Table 5.

Inter-observer error between the main and second observer (n = 36).

	Mean (bias)	Mean 95%CI	Standard deviation- sd	Lower LE = Mean-1.96*sd	Higher LE = Mean+ 1.96*sd	Regression bias
M GAT pre - L GAT pre	0.028	−0.364–0.420	1.158	−2.243	2.298	b = −0.014; p = 0.829
M GAT post - L GAT post	0.056	−0.372–0.483	1.264	−2.421	2.532	b = −0.154; p = 0.137
M CT1 pre - L CT1 pre	0.472	−0.152–1.096	1.844	−3.141	4.086	b = 0.000; p = 0.999
M CT1 post - L CT1 post	0.333	−0.165–0.832	1.474	−2.555	3.222	b = −0.089; p = 0.346
M CT2 pre - L CT2 pre	0.028	−0.493–0.549	1.540	−2.990	3.046	b = −0.045; p = 0.493
M CT2 post - L CT2 post	−0.389	−0.968–0.190	1.712	−3.744	2.966	b = −0.183; p = 0.069

GAT, Goldmann applanation tonometer; CT1-CT2, convex tonometer; M, main observer; L second observer. LE, limit of agreement; p, probability value for regression; b, regression coefficient.

The mean difference between measurements was computed and the absence of differences was tested with the Wilcoxon or paired sample t-test; the Bland-Altman analysis was used; and the ICC was calculated based on absolute-agreement.

Present significant differences are marked in cursive bold.