Table 4.

Parameter estimates for final model (# 3) and effect sizes of observed changes based on ordinal 5-point response categories and of true changes when item means were modeled by specifying a vector of model intercepts in final CFA model

Item #	Parameter estimatesa					Effect sizes
	Γ 1 = Γ 2	τ 1 = τ 2	diag(Ω 12)	diag(Ω 11)	diag(Ω 22)	Observed item changes	True item changes
Item 2	1.000	0.762	.155	.701	.365	−.18	−.24
Item 6	0.731	0.594	.217	.565	.403	−.14	−.22
Item 10	0.872	1.042	.172	.816	.618	−.21	−.19
Item 16	1.439	1.202	.085	.731	.438	−.26	−.28
Item 20	1.394	0.993	.072	.903	.257	−.41	−.28
Item 23	1.236	0.942	.023	.570	.295	−.33	−.27
Item 29	1.188	0.719	.119	.550	.206	−.23	−.31
Item 32	1.047	0.735	.048	.579	.358	−.20	−.24
Item 35	1.010	0.784	.168	.736	.351	−.29	−.25
Item 38	1.219	0.811	.005	.484	.160	−.35	−.30
Item 42	0.752	0.480	.088	.308	.233	−.14	−.26
Item 43	0.787	0.471	.058	.265	.227	−.14	−.26
Item 47	1.064	0.760	.140	.490	.287	−.34	−.29
Item 48	0.600	0.332	.092	.222	.203	−.09	−.25
						OHIP-14 sum score	Latent factor mean
	α 1	α 2	Φ 12	Φ 11	Φ 22	Observed change	True change
	.000	−.246	.229	.441	.284	−.34	−.37

^aSubscripts refer to Time 1 and Time 2, respectively

Γ: factor loadings; τ: item intercepts; Ω: across-occasion and occasion-specific residual variances; α: latent factor means; Φ: variances and covariances among the common latent factors

Note: For the factor loadings (Γ), standard errors were ≤ .095. For the intercepts (τ), standard errors were ≤ .049. For the residual covariances (diag(Ω₁₂)), standard errors were ≤ .039. For the residual variances at Time 1 (diag(Ω₁₁)), standard errors were ≤ .067; at time 2 (diag(Ω₂₂)), standard errors were ≤ .054. The standard error of α ₂ equals .029; the standard errors of Φ ₁₂, Φ ₁₁, and Φ ₂₂ equal .033, .058, and .042, respectively