Homework self-regulation strategies: a gender and educational-level invariance analysis

Psicologia: Reflexão e Crítica

Psychology: Research and Review

Table 4 Measurement invariance across genders (N = 1387) and educational levels (N = 1398)

Model	χ ²	df	p	χ ²/df	CFI	RMSEA	90% CI RMSEA	SRMR	BIC	ΔSB − χ ² (df)	ΔCFI	ΔRMSEA
Gender
Model 0: configural	1378.467	372	<.001	3.706	.903	.062	.059–.066	.059	80267.581	–	–	–
Model 1: metric	1408.690	393	<.001	3.584	.902	.061	.058–.065	.063	80141.507	24.302 (21)	.001	.001
Model 2: scalar	1545.190	414	<.001	3.732	.891	.063	.059–.066	.077	80138.773	147.374 (21)^***	.011	.002
Model 3^a: partial scalar	1528.940	413	<.001	3.702	.892	.062	.059–.066	.077	80127.547	128.712 (20)^***	.010	.001
Educational level
Model 0: configural	1392.982	372	<.001	3.745	.904	.063	.059–.066	.059	80759.100
Model 1: metric	1451.976	393	<.001	3.695	.901	.062	.059–.066	.068	80666.025	56.208 (21)^***	.003	.001
Model 2: scalar	1749.108	414	<.001	4.225	.875	.068	.065–.071	.080	80851.854	339.662 (21)^***	.026	.006
Model 3^b: partial scalar	1537.974	410	<.001	3.751	.894	.063	.059–.066	.071	80632.473	90.717 (17)^***	.007	.001

CFI comparative fit index, RMSEA root mean square error of approximation, SRMR standardized root mean square residual, BIC Bayesian information criterion, ΔSB − χ ² Satorra–Bentler scaled chi-square difference test
***p < .001
^aEstimating freely the intercept of item 14 in both samples
^bEstimating freely the intercepts of items 12, 14, 16 and 18 in both samples