A statistical model of dietetic intern selection was developed from a profile of selection criteria that was obtained in a 1988 survey of Canadian graduate dietetic internship directors. The model was composed of four clusters of variables that resulted from the most frequently used selection criteria: academic performance, work experience, communication skills, and extracurricular activities. Data from a convenience sample of 39 dietetic intern applicants were analyzed, using principal components analysis and discriminant analysis, to test the model's power to predict success in obtaining an internship appointment. In descending order, the criteria with the greatest predictive powers were: academic performance; extracurricular activities; and supervisory, teaching, or instructing types of work experience. The model accounted for 41% of the differences between those who were successful and those who were not successful in obtaining internship appointments in 1989 and correctly classified 30 of 39 subjects. These results provide baseline data on the predictive power of some criteria used for selecting dietetic interns. These findings suggest the need for a replication study with a randomized national sample to crossvalidate the results obtained in this exploratory research.