Background: Exposure to environmental tobacco smoke (ETS) presents substantial health risks for pregnant women and newborn infants. Measurements of ETS include invasive and expensive biochemical tests, as well as less invasive and lower-cost, self-reported exposure and avoidance measures. Better understanding of self-report measures will help to select ETS assessments for evaluation. Methods: This analysis was conducted within the context of a tailored video intervention to reduce tobacco smoking and ETS exposure during pregnancy and after delivery in the control group sample of 147 nonsmoking women. Measurements of salivary cotinine concentration, self-reported ETS exposure, and avoidance behaviors were captured at 32 weeks’ gestation and 6 months postpartum. Results: Salivary cotinine concentration was significantly related to ETS avoidance among pregnant nonsmokers at 32 weeks’ gestation, but not ETS exposure. At 6 months postpartum, both the reported ETS exposure of the infant and maternal avoidance behaviors to reduce her infant’s exposure were associated with the infant’s salivary cotinine concentration. At 32 weeks’ gestation and 6 months postpartum, avoidance behaviors decreased as exposure increased. Discussion: This study suggests that for nonsmoking women during pregnancy, reports of tobacco smoke avoidance are more valid than reports of exposure. After delivery, self-reported ETS exposure or avoidance are associated with each other and the biochemical measurement of salivary cotinine. These results provide researchers and clinicians with evidence to support the inclusion of avoidance behaviors in the selection of ETS measures.