To aid in the construction of effective task analysis inventories, this technical report discusses: (1) an optimum questionnaire length that adequately covers Marine tasks without unduly fatiguing respondents; (2) procedures for the phrasing of task statements to avoid ambiguities and be understandable to as broad a range of Marines as is possible; and (3) recommendations for inventory design and format. lt is recommended that the Office of Manpower Utilization experiment with a "mini-booklet" that would substantially reduce the number of items, since some Marine Corps inventories have contained as many as 1,000 items. An experimental design is given for dividing a lengthy task inventory into a series of shorter inventories. There is an overlap of task statements in each small questionnaire booklet to provide adequate samples of responses to each item. Major attention is given to the wording of task statements and inventory instructions. Data are presented from the application of six readability formulas to nine task inventories. Reading comprehension levels of Marines at three bases are described and are compared with the comprehension levels required to understand task statements and instructions in the inventories. Guidelines for simplified wording of task inventory items and methods for measuring readability are outlined. A single booklet containing both test statements and answers is presented to replace the traditional two-booklet format. Also discussed are studies of the effects of anonymus versus identified responses to task inventories. The Appendix includes the Dale list of 3,000 most familiar words and an activity verb list. (Author/CP)