Table IV.

Adjusted odds ratio (OR) and 95% confidence interval (CI) for education, household income and self-rated health when 4 partially overlapping Sámi populations aged 36–79 years as of 2003–2004 in 17 municipalities in Norway north of the Arctic Circle are related to each population's respective non-Sámi proportion of the study sample^*

	OR	95% CI	p
Self-reported length of education: >12 vs. ≤12 years
G1) Resident in the Language Area – 8 municipalities (n=5,677)	0.96	0.89–1.05	0.386
I1) Sámi linguistic connection≈objective criterion# (n=5,249*)	0.81	0.75–0.88	**
I2) Self-identification as Sámi=subjective criterion (n=3,112*)	1.09	0.99–1.21	0.070
I3) Sámi as active language (n=2,614*)	0.87	0.78–0.97	***
Self-reported household income: >300,000 vs. ≤300,000 NOK
G1) Resident in the Language Area – 8 municipalities (n=5,677)	0.69	0.63–0.74	**
I1) Sámi linguistic connection≈objective criterion#3 (n=5,249*)	0.57	0.53–0.62	**
I2) Self-identification as Sámi=subjective criterion (n=3,112*)	0.60	0.55–0.67	**
I3) Sámi as active language (n=2,614*)	0.55	0.50–0.61	**
Self-rated health: good vs. not good
G1) Resident in the Language Area – 8 municipalities (n=5,677)	0.98	0.91–1.06	0.641
I1) Sámi linguistic connection≈objective criterion# (n=5,249*)	0.83	0.77–0.89	**
I2) Self-identification as Sámi=subjective criterion (n=3,112*)	0.89	0.81–0.98	***
I3) Sámi as active language (n=2,614*)	0.88	0.80–0.98	***

^*Assignment to I1–I3 is based on self-reported data. All OR adjusted for sex and age and take into account that n varies because of missing values.

^#The criterion for enrolment in the Sámediggi electoral roll applies to 4 generations, our data cover 3.

^**p<0.0001

^***p<0.05.