Purpose An increasing percentage of children are born to couples who cohabit but are not legally married. Using data from a nationally representative Canadian sample, we estimated associations of maternal marital and cohabitation status with stillbirth, infant mortality, preterm birth (PTB), and small- and large-for-gestational-age (SGA and LGA) birth. Methods The 2006 Canadian Birth-Census Cohort was created by linking birth registration data with the 2006 long-form census. We used log-binomial regression to estimate risk ratios (RRs) for adverse birth outcomes associated with being single or living with a common-law partner. Analyses were adjusted for maternal age and education. Results Data were analyzed for 130,931 singleton births. Adjusted RRs (95% confidence intervals) for single mothers compared with married mothers were 1.92 (1.51-2.42) for stillbirth, 2.08 (1.55-2.81) for infant mortality, 1.36 (1.27-1.46) for PTB, 1.31 (1.22-1.39) for SGA birth, and 0.95 (0.90-1.01) for LGA birth. Adjusted RRs for cohabiting mothers compared with married mothers were 0.93 (0.74-1.16) for stillbirth, 1.05 (0.81-1.35) for infant mortality, 1.09 (1.03-1.15) for PTB, 1.05 (0.99-1.10) for SGA birth, and 0.96 (0.92-1.00) for LGA birth. Conclusions In a nationally representative Canadian birth cohort, cohabiting and legally married women experienced similar birth outcomes, but most outcomes for single women were substantially worse.