| Literature DB >> 2669989 |
Abstract
Controlled experiments can be used to study the effects on health of behaviors that cannot be perfectly controlled. A simple statistical technique allows causal effects to be distinguished from selection effects. The technique is applied to measure the effect of maternal smoking on birth weight.Mesh:
Year: 1989 PMID: 2669989
Source DB: PubMed Journal: Biometrics ISSN: 0006-341X Impact factor: 2.571