. 2007 Sep 24;7:151. doi: 10.1186/1472-6963-7-151

Table 4.

Logistic regression models of the non-agreement between adolescents and proxy respondents (95%CI) in responses on different socio-economic variables

		Non-agreement on paternal education	Non-agreement on maternal education	Non-agreement on highest education	Non-agreement on paternal OSC	Non-agreement on maternal OSC	Non-agreement on highest OSC
Age	> 15	1^a	1^a	1^a	1^a	1^a	1^a
	12–15	0.83 (0.3 – 2.2)	0.47 (0.2 – 1.4)	0.76 (0.3 – 2.0)	0.73 (0.3 – 2.0)	0.44 (0.2 – 1.2)	0.38 (0.1 – 1.2)
Sex	Male	1^a	1^a	1^a	1^a	1^a	1^a
	Female	1.30 (0.5 – 3.3)	1.2 (0.4 – 2.9)	0.90 (0.3 – 2.3)	2.6 (0.9 – 6.9)	0.95 (0.4 – 2.5)	1.22 (0.4 – 3.5)
Family type^b	Standard	1^a	1^a	1^a	1^a	1^a	1^a
	Non-standard	1.15 (0.3 – 3.9)	2.28 (0.6 – 7.8)	0.63 (0.2 – 2.2)	4.98 (1.3 – 18.8)	3.0 (0.9 – 10.4)	4.63 (1.1 – 19.2)
Resilience	High (>p25)	1^a	1^a	1^a	1^a	1^a	1^a
	Low (≤p25)	0.65 (0.2 – 1.8)	1.25 (0.4 – 3.7)	0.95 (0.3 – 2.7)	1.31 (0.4 – 4.01)	0.69 (0.2 – 2.09)	0.85 (0.2 – 3.01)
Risks	Low (>p25)	1^a	1^a	1^a	1^a	1^a	1^a
	High(≤p25)	0.62 (0.2 – 1.8)	0.6 (0.2 – 1.9)	0.88 (0.3 – 2.6)	0.67 (0.2 – 2.1)	0.97 (0.3 – 2.8)	0.64 (0.2 – 2.2)
Academic achievement	High (>p25)	1^a	1^a	1^a	1^a	1^a	1^a
	Low (≤p25)	2.01 (0.6 – 6.2)	1.69 (0.5 – 5.6)	2.22 (0.7 – 6.8)	0.72 (0.2 – 2.2)	1.04 (0.3 – 3.2)	1.1 (0.3 – 3.8)

The dependent variables have been calculated as: 0 = agreement between adolescents and proxies, 1 = non-agreement. ^aReference category. ^bThe category of "non-standard families" comprises both "single-parent" and "other" families. Each model is adjusted for the rest of variables in the table. Higher scores in the CHIP-AE domains means better health (better Resilience and academic achievement, and low Risks)