Research has established the multidimensional nature of student ratings of teaching, but debate continues concerning the use of multiple- versus single-item ratings for summative evaluation. In this study the usefulness of global items in predicting weighted-composite evaluations of teaching was evaluated with a sample of 17,183 classes from 105 institutions. In separate regression analyses containing 2 global items-one concerning the instructor, the other concerning the course-each global item accounted for more than 50% of the variance in the weighted-composite criterion measure. Student, class, and method items generally added much less variance. These results suggested that, because global items accounted for a substantial amount of the variance, a short and economical form could capture much of the information needed for summative evaluation and longer diagnostic forms could be reserved for teaching improvement.