This paper examines the hypothesis that parents exaggerate their reading with children aged 3 to 5 when asked typical single-item questions and that the extent of exaggeration is greater for better-educated parents. It examines differences in parental reporting of reading to children that may result from differences in response bias. It examines whether differences in reading with children by race/ethnicity, income, and family structure hold up after controlling for maternal education and other factors. Finally, it examines whether any bias we find affects the relationship between reading and achievement test scores. Data are from the Child Development Supplement to the Panel Study of Income Dynamics, a nationally representative sample of children and their parents who were asked detailed questions about their lives and activities in 1997.