SUMMARY The construction and development of an inventory of 69 questions dealing with the subjective assessment of obsessional traits and symptoms is described. A card-sorting procedure requiring supervision is used, which gives rise to information about feelings of resistance and interference with other activities, in addition to the straightforward answers to the questions. The resulting scores are shown to differentiate well between a group of selected obsessional patients and normals. The problem of to what extent replies to apparently specific questions in inventories and questionnaires can be interpreted as having only general significance is discussed.

The Leyton Obsessional Inventory