The purpose of the present investigation was twofold: first, to examine the relationship between goal difficulty, goal specificity, and endurance performance in a physical activity setting, and second, to determine the relationship between different types of information feedback, goals, and performance. Subjects (N = 94) performed on a hand dynamometer endurance task, being asked to hold a one-third maximum contraction for as long as possible. The subjects were randomly assigned to one of the following goal-setting conditions: (a) Do your best, (b) improve by 40 s, or (c) improve by 70 s. They were provided with either concurrent or terminal feedback in a 2 x 3 x 2 (feedback x goals x trials) design. Performance results indicated a significant goals-by-trials interaction with the 40- and 70-s goal groups exhibiting significantly more improvement than the "do your best" group. No significant performance differences were found between the two feedback groups. However, significant differences in the performance-associated cognitions of the feedback groups indicated a preference for concurrent feedback as an adjunct to goals. Results are discussed in terms of Locke's goal-setting theory as well as some recent field research investigating the goal-setting performance relationship in physical education settings.