Table 7. Multiple hierarchical linear regression results with A1C as dependent variable.

Analysis was conducted after the logarithmic transformation of the dependent variable. +regression coefficient (Standard Error) ¹scale from 1(None) to 4(University/MSc) ²scale from 1 (< 400) to 4 (> 1,500). Statistical significance was set at p < 0.05. A1C- Glycosylated hemoglobin, BMI - Body mass index, HL - Health literacy, SE - Self-Efficacy.

Step		β(SE)+	P
1: F(8,111) = 7.49; p < 0.001; R2 = 0.30	Age	-0.003 (0.001)	< 0.001
	Gender (women vs men)	-0.031 (0.013)	0.020
	Family status (Unmarried/ Divorced/ Widowed vs Married/ Living with partner)	0.049 (0.016)	0.003
	Educational status1	-0.009 (0.010)	0.357
	Monthly income2	-0.012 (0.008)	0.133
	ΒΜΙ (Kg/m2)	0.002 (0.001)	0.023
	Chronic disease (yes vs no)	-0.008 (0.024)	0.730
	HL	-0.009 (0.003)	0.003
2: F(9,110) = 7.20; p < 0.001; R2 = 0.32, R2 change =.02	Age	-0.003 (0.001)	< 0.001
	Gender (women vs men)	-0.031 (0.013)	0.018
	Family status (Unmarried/ Divorced/ Widowed vs Married/ Living with partner)	0.050 (0.016)	0.002
	Educational status1	-0.010 (0.010)	0.309
	Monthly income2	-0.012 (0.008)	0.124
	ΒΜΙ (Kg/m2)	0.002 (0.001)	0.119
	Chronic disease (yes vs no)	-0.016 (0.025)	0.503
	HL	-0.007 (0.003)	0.039
	SE total	-0.002 (0.001)	0.030