The handedness of 1599 college students was assessed using a modification of Annett's (1967) inventory. The addition of a strength of preference scale makes the "mixed" classification a better indicator of ambidexterity than in the original inventory by eliminating from it subjects who show preference for the same hand for nearly all tasks. Contrary to previous reports, sex was not significantly related to handedness. It was found that a family history of left-handedness was significantly related to the handedness of the subject. The present form of the Annett inventory provides an easily scorable inventory for determining handedness of large groups as well as for use in individual testing situations. A further advantage of the present inventory is the ease with which data may be transferred by key-punch operators prior to special analysis or sorting procedures.