Hand preference (HP) is a major behavioural variable for inferring individual differences in neurological organisation. Yet, despite the extensive use of HP measures for both research and clinical purposes, there is little research on the properties of the structural features of HP questionnaires, the most common method of HP measurement. We study here the effects of different response formats of HP questionnaires on participants' responses. A total of 200 volunteers (100 left-handers, 50 male; 100 right-handers, 50 male) completed two versions of a composite HP questionnaire containing the items of the Annett, Edinburgh, Healy, and Waterloo instruments, one with a binary response format and the other with a 5-point graded response format. The tendency towards extreme responses and the mapping between a graded either response and a binary left/right response were investigated. It was found that both patterns were significantly influenced by participant handedness, but not by sex. Finally, in terms of the overall ordering of participants by HP measures, considerable consistency was found across questionnaire content and response format. © 2013 Copyright Taylor and Francis Group, LLC.