Skip to main content
. 2014 Oct 2;14:1027. doi: 10.1186/1471-2458-14-1027

Table 3.

Multi-level logistic models for any physical violence against unmarried female adolescents in Bangladesh

aOR (95% CI) Unmarried female adolescents
n = 4370
Intercept 0.50 (0.41-0.61)*
Community-level variables
Community wealth (poor ref)
Non-poor 1.28 (0.96-1.72)
% literate (female) 0.68 (0.26-1.79)
% accepting of physical punishment of adolescents 5.42 (2.52-11.64)*
% households belonging to NGO 1.99 (0.88-4.54)
Area (rural ref)
Urban 0.98 (0.69-1.39)
Individual-, family-, and household variables
Age 0.65 (0.63-0.68)*
Religion (non-Muslim ref)
Muslim 1.20 (0.91-1.58)
Household wealth (poor ref)
Non-poor 1.00 (0.84-1.19)
NGO membership (non-membership ref)
Household member of NGO 1.05 (0.88-1.25)
Random effects tau (SE)
Community level 0.53 (0.10)
District level 0.43 (0.12)

*p < 0.01, Significance of random effects not estimated.