Table 4.

Logistic regression models on food insecurity by set of variables including household composition, age and education.

Variables	2014–2015		2016–2017		2018–2019
	OR	SE	OR	SE	OR	SE
Household composition (reference: two-adults household)
One-person	2.457 *	0.402	1.823 *	0.318	1.740 *	0.364
Three or more adults	0.759	0.160	0.646	0.145	0.562 *	0.153
H. with one child	0.888	0.206	0.865	0.202	0.351 *	0.109
H. with two children	0.610	0.158	0.745	0.197	0.203 *	0.077
H. with at least three children	0.824	0.318	0.803	0.366	0.193 *	0.097
Gender (reference: male)
Female	1.323 *	0.182	1.371 *	0.189	1.660 *	0.277
Age	1.091 *	0.022	1.107 *	0.025	1.071 *	0.029
Age squared	0.999 *	0.0002	0.999 *	0.0002	0.999 *	0.0003
Education (reference: tertiary)
Secondary	2.768 *	0.599	2.069 *	0.376	2.616 *	0.730
Elementary	5.260 *	1.471	3.599 *	0.994	3.927 *	1.486
Income quintile group (reference: first quintile group)
Second quintile group	0.586 *	0.115	0.360 *	0.071	0.420 *	0.099
Third quintile group	0.324 *	0.070	0.357 *	0.072	0.319 *	0.081
Fourth quintile group	0.284 *	0.064	0.343 *	0.074	0.266 *	0.076
Fifth quintile group	0.099 *	0.027	0.114 *	0.032	0.111 *	0.033
Constant	0.041 *	0.022	0.039 *	0.023	0.049 *	0.036

OR–Odds Ratio, SE–robust standard errors (with heteroscedasticity-robust asymptotic variance), * means statistical significance at 0.05.