In order to inform discussions about possible changes to Medical Education's blinding policy, members of the journal's editorial board were interested in discovering reviewers' and authors' preferences with regard to the current double-blind policy and various alternatives. In September 2005, an 8-question, web-based survey was sent to all authors and reviewers who had submitted or reviewed a manuscript for Medical Education in 2003 and 2004 (n = 2632). The questions asked about authorship and reviewing experiences and preferences regarding 5 types of blinding procedure, from double-blinding to fully unblinded, open reviews. Following 2 electronic mailings, 838 surveys were completed. There was a range of experience among respondents, with a high proportion of experienced authors (49% with over 20 publications) and reviewers (41% with over 20 reviews). Overall, 68% of respondents preferred a review process that concealed author names and 72% preferred a process that allowed for concealment of reviewer names. Less experienced authors and reviewers were significantly more likely to prefer concealing author names, but even the most experienced respondents had a 54% preference for author concealment. Reasons for concealing identities included facilitating fairness and honesty in reviews and acknowledging the need to avoid personal conflicts or rivalries. Reasons for revealing identities included facilitating greater transparency and accountability, and a better understanding of the author's and reviewer's contexts and credentials. The Medical Education authors and reviewers who chose to respond to the survey voted strongly in favour of continuing the double-blinding procedure of concealing both author and reviewer identities during the review process.