Introduction: Since January 2000, standard presenter evaluation forms have been made available to grand rounds organizers in the Department of Medicine, University of Toronto. During the 2000-2001 academic year, effort was directed at the accumulation of evidence for the validity of the results generated.
Methods: Two issues were addressed: the integrity or coherence of the form itself and the number of forms or evaluations required to achieve a stable estimate of the construct "presenter effectiveness" for an individual presenter.
Results: Positive evidence relating to the integrity of the form is presented and the number of evaluations or ratings required to provide a stable estimate of presenter effectiveness is suggested.
Discussion: Most presenters' ratings were distributed in a narrow range. Ranking of individual presentations would require exceptionally high precision. Separation into groups requires less precision. This type of classification appears sufficient to enable planning decisions.