Matching IRT Models to Patient-Reported Outcomes Constructs: The Graded Response and Log-Logistic Models for Scaling Depression

Table 2 Item-scale correlations (corrected for item overlap), means, and response proportions

Item	R.drop	Mean	Response proportions
			0	1	2	3	4
1	.83	1.05	.47	.21	.18	.09	.05
2	.84	1.20	.42	.20	.21	.11	.06
3	.85	1.21	.40	.21	.22	.12	.05
4	.83	1.44	.28	.25	.28	.14	.06
5	.87	1.15	.43	.20	.21	.10	.06
6	.85	1.33	.35	.22	.24	.14	.06
7	.84	1.43	.28	.25	.28	.13	.06
8	.87	1.14	.44	.19	.20	.12	.05
Mean	.85	1.24	.38	.22	.23	.12	.06
SD	0.01	0.05	.07	.02	.03	.02	.00