Visual analogue scales (VAS) and verbal rating scales (VRS) are widely used to assess pain. This paper presents a secondary analysis of a subsection of data collected as part of an evaluation of an intervention to improve nighttime pain. The aims were to describe the relationship between the VAS and VRS; to compare characteristics of the noncompliant groups in terms of age, gender, type of surgery, and analgesics; to explore the reasons for noncompletion of the VAS or VRS; and to determine the noncompliance rates with each assessment scale. Worst and average pain scores were obtained for the second postoperative night for 417 patients. Fifty-nine patients (14.2%) did not complete the VAS. In contrast, two patients did not complete the VRS (0.5%). The scores generated from the VAS and VRS correlated well, although the range of VAS scores corresponding to each VRS category was wide. We conclude that the VRS was more suitable for use in this clinical setting.