This study investigates the factors that determine the relative household poverty which prevents the growth of social welfare in Turkey. For this purpose, Turkish Statistical Institute (Turkstat) (2017) Income and Living Conditions micro dataset was used. The poverty line was calculated by using 60% of the household median income. In the binary logistic regression analysis, poor and non-poor shares of population was determined according to the poverty line and used as the dependent variables. Household demographic characteristics, labour market conditions and geographical characteristics were employed as explanatory variables. The estimation results indicate that education, age, formal employment, and employment in industry sector decreases relative household poverty. On the contrary, unemployment, causal employee, marriage and household size increase the risk of household poverty.