Table 8.

Measurement invariance tests of science identity latent growth curves for Latinx students

Time invariance
Invariance Model	Robust χ2 goodness of fit			DIFFTEST			CFI	ΔCFI	RMSEA	ΔRMSEA
	Value	df	p	Value	df	p
Configural	8.87	2	0.01	—	—	—	1.00	—	0.032	—
Weak	7.87	4	0.10	0.94	2	0.63	1.00	<0.001	0.017	0.015
Strong	74.84	10	<0.001	59.69	6	<0.001	0.999	0.001	0.044	0.027
Partial Stronga	27.59	6	0.0001	15.81	2	0.0004	1.00	<0.001	0.033	0.016
Gender invariance
Configural	33.08	12	0.001	—	—	—	1.00	—	0.033	—
Weak	30.59	15	0.01	1.34	3	0.72	1.00	<0.001	0.025	0.008
Strong	86.54	24	<0.001	47.30	9	<0.001	0.999	0.001	0.040	0.015
Partial Strongb	69.84	23	<0.001	33.96	8	<0.001	0.999	0.001	0.035	0.010
College generational status invariance
Configural	12.14	4	0.02	—	—	—	1.00	—	0.035	—
Weak	12.58	7	0.08	2.87	3	0.41	1.00	<0.001	0.022	0.013
Strong	22.38	16	0.13	11.20	9	0.26	1.00	<0.001	0.016	0.006

Note. ^aOne threshold for indicator 1 and one threshold for indicator 2 were freely estimated across time. ^bOne threshold for indicator 1 was freely estimated across groups

SOURCE: U.S. Department of Education, Institute of Education Sciences, National Center for Education Statistics, High School Longitudinal Study of 2009 (HSLS:09), Base Year, First Year Follow-Up, Second Year Follow-Up