Trait and other sources of variance in the S-R inventory of anxiousness.

The S-R Inventory of Anxiousness has a reliability of .95 but shows only 5% of its variance due to persons (traits). This strange combination of results is explored logically and empirically. A distinction is made between the Inventory as a test and as an experimental design. Some characteristics and problems of the Inventory as an experimental design are examined. These problems suggest that it is unwarranted to include variance from situations and modes of response in the total variance against which the component of variance due to individual differences is evaluated. It is such inclusions that have yielded the result of 5% of variance due to persons. It is concluded that results from the Inventory, when generalized to other omnibus inventories of anxiety, do not have the drastic implications that have been previously claimed.