Title: The effect of socioeconomic status on overall mortality after diagnosed with colorectal cancer. Background: Social inequality in health persists even in egalitarian and highly developed countries. There is a social gradient found in incidence and mortality from several non-communicable diseases, cancer being one of them. Colorectal cancer is the third most common form of cancer and one of the most common causes of cancer death worldwide. Survival analysis is an appropriate way of analysing time and cause to such an event. Objective: To estimate the effects of socioeconomic status on overall mortality after diagnosed with colorectal cancer in a Norwegian cohort, and to compare results from the Cox proportional hazard and the Weibull models to see how they differ. Method: An individual level dataset conducted from a randomized controlled trial for colorectal cancer, Norwegian colorectal cancer prevention (NORCCAP), was used. 1298 individuals diagnosed with colorectal cancer were included. The Cox Proportional Hazard model and the Weibull model were applied to calculate hazard ratios for mortality risk by level of education, income, family status and place of residence, adjusted for age, gender and period. The multivariable models were compared regarding their predictive power using Harrell`s C-statistic and by survival curves. Results: An education level of 10-12 years and an income of 200 000 – 300 000 NOK and 400 000 – 600 000 had a positive impact on mortality compared to a low level of education and lower income. Family status and place of residence did not have a significant effect on mortality. The Cox Proportional Hazard model and the Weibull Proportional Hazard model gave almost identical hazard ratios and p-values, but the Cox model gave a slightly higher predictive power than the Weibull model. The survival curves gave quite different results even though the multivariable analyses gave virtually identical hazard ratios and p-values. Conclusion: Socioeconomic status seems to have an impact on mortality after diagnosed with colorectal cancer given that education and income had a significant effect on mortality. Similar hazard ratios from the Cox model and the Weibull model nevertheless gave some deviations in estimated survival over time in these data.