Purpose This study assessed the effect of marital status on stage at diagnosis and survival in women with cervical cancer. Methods Cervical cancer cases diagnosed between 2000 and 2010 were identified from the Surveillance, Epidemiology and End Results (SEER) program. Patient demographic and clinical characteristics were compared by marital status. Multivariate logistic and Cox proportional hazard regression models were performed to calculate odds ratios of advanced stage at diagnosis and hazard ratios of death risk respectively. Results Among 31 425 women, 46% of cases were married at the time of diagnosis. Married women were more commonly diagnosed at a localized stage (55%) compared to other non‐marital groups (47% of singles, 42% of separated/divorced, and 28% of widowers, p < 0.001). After controlling for age, race/ethnicity, period of diagnosis, histology, and SEER area, single [adjusted odds ratio (aOR) 1.41; 95% Confidence Interval (CI) 1.33–1.49], separated/divorced [aOR 1.44; 95% CI 1.34–1.55], and widowed women [aOR 1.43; 95% CI 1.31–1.58] were all more likely to be diagnosed at an advanced stage compared to married women. In terms of prognosis, single (adjusted hazard ratio (aHR) 1.35; 95% CI 1.28–1.43), separated/divorced (aHR 1.22; 95% CI 1.15–1.29), and widowed women (aHR 1.28; 95% CI 1.19–1.36) had significant increased risk of death compared to married women. Adjusting for insurance status did not change the findings. Conclusion Being married is associated with earlier diagnosis and a more favorable prognosis for cervical cancer among US women. Interventions to improve prognosis for unmarried women, including increasing use of cervical cancer screenings, are warranted.