Various methods can be used for assessing symptoms of depression. These include check lists, scales for patient self-rating and rating scales completed by trained professional observers. Each technique has its special place in psychiatric research and each technique has its own strengths and weaknesses. A common concern on the part of psychiatrists is the fear that numerical assessment is dehumanizing. This is not the case as this method is intended only to make statistical analysis convenient and in no way reduces the concern for the subject's individuality. A commonly used depression rating scale, the Hamilton, has been extensively used in clinical trials of antidepressant drugs and for other purposes in clinical research. The various items of that scale are discussed and the system of rating explored from a clinician's point of view.