Skip to main content
. Author manuscript; available in PMC: 2010 Jan 14.
Published in final edited form as: Learn Disabil Res Pract. 2005 Aug 1;20(3):142–155. doi: 10.1111/j.1540-5826.2005.00129.x

TABLE 4.

Illustration of the Association Between the Four Indices of Prediction Accuracy, Using Grade K or 1 Data and Grade 2 and 3 MLD Status as the Gold Standard for Presence of MLD

Test Results = Positive (MLD) Test Results = Negative (Not MLD)
Part 1: Definitions of Accuracy Indices
MLD True positive False negative Sensitivity: P(test+|MLD+)
Not MLD False positive True negative Specificity: P(test−|MLD−)
Positive predictive value P(MLD+| test+)* Negative predictive value P(MLD−| test−)

MLD Status at Grade 3 Test Positive (MLD) Test Negative (Not MLD)

Part 2: Accuracy Based on Kindergarten Predictor Variables
MLD 26 15 Sensitivity: 0.63
Not MLD 61 104 Specificity: = 0.63
Positive predictive value = 0.30 Negative predictive value = 0.87

MLD Status at Grade 3 Test Positive (MLD) Test Negative (Not MLD) Based on First Grade Data

Part 3: Accuracy Based on First-Grade Predictor Variables
MLD 12 29 Sensitivity: 0.29
Not MLD 11 159 Specificity: 0.94
Positive predictive value = 0.52 Negative predictive value = 0.85
*

P(MLD+|test+) refers to the probability that a child is truly positive for MLD given that the test result for MLD is positive.