Working Paper: CEPR ID: DP5362
Authors: Marco Francesconi; Stephen P. Jenkins; Thomas Siedler
Abstract: We analyse the impact on schooling outcomes of growing up in a family headed by a single mother. Growing up in a non-intact family in Germany is associated with worse outcomes in models that do not control for possible correlations between common unobserved determinants of family structure and educational performance. But once endogeneity is accounted for, whether by using sibling-difference estimators or two types of instrumental variable estimator, the evidence that family structure affects schooling outcomes is much less conclusive. Although almost all the point estimates indicate that non-intactness has an adverse effect on schooling outcomes, confidence intervals are large and span zero.
Keywords: childhood family structure; educational success; instrumental variables; lone parenthood; sibling differences; treatment effects
JEL Codes: C23; D13; I21; J12; J13
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Nonintact family structure (J12) | Worse educational outcomes (I24) |
| Nonintact family structure (J12) | Less likely to be in top secondary school track at age 14 (I24) |
| Maternal employment and income (J12) | Adverse effects of family structure on educational attainment (J12) |
| Unobservable family background characteristics (J79) | Educational outcomes (I21) |
| Instrumental variable methods (C36) | Less conclusive evidence for nonintact family structure's effect on educational outcomes (J12) |
| Sibling-difference estimators (J79) | Less conclusive evidence for nonintact family structure's effect on educational outcomes (J12) |