OBJECTIVE This article examines the association between maternal education, smoking and other risk factors and small-for-gestational-age (SGA) births. METHODS The data are from the 1994/95 National Longitudinal Survey of Children and Youth. The analysis was restricted to a subsample of 4,181 children younger than age 2 and was based on information provided by their biological mothers. METHODS Logistic regression was used to estimate the odds ratios for SGA by maternal education, controlling for maternal smoking during pregnancy, household income, family status, maternal age at birth of child, and use of prenatal care. RESULTS Maternal education and smoking during pregnancy appear to have independent effects on SGA, after controlling for other risk factors. The effects of maternal education, smoking and other risk factors are likely underestimated, as the analysis pertains only to children who had survived at the time of the interview.