The purpose of this study was to determine the efficacy of three motoric conditions (writing, tracing, and computer keyboarding) on the spelling performance of 24 third- and fourth-grade students without learning disabilities (NLD) (15 males, 9 females) and 24 third- and fourth-grade students with learning disabilities (LD) (16 males, 8 females). This study applied empirically based procedures for teaching spelling, examined student performance over time, and incorporated student interviews concerning their preference for motoric condition. For number of words spelled and proportion of bi-grams (correct letter sequences) used correctly, significant differences were found between the LD and NLD groups, both at posttest and follow-up, with the NLD students learning to spell more words and apply more correct bi-grams than the students with LD. There was no significant effect for condition for either words spelled or bi-grams, indicating that students did not learn significantly more words in the writing, tracing, or computer condition. There was also a significant time effect indicating that the accuracy of both groups decreased over time from posttest to follow-up for both words and bi-grams. Interviews revealed that students in both groups preferred the computer condition; however, they believed that they learned best in the writing and tracing conditions.