Results

TestROC

Procedure Notes

The TestROC optimal cutpoint analysis has been completed using the following specifications:

Measure Variable(s): THRIL

Class Variable: outcome

Positive Class: Yes

Method: maximize_metric

All Observed Cutpoints: TRUE

Metric: sum_sens_spec

Direction (relative to cutpoint): >=

Tie Breakers: c

Metric Tolerance: 0.05

For more information on how calculations are performed and interpretting results, please see the documentation

Results Table

Scale: THRIL
Cutpoint	Sensitivity (%)	Specificity (%)	PPV (%)	NPV (%)	Youden's index	AUC	Metric Score
3.55	100%	0%	50%	NaN%	0.0000	0.8806	1.0000
3.8	100%	3.33%	50.85%	100%	0.0333	0.8806	1.0333
4.45	100%	6.67%	51.72%	100%	0.0667	0.8806	1.0667
4.5	96.67%	10%	51.79%	75%	0.0667	0.8806	1.0667
4.58694426685242	96.67%	16.67%	53.7%	83.33%	0.1333	0.8806	1.1333
4.8	96.67%	20%	54.72%	85.71%	0.1667	0.8806	1.1667
4.9	93.33%	20%	53.85%	75%	0.1333	0.8806	1.1333
5.3	93.33%	23.33%	54.9%	77.78%	0.1667	0.8806	1.1667
5.30659641660015	93.33%	26.67%	56%	80%	0.2000	0.8806	1.2000
5.5	93.33%	30%	57.14%	81.82%	0.2333	0.8806	1.2333
5.58	93.33%	50%	65.12%	88.24%	0.4333	0.8806	1.4333
5.66006390778403	93.33%	53.33%	66.67%	88.89%	0.4667	0.8806	1.4667
5.74836209039055	93.33%	56.67%	68.29%	89.47%	0.5000	0.8806	1.5000
5.9	93.33%	60%	70%	90%	0.5333	0.8806	1.5333
6	93.33%	66.67%	73.68%	90.91%	0.6000	0.8806	1.6000
6.21590313071122	93.33%	70%	75.68%	91.3%	0.6333	0.8806	1.6333
6.6	90%	70%	75%	87.5%	0.6000	0.8806	1.6000
6.65	90%	73.33%	77.14%	88%	0.6333	0.8806	1.6333
6.8	90%	76.67%	79.41%	88.46%	0.6667	0.8806	1.6667
6.80946205635812	86.67%	86.67%	86.67%	86.67%	0.7333	0.8806	1.7333
6.9	83.33%	86.67%	86.21%	83.87%	0.7000	0.8806	1.7000
7	76.67%	86.67%	85.19%	78.79%	0.6333	0.8806	1.6333
7.04067316411162	76.67%	90%	88.46%	79.41%	0.6667	0.8806	1.6667
7.2	73.33%	90%	88%	77.14%	0.6333	0.8806	1.6333
7.25	70%	90%	87.5%	75%	0.6000	0.8806	1.6000
7.32	63.33%	90%	86.36%	71.05%	0.5333	0.8806	1.5333
7.5	56.67%	90%	85%	67.5%	0.4667	0.8806	1.4667
7.54	53.33%	93.33%	88.89%	66.67%	0.4667	0.8806	1.4667
7.58054196680174	53.33%	96.67%	94.12%	67.44%	0.5000	0.8806	1.5000
7.61469098223	50%	96.67%	93.75%	65.91%	0.4667	0.8806	1.4667
7.70671241715943	46.67%	96.67%	93.33%	64.44%	0.4333	0.8806	1.4333
7.9	43.33%	96.67%	92.86%	63.04%	0.4000	0.8806	1.4000
7.96792992880786	40%	96.67%	92.31%	61.7%	0.3667	0.8806	1.3667
8.18291986862297	36.67%	96.67%	91.67%	60.42%	0.3333	0.8806	1.3333
8.40157557717447	33.33%	96.67%	90.91%	59.18%	0.3000	0.8806	1.3000
8.41842667276905	26.67%	96.67%	88.89%	56.86%	0.2333	0.8806	1.2333
8.5	23.33%	96.67%	87.5%	55.77%	0.2000	0.8806	1.2000
8.85	23.33%	100%	100%	56.6%	0.2333	0.8806	1.2333
8.9	20%	100%	100%	55.56%	0.2000	0.8806	1.2000
9.25	16.67%	100%	100%	54.55%	0.1667	0.8806	1.1667
9.5	3.33%	100%	100%	50.85%	0.0333	0.8806	1.0333

ROC Curves

ROC Curve: THRIL

Binomial Logistic Regression

Model Fit Measures
Model	Deviance	AIC	R²_McF	R²_T
1	53.3593	57.3593	0.3585	0.4401

Model Coefficients - outcome
Predictor	Estimate	SE	Z	p
Intercept	8.9558	2.2892	3.9122	< .0001
THRIL	-1.3309	0.3353	-3.9696	< .0001
Note. Estimates represent the log odds of "outcome = No" vs. "outcome = Yes"

Prediction

Classification Table – …
	Predicted
Observed	Yes	No	% Correct
Yes	27	3	90.0000
No	7	23	76.6667
Note. The cut-off value is set to 0.5

Predictive Measures
Accuracy	Specificity	Sensitivity	AUC
0.8333	0.9000	0.7667	0.8806
Note. The cut-off value is set to 0.5

ROC Curve

Independent Samples T-Test

Independent Samples T-Test
		Statistic	p
THRIL	Mann-Whitney U	107.5000	< .0001
Note. Hₐ μ _No ≠ μ _Yes

R

    AUC           p-value 
_____________________________ 
   0.8806 	 4.177e-07