Child and Adolescent Psychiatry and Mental Health

Table 2 Measurement invariance models between parent and self-report

From: Parent-adolescent informant discrepancy on the Strengths and Difficulties Questionnaire in the UK Millennium Cohort Study

	Fit			Fit difference
	CFI	RMSEA	SRMR	ΔCFI	ΔRMSEA	ΔSRMR
Conduct
Configural	0.934	0.038	0.030	–	–	–
Metric	0.901	0.044	0.043	0.032	0.006	0.014
Metric partial	0.932	0.037	0.032	0.002	0.001	0.003
Scalar	0.862	0.050	0.044	0.070	0.013	0.011
Scalar partial	0.925	0.037	0.033	0.006	0.001	0.001
Emotion
Configural	0.968	0.048	0.028	–	–	–
Metric	0.967	0.046	0.030	0.001	0.002	0.002
Scalar	0.883	0.082	0.067	0.084	0.036	0.037
Scalar partial	0.962	0.048	0.034	0.005	0.002	0.004
Peer
Configural	0.973	0.028	0.019	–	–	–
Metric	0.955	0.034	0.030	0.017	0.006	0.011
Scalar	0.766	0.074	0.058	0.189	0.040	0.028
Scalar partial	0.949	0.036	0.031	0.007	0.001	0.001
Pro-social
Configural	0.984	0.027	0.018	–	–	–
Metric	0.969	0.035	0.031	0.015	0.008	0.014
Scalar	0.865	0.068	0.052	0.104	0.033	0.021
Scalar partial	0.957	0.039	0.034	0.012	0.004	0.003

Bold indicates selected model; Conduct problems: the loadings for item 12 (“often fights”) were freed from the metric model, the intercepts for item 5 (“temper/angry”) and item 18 (“lies/cheats”) were freed from the scalar model; Emotion symptoms: the intercepts for item 8 (“many worries”) and item 16 (“nervous/clingy”) were freed from the scalar model. Peer problems: the intercepts for item 14 (“generally liked”) and item 19 (“picked on”) were freed from the scalar model; Pro-social behaviour: the intercepts for item 1 (“considerate/nice”) were freed from the scalar model

Back to article page

ISSN: 1753-2000

Contact us

Submission enquiries: journalsubmissions@springernature.com