Empirical analyses of the determinants of life satisfaction routinely include the number of children as one of the socio demographic controls, without explicitly considering that, for a given household income, more children imply a lower level of income per family member. The variable "number of children" then often attracts a negative or insignificant coefficient. Using data from the German Socio Economic Panel 1984-2007 we confirm that the sign of the coefficient for the variable "number of children in the household" is negative when introducing household income without correction for the number of members in a life satisfaction regression. On the contrary, when we equivalise income with the most commonly adopted equivalence scales, so eliminating the monetary cost of raising children, the impact of the variable is positive and significant when a high level of economies of scale is assumed. Our results however lead us to reject slope homogeneity as we find strong differences by gender and region. In particular, the positive effect of children on life satisfaction is stronger for males and East Germans. We interpret these subsample split results as driven by heterogeneous opportunity costs and cultural traits.