University students’ intentions to learn artificial intelligence: the roles of supportive environments and expectancy–value beliefs

Table 2 Measurement invariance across field of study, gender, and year level

Models	χ²(df)	CFI	TLI	RMSEA	SRMR	ΔCFI	ΔRMSEA	ΔSRMR
Measurement invariance across field of study
Humanities and social sciences	349.814 (159)***	0.920	0.904	0.075	0.057	–	–	–
Sciences	304.985 (159)***	0.945	0.934	0.061	0.049	–	–	–
Configural invariance	654.799 (318)***	0.933	0.920	0.068	0.053	–	–	–
Metric invariance	677.229 (333)***	0.931	0.922	0.067	0.062	0.002	0.001	− 0.009
Scalar invariance	693.778 (348)***	0.931	0.925	0.066	0.063	0.000	0.001	− 0.001
Measurement invariance across gender
Female	402.161 (159)***	0.928	0.914	0.070	0.049	–	–	–
Male	377.063 (159)***	0.891	0.870	0.087	0.065	–	–	–
Configural invariance	779.224 (318)***	0.914	0.897	0.077	0.055	–	–	–
Metric invariance	791.171 (333)***	0.915	0.903	0.075	0.060	− 0.001	0.002	− 0.005
Scalar invariance	852.574 (348)***	0.906	0.897	0.077	0.063	0.009	− 0.002	− 0.003
Measurement invariance across year level
Junior (year 1 and 2)	368.647 (160)***	0.917	0.901	0.078	0.055	–	–	–
Senior (year 3 and 4)	377.135 (160)***	0.926	0.913	0.070	0.053	–	–	–
Configural invariance	745.782 (320)***	0.922	0.907	0.073	0.054	–	–	–
Metric invariance	780.434 (335)***	0.918	0.907	0.073	0.067	0.004	0.000	− 0.013
Scalar invariance	793.544 (350)***	0.919	0.912	0.072	0.066	− 0.001	0.001	0.001

***p < 0.001; CFI = comparative fit index; TLI = Tucker Lewis index; RMSEA = root mean square error of approximation; SRMR = standardized root mean square residual