. Author manuscript; available in PMC: 2009 Aug 21.

Published in final edited form as: Brain. 2008 Feb 7;131(Pt 3):665–680. doi: 10.1093/brain/awm336

Table 6.

Summaries of performance of proportional odds ordinal logistic regression models

Model	Likelihood ratio chi-squared (p-value) ^a	Concordance/ generalized AUC ^b	Generalized R²^c
Global cortical PiB only	12.2 (<0.0001)	0.75	0.27
Hippocampal W score only	29.2 (<0.0001)	0.84	0.55
Global PiB and W score^d	33.4 (<0.0001)	0.86	0.60

Note: Due to skewness, global PiB was log transformed

Likelihood ratio test versus the null model

This can be interpreted as the probability of correctly identifying which is the more clinically impaired patient from a pair of patients having different diagnoses using only the imaging measure(s) in the model

This can be interpreted as the model likelihood divided by the likelihood from a saturated, or "perfect fitting" model, after adjusting for model complexity. In some sense, what proportion of the observed data is "explained" by the model

Significantly better than PiB only model (p<0.001) and hippocampal W score only model (p=0.040) by likelihood ratio test