Objective Studies to date have found little correlation between subjective and objective measures of cognitive function in cancer patients, making it difficult to interpret the significance of their cognitive complaints. The purpose of this study was to determine if a stronger correlation would be obtained using measures of cognitive change rather than static scores. Methods Sixty women with early stage breast cancer underwent repeated cognitive assessment over the course of chemotherapy with a neuropsychological test battery (objective measure) and with the FACT‐Cog (subjective measure). Their results were compared to 60 healthy women matched on age and education and assessed at similar intervals. We used multilevel modeling, with FACT‐Cog as the dependent measure and ordinary least squares slopes of a neuropsychological summary score as the independent variable, to evaluate the co‐variation between the subjective and objective measures over time Results Measures of both objective and subjective cognitive function declined over the course of chemotherapy in the breast cancer patients but there was no significant relationship between them, even when using change measures. Change in objective cognitive function was not related to change in anxiety or fatigue scores but the decline in perceived cognitive function was associated with greater anxiety and fatigue. Conclusions The discrepancy in objective and subjective measures of cognition in breast cancer patients cannot be accounted for in terms of a failure to use change measures. Although the results are negative, we contend that this is the more appropriate methodology for analyzing cancer‐related changes in cognition.