Early Spelling Acquisition: Does Writing Really Beat the Computer?

Abstract

This research evaluated the efficacy of handwriting, letter tiles, and computer-based instruction on the early spelling acquisition of normal achieving (NLD) and learning disabled (LD) elementary students. The study replicated and extended the Cunningham and Stanovich (1990) study. Forty-eight primary-grade students (24 non-learning disabled; 24 learning disabled) received spelling training under three experimental conditions that involved different types of instructional activity: writing, sorting letter tiles, or typing on the computer. Results indicated no significant differences between the LD and NLD groups on words spelled correctly for any of the three conditions. Since the number of spelling words learned was low across conditions and groups, the number of correctly learned bigrams was examined. No within-group differences emerged for condition (writing, tile, computer); however, significant differences between the groups were found on number of bigrams learned for writing, tile, and computer, with the NLD group outperforming the LD group on all three conditions. Statistically significant time effects for bigrams were also noted for all three conditions. Unlike the Cunningham and Stanovich (1990) study, results did not replicate the superiority of the handwriting condition for the NLD group, nor was the handwriting condition found to be significantly more effective for the LD group.