Perceived coercion in psychiatric hospital admission: validation of the French-language version of the MacArthur Admission Experience Survey

Table 3 Comparisons of model fit and IRT parameterisation for the AES

Model	χ²	df	p-value	RMSEA	90% C.I. for RMSEA	CFI	TLI
One-factor model	112.326	77	.005	0.055	0.031–0.076	0.987	0.985
Three-factor model	94.235	74	.056	0.042	0.000–0.066	0.993	0.991
IRT parameterisation		One-factor model			Three-factor model
		Item difficulty	Item discrimination		Item difficulty	Item discrimination
Item 1		−0.347	1.130		−0.334	1.238
Item 2		0.374	2.223		0.358	3.110
Item 3		0.613	−2.018		−0.591	2.510
Item 4		0.241	2.682		0.230	5.199
Item 5		0.733	−1.401		−0.713	1.530
Item 6		1.681	0.796		1.633	0.835
Item 7		−0.100	1.530		−0.097	1.724
Item 8		1.023	1.333		0.994	1.449
Item 10		0.503	2.315		0.495	2.614
Item 11		0.538	1.580		0.527	1.708
Item 12		0.070	1.530		0.068	1.724
Item 13		0.312	−1.649		−0.302	1.891
Item 14		−0.059	0.742		−0.057	0.788
Item 15		−0.180	1.151		−0.175	1.242

Note. IRT Item Response Theory, df degree of freedom, RMSEA Root Mean Square Error of Approximation, C.I. Confidence Interval, CFI Comparative Fit Index, TLI Tucker–Lewis Index

ISSN: 1471-244X