In this paper, the importance of a pressure correction of viscosity data obtained in capillary melt rheology is demonstrated. A linear polycarbonate has been chosen as a highly pressure-sensitive material for which data obtained by rotational rheometry does not overlap with capillary data. This apparent problem with the Cox-Merz relation is attributed to the existence of a mean pressure inside the capillary which is significantly different from atmospheric conditions. Different methods to determine the pressure coefficient of polycarbonate have been evaluated based on experiments performed with a capillary rheometer equipped with a pressure chamber. It is demonstrated that the pressure coefficient obtained at constant shear stress and the pressure coefficient obtained by the superposition method represent accurate pressure coefficient values. Two approaches are proposed to correct the original capillary data. In the direct methodology, the pressure coefficient is used to rescale the mean pressure inside the capillary to atmospheric conditions. The indirect approach consists of first constructing a mastercurve at a certain reference pressure using capillary data obtained with a pressure chamber. The resulting mastercurve can then be rescaled to atmospheric conditions. It is shown that both methods lead to viscosity curves on which both rotational and capillary data overlap, hence confirming the Cox-Merz relationship for polycarbonate. The indirect method is proven to be advantageous since it opens the possibility to significantly extend the shear rate window in which viscosities can be measured.