Using the generalized residual method, this paper investigates the assumption that children are exogenous in the labour supply equations of married women via alternative regression specifications for the children variables. The empirical results of this paper suggest that children are endogenous to the participation decision and likely to be exogenous to the hours of work decision for married women. Children under age six have a dramatic negative impact on the labour force participation of married women. These findings support the implication of the literature relating fixed costs of employment, including child care costs, to female labour supply.