence 0.79
Renal outcome mastery lowest at 58% — consider blueprint reinforcement or remediation. Confidence 0.83
Overall reliability stable at 0.89, up from 0.87 prior sitting.