This study used data from the Panel Study of Income Dynamics and its 1997 Child Development Supplement to examine how family income matters for young children's development. The sample included 753 children who were between ages 3 and 5 years in 1997. Two sets of mediating factors were examined that reflect two dominating views in the literature: (1) the investment perspective, and (2) the family process perspective. The study examined how two measures of income (stability and level) were associated with preschool children's developmental outcomes (Woodcock-Johnson [W-J] Achievement Test scores and the Behavior Problem Index [BPI]) through investment and family process pathways. Results supported the hypothesis that distinct mediating mechanisms operate on the association between income and different child outcomes. Much of the association between income and children's W-J scores was mediated by the family's ability to invest in providing a stimulating learning environment. In contrast, family income was associated with children's BPI scores primarily through maternal emotional distress and parenting practices. Level of income was associated with W-J letter-word scores and income stability was associated with W-J applied problem scores and BPI, even after all controls were included in the models.