Table 6.

Diagnostic accuracy of the past tense marking index measure as a function of different reference standards.

Reference standard	Area under the curvea	Optimal cutoffb	Sensitivity	Specificity	Positive likelihood ratioc	Negative likelihood ratiod
I. Receiving SLP services	.671	95.50	.529	.776	2.36	.607
II. Parental ratings
CCC-2 ≤ 85	.752	89.00	.585	.829	3.42	.501
CCC-2 ≤ 80	.736	89.00	.657	.806	3.39	.426
III. Standardized tests and measures
CELF-4 ≤ 85	.840	92.50	.750	.835	4.55	.299
TEGI ≤ 85	.855	96.50	.713	.880	5.94	.326
NWR ≤ 85	.757	96.50	.652	.837	4.00	.416
CELF-4 ≤ 80	.855	92.50	.796	.803	4.04	.254
TEGI ≤ 80	.843	96.50	.716	.860	5.11	.330
NWR ≤ 80	.736	96.50	.658	.788	3.10	.434
IV. At least two of the following:
Receiving SLP services, CCC-2 ≤ 85,  CELF ≤ 85, TEGI ≤ 85, NWR ≤ 85	.803	96.50	.684	.877	5.56	.360
V. All of the following:
Receiving SLP services, CCC-2 ≤ 80,  CELF ≤ 80, TEGI ≤ 80, NWR ≤ 80	.909	89.00	1.0	.789	4.74	0

Note. SLP = speech-language pathology; CCC-2 = Children's Communication Checklist–Second Edition; CELF-4 = Clinical Evaluation of Language Fundamentals–Fourth Edition; TEGI = Test of Early Grammatical Impairment; NWR = Dollaghan and Campbell's (1998) nonword repetition task.

^a All areas under the receiver operating characteristic curve were significant at p < .001.

^b Determined using Youden index (J) where J = maximum {Sensitivity + Specificity − 1}

^c Positive likelihood ratio = Sensitivity / (1 − Specificity): values of 1 = “neutral,” 3 = “moderately positive,” ≥ 10 = “very positive.”

^d Negative likelihood ratio = (1 − Sensitivity) / Specificity: values of 1 = “neutral,” ≤ 0.30 = “moderately negative,” ≤ 0.10 = “extremely negative.”