We investigated the effectiveness of two teaching formats that fall under the canopy of Mosston and Ashworth’s (2008) practice style, on fifth grade students’ motor skill performance and task engagement. Both formats are also known as station teaching or learning centers. In the teacher-rotated format (TR), the teacher decides the amount of time apportioned during practice at each station, whereas in the learner-rotated format (LR), each learner decides on task order and the amount of time spent at each station. Ten-year-old children (N = 60) were randomly assigned to the TR group (n = 20), the LR group (n = 25), and a control group (n = 15). A soccer dribbling test was employed to evaluate the soccer dribbling skill prior to and after the instructional intervention. The same soccer dribbling tasks were taught to the learners in both treatment groups in eight 30-min sessions. ANCOVA on the posttest scores showed a significant difference between the experimental groups and the control group (p < .001) and between the two experimental groups, favoring the LR group (p < .001). A 3 × 2 (Group × Test) repeated measures ANOVA showed a significant improvement of the soccer dribbling skill for both teaching formats (p < .001) but not for the control group. An ANOVA on the overall practice trial data yielded significant differences between the two formats, favoring the LR group (p < .001). Both formats were found to be effective, but the results indicate that given the opportunity to reapportion their practice time, learners in the LR format took advantage of this opportunity and improved their performance further.