Duckling survival is an important component of mallard (Anas platyrhynchos) recruitment and population growth, yet many factors regulating duckling survival are poorly understood. We investigated factors affecting mallard duckling survival in the drift prairie of northeastern North Dakota, 2006–2007. Mammalian meso-predators were removed by trapping on 4 92.3 km² study sites and another 4 study sites served as controls. We monitored 169 broods using telemetry and periodic resighting, and we modeled cumulative survival to 30 days of age (Ŝ_30d) using the nest survival module in Program MARK. Duckling survival was not affected by predator removal (Ŝ_30d-Control = 0.208, 85% CI: 0.182–0.234; Ŝ_30d-Trapped = 0.183, 85% CI: 0.155–0.211) and was only weakly negatively correlated with duckling density. Duckling survival was higher in 2007 (Ŝ_30d-2007 = 0.264, 85% CI: 0.193–0.355) than 2006 (Ŝ_30d-2006 = 0.157, 85% CI: 0.084–0.252) and increased with total seasonal and semipermanent wetland area and declined with perennial cover in the surrounding landscape. Broods that hatched earlier in the season (especially in 2006) and ducklings that were heavier at hatch also had higher survival. Our estimates of duckling survival are among the lowest reported for mallards and contradict previous research in Saskatchewan that found predator removal increased duckling survival. However, our results are consistent with other studies suggesting that earlier hatch date, increased wetland availability, and better duckling condition lead to increased survival. Management actions that increase wetland density, improve nest success early in the season, and potentially target brood-specific predators such as mink (Neovison vison) would likely lead to higher duckling survival.