Table 2 Logistic Regression Analysis Results
From: Predicting Relapse to Alcohol and Drug Abuse via Quantitative Electroencephalography
Analysis #1: Variables Entered: MAST, DAST-10, FHA × Conduct Disorder | |||
|---|---|---|---|
Observed Outcome → | Abstinent | Relapse | |
Predicted Outcome | |||
Abstinent | 61% | 48% | |
Relapse | 39% | 52% | |
Avg. Predictive Accuracy = 56.5%, κ = 0.10 (P = N.S.) | |||
Analysis #2: Variables Entered: Fast β Relative Power, MAST, DAST-10, FHA × Conduct Disorder | |||
F-to-Enter Statistics | |||
Fast β relative power: F = 7.74, P = 0.007 | |||
FHA × Conduct Disorder: F = 2.42, P = N.S. | |||
MAST: F = 2.02, P = N.S. | |||
DAST-10: F = 0.01, P = N.S. | |||
Observed Outcome → | Abstinent | Relapse | |
Predicted Outcome | |||
Abstinent | 84.7% | 39.2% | |
Relapse | 15.3% | 60.8% | |
