Background: Ventilation is the key intervention to resuscitate non-breathing newborns. Positive end-expiratory pressure (PEEP) may facilitate lung-liquid clearance and help establish functional residual capacity. Objectives: The aim of this study was to describe how mask leak and ventilation rates affect delivered PEEP and tidal volumes during newborn resuscitations using a self-inflating bag with an integrated PEEP valve. Methods: This was an observational study including near-term/term newborns who received bag-mask ventilation (BMV) with a new self-inflating bag with a novel 6 mbar PEEP valve, without external gas flow, between October 1, 2016 and June 30, 2018 in rural Tanzania. Helping Babies Breathe-trained midwives performed most of the resuscitations. Pressures and flow were continuously measured and recorded by resuscitation monitors. Results: In total, 198 newborns with a median gestation of 39 weeks (25th, 75th percentiles 37, 40) and birth weight of 3,100 g (2,580, 3,500) were included. The median delivered PEEP and expired (tid-al) volume at different levels of mask leak were 6.0 mbar and 11.3 mL/kg at 0-20% mask leak, 5.5 mbar and 9.3 mL/kg at 20-40%, 5.2 mbar and 7.8 mL/kg at 40-60%, 4.6 mbar and 5.0 mL/kg at 60-80%, and 1.0 mbar and 0.6 mL/kg at 80-100% mask leak. A high ventilation rate (> 60/min) nearly halved expired volumes compared to < 60/min for 0-60% leak. The BMV rate had a negligible effect on peak inflation pressure (PIP) and PEEP. Conclusions: Mask leak up to 80% did not impair the provision of recommended PEEP or tidal volumes during BMV with a self-inflating bag. High or low ventilation rates did not significantly affect PIP or PEEP. Expired volumes were reduced at ventilation rates > 60/min.