This study explored an alternative self-report approach to the measurement of depressive symptoms in children that was hypothesized to be less prone to the distorting influences of defensiveness. Children with cancer (n = 107) and healthy controls (n = 442) completed measures of adaptive style (defensiveness, anxiety), a standard depression inventory, and an anhedonia measure used as a proxy estimate of depressive symptoms. As predicted, children with cancer reported significantly fewer depressive symptoms than did healthy controls on the depression inventory, whereas no differences were found on the measure of anhedonia. However, self-report of anhedonia was found also to be subject to the influence of defensiveness, and neither the depression inventory nor the anhedonia measure was significantly related to parent and physician ratings of depression. An approach that combined self-report measures of depression and anhedonia did not significantly improve the identification of children rated as depressed by parents or physicians. Measurement of anhedonia may provide an interesting avenue for further research, but there is still no adequately validated self-report instrument for the measurement of depressive symptoms in children with cancer.