. Author manuscript; available in PMC: 2018 Sep 13.

Published in final edited form as: Appl Res Qual Life. 2017 Apr 4;12(2):251–288. doi: 10.1007/s11482-017-9516-9

Table 8.

Feeling Tone Questionnaire (FTQ) 13 item set: Differential item functioning (DIF) results Race/Ethnicity subgroup comparisons

Item Description	IRTPRO		lordif		Magnitude (NCDIF)		Effect Size T1
Item Description	NHW vs. NHB	NHW vs. Hispanic	NHW vs. NHB	NHW vs. Hispanic	NHW vs. NHB	NHW vs. Hispanic	NHW vs. NHB	NHW vs. Hispanic
Are you feeling well?	U^*	U^*	NU; U	NU^*; U	0.0041	0.0143	0.0613	0.1066^†
Do you like being here?					0.0013	0.0002	0.0121	−0.0020
Are you feeling happy today?	U	U^*	U	NU; U^*	0.0028	0.0120	−0.0518	−0.1025^†
Do you have any pain?	Item excluded
Are you feeling bored?	U		U^*		0.0133	0.0015	−0.1068^†	−0.0301
Are people helpful here?	U^*	U	U^*	U	0.0089	0.0063	0.0822	0.0712
Do you feel lonely?	NU; U^*		NU; U^*	NU	0.0297	0.0050	−0.1622^†	−0.0426
Is there anything that stops you from doing what you want to do?	Item excluded
Do you have a good appetite?	U^*	U^*	NU; U^*		0.0111	0.0064	0.0984	0.0579
Do you like the people here?	U^*		NU; U^*	NU	0.0088	0.0060	0.0706	0.0582
Is anybody nasty to you?			U		0.0021	0.0031	−0.0455	−0.0494
Do you sleep well?	U^*	U^*			0.0009	0.0103	−0.0090	0.0044
Did you have a nice day yesterday?	U^*	U	U		0.0035	0.0013	−0.0444	0.0125
Did anything upset you yesterday?	U^*	U	NU^; U^	U^*	0.0137	0.0132	−0.0970	−0.1071^†
Do you have any trouble with your health?	Item excluded
Do you feel good about tomorrow?					0.0015	0.0006	−0.0279	0.0231

Item 7 – ‘Lonely’ had NCDIF values larger than the threshold (0.0240)

Asterisks indicate significance after adjustment for multiple comparisons.

^†

Indicates value above threshold of 0.10.

NU= Non-uniform DIF involving the discrimination parameters; U=Uniform DIF involving the location parameters.

For the lordif analyses, uniform and non-uniform DIF was determined using the likelihood ratio chi-square test. Uniform DIF is obtained by comparing the log likelihood values from models one and two. Non-uniform DIF is obtained by comparing the log likelihood values from models two and three. DIF was not detected using the pseudo R² measures of Cox & Snell (1989), Nagelkerke (1991), and McFadden (1974) or with the change in Beta criterion.