A meta-analysis found that attending full-day (or all-day) kindergarten had a positive association with academic achievement (compared to half-day kindergarten) equal to about one quarter standard deviation at the end of the kindergarten year. But the association disappeared by third grade. Reasons for this fade-out are discussed. Social development measures revealed mixed results. Evidence regarding child independence was inconclusive. Evidence was suggestive of a small positive association between full-day kindergarten and attendance and a more substantial positive association with the child's self-confidence and ability to work and play with others. However, children may not have as positive an attitude toward school in full-day versus half-day kindergarten and may experience more behavior problems. In general, the research on full-day kindergarten would benefit from future studies that allow strong causal inferences and that include more nonacademic outcomes. The authors suggest that full-day kindergarten should be available to all children but not necessarily universally prescribed. (Contains 7 notes and 6 tables.)